Vol. 96, No. 5 Archives – Southern California Law Review

The Healthcure System: A Regional Accountable Care Model to Remedy Healthcare’s Pricing Problem

March 2024March 2024Christopher A.G. LoCascio*

INTRODUCTION

The most sinister game show in American life commences every time a hospital provides care, draws up an eye-popping bill, and asks its patient how it will be paid.¹ Imagine waking up from a medically induced coma to the words, “Will that be cash or card?” In its own sick twist of the three-legged race, the healthcare system effectively binds patients’ ability to navigate the costs of their care, such that even those with insurance are often left hobbling for answers to the questions, what must I pay and why? The stakes are ever graver for those underinsured or uninsured. In this game, the winners are not those who make it out alive, but those who can afford to keep on living.

In a properly functioning market, supply and demand would theoretically prevent a hospital from wildly inflating its prices, such as charging a patient approximately $200.00 for a routine blood test that would otherwise cost $13.00, but the reality is that healthcare is no such market.² As Princeton Professor Uwe Reinhart put it, “In effect, [patients] enter that market like blindfolded shoppers pushed into a department store to shop around smartly for whatever item they might want or, in the case of health care, need.”³

Patients cannot expect to make informed decisions when they lack reliable pricing information before seeking care, but this is just one of the many market failures that drives the ever-skyrocketing costs of American healthcare.⁴ Hospital administrators themselves often fail to grasp the true costs of their services, as do doctors when ordering tests and writing prescriptions.⁵ In fact, healthcare providers often have financial and legal incentives to overtreat their patients, at the patients’ expense.⁶ When providers are paid based on the services they render, in what is known as the “fee-for-service” payment model, providers that do more, make more.⁷ In the fear of the dreaded medical malpractice lawsuit, providers have an incentive to cover their bases and test for everything, no matter the cost.⁸ Even insurance companies, which foot the providers’ bills, stand to gain from exaggerated costs.⁹ Unlike insurance companies in other sectors, which derive profit by spending as little of policyholders’ premiums as possible, health insurance companies have incentives to maximize their spending because regulations cap their profits at a certain percentage of their expenditures.¹⁰ Essentially, insurers earn more when they spend as much of their beneficiaries’ premiums as possible.¹¹ In what is perhaps the most perplexing market failure of them all, individuals continue to pay the rising premiums, copays, coinsurance, and taxes that feed the hungry, hungry healthcare hippo.¹² Yet, the pricing problem snuck up on no one—the rubbery, rotund river beast of a healthcare system has slowly barreled through America’s regulatory swamp for a century as landlocked policymakers repeatedly tried and failed to halt its growth by trying different reimbursement models, competition enhancements, and delivery programs.¹³

The most promising opportunity to impose downward cost pressure on the healthcare system came in 2010 with the advent of the Accountable Care Organization (“ACO”) concept as part of the Patient Prevention and Affordable Care Act (“ACA”).¹⁴ ACOs are networks of healthcare providers that coordinate care, integrate finance and delivery, and share in financial gains and losses.¹⁵ A primary goal of the ACO is to achieve a more cost-efficient system that incentivizes preventive care and integrated treatment while limiting incentives to drive up costs.¹⁶ While they come in many forms, ACOs often accomplish cost savings by allocating a set amount of money to providers for each patient they treat, known as capitated payments, thereby exposing providers to the risk of outspending that amount if they do not keep costs down.¹⁷ In essence, ACOs are financially accountable for the care of a particular population.¹⁸

ACOs remain a largely underdeveloped concept with as-yet-unsolved complications.¹⁹ For instance, the Center for Medicare & Medicaid Innovation (“CMMI”) pilots a wide variety of ACO models, ranging from “one-way risk” models with no downside and modest upside to “two-way risk” models with varying levels of risk and reward.²⁰ Providers are free to choose how much risk of cost overruns they would like to take on, and they get to keep a proportionate amount of any cost savings.²¹ For example, one ACO may choose a one-way risk model with a 0% risk of losing money and a 3% share of cost savings, whereas another ACO may choose a two-way risk model that exposes it to a 10% risk of any cost overruns but entitles it to 30% of any cost savings.²² Naturally, the more risk a provider faces, the greater the incentive to cut costs.²³ The prospect of a greater reward has not proven persuasive for ACOs to adopt riskier models, however, and all but the least risky models have struggled to attract provider participation.²⁴ Moreover, ACOs fail to address the demand-side concern of consumers’ continued payment for health insurance despite increases in rates—a phenomenon known as the price inelasticity of health insurance premiums.²⁵ Equally alarming among these concerns is the antitrust component.²⁶ In an environment in which providers are already consolidating, ACOs stand to exacerbate a shrinking market and empower consolidated provider networks to wield unmatched pricing power.²⁷

Despite the promise of new models, the healthcare system remains in a precarious position. The ACA has been left on unstable footing following the repeal of the individual mandate tax penalty in 2017²⁸ and efforts to either fully repeal or replace the law altogether.²⁹ Meanwhile, insurance premiums have continued to rise.³⁰

This Note proposes a novel framework from which to develop pilot programs for future healthcare regulations and legislation. In doing so, this Note will identify certain regulatory factors that contribute to, or at least fail to stop, the upward march of healthcare prices, and propose a novel alternative model that delivers on the three pillars of healthcare: broad access, low cost, and high quality.³¹ Here, access refers to both access to coverage of costs and access to care. Quality refers to both the breadth of covered benefits and health outcomes.

This Note takes a law and economics approach to healthcare, focusing on the information asymmetry, moral hazards, principal-agent problems, adverse selection, and misaligned incentives that contribute to healthcare’s current market failures.³² To solve these problems, this Note prescribes a new outcomes-based model that aligns the incentives of patients and providers by tying provider funding to certain health indicators. The proposed healthcare model, titled the Healthcure System, achieves universal coverage through regional healthcare districts that draw on the funding model of employer-based insurance, the cost-cutting features of ACOs, the monopoly regulation of public utilities, the accountability of special districts, the mixed public and private partnership of government-sponsored enterprises, and the structure of the corporate form. Under this approach, regional healthcare districts replace private insurance companies, and the districts offer universal coverage to all within the region in return for a direct healthcare tax. The districts pay providers in a capitated payment model, similar to paying a lump sum for each patient, instead of the fee-for-service model that pays per service rendered. The outcomes-based component consists of back-end, per-event incentive payments—which reward providers for each successful treatment—and additional payments that resemble dividends based on the overall health of the region. Providers get additional funding through government adjustment payments if they operate in underserved communities. The result is a synthesis of burgeoning knowledge on finance and governance in healthcare law and economics into the first model of its kind.

This Note proceeds in six parts. Part I traces the development of healthcare regulations from their inception in the early twentieth century and outlines the corresponding rise in healthcare costs. Part II discusses the various economic concepts and challenges that underlie the increase in costs. Part III explains how the Healthcure System achieves access by establishing universal coverage risk pools based on region and price elasticity by reducing individual healthcare expenditures to one income-based payment. Part IV describes the model’s downward price pressures through a new governance model that combines integrated finance and delivery with public electoral accountability. Part V explores how the Healthcure System enhances quality by aligning the incentives of patients and providers through a capitation and incentive payment model. Part VI considers the legal path and obstacles facing the implementation of the Healthcure System before concluding the Note.

I. BACKGROUND

Healthcare was once an unregulated and uninsured marketplace consisting of independent doctors making house calls in exchange for modest out-of-pocket fees.³³ The low cost of this relatively unsophisticated care sustained a functioning market until the early 1900s, when a combination of increasingly complex medical care, growing demand, and rising quality standards led to a surge in the average family’s medical expenses.³⁴

An early insurance market grew organically out of a need to spread out costs and risks by making regular payments to guarantee access to care without financial barrier when it was needed.³⁵ As insurers increasingly became intermediary payers between patients and healthcare providers, a “cost plus” reimbursement methodology emerged that paid doctors whatever “reasonable and customary charges” they set and covered hospital costs plus an additional negotiated rate payment.³⁶ The cost plus model supercharged the already-upward trend in healthcare costs by creating incentives to treat more and charge more.³⁷

The 1940s saw the addition of employers as an integral layer to the increasingly complicated healthcare funding landscape. Amid World War II’s labor shortage and inflation, Congress enacted the Stabilization Act in 1942 to place limits on wage increases, but it carved out an exception that allowed employers to offer fringe benefits like health insurance up to the value of five percent of wages.³⁸ In 1951, the Internal Revenue Service (“IRS”) adopted a rule making employer-paid insurance premiums a tax-deductible business expense.³⁹ Health insurance thus became a form of tax-free compensation that employers could offer their employees.⁴⁰ Once private health insurers instituted provisions requiring that a substantial majority of employees participate in the employer-sponsored plan, insurers had a risk pool of working-age adults that avoided disproportionate inclusion of higher-risk individuals who tend to consume more in medical expenses, such as those in the general population who are too old or ill to work.⁴¹ Using the employee risk pool as a guide, insurers then set a standardized premium rate for all participants, regardless of participants’ individual health histories, under what is known as “community rating.”⁴² Essentially, an employee with a clean bill of health paid the same premium as an employee who previously battled cancer. At the same time, labor unions negotiated rapidly increased employer-paid percentages of insurance premiums, achieving 100% coverage at some of the largest automobile manufacturers by 1961.⁴³ In 1974, the Employee Retirement Income Security Act of 1974 (“ERISA”) further solidified employer-provided healthcare by creating a nationally uniform regulatory scheme for multistate employers, imposing fiduciary duties on employer health plans, and providing beneficiaries with a positive right to sue for recovery of denied benefits.⁴⁴

The rise in healthcare costs significantly impacted two populations that tend to lack employer-based health insurance: the elderly and the poor. Congress responded with the Social Security Amendments of 1965 that established the Medicare and Medicaid programs, a pair of national insurance programs that positioned the federal government as the single largest third-party payer in healthcare.⁴⁵ Despite its outsize role in the industry, the federal government initially made no changes to the healthcare business model and adopted the same cost plus reimbursement model that had driven up costs in the private insurance industry.⁴⁶

Lawmakers have struggled to reign in the cost of the healthcare fee-for-service model since the 1970s, when they sought to incentivize adoption of the health maintenance organization (“HMO”) model that had been pioneered by the Ross-Loos Medical Group and the Kaiser Foundation Health Plan.⁴⁷ The HMO is the archetypal organizational form of coordinated care in which a network of providers deliver a comprehensive benefit package for a fixed premium.⁴⁸ The primary advantage of HMOs is the integration of finance and delivery of healthcare within the defined network of providers who cut down on costs by managing utilization and provider payments.⁴⁹ A key component of the HMO model is “managed care,” which comprises of “gatekeeping, capitation reimbursement, utilization review, clinical practice guidelines, and selective physician contracting.”⁵⁰ The Health Maintenance Organization Act of 1973 encouraged adoption of HMOs by funding the expansion of HMOs and requiring large employers to offer an HMO benefit option in addition to fee-for-service plans.⁵¹ Despite HMOs’ success at cutting costs, concerns over provider incentives to reduce access to services or diminish patient control ultimately led to the downfall of most HMOs and a return to healthcare cost inflation within two decades.⁵²

The nature of healthcare evolved in the mid-twentieth century as policyholders sought coverage of medical expenses beyond hospital visits and catastrophic illness. Insurers implemented new forms of sharing the increased costs of new “major medical” coverage with individuals through deductibles, an annual dollar amount that a policyholder must pay before insurance begins covering costs, and co-payments, a share of healthcare service costs paid by policyholders each time they use a service.⁵³ On paper, insurers had strong reasons for implementing cost sharing in healthcare. As individuals took on more of their healthcare costs, insurers could not only offset some of their expenditures, but also adjust their offerings with lower premiums or higher annual coverage limits.⁵⁴ Cost-sharing measures also looked to solve the “moral hazard” problem that arises when individuals seek medical care that they may not need because they do not bear any of the cost.⁵⁵ Once individuals had to pay each time they visited a doctor or got an X-ray, they would “think twice” and presumably seek fewer services.⁵⁶

Healthcare coverage reached an inflection point in the 1980s, at which point the rapid growth in access to health insurance and care began to move in the opposite direction. Employer-sponsored health coverage reached its peak in 1980, when it covered 79.4% of the U.S. population under sixty-five.⁵⁷ By 2018, employer-sponsored coverage of the same population had fallen to 58.1%.⁵⁸ As fewer employers offered insurance, access to private plans did not grow to cover the difference—the rate of uninsured grew from 12% in 1980 to 18.2% in 2010, with the majority of the growth occurring in Medicare and Medicaid enrollment.⁵⁹

Regulators saw an opportunity to cut healthcare costs with market-based interventions that realign competition across the industry. In the early 1990s, President Clinton introduced the Health Security Act that built on economist Alain Enthoven’s concept of managed competition.⁶⁰ Under managed competition, sponsor agencies or “alliances” (such as employers, Medicare, or Medicaid) act as referees between the competing health plans available to the sponsors’ members, determining benefits, prices, enrollment, and more.⁶¹ Sponsors focus competition on the price of annual premiums rather than individual services, with the goal of creating price-elastic demand.⁶² Price-elastic demand occurs when individuals reduce demand as prices go up, and this incentivizes sellers to keep prices as low as possible.⁶³ Naturally, regulators try to avoid price inelasticity, which occurs when a seller can increase prices without reducing demand.⁶⁴ Managed competition also pursues cost cutting by dividing providers into competing economic units and imposing market forces to compel them to become efficient delivery systems.⁶⁵ While President Clinton’s proposal would have introduced the American healthcare system to a new phase of managed care, the bill failed, and it would be another sixteen years before Congress would pass large-scale healthcare reform.⁶⁶

When President Obama signed the ACA in 2010, it represented the most significant healthcare reform package since President Johnson’s Great Society gave Americans Medicare and Medicaid.⁶⁷ Rather than deconstructing the healthcare system to cut costs as President Clinton had attempted to do a generation prior, the ACA primarily focused on increasing access to health insurance and improving the quality of health benefits. The ACA created a new marketplace for health insurance plans that aimed to streamline the insurance purchase process and required that plans offer ten essential health benefits to all who sign up.⁶⁸

The ACA took a carrot and stick approach to expanding health coverage in what is known as the “three-legged stool.”⁶⁹ The first leg required insurance companies to adopt community rating with guaranteed issue of ten essential health benefits for all who seek coverage.⁷⁰ Those who did not buy into the health insurance market, either through the marketplace or another avenue such as an employer, were subject to the second leg: a tax penalty known as the “individual mandate.”⁷¹ Many who signed up, however, enjoyed tax credits—the third leg—to help cover their premiums and cost sharing, such as co-pays, deductibles, and coinsurance.⁷² As a result, twenty million individuals gained health insurance in its first five years.⁷³

More than any other aspect of the ACA, the individual mandate faced intense legal and political scrutiny. An array of court battles culminated in NFIB v. Sebelius, a 2012 Supreme Court decision that upheld the constitutionality of the individual mandate.⁷⁴ In his opinion, Chief Justice Roberts wrote that although the individual mandate fails as an exercise of Congress’s Commerce Clause power, “it is reasonable to construe what Congress has done as increasing taxes on those who have a certain amount of income, but choose to go without health insurance. Such legislation is within Congress’s power to tax.”⁷⁵ Constitutionality was not enough to save the individual mandate, however, and Congress repealed the tax penalty in 2017.⁷⁶ The ACA itself came just one Senate vote short of repeal,⁷⁷ and there remain efforts to replace it.⁷⁸

II. ECONOMIC CHALLENGES AND SOLUTIONS IN HEALTHCARE: FROM ACOS TO HEALTHCARE DISTRICTS

A. Diagnosing Healthcare’s Market Failures

Healthcare represents not only a lifeline for individuals, but also for the American economy.⁷⁹ In 1960, healthcare expenditures accounted for 5% of the nation’s gross domestic product (“GDP”).⁸⁰ By 2022, that figure had risen to 17.3%.⁸¹ Some of the increase can be attributed to positive developments in care and coverage. But economists point out that healthcare spending also results from fundamental problems in the healthcare market. In an efficient healthcare market, rational and fully informed individuals could purchase healthcare services they need from fair, perfectly competing sellers.⁸² Healthcare resources could be allocated efficiently in a world in which people can shop around for healthcare, with the full scope of information on the prices and quality of each provider’s services, and the ability to then pay for those services directly. As Americans learned in the early-twentieth century, when complex healthcare emerged and insurance developed to pay for it, such a world is a fiction. The healthcare system that resulted was one fraught with market failures that have driven costs upward, and healthcare reform to this point has failed to stem the tide.

For many, the loss of the individual mandate spelled the end of the ACA.⁸³ In theory, the less-risky population of younger, healthier individuals could pull themselves out of risk pools and skip health insurance in a phenomenon that economists call “adverse selection.” With risk pools more heavily concentrated with older and sicker individuals, as the theory goes, prices would increase.⁸⁴ Increased prices would lead more people to withdraw from the health insurance market, and the so-called “adverse selection death spiral” would lead to a collapse of the market altogether. It turns out that one of healthcare’s greatest problems is what has propped up the system post-mandate: price inelasticity.

Healthcare suffers from price inelasticity because when healthcare costs go up, individuals do not drop insurance coverage, they just drop going to the doctor. By 2010, the uninsured non-elderly population reached its peak at 17.8% before the passage of the ACA.⁸⁵ The law’s drafters understandably made it a priority to bring the number of uninsured down, and on that front the law has been largely successful to date.⁸⁶ In 2018, the uninsured rate dropped to 11%,⁸⁷ and by 2022 the non-elderly uninsured rate reached 9.6%, the lowest level on record.⁸⁸

Yet, increased coverage did not spell increased access to care. One survey found that in 2001, 19% of adults reported putting off needed care due to costs, but even with the passage of the ACA, by 2022, that figure had risen to 38%.⁸⁹ Another survey reported that 40% of Americans skipped a recommended medical test or treatment due to cost, and 40% of Americans have cited cost as the reason for going without routine physicals or other preventive care.⁹⁰

Not all insurance plans are the same, and the differing approaches to cost sharing exposes the disparity in access to care across the healthcare system and the problems that arise from a lack of reliable pricing information. Cost sharing comes in various forms, including percentages of medical service costs or fixed rates set by insurance companies according to a particular service, such as $20 for a physician visit or $150 for a hospital stay. Alternatively, one might pay $150 or 20% for a hospital stay, depending on the type of plan one has. The disparity in insurance can be seen in the enrollment trends in the ACA insurance marketplace, in which the middle “silver” tier has seen declines in enrollment, the “gold” tier has seen modest gains, and the lowest “bronze” tier has seen significant increases.⁹¹ Whereas set rates, rather than percentages, for healthcare services shields individuals from unexpected costs, it simultaneously hides the complex and mysterious world of medical billing.

The key takeaway is this: price elasticity of demand occurs in the provision of healthcare services, rather than in the provision of insurance coverage.⁹² More Americans than ever have health insurance, but a great deal of those with coverage forgo the added out-of-pocket costs that come with seeking healthcare services.⁹³ And while that reduced demand for services might compel providers to reduce prices in an efficient market, they have made up the difference by continuing to increase prices and extracting more money per service for those who do seek treatment.⁹⁴ Individuals with reduced cost sharing, such as set rates for services, face a reduced barrier to services and do not encounter the prices that keep others away.

Because individuals are not the only payers in the health insurance market, price elasticity could ostensibly come from insurers or employers. In theory, employers ought to balk at rising healthcare costs, but economists have suggested that they pass on the additional costs to employees.⁹⁵ In an era of high inflation, employers can pass on healthcare costs easily through smaller nominal wage increases.⁹⁶ Between 2009 and 2019, worker contributions to employer-sponsored premiums rose 59% and employer contributions rose 54%, while employers’ share of the total premium held steady at 73%.⁹⁷ Employees absorb the increased prices charged by providers in a way that is largely hidden from them—employees cannot readily know how much they would earn in the absence of healthcare cost increases.⁹⁸ Evidence suggests the passing of these costs to employees, combined with rising wage inequality, “significantly reduced the percentage of compensation.”⁹⁹ If providers can increase prices that individuals will ultimately bear, without some other market basis like a proportionate loss in demand or increase in value, the provider gets away with earning what economists call “rents,” or excess prices beyond the minimum price a seller would otherwise be paid in the market.¹⁰⁰ Not only is this harmful to individuals, in that it represents an inefficient allocation of resources, rent-seeking behavior evinces a level of monopolistic power by providers in the market.¹⁰¹ Were prices elastic, the number of insured individuals would decrease as prices rose. Over that same period, however, more people have gained insurance.¹⁰² While the growth in healthcare coverage is a positive development, it represents a worrying trend when paired with increased prices. The brakes that traditionally keep prices low—the threat of losing paying customers once prices exceed what they are willing to pay—either do not exist or have not been reached.

With market failures taking out downward cost pressures from individuals and employers, insurers stand as the apparent last line of defense against rising healthcare costs. The traditional insurance business model incentivizes cost efficiency—policyholders pay insurers set premiums, and insurers have the incentive to pay as little of those premiums out as reimbursement for services in order to retain the greatest possible profit margin. The ACA turned this model on its head when it mandated a “medical loss ratio”—a requirement that insurance companies spend 80–85% of premium dollars on medical care-related expenses, thereby tying the amount they get to keep (including profits) to a percentage of care dollars spent.¹⁰³

The healthcare marketplace insulates individuals from many of the direct costs of healthcare, and the lack of robust price competition for insurance means that insurers may continue to raise premiums to accommodate the high healthcare prices that net them greater profits.¹⁰⁴ Insurers can take advantage of market failures to pursue the perverse incentives of a medical loss ratio policy that was meant to decrease costs but instead incentivizes them to spend as much as they can.¹⁰⁵ Furthermore, insurers have an additional incentive to keep costs as high as their premiums can bear because high prices set a high barrier to entry for other potential competing insurers.¹⁰⁶ These incentives expose the pitfalls of cost-based regulations instead of incentive-based ones, as proposed here.¹⁰⁷

Insurers cannot overpay for services if they are not charged such high prices in the first place. Whereas cost sharing effectively curbed the threat of moral hazard when individuals seek out more medical services than they need because they do not bear the cost, another moral hazard problem arose in the form of provider billing and overtreatment.

The cost-sharing measures implemented by insurers essentially traded one moral hazard problem for another. Insurers overcorrected the moral hazard problem by disincentivizing individuals from seeking treatment, instead incentivizing doctors to overtreat those that do come in.¹⁰⁸ In its worst incarnation, physician-induced demand can rack up healthcare expenses when patients do not know any better, and insurers put few, if any, brakes on unnecessary charges.¹⁰⁹ A major factor that has enabled healthcare providers to charge ever-higher prices is the lack of transparency around the prices of services between patients, insurers, and even providers.¹¹⁰

One contributor to healthcare’s price inelasticity is the varying difficulty that consumers, providers, employers, and insurers have with understanding the price of healthcare services, or what economists call “information asymmetry.” Throughout healthcare, there are discrepancies in the amount of information available to transacting parties.¹¹¹ For instance, doctors typically have more information about the care they can provide than patients. Similarly, healthcare providers generally have more information about the costs of their services than insurers. The discrepancies in information create inefficiencies that drive up costs. When patients seek healthcare services, they likely do not know the cost of the services beforehand.¹¹² Depending on their insurance plan, patients can either anticipate paying their insurance plan’s set rate co-pays or face the surprise bill for a percentage of the services they incurred. Information asymmetry gives rise to a principal-agent problem, in which the physician-agent has incentives to use the information asymmetry for the physician’s financial benefit.¹¹³

A primary culprit of increased healthcare prices is the mysterious and unscientific hospital “chargemaster” system. Chargemasters are automated systems traditionally employed by large healthcare providers such as hospitals to set price markups and generate the sticker prices for their various services.¹¹⁴ Few, if any, in the hospital administration are involved with the setting of the prices, and even doctors generally are not informed of the prices of the services they perform.¹¹⁵ Here, a type of information asymmetry even occurs within providers. Yet, chargemasters set the baseline price from which insurance companies negotiate down to a level the insurer is willing to pay. Those with less bargaining power, such as uninsured individuals, may face the full chargemaster price without the assistance of medical billing consultants to negotiate on their behalf. The result often leads to newsworthy charges such as 675% price hikes.¹¹⁶

Providers capitalize on information asymmetry by using chargemasters to extract the highest possible prices, and they have few disincentives to do otherwise. As discussed earlier, many patients never see their medical bills aside from standardized, insurer-set co-pays. One might reasonably expect the other payers in the healthcare system—employers and insurers—to thus bear the sensitivity to providers’ exorbitant costs and apply downward price pressures. Unfortunately, the confluence of healthcare’s market failures renders those pressures toothless.

Efforts over the past decade to improve price transparency will not likely affect individuals’ healthcare decision-making process in a significant enough way because patients are not the ones making many of the decisions about their care.¹¹⁷ Revealing chargemaster prices, for instance, likely matters more to insurers and employers, as well as smaller hospitals that seek to charge comparable prices to larger competitors, than it would to consumers.¹¹⁸

The goals of universal access, low cost, and high quality can be achieved with a model that addresses the information asymmetry, principal-agent problem, misaligned incentives, adverse selection, poor competition, price inelasticity, and antitrust concerns that plague healthcare today.

B. Finding a Cure in Accountability

Enter the Healthcure System, an entirely new healthcare model proposed by this Note that adapts the best features of ACOs, incentive payment models, and employer-provided health plans while abandoning fee-for-service cost plus payments, private health insurers, complexity of multiple payment sources, incentives for providers to overtreat, and the power of providers to increase prices by consolidating and reducing competition.

Historically, healthcare reform has consisted of attempts to achieve two of the three pillars: cost, quality, and access. The ideal healthcare system keeps costs low for individuals and providers (whether someone can afford their healthcare costs and whether providers reign in the costs of their services), maintains high quality by making available a breadth of healthcare benefits with strong outcomes (whether a particular ailment is covered and whether a treatment has a high likelihood of success), and ensures that the greatest possible population has access to care and coverage of costs (whether most are able to easily visit a doctor or hospital and will they have a means of paying for their treatment). Medicare focused on cost and access by providing insurance to all who reach a certain age, but it has historically lacked important quality indicators such as coverage of prescriptions, dental, vision, long-term care, and nursing home care.¹¹⁹ The proposed Clinton health reforms of the 1990s focused on cost and quality by managing competition and guaranteeing benefits, but they did not address a growing number of uninsured.¹²⁰ The ACA focused on quality and access by guaranteeing ten essential health benefits and expanding access to Medicaid and a private insurance marketplace, but it left intact the payment methods that have driven prices upward.¹²¹ The Healthcure System rebuts the presumption that no model can achieve all three.

Under the Healthcure System, all residents within a particular region make payments (essentially a healthcare tax) to a healthcare district—a corporate entity that encompasses all of the healthcare providers in the region. Each region would be determined by the state legislature based on population and concentration of providers and should account for the equitable distribution of resources when doing so.¹²² The healthcare district coordinates payment and care for residents of the region, employing a front-end per capita payment and back-end incentive payments system that encourages providers to not rack up unnecessary costs but still have an incentive to provide cost-effective quality care. Healthcare districts largely take the place of private health insurers and employer-paid plans, instead centralizing each individual’s healthcare costs into one monthly income-based payment. Each healthcare district’s board of directors sets this progressive healthcare tax rate for the region, and if that percentage exceeds what residents are willing to pay, individuals may vote out the directors during staggered biennial elections or, if feasible, avoid living in the district.

This Note makes some acknowledgments from the start. First, healthcare delivery is inherently local—patients are realistically constrained to choosing providers near them, and healthcare costs are significantly influenced by local factors. Rather than share the struggles Medicare has had with accounting for regional differentiation in markets for medical services, products, and employment when calculating reimbursement, each healthcare district only concerns itself with negotiating local pricing with local providers for local beneficiaries. Districts are therefore organized around a large enough population to distribute risk in a risk pool while also encompassing the entirety of a local market (that is, all of the providers that would compete with one another for individuals in the region). In densely populated states, there may be healthcare districts proportionate to the number of counties, but in more rural areas, districts could theoretically be the size of a state. Second, provider consolidation is inevitable and comes with benefits despite its drawbacks. The upside of a fully functional ACO is the imposition of cost sensitivity on the providers who are responsible for the decision-making behind those costs. Providers would need to evaluate, for instance, whether to use an expensive treatment based on its likelihood of success, rather than indiscriminately prescribing costly remedies, because they can no longer rely on a third-party payer to just foot the bill. Such integration of finance and delivery can act as a powerful downward cost pressure that can alleviate the current upward incentives to overtreat and overprice. Although consolidation also breeds upward price pressures in the form of reducing competition, Healthcure makes use of other downward cost pressures such as regional competition, price elasticity, public accountability, and a market for supplemental benefits to counterbalance antitrust pricing concerns. Third, incentive payments must account for both event-specific outcomes (rewarding providers when a particular treatment works, for example) and community outcomes (rewarding providers for overall health gains in a region) in order to fully align the incentives of payments and providers.

III. ACCESS: ADOPTING A REGIONAL FOCUS TO DELIVER EMPLOYER-STYLE UNIVERSAL COVERAGE

In a sense, access is the ultimate aim of healthcare reform: it encompasses both cost and quality to a certain degree. High cost is often a barrier to access to coverage, and poor-quality coverage often keeps people from access to care. Any discussion of access in the healthcare system must account for both coverage (as in an individual’s participation in some support system to cover healthcare costs) and care (which accounts for the kinds of healthcare services available to an individual). Both are of major concern in the current United States healthcare system. In early 2023, 7.7% of Americans did not have access to health insurance.¹²³ Furthermore, studies have shown that low-income and marginalized communities see a disproportionate decline in access to care from higher odds of losing healthcare facilities.¹²⁴ Healthcure addresses access by co-opting one of the original drivers of health insurance adoption: employer-sponsored health plans.

As evidenced by the public concerns over whether individuals would be able to keep their insurance plans at the time the ACA was passed,¹²⁵ such employer plans have achieved a level of ubiquity in American society that any reform that upends the status quo would be well-advised to consider. The Healthcure System shares two key elements with employer-sponsored health plans: community rating and income contributions.

A. Expanding the Risk Pool from Coworkers to Neighbors: A Regional Approach

Today’s healthcare landscape is filled with geographically mismatched systems and entities. When individuals seek care, they are generally constrained to providers within their immediate vicinity. These individuals may participate in health plans through employers that also provide coverage to employees located in other states. Insurance marketplaces through the ACA, on the other hand, are constrained to markets within each state. Meanwhile, the single largest payer in the country—the Medicare system of the federal government—operates nationwide but uses a litany of processes to account for regional disparities in costs. To put it simply, healthcare is inescapably local, and the first step to crafting an efficient healthcare system is recognizing the value of a regional approach.

At its core, the Healthcure System consists of smaller, regional systems that naturally incorporate the idiosyncrasies of their localities in healthcare finance and delivery. Although healthcare districts do not currently exist as envisioned for the Healthcure System here, the concept of regional healthcare districts is not entirely new. Regional health districts under the Healthcure System trace their roots to a different kind of healthcare district employed by state governments to coordinate healthcare in rural areas.¹²⁶ Funded by either general taxes or special taxes, traditional healthcare districts are local governments that operate healthcare facilities, establish managed care, contract with providers, or take on any other health-related service for a community.¹²⁷

Any third-party payer that collects funds from a population to distribute medical services must consider risk. The delicate balance between risk and ratemaking can have significant implications for access if those with higher health risks face significantly higher rates (or exclusion from the market altogether). Such was the consequence of experience rating, a method used by health insurers to determine eligibility and rates based on an assessment of one’s health risks. Insurers engaged in favorable selection, attempting to insure healthier individuals and avoid sicker people. Prior to the ACA, those with certain preexisting conditions would face unaffordable premiums or even exclusion from certain health plans. To avoid locking out those with the most medical need from the insurance system, and thereby exposing them to insurmountable out-of-pocket costs, the ACA banned experience rating in all circumstances except age and tobacco use. To counteract the risk imbalance that would result from removing the experience rating mechanism plans could use to reduce risk, especially on small plans, the ACA promised economies of scale that could accommodate high-risk individuals but spread exposure across a larger risk pool. Insurers had the individual mandate to thank in part for the uptick in larger, healthier risk pools, as it compelled those who might otherwise go without health insurance to do so or face tax penalties.¹²⁸

While the ACA expanded coverage for those who did not otherwise have access, employer-sponsored health coverage still dominates. Employers cover approximately 50% of the U.S. population.¹²⁹ Of those who have access to employer-sponsored coverage, 77% participate in the employer’s group.¹³⁰ In the mid-twentieth century, employer-sponsored plans drove rapid growth in health coverage, but today they are fraught with drawbacks. Employer plans are responsible for additional complications in the payment structure, buffering individuals from the cost of their care by passing on the costs through reduced wages, shifting compensation from paychecks to health plan contributions, limiting coverage to those employed, and constraining risk pools. Studies have shown that the upward cost pressure problems of employer plans are exacerbated in markets with fewer insurance carriers because carriers take advantage of the lack of competition by negotiating higher premiums, especially with employers experiencing profit shocks.¹³¹ Moreover, the costs of researching different plans and transitioning to new ones often keep those in employer plans from switching, thereby reducing competition that could impose downward price pressure.¹³²

Even before it was mandated by the ACA, individuals were exposed to community rating by their employer health plans. Rather than engaging in experience rating, which evaluates risk based on the individual health histories of each ratepayer and charges them accordingly, traditional employer plans distribute risk across a group and charge each member the same rate.¹³³ Because community rating does not take into account health histories, the primary mechanism by which to manage risk is to achieve scale.¹³⁴ Employer plans are limited by the group of enrollees at a particular firm.

Given the local limitations of healthcare delivery, the Healthcure System’s answer to risk distribution is to expand it from the silos of individual employers to the very limits of a particular region. The general wisdom is that when allocating risk, the larger the risk pool, the better. Although community rating across organizations provides some semblance of risk distribution, it pales in comparison to the distribution across an entire region at all employers except for the largest of firms. The Healthcure System’s expansion of risk pools enables healthcare districts to enjoy the cost savings of the risk distribution.

B. Simplifying Beneficiary Healthcare Costs to Induce Price Elasticity and Protect Affordability

The “Affordable” in Affordable Care Act may be the most significant undelivered promise of the law. What the ACA did not address is the complex web of payment sources that underlies the entire system. Whereas insurance premiums are the primary focus of the cost debate, and deservedly so, they do not account for the entirety of an individual’s healthcare expenses.

It is this mixture of cost types that shields individuals from grasping the full scope of their healthcare expenditures and buffers their sensitivity to the exorbitant charges providers make for their services.¹³⁵ The Healthcure System narrows each individual’s health expenses to a percentage of one’s income—the starkest representation of one’s ability to pay. The percentage is set by each healthcare district and may be adjusted annually by its board of directors. Income contributions have a low-income threshold—individuals with an income below a certain dollar amount pay nothing, and as one’s income crosses the threshold, the percentage gradually rises up to the base percentage that is broadly applied across the district. All residents in a region have access to the district’s health benefits, regardless of income level. To avoid adverse selection of high earners fleeing from risk pools, a high-income cap places a limit on the dollar amount a family may contribute to the healthcare district’s base benefits. While healthcare districts adopt the ten essential health benefits as defined by the ACA and offer them to all residents, those who wish to supplement their benefits may purchase a supplemental plan on an open market similar to Medigap supplemental coverage for Medicare.¹³⁶

The extent of the current healthcare system’s price inelasticity was tested when Congress repealed the individual mandate. With the flick of President Trump’s pen, those who would otherwise leave the market due to high prices could all of a sudden do just that. However, as healthcare costs continued to rise, enrollment did not precipitously decline.¹³⁷ Even with the loss of the individual mandate, price inelasticity may still rely on the buffer between individuals and their healthcare costs to insulate them from the price sensitivity that might otherwise drive them from the market.

The Healthcure System’s approach to individual payment achieves price sensitivity by wrapping up all healthcare costs into one tax payment that is broadly applied to all residents in the healthcare district. The current mix of individual premiums, employer contributions, co-pays, deductibles, coinsurance, and government tax credits all obscure the cost of healthcare to the individual. However, when all that individuals must consider is a particular tax, their understanding of health expenses gains a clarity with which they can impose downward price pressures.

Downward price pressures on the demand side of the Healthcure System consist of individual choice between healthcare districts and public election of district boards of directors. Instead of imposing cost sharing through co-pays for each medical service rendered, Healthcure incentivizes individuals to consider healthcare tax costs when choosing where to live. If a particular healthcare district sets too high of an income contribution rate for its residents, individuals may choose not to move to that particular district. But moving may not always be feasible or desirable, so those who reside in the district can express their objection to high income percentages by voting out the board of directors and electing a board committed to cutting costs.

Traditional cost sharing does not sufficiently cause individuals to better consider which healthcare services they should seek because they lack the information needed to make that determination and will always lack that information in the absence of the training of a medical professional.¹³⁸ Cost sharing on the service level thus results not in cost savings, but in worse health outcomes when people forgo treatment due to cost. Instead, price sensitivity should be focused on what individuals can make informed decisions about and fully understand. For most, that is the total amount they are able to pay for their healthcare tax.

C. Possible Equity Implications of a Regional Model

The deontological debate over whether healthcare constitutes a commodity, a right, or something in between extends far beyond the scope of this Note. But the law and economics approach taken here does not ignore the ethical implications of equity in its design of a more efficient healthcare system. Policymaking necessarily has ethical and economic consequences, and this Note presents the Healthcure System as a means of compelling providers to become fairer market actors, guaranteeing egalitarian access to healthcare, enhancing competition, and reducing costs.

The general thrust of the Healthcure System model is its introduction of new accountability measures into a healthcare landscape where there are currently few. This lack of accountability, whether it be in hospital pricing or insurers’ willingness to pay, affects every individual regardless of whether they choose to participate in the insurance market. This is true of any market—the actions of firms do not exist in a vacuum without influencing supply and demand for all market participants.¹³⁹ The imperative for broad public accountability in the market for widgets, however, does not measure up to the imperative in a healthcare industry that every individual is likely to transact with at some point. For healthcare providers to be truly accountable, they must be accountable to everyone, not just their customers. By bringing all individuals under a universal coverage model, the Healthcure System’s accountability measures can internalize the externalities of the healthcare market. In other words, no longer will insurers and providers suscept the uninsured to price inflation.

In its effort to provide universal coverage, the Healthcure System sets forth regional healthcare districts that collect taxes from residents and distribute those funds to providers for the provision of services. Notwithstanding its unique policy prescriptions, the Healthcure System would be wise to adopt some of the innovations of the ACA, such as the guaranteed issue of essential health benefits. All health insurance plans were required to offer ten services as part of their benefit packages under the ACA: (1) ambulance; (2) emergency; (3) hospitalization; (4) maternity and newborn care; (5) mental health and substance use disorder; (6) prescription drugs; (7) rehabilitative and habilitative services and devices; (8) lab work; (9) preventive, wellness, and chronic disease management; and (10) pediatric services including oral and vision care.¹⁴⁰ Additionally, the ACA ensured that insurers could not turn away or even raise premiums for those with preexisting health conditions. Although the ACA may not have done enough to stem rising costs, it did take significant steps toward improving access to health coverage and the quality of the benefits provided.

A regional, tax-funded healthcare system raises concerns over provider concentration and equity between regions with different income levels. These concerns predate the proposed reforms offered here, but this Note does not seek to entrench an already inequitable system. On the contrary, the Healthcure System can act as a vehicle for identifying areas with inequitable access to healthcare and delivering targeted financial support in the form of supplemental government adjustment payments.

The Health Resources & Service Administration (“HRSA”) currently uses two designations to identify areas of need in the healthcare system: Medically Underserved Areas/Populations (“MUA/P”) and Health Professional Shortage Areas (“HPSAs”).¹⁴¹ Additionally, Medicaid identifies hospitals that serve large numbers of Medicaid and uninsured individuals and directs supplemental payments through the Disproportionate Share Hospital (“DSH”) program.¹⁴² These programs may be rolled into healthcare districts, which can more easily collect data on the community’s needs due to the coordinated care of all the providers in the region. Federal and state governments can then issue adjustment payments to those districts with the most need, and the payments can come in the form of block grants, incentives to recruit physicians, infrastructure and capital improvement funding, equipment, and even technical assistance and consultation. Further, government payments could subsidize the costs of patients visiting other healthcare districts to make use of equipment, services, or specialists that their underserved district may lack. The district could evaluate the cost of purchasing access to equipment, services, or specialists in other districts and the demand for them, and if cost-effective, petition the government to subsidize the acquisition of them for the underserved district. In this way, the Healthcure System can play an active role in improving health equity in a given region.

Another concern arises over the district’s control over pricing and the potential for monopsony. Whereas monopoly power enables a seller to set prices above what a perfect market would dictate, monopsony power on the buyer side enables one to counterbalance monopoly rent seeking.¹⁴³ But a monopsonist need not face a monopoly on the other side of the transaction in order to wield its power, and in the healthcare context, a single-payer such as a government can lead providers to withdraw supply in response to lack of price power.¹⁴⁴ It is not difficult to imagine how low-income districts may be disproportionately susceptible to constrained supply if they pay providers less than neighboring districts. Government adjustment payments are just one way districts can avoid these circumstances.

Here, the board of a healthcare district also has work to do to protect individuals’ access to care. One way to fight the possibility of constrained supply is through the system’s per-event incentive payment structure that rewards providers for seeing more patients, rather than rewarding them based on the number of services administered through fee-for-service.

An additional safeguard would be a requirement that providers offer essential health benefits if they are to participate in a supplemental coverage market not subject to price controls by the district. Providers could offer supplemental coverage in a private market for services not otherwise available through the district. The districts must ensure there is no overlap of services between the essential benefits they offer and the supplemental benefits offered by providers because an overlap would open the door for a conflict of interest. If providers’ supplemental services competed with those offered by the district, then providers will have incentives to reduce district essential benefits made available to everyone, offer lesser-quality district benefits, or otherwise push individuals toward its supplemental offerings (which would ostensibly be more lucrative for the providers).

IV. COST: INNOVATING A NEW HEALTHCARE DISTRICT GOVERNANCE MODEL WITH PUBLIC AND FINANCIAL ACCOUNTABILITY FROM STAKEHOLDERS

Although not a centerpiece of the ACA, a modest provision in section 3022 now stands as one of the most promising elements of the ACA left standing, one that opened the door to a new organizational form dedicated to accountable care: the ACO.¹⁴⁵ When providers come together to form an ACO, they commit to coordinating care and sharing in the responsibility for financial and quality outcomes for a certain population of patients.¹⁴⁶ Not only can ACOs deliver improved health outcomes by sharing information between doctors, hospitals, and other providers, thereby reducing unnecessary or redundant treatment, but they can also reduce costs.

The healthcare market traditionally imposes on insurers—not providers—the risk of losing money when healthcare costs exceed the amounts collected from individual premiums. As discussed earlier, market failures have led to a system in which healthcare providers face no downside to overtreating patients and often have incentives to charge as high of prices as possible.¹⁴⁷ Thus, the system imposes few, if any, brakes on runaway healthcare costs, and those costs are ultimately passed to individuals in the form of increased premiums, greater cost sharing, and reduced wage gains.¹⁴⁸ ACOs flip this relationship on its head by shifting risk to the providers who are largely responsible for making decisions about and incurring the costs of care in the first place. Rather than rewarding providers based on the number of services provided, as the traditional fee-for-service model incentivized, ACOs offer providers financial incentives for hitting cost-savings targets and meeting quality benchmarks. In theory, the goal is to align incentives in the healthcare market so that all benefit from low-cost, high-quality care.

In concert with the Medicare Shared Savings Program (“MSSP”), the federal government incentivizes cost cutting by offering providers who organize under an ACO model a share of any cost savings they generate from efficient service delivery.¹⁴⁹ The Centers for Medicare & Medicaid Services (“CMS”) offer several levels of shared savings, and the ACO shares a proportionate amount of risk of cost overages according to the level it joins.¹⁵⁰

The upside of the ACO is so great that it could—with the right adjustments—avoid the upward cost pressures that come with traditional private insurers.¹⁵¹ MSSP reported $1.8 billion in net savings in 2022 after making incentive payments to participating ACOs.¹⁵² On its face, that number represents a laudable achievement and proof of concept for ACOs, but it does not fully capture the state of the program. The number of participating ACOs shrank from 561 in 2018 to 456 in 2023,¹⁵³ and economists point out that the savings figure is largely the product of selective participation.¹⁵⁴ Due to the voluntary nature of MSSP, higher-spending ACOs have disproportionately exited the program and lower-spending ACOs have entered. The result is a skewed program that provides incentive payments to smaller, potentially less-efficient providers and falls short of imposing meaningful cost savings on the large providers most responsible for the healthcare system’s increasing costs.¹⁵⁵

A key selling point of ACOs is the integration of finance and delivery among a network of providers, but adverse selection stands in the way of achieving that integration at scale. Healthcare districts may solve the selective participation problem by automatically enrolling all providers and individuals to participate in the district’s network. Such a mandate creates a new set of challenges when considering the varying interests of individual, hospital, and physician stakeholders.

On one side, individuals are responsible for paying healthcare taxes to pay for their access to the district’s provider network and essential health benefits. In their pursuit of quality care at the lowest possible cost, individuals may voice their preferences through choice of provider, district in which to live, and elected directors of the district. In this regard, districts take on qualities resembling special districts.

On the other side, a mix of independently run private for-profit and not-for-profit healthcare providers compete with one another for patients, and they may make different business decisions in light of that competition. To require these providers to join forces might exacerbate provider consolidation and raise antitrust concerns that they might wield too much pricing power in their regional market. In light of this, healthcare districts could take cues from public utilities in the administration of a regulated monopoly on healthcare.

A reasonable objection to the mandated participation of healthcare providers is the restriction on the freedom of private businesses to transact in an open market. With traditional ACOs, providers choose with whom they would form a network; in contrast, healthcare districts bring together all neighboring providers to coordinate. Concerns over the agency of providers are warranted, and it is precisely these concerns that encourage the consideration of the corporate form as a source of inspiration to protect these interests.

The formation of special districts and regulation of corporations are largely functions of state law.¹⁵⁶ When legislating the creation of healthcare districts, state lawmakers will need to consider an organizational form tailored to the idiosyncrasies of healthcare. Whereas the ACO is the archetypal structure from which healthcare districts are designed, the districts must take on characteristics of both public and private entities to overcome the market failures and regulatory shortcomings that otherwise keep ACOs from achieving the integration of finance and delivery at scale today.

A. Public Governance as a Counterweight to Consolidated Provider Market Power

At the center of the ACO adverse selection problem is the gap between, on the one hand, the good governance principles that CMS inspires through incentives for cost-savings, and on the other, the financial motives for avoiding participation in ACOs due to the risk of cost overages. As a means of broad institutional change, MSSP’s ACO program is a portrait in weak governance.¹⁵⁷ In the tripartite scheme of healthcare governance between individuals, providers, and government, the government’s role in enacting ACOs is more akin to making a series of suggestions than outright rulemaking.¹⁵⁸

Despite the federal government’s soft touch when it comes to pushing providers to participate in a two-way risk MSSP model, industry watchers have raised antitrust concerns over ACOs’ potential to exacerbate provider consolidation.¹⁵⁹ Therein lies an inherent conflict within the very concept of the ACO: while ACOs strive to achieve a reduction in costs, they simultaneously enable increased prices by encouraging providers to join forces and reduce competition.¹⁶⁰ With fewer competitors on price, ACOs can theoretically wield more market power and raise prices with impunity.¹⁶¹

A conflict thus arose between CMS’s encouragement of ACO formation and the Department of Justice’s (“DOJ”) antitrust enforcement role. Where the two federal agencies stood diametrically opposed, the DOJ capitulated to CMS and issued a policy statement effectively taking a hands-off approach to antitrust enforcement when providers are organized in an ACO.¹⁶² Notwithstanding the multitude of market failures that have largely eviscerated price competition in healthcare, the fact remains that a traditional ACO could adopt a one-sided risk model that imposes no risk for cost overages while enjoying the lax antitrust rule enforcement that comes with participating in MSSP.

Provider consolidation was a growing trend in healthcare even before the ACA took effect, but it has ramped up in the years since.¹⁶³ In the two-year span between 2016 and 2018, physician affiliation with vertically integrated health systems jumped 11%,¹⁶⁴ and the number has doubled over the past decade.¹⁶⁵ The movement toward vertical integration, in which entities along a supply chain such as doctors and hospitals align under one entity, can be a positive development for cost efficiencies and care coordination. After all, coordination is a central component of the ACO model. But consolidation inevitably results in price increases in the absence of downward cost pressures such as ACO cost sharing and strong antitrust regulation, especially when entities such as nearby hospitals horizontally integrate and cease competition.¹⁶⁶ With ACOs increasingly opting for one-way risk models and the DOJ relaxing antitrust enforcement, the current system does not pose much of a barrier to increased prices.

The Healthcure System’s mandate to form provider networks by region would not mark the first time a government sanctioned a geographic monopoly over a particular market. The regulation of public utilities arose from the recognition that some businesses operate in the public interest and regulations would ensure that “all must be served, adequate facilities must be provided, reasonable rates must be charged, and no discriminations must be made” when the free market alone would not.¹⁶⁷ Historically, sectors of public interest, such as transportation, communications, electricity, and water, invited regulation as utilities when they were dominated by large businesses with enough market power to exploit customers.¹⁶⁸ The similarity of circumstances in healthcare today has raised the question of whether medicine should be regulated as a public utility.

The difficulty with applying public utility regulation to medicine lies with its traditional business model: while its rates and service may be mandated by the government, a utility remains a private business that charges a rate based on use.¹⁶⁹ For instance, an electricity company may be required to supply energy to all homes in a given region. Not all of those homes may choose to purchase that energy, however, and if they do, they pay an amount proportionate to the amount they consume.¹⁷⁰ In healthcare, pay-per-use is fraught with complications, most prominently the information asymmetry problem that keeps patients from fully understanding their healthcare services and costs.¹⁷¹ Patients are simply not equipped to make many choices about constraining their use of healthcare services, especially when they occur in emergency situations.

In the bargain between public utilities and their regulating agencies is the grant of a monopoly to provide services in a given region in exchange for a duty to serve everyone, often at certain price levels.¹⁷² The monopoly bestows exclusive access to the market and ensures supply to the populace, but it does not compel demand. Consumers are not required to purchase the services or purchase a certain amount. To do so could run afoul of the Supreme Court’s Commerce Clause analysis in Sebelius, which would have struck down the individual mandate to purchase health insurance as an overstepping of Congress’s authority had it not been deemed a form of taxation.¹⁷³ It is in this respect that the public utility model falls short of its usefulness in a healthcare system that provides universal coverage—the model must reckon with the public’s expectations of electoral accountability in the face of taxation.

The inextricable relationship between taxation and voting traces its roots to an early moment in American history when colonists protested their lack of electoral representation by turning Boston Harbor into the world’s largest teacup.¹⁷⁴ One can imagine the fervor with which Americans might destroy crates of stethoscopes and gauze bandages in protest of healthcare taxes without a say in how those taxes are spent. The establishment of a specialized entity that lays taxes and delivers services to a particular population naturally invokes a governmental structure—in particular, the ubiquitous local public entities such as school districts and water districts. These special districts are generally governed by a board of elected representatives charged with hiring managers, monitoring the quality of services rendered, and tending to the responsible expenditure of public funds.¹⁷⁵ When applied to the concept of a healthcare district, the election of directors carries the potential to apply accountability measures that, in tandem with improved consumer price sensitivity, can impose downward cost pressures that counteract the market power of consolidated providers.

The public election of a district’s board of directors is a step toward accountability, but it begs the question: Accountability to whom? In theory, popularly elected directors stand to lose their jobs should taxes exceed what constituents deem acceptable or services fail to meet the desired level of quality. Directors always face the possibility that constituents may vote them out of office. These are conditions under which directors are seemingly incentivized to govern in a manner that is responsive to voter concerns. Indeed, free and fair elections are a hallmark of American democratic governance but so is the pervasive influence of special interests in the electoral process. Individuals do not comprise the entirety of the stakeholder population with an interest in a healthcare district’s decision-making. Providers, and the healthcare industry as a whole, will almost assuredly seek political influence to promote their interests. Such influence peddling can range from the standard fare of lobbyists sharing their expertise to financial contributions to candidates. It is the latter activity that concerned Reinhardt and led him to reject a single-payer model at the federal level “because [the United States federal] government is too corrupt. Medicare is a large insurance company whose board of directors (Ways and Means and Senate Finance) accept payments from vendors to the company. In the private market, that would get you into trouble.”¹⁷⁶ A purely governmental form does not appear wise when a taxing entity is susceptible to corrupt influence and capable of generating profits. Additional measures are in order to best ensure individuals’ interests remain protected.

B. Drawing on Private Organizational Forms to Balance Stakeholder Interests

A board of directors elected at the local level can be responsive to the ethical concerns of the electorate and its preferences regarding the allocation of resources.¹⁷⁷ Yet corruption, or even just the undue influence of interests other than that of individual voters, can stand in the way of public accountability.¹⁷⁸ This is only of concern if the interests of the public and the third-parties are not aligned, and providers’ interest in financial rewards stand to do just that.

As envisioned here, healthcare districts do not force providers to take on nonprofit status. The Healthcure System takes a general approach of noninterference with for-profit providers, recognizing that the pursuit of profits, when earned legitimately and not by taking advantage of market failures, may incentivize innovation and new efficiencies.¹⁷⁹ In Part IV, this Note instead proposes an aligned incentives payment structure that rewards providers for delivering on cost savings and outcomes measures.

While purely governmental entities are not profit-making ventures, it is not as though governments do not transact with private, for-profit entities.¹⁸⁰ Taxpayer dollars provide profits to contractors regularly.¹⁸¹ Yet, contractors willingly enter the public bidding process for government contracts with some expectation of their profit margins at the outset, and if the business proposition does not meet their business objectives, they can refrain from participating in the competitive bid.¹⁸² Here, a governmental structure does not suffice for healthcare districts because the relationship between districts and providers is not a contractual one entered into voluntarily. To realize the goals of aligned incentives and accountability, the Healthcure System devises that all providers retain their organizational form while uniting under the single healthcare district entity in which they share a financial stake. Because healthcare districts compel providers to provide services (to avoid the adverse selection problem of ACOs) and control their payment, providers lose a great deal of their pricing power. Although the districts can attempt to offset the pricing restrictions with promises of increased scale, it is nevertheless reasonable to expect that providers will seek to maximize their financial reward with the highest possible incentive payments.

Thus, an incentive payment model cannot fully rectify the tension between public interest and provider profit motive. The public utility model excels in accommodating a regulated monopoly that provides services in the public interest, but it fails to fulfill the need for electoral representation of a public taxed for their healthcare costs. The special district model does a better job of incorporating democratic ideals into the provider of specialized public services, but it cannot accommodate the profit-generating motive that some providers inevitably seek, such as individual physicians’ practices.

Where public entities fall short of balancing stakeholder interests, state legislators can look to variations on private organizational forms for guidance.

1. Allocating Rights and Responsibilities Between Patients and Providers

The underpinnings of the Healthcure System’s financial model are an amalgamation of seemingly conflicting revenue collection and distribution streams that do not lend themselves to a cut-and-dry organizational form. Starting with the residents of a healthcare district, individual tax payments form the inflow of capital with which a district procures healthcare services. All told, a district’s residents are stakeholders as (1) financial supporters, (2) beneficiaries of its services, and (3) electors of its board of directors. Once a healthcare district collects taxes, it distributes funds to private provider entities. All healthcare providers in the region are likewise stakeholders, as they (1) provide services to residents, (2) operate under the governance of the district’s board of directors, (3) receive funding from the district, and (4) share in the gains and losses of the district.

Yet, where one party has voting rights to elect a board of directors, and another lacks that right but bears exposure to the financial decisions the board makes, there arises a principal-agent problem.¹⁸³ The question then turns to how to legally organize such an entity.

In the private sector, the corporate form offers a number of options for interested parties to organize around a shared mission. Like governments, corporations are governed by representatives elected by a constituency of interested parties.¹⁸⁴ Those parties—the holders of shares of ownership in the corporation—need not retain the same rights and responsibilities as one another.¹⁸⁵ When for-profit corporations wish to assign different rights to different shareholders, they may issue preferred stock or create a dual-class share structure.¹⁸⁶ As such, certain shareholders may have priority over others when it comes to receiving financial distributions from the corporation, or they may have the right to vote on certain business matters that other classes of shareholders do not.

Ultimately, though, shareholders are owners of the corporation, and to regard individuals and providers as “owners” of a healthcare district opens the door to questions about the relative amounts of shares they hold and whether they can be transferred.¹⁸⁷ The concept of owning an entity that has the power to tax and regulate an industry confers power that would undermine electoral accountability to the public.

Other forms also offer distinct benefits but ultimately fail in their application to healthcare districts. A partnership, for instance, offers even more flexibility to cleave apart interested parties into distinct, nonoverlapping roles. Hypothetically, a healthcare district could make individuals partners or members of the manager-managed organization. These members can then assign their rights to distributions to providers while retaining their management rights.¹⁸⁸ An analog to the public’s role in a corporation might be a limited shareholder with a subscription (accounting for tax payments as the subscribed consideration) and retained voting rights but assigned distribution rights to providers. Another possible route for potential exploration is the treatment of providers as creditors to the healthcare district, or a complex contractual (or “contractarian”) relationship that binds individuals, providers, and districts, thereby avoiding the corporate form altogether.¹⁸⁹

This futile exercise represents a microcosm of an ongoing debate in health law. Clearly, a healthcare district is not conducive to cleanly applying a preexisting public or private form, but in light of the struggle between patient and provider interests, the healthcare industry has grappled with the limits of organizational forms for decades.¹⁹⁰ This is where a public-private partnership delivers useful inspiration.

Federal and state governments have established a variety of instrumentalities that skirt the line between public entity and private business. Although the United States has traditionally shunned the kinds of government-owned corporations that are prevalent in other parts of the world,¹⁹¹ federal corporations are still prominent fixtures in American life.¹⁹² Amtrak is one such quasi-corporation,¹⁹³ as are government-sponsored enterprises like the Federal National Mortgage Association (Fannie Mae) and the Federal Home Loan Mortgage Corporation (Freddie Mac).¹⁹⁴ Unlike a special district, a government corporation can be “[a] self-funding, self-perpetuating, profit-making corporation [that] enjoys a degree of potential, and perpetual, independence undreamed of in most agencies.”¹⁹⁵ Some of these entities feature characteristics particularly useful for the conception of healthcare districts, such as the ability to distribute dividends and a mixed ownership structure split between a preferred-stock-holding government and common-stock-holding private investors.¹⁹⁶ While the latter opens up a world of possibilities with respect to organizing private healthcare providers in healthcare district, it simultaneously raises questions over who and what guides the district’s decision-making.

The distribution of interests and rights presents healthcare districts with a distinct principal-agent problem, or perhaps stated more accurately, a principal-agent-principal problem.¹⁹⁷ When those principals have different interests, or “heterogenous preferences,” they must reckon with coordination costs in the form of agents’ difficulty with determining the right goals.¹⁹⁸ Confronted with such agency costs, organizations generally look to legal constraints on agents and corresponding enforcement mechanisms.¹⁹⁹ This Note will next explore the bounds of a healthcare district’s director’s role in light of their principals’ heterogenous preferences, asking whether directors can simultaneously act in the best interests of both individuals and providers.

2. Fiduciary Duties of Directors and Stakeholder Health Maximization

The election of a public official often involves what political scientists call a “mandate,” or the set of policy priorities that form a candidate’s platform and which the candidate is expected to implement upon election.²⁰⁰ Elected officials are often judged by how they deliver on the promises they made to voters on the campaign trail. Should they fail, they may lose reelection or even a recall election.

The elected representatives of corporations are guided by the ubiquitous and often legally enforced commitment to “shareholder wealth maximization” or “shareholder primacy.”²⁰¹ Delivering value to shareholders is the utmost concern, and a failure to adhere to this principle could result in liability for breach of fiduciary duty.²⁰² The representatives of nonprofit corporations, on the other hand, are guided by the organization’s mission.²⁰³ Directors are generally motivated, at least in large part, by an interest in maintaining tax-exempt status and adhering to the nonprofit’s stated mission in its articles of incorporation and IRS filings.²⁰⁴

The goals of a for-profit corporation and a nonprofit organization are not difficult to discern. But what if an entity comprises both? How should a board of directors square the push and pull of seemingly competing goals?

Healthcare today is dominated by entities that, on paper, appear to be either for-profit or nonprofit, but the rapid growth of nonprofit hospitals has blurred the line between them. The lack of an explicit profit motive does not stop nonprofit hospitals from generating enormous revenue exceeding their costs—but instead of distributing the funds to shareholders, they must instead reinvest the funds.²⁰⁵ ACOs particularly struggle with the conflicts of differing business models because their provider networks may consist of a mixture of for-profit and nonprofit entities.²⁰⁶ The Healthcure System proposed here is no different.

In an era of skyrocketing healthcare costs, health law experts have begun to reevaluate organizational forms in healthcare to better accommodate missions beyond shareholder wealth maximization. The public benefit corporation is a form available in some states, and it allows corporations to augment shareholder wealth maximization and express an additional mission.²⁰⁷ The form essentially provides “cover” to directors in the event that shareholders bring a lawsuit alleging a breach of fiduciary duty for taking actions not in the interest of shareholder wealth maximization.²⁰⁸ If the directors can point to a stated mission of the public benefit corporation as the motivating factor behind a challenged action, they will be shielded from liability.²⁰⁹

A full exploration of alternative corporate forms reaches far beyond the scope of this Note, but to the extent that some bear on healthcare, it will be useful to acknowledge prior efforts to develop novel forms. Dayna B. Matthew argued for a “fiduciary medicine model” that imposes new fiduciary duties on health care organizations, such as considerations of larger systemic duties and an expansion of fiduciary law to modern health care delivery systems.²¹⁰ Terry Corbett articulated a new legal form for ACOs based on the benefit corporation form. This form, the health care benefit corporation (“HCBC”), promotes accountability through legally enforceable mission primacy that can supersede the pursuit of profits.²¹¹

Accountability measures may even be found in antitrust. Rather than combat consolidation outright, the Healthcure System brings providers together and seeks downward price pressures elsewhere. This would seem to throw antitrust enforcement mechanisms out the window, such as blocking mergers or forcing divestitures in closely competing entities, but antitrust principles may serve some use when considering the role of the board of directors.²¹² Accordingly, “conduct remedies” that pertain to the behavior of consolidated entities can protect against price increases in the absence of traditional structural remedies.²¹³ For instance, states may direct healthcare districts to set an overall expenditure growth target that can be tied to the region’s economy and enforced by the state attorney general’s office as part of its antitrust enforcement.

Ultimately, though, the Healthcure System offers its own theory on which to base director fiduciary duties: what this Note calls “stakeholder health maximization.” Intentionally analogous to shareholder wealth maximization, stakeholder health maximization would be the paramount aim of healthcare district directors and is intended to orient their decision-making toward the constant improvement of health outcomes. Every decision, at its core, must be grounded in an effort to positively impact the health of the community. Cost savings, for instance, can be justified under stakeholder health maximization because the reduction of costs allows for improved allocation of scare resources: additional funds mean more people can get more care and better care. Conversely, efforts by providers to extract higher incentive payments without a justification based on health outcomes result in waste that would otherwise fund care.

With stakeholder health maximization as its guide, a healthcare district’s board of directors can navigate the distinct interests of its stakeholders with a potentially reduced risk of conflict. It is a path forward for balancing the three pillars of cost, quality, and access, even when stakeholders might push to prioritize one pillar over the rest. That is, as long as boards remain answerable to their constituents.

Director fealty to the public is as much an open question in elections of Congress as it is in the proposed healthcare district. There is no easy answer. The Supreme Court has made it clear that there is no path for constraining the rights of business entities such as for-profit healthcare providers to engage in political speech.²¹⁴ Ultimately, representative governance requires trust in the democratic process, and while the Healthcure System certainly relies on that trust, it is bolstered by protections such as stakeholder health maximization and an incentive payment model to help ensure that patient and provider interests are aligned.²¹⁵

V. QUALITY: ALIGNING THE INCENTIVES OF PROVIDERS AND PATIENTS TO DRIVE COST-EFFECTIVE, VALUE-BASED CARE

For the purposes of this Note, quality refers to both the scope of benefits available to individuals as well as the health outcomes of those benefits. Up to this point, this Note has explored the tax-based revenue stream and novel governance model of healthcare districts, but it has yet to describe how these features translate to cost savings and outcomes improvements. The third and final piece of the Healthcure System directly addresses the misaligned incentives of the traditional fee-for-service provider reimbursement model by replacing it with a three-part outcomes-based incentive structure.

A century’s worth of refinement of the managed care model and the recent piloting of ACOs have led to a moment in which the healthcare system can finally capitalize on the cost-efficiencies of vertical integration. Doing so will require the full participation of providers in the risk and reward to counteract the adverse selection and moral hazard problems that plague the system today.²¹⁶ The Healthcure System’s approach to healthcare payment reform draws on an integrated model that has been successfully implemented in California for over seventy-five years by the nonprofit health system Kaiser Permanente (“KP”).²¹⁷

KP is made up of three separate entities linked together by exclusive contract to share financial incentives, coordinate care, and manage the health of a population. These entities—the not-for-profit Kaiser Foundation Hospitals (“KFH”), the for-profit medical groups (made up of physicians), and the not-for-profit Kaiser Foundation Health Plan (the insurance company)—share board members and leadership to seamlessly coordinate the allocation of capital between them.²¹⁸ The KP model explicitly rejects fee-for-service and its unnecessary incentive on increasing quantity of services over quality, instead using a capitated model of payment. The capitation system consists of the health plan making monthly payments of a set dollar amount calculated per enrollee, regardless of whether they seek services.²¹⁹

If this arrangement sounds familiar, it is because the KP model is one of the earliest in a long lineage of managed care models that counts ACOs and the Enthoven-inspired Health Security proposal of the 1990s as siblings.²²⁰ This Note proposes a further refinement on the model, adding the layers of electoral accountability and novel governance as previously discussed, as well as a dividend payment scheme that emphasizes the shared responsibility of health within a particular region.

Under the Healthcure System’s stakeholder health maximization model, payments to providers consist of capitated payments as well as back-end payments based on outcomes and cost savings. The capitated payments provide hospitals a lump sum per event, and the provider then bears the cost of all services rendered during the patient’s event. But incentives to keep costs as low as possible can only serve a partial role in delivering an aligned-incentives payment structure. On their own, capitated payments can incentivize “bare minimum” treatment that would fulfill a district’s essential health benefits guarantee but disincentivize a great deal of innovative, risky, or even preventive treatments. Furthermore, there will surely be instances when an individual’s course of treatment exceeds the capitated payment a provider receives. While this is a risk the system is designed to handle, the system should equally incentivize providers to go above and beyond with their treatment and reward them when successful. Accordingly, the Healthcure System includes back-end payments in two forms: (1) event-based outcomes incentive payments and (2) regional dividends based on the overall population’s health benchmarks. Event-based incentives offer payments based on each service provided to a patient. Suppose a doctor’s treatments for a patient exceed the capitated payment for a particular illness, but they successfully treat the illness. The event-based outcomes incentive makes the treatment a financially sustainable one. The incentive payments may be calculated using similar formulas to those employed by the Medicare Quality Payment Program Merit Based Incentive Payments System.²²¹ But even worthwhile treatments may not always be successful, and there are strong policy reasons for encouraging best practices on a macro scale. Regional dividends reward providers for positive health trends across a community. Such dividends can encourage providers to collaborate and promote wellness beyond their particular practice. Moreover, the state or federal government may make additional adjustment payments in certain cases, such as rural healthcare districts, low-income districts, or catastrophic regional events such as natural disasters.

The result is a multi-payer, universal access healthcare system that ensures providers’ basic costs are covered, with downward price pressure to keep those costs as low as possible, while rewarding positive health outcomes and best practices with incentive payments from healthcare districts.

VI. PRACTICAL CONCERNS OF IMPLEMENTING HEALTHCARE DISTRICTS

The Healthcure System model proposed in this Note is nothing short of a radical rethinking of the overall structure of the healthcare system. To implement it on a state or even national scale would likely mean a reform package even larger than the ACA. As such, there are a number of practical barriers that would need to be overcome for a successful transition to take place.

This Note provides a general overview of the basic structure of the Healthcure System and a broad survey of the economic, legal, and policy considerations it implicates. Additional study will yield a better understanding of its practical applications, particularly through empirical modeling of its operation in various regions. Another topic for exploration is the refinement of formulas with which to draw healthcare districts, calculate healthcare tax rates, set incentive and dividend payment rates, and determine the startup funding necessary for regions to transition into healthcare districts.

The Healthcure System involves a litany of issues of state law, further increasing the challenge of consistent deployment across the United States. One can look to the resistance of states to participate in the Medicaid expansion of the ACA as a preview of the challenge ahead.²²² Even financial support from the federal government may not be enough to convince some states to adopt the plan, especially if it is seen as a comparable expansion of coverage through a public program. A related objection is the Healthcure System’s reliance on community rating instead of an actuarial fairness approach that attempts to price healthcare based on use.

Another topic ripe for exploration is the role of Medicare and Medicaid in the model proposed here. The MSSP’s ACO program served as a jumping-off point for the Healthcure System, but the scope of this Note does not include how it might incorporate or augment Medicare or Medicaid. Further study could evaluate whether there are additional efficiencies to be found by merging the public insurance programs with the Healthcure System.

Perhaps the single most significant objection to the Healthcure System is the general resistance to a substantial disruption to the healthcare system. Voters were so concerned that the ACA would make them change their health insurance that President Obama made the promise, “If you like your health care plan, you’ll be able to keep your health care plan” a central part of his pitch.²²³ This Note’s proposal upends healthcare by design. It is an effort to stave off unsustainable increases in healthcare costs by correcting market failures that are endemic to the status quo. A possible approach to easing the transition might be a gradual implementation of healthcare districts over a period of years, offering incentives to individuals who join early before all are eventually enrolled. Of course, this presents its own host of problems, namely a selective participation problem in which the risk pool may be concentrated with higher-risk individuals. A sound implementation plan would stave off such concerns by making every effort to allow individuals to keep their doctors, thereby reducing an otherwise significant barrier to enrollment.

CONCLUSION

This Note calls its proposed model the Healthcure System because it represents a fundamental fix to some of the most pervasive economic failings of healthcare in the United States. The Healthcure System aims to create downward pressures on cost by introducing three accountability measures: (1) accountability through price-elastic demand; (2) accountability to a population of voters; and (3) accountability through the aligned interests of stakeholders in an incentive payment structure. It does so by setting forth a novel organizational form that uniquely caters to the interest of patients and providers, and guides healthcare district directors to govern in the name of stakeholder health maximization. Although it would represent a monumental reform with an undoubtedly difficult challenge of federal and state lawmaking, the Healthcure System’s regional approach to universal healthcare access and reduced costs could finally deliver broad access, low cost, and high-quality healthcare to an ailing and priced-out America.

96 S. Cal. L. Rev. 1251

Download

Christopher A.G. LoCascio*

* Editor-in-Chief, Southern California Law Review, Volume 96; J.D. 2023, University of Southern California Gould School of Law; B.A. Political Science and English 2014, University of California, Riverside. Thank you to Professor Ankit Shah, Professor Jonathan Barnett, the editors of the Southern California Law Review, my family, friends, and classmates for your support throughout this Note’s writing and editing process.

Red, White, and Blue—And Also Green: How Energy Policy Can Protect Both National Security and the Environment

March 2024March 2024David M. Schizer*

Too often, energy policy protects the environment while neglecting national security, or vice versa. Since each goal is critical, this Article shows how to advance both at the same time.

For national security, the key is to avoid depending on the wrong suppliers. If they are vulnerable to attack (like some Middle Eastern producers), they need to be defended. Or, if they are themselves geopolitical threats (like Russia and Iran), their energy exports fund harmful conduct. This Article breaks new ground in showing why suppliers tend to be insecure or menacing: authoritarian regimes—which are more likely to pose these risks—have a comparative advantage in producing oil and gas, since they are less responsive to opposition from environmentalists, local residents, and other groups.

To avoid depending on the wrong suppliers, the U.S. and its allies should pursue two strategies. First, they should cut demand for fossil fuel. Along with making it easier to stop buying from the wrong suppliers, slashing demand also reduces greenhouse gas emissions and pollution. Yet although these are significant national security and environmental advantages, there is an offsetting national security risk: like fossil fuel, the main alternative—clean energy—also can foster dependence on insecure or potentially hostile suppliers (like Congo and China). In response, the U.S. and its allies should ramp up domestic production of clean energy technology, while also encouraging households and businesses to use it.

Second, since the transition to clean energy will take time, the U.S. and its allies also need to tap new sources of fossil fuel in countries that are secure and friendly. Yet since new fossil fuel development raises familiar environmental concerns, this Article proposes three ways to do it while still reducing emissions and pollution. First, these new sources should be as “clean” as possible (for example, natural gas instead of coal). Second, in adding new capacity, the goal should be to replace other fossil fuel sources, not to add to them (for example, so more production in the U.S. means less production in Russia). Third, new sources should be flexible, so they can ramp up and scale back as needed. Fortunately, these shifts are relatively easy for U.S. shale producers—indeed, more so than for others—and can be encouraged with the right regulatory approach.

While government intervention is needed to pursue these goals, policymakers should strive to harness the private sector’s capacity to innovate, cut costs, and enhance quality. A moratorium on new fossil fuel development is counterproductive, entrenching a status quo that depends too much on coal, as well as on insecure and hostile energy suppliers. Instead, the best approach is to “price” the relevant national security and environmental costs with Pigouvian taxes, motivating businesses and consumers to mitigate these costs and letting them choose how to do it. Yet if Pigouvian taxes are not politically feasible, this Article recommends a heuristic called “the marginal efficiency cost of energy”: policymakers should account for all the social costs of each source—private costs, national security costs, and environmental costs—and strive to replace high-cost sources with low-cost sources. This framework should guide all aspects of energy policy—from permits and regulations to rate-setting, mandates, moratoriums, subsidies, and government leases—so policymakers stay focused on both environmental and national security goals.

In December 2021, the Biden Administration blocked the construction of a natural gas pipeline from the Eastern Mediterranean to Europe. “Why would we build a fossil fuel pipeline,” the Administration’s senior energy advisor asked, “when our entire policy is to support new technology . . . and new investments in going green and in going clean?”¹

The answer to this rhetorical question should have been clear. In a word, it was “Russia.” Just three months later, Russia would invade Ukraine, and troops were already amassing on the border. To fund this military build-up, the Russian government depended heavily on energy exports, which accounted for a whopping 45% of its revenue.² Since Russia’s main market was Europe, one way to weaken Russia was to wean Europe off its energy. A new pipeline from the Eastern Mediterranean would help (although it would take years to complete). Even so, the Biden Administration nixed this pipeline,³ prioritizing the environment over national security.

As this example illustrates, energy policy has a profound impact on both national security and the environment, but too often the focus is on one or the other. Indeed, several prominent scholars have chosen to omit national security from their analysis altogether.⁴ For example, an influential 1996 book on energy security pointedly ignores the cost of defending Middle Eastern oil, arguing that energy is not the only reason for the U.S. to intervene in the region.⁵ But oil surely is one of the reasons. Ignoring it renders their analysis incomplete. Instead, energy policy needs to account for—and, indeed, to protect—both national security and the environment. This Article shows how to do it.

To enhance national security, the key is to avoid depending on the wrong suppliers. If they are vulnerable to attack (like some Middle Eastern suppliers), they need to be defended. Or, if they are themselves geopolitical threats (like Russia and Iran), their exports fund harmful conduct.

Unfortunately, it is no accident that fossil fuel suppliers often are insecure or hostile. This Article breaks new ground in explaining why. In democracies, fossil fuel production regularly faces staunch opposition from local residents, economic competitors, and environmental groups. But interest groups have less influence in authoritarian regimes, so production gravitates to these countries. This “authoritarian comparative advantage,” as the dynamic is called here, renders the U.S. and other democracies more dependent on authoritarian suppliers, which are more likely to be insecure or hostile.

To mitigate these national security risks, the U.S. and its allies should rely less on these suppliers. In general, there are two ways to do this. The first is to reduce demand for their product, while the second is to find other suppliers. This Article analyzes both alternatives, evaluating their implications for national security and the environment.

The first strategy—cutting demand for fossil fuel—lessens the stakes. There is less economic disruption when the U.S. and its allies stop buying from insecure or hostile suppliers. Finding other suppliers also is easier, since there is more slack in the system. Along with these national security advantages, reducing demand also yields familiar environmental benefits, reducing greenhouse gas emissions and pollution.

Yet although it is important to reduce demand for fossil fuel—for instance, by depending more on clean energy—this strategy poses national security risks of its own. Unfortunately, as with oil and gas, many clean energy suppliers are insecure or potentially hostile. For example, China is a leading supplier of EV batteries, solar panels, and minerals needed for clean energy. Arguably, replacing Russian hydrocarbons with Chinese clean energy is like jumping out of the frying pan into the fire.

In response, the U.S. and its allies should ramp up domestic production of clean energy technology, while also encouraging households and businesses to use it. To incentivize this effort, Congress offered a range of subsidies in the Inflation Reduction Act of 2022, although it remains to be seen how effective these subsidies will be; as I have emphasized elsewhere, targeted subsidies require Congress to pick which technologies to fund, but Congress often lacks the expertise and incentives to make the right choices.⁶ Even with these subsidies, moreover, clean energy still faces daunting regulatory barriers. For example, the permitting process for wind projects, mines, and solar farms is expensive, slow, and risky, but Congress failed to pass a 2022 bill on permitting reform.⁷ For this reason (and others as well), the “friend-shoring” of supply chains—and, more generally, the transition to clean energy—is likely to take many years.

Meanwhile, the U.S. and its allies should also pursue a second strategy. To ease their dependence on insecure and hostile fossil fuel suppliers, they should find other suppliers. Yet this effort, which involves adding new wells, pipelines, and infrastructure, raises familiar environmental concerns.

To square this circle, this Article proposes three ways to develop new sources of fossil fuel while still reducing emissions and pollution. First, these new sources should be as “clean” as possible; for example, natural gas generally is preferable to coal. Second, in adding new capacity, the goal should be to replace other fossil fuel sources, not to add to them. For instance, the point of increasing U.S. exports should be to reduce Russian exports. Third, new sources should be flexible, so they can ramp up and scale back, as needed. Fortunately, these shifts are relatively easy for U.S. shale producers—indeed, more so than for others—and can be encouraged with the right regulatory approach. For example, in awarding permits for a new pipeline or Liquefied Natural Gas (“LNG”) facility, the government should reserve (and pay for) the right to shut it down after a specified period. This would be much better than the Biden Administration’s decision in January 2024 to “pause” decisions on new export permits.⁸

To protect both the environment and national security, then, the U.S. and its allies need to reduce demand for fossil fuel, while also tapping new supply. But who is supposed to pursue these twin goals? After all, producing energy is not a government responsibility—at least not in the U.S. Rather, this is the job of private firms, and rightly so. They have the expertise and incentives to innovate, cut costs, and enhance quality. So even though authoritarian systems are better at overcoming interest group opposition, free societies have their own edge—economic dynamism—which they should harness.

Yet the private sector can do only what it is allowed to do. New wells and pipelines require permits, as do wind farms and mines for clean energy minerals, while extensive regulations also apply. The wrong government policies would thwart the approach recommended here. For example, a moratorium on new fossil fuel development—a step with influential supporters, including the International Energy Agency (“IEA”)⁹ —would be counterproductive, entrenching a status quo that depends too much on coal, as well as on insecure and hostile suppliers of oil and gas.

Yet the point is not for the government to leave these issues to the market, but to intervene the right way. The most efficient response is a Pigouvian tax. By adding environmental and national security harms to market prices, it creates financial incentives to mitigate them. At the same time, a Pigouvian tax lets consumers and businesses pick the solutions that are best for them, from electric vehicles and energy efficient appliances to shorter commutes, mass transit, better home insulation, and much more. The government does not have to pick specific responses to support—a key advantage because the government is not good at “picking winners.”¹⁰

Unfortunately, Pigouvian taxes have encountered stiff political resistance in the U.S.¹¹ If they are not available, policymakers should use other policy instruments to reduce the demand for fossil fuels and change the ones we use. To guide this effort, this Article proposes a heuristic called “the marginal efficiency cost of energy”: policymakers should account for all the social costs of each source—private costs, national security costs, and environmental costs—and then seek to replace high-cost sources with low-cost sources. This framework should guide all aspects of energy policy—from permits and regulations to rate-setting, mandates, moratoriums, subsidies, and government leases.

Admittedly, this agenda faces political challenges. Generating the requisite political support will require compromise, as well as an alliance between advocates for the environment and for national security.

Part I analyzes the national security costs of defending insecure fossil fuel suppliers, showing that these costs can be reduced by cutting demand for fossil fuel and adding secure new sources. Part II shows that this two-part strategy also addresses another national security cost of fossil fuel: strengthening hostile exporters. Since energy policy should also protect the environment, Part III briefly surveys two familiar environmental goals: limiting climate change and pollution. To identify synergies and tensions among the various national security and environmental goals, Part IV focuses on strategies to reduce demand for fossil fuel, while Part V considers strategies to tap secure new sources. Part VI generalizes these insights into a regulatory strategy, and Part VII is the conclusion.

I. DEFENSE EXTERNALITIES: A COST OF DEPENDING ON THE WRONG SUPPLIERS

When energy exporters are vulnerable to attack, they may need to be defended; indeed, the U.S. and its allies have protected Middle Eastern oil producers for decades. But this Part argues that instead of defending insecure suppliers, the U.S. and its allies should find ways to depend less on them. By tapping new sources of supply and reducing demand, the U.S. and its allies could cut their defense budgets.

Even so, some commentators and government agencies dismiss this potential benefit, deeming it too speculative to consider in energy policy. The last Section in this Part responds to their claims.

A. The National Security Implications of Energy

1. Defining National Security

Before considering this link between energy policy and national security, it is important first to clarify what the phrase “national security” means here. This Article uses the classical “realist” definition, which focuses on physical security and material well-being, rather than on the advancement of ideals.¹²

The goal here is to enhance the security of the U.S. and its allies, not to maximize global welfare. For example, conduct that is dangerous to U.S. citizens is considered harmful, even if it is beneficial to adversaries of the U.S.

To identify threats, this Article relies on the U.S. Intelligence Community’s annual threat assessments.¹³ The 2022 analysis highlighted four threats—China, Russia, Iran, and North Korea¹⁴—while earlier assessments also focused on terrorism.¹⁵

In mentioning “allies” of the U.S., this Article refers to countries that feature prominently as “allies and partners” in the Biden-Harris 2022 National Security strategy, including the U.K., Germany, France, and other NATO allies in Europe; Canada and Mexico in North America; and Japan, Australia, and South Korea in the Indo-Pacific.¹⁶

2. Links Between Energy and National Security

As the U.S. and its allies strive to counter security threats, energy is relevant in a number of ways. This Part focuses on the cost of defending suppliers, while the next considers the cost of empowering them.

But admittedly, these are not the only links between national security and energy. The military needs fuel to fight wars, just as it also needs weapons, rations, and other materials.¹⁷ The familiar response is to stockpile these supplies.

Energy policy also can cause environmental harms, which some classify as security threats.¹⁸ Yet as a matter of terminology, this Article classifies them instead as environmental harms, discussing them in Part III’s analysis of climate change and pollution.¹⁹

Energy also affects national security through the economy. Without cheap and reliable energy, it is harder to produce and deliver food, medicine, and other essentials; heat homes; enforce the law; maintain effective communications and transportation networks; and engage in a range of other indispensable activities. In short, energy is a fundamental ingredient of modern life.

To avoid severe economic and social disruptions, countries need to protect their electrical grids, pipelines, and power plants.²⁰ This is no different from the need to defend other vital infrastructure.

For similar reasons, countries also need reliable sources of fuel. Recognizing the importance of this precious resource, the literature used to focus on another security challenge: the economic drain from energy imports.

In the U.S., this was mainly an issue for oil, not for natural gas or coal. The U.S. became a net importer of petroleum in the 1950s, and these imports generally increased every year after 1954 until they peaked in 2005.²¹ Since the U.S. was the world’s largest oil importer for decades, the cost of these imports loomed large in the literature on energy and national security.²² In contrast, the U.S. did not depend on imports for natural gas during this period, although experts worried that this would change as U.S. reserves dwindled.²³ There was no such concern about coal, though. With the largest reserves in the world,²⁴ the U.S. has been a net exporter for decades.²⁵

Yet the economic drain from energy imports is no longer a concern in the U.S. In the past fifteen years, U.S. firms have unlocked vast oil and gas reserves in shale formations, using hydraulic fracturing.²⁶ This innovation has turned the U.S. into the world’s largest producer of oil and gas.²⁷ U.S. oil production surged 145% from 2008 to 2019, from 5,000 to 12,289 barrels per day.²⁸ Likewise, U.S. natural gas production increased 88% between 2005 and 2019.²⁹ After a dip during the coronavirus pandemic, U.S. gas production set a record in the summer of 2022,³⁰ reaching a level that once was unimaginable.³¹ This “shale boom” has turned the U.S. into a net exporter of oil and gas,³² so energy no longer contributes to the U.S. trade deficit. The economic drain of energy imports still burdens many U.S. allies, but not the U.S.

B. National Security Risks from Insecure Suppliers

Even so, the U.S. still faces another important energy security challenge, which is the focus of this Section: the risk of sudden contractions in the global supply of energy. This can happen if a supplier suddenly stops producing because of a revolution, war, or other geopolitical crisis. Notably, these supply shocks can still harm the U.S.—even though it is a net exporter—by triggering economically damaging spikes in energy prices.

1. Supply Shocks

These supply shocks can trigger both inflation and recessions in the U.S. and across the globe. For example, when Arab nations slashed oil production in 1973 and embargoed the U.S. to protest U.S. support of Israel, the spike in energy prices triggered nearly a decade of “stagflation.”

Soaring energy prices are especially painful for low-income households. Energy represents a larger percentage of their budgets, so price spikes are even more noticeable, causing difficult tradeoffs between oil, gas, and electricity, on the one hand, and necessities like food, medicine, rent, and education, on the other. Unlike wealthier households, families with low incomes do not have the liquidity to invest in more energy efficient cars, homes, and appliances or, in many cases, the flexibility to move closer to work or telecommute.

To head off these dire economic consequences, policymakers need “to ensure that the United States . . . is more resilient to inevitable global energy shocks,” Jason Bordoff and Meghan O’Sullivan have observed.³³

2. Cost of Defending Access to Energy

The traditional way to avoid energy shocks is to police access to fossil fuel, especially oil. For decades, the U.S. armed forces have “maintain[ed] the security of international oil flows for the global market,” a RAND Corporation analysis explained in 2009.³⁴

Like climate effects and pollution, this cost is not included in the price at the pump, so consumers do not consider these “defense externalities” in deciding how much fuel to use. Instead, “[t]he cost of those forces . . . generates a burden on the U.S. taxpayer.”³⁵

For many years, the U.S. has defended oil suppliers in the Middle East. For example, when the Soviet Union invaded Afghanistan in 1979, President Jimmy Carter warned that “[a]n attempt by any outside force to gain control of the Persian Gulf region will be regarded as an assault on the vital interests of the United States of America.”³⁶ Two years later, President Ronald Reagan pledged to defend oil producers from their neighbors as well.³⁷

Honoring this commitment, President George H.W. Bush protected Kuwait from an invasion by Iraq in 1990, invoking U.S. reliance on fossil fuels, among other things, to justify a military response. “[M]y administration, as has been the case with every President from President Roosevelt to President Reagan, is committed to the security and stability of the Persian Gulf,” he told the American people.³⁸ “Our country now imports nearly half the oil it consumes and could face a major threat to its economic independence.”³⁹

To secure the Middle East (and its oil) after the First Gulf War, the U.S. permanently stationed troops there for the first time. This U.S. military presence, especially in Saudi Arabia, was one of the reasons invoked by Osama Bin Laden to rally support for terrorist strikes against the U.S.⁴⁰

Bin Laden’s attacks on September 11, 2001 prompted the U.S. to invade Afghanistan. So, although this invasion was a response to terrorism, the terrorism itself was motivated (at least in part) by U.S. efforts to defend fossil fuels. “You can draw a thread through the whole thing with oil,” argued Admiral Dennis C. Blair, former director of National Intelligence.⁴¹

Similarly, although the U.S. invaded Iraq in 2003 for a number of reasons, energy was a key motivation for Vice President Dick Cheney. “Armed with an arsenal of these weapons of terror, and seated atop ten percent of the world’s oil reserves,” he observed six months before the invasion, Iraq’s leader Saddam Hussein could then be expected to “seek domination of the entire Middle East” and “take control of a great portion of the world’s energy supplies.”⁴²

Admittedly, the U.S. has intervened in the Middle East not only to protect its oil, but also to counter terrorism, support allies, contain rivals, and defend the principle of sovereignty. But although energy is not the only factor shaping U.S. defense policy, it is an important one, as a group of retired senior military planners affirmed in interviews for a 2018 study.⁴³ “We are not in the Persian Gulf because we are benevolent. We want oil to flow out of there,” one observed.⁴⁴ “Since the end of [the] Cold War, the only real threats we have are threats to the oil supply,” another said.⁴⁵ “[M]ore than half the Defense budget is for the security of Persian Gulf oil.”⁴⁶

These military efforts have long been reinforced by diplomacy. The U.S. has maintained close ties with oil-producing regimes, including ones that do not share U.S. values.⁴⁷ Propping up these authoritarian “petrostates” is all the more costly because they often are unstable.⁴⁸

3. As a Net Exporter, Can the U.S. Stop Worrying About Supply Shocks?

Can the U.S. stop supporting these regimes now that it has become a net exporter of petroleum?⁴⁹ Are Americans protected from oil shocks, as long as U.S. wells keep pumping? Unfortunately, the answer is “no.”

For one thing, key allies and trading partners still import oil, and their economic health affects the U.S. Energy shocks drain away money they otherwise would spend on U.S. goods and services, as well as on shared strategic interests.

Supply shocks also still affect the U.S. more directly: when consumers in Europe or Asia cannot buy from their usual supplier, they try to buy from U.S. producers, bidding up the price.⁵⁰ This is why U.S. gasoline prices spiked after Russia invaded Ukraine, even though the U.S. was importing very little oil from Russia.⁵¹ In a global market, a war or revolution thousands of miles away—involving suppliers who do not sell to U.S. consumers—can still cause U.S. prices to spike.

Does the U.S. have the same sort of exposure with natural gas? The answer is “yes, but not nearly as much.” The difference is that oil is easier to ship. Since a tanker can take Texas oil to either Athens or Alabama, buyers in both places can bid for it, yielding a (relatively) uniform global price.

In contrast, the price of natural gas is set locally because it is harder to transport. The cheapest way is a pipeline, but then the destination is fixed. If the pipeline goes to Alabama, Athenians cannot easily buy this gas. They would have to turn it into liquid, ship it on a tanker, and then turn it back into gas. This costly process requires a sophisticated infrastructure. At the moment, the U.S does not have enough liquefaction facilities to satisfy European demand. This constraint on exports leaves more gas for domestic consumption. As a result, prices in the U.S. are much lower than in Europe and Asia.

Even so, U.S. natural gas prices are still affected by global supply shocks, at least to an extent. When prices spike in other markets, U.S. suppliers can export at least some gas, a choice that reduces domestic supply. This helps explain why U.S. natural gas prices spiked for several months after Russia invaded Ukraine (while European prices went much higher).⁵² Looking ahead, global demand will have even more impact on U.S. prices as the U.S. builds more LNG facilities.⁵³

To sum up, energy supply shocks are still a challenge, even though the U.S. has become a net exporter of petroleum and gas. The traditional U.S. response has been to protect oil and gas suppliers, especially in the Middle East. The costs of defending these suppliers are a hidden price of fossil fuels.

C. Depending Less on Insecure Suppliers

Is the U.S. stuck bearing these defense externalities? Or can these costs be reduced over time? In principle, there is another way to deal with supply shocks: instead of defending insecure suppliers, we can depend less on them. Admittedly, reducing reliance on these suppliers can be difficult, especially on short notice. Yet easing this dependence has become much more plausible than it used to be, and the right policies can accelerate this progress.

1. An Illustrative Example

To illustrate different responses to defense externalities, assume that two neighboring countries, Emirate and Warmonger, are both oil exporters. Unfortunately, Warmonger has been threatening to invade Emirate. If war breaks out, the two countries’ combined exports of 4.3 million barrels per day will suddenly become unavailable.

The traditional way to avert this oil shock is for the U.S. and its allies to protect Emirate by issuing security guarantees and, if necessary, dispatching troops. If we replace “Emirate” with Kuwait and “Warmonger” with Iraq, this is precisely what happened in 1990.

Yet there are two other ways to avoid this oil shock. First, if another supplier can increase production by 4.3 million barrels per day, it can replace the exports from Emirate and Warmonger. To avoid extra defense costs, this supplier should be easy to defend. The U.S. obviously fits the bill, as do Canada, Brazil, Norway, Mexico, the U.K.,⁵⁴ and other secure jurisdictions. If enough additional supply can be coaxed from secure countries, there is less need to defend insecure ones. In this way, energy development can be a substitute for military spending. National security is protected with wells and pipelines, instead of troops and fighter planes.

Second, the same is true of reductions in demand. There would be no oil shock if the global economy could cut consumption by 4.3 million barrels per day through fuel efficiency, renewable energy, mass transit, and the like. “Meeting more of the energy needs of the United States through alternative sources of energy,” Bordoff and O’Sullivan have observed, “can lessen exposure to global markets by reducing U.S. consumption of oil and gas overall . . . .”⁵⁵ So, like new supply, efforts to reduce demand can be an alternative to military spending.

2. Reducing Dependence on Insecure Suppliers: Promising Trends

Admittedly, severing ties with a problematic supplier is sometimes quite challenging, especially in the short run. Indeed, after Russia invaded Ukraine in 2022, Europe’s efforts to stop buying oil, gas, and coal from Russia were painful. Even so, replacing—instead of defending—a supplier has become more realistic in recent years, and the right policies can make this alternative even more plausible.

For example, if some oil from the Middle East becomes unavailable, do other suppliers have the potential to replace it? A promising candidate is the U.S., where production has surged in recent years, as noted above. This increase (about seven million barrels per day) is much more than the 4.3 million barrels per day that Iraq and Kuwait were exporting in 1990 when Iraq invaded. In other words, the recent U.S. increase is almost twice the size of the disruption the U.S. intervened to prevent in the First Gulf War.

The U.S. also has ample reserves of natural gas and coal. As long as the U.S. has the necessary LNG terminals and other infrastructure to transport these fossil fuels, it may be able to replace other gas and coal producers in a crisis, even if they serve markets far from the U.S.

Obviously, any effort to replace other suppliers is more effective, and thus more likely to reduce defense externalities, when the products are fungible. For example, LNG can replace natural gas from a pipeline (though LNG is more expensive). Likewise, coal and natural gas are plausible substitutes for each other since both generate electricity. Yet neither can

substitute for oil, as long as oil (not electricity) is the main fuel for transportation.

Looking ahead, could the U.S. develop the potential to increase fossil fuel production even more in a crisis? Do U.S. firms have the capacity and incentives to ramp up? Could policymakers encourage them to do so? If the answer is “yes,” this backup capacity would reduce the pressure to defend other suppliers.

This pressure would ease not only if the U.S. and its allies could produce more fossil fuel, but also if they used less. In a supply shock, prices do not spike as much if demand also contracts. Even if prices do rise, there is less harm if the economy is less reliant on fossil fuel.

This brings us to a second promising trend: even as demand for fossil fuel has increased worldwide (and especially in the developing world), the U.S. and its allies have become less dependent on fossil fuel in recent years. For example, the “energy intensity” of the U.S. economy—a measure of how much energy is needed to produce a given level of economic output—is less than half of what it was forty years ago.⁵⁶ Even as the population and economy have grown significantly, U.S. oil consumption has held steady at about 18.5 million barrels per day.⁵⁷ Vehicles and appliances have become more energy efficient,⁵⁸ and the U.S. economy depends less on energy intensive industries, such as manufacturing. In addition, renewable energy has become less expensive, and thus more common. It generated 12.61% of all energy used in the U.S. in 2021—a new high⁵⁹—as well as 67% of new electric power generation in the first half of 2022.⁶⁰

Can U.S. firms and households build on this progress? Can policymakers encourage this trend? Again, if the answer is “yes,” there would be less pressure to defend insecure fossil fuel suppliers.

3. Encouraging Extra Supply

How can the U.S. and its allies keep reducing defense externalities? What else can they do to tap more supply from secure sources, while also cutting demand? How can policymakers reinforce these trends?

i. Stockpiles

A key challenge is timing. Supply shocks come on suddenly, but it takes time to tap new supply and reduce demand. Until these efforts bear fruit, the U.S. and its allies are exposed to higher prices. If these responses take years to implement, instead of weeks or months, there could be significant economic disruptions in the interim.

As a (partial) response, the U.S. and its allies can rely on stockpiles of fossil fuel, such as the Strategic Petroleum Reserve (“SPR”) for oil.⁶¹ “[E]mergency stocks could smooth economically harmful price spikes until markets are able to adjust,” observed Jason Bordoff, Antoine Halff, and Akos Losz.⁶²

Even so, a stockpile is more effective when the supply shock is temporary. Since a stockpile’s supply is finite, the market knows it eventually will run out. The key question, then, is whether the stockpile can outlast the supply shock. If the answer is “yes”—for instance, while a pipeline is being repaired—prices should remain stable. But a stockpile is less effective when the shock is expected to persist, which is likely for a revolution, an invasion, or another geopolitical crisis.⁶³ Since market prices are forward-looking, they will still rise, even when supply from the stockpile is released, because everyone knows the extra supply is only temporary.⁶⁴

As a result, a stockpile alone cannot address supply shocks. Other measures are also needed, which either increase supply or reduce demand. A shock is averted only if the market expects these measures to kick in before the stockpile runs out.

ii. Spare Capacity

When the solution is new supply, it needs to get to market quickly. Yet, although firms have economic incentives to ramp up production when prices rise, a rapid pace often is not feasible.

“Generally speaking, the oil industry is highly capital intensive and relatively slow moving,” observed Bordoff, Halff, and Losz.⁶⁵ “Most oil development projects cost billions of dollars and take years to bring into production.”⁶⁶ Natural gas projects have an added challenge, emphasized above: transporting gas requires either pipelines or liquefaction facilities, which take years to build.

Fortunately, some suppliers can respond more quickly. In the oil market, the fastest response is what the International Energy Agency calls “spare capacity”: additional production that comes online within thirty days and lasts for more than ninety days. This pace usually is feasible only for Saudi Arabia. It “maintains the largest spare capacity and has historically played the role of ‘swing’ supplier,” explained Bordoff, Halff, and Losz, “adjusting production in line with market conditions.”⁶⁷

Even so, Saudi Arabia is not always able (or willing) to ramp up oil production. For example, after Russia invaded Ukraine, the Saudis agreed to only a minor increase.⁶⁸ A few months later, they cut production, disregarding a U.S. request to pump at capacity.⁶⁹

As this disagreement highlighted, Saudi and U.S. interests sometimes diverge. For one thing, the Saudis benefit from high oil prices. The U.S. and the Saudis also have clashed over Saudi ties to Russia, U.S. diplomatic approaches to Iran (the Saudis’ main regional rival), and the murder of a dissident Saudi journalist. The relationship was further strained by Joe Biden’s comments on the kingdom while running for President: asserting that there was “very little social redeeming value in the present government in Saudi Arabia,” he pledged to make them “the pariah that they are.”⁷⁰ Indeed, ties between the Biden Administration and the Saudi leadership were so frayed that when the Saudis restored diplomatic relations with Iran in 2023, they worked through China instead of the U.S., a step that was “a real slap in the face to Biden.”⁷¹

iii. Increasing Supply in Other Ways and Reducing Demand

Instead of relying on Saudi Arabia to stabilize global oil markets, the U.S. would be better off developing its own backup capacity, which could be tapped in a crisis. But is this feasible? Can U.S. oil producers ramp up quickly enough to play this role? What about the U.S. natural gas industry? How can policymakers encourage faster responses?

In general, the answer depends on the type of well and the availability of key infrastructure. Offshore wells take years for permitting, construction, and drilling, costing billions of dollars. But fortunately, drilling in shale is different.⁷² The “ability of US shale producers to ramp output up or down relatively quickly in response to price signals or changing market conditions,” Bordoff, Halff and Losz have explained, “could be seen as a form of insurance against disruption risks . . . .”⁷³ The same is true of new natural gas wells in shale.

Yet even if wells can be drilled quickly, pipelines and other infrastructure are needed to bring oil and gas to market. Even so, with the right infrastructure in place—and, more generally, with the right policies—the U.S. could take advantage of the elasticity of shale production to respond to supply shocks. Part V of this Article explores this possibility, and the synergies and tradeoffs it presents for national security and the environment.

To become less dependent on insecure suppliers, the U.S. and its allies also should reduce demand. Like new supply, this response takes time but, again, the right policies can accelerate it. Part IV explores the national security and environmental implications of promoting energy efficiency and renewable energy.

Admittedly, neither of these strategies—increased supply or reduced demand—is easy to execute on short notice.⁷⁴ But the same is true of an effective military response. All of these efforts require long-term investment and preparation.

The fundamental question, then, is which response maximizes welfare. To head off supply shocks—and, more generally, to access energy at the lowest social cost—is it better to build aircraft carriers, drill new wells, or install electric vehicle charging stations? The answer is a combination of measures—not just military responses, but also new sources of fossil fuel, as well as efforts to use less of it.

D. Division of Labor Between the Private and Public Sectors

Which institutions are supposed to pursue these various goals? Unlike in some countries, the U.S. does not have government-owned energy companies, which could be tasked with implementing government policy along with earning profits.

By relying instead on the private sector, the U.S. reaps familiar benefits. In a competitive market, private firms have strong incentives to cut costs and experiment with new approaches. In this way, the private sector sometimes delivers transformative innovations, such as the U.S. shale boom.⁷⁵

Yet a familiar downside of private firms is that they do not minimize negative externalities, such as the national security and environmental costs in this Article. Rather, addressing these externalities requires a government response. Policymakers can choose from a range of policy instruments, including Pigouvian taxes, permitting policies, subsidies, moratoriums, and mandates. Part VI surveys various options, highlighting their advantages and disadvantages.

E. Objections to Considering Defense Externalities in Energy Policy

So far, this Part has argued that depending on fossil fuel adds to the defense budget, and that policymakers need to account for this cost in evaluating the merits of different energy sources. However, other commentators have taken the opposite view, urging policymakers to omit defense externalities from this analysis. Douglas Bohi and Michael Toman made this case in an influential 1996 book.⁷⁶ Several other commentators have followed their lead,⁷⁷ as have a number of U.S. government agencies. For example, a 2018 analysis of tougher fuel economy standards omitted the national security advantages of using less petroleum,⁷⁸ as did a 2009 National Research Council study⁷⁹ and a 1992 Congressional Research Service report.⁸⁰ The literature has offered two reasons to ignore defense externalities, and this Section shows why neither is persuasive.⁸¹

1. Many Factors Influence Defense Policy

The first argument is that energy is just one of many factors affecting the defense budget, so its impact is too hard to isolate.⁸² “Until an effort that yields a credible measure of the externality involved is completed,” Bohi and Toman wrote, “this externality is too uncertain to be used in determining energy policy.”⁸³

But even when a cost is difficult to quantify, we should not simply ignore it. After all, we do not do this with climate externalities, even though they are hard to measure.⁸⁴ Instead, the right approach is to use the best available estimate, however imperfect it is.

For example, a 2018 study concluded that the Pentagon spends 16% of its general operating budget to protect Middle Eastern oil.⁸⁵ To cover this cost, U.S. gasoline taxes would have to increase by 28 cents per gallon⁸⁶ (and by an additional 70 cents to fund the wars in Afghanistan and Iraq).⁸⁷ Admittedly, this estimate may be off the mark. Other studies propose estimates of their own—some higher, some lower—by using different methodologies and assumptions.⁸⁸ The goal here is not to defend a particular estimate, but to show that these costs are too substantial to ignore.

2. Can Shifts in Energy Markets Change the Defense Budget?

Second, these commentators argue that energy policy should ignore the cost of defending fossil fuel, not only because this cost is hard to measure, but also because it is fixed. In their view, the U.S. is stuck protecting insecure suppliers, and energy policy cannot do anything about it. For example, even if U.S. energy policy delivers modest increases in energy efficiency and in domestic oil and gas production, the U.S. would still have to defend the Middle East.⁸⁹

While this argument is persuasive for limited changes, there have been major shifts since Bohi and Toman made this claim in 1996. As emphasized above, domestic production of oil and gas has surged, while energy efficiency and renewable energy have enabled the U.S. and its allies to depend less on fossil fuel. If government policy can accelerate these trends, there will be less pressure to defend insecure suppliers.

To sum up, American dependence on fossil fuels has added to U.S. military and foreign policy burdens. These “defense externalities” are a hidden price of oil and gas. To reduce the cost of defending insecure suppliers, the U.S. and its allies need to depend less on them. The answer is a two-pronged strategy, which reduces demand for fossil fuel, while also tapping new supply in the U.S. and other secure locations.

II. FUNDING EXTERNALITIES: ANOTHER COST OF DEPENDING ON THE WRONG SUPPLIERS

While the last Part analyzed the cost of protecting fossil fuel suppliers, this Part considers the cost of strengthening them. Unfortunately, some energy exporters use fossil fuel revenue to pay for harmful conduct. For example, Russia’s energy exports have financed its war in Ukraine. Buying from bad actors can facilitate their threatening behavior. Like defense externalities, these “funding externalities”—the national security costs of providing revenue to dangerous suppliers—do not appear in the price of energy.

So how should policymakers respond? As with defense externalities, the key is to depend less on the wrong suppliers. Again, the U.S. and its allies should use less fossil fuel, while also finding other (friendly) suppliers. Even so, some commentators urge policymakers to ignore funding externalities, so the last Section of this Part responds to their claims.

A. National Security Risks from Dangerous Suppliers

There are national security risks from depending not only on insecure suppliers, but also on threatening ones. “Governments of some countries openly hostile to the United States . . . ,” the RAND Corporation observed, “rely on oil exports for most of their budget revenues.”⁹⁰ They use this revenue to pursue harmful agendas both abroad and at home.⁹¹ Unfortunately, buying from these suppliers undermines national security by facilitating this threatening behavior.

1. Funding War, Terrorism, and Other Threats

Let us begin with the harm these suppliers cause outside their borders. Russia is the quintessential example. In 2022, its energy exports paid for its invasion of Ukraine. Before the war, Russia was the world’s largest exporter of oil and natural gas⁹² and Europe’s main supplier.⁹³ Russia was also the world’s third largest exporter of coal in 2021 (behind Australia and Indonesia and ahead of the U.S.).⁹⁴ Since Russia’s largest energy companies were state owned and private producers were heavily taxed,⁹⁵ nearly half of the Russian government’s revenue in 2021 came from energy exports.⁹⁶ “Increased production and long periods of high prices gave President Vladimir Putin the resources to beef up Russia’s army and throw his weight around,” Ricardo Hausmann has observed.⁹⁷

Since the costs of “throwing his weight around” did not appear in the market price, consumers did not account for them in deciding how much fossil fuel oil to buy. Europe kept paying Russia for energy, even as troops amassed on the Ukrainian border. Once the war began, the human toll from these purchases became clear: hundreds of thousands of casualties, millions of displaced civilians, and massive economic dislocation.⁹⁸ As the foreign minister of Lithuania put it, “buying Russian oil and gas is financing war crimes.”⁹⁹ Yet even as the war raged, Russia still earned nearly $1 billion per day exporting energy. Although the U.S. and its allies tried to stop buying from Russia, soaring oil prices—stoked in part by the war itself—kept Russia’s coffers full in the months following the invasion.¹⁰⁰

Russia is not the only fossil fuel exporter that threatens the U.S. and its allies. Iran uses export revenue to finance its nuclear program,¹⁰¹ as well as terrorist organizations such as Hamas and Hezbollah.¹⁰² Venezuela poses a threat to its neighbors,¹⁰³ and Saudi Arabia funds institutions that promote Islamic extremism.¹⁰⁴

2. Extortion

Hostile fossil fuel exporters can threaten others by harnessing not just the revenue they earn, but also the leverage they exert.¹⁰⁵ To influence their buyers’ policies, exporters can use fossil fuel as a carrot (by dangling favorable terms) or a stick (by threatening to stop selling).

Again, Russia’s invasion of Ukraine is a paradigmatic example. When Europe supported Ukraine, Russia retaliated by reducing the flow of natural gas, causing prices in Europe to skyrocket.¹⁰⁶ The goal was to pressure Europeans to “vote their pained pocketbooks,” Daniel Yergin and Michael Stoppart explained.¹⁰⁷ “The ultimate aim is to bring governments to power in Europe that aren’t committed to supporting Ukraine . . . .”¹⁰⁸

Unfortunately, Europe could not easily replace Russian gas on short notice. The price of LNG skyrocketed—even as U.S. firms redirected their LNG exports from Asia to Europe—because there were not enough liquefaction facilities to meet Europe’s needs.¹⁰⁹ These shortages exacted a severe economic toll as inflation spiked and the economy slowed.¹¹⁰ This energy crisis exposed the geopolitical cost of depending on Russia, empowering it not only with export revenue, but also with the ability to turn off the spigot.

3. Entrenching Repressive Regimes

Buying from hostile petrostates empowers them to harm not just the U.S. and its allies, but also their own people. Authoritarian leaders often use this revenue to line their own pockets and stay in power.¹¹¹ Again, this cost does not appear in the price at the pump.

Venezuela is a tragic example. Despite its vast oil reserves,¹¹² Venezuela has faced hyperinflation and a steep decline in GDP in recent years.¹¹³ The poverty rate is 90%, and food shortages caused the average citizen to lose 24 pounds in 2017.¹¹⁴ Meanwhile, President Nicolás Maduro’s “policies are marked by authoritarianism, intolerance for dissent, and violent and systematic repression of human rights and fundamental freedoms.”¹¹⁵

Even so, Maduro clings to power by exporting oil. These sales represent 99% of the nation’s export earnings and about 25% of its GDP.¹¹⁶ Maduro controls this revenue, using it to maintain the military’s support and reward political allies.¹¹⁷

To sum up, there are national security risks from buying oil and gas from Russia, Iran, and other threatening suppliers. These “funding externalities” arise when suppliers use export revenue to finance wars and terrorism abroad and repressive policies at home.

B. Depending Less on Hostile Suppliers

So what should the U.S. and its allies do? How can they reduce these funding externalities? As with defense externalities, the key is to depend less on the wrong suppliers. Indeed, since the responses are so similar, the discussion here can be brief.

To avoid empowering hostile suppliers, the U.S. and its allies should stop buying from them, while encouraging others to do the same. But if these commercial ties need to be severed quickly, there is a risk of a supply shock. Indeed, this is what happened after Russia invaded Ukraine, prompting Europe to wean itself off of Russian oil, gas, and coal.¹¹⁸

How can the U.S. and its allies mitigate these supply shocks? As with defense externalities, the answer is a two-part effort. Along with reducing demand, they should tap more supply in friendly countries, such as the U.S., Canada, Mexico, Brazil, Norway, Israel, Cyprus, and the U.K.

In responding to funding externalities, the U.S. and its allies face an additional challenge, which does not arise with defense externalities: persuading other countries to stop buying from the relevant supplier.

Why the difference? Either way, the U.S. and its allies do not buy from the supplier, but the reason is different. With defense externalities, the supplier cannot sell (for example, because it has been invaded). In contrast, with funding externalities, the supplier might still want to sell, but the U.S. and its allies do not want to buy from it (for example, because it has launched an invasion).

In refusing to buy, their goal is to deprive the supplier of revenue, and thus to reduce its military and economic power. Yet this goal will not be achieved if the supplier can simply sell to other buyers. To discourage these other buyers, the U.S. and its allies can try a range of policies, including embargoes, tariffs, price caps, restrictions on financing and insuring cargoes, sanctions on buyers, and the like.

Admittedly, these policies can be hard to enforce. Some countries will not adopt them. Hostile suppliers also might evade them with deception (for example, by selling through intermediaries, falsifying records, and so forth).

Fortunately, however, even porous sanctions can still reduce funding externalities, as long as they force the hostile supplier to sell at a discount, eroding the funding for its harmful agenda. For example, even though China and India did not join the U.S.-led embargo of Russian oil in 2022, they bought this oil at a steep discount.¹¹⁹

C. Objections to Considering Funding Externalities in Energy Policy

While this Article warns about risks from buying fossil fuel from hostile suppliers and offers a strategy to deal with these funding externalities, some commentators are not troubled by these risks. In their view, the real problem is with the hostile regime itself, not with the commodities it exports, and that sometimes the best way to moderate the regime is to buy its energy. This Section responds to these claims.

1. The Problem is the Regime, Not its Fossil Fuel Exports

Even though a number of fossil fuel exporters pose a threat to the U.S. and its allies, some commentators urge us to distinguish between these regimes, on the one hand, and their exports, on the other.¹²⁰ This argument is a bit like the mantra of some gun rights advocates: “Guns don’t kill people, people kill people.”¹²¹ In this spirit, “fossil fuel exports don’t harm people, exporting regimes do.”

They are right that not all fossil fuel exporters are threatening. After all, Canada and Norway are major exporters.¹²² At the same time, some potentially threatening regimes are not fossil fuel exporters, including China and North Korea.¹²³

But unfortunately, some fossil fuel exporters clearly do pose a threat to the U.S. and its allies, including Russia and Iran. Their fossil fuel exports give them more power to pursue their threatening ambitions. Indeed, if the invasion of Ukraine in 2022 has taught us anything, it has revealed the folly of ignoring defense costs in energy policy. The willingness of Europe—and of Germany in particular—to become so dependent on Russian energy has turned out to be a grave mistake.

In Russia, Iran, and other energy producers, energy exports do not just facilitate harmful behavior; in some cases, they actually cause it. As Michael Ross has argued, a government funded by energy exports is less accountable,¹²⁴ and thus is more likely to pursue reckless policies. To extract oil and gas, the government can rely on a small fraction of the population (or on foreign partners).¹²⁵ So, instead of depending on the labor, tax dollars, and the good will of its people, the regime can use export revenue to fund a police state, buy off dissent, and control the press.¹²⁶ This revenue also can cause a “resource curse,” undermining entrepreneurship, diversified growth, and the social rights they facilitate.¹²⁷ For a regime presiding over this sort of stagnant economy, an aggressive foreign policy can rally domestic support, tapping into nationalist sentiment, justifying military expenditures that keep the military on its side, and distracting citizens from the regime’s failings. As a result, it is no accident that petrostates tend to be unstable or aggressive (or both).

In short, it is not always persuasive to distinguish between a threatening regime and its fossil fuel exports. These exports facilitate (and sometimes may even motivate) its aggressive and repressive policies.

2. Target the Harmful Conduct, Not the Revenue That Funds It

Even if fossil fuel exports contribute to harmful conduct, some commentators argue that the right response is to target the conduct, not the revenue that pays for it.

It would be better “to address the foreign policy problem directly,” the National Research Council argued in its 2006 report, instead of “reduc[ing] oil consumption to lower world prices,” since “such an effort would be an imperfect proxy for better targeted instruments and would hurt oil producing friends and foe alike.”¹²⁸

Admittedly, targeting the behavior is sometimes more efficient, but this is not always true. In some situations, going after the revenue stream could be cheaper, less risky, more technologically feasible, or otherwise more effective. Nor are these approaches necessarily alternatives.¹²⁹ For example, when Russia invaded Ukraine, the U.S. and its European allies reduced their purchases of oil and gas from Russia, while also supplying military and humanitarian aid to Ukraine. Ultimately, the right answer is to pick the response—or, indeed, the combination of responses—that is most efficient under the circumstances.

3. Exports Might Moderate the Regime

In this spirit, there may be times when the most efficient course is not to stop buying from a hostile regime, but to buy more from them. If trade would moderate a hostile regime, persuading it not to pursue aggressive or repressive policies, the externalities for these purchases actually would be positive, instead of negative.¹³⁰

How can trade have this beneficial impact? For one thing, it gives a regime’s leaders something to lose. They may shy away from an aggressive policy if they think it will jeopardize export revenue (but not if they expect this revenue to continue anyway). In addition, trade can moderate a regime by empowering constituencies that press for change, such as a pro-Western middle class. Unfortunately, fossil fuel exports often have the opposite effect of strengthening regime loyalists.¹³¹

For decades, Germany hoped to moderate Russia through trade (and, a cynic would add, to reap the commercial advantages of cheap Russian energy). Angela Merkel doggedly pursued this policy as chancellor.¹³² Her predecessor, Gerhard Schröder, struck the deal to build the Nord Stream 1 pipeline and then earned sizable sums after leaving office as chair of the pipeline’s shareholder committee and a board member of Russia’s state-controlled oil company.¹³³ Schröder has not been willing to concede error even after Russia invaded Ukraine. “I don’t do mea culpa,” he said.¹³⁴ “It’s not my thing.”¹³⁵ Nevertheless, his approach to Russia has been thoroughly discredited. “Obviously, this policy has totally failed,” said Marcel Dirsus, a German security expert, articulating a widely shared view.¹³⁶

Yet, although commercial ties have not moderated Russia (but, on the contrary, have made the country a more dangerous foe), this will not be true of every regime. To predict the effect of export revenue, policymakers need to make context-specific judgments about a country’s leadership, the potential influence of other constituencies, and the like. But the general assumption in this Article is that policymakers will want to weaken and deter hostile regimes, not to rely on trade to moderate them.

To sum up, the key to weakening hostile suppliers is to depend on them less. In this spirit, the U.S. and its allies should reduce demand for oil and gas, while also looking for new (friendly) suppliers. Through this two-part strategy, energy policy can enhance national security by reducing both defense and funding externalities.

III. ENVIRONMENTAL GOALS AND CONSTRAINTS: CLIMATE, POLLUTION, AND THE AUTHORITARIAN COMPARATIVE ADVANTAGE IN ENERGY PRODUCTION

So far, this Article has focused exclusively on national security. Yet environmental goals are also important in energy policy, so we need to understand how these goals affect the analysis. What are the environmental implications of the proposal in Parts I and II to reduce demand for fossil fuels and tap new supply? What changes, if any, are needed to ensure that this strategy protects the environment, as well as national security? The rest of this Article focuses on these issues.

To lay the groundwork for this analysis, this Part briefly surveys two familiar environmental goals in energy policy: limiting climate change and pollution. How can we advance these goals, while also reducing demand for fossil fuel and tapping new supply? Parts IV, V, and VI identify synergies and tensions among these various goals, showing how to make progress on all fronts.

But before the rest of this Article digs into these policy details, this Part identifies a blunter tension between the environment and national security, which is rooted more in political economy than in policy. To enhance national security, the U.S. and its allies should produce more energy domestically, so they depend less on insecure and hostile suppliers. But unfortunately, democracies are not easy places to produce energy. Opposition to energy production—whether from local residents, environmental organizations, or other groups—gains more traction in democracies than in authoritarian regimes. As a result, the production of fossil fuel gravitates to authoritarian countries, as do some aspects of the production of clean energy. This unfortunate reality, which this Article calls the “authoritarian comparative advantage,” can harm both national security and the environment.

A. Climate Harm from Fossil Fuels

The connection between energy policy and climate change is familiar: fossil fuel is the key driver of rising temperatures.

1. Fossil Fuel, Emissions, and the Social Cost of Rising Temperatures

The concentration of CO₂ in the atmosphere has increased by 50% in the 250 years since the industrial revolution began.¹³⁷ A scientific consensus has emerged that these emissions are raising global temperatures.¹³⁸

“The largest source of CO₂, and of overall greenhouse gas emissions,” the EPA recently reported, “was fossil fuel combustion primarily from transportation and power generation.”¹³⁹ For example, petroleum represents 90% of transportation fuel in the U.S., accounting for 27% of total U.S. emissions.

According to the U.N. Intergovernmental Panel on Climate Change (“UN IPCC”), rising emissions are already causing a range of harms, including extreme weather, wildfires, water shortages, rising sea levels, more heat-related deaths, and species extinctions.¹⁴⁰ Looking ahead, the UN IPCC warns of significant economic losses from submerged coastal property, damaged infrastructure, effects of heat on health and productivity, storm damage, and reduced crop yields.¹⁴¹

These costs do not appear in the market price of energy. So, like the funding and defense externalities discussed above, consumers do not account for them in deciding how much energy to use.

2. Climate Change as a National Security Threat

In addition to the costs described above, the Biden Administration has emphasized that some climate harms affect national security. When the administration requested an analysis of this question, the intelligence community highlighted three issues. First, there will be geopolitical tension about how to respond to climate change, including the speed of the response, who will pay for it, and whether China and India will join the effort.¹⁴²

Second, nations may clash over resources and refugees. For instance, there will be competition for fresh water, as well as for resources in the Arctic, a region that will become more accessible as temperatures rise.¹⁴³ In addition, when areas become uninhabitable, the flight of refugees will stoke tensions along borders.¹⁴⁴

Third, climate change will be especially costly in warmer regions. This could “increase the potential for instability and possibly internal conflict” in central Africa, Latin America, South and East Asia, and island nations in the Pacific.¹⁴⁵

In principle, this Article could label these climate-related national security risks either “climate” costs or “national security” costs. For clarity of exposition, this Article calls them “climate” costs, as noted above,¹⁴⁶ but this choice should not affect the analysis. After all, a cost is a cost, regardless of what we call it. Either way, energy policy is more efficient if it accounts for these externalities, as well as the others flagged in this Article.

3. Reducing Climate Externalities from Fossil Fuel

To mitigate climate harms, the UN IPCC has called for “[n]ear term actions that limit global warming to close to 1.5°C.”¹⁴⁷ This step, the UN IPCC has said, “would substantially reduce projected losses and damages related to climate change.”¹⁴⁸

Given the role of fossil fuel in climate change, energy policy needs to feature prominently in this effort. The right policies can reduce climate change in three ways. The first is to dial back the use of fossil fuel. Greater energy efficiency reduces the need for it, as does the wider use of clean energy, such as solar, wind, and nuclear power.

Second, since fossil fuels vary in their climate impacts, it is better to use ones with lower carbon footprints. Coal is the worst offender, since burning it produces nearly twice as much CO₂ as burning natural gas.¹⁴⁹ Admittedly, natural gas has a limitation of its own: its main component, methane, is a potent greenhouse gas that can leak into the atmosphere.¹⁵⁰ But as long as these leaks are prevented—and they are, indeed, preventable—emissions can be slashed by replacing coal with natural gas.¹⁵¹

This is precisely what has happened in the U.S., causing U.S. emissions to fall even as they have increased in China, India, and the developing world (and thus overall).¹⁵² In the U.S., the percentage of electricity generated by coal fell from about 50% to 24% between 2007 and 2019, with natural gas picking up most of the slack. “That was the main reason,” Dan Yergin observed, “why U.S. carbon dioxide (CO₂) emissions dropped down to the levels of the early 1990s, despite a doubling in the U.S. economy.”¹⁵³

Third, along with using less fossil fuel and changing the ones we use, another strategy is to offset or capture emissions. Planting trees and reclaiming land reduces the concentration of CO₂, as does trapping emissions underground or converting them into chemicals or plastics.¹⁵⁴

B. Pollution from Fossil Fuel

Along with climate change, energy policy also needs to account for pollution. It is well understood that fossil fuel is dirty to extract, transport, and burn, and that these costs are not always reflected in the market price.

1. Polluting Air, Water, and Soil

Extracting fossil fuel can damage the air, water, and land, harming human health, disrupting local economies, and disturbing animal habitats. Perhaps the most extreme example was the accident at Deepwater Horizon, an offshore oil rig, that released 130 million gallons of oil into the Gulf of Mexico in 2010. Tragically, it took eighty-seven days to stop the oil from flowing. During those long weeks, the spill caused $17.2 billion of environmental damage to animals, beaches, coral, fish, and marshes.¹⁵⁵ The spill killed millions of marine mammals, sea turtles, birds, and fish,¹⁵⁶ while also causing lasting health problems among workers who cleaned up the spill.¹⁵⁷

Extracting fossil fuel can cause pollution in more mundane ways as well. Coal mining causes miners to contract black lung disease and other health problems,¹⁵⁸ while also polluting streams and disfiguring landscapes. For example, “[m]ountaintop removal, a particularly destructive form of surface mining, involves stripping all trees and other vegetation from peaks and hilltops,” the Union of Concerned Scientists has explained, “and then blasting away hundreds of feet of the earth below with explosives.”¹⁵⁹ Likewise, extracting oil and gas also can cause pollution. For example, wastewater from hydraulic fracturing can contaminate water or induce seismic activity if not disposed of properly.¹⁶⁰

Transporting fossil fuels can also cause pollution, for instance, when pipelines leak or there are accidents involving tankers, barges, trains, and trucks. In March of 1989, for example, the Exxon Valdez, an oil supertanker, ran aground in Prince William Sound, releasing 11 million gallons of oil.¹⁶¹ While the ship’s hungover captain slept, his third mate missed a turn. This careless mistake dealt a devastating blow to local wildlife, with some effects lasting for years.¹⁶²

Likewise, a train carrying oil exploded in the small Canadian town of Lac-Mégantic in July of 2013, killing forty-seven people, destroying over forty buildings, and releasing millions of gallons of oil into the soil and the nearby Chaudière River. Sadly, another seven trains carrying oil derailed in Canada between 2013 and 2020.¹⁶³

Even if there are no mishaps in extracting or transporting fossil fuel, burning it is a familiar source of pollution. For example, coal-fired power plants and factories cause smog and acid rain, which can affect air quality thousands of miles away.¹⁶⁴ Auto exhaust also degrades air quality.¹⁶⁵ Indeed, air pollution from fossil fuel harms human health in a range of ways.¹⁶⁶

2. Reducing Pollution from Fossil Fuel

How can energy policy reduce pollution from fossil fuel? Although the literature on this topic is vast, and the details are beyond this Article’s scope, it is worth emphasizing that the three responses to climate change, noted above, also reduce pollution.

First, using less fossil fuel generally reduces the pollution it causes. Again, energy efficiency and clean energy can help, although some types of clean energy have pollution risks of their own (such as the radioactive waste from nuclear power).¹⁶⁷

Second, some types of fossil fuel are dirtier than others. Again, coal is the worst of them. It produces the most pollution, and its pollutants are most harmful to human health¹⁶⁸ At the same time, some types of coal—and, indeed, some types of mining—are worse than others.¹⁶⁹ As a result, replacing coal with natural gas reduces pollution, as well as emissions.

Third, when fossil fuels are used, there are ways to keep pollutants from being released. For instance, thick well casings prevent fracking fluid from seeping into drinking water when oil is extracted, while tankers with two hulls prevent oil spills when oil is transported.¹⁷⁰ Likewise, catalytic converters in cars and scrubbers in power plants contain some pollutants when fossil fuel is burned.¹⁷¹

C. National Security and the Environment: Synergies and Tensions

So far, this Article has showed that the U.S. and its allies need to pursue a range of goals in energy policy. Along with encouraging firms to generate and transport energy efficiently and reliably—goals that a competitive market usually is well suited to advance—policymakers also need to address four externalities, which require government intervention: first, the cost of protecting insecure suppliers; second, the cost of funding hostile suppliers; third, the cost of climate change; and fourth, the cost of pollution.

As Parts I and II showed, the key to addressing the first two externalities—and, thus, to protecting national security—is to reduce demand for fossil fuel, while also tapping new supply in secure and friendly countries. Yet what effect does this two-part strategy have on the environment?

In principle, the first strategy—reducing demand—has the potential to advance environmental goals. After all, using less fossil fuel can reduce emissions and pollution. But in fact, reducing demand is not always a “win-win” for national security and the environment. For example, even as clean energy eases dependence on problematic fossil fuel suppliers (such as Russia), it increases dependence on problematic clean energy suppliers (such as China). Part IV analyzes various options to reduce demand for fossil fuel, highlighting synergies and tensions between national security and environmental goals.

What about the second part of the two-part strategy? Is it feasible to tap new sources of fossil fuel while also protecting the environment? At first blush, these goals seem to conflict. Instead of drilling new wells (to protect national security), aren’t we supposed to phase out fossil fuel (to protect the environment)? But in fact, this tension can be resolved with the right policies, which tap new sources of fossil fuel while still reducing emissions and pollution. The key is for the new sources to be lower-carbon fossil fuels (for example, natural gas instead of coal) and for them to replace, instead of adding to, existing sources. Part V considers a range of strategies to tap new supplies of fossil fuel, identifying synergies and tensions between national security and environmental goals.

D. Authoritarian Comparative Advantage

The rest of this Article shows how nuanced policy judgments, with careful attention to the relevant tradeoffs, can deliver gains for both national security and the environment. But before turning to this challenge of policy, it is important to highlight a challenge of political economy that complicates efforts to adopt better policies: compared with authoritarian regimes, democracies are at a disadvantage in producing and transporting energy.

In a nutshell, the problem is interest group opposition. When democratic governments are asked to approve new wells, pipelines, or other fossil fuel infrastructure, there is almost always opposition from environmental groups, local residents and businesses (motivated by “not in my backyard” or “NIMBY” concerns), regulators who protect culturally significant sites, and economic competitors (such as coal companies, which regularly oppose natural gas pipelines).¹⁷² In the U.S. and Europe, these coalitions have banned fracking in several jurisdictions, halted drilling in some places, and blocked pipelines and LNG terminals.

Ironically, similar dynamics also have thwarted clean energy projects, including nuclear power plants and wind and solar facilities.¹⁷³ For example, after a failed effort to secure approval, which took sixteen years and cost $100 million, a clean energy company gave up on installing wind turbines off the coast of Cape Cod.¹⁷⁴ “The project unfortunately demonstrated,” observed a Massachusetts regulator who supported the project, “that well-funded opposition groups can effectively use the American court system to stop even a project with no material adverse environmental impacts . . . .”¹⁷⁵

In contrast, this sort of interest group pressure gains much less traction in authoritarian regimes. For instance, environmental campaigns in Russia often “butt up against political realities,” observed a 2021 report by a U.S. think tank, “leading to the prosecution of activists and even physical threats and abuse toward . . . them by state institutions, often on behalf of a private company.”¹⁷⁶ Likewise, “it can be said that there is no green movement in Iran,” concluded a 2019 study by Iranian academics.¹⁷⁷ “Policy makers in Iran still don’t cooperate with ENGOs [environmental NGOs] and even newspapers consider them as marginal issues.”¹⁷⁸

As a result, fossil fuel production (and, indeed, some clean energy initiatives) gravitate to authoritarian countries. In a sense, their insulation from political pressure gives them an edge, which this Article calls “authoritarian comparative advantage.”

This is the mirror image of a more familiar idea, noted above, that extractive industries encourage authoritarianism (for example, by freeing governments from depending on citizens for tax revenue and labor).¹⁷⁹ The point here is that the causal link can run in the other direction as well: not only do extractive industries facilitate authoritarianism, but authoritarianism also can facilitate extractive industries.

To weaken the competition, some authoritarian leaders cynically encourage environmental opposition in democracies. For example, Vladimir Putin regularly warns western audiences of the risks of fracking. “Today’s technology of shale oil production and shale gas,” he said at a 2019 business conference, “are without any doubt . . . barbaric.”¹⁸⁰ He made the same point quite heatedly at an earlier conference. “I was going to ask him a normal question about diversifying your economy,” recalled energy expert Daniel Yergin, whose question prompted this outburst.¹⁸¹ “And I said ‘shale,’ and to be shouted at by him in front of 3,000 people [was] a really unpleasant experience.”¹⁸² Putin’s vehemence presumably stems not from concern for the planet, but from economics and geopolitics. “[S]hale was a challenge for Russia,” Yergin explained.¹⁸³

To protect Russia’s market share, Putin has allegedly funded groups opposing shale development in Europe, as Hillary Clinton and the Secretary General of NATO each have claimed.¹⁸⁴ “We were . . . up against phony environmental groups, and I’m a big environmentalist,” Clinton asserted, “but these were funded by the Russians to stand against any effort, ‘Oh that pipeline, that fracking, that whatever will be a problem for you,’ and a lot of the money supporting that message was coming from Russia.”¹⁸⁵

As Putin understands, when democracies abstain from energy production, authoritarian regimes fill the gap. Unfortunately, this makes the U.S. and its allies more dependent on authoritarian suppliers, which often are insecure or hostile.¹⁸⁶ It would be far better to rely on production in democracies, which usually are more secure and friendly.¹⁸⁷ But as this Section has showed, this is an uphill climb politically.

Even so, this climb must be attempted. After all, relying on authoritarian suppliers is problematic not just for national security, but also for the environment. Who is more vigilant in regulating emissions and pollution? Do we trust Russia and Iran more than the U.S. and the EU? Admittedly, when environmental harms are localized, democracies can deflect these costs to the citizens of authoritarian countries—in effect, a form of global nimbyism. But two of the most important risks—climate change and air pollution—are global, not local. So instead of simply farming out fossil fuel development (and other energy initiatives) to authoritarian regimes, democracies should rely more on domestic production.

More specifically, the U.S. and its allies should pursue a two-part strategy that protects both national security and the environment: first, they should reduce the demand for fossil fuel; second, they should tap new supplies of fossil fuel in environmentally responsible ways. The next two Parts consider these strategies in turn.

IV. REDUCING DEMAND FOR FOSSIL FUELS: SYNERGIES AND TENSIONS

Let us begin with reducing demand. This Part shows how efforts to use less fossil fuel have the potential to be a “win-win,” protecting both national security and the environment. Yet the devil is in the details. Does an initiative actually reduce demand, once all the relevant fuel consumption is considered? If it does, are there offsetting costs, such as new risks to national security or the environment? Some strategies to reduce fossil fuel demand are better than others. Finding and implementing the right ones is critical.

At the same time, reducing demand should not be our exclusive focus. Since the transition away from fossil fuel will take years, the U.S. and its allies also need to find new sources that are secure, friendly, and can be tapped in environmentally responsible ways.

A. Potential to be a “Win-Win”

In general, the demand for fossil fuel can be cut in two ways: energy efficiency, and wider use of clean energy. Both are promising and should be pursued vigorously.

1. Energy Efficiency

A key step is to change the habits of consumers. For example, instead of driving to work, they should walk, carpool, take mass transit, or work from home. The right policies can encourage these shifts, including congestion pricing, bike lanes, cheaper mass transit fares, lower speed limits, and the like.

Technological innovations—and policies that encourage them—also can enhance energy efficiency. For example, switching from incandescent to more efficient LED bulbs—as the Biden Administration mandated in 2022—should reduce U.S. emissions by 222 million metric tons over thirty years and save nearly $3 billion annually in electricity costs.¹⁸⁸ Likewise, changing the idle power settings on computers also can “save $3 billion a year . . . and reduce CO₂ emissions by 20 million metric tons,” Kit Kennedy has explained, “without any impact on computer performance. . . .”¹⁸⁹ The same is true of better heating and cooling systems. For example, “heat pumps” use 50% less energy because they do not actually generate heat; instead, they extract it from the air.¹⁹⁰

Energy efficiency has obvious national security advantages, as German Vice-Chancellor Robert Habeck emphasized a few weeks after Russia invaded Ukraine. “If you can take the train or bike . . . , that’s good,” he said.¹⁹¹ “[I]t’s easy on the wallet and annoys Putin.”¹⁹² Obviously, there are parallel environmental advantages as well.

2. Clean Energy

Along with energy efficiency, another way to use less fossil fuel is to rely more on clean energy, including wind, solar, geothermal, nuclear, hydroelectric, and hydrogen. Fortunately, the cost of wind and solar has declined significantly in recent years, making them increasingly competitive even without subsidies.¹⁹³

So far, clean energy has been used mainly to generate electricity. This has been an effective way to burn less fossil fuel, since electric power plants are responsible for about 38% of all energy generated in the U.S.¹⁹⁴ The progress so far has been significant: in 2021, only 60% of electricity in the U.S. came from fossil fuels, compared with 21% from nuclear and 19% from renewable energy.¹⁹⁵

Another 37% of energy in the U.S. is used for transportation.¹⁹⁶ For decades, virtually all of this energy has come from fossil fuel—and, specifically, from petroleum.¹⁹⁷ Yet EVs can break petroleum’s monopoly, since the electricity powering them can come from clean energy (or, for that matter, from coal or natural gas). Fortunately, the performance and range of EVs has improved significantly. There also is a growing network of charging stations,¹⁹⁸ although this effort has a long way to go.¹⁹⁹

Like energy efficiency, clean energy has the potential to offer national security and environmental advantages. Again, using less fossil fuel not only reduces emissions and pollution, but also eases dependence on the wrong fossil fuel suppliers.

B. Are We Really Using Less Fossil Fuel?

Even so, before policymakers conclude that a policy or technology really is a “win-win,” they need to dig deeper. A key question is how much, if at all, it actually reduces the demand for fossil fuel.

1. Rebound

For example, a fuel-efficient car is supposed to use less fuel. But what if drivers respond by putting more miles on the car, since each additional mile is cheaper? Similarly, what if homeowners with heat pumps turn up the thermostat? If energy efficient products are used more, they do not save as much energy. This “rebound effect,” as it is called, reduces the national security and environmental advantages of energy efficient technology, since fossil fuel consumption declines less than expected.²⁰⁰

2. Life Cycle Calculations

Similarly, to determine how much fuel a new technology actually saves, we need to know how much is used not just in operating it, but also in manufacturing and powering it. For example, compared with a gasoline-powered car, less energy is needed to run an EV, but more is required to manufacture it because extra energy is needed to build the battery.²⁰¹ EVs make up for this disadvantage by using less energy when driving.²⁰² But how much less depends on how the electricity was generated. If it comes from solar or wind, the EV cuts fossil fuel demand more than if it comes from coal.²⁰³ This sort of “life cycle” analysis is needed to determine how effective a new technology is in cutting demand for fossil fuel, and thus in reducing emissions, pollution, and the national security risks from depending on the wrong fossil fuel suppliers.

C. New National Security Risks: “Just When I Thought I Was Out . . .”

Using less fossil fuel elicits another challenge as well: the transition to clean energy poses national security risks of its own. So as much as the U.S. and it allies would like to stop protecting insecure suppliers and funding adversaries, they will not necessarily get their wish. This Section shows that clean energy imposes parallel burdens. One is reminded of a famous line from The Godfather: “Just when I thought I was out, they pull me back in.”²⁰⁴

1. Defense Externalities: The Electrical Grid

For one thing, the U.S. and its allies become even more vulnerable to attacks on power plants, power lines, and other infrastructure. Since electricity is the most effective way to harness clean energy, using more of it means depending more on this grid.

There already are familiar risks from relying on the grid. Blackouts disrupt communications, finance, business, law enforcement, health care, and the delivery of water, food, and other essential goods and services.²⁰⁵ To avoid these hardships, the grid must be protected from extreme weather and natural disasters, as well as from cyber and physical attacks²⁰⁶—a lesson emphasized, sadly, by Russia’s repeated attacks on Ukraine’s grid in 2022.²⁰⁷

Yet these risks are bounded today because electricity is not the only game in town. It provides only 38% of the energy consumed in the U.S.²⁰⁸ Petroleum and other fossil fuels are the main sources for transportation,²⁰⁹ heating, and industrial processes.²¹⁰ Admittedly, the infrastructure for these fuels is also vulnerable. Pipelines can be hacked²¹¹ or sabotaged,²¹² refineries can be damaged in fires²¹³ or natural disasters,²¹⁴ oil depots can be attacked,²¹⁵ and the like. Yet the fact that this infrastructure is separate from the grid—and, for that matter, scattered across the country—offers useful diversification. If some pipelines, refineries, and gas stations go offline, others still function. Likewise, if the grid fails today, most homes will still be heated and most cars will still work.²¹⁶ But if all homes are heated with electricity and all cars are EVs, this will no longer be true.

In short, tapping clean energy means depending more on the grid. As a result, its security—and, more generally, its effectiveness—become even more essential.

2. Defense Externalities from Nuclear Power

The grid’s vulnerability is an example of a broader point: although fossil fuels have national security costs, so do other energy sources, and the risks from these new sources need to be addressed.

This is certainly true of nuclear power. On the one hand, it has national security advantages in easing dependence on problematic fossil fuel suppliers. For example, France has relied less on Russian natural gas because 70% of its electricity comes from nuclear plants.²¹⁷ In contrast, Germany has been in a weaker position because it started phasing out nuclear power in 2011.²¹⁸

On the other hand, nuclear power requires uranium. While some uranium suppliers are secure and friendly (such as Canada, Australia, and India),²¹⁹ others are not. For example, Kazakhstan (the world’s largest producer) shares a border with Russia, as do Uzbekistan and Ukraine.²²⁰ Meanwhile, Russia itself is a “top ten” producer, as is China.²²¹

Along with dependence on uranium, another risk is the security of the reactor itself. During a war, its core could be breached by missiles or artillery, or staff responsible for safety protocols could be incapacitated or driven away. Unfortunately, these risks became all too real in 2022 when Russian troops captured the Zaporizhzhya nuclear power plant in Ukraine.²²² Reactors also need protection from terrorist attacks, including truck bombs, plane crashes, and attempts to trigger a meltdown. The waste from reactors also must be secured, so terrorists cannot build dirty bombs. After the attacks on September 11, 2001, security at U.S. nuclear facilities was upgraded to address these threats.²²³

3. Defense Externalities: Clean Energy Raw Materials

Like nuclear power, other types of clean energy also ease some national security burdens, while creating others. A key challenge is the need for specialty minerals, such as cobalt and lithium.²²⁴ More are needed for solar panels, wind turbines, electricity networks, and EVs than for fossil fuel systems. “A typical electric car requires six times the mineral inputs of a conventional car,” the IEA has observed, “and an onshore wind plant requires nine times more mineral resources than a gas-fired power plant.”²²⁵ To scale up clean energy, the global economy will need far more of these minerals—four times more in 2040 to meet the goals of the Paris Accords and six times more to hit net-zero globally by 2050.²²⁶

Yet to source these minerals, the U.S. and its allies rely heavily on imports.²²⁷ Most come from only a handful of suppliers, since production “is more concentrated than that of oil or natural gas.”²²⁸ Unfortunately, “[m]uch of it comes from countries that are not our friends,” as Mark Mills has observed.²²⁹

For example, more than two-thirds of the world’s cobalt comes from the Democratic Republic of Congo,²³⁰ which has been wracked by factional violence and protests in recent years.²³¹ Likewise, the U.S. and its allies depend heavily on Gabon for manganese, as well as on Chile and Argentina for lithium.²³² China is the main supplier of rare earths,²³³ the global leader in processing other clean energy minerals,²³⁴ and a determined buyer of mines all over the world.²³⁵

Depending on other countries for these minerals poses familiar national security risks. Like with oil, the U.S. and its allies will feel pressure to defend insecure suppliers in a crisis, while also funding hostile suppliers’ harmful policies. “New geopolitics around the minerals for net zero may well emerge,” S&P Global has warned, “which will echo the geopolitics that have long surrounded oil and natural gas.”²³⁶

Fortunately, there are ways to reduce these national security costs. For one thing, the U.S. and its allies should ramp up efforts to recycle minerals, so newly-mined sources are less critical. In addition, they should try to replace minerals from an insecure supplier with alternatives that are easier to access (for example, by building batteries with lithium iron phosphates instead of cobalt).²³⁷

Likewise, the U.S. and its allies should encourage domestic mining. The good news is that they actually have deposits of most of the relevant minerals. But the bad news is that opening new mines is quite challenging, especially in democracies. Many projects whither on the vine, while successes usually require a decade or more of planning, negotiation, and construction to complete.²³⁸ “[D]isruptions from labor strikes, protests, environmental activism, domestic political rivalries, governmental shifts, and contractual disputes and renegotiations . . . delay projects and investment,” S&P Global has warned.²³⁹ “Brownfield and greenfield development of new projects turn on the complex interaction of permitting and policy, contracts and politics, and businesses and civil society . . . .”²⁴⁰ Unfortunately, some environmental groups are adamantly opposed to mining, even for minerals needed to reduce emissions.²⁴¹ Like when they oppose fossil fuel projects, these activists will find allies among local residents and economic competitors. Again, we see the authoritarian comparative advantage in extractive industries.²⁴²

Seeking to change this dynamic, the Inflation Reduction Act offered substantial subsidies for domestic mining and recycling. For example, half of the EV tax credit is available only if enough of the minerals in the battery were extracted and processed in the U.S. (or in a country with a free trade agreement with the U.S.) or were recycled in North America.²⁴³

Yet even with these subsidies, domestic mining and recycling cannot proceed without permits. Although Congress considered a permitting reform bill in 2022, it was not enacted.²⁴⁴ “[T]he United States is still in some early days of scrutinizing the existing web of federal and local mining permits and addressing NIMBY-ism,” Jane Nakano observed, “[so] a domestic supply chain is years away at best.”²⁴⁵

4. Funding Externalities: Clean Energy Manufacturing

The U.S. and its allies depend on other countries not just for raw materials, but also for the finished product. The main concern is China, a formidable geopolitical rival, ²⁴⁶ which has become the world’s leading manufacturer of clean energy technology.²⁴⁷

China has seven of the world’s top ten solar manufacturers,²⁴⁸ and produces over 75 percent of the world’s EV batteries.²⁴⁹ While the U.S. and its allies are less dependent on China for wind turbines, this could change. For land-based turbines, the U.S. is a leading manufacturer (though it faces stiff competition).²⁵⁰ For offshore turbines, however, the U.S. relies on European suppliers,²⁵¹ but China is ramping up, “buil[ding] more offshore wind turbines in 2021 than every other country did in the past five years.”²⁵²

Unfortunately, depending on China for clean energy could pose the same three risks, discussed above, as depending on rivals for fossil fuel. First, these purchases could fund policies that undermine U.S. interests, such as China’s efforts to control the South China Sea and Taiwan. Second, clean energy exports could give China leverage (for example, to stop exporting to countries that oppose its geopolitical agenda). Thus, just as Russia has weaponized natural gas while invading Ukraine, China might do the same with EV batteries and solar panels while attacking Taiwan. Third, China’s exports could also fund repressive policies at home, such as the use of forced labor to produce solar panels.²⁵³

Admittedly, even without clean energy, the U.S. and its allies already depend on China for other important products, ranging from semiconductors and cell phones to surgical masks. Even so, adding clean energy to this list is still significant—not just because the list grows longer—but also because energy is so fundamental in (literally) powering a modern economy.

While this sort of economic interdependence is not always bad—indeed, it can moderate potentially hostile regimes, as noted above—these benefits do not always materialize, as Germany learned in buying fossil fuel from Russia. Is China a safer bet? Although President Xi’s assertive foreign policy is not reassuring in this regard,²⁵⁴ the relationship between China and the West is complex. Over time, it could become either more confrontational or more cooperative. The goal here is not to offer a definitive prediction, but to highlight a meaningful risk.

Again, like with fossil fuel, the best way to mitigate funding externalities is to depend more on production in the U.S., Europe, and other friendly countries. In fact, Congress tried to encourage this sort of “friend-shoring” in the Inflation Reduction Act of 2022. Along with providing billions of dollars in subsidies directly for clean energy manufacturing in the U.S., Congress set conditions on other subsidies, so they would be available only for products with supply chains in the U.S. or other treaty partners.²⁵⁵

This is not to say that the Inflation Reduction Act has executed this policy flawlessly. For one thing, the definitions of which countries count as “friend-shoring” vary by provision,²⁵⁶ and have prompted complaints from U.S. allies.²⁵⁷

In addition, the Inflation Reduction Act also tries to “pick winners,” choosing which technologies get larger subsidies, which get smaller ones, and which get none at all. As I have emphasized elsewhere, government institutions often lack the expertise and incentives to make this sort of judgment effectively.²⁵⁸ As Part VI emphasizes below, a better strategy is to rely on policies that do not depend on the government to make these judgments, such as Pigouvian taxes.²⁵⁹

To sum up, clean energy can mitigate one national security risk (depending on the wrong fossil fuel suppliers) while creating another (relying on the wrong clean energy suppliers). To address this new risk, the key is to “friend-shore” the relevant minerals and manufacturing (and, of course, to do this the right way).

D. New Environmental Harms

Just as clean energy has national security costs, as well as benefits, the same is true of its environmental effects. While the benefits (such as reducing emissions and pollution) are very significant, they are not free.

1. Avian Impacts of Wind and Solar

For one thing, wind turbines and solar panels can harm birds and bats. These risks are regularly invoked to block clean energy projects.²⁶⁰ Although regrettable, these species impacts should not keep the U.S. and its allies from using renewable energy. As Professor Michael Gerrard has explained, we face a choice between saving “birds now or birds later.”²⁶¹ Halting these projects may save some birds now, but “won’t some of the animals we are trying to protect be gone anyway [because of climate change], together with untold numbers of others?”²⁶² Faced with this tradeoff, policymakers should target the greater threat, which is climate change.²⁶³ Unfortunately, some environmentalists resist this logic. “Rather than climate denial, the environmental community has tradeoff denial,” Professor Gerrard has observed.”²⁶⁴

2. Mining for Clean Energy Minerals

Clean energy has another set of environmental costs as well: mining for the necessary minerals can cause pollution, water shortages, accidents, and disruption of local communities and habitats.²⁶⁵ Invoking these risks, some environmental groups oppose mining for these minerals in the U.S. and E.U.²⁶⁶

Again, this is a mistake. As with the impact on birds, pollution from these mines is an unavoidable cost of combatting climate change. “[W]e need to be in an era of triage,” Michael Gerrard has urged, “where we save what we can but recognize that there are things we’ll have to give up.”²⁶⁷ Moreover, if these minerals are going to be extracted somewhere, isn’t it better to do it in jurisdictions with meaningful environmental regulation? The U.S. and E.U. are likely to be more responsible than many current suppliers.

3. Pollution and Accidents From Nuclear Energy

Like wind and solar power, nuclear power also poses environmental tradeoffs. On the one hand, the advantages are quite significant. No greenhouse gas is emitted²⁶⁸ and, unlike with wind and solar, output does not vary with the weather.²⁶⁹

On the other hand, nuclear power poses two familiar risks, which prompt staunch opposition from some environmentalists. First, there is radioactive waste, which must be disposed of safely. Second, there also is a risk of accidents. Fortunately, these have been rare. In the U.S., the most significant one, a partial meltdown at the Three Mile Island plant in 1979, had only negligible effects on public health and the environment.²⁷⁰ In contrast, the meltdowns at Fukushima in Japan in 2011 and at Chernobyl in the Soviet Union in 1986 were quite serious, causing disease and death from radiation exposure and also rendering some areas uninhabitable for decades.²⁷¹ Yet these tragedies were the product of unique circumstances: a tsunami in Japan, and a blend of human error and dysfunctional efforts to conceal the incident in the Soviet Union.²⁷²

With the right safety features and regulation, the risks from nuclear power should be quite limited. For example, new reactors have passive cooling systems that work even if power has been knocked out—the problem at Fukushima—as well as “core catchers” that contain radiation more effectively.²⁷³

Arguably, then, the risks of not using nuclear power are greater than the risks of using it. For example, after Germany began phasing out nuclear power as a response to Fukushima, its economy became even more dependent on Russian natural gas. This choice turned out badly not just for national security, but also for the environment: when the gas stopped flowing, Germany ramped up its use of coal, spewing more emissions and pollution into the air.

E. Timing: A Gradual Transition

Although replacing fossil fuel with clean energy has the potential to be a “win-win”—protecting both national security and the environment—a number of new national security and environmental risks must be addressed. In addition, there is another daunting challenge: for now, the world’s supply of clean energy is not even close to adequate. In 2021, renewables generated only 7% of the world’s energy. Paired with nuclear and hydroelectric power, the non-carbon total rose to just 18%. As Table 1 shows, fossil fuel still provided 82% of the world’s energy:

Table 1. Share of Global Energy By Source in 2021
Fuel	Amount (Exajoules)	Percentage
Oil	184.21	31%
Natural Gas	145.35	24%
Coal	160.10	27%
Nuclear	25.31	4%
Hydroelectric	40.26	7%
Renewables	39.91	7%
Total	595.15	100%
Notes: This data comes from BP Statistical Review of World Energy 9 (71st ed. 2022).

The market share of clean energy can increase, to be sure, but this progress inevitably will be gradual. Consumers do not replace cars and heating systems all that frequently. Even if they did, there are not enough EVs, heat pumps, solar panels, and wind turbines to accommodate everyone at once. As of now, we do not have enough factories to build them—or, for that matter, enough raw materials.

For example, replacing all petroleum-powered cars on the road today would require 1.39 billion EVs, whose batteries would need massive quantities of lithium, cobalt, and other minerals.²⁷⁴ But according to the Geological Survey of Finland, “[p]reliminary calculations show that global reserves, let alone global production, may not be enough to resource the quantity of batteries required.”²⁷⁵

There is a similar challenge with copper. Since it is essential for power infrastructure, renewable generation, and EVs, global demand is expected to double by 2035.²⁷⁶ According to S&P Global, the global supply cannot grow fast enough to meet the goal of net-zero global emissions by 2050.²⁷⁷ “Unless massive new supply comes online in a timely way,” they warn, “the goal of Net-Zero Emissions by 2050 will be short-circuited and remain out of reach.”²⁷⁸

The bottom line, then, is that the world has no realistic choice but to keep depending on fossil fuel for years to come. So although it is important to reduce demand for fossil fuel over time—the first part in this Article’s two-part proposal—it also is necessary to implement the second part: tapping new sources of supply.

V. TAPPING NEW SOURCES OF FOSSIL FUEL IN ENVIRONMENTALLY RESPONSIBLE WAYS

As Parts I and II showed, new sources of fossil fuel enhance national security when they come from the U.S. and other secure and friendly countries, and thus ease dependence on insecure or hostile suppliers. Yet notwithstanding these advantages of new sources, the IEA and other influential voices have called for an end to fossil fuel development.²⁷⁹ At times, the Biden Administration has also gestured in this direction, although at other times it has supported more drilling in response to rising energy prices, legal constraints, geopolitical imperatives, and political concerns.²⁸⁰

It is naïve—and, ultimately, misguided—to end fossil fuel development in the near term. On the contrary, to protect national security, the U.S. and its allies need to keep adding new wells and infrastructure in the right countries. Even so, this needs to be done in an environmentally responsible way. How can the U.S. and its allies tap new fossil fuel sources while still reducing emissions and pollution? This Part proposes three strategies to do both at once: first, new sources should be as “clean” as possible; second, they should replace, instead of adding to, sources that pose national security risks; and third, new sources should be temporary instead of permanent.

A. Increase Carbon Efficiency of Secure and Friendly Sources

As emphasized above, a key national security goal is to use less oil and gas from Russia, Venezuela, and Iran, and more from countries like the U.S., Brazil, Mexico, and Canada. If these friendly and secure sources also offer environmental advantages, developing them advances both national security and environmental goals. To pursue this “win-win” scenario, policymakers should look for ways to reduce the carbon and pollution footprints of these sources. How can we get more energy from them, while generating the same levels of emissions and pollution—or, ideally, reducing these levels?

1. Flaring

For one thing, we should get more energy from fuel we already burn. Unfortunately, massive amounts of natural gas are burned (or “flared”) at the wellhead. No one uses this energy, but it still produces significant emissions and pollution.²⁸¹

This means a great deal of energy is wasted: about eight percent of global natural gas production (accounting for six percent of global emissions).²⁸² “If half of the amount of gas flared annually [across the globe] was used for power generation,” Zubin Bamji observed, “it could provide about 400 billion kilowatt-hours of electricity – that’s roughly the annual electricity consumption of Sub-Saharan Africa.”²⁸³ In the U.S. alone, gas worth $10.6 billion was flared between 2012 and 2020.²⁸⁴

Why is so much natural gas wasted? Unfortunately, there is no infrastructure to bring it to market. This gas comes from oil wells, which have infrastructure to transport oil, but not gas.²⁸⁵ When gas cannot be delivered to consumers, the easiest alternatives are either to burn it or—even worse—to release it into the atmosphere.²⁸⁶

But instead of wasting this gas, we should find ways to use it—and, thus, to increase the supply of energy without increasing emissions (since this gas will be burned anyway). One option is to build pipelines to take it to market.²⁸⁷ Indeed, flaring is less common in Texas than in North Dakota because there are more pipelines.²⁸⁸ Where pipelines are not economical, facilities to use this gas can be added near the well, including small-scale generators, “micro” compression and liquefaction facilities, and petrochemical plants.²⁸⁹ If these solutions are not viable, the gas can be stored underground.²⁹⁰

2. Methane Leaks

Natural gas also is wasted when it leaks from wells and pipelines. Like flaring, these leaks increase emissions without generating useful energy,²⁹¹ so plugging them should be a priority. EPA proposed new rules on leaks in November of 2021, as well as supplemental rules a year later.²⁹² The Inflation Reduction Act also introduced a charge on methane leaks in some circumstances.²⁹³ Analyzing the details of these measures is beyond this Article’s scope. The goal here is not to determine whether they are the best ways to target leaks, but to emphasize the importance of addressing this issue.

3. Carbon Capture, Utilization, and Storage (“CCUS”)

Still another way to reduce environmental harms from fossil fuel is to capture and store CO₂, so it is not released into the atmosphere. For example, emissions from power plants can be piped to old oil and gas wells.²⁹⁴ “Carbon capture, utilisation and storage (CCUS) so far has not lived up to its promise,” the IEA has observed, but “[s]tronger climate targets and investment incentives are injecting new momentum into CCUS.”²⁹⁵ The U.S. tax code offers a tax credit for carbon capture, which the Inflation Reduction Act made more generous.²⁹⁶ Again, the details of this credit are beyond this Article’s scope.

4. Replace Coal With Natural Gas

Along with reducing emissions and pollution from specific types of fuel, policymakers also should change the mix of fuel. Specifically, a determined effort is needed to replace coal with natural gas.

Since the U.S. has ample reserves of both, they offer similar national security advantages.²⁹⁷ Yet burning coal produces nearly twice as many emissions as burning natural gas, as well as more pollution.²⁹⁸ Admittedly, natural gas poses the additional risk of methane leaks, as noted above.²⁹⁹ But as long as this problem is addressed, replacing coal with natural gas reduces emissions and pollution.³⁰⁰

Indeed, this switch has helped U.S. emissions decline substantially in recent years, as noted above.³⁰¹ Yet there is a lot of room for improvement, since coal still accounts for about 27% of the world’s energy,³⁰² as well as 11% of all energy used in the U.S.³⁰³

As a result, U.S. exports of natural gas do double duty. Not only do they enhance national security (by replacing Russian gas), but they also protect the environment (by replacing coal). For the same reasons, bringing natural gas to Europe from the Eastern Mediterranean is also good for both national security and the environment, so it is unfortunate that the Biden Administration has impeded this effort, at least initially, as noted above.³⁰⁴

B. Extra Supply Should Replace, Instead of Adding to, Existing Sources

As the previous Section showed, when the U.S. and its allies tap new sources of fossil fuel, they should favor cleaner ones. This Section adds a second environmental safeguard: in tapping new sources, the goal should be to stabilize—not increase—global supply.

Fortunately, increasing supply should not be necessary. Rather, to protect national security, the key is to fill a gap. If one supplier suddenly stops selling (because of an invasion or revolution) or should not be allowed to sell (because its revenue funds harmful conduct), another supplier needs to step in—not to add to global supply, but to replace the insecure or hostile source.

This reality reduces the tension with environmental goals. As long as new sources just fill a gap in the market, emissions and pollution should not increase. Overall, the same quantity of fossil fuel is burned; it just comes from different countries. Indeed, if friendly and secure sources are cleaner than the oil and gas they replace, as the previous Section recommended, global emissions and pollution would decline.

To be clear, this happens only if the insecure or hostile supplier’s oil and gas actually come off the market. This is quite likely for insecure sources, whose production is disrupted by an invasion or revolution. But with a hostile exporter, exiting the market is not automatic. If they want to keep selling, boycotts or other sanctions are needed to stop them. The key question, noted above, is whether these sanctions are effective.³⁰⁵ If not, adding new wells would increase global supply, instead of just stabilizing it, and thus could generate additional emissions and pollution.

To avoid this problem, new sources ideally should be elastic. They should increase production as insecure or hostile producers cut back, while reducing production as these other producers ramp up.

Is this feasible? The good news is that, at least to an extent, it happens automatically in response to market prices. On the one hand, if insecure or hostile producers cut production, prices rise, motivating secure and friendly suppliers to increase production. On the other hand, if insecure or hostile producers maintain their production, prices do not rise (or, at least, they revert after an initial panic). As a result, there is no market-based incentive to produce more (and, obviously, policymakers do not need to intervene with other incentives).

Yet, although prices provide some protection from oversupply (and the extra emissions and pollution it triggers), they are not a sure-fire solution. Suppliers sometimes respond slowly to changes in price, expecting prices to be volatile and waiting to see whether a trend endures.³⁰⁶ In addition, some suppliers are able to adjust more quickly than others, as emphasized above.³⁰⁷ When prices rise, how fast can they bring more product to market? When prices fall, how rapidly can they cut production? Producers vary widely in this regard; for example, Saudi Arabia is fast, U.S. offshore wells are slow, and so forth.³⁰⁸

The bottom line, then, is that tapping new sources of fossil fuel does not increase emissions or pollution if this new supply stabilizes (but does not increase) global supply. The same amount of fossil fuel is used, but it comes from different countries. This outcome is more likely when the new suppliers are flexible, so they can respond more quickly to market prices (and, therefore, to underlying shifts in global supply).

C. In Adding New Capacity, Retain Flexibility to Make It Temporary

This brings us to a third way to minimize environmental harm from tapping new sources of oil and gas: ideally, this extra supply should be reversible, producing only as long as it is needed. This Section shows how U.S. shale can offer this flexibility and explains how policymakers can leverage it.

1. U.S. Shale Has the Potential to Ramp Up Quickly

For one thing, shale producers can increase production fairly quickly, enabling them to replace suppliers that have become unavailable. Shale producers are not as fast as Saudi Arabia—which, as noted above, often can bring additional barrels to market within thirty days—but they are quicker than most other producers.³⁰⁹ In general, shale producers take between six and twelve months to react to price changes.³¹⁰ So when prices spike, these suppliers have the ability (and usually also the incentive) to increase production.

“Shale oil developments require relatively low amounts of initial capital and can be developed in relatively short order, making shale oil highly sensitive to price increases,” Bordoff, Halff and Losz have explained.³¹¹ “In contrast with the rest of the industry, which is highly concentrated, the shale oil industry is fragmented and made up of myriad small companies that are nimble, dynamic, innovative, and responsive to market changes.”³¹²

Admittedly, before a well is drilled—in a shale formation or anywhere else—years may be required to find the right location and secure permits. But this does not slow down well-run energy companies. They constantly explore and lease new sites and secure permits, sometimes years before they ever intend to drill there. As a result, these firms have an inventory of sites already approved. “The idea is that if there are delays with permitting or other land issues,” one analyst explained, “that the management team will have flexibility in deciding where they want to drill and not run short on options.”³¹³

Obviously, they still have to drill the well, but drilling in shale is a lot faster than drilling offshore.³¹⁴ To shorten the time even further, companies can drill the well, cap it, and then complete the process later when they actually need the oil or gas.³¹⁵

2. U.S. Shale Also Can Slow Production Quickly

Not only can U.S. shale producers ramp up quickly to fill a gap in the market, but they also can ramp down when extra production is no longer needed. As a result, shale wells are well positioned to replace, instead of supplementing, other production.

A more traditional well, whether drilled onshore or offshore, usually lasts twenty years or more.³¹⁶ As a result, it has more potential to become a stranded asset, which keeps pumping even after it is no longer needed.

In contrast, shale wells have much shorter useful lives. Hydraulic fracturing usually enables them to pump only for a year or two.³¹⁷ To keep producing, the firm needs another round of hydraulic fracturing or a new well. If demand has declined—so prices have fallen—the firm will not have the incentive to make this additional investment. “Unlike other types of oil projects, shale oil production declines steeply after initial production,” Bordoff, Halff and Losz have explained.³¹⁸ “Another distinctive quality of shale projects is their relatively high ongoing costs, which makes shale production sensitive to price declines as well.”³¹⁹

These differences give shale a significant edge over conventional wells, such as the Willow Project in Alaska, which the Biden Administration green-lighted in March 2023.³²⁰ Unlike shale wells, Willow is expected to have a thirty-year useful life.³²¹ Will the world still need Willow’s oil in thirty years? With a fast enough transition to clean energy proceeds, the answer could well be “no.” Arguably, then, approving Willow and other long-lived conventional projects is a mistake. Instead, the better approach is to rely more on shale to increase U.S. production.

In short, the geology of shale wells makes them much less likely to become stranded assets. They can ramp up to fill gaps in the market, and then slow production when there is excess supply.

3. Reversible Infrastructure

While shale producers can get oil and gas out of the ground quickly, it still needs to be refined and brought to market. This requires an elaborate infrastructure, including pipelines, refineries, and LNG terminals. Unlike shale wells, infrastructure take years to build.

i. Infrastructure Approvals: A Key Lever for the Government

A key question is whether this infrastructure is already in place. The answer varies with the location of the drilling, as well as the type of fossil fuel. In places where the infrastructure is adequate, shale producers already have the potential to serve as swing producers. But in other places, costly infrastructure investments are still needed, especially for natural gas, to take full advantage of their rapid reaction time. For example, the U.S. needs more LNG terminals to do more to replace Russian natural gas in Europe. Yet although a number of new export terminals have been built and more are under construction, the Biden Administration decided in January 2024 to “pause” the issuance of permits for new projects.³²²

In these efforts to ramp up production, the division of labor between the private sector and the government should be emphasized. In the U.S., private firms have significant discretion about how many wells to drill. In contrast, the government plays a critical role in major infrastructure projects. Under current law, the permitting process is as protracted as it is significant; a project cannot proceed unless the relevant agencies sign off. Deciding whether to do so is one of the main ways the government can either encourage or slow U.S. production.

In making these judgments, policymakers should balance the various considerations highlighted in this Article. How important is it to replace Russia? Or to have greater potential to replace Middle Eastern suppliers? Is the new source of oil and gas cleaner or dirtier than other sources? Would it really replace these other sources, or just supplement them?

ii. Investing in Infrastructure to Leverage the Flexibility of Shale

While these are context-specific judgments, the recommendation here is to err on the side of building more infrastructure. We should build it, even if we will not always use it.

This approach enables the U.S. and its allies to leverage the flexibility of shale production. On the one hand, when more fossil fuel is needed to replace insecure or hostile suppliers, the infrastructure is there to bring it to market. On the other hand, when there is too much supply—so shale producers are ramping down—the infrastructure does not need to be fully utilized.

Admittedly, this approach raises two potential concerns. First, once this infrastructure is built, we might keep using it longer than we should. For instance, what if we hope to stop using fossil fuel in fifteen years, but a pipeline has a thirty-year useful life? If the pipeline is built, there will be a temptation to use it for thirty years, instead of just fifteen.³²³

The solution to this problem is for infrastructure approvals to be contingent. In licenses and permits, the government should reserve the right to shut the infrastructure down before the end of its useful life. In the above example, the permit could allow the government to close the pipeline after fifteen years. This provides flexibility to respond to changed circumstances. In fifteen years, if the primary concern is climate and pollution, policymakers can shutter the pipeline. But if national security looms especially large, it can remain in use. Put another way, it is better to make permits contingent than to refuse to issue them at all—as the Biden Administration has done in “pausing” new LNG export permits—so the infrastructure is there when it is needed.

Admittedly, even if a contingent permit is granted, this does not mean that the project will proceed. This brings us to the second issue: limits on the use of infrastructure reduce its value to the private sector. The risk of an early shut down may keep projects from being built, even if they are urgently needed now. For example, the U.S. should build LNG terminals to be able to support its European allies and weaken Russia. But what if the risk of an early shut down discourages the private sector from building them?

The answer is for the government to help fund some projects. This is not meant as a handout to the fossil fuel industry, but as a response to the national security and environmental externalities highlighted in this Article. In some cases, investing in an LNG facility is more cost-effective than supplying military aid to Russia’s neighbors. Likewise, shutting down this facility when it is no longer needed may well be cheaper than building seawalls or repairing damage from storms. In short, the market failures discussed in this Article justify a government role, along with government expenditures.

“[G]overnments,” Bordoff and O’Sullivan have observed, “could develop innovative tools to plan for obsolescence.”³²⁴ For example, one approach would be for the government to pay compensation (for example, the infrastructure’s appraised value) in shutting down infrastructure after a period of time (for example, fifteen years). Another would be for the government to cover a share of the cost up front in exchange for acquiring ownership (or the right to shut the infrastructure down) after a preset term of years.³²⁵

In deciding which infrastructure projects to approve, policymakers should also consider whether they can be retooled for another purpose. An advantage of natural gas pipelines, for instance, is that they might someday transport hydrogen, a potential source of clean energy.³²⁶

Is investing in pipelines and other infrastructure the best use of public money? Obviously, rigorous judgments are needed about how this investment stacks up against other ways to address national security and environmental risks. But in some cases, reversible infrastructure is likely to be the most cost-effective way to pursue these goals.

4. So Why Did U.S. Shale Production Not Increase More Rapidly After Russia Invaded Ukraine?

With the right infrastructure, U.S. shale producers have the potential to avert (or at least dampen) supply shocks. Yet admittedly, they did not play this role after Russia invaded Ukraine in 2022. Even as prices spiked, they were slow to increase production.³²⁷

Even so, there were context-specific reasons for their hesitation, which will not necessarily recur. For one thing, the industry had just weathered the coronavirus pandemic, which slashed global demand, requiring steep production cuts, and plunged a number of producers into bankruptcy. This bruising experience made firms cautious about ramping up quickly.³²⁸

For firms that were willing to increase production, there was another barrier: the pandemic caused a host of supply chain bottlenecks. Like in other industries, shale producers struggled to get enough equipment and employees to increase production.³²⁹

Economic losses during the pandemic also burned investors. For years, they had provided capital even though shale producers were not (yet) profitable, prioritizing production increases and accepting that profits would come eventually. But the pandemic changed Wall Street’s attitude. After a wave of losses and bankruptcies, earnings—not increased production—became the priority.³³⁰ Executive compensation was adjusted to reflect this shift,³³¹ and the inventory of wells declined.³³² Yet the good news is that shale producers delivered record profits in 2022, which were turbocharged by a surge in oil and gas prices. Hopefully, these profits will ease investor concerns about expansion going forward.

Even so, there is still another barrier to overcome—one rooted in policy and perception, rather than in market dynamics. Before Russia invaded Ukraine, the Biden Administration’s rhetoric and policies sent a clear signal that fossil fuel production should decline.

“I want you to just take a look . . . I want you to look in my eyes,” Joe Biden said as a presidential candidate.³³³ “I guarantee you, I guarantee you we are going to end fossil fuel and I am not going to cooperate with them, OK?”³³⁴ In this spirit, he pledged to stop auctioning oil and gas leases on federal land. “And by the way,” he said, “no more drilling on federal lands, period. Period, period, period.”³³⁵ On his first day in office, President Biden canceled the Keystone Pipeline.³³⁶ Less than a week later, he imposed a “pause [on] new oil and natural gas leases on public lands or in offshore waters . . . .”³³⁷

Yet as energy prices started to rise—and then spiked after Russia invaded Ukraine—the Biden Administration began walking back this message. They resumed leasing federal land after a district court enjoined the “pause,”³³⁸ and agreed to allow more leases as a compromise to pass the Inflation Reduction Act.³³⁹ President Biden also began urging U.S. companies to increase production (while also criticizing them for profiting from higher prices).³⁴⁰

Even so, U.S. oil and gas producers were skeptical about the Administration’s shift in policy and rhetoric. “The Biden administration’s anti-fossil fuel policies and messaging have not helped the investment environment,” observed the CEO of an oil services firm.³⁴¹ “The White House may ask producers for more supply today, but their policy priorities seek to eliminate the need for that additional supply within five years.”³⁴² This pessimistic assessment was reinforced when the Biden Administration stopped issuing new LNG export permits early in 2024.

This chilling effect was unfortunate. Since the U.S. and its allies will rely on fossil fuel for years to come, discouraging new development comes at a cost.

To sum up, developing extra supply in the market for oil and gas has advantages for national security, but potential costs for the environment. Yet there are three ways to square this circle. First, policymakers should aim to make these new fossil fuel investments as “clean” as possible. Second, in adding new capacity, the goal should be to replace other fossil fuel sources, not to add to them. Third, the new sources should be flexible, so they can be ramped up and dialed back, as needed. In these ways, the U.S. and its allies can bring new oil and gas online while still reducing emissions and pollution.

VI. REGULATORY STRATEGY

As the last two Parts have shown, the U.S. and its allies need to reduce demand for fossil fuels, while also tapping new sources in environmentally responsible ways. This Part outlines a regulatory strategy to advance these goals. The best approach is a mix of Pigouvian taxes, targeting the various national security and environmental costs discussed in this Article. Unfortunately, this strategy has not gained any political traction in the U.S., at least so far.

As a fallback, some commentators (and, indeed, a number of celebrities) have called for a moratorium on new fossil fuel development. Yet this would be a mistake, as a moratorium would actually harm both national security and the environment.

Instead, the better approach is an incremental effort to alter the mix of energy sources over time. To guide this effort, this Part proposes a heuristic called “the marginal efficiency cost of energy”: policymakers should account for all the social costs of each source (for example, U.S. oil, Russian natural gas, U.S. coal, nuclear, and so forth)—not just private costs, but also national security and environmental costs—and then proceed step-by-step, looking for opportunities over time to replace high social cost sources with low social cost sources. To advance this agenda, policymakers can rely on whatever policy instruments are available, including permits, licenses, regulations, mandates, and subsidies.

A key challenge in implementing this agenda is regulatory fragmentation. A policymaker responsible solely for environmental risks will not have the incentives (and possibly also the expertise) to consider national security risks, and vice versa.

Lining up political support is also a challenge, but this Article’s approach—emphasizing both the environment and national security—could prove helpful. At the risk of dramatically oversimplifying U.S. politics, the environment tends to be more of a priority for the left, while national security tends to be more of a priority for the right. The key to bipartisan support could well be policies that advance both sets of goals. In other words, the right coalition could be both green and red, white, and blue.

A. Pigouvian Taxes

As I (and many others) have written elsewhere, arguably the best way to deal with negative externalities in energy is with Pigouvian taxes, which add these third-party costs to the price.³⁴³ This Section outlines the advantages of this regulatory approach, and briefly discusses how it can be used to target threats not only to the environment (which are well understood), but also to national security (which have received less attention). Yet since carbon taxes have attracted very little political support in the U.S., the discussion of Pigouvian taxes here is brief.

For the same reason, this Article does not offer a separate discussion of cap and trade. It is well understood that this regulatory strategy—which sets limits on an activity and issues tradable permits that authorize a designated level of it—offers similar benefits as Pigouvian taxes, so there is no need for a separate analysis here.

1. Efficiency of Pigouvian Taxes

Pricing externalities is a very efficient way to mitigate them. With a carbon tax, for example, if emissions from a gallon of gasoline cause fifty cents of harm to the climate, a tax of fifty cents per gallon is added to the price at the pump.³⁴⁴ A tax also is imposed on other sources of climate harms, including natural gas, coal, jet fuel, propane, livestock, chemicals, and so on. Since the harms from these various activities are not the same, a well-crafted carbon tax is calibrated to reflect these variations.

In implementing a Pigouvian tax, the regulator’s most important job is to estimate the externalities as accurately as possible—a responsibility that is difficult, to be sure, but also limited.³⁴⁵ The good news is that adding these costs to the price of goods and services fixes the market failure. Once regulators accomplish this, they can rely on the market to address the externality as efficiently as possible.

Instead of a “one-size-fits-all” approach, consumers have broad discretion to mitigate the relevant harm in whatever way is easiest for them. In response to a carbon tax, for instance, consumers can adjust their behavior in a host of ways: they can take mass transit, carpool, telecommute, move closer to work, get a car with a more fuel-efficient internal combustion engine, drive a hybrid, buy an EV, install solar panels on their roof, lower the thermostat in the winter, buy heat pumps and energy-efficient appliances, turn off the lights when they leave the room, use energy efficient bulbs, install better insulation, eat less meat, and much more. In dozens of choices every day, they can reduce their carbon footprint.³⁴⁶

Pigouvian taxes also offer similar flexibility to businesses. For example, by increasing gasoline prices, a carbon tax motivates auto manufacturers to prioritize fuel efficiency. Again, there are a host of ways to do this, including lighter materials, more efficient internal combustion engines, hybrids, EVs, and so on.³⁴⁷

With a subsidy, the government would have to pick which approaches to support—something the government usually lacks the incentives and expertise to do well.³⁴⁸ With a carbon tax, by contrast, the government does not have to make this sort of a judgment. Instead, the tax motivates businesses to respond to the problem. They compete for customers by experimenting with different approaches.³⁴⁹

2. The Perils of Picking Winners: EVs versus Hybrids

Sadly, the problems with “picking winners” were on full display in the Inflation Reduction Act. For example, it offers a generous subsidy for EVs, but no subsidy for hybrids, which have both a battery and a gas tank. At one level, this makes sense. EVs have a smaller carbon footprint, so the switch from a gasoline-powered car to an EV reduces emissions approximately twice as much as the switch to a hybrid.

But this analysis does not take account of an important downside of EVs: their batteries are a lot larger because, unlike hybrids, they can’t run on gasoline as a backup power source. So compared with a Toyota Camry hybrid, a Chevy Bolt’s battery is sixty times larger, and thus requires sixty times more lithium and other minerals. Hopefully, this differential won’t matter over time, as the global supply of the relevant minerals expands.

But for now, this supply is quite constrained. This means that the same quantity of scarce minerals can produce either one Chevy Bolt or sixty Toyota Camry hybrids. Although a Bolt is about twice as effective at reducing emissions as a hybrid on a one-for-one basis, the analysis is very different when one Bolt is compared—not to a single hybrid—but to sixty of them. Are emissions reduced more by replacing one gasoline-powered car with a Bolt, or sixty gasoline-powered cars with sixty Toyota Camry hybrids? The sixty hybrids reduce emissions twenty-nine times more than a single Bolt!³⁵⁰

In other words, once the analysis incorporates the scarcity of minerals—and thus the number of cars that actually can be produced—Congress’s decision to subsidize only hybrids, and not EVs, is questionable. The broader point, of course, is that Congress is not well positioned to pick one technology over another. Again, this is the great advantage of a carbon tax. It spares Congress from making these choices. After setting a price for emissions, Congress can rely on the market to develop the most cost effective ways to reduce them, such as hybrids in the short term and EVs in the long term.

3. A Menu of Pigouvian Taxes on Energy

Like carbon taxes, Pigouvian taxes on pollution and national security harms have the same advantages. For example, since coal causes more pollution than other fossil fuels, adding this cost to the price of coal motivates consumers and businesses to use less of it and favor cleaner alternatives.

To internalize the externalities discussed in this Article, four types of Pigouvian taxes are needed: first, a carbon tax; second, a tax on pollution; third, a tax to cover the cost of defending access to energy from insecure or unstable sources (including petroleum, specialized minerals used in clean energy, and uranium); and, finally, the cost of funding exporters that engage in aggressive or repressive conduct (including oil and gas from Russia and Iran, clean energy from China, and so forth).

4. Defense Externalities: A Tax on Oil

As an example of a tax on defense externalities, consider the case of oil. Should this tax apply to all oil, or only to barrels imported from insecure suppliers? In other words, should it be a version of a gasoline tax, or a tariff?

The argument for a broader tax, which would apply even to domestic production, is that oil is fungible. Using it exposes the U.S. to supply shocks, and the prospect of these shocks motivates the U.S. to defend insecure suppliers (even ones that do not sell oil in the U.S.). By taxing all oil, policymakers would reduce demand for oil overall, thereby mitigating these risks.

In contrast, the case for the narrower tax, which would apply only to imports from insecure suppliers (such as those in the Middle East), is that the U.S. incurs extra defense costs only to protect these suppliers, not suppliers in the U.S., Canada, and other secure countries. Favoring the latter (with either an exemption or a lower rate) would encourage more production in North America and other secure locations.

Notably, the U.S. could probably differentiate among these suppliers without violating its trade commitments. Under the General Agreement on Tariffs and Trade (“GATT”), countries have significant latitude to protect national security.³⁵¹

Even so, distinguishing among suppliers poses a number of administrability challenges. For example, how feasible would it be to trace the origin of crude oil?³⁵² What if crude from different sources is blended together?³⁵³

If these administrability issues can be addressed, the right approach may be to impose two taxes: one on all oil used in the U.S., regardless of where it is produced, and another on imports from insecure suppliers (so these imports are subject to both taxes). Yet a definitive analysis of this issue is beyond this Article’s scope.³⁵⁴

5. Funding Externalities: A Tariff on Russian Oil

A tax can be used to internalize the cost not only of defending insecure exporters (such as Kuwait), but also of funding threatening exporters (such as Russia). This sort of tax is supposed to reduce the revenue of rogue exporters, but this does not always happen. In some cases, the tax ends up hurting consumers, instead of the hostile exporter.³⁵⁵

The key question is, “who has more bargaining power?” To illustrate the difference, let us use a stylized example in which the global price of oil is $75 per barrel and the U.S. and E.U. impose a $25 per barrel tariff on Russian oil.

Let us start with the optimistic scenario. Assume that consumers have a lot of bargaining power (because, for example, they can either use less oil or buy it from other producers), while Russia cannot afford to cut production. In this case, Russia absorbs the tariff: the global price stays at $75, forcing Russia to cut its price to $50 to remain competitive. In this situation, the tariff does its job. It reduces Russia’s revenue—shifting Russia’s producer surplus to the taxing jurisdictions—so Russia has less money for the war in Ukraine.³⁵⁶

Unfortunately, this successful outcome is not inevitable. Instead, another possibility is that the tariff ends up hurting U.S. and E.U. consumers, without reducing Russia’s export revenue very much. This happens if Russia is the one with the bargaining power (for example, because it can afford to stop selling or can sell to other buyers, but consumers cannot cut their consumption or rely on other suppliers). In this case, consumers bear the economic burden of the tax.³⁵⁷ Russia raises its price to $100 (so it still gets $75 pre-tax per barrel), and the global price rises to $100. In this situation, the tariff does not reduce Russia’s revenue, at least by much. The modest advantage of this policy is that higher prices should reduce demand a bit in the short term—and, presumably, more over time as consumers find ways to adjust—so Russia sells fewer barrels.³⁵⁸

What does the evidence suggest about Russia? Instead of a tariff, the U.S. immediately stopped buying Russian oil, while the E.U. phased down its purchases more gradually. In response, Russia redirected its exports to China, India, and other countries that have not joined these boycotts. But as noted above, Russia has had to sell at a discount of approximately 25%. This lack of bargaining power suggests that if the U.S. and E.U. decided to impose a 25% tariff, instead of a boycott, Russia would be willing to keep selling at a 25% discount (as they already do in selling to China and India).³⁵⁹

So, what is the difference? Either way, Russia loses this 25%. The question is who gets it. With the tariff, it would go to the taxing countries (for example, the U.S. and members of the E.U.). With the embargo, it goes to buyers in China, India, and other countries that keep buying discounted Russian crude.

In December of 2022, the U.S. and its allies imposed another sanction: a cap on the price of Russian oil. Instead of a total ban on insuring and transporting Russian oil—a policy that was about to go into effect, and might have triggered a supply shock—the U.S. and its allies made an exception for Russian oil, as long as it was selling below $60 per barrel.³⁶⁰ Notably, Russian crude was already trading below this level because of the discount, as discussed above.³⁶¹ This price cap presumably gave India and China even more leverage to demand discounts, while also avoiding a supply shock by allowing Russia to keep selling crude. But eventually, Russia found ways to avoid this cap, for instance, by cobbling together its own fleet to ship oil (and overcharging on shipping as a way to make up for the discount).³⁶²

Although Russia was forced to accept discounts in selling oil, it has more bargaining power in selling natural gas. Because it is much harder to reroute, as noted above, Europe cannot easily replace Russian natural gas, at least in the short term. This means a tariff on Russian natural gas is likely to hurt E.U. consumers, not Russia, at least in the near term.³⁶³ Indeed, even without a tariff, Russia has dramatically cut its gas shipments to Europe, as noted above, causing prices to spike and pressuring Europe to ration natural gas.³⁶⁴

6. Political Constraints

While Pigouvian taxes have obvious advantages, which have prompted most of our allies to adopt carbon taxes,³⁶⁵ the political track record in the U.S. is discouraging. Indeed, few U.S. politicians have been willing even to propose carbon taxes.³⁶⁶ Instead, the Obama and Biden Administrations usually favored subsidies for green technology—a choice I have criticized elsewhere.³⁶⁷

Since a carbon tax still seems to be a political dead letter, a national security tax presumably also is a hard sell, at least in ordinary circumstances. Yet perhaps the idea could gain traction in a time of crisis.

For example, what if President George W. Bush had proposed a tax on petroleum (or on petroleum imports) in response to the terror attacks on September 11? Like Nixon going to China, a former oil executive like President Bush had added credibility in making this case. To rally support, he could have argued that the tax would weaken regimes that fund terrorism. Given the groundswell of support for a vigorous response to 9/11, one wonders whether a promising opportunity was missed.

A more recent crisis—Russia’s invasion of Ukraine—could also have justified a different national security tax: a tariff on Russian oil, like the one discussed above. The President already had statutory authority to impose this tariff.³⁶⁸ But instead, the Biden Administration initially opted to ban imports. A few weeks later, Treasury Secretary Janet Yellen floated the idea of a tariff with U.S. allies,³⁶⁹ but the focus quickly shifted to the price cap, discussed above.³⁷⁰ Maybe the concern was that a tariff might raise prices—and thus hurt U.S. and E.U. consumers, instead of Russia—but as noted above, this seems unlikely for oil.³⁷¹

In any event, the glaring absence of these proposals—even in times of crisis—is not an encouraging sign. Since the political prospects for Pigouvian taxes in the U.S. seem to be dim, at least for now, let us turn to potential alternatives.

B. Moratorium on New Exploration and Infrastructure: A Flawed Strategy

In principle, one option is a moratorium on new fossil fuel development and infrastructure. We may be stuck using existing wells and pipelines for many years, the logic goes, but let’s at least stop adding more.

This idea has gained significant traction. The International Energy Agency supports it.³⁷² The Biden Administration showed some sympathy for this approach early on, and returned to it in pausing LNG export permits, as noted above.³⁷³ Seattle and Vancouver have banned new fossil fuel infrastructure and development,³⁷⁴ while other state and local governments have taken more limited steps.³⁷⁵ A number of advocacy groups have also urged a moratorium. Calling for a “nonproliferation treaty” for fossil fuels, Fossil Fuel Treaty.org claims endorsements from 101 Nobel Laureates, 2900 scientists, hospitals representing over 100,000 doctors, 230 legislators from 60 countries, Hawaii’s state legislature, London’s City Council, the Foreign Minister of Tuvalu, and the Vatican.³⁷⁶ “Keep it in the ground” has drawn support from a number of prominent celebrities.³⁷⁷ The same drumbeat has been sounded also by Oil Watch,³⁷⁸ Clean Water Action,³⁷⁹ and “LINGO,” which is short for “Leave it in the ground.” “What is clear today is that looking for more fossil fuels needs to stop,” LINGO urged.³⁸⁰ “Allowing it to continue is like allowing a child to buy more sweets, when we already know its teeth are rotten and it has diabetes.”³⁸¹

Nevertheless, a moratorium on new development and infrastructure is a bad idea. The risks to national security are obvious. The world would have to depend solely on existing production and, as emphasized above, too many wells are in countries that either have to be defended or are themselves threats.³⁸² The decades-long useful life of these wells, moreover, is much longer than the typical two-year life of a well in U.S. shale. If new U.S. wells could not be drilled, shale production would fall dramatically, and the global economy would become even more dependent on the wrong producers.

Instead, the better course for national security, as emphasized above, is to rely increasingly on new production in the U.S. and other secure and friendly countries, while cutting back purchases from rogue exporters (for example, Russia and Iran) and insecure sources (for example, in the Middle East). This would not be possible with a moratorium.

Ironically, a moratorium also would harm the environment, locking us into a status quo that wastes energy and uses the wrong fossil fuels. As emphasized above, we need new pipelines to help end flaring.³⁸³ Likewise, we should keep replacing coal with natural gas, an effort that requires more natural gas wells and infrastructure, including more LNG export terminals.³⁸⁴ In short, changing the mix of the fossil fuels we use would help the environment, but a moratorium would stand in the way.

C. A Better Approach: Incremental Substitutions Based on the Marginal Efficiency Cost of Energy

Instead of a moratorium, a better strategy is to rely on incremental change. To vet these reforms, this part offers a heuristic called “the marginal efficiency cost of energy.” In essence, the idea is to consider all the social costs of energy—environmental and national security costs, along with private costs—and to hunt for ways to replace costlier sources with more efficient ones.

To be clear, the goal here is not to grant new power to regulators, but to help them make wiser use of the power they already have. They should use this framework in all the choices they are called upon to make, including decisions about permits, regulations, rates, leases, moratoriums, and subsidies. Whenever regulators make these judgments, they should compare alternative sources of energy, account for all their social costs, and favor the most (socially) efficient ones.

1. Parallel Problems: Tax and Energy

In offering this approach, this Article applies an idea from public finance, developed by Joel Slemrod and Shlomo Yitzhaki, called “the marginal efficiency cost of funds.”³⁸⁵ Notably, the problem they addressed—how to determine which marginal changes in the tax system improve efficiency³⁸⁶—resembles the challenge here in four important ways.

First, in each case, the goal is to figure out how to provide an additional unit of output at the lowest possible cost. In one case, the output is tax revenue, while in the other it is energy.

Second, in each case, there are several options for producing this additional output. Another dollar of tax can be collected with an income tax, wealth tax, value added tax, estate tax, carbon tax, or some other tax. For each type of tax, a range of adjustments can be considered, including in rates, audits, penalties, and particular rules. Like tax revenue, additional energy also can be generated in many ways. The next kilowatt hour can come from Russian natural gas, German coal, U.S. oil, solar panels from China, wind turbines from the U.S, or a host of other sources.

How do we know which option is most efficient? This brings us to the third parallel between tax and energy: each option has its own unique mix of costs, which often involve tradeoffs. In tax, there are administrative costs (such as when staffers write rules and auditors check returns), compliance costs (when accountants prepare returns), substitution effects (when taxpayers respond by working fewer hours or saving less), evasion costs (when taxpayers cheat), and avoidance costs (when taxpayers pursue legal tax minimization strategies). Likewise, in energy, there are different types of environmental and national security externalities, as well as private costs.

Fourth, making a change can increase some costs, while reducing others. For example, if Congress starts requiring foreign banks to share information about U.S. depositors, this change in the tax system increases compliance costs (as banks prepare these reports) and administrative costs (as the IRS reviews them), but (hopefully) reduces evasion costs (as taxpayers stop hiding money in offshore banks). Likewise, switching from German coal to Russian natural gas reduces environmental harms, while increasing national security risks.

2. The Answer in Tax Policy: Marginal Efficiency Cost of Funds

When there are a host of options, and each offers a unique mix of different costs, what should policymakers do? For one thing, they need to account for all the relevant costs. “[I]f an essential part of the problem is overlooked,” Slemrod and Yitzhaki observed, “partial models may give incorrect answers.”³⁸⁷

Policymakers then should strive to reduce the sum of these various costs, so they can collect a specified amount of revenue as efficiently as possible. The key is to figure out which features of the tax system are more costly, and to replace them with more efficient alternatives. “In reality, the MECF [marginal efficiency cost of funds] of different instruments can differ,” Slemrod and Yitzhaki showed, “and it is feasible to raise revenue utilizing only those policy instruments with a relatively low MECF.”³⁸⁸

For example, what if the same amount of revenue can be raised either by eliminating a deduction or by raising the tax rate? Policymakers should pick the one with the lowest total social costs, including administrative costs, compliance costs, and tax-motivated changes in taxpayer behavior. “One can calculate the MECF for alternative ways of raising revenue,” Slemrod and Yitzhaki explained, “and other things being equal, the one with the lowest MECF is the one that should be recommended.”³⁸⁹

3. The Answer in Energy Policy: Marginal Efficiency Cost of Energy

The same approach should be used in energy policy. Like another dollar of revenue, another kilowatt hour can be generated in various ways. What is the social cost of each alternative?

Like in tax policy, it is essential to account for all the costs. Again, the established practice among some commentators and government agencies to omit national security costs is simply wrong.³⁹⁰ Hopefully, the invasion of Ukraine in 2022—and the ensuing scramble to replace Russian oil and gas on short notice—has discredited this misguided approach.

Instead, policymakers should consider the five different types of costs emphasized in this Article: first, private costs (X); second, climate externalities (C); third, pollution externalities (P); fourth, defense externalities (D); and, fifth, funding externalities (F). (In accounting for all these costs, this heuristic seeks to replicate the effect of a menu of Pigouvian taxes, which was recommended above.)³⁹¹

The cost of producing another kilowatt hour from a specific source—that is, this source’s “marginal efficiency cost of energy” (“MECE”)—must include all of these costs. For example, assume that two sources of energy are available, A and B. To decide which to favor, policymakers should calculate the MECF of each one:

MECE_A = [X_A + C_A+ P_A + D_A + F_A] / kWh

MECE_B = [X_B + C_B+ P_B + D_B + F_B] / kWh

After comparing these two options, policymakers should favor the one with the lowest total social cost. So, if MECE_A> MECE_B, policymakers should favor B.

For example, if A is U.S. coal and B is U.S. natural gas, policymakers should replace coal with gas. More generally, within the set of sources with comparable national security impacts (such as energy produced in the U.S.), policymakers should favor ones with environmental advantages (for example, natural gas instead of coal). This is analogous to a Pareto improvement: one goal is advanced, without setting the other back.³⁹²

The same analysis holds if A is Russian natural gas and B is U.S. natural gas. Within a set of energy sources with comparable environmental impacts (for example, natural gas), policymakers should favor ones with national security advantages (gas produced in the U.S., instead of in Russia). Again, policymakers can advance one goal, without losing ground on the other.

Policymakers also can trade off environmental and national security benefits. Since the goal is to minimize the sum of the relevant costs, it usually makes sense to accept a modest increase in some costs in exchange for major reductions in others.³⁹³

D. Regulatory Expertise and Stability

To make these judgments effectively, policymakers need the right information, expertise, and incentives—but this is a tall order. Just understanding the relevant technology and markets is hard enough. Yet energy policy is even harder because of its implications for the environment and national security. Therefore, a truly interdisciplinary effort is needed. Wise decisions require a keen understanding not only of the relevant science, markets, and law, but also of defense strategy and foreign policy.

While the U.S. government as a whole has expertise on this diverse range of issues, these experts are not all in the same agency. On the one hand, EPA (and their counterparts at the state level) understand environmental challenges and the laws governing them. On the other hand, the Pentagon, State Department, and various intelligence agencies know the nuances of defense and foreign policy. Meanwhile, other institutions master the details of trade and industrial policy (for example, Treasury, Commerce, and Department of Energy (“DOE”)), oil and gas drilling (state and local agencies), approval of oil and gas exports (DOE and Federal Energy Regulatory Commission (“FERC”)), fuel economy standards and vehicle emissions (the National Highway Traffic Safety Administration, EPA, state regulators), the regulation of nuclear power (Nuclear Regulatory Commission), disposal of nuclear materials (DOE), the regulation of electricity (FERC and state public utility commissions), and the regulation of pipelines (FERC, Department of Transportation, and state agencies).

There is room to wonder whether this fragmented structure serves us well enough. Are these various regulators accounting for all the relevant costs? Are they valuing them the same way? For example, when regulators develop U.S. fuel economy standards, they should account not just for pollution and emissions, but also for the national security costs of defending access to petroleum. Yet unfortunately—and, indeed, somewhat unfathomably—they have omitted this important national security cost, as noted above.³⁹⁴ National security costs are not easy to value, to be sure. But like with the social cost of carbon, the best available estimate should be developed and periodically updated. The Office of Information and Regulatory Affairs (or some other body of experts) should ensure that the same estimate is used throughout the government.

The problem with our fragmented system of energy regulation is not only one of information and expertise, but also of decision-making authority. There are different ways to pursue our various policy goals, but no single agency has broad enough jurisdiction to compare them all and pick the best one. For example, as this Article has emphasized, one way to counter Russian influence is with diplomacy, covert capabilities, and military force. Another is to “starve the beast” by weaning Europe off Russian energy, whether with the right fossil fuel infrastructure (for example, to deliver U.S. natural gas) or with the wider use of alternative energy and energy efficient technology. Yet these various alternatives each fall under the jurisdiction of a different cluster of government institutions. If some are more promising than others, does anyone actually have the authority—and, for that matter, the incentives—to compare all the relevant options and pick the best ones?

As if this were not hard enough, still another challenge is worth emphasizing. Energy policy goals cannot be achieved overnight. They require sustained effort and investment over the course of years, or even decades. This means that a measure of stability is needed in U.S. policy.

But unfortunately, there have been wild gyrations from one administration to the next. For example, nurturing renewable energy was a high priority under President Obama, a lower priority under President Trump, and a high priority again under President Biden. Likewise, tapping domestic oil and gas was a high priority under President Trump, but not under President Biden, at least initially, as noted above.³⁹⁵ Unfortunately, mixed signals and constant changes in priorities come at a cost; without certainty, the private sector is less likely to invest, experiment, and innovate.

To sum up, two things should be clear. First, whoever is responsible for energy policy needs to have a broad enough mandate to consider all the relevant issues. Second, there should be a measure of policy stability from one administration to the next, so long-term goals can be pursued effectively.

The good news is that there is an institution that checks both of these boxes and, of course, is charged with these responsibilities under the constitution: the U.S. Congress. Although individual committees have specialized mandates, Congress as a whole has more general jurisdiction, so members are supposed to see “the big picture.” Their decisions also have unique legitimacy, since they answer directly to the people.

Admittedly, there is a familiar challenge in relying on Congress: political deadlock often prevents it from acting. But in a way, this weakness is also a strength: once legislation is enacted, it is quite hard to repeal, so a measure of stability is assured even as the White House changes hands. As a result, congressional action on these issues is especially valuable.

To administer the relevant statutes, Congress should consider consolidating more responsibilities under a single energy regulator with broad jurisdiction. On the one hand, if the priority is political accountability, the model could be a cabinet-level department like Homeland Security. On the other hand, if the priority is independence and policy stability, the model could be an independent agency like the Federal Reserve. For instance, just as the Federal Reserve has a dual mandate to target both inflation and unemployment, this energy regulator could have a triple mandate to (1) assure that the supply of energy is cheap and reliable; (2) strengthen national security; and (3) protect the environment.

In any event, an analysis of the right institutional division of labor is beyond this Article’s scope. The goal here is to flag these issues, not to resolve them. After all, designing the right structure for crafting energy policy—one that accesses all the relevant information, creates the right incentives, and accords with constitutional norms—is a complex task. It warrants hundreds of pages of analysis, not just a few paragraphs.

E. Political Economy: A “Red, White, and Blue—and Green” Coalition

The same is true of the political dynamics driving energy policy—another issue that is beyond this Article’s scope, but still is critically important. After all, in reflecting on the recommendations here, one might easily say, “This is all fine in theory, but could any of this ever actually happen in our polarized political environment?”

While Congress did pass climate legislation in 2022, it used reconciliation instead of its regular process so the Vice President could cast the deciding vote in the Senate. Doesn’t this suggest that the prospects for more robust legislation are dim?

Not necessarily. For one thing, the legislation made it past the finish line because Joe Manchin, the deciding vote in the Senate at the time, insisted that the bill should include support for both clean energy and fossil fuels. This is not to say that the relevant provisions were the right ones. Rather, the point is that there could be a coalition—even a bipartisan one—for efforts to promote clean energy, while also encouraging environmentally responsible fossil fuel development, as this Article has urged.

How does this sort of effort, which pursues multiple goals at once, help to attract political support? A cynic would observe that it appeals to more interest groups, and thus may draw a measure of support from both environmental groups and fossil fuel producers.

But there is another political advantage as well: national security has broad political appeal, especially in times of crisis. Invoking this goal allows legislation to resonate not only with voters who are passionate about the environment, but also with voters who want to thwart terrorism, block the global ambitions of America’s adversaries, and support our troops. As emphasized above, these are not necessarily the same voters. As a result, energy policy that is grounded in both the environment and national security is likely to attract a broader coalition.

CONCLUSION

This Article has shown that energy policy must consider risks not only to the environment, but also to national security. It is important to account for the costs of securing access to energy (defense externalities) and of funding exporters that engage in harmful conduct (funding externalities), even though a number of commentators have argued over the years that these costs should not be considered.

This Article has offered guidance about how energy policy can protect both national security and the environment. The key goal for national security is to depend less on insecure or hostile suppliers. To do so, while also reducing emissions and pollution, policymakers need to pursue a two-part agenda: they should reduce the demand for fossil fuels, while also tapping new sources of supply in environmentally responsible ways. Pigouvian taxes would be an effective way to implement this agenda. Alternatively, policymakers could use the heuristic proposed in this Article, the marginal efficiency cost of energy, to replace high (social) cost energy sources with more efficient alternatives.

Generating the requisite political support will require compromise, as well as an alliance between supporters of the environment, on one hand, and national security, on the other. Ultimately, the policy goals, as well as the political coalition supporting them, need to be red, white, and blue—and also green.

96 S. Cal. L. Rev. 1157

Download

David M. Schizer*

Dean Emeritus & Harvey R. Miller Professor of Law and Economics, Columbia Law School. Helpful comments were received from Jason Bordoff, Mike Gerrard, Miryl Hilibrand, Erica Hur, Tom Merrill, Kira Patterson, Alex Raskolnikov, Joel Slemrod, Matt Waxman, Dan Yergin, and participants at the annual conference of the American Law and Economics Association, the N.Y. State Bar Association Tax Section’s Annual Meeting, as well as at workshops at Columbia Law School, Hebrew University, IAC-Edge, the Institute for Law and Strategy, the University of British Columbia, the World Law Congress, the Center on Capitalism and Society at Columbia University, and the Louis D. Brandeis Law Society.

The Court’s Morality Play: The Punishment Lens, Sex, and Abortion

March 2024March 2024June Carbone* & Naomi Cahn†

This Article uncovers the hidden framework for the Supreme Court’s approach to public values, a framework that has shaped—and will continue to shape—the abortion debate. The Court has historically used a “punishment lens” to allow the evolution of moral expression in the public square, without enmeshing the Court itself in the underlying values debate. The punishment lens allows a court to redirect attention by focusing on the penalty rather than the potentially inflammatory subject for which the penalty is being imposed, regardless of whether the subject is contraception, abortion, Medicaid expansion, or pretrial detention.

This Article is unique in discussing the circumstances in which the Court has simultaneously concluded that the state could regulate but could not punish, even if that means redefining a sanction as not punitive. By making visible this framework, we offer the Court and the states a potential off-ramp from the continuation of an ugly and litigious future on abortion access. If the Supreme Court seeks to deflect the outrage over Dobbs, the simplest way to do so would be to take seriously the statement that all it has to do is to return the issue to the states. In that case, the Court’s focus should be, as Justice Kavanaugh suggested in his concurrence, on the impermissibility of punishment that infringes on established rights, independent of a right to abortion, such as the right to travel, the First Amendment right to communicate accurate information about abortion availability, or doctors’ efforts to perform therapeutic abortions necessary to preserve a pregnant person’s health. The Court would not pass judgment on the permissibility of abortion, and it could affirm the propriety of state bans, but still strike down heavy-handed prosecutions and ill-defined prohibitions that impose undue penalties.

After Dobbs v. Jackson Women’s Health Organization, this Article is particularly important for three reasons. First, this Article examines the ways in which the Court has used considerations of punishment to deflect irreconcilable values clashes. Second, a focus on punishment often illuminates the “dark side” of government action, justifying limits on such actions. Third, a focus on “punishment” often illustrates the consequences of government actions, consequences that may be an indirect result of statutes or regulations but that have disproportionate effects on marginalized communities. Understanding how the Court has used this elusive concept in the past may thus help shape the response to Dobbs.

INTRODUCTION

The concept of punishment is central to the Supreme Court’s jurisprudence on abortion—and, beyond abortion, to the expression of moral values in the public square. In Dobbs v. Jackson Women’s Health Organization, Justice Alito found “an unbroken tradition of prohibiting abortion on pain of criminal punishment” throughout the common law until the Court’s decision in Roe v. Wade in 1973.¹ He noted that “the great common-law authorities—Bracton, Coke, Hale, and Blackstone—all wrote that a post-quickening abortion was a crime”² and he traced these developments from the thirteenth century forward.³

The Dobbs opinion, like most criminal law discussions, assumes that the power to prohibit includes the power to punish violations of those prohibitions. And, indeed, criminal law scholars have produced an extensive literature on the justifications for the imposition of criminal sanctions and the constitutional limits on that imposition.⁴

What neither that vast literature nor the Dobbs opinion addresses, however, is the role of punishment in the evolution of the jurisprudence addressing the expression of public values, separate and apart from the existence of the laws prohibiting conduct. As this Article shows, when the Supreme Court has focused on the state’s justification for punishment independently from the underlying policy, it has often used the nature of punishment as a justification for striking down legislation—even when the Court concedes that the state purpose is otherwise legitimate.⁵ And it sometimes uses the declaration that onerous provisions are not “penalt[ies]” to uphold coercive legislation that, as a practical matter, limits access to what the Court otherwise recognizes as important rights.⁶ As these cases show, outside of the narrow context of whether a criminal prohibition justifies the imposition of a particular sentence,⁷ punishment has an ill-defined life of its own in Supreme Court jurisprudence.

This Article is the first to detail how the Supreme Court has viewed the concept of “punishment” as a justification for upholding or invalidating government acts in the context of issues involving contested values.⁸ While an intense debate raged at mid-century over whether the state should regulate morality, that debate generally assumed that if the state could regulate, it could also punish.⁹ This Article is unique in discussing the circumstances in which the Court has simultaneously concluded that the state could regulate but could not punish. For example, the Court held that a state could discourage teen sex but not by encouraging pregnancy as the consequence¹⁰ or could adopt restrictive measures,¹¹ such as blanket refusals to fund medically necessary abortions, so long as the statute did not prohibit abortion or penalize those seeking one.¹²

This Article is particularly important following Dobbs for three reasons. First, it illustrates the ways in which the Court has used considerations of punishment to deflect irreconcilable values clashes. For those who would like to extricate the Court from the conflicts Dobbs has inflamed, limiting punishment, for example, of those exercising a constitutionally protected right to travel, offers a potential off ramp.

Second, a focus on punishment often illuminates the “dark side” of government action. The opinion in Griswold v. Connecticut placed great weight on the intrusiveness of policing the use of contraceptives in the marital bedroom.¹³ The ugliness of imposing punishment may similarly become a focal point for organization in response to the patchwork of state laws after Dobbs.¹⁴

Third, a focus on “punishment” is often used to illustrate the consequences of government actions, consequences that may be an indirect result of statutes or regulations but that have disproportionate effects on marginalized communities.¹⁵ Abortion bans may aggravate race and class-based differences,¹⁶ prompting greater recognition of the rights of the pregnant to obtain the medical care needed to safeguard their health.¹⁷ Understanding how the Court has used this elusive concept in the past can thus help shape the response to Dobbs.

The Supreme Court’s conception of “punishment” underlying these considerations is slippery, perhaps intentionally so.¹⁸ The Court uses the concept at both an expressive level, reinforcing public norms, and a practical level, specifying the consequences for the violation of government mandates, both civil and criminal. Most critically for this Article, it provides the Court with a way to shape emerging norms in the context of public unease.

After describing the multidisciplinary literature on punishment’s multiple roles, we examine the way that the Court has deployed punishment as a rationale for invalidating government action, particularly in the context of cases involving sexual morality. Eisenstadt v. Baird, which stuck down bans on the sale of contraceptives to single women, provides a classic case: the Court simultaneously “conceded” that “the State could . . . regard the problems of extramarital and premarital sexual relations as ‘(e)vils’ ”¹⁹ but still held that this could not be the purpose of the Massachusetts legislation because it “would be plainly unreasonable to assume that Massachusetts has prescribed pregnancy and the birth of an unwanted child as punishment for fornication . . . .”²⁰ The irrationality of the punishment, not the permissibility of unmarried sex, provided the basis for the decision—and implicitly for the limitation of state power to regulate sexual morality.

We then explore how the Court has used the determination of what is a punishment to affirm state decision-making power in a federal system. The question of when the state is inflicting a punishment as opposed to imposing a reasonable condition or proceeding on appropriate administrative grounds arises in contexts ranging from welfare “home visits” to detention to Medicaid expansion, with the Court using the punishment lens to sidestep the substantive bases for these decisions.²¹

Finally, this Article considers cases that directly engage the relationship between punishment and the underlying values debate. Lawrence v. Texas, which invalidated Texas’s same-sex sodomy statute, provides the most striking example. Justice Kennedy’s majority opinion did not just strike down the criminalization of the sexual conduct.²² It affirmed the dignity and worth of the expression of intimacy in the case. Justice Scalia’s dissent, by contrast, saw punishment as the point, with both the majority and dissent agreeing that the values debate was central to the discussion.²³

This Article observes that the “punishment lens” provides a powerful tool for shaping the evolution of public values without enmeshing the Court in the underlying values debate. We consider whether the punishment lens can be successful in two ways: guiding the evolution of public values without triggering a backlash that further entrenches polarized opposition²⁴ or, failing that, reaching decisions in controversial cases that do not undermine the Court’s own legitimacy and authority.²⁵ By this standard, Eisenstadt v. Baird, which used the punishment lens to avoid the underlying values questions while striking down barriers to contraceptive access, and Lawrence v. Texas, which instead of relying on the punishment lens directly engaged the values questions, both succeeded in resolving issues in ways that helped move public opinion and lock in legal conclusions that remain embedded in American law. Whether applying the punishment lens to abortion can enjoy similar success remains to be seen, but this Article concludes by outlining the possibilities a focus on punishment can offer.

I. PUNISHMENT AND THE RULE OF LAW

The role of state-administered punishment is much studied—and much contested. Existing literature addresses the questions of what might justify the ability of the state to inflict intentionally burdensome treatment on its citizens,²⁶ what purposes such punishment should serve,²⁷ and what constitutes appropriate punishment.²⁸ As this scholarship establishes, law enforcement—and punishment of egregious crimes—is essential to a state’s legitimacy. Without punishment of criminal acts, a state cannot govern—and command either the support of its constituents or deference from the international order.²⁹ This literature, however, in its most idealized form, tends to assume a straightforward relationship between crime and punishment: the state prohibits certain acts, imposes prescribed penalties for the violation of the law, and administers the penalties in accordance with principles of procedural and substantive justice, emphasizing due process rights for the accused and fairness defined in terms of proportionality between the crime and the punishment.³⁰

This Section goes beyond the conventional analysis of criminal punishment to explore the expressive role that punishment serves. It shows that the judicial oversight of punishment serves four roles that pose difficult challenges in the face of contested or changing values: establishing shared societal values, maintaining or dismantling social hierarchies, mediating disputes over the authority of governmental actors to impose punishment, and channeling the individual desire for vengeance into state-approved channels.

First, the administration of punishment defines and reinforces societal values, often in symbolic ways. For example, with the recognition that smoking caused cancer and other health risks,³¹ the perceived acceptability of smoking changed.³² In the United States, the state did not respond by prohibiting smoking. Instead, government entities gradually limited the places where smoking was permitted, first, creating “no smoking” areas and ultimately banning smoking in restaurants, offices, and other places.³³ Over time, enforcement of these rules—and the imposition of sanctions on violators—did not just shift norms of politeness; they expressed moral disapproval of smoking as undesirable and deviant.³⁴ In 1993, the Supreme Court of the United States embraced the shift in attitudes in a decision that held placing a nonsmoking prison inmate in a cell with a five-pack-a-day smoker could constitute constitutionally impermissible “cruel and unusual punishment.”³⁵ In so ruling, the Court did not limit the word “punishment” to the prescribed penalties for a criminal act.³⁶ Instead, its finding of “cruel and unusual punishment” reflected and reinforced the changed social meaning of smoking from an acceptable activity to one that violated evolving “standards of decency,”³⁷ and concluded that violating this new moral sensibility could constitute “punishment” within the meaning of the Constitution. The act of placing a nonsmoker with a smoker thus became punishment because of the changed moral status of smoking.

Second, legal scholars have argued that beyond merely maintaining order, much of the power of state-administered punishment comes from this expression of “moral condemnation”³⁸ and its role in establishing social hierarchies within a society. In accordance with this analysis, moral condemnation does not just declare particular conduct to be illegal; it establishes and reinforces social order and social standing in a society.³⁹ Criminal acts threaten to upend the social order, as the person committing the crime asserts the right to defy established law and norms. Imposing punishment that carries moral condemnation with it restores the moral order,⁴⁰ affirming the victim’s superior status to that of the violator.⁴¹ Punishment can thus signal that “the community values the victim”⁴² while the failure to punish can indicate indifference, or even disdain, toward the victim.⁴³ Accordingly, both imposing punishment and failing to punish send important messages about what a society values. State-administered punishment can thus establish and reinforce norms in ways that contribute to social cohesion,⁴⁴ cohesion operating at the group as well as the individual level.⁴⁵ Disturbingly, the imposition of punishment can breed cohesion even if there is no crime;⁴⁶ nonetheless, the state reaffirms its legitimacy and authority when it punishes in the name of a value or ideal, rather than simply because it can.⁴⁷

The role of punishment in establishing social hierarchies, particularly when it operates at a group-based level, contributes to the dark side of punishment.⁴⁸ Brain imaging studies⁴⁹ show that the act of punishing engages the part of the brain that produces feelings of reward—the same area of the brain involved in drug addiction.⁵⁰ Individuals may thus derive pleasure from imposing punishment on others even when imposing punishment makes the punisher worse off.⁵¹

This psychological dimension corresponds to some descriptions of the retributivist purpose of punishment. Nietzsche argues that cruelty—and the satisfaction some derive from it—is the point of punishment.⁵² Even Oliver Wendell Holmes agreed that at least in some cases, punishment “is inflicted for the very purpose of causing pain” and “one of its objects is to gratify the desire for vengeance.”⁵³ And the anger and moral outrage that fuels demand for punishment can be manipulated.⁵⁴ Research has tantalizingly suggested that the act of punishment itself reinforces perception of harm.⁵⁵ Cultural cognition studies further show that people associate behavior contrary to their moral norms with socially detrimental consequences.⁵⁶

Precisely because the administration of punishment reinforces social standing at both the individual and the group level, it has implications that go beyond the punishment administered to any particular individuals. The decisions about which punishments to implement (such as firing an employee who refuses to be vaccinated or imposing work requirements as a condition of eligibility for state subsidized health insurance benefits) can create group-based winners and losers, elevating the status of one group at the expense of another.⁵⁷ Yet, denying the legitimacy of such demands for punishment—or imposing them too harshly—can also undermine respect for law.⁵⁸

This leads to the third role of judicial oversight of punishment: mediating conflicts that involve the authority of different governmental actors to impose punishment. In the United States, for example, the Supreme Court has overseen evolving conflicts between the states and the federal government in the administration of family law. The U.S. Constitution has historically been viewed as entrusting family law to the states,⁵⁹ but the Supreme Court has selectively intervened, at times to enhance or restrain state authority to impose punishment. In Stanley v. Illinois,⁶⁰ for example, the Supreme Court held that Illinois could not treat Peter Stanley as an unfit parent⁶¹—and thus deprive him of standing to seek the custody of his children after their mother’s death—solely because he had not married the mother.⁶² The Court intervened to limit the power of the state to punish unmarried fathers, at a time when attitudes were changing toward unmarried relationships.

Finally, the courts have historically overseen punishment in order to channel vengeance into socially constructive venues. The failure to punish perceived wrongs may persuade wronged individuals or groups to “take the law into their own hands” or to impose punishments out of proportion to the wrongful act.⁶³ The courts, in contrast, are supposed to act “judiciously” in administering punishment in a neutral manner, not just on behalf of the wronged individual, but because the assertion of the moral values of the social order can contribute to a sense of social order and cohesion.⁶⁴

The challenge of serving these four roles increases as social norms change.⁶⁵ The tension between maintaining order and imposing destabilizing punishments is particularly difficult if some social groups reject the norms, while others respond to the increasing defiance of the first group by calling for greater punishment as violations increase.⁶⁶ The imposition of punishment thus involves an “ever-shifting relationship between a regime and a given population that makes up the most essential element in any political order.”⁶⁷

These four roles make the administration of punishment central to the rule of law. They are also evident as a longstanding aspect of Supreme Court jurisprudence. Yet, managing the tensions between these objectives can undermine as well as maintain social cohesion. Congress and various state legislatures, for example, have attempted to shift norms surrounding intimate relationships by changing the laws governing sexual assault to make date rape easier to prosecute.⁶⁸ Imposing more serious penalties, however, may make judges and juries more reluctant to convict—and failures to impose punishment can undermine, in turn, the efforts to shift norms and also lead victims to feel even more isolated and aggrieved.⁶⁹ Expressing moral condemnation while keeping punishments commensurate with the perceived seriousness of the offenses thus requires walking a tightrope, one that sways with changing public sensibilities. Abortion, perhaps as much as if not more than any other issue, involves “irreconcilable disagreement” that challenges the legitimacy of the judicial system itself. The issues of punishment in the abortion context will test whether the judiciary generally, and the Supreme Court in particular, retain any capacity for guiding the recreation of shared social values.⁷⁰

II. SEX AND PUNISHMENT: RECOGNIZING REPRODUCTIVE RIGHTS

At the time Roe v. Wade was decided in 1973, the Supreme Court was carefully navigating a revolution in sexual mores.⁷¹ Sexual morality presents a classic case for the expressive role of punishment,⁷² with punishment serving to reinforce what are seen as consensus-based moral values broadly shared by the public.⁷³ Enforcing such norms also involves, however, punishment of private consensual conduct.

This Part shows how the Supreme Court focused on the acceptability of punishment as a rationale for state action rather than on the changing norms themselves. It did so through a series of cases that addressed contraception, nonmarital children’s legitimacy status, welfare benefits, parentage—and ultimately abortion—though the lens of punishment for sexual conduct. Within this new jurisprudence, the Court carved out a right to privacy that did not address the propriety of intimate conduct, but rather evaluated the permissibility of state action designed to shape private conduct.

A. Contraception and the Propriety of Pregnancy as Punishment for Sex

Starting with Griswold v. Connecticut⁷⁴ in 1965, the Supreme Court began to strike down legislation that regulated sexuality in ways that the Court deemed needlessly punitive. In doing so, the Court never waged a frontal assault on the moral order that channeled sexuality into marriage.⁷⁵ Instead, the Court examined the rationales underlying the laws and the consequences of imposing punishment.

Griswold addressed the constitutionality of a law that forbade the use of contraception. Anthony Comstock had spearheaded prohibition of contraceptives in the nineteenth century, convinced that they “facilitate[d] immoral conduct” because they “reduce[d] the risk that individuals who engage[d] in premarital sex, extramarital sex, or prostitution [would] suffer the consequences of venereal disease or unwanted pregnancy.”⁷⁶ Comstock persuaded Congress to outlaw “print and pictorial erotica, contraceptives, abortifacients, information about contraception or abortion, sexual implements and toys, and advertisements” in 1873⁷⁷ and the states adopted their own “Little Comstock laws” thereafter.⁷⁸ Connecticut’s statute, adopted in 1879,⁷⁹ was one of the most restrictive, banning not just the advertising and sale of contraceptives, but also the use of contraception.⁸⁰

The Court framed the case as one against defendants who “gave information, instruction, and medical advice to married persons as to the means of preventing conception.”⁸¹ In resolving the matter, the Court conceptualized a right to privacy, a right that justified looking the other way at sexual conduct. The Court wrote that “[w]e deal with a right of privacy older than the Bill of Rights older than our political parties, older than our school system.” ⁸² The Court did not mention married couples’ efforts to limit the number of children they had directly, although it did refer to the marital relationship as “intimate to the degree of being sacred” and suggested that enforcing a ban on contraceptive use would have “a maximum destructive impact upon [the marital] relationship.”⁸³ By contrast, the Court acknowledged the validity of the state’s purported rationale for the regulation: “the discouraging of extra-marital relations.”⁸⁴ While the Court stated that this rationale “is admittedly a legitimate subject of state concern,”⁸⁵ banning contraceptive use by married couples was simply too far removed from the purported subject of the statute to pass constitutional muster.⁸⁶ The Court suggested that the state could regulate the manufacture or sale of contraceptives but not their use within marital unions.⁸⁷ In short, the Court focused on the ugliness of enforcement⁸⁸ rather than on the permissibility of the underlying conduct—the use of contraception.⁸⁹

Connecticut did not often enforce its ban on married couples’ contraceptive use, but the fact that the law was on the books effectively limited the ability to use contraception to those with access to doctors and pharmacists.⁹⁰ While the Griswold decision did not mention the issue,⁹¹ a major reason for challenging the ban on contraception was the unequal nature of contraceptive access.⁹² By striking down criminal penalties for contraceptive sales, the Court effectively allowed doctors and clinics to make contraception more broadly available. The implicit principle at the core of this decision was that, while the state could steer sexuality into marriage, it could no longer seek to ensure that pregnancy be the unavoidable consequence of sexual relationships.

Eisenstadt v. Baird,⁹³ decided in 1972, expanded the principle—that pregnancy was an unreasonable punishment—beyond marriage. Eisenstadt struck down a Massachusetts statute that prohibited supplying contraception to single, as opposed to married, individuals. As in its decision in Griswold, the Court “conceded” that “the State could, consistently with the Equal Protection Clause, regard the problems of extramarital and premarital sexual relations as ‘(e)vils.’ ”⁹⁴ Nonetheless, it concluded that this could not be the purpose of the Massachusetts legislation because it “would be plainly unreasonable to assume that Massachusetts has prescribed pregnancy and the birth of an unwanted child as punishment for fornication.”⁹⁵ The Court acknowledged, as it did in Griswold, that notwithstanding the law, contraceptives are widely available, and thus “the rationality of this justification is dubious.”⁹⁶

By 1977, the Supreme Court was willing to say that the state could not prescribe pregnancy as the punishment for sex even where the state had a clear interest in discouraging sex between minors. In striking down a state law that prohibited selling contraceptives to minors under the age of sixteen, the Court noted the state interest in regulating the “morality of minors” in its efforts to promote “the State’s policy against promiscuous sexual intercourse among the young.”⁹⁷ Again, however, the Court accepted the legitimacy of the state interest, but rejected the connection between such a state interest and the prohibition on sales of contraceptives to minors. The Court observed that, “with or without access to contraceptives, the incidence of sexual activity among minors is high, and the consequences of such activity are frequently devastating,” but observed that there was little evidence that banning contraception had much impact.⁹⁸ The Court thus concluded that the state could not promote an otherwise legitimate objective—discouraging “promiscuous sexual intercourse among the young”—by making pregnancy the punishment for sex and criminalizing efforts to avoid the consequences.⁹⁹ And it emphasized that the justification for banning contraceptive sales became that much weaker as the evidence mounted that the laws on the books did not have the desired effect. It thus concluded that the “punishment” (pregnancy) did not serve the interests of either deterrence (teens with still have sex) or an appropriate desert (a child) for a wrongful act.

B. Public Recognition and the Removal of the Scarlet Letter from Children

The Supreme Court relied on similar reasoning in dismantling the distinctions between “legitimate” and “illegitimate” children, with the Court ultimately concluding that the states could not seek to channel childbearing into marriage by punishing children for their parents’ conduct. In the “seminal” case of Levy v. Louisiana,¹⁰⁰ the Court considered a Louisiana law that restricted the ability to bring a tort action for the wrongful death of a parent to “legitimate children.”¹⁰¹ As a result, an unmarried mother’s five children, who lived with her, and whom she raised on her own earnings, had no right to sue for their mother’s allegedly wrongful death. The Court, in striking down the statute in a brief opinion, observed that the Court could imagine no reason “why, in terms of ‘equal protection,’ should the tortfeasors go free merely because the child is illegitimate?”¹⁰² The Court reasoned that the circumstances of the birth had “no relation to the nature of the wrong allegedly inflicted on the mother;” the children, “though illegitimate, were dependent on her.”¹⁰³ The Court even recounted how the mother in the Levy case supported her children by working as a domestic servant, “taking them to church every Sunday and enrolling them, at her own expense, in a parochial school.”¹⁰⁴ In this opinion, the Court identified no countervailing state interest; the children were deprived of the right to sue for the loss of their mother simply because of the circumstances of their birth.

The Supreme Court in Levy did not mention the issue of race, but amicus briefs filed in the case emphasized that, particularly in Louisiana, the distinctions between marital and nonmarital children had a significant racial impact. Indeed, an amicus brief filed by Illinois law professor Harry Krause (and others) argued explicitly that the statute “discriminates on the basis of race.”¹⁰⁵ The brief maintained that the discrimination stemmed partly from the fact that “disproportionately more Negro children than white children are born out of wedlock,” and, partly from the fact that “a high percentage (70%) of white illegitimate children are adopted . . . whereas very few (3-5%) Negro illegitimates find adoptive parents.”¹⁰⁶ As a result, “95.8 percent of all persons affected by discrimination against illegitimates under the statute are Negroes.”¹⁰⁷ The brief concluded, “the classification of illegitimacy . . . is a euphemism for discrimination against Negroes.”¹⁰⁸

Louisiana denied that it sought to punish the children for immorality in sexual behavior, but it nonetheless maintained that it sought to encourage marriage.¹⁰⁹ And the state asserted: “If the community grants almost as much respect for non-marriage as for marriage, illegitimacy increases” and that “illegitimate daughters tend to err in the manner of their illegitimate mothers, producing more illegitimate children.”¹¹⁰ In short, Louisiana did argue that it was necessary to punish the children to deter their parents, if not quite in so many words. And the children who would be punished as a result were overwhelmingly Black.¹¹¹ Louisiana’s efforts to punish nonmarital births thus reinforced a racial as well as moral line, though the majority opinion for the Court did not directly address the racial issue.

In subsequent cases, the Court made the role of punishment even more explicit. In 1972, the Court reaffirmed Levy in striking down a Louisiana statute that defined “child” so that only marital children were eligible for insurance benefits resulting from their father’s death.¹¹² Justice Powell’s majority opinion held that the “status of illegitimacy has expressed through the ages society’s condemnation of irresponsible liaisons beyond the bonds of marriage,” but still concluded that imposing “this condemnation on the head of an infant is illogical and unjust.”¹¹³ Powell concluded that the distinction between marital and nonmarital children was not justified by any state interest.¹¹⁴

In 1977, the Court revisited the issue of inheritance,¹¹⁵ invalidating an Illinois statute that permitted nonmarital children to inherit only from their mothers, not their fathers.¹¹⁶ In a 5-4 decision, Justice Powell reiterated that “visiting this condemnation on the head of an infant is illogical and unjust.”¹¹⁷ He emphasized that, while the parents’ behavior might have been immoral, that was not the fault—nor the responsibility—of the children. The opposition to the punishment of children commanded a majority of an even more conservative Court than the Warren Court that had initially struck down such classification.¹¹⁸

C. The Right to Abortion: Part I

The Supreme Court’s 1973 decision in Roe v. Wade situated the case within the punishment lens the Court had constructed to deal with reproductive rights more generally. The case never squarely fit there, however, because abortion did not just involve the regulation of sexual behavior between consenting partners; it also raised issues about the involvement of the medical profession and the status of the fetus. Nonetheless, the Court framed the decision as a right centered on the irrationality of the state prescription of childbirth as a way to prevent illicit sex and a jurisprudence conscious of the consequences, intended and unintended, of regulating sexual morality. It thus treated laws banning abortion as imposing punishment—on the pregnant for incurring an unwanted pregnancy, on doctors for exercising medical judgment in treating patients, and on those who felt compelled to seek illegal abortions in unsafe circumstances.

Among the telling aspects of this analysis is the way the Court articulated the state interests at stake. The Court identified the first such interest as one based on “a Victorian social concern to discourage illicit sexual conduct.”¹¹⁹ Curiously, though, the Court acknowledged that Texas did not articulate that justification in Roe, and it appeared that courts and commentators had not actually taken the argument seriously.¹²⁰ On the other hand, however, the Comstock laws, which banned abortifacients along with pornography and contraception, treated the regulation of sexual morality as of a piece with abortion.¹²¹ The Court thought the connection between an abortion ban and the regulation of morality sufficiently important to mention—and dismiss.

Second, the Court acknowledged that forcing a woman to carry an unwanted pregnancy to term is cruel.¹²² It referred to the burdens of pregnancy and childbirth, including the possibility that childbirth “may force upon the woman a distressful life and future,” her “[m]ental and physical health may be taxed by child care,” and the unwanted child may cause “distress, for all concerned.”¹²³ The opinion acknowledged the hardship involved in bringing a child into a family that could not care for the child, and the potential for stigmatizing a nonmarital mother.¹²⁴ The Court accordingly echoed earlier cases treating avoidable pregnancy and childbirth an inappropriate way to advance state purposes because of the burden imposed.

Third, the Court was aware that the states often brought criminal actions against doctors.¹²⁵ One of the parties in Roe, Dr. James Hubert Hallford, allegedly had faced prosecutions for violations of the Texas abortion statutes.¹²⁶ Hallford maintained that the applicable statutes were unconstitutionally vague because he could not determine whether his patients’ situations would qualify as exceptions to the abortion ban,¹²⁷ so he faced punishment for exercising a good faith medical judgment about his patients’ therapeutic needs. Justice Blackmun’s initial draft proposed striking down Texas’s anti-abortion law as unconstitutional only on the grounds that it was void for vagueness.¹²⁸ The punishment that doctors faced in making delicate judgements was clearly a factor in the subsequent Roe decision and in its declaration that abortion decisions should be left to “the woman and her responsible physician.”¹²⁹

Fourth, the Court dismissed state assertions that banning abortion was necessary to protect women’s health, observing that mortality rates during the first trimester of pregnancy “appear to be as low as or lower than the rates for normal childbirth” in contrast with the “prevalence of high mortality rates at illegal ‘abortion mills.’ ”¹³⁰ While less explicit than the Court’s acknowledgment of the burdens of pregnancy, the Court recognized that resort to unsafe abortions was a punitive consequence of the prohibition of legal abortions.

In the background of the case, states’ law on abortion had begun to change, with some states repealing their anti-abortion statutes entirely and others reforming their law to expand the availability of therapeutic abortions.¹³¹ A practical consequence was that, as with contraception, the availability of abortion, particularly safe abortion, differed significantly by race, location, and class.¹³² Partly as a result, women of color were substantially more likely—by some estimates twelve times more likely¹³³—to die from illegal abortion than white women.¹³⁴

In limiting the state ability to restrict abortion, the Court treated these restrictions as imposing impermissible penalties on those seeking abortion. The penalties were not so much the criminal sanctions themselves; these were rarely imposed on the individuals who secured abortions.¹³⁵ Instead, states banning abortion were making childbirth the consequence of unprotected sex—and the risk of death the price of seeking an illegal abortion. The Court found that unacceptable. And while the Court recognized the state interest in protecting fetal life, it balanced that interest against the woman’s interest in deciding whether to give birth. Fetal life, as an interest separated from the sexuality (and women’s bodies) that produced it, would become more prominent as an issue only after Roe was decided.¹³⁶

In these cases, the Supreme Court helped oversee a shift in sexual mores during a period where nonmarital sexuality was becoming more common and accepted. In focusing on the acceptability of the punishment, the Court did not endorse the changes directly; instead, it addressed the rationality of widely violated restrictions that imposed serious, arbitrary and discriminatory harms. The Court’s use of the term “punishment” was not, however, consistent or the subject of a coherent jurisprudence. Sometimes, it referred to the state rationales (deterring sex by limiting access to contraception, making pregnancy the “punishment” for fornication), sometimes it referred to the intrusive nature of criminal enforcement (searching the marital bedroom) rather than the imposition of criminal sanctions, and sometimes it considered the collateral consequences of government action (the stigma and limitations associated with nonmarital births). In the process, however, the Court used the punishment lens to oversee a wholesale effort to strike down what it saw as the outdated remnants of “Victorian” sexual mores without disavowing the legitimacy of state efforts to channel sexuality into marriage.

III. PUNISHING PARENTS

The era that produced Roe involved overlapping interests reducing the support for a punitive approach to sexual morality: a change in sexual norms, a remaking of women’s roles, and more urgent calls for racial equality.¹³⁷ In addition, the parties were less ideologically polarized, with greater elite consensus.¹³⁸

Nonetheless, by the mid-seventies, another jurisprudential revolution was taking place: one embedding a neoliberal¹³⁹ view of the state into Supreme Court jurisprudence. The Warren Court had been sympathetic to calls not just for racial equality, but also for greater economic rights.¹⁴⁰ These claims often took the form of calls to treat government benefits as entitlements, with more equal access to the benefits and more obstacles to denying eligibility. The neoliberal era taking hold by the late seventies rejected these claims. The Court embedded this perspective in the same way it had overseen the change in sexual mores: by using the punishment lens to resolve issues that involved farther reaching clashes in values. The Court did so by denying the very fact of punishment. It concluded that if a given regulation did not penalize the individuals subject to it for protected activity, no constitutional issue arose at all. In the process, the Court upheld regulations that supervised poor women’s sexuality and denied access to abortion funding.

This Section focuses on how the punishment lens applies in more varied civil settings, tracing the evolution of the Court’s treatment of government benefits. The first part of this Section describes how the Court deemed public benefit requirements non-punitive in order to uphold limitations on government benefits under the Aid to Families with Dependent Children (AFDC) and Medicaid programs; the second part of the Section shows how the punishment lens applies outside of the sexual-morality context, analyzing how it has been used to limit access to benefits under the Affordable Care Act.

A. Welfare Benefits and the Rejection of Positive Rights

In the 1960s, the Supreme Court addressed the relationship between sexuality and eligibility for government benefits during a period in which the Court was enhancing access to government benefits more generally.¹⁴¹ The original Aid for Dependent Children (ADC) program was adopted in the 1930s as part of the New Deal’s far-reaching social legislation. The United States, unlike many European nations, had never adopted a universal system of family allowances to support childrearing but instead had a variety of state programs designed to provide widows’ pensions to support children who would otherwise land in orphanages because their mothers could not support them.¹⁴² In the 1930s, Congress nationalized these efforts, providing federal funding for a state-run system to compensate for the loss of a male breadwinner.¹⁴³ Congress limited aid to children who had “been deprived of parental support or care by reasons of the death, continued absence from the home, or physical or mental incapacity of a parent” and allowed the states to impose additional eligibility standards, such as “moral character” requirements that excluded the children of unmarried parents from the program.¹⁴⁴

As early as the 1940s, critics argued that the moral requirements “were habitually used to disguise systematic racial discrimination; and that they senselessly punished impoverished children on the basis of their mothers’ behavior.”¹⁴⁵ The federal government sought to discourage the moral requirements.¹⁴⁶ By the late 1960s, the states had shifted from outright prohibition of benefits to “man in the house rules” that deemed the income of a man who cohabited with a welfare recipient to be available to the family, thereby affecting the family’s qualification for public welfare.¹⁴⁷ These regulations were understood to serve the dual purpose of punishing African Americans and privatizing dependency by withholding public benefits from nonmarital families.¹⁴⁸

In King v. Smith, the Supreme Court examined the punitive nature of these requirements. The Court sidestepped the constitutional issues in the case, striking down the Alabama regulation at issue on statutory grounds, noting that federal law precluded states from denying public welfare to children because “of their mothers’ alleged immorality or to discourage illegitimate births.”¹⁴⁹ The Court concluded that “Congress has determined that immorality and illegitimacy should be dealt with through rehabilitative measures rather than measures that punish dependent children, and that protection of such children is the paramount goal of AFDC.”¹⁵⁰ Justice Douglas’s concurrence, however, would have reached the constitutional issue. He saw Alabama officials as discriminating against children on the basis of illegitimacy and therefore acting at odds with the ruling in Levy v. Louisiana, decided during the same term.¹⁵¹ He wrote that “the Alabama regulation is aimed at punishing mothers who have nonmarital sexual relations.”¹⁵² In administering the provisions, the “economic need of the children, their age, their other means of support, are all irrelevant. The standard is the so-called immorality of the mother.”¹⁵³ He viewed that standard—and the attendant punishment—inflicted on the mother to be constitutionally impermissible.¹⁵⁴

By the time the Supreme Court decided the case in 1968, the nature of the AFDC program had changed. While 43% of the ADC caseload in 1937 consisted of widows, only 7% were in 1961.¹⁵⁵ And as documented in an amicus brief in Levy v. Louisiana, decided the same term, the statute was both “overt discrimination on the basis of the criterion of illegitimacy,” and “covertly discriminate[d] on the basis of race.”¹⁵⁶ The Court almost certainly saw the two cases as linked, although only Justice Douglas’s concurrence in King made the connection directly.¹⁵⁷

In deciding King v. Smith, the majority opinion, however, dealt with these issues only obliquely. Instead, it focused on the irrationality of the punishment imposed—the denial of benefits in a program intended to help children that would disproportionately disadvantage the very children the program was intended to help. The Court did not endorse a right to nonmarital sexuality.¹⁵⁸ It did not discuss the discriminatory motive and effect underlying the regulations.¹⁵⁹ It did not recognize an affirmative “right” to federal benefits¹⁶⁰ nor a right to privacy for benefit recipients.¹⁶¹ Instead, it focused solely on the legitimacy of the punishment, concluding that children could not be deprived of benefits in an effort to change their mothers’ conduct. It treated the man-in-the-house rules not as a rational effort to determine the resources available to the family, but as a subterfuge to continue morals regulation in the face of federal disapproval.¹⁶² The case was thus of a piece with the contraception and legitimacy cases in challenging irrational punishments: punishments that were irrational because once they failed to deter nonmarital sexuality in an era of changing mores, their application became arbitrary and discriminatory.

In subsequent cases, however, the Supreme Court upheld provisions that burdened the poor and their children by deeming such provisions non-punitive. Thus, in Wyman v. James, the Court found constitutional a New York statute mandating home visits,¹⁶³ that were in line with federal law’s requirements that aid be provided only after consideration of the family’s resources and only to children who were not being neglected.¹⁶⁴ The “visits” could prove embarrassing in front of children and guests, and could serve to police sexual relationships.¹⁶⁵ The Court refused to find that mandated visits were a penalty at all, terming them instead a condition of benefit eligibility¹⁶⁶ and not a substantive, much less punitive, standard tying loss of benefits to impermissible or arbitrary considerations.¹⁶⁷

The dissent objected on the grounds that welfare rights should be seen as entitlements.¹⁶⁸ While both the majority and the dissent focused on the status of welfare benefits, Justice Blackmun’s majority opinion used the conclusion that the “conditions” on receipt of benefits were not penalties to lock in a neoliberal view of government action: because there is no right to benefits, the state could impose whatever standards it chooses as preconditions for eligibility, and those conditions never become punishment subject to constitutional scrutiny.¹⁶⁹

B. Punishing Sex

In subsequent cases, the Court’s characterization of a particular government action as non-punitive allowed it to uphold conditions that were challenged as discriminatory, cruel, or unjust. The results were particularly striking when the issue turned to abortion. Legislators who opposed abortion and who could not overturn Roe v. Wade directly sought to express their disapproval of abortion by prohibiting the use of public funds to pay for abortions, while permitting those funds to be used for pregnancy and childbirth.¹⁷⁰ Were these bans penalization of a constitutionally protected right—the right to elect abortion to terminate a pregnancy—or were they simply the exercise of legislative policy preferences to allocate public funds to support some activities and not others? The Supreme Court used the punishment lens to resolve the issue. Since individuals enjoyed no positive right to health care—or to abortion funding—the denial of funding could not constitute a penalty and thus had no constitutional implications.¹⁷¹

An initial case upheld Connecticut regulations limiting public funding of abortions to medically necessary abortions during the first three months of pregnancy.¹⁷² Justice Powell wrote for the 6-3 majority that the Constitution did not impose any obligation on the states to pay pregnancy-related medical expenses of low-income women or any other medical expense.¹⁷³ He noted that the Court had not found in previous cases that wealth was a suspect class¹⁷⁴ and that Connecticut was accordingly free to subsidize childbirth and not abortion as an expression of state policy designed to encourage the former.¹⁷⁵

By 1980, Congress had gone further, adopting the Hyde Amendment, a prohibition on the use of federal funds to reimburse the cost of abortions under the Medicaid program, including abortions that were the result of rape or incest or medically indicated.¹⁷⁶ In a 5-4 opinion later that year, the Supreme Court upheld the constitutionality of the Amendment. The majority opinion treated the issue as a classic one of negative liberty, explaining that the freedom to choose to have an abortion, even a medically necessary one, does not carry with it a government obligation to fund the abortion.¹⁷⁷ It then explained that a woman’s poverty was “the product not of governmental restrictions on access to abortions, but rather of her indigency.”¹⁷⁸ Accordingly, the Court concluded that the failure to pay for abortions was not punishment and thus not subject to constitutional review.

The four dissenters viewed the Hyde Amendment as punitive and cruel. Justice Blackmun made the point that the legislators championing the Hyde Amendment cynically sought to express their own views on the morality of abortion by imposing those views “only upon that segment of our society which, because of its position of political powerlessness, is least able to defend its privacy rights from the encroachments of state-mandated morality.”¹⁷⁹ He would have accordingly subjected the legislation to more exacting judicial review.¹⁸⁰ Justice Stevens emphasized that “the Court expressly approves the exclusion of benefits in ‘instances where severe and long-lasting physical health damage to the mother’ is the predictable consequence of carrying the pregnancy to term” and, indeed, “even if abortion were the only lifesaving medical procedure available.”¹⁸¹ He concluded that the result “is tantamount to severe punishment” for wanting an abortion.¹⁸² Justice Marshall emphasized the racial impact of denying abortion funding and also noted that the Hyde Amendment resulted in “excess deaths.”¹⁸³

In Harris, the Court upheld the validity of an extraordinarily cynical statute. Congress, in effect, limited poor women’s abortion access because it could—it could allow expression of the anti-abortion sentiments of members of Congress at the expense of a relatively powerless group.¹⁸⁴ By declaring that forced birth due to the failure to secure funding for an abortion was not a punishment, the Court avoided addressing the question of whether it burdened a constitutional right.

In Wyman and Harris, neither the majority nor the dissenting opinions treated these cases as imposing punishment for sex, and the majority opinions rejected even the premise that the aid recipients had been punished for the exercise of constitutional rights (privacy in Wyman, abortion in Harris). The reasoning in the cases backtracked on the entitlement language that had been building in the welfare rights era, leading to the conclusion that if the benefits at issue were not entitlements, the failure to provide them could not be seen as punishment—effectively ending the discussion of whether the provisions at issue were unduly cruel or whether they reinforced class- or race-based social hierarchies.¹⁸⁵

C. The Punishment Lens Beyond Sex

The litigation over the Affordable Care Act (ACA)¹⁸⁶ involves the clash of values we have described in this Article and the use of the punishment lens to resolve some of the challenges. The ACA was the largest expansion of public largesse in a half century and therefore a direct challenge to neoliberal values. The legislation’s principle of universal health insurance coverage clashed with those who wished to limit government benefits altogether or to withhold them from those deemed unworthy, such as those who were not working, reinforcing class and racial hierarchies. In addition, by treating contraception as an integral part of women’s health care, the ACA conflicted with the views of some Christian employers who opposed contraception. The legislation thus involved, on a much larger scale, the clash of values underlying the characterization of government benefits in Wyman and Harris.

In the cases discussed in this Section, the Supreme Court returns to the issue of punishment, though without any more precise a definition of the concept. Instead, the Court repeatedly faced the question of whether the ACA provisions operated as a tax or a penalty, a condition or a penalty, and a provision of alternative means of compliance or a penalty, and used the characterization of the actions as penalty or not to resolve the cases.¹⁸⁷ The net effect for the ACA was a compromise: the ACA endured but on somewhat more neoliberal terms than the Obama Administration and the Congress that enacted the legislation might have intended.¹⁸⁸

The ACA, in attempting to provide universal health care access,¹⁸⁹ included a series of alternatives that were designed to balance the principles of expanded access, adequate funding, and reasonable private choice.¹⁹⁰ In National Federation of Independent Business v. Sebelius, the most prominent of the ACA cases,¹⁹¹ the Court addressed two issues that turned on the concept of a penalty. The first involved the “individual mandate,” which required an individual who did not otherwise receive health insurance through their employers or other state provisions, to purchase health insurance on state exchanges or pay what the legislation described as a “penalty” collected by the Internal Revenue Service with the filing of individual tax returns. The Court rejected the government’s claim that the commerce clause authorized the mandate,¹⁹² but upheld it instead as a “tax.”

The Court reasoned that under the ACA, “if an individual does not maintain health insurance, the only consequence is that he must make an additional payment to the IRS when he pays his taxes.”¹⁹³ The Government accordingly argued that the mandate could “be regarded as establishing a condition—not owning health insurance—that triggers a tax—the required payment to the IRS.”¹⁹⁴ Under this theory, the legislation does not establish “a legal command to buy insurance,” just a trigger for owing taxes, like “buying gasoline or earning income.”¹⁹⁵ Therefore, the Court concluded that the ACA was within the Congressional tax power.

Critical to the Court’s reasoning was its decoupling of a requirement to buy insurance, which the Court concluded that Congress could not do, and a requirement to pay an amount, deemed by the Court a “tax,” intended to finance the program. In reaching this conclusion, the Court explained that “[i]n distinguishing penalties from taxes, this Court has explained that ‘if the concept of penalty means anything, it means punishment for an unlawful act or omission.’ ”¹⁹⁶ The ACA mandate was not a penalty (or punishment) because while the mandate sought to incentivize health insurance purchases, it did not make the failure to do so “unlawful.”¹⁹⁷ The fact that Congress sought to influence individual behavior did not matter, just as Congress’s efforts to encourage childbirth rather than abortion did not matter in Harris v. McRae; so long as the federal government did not outlaw the failure to buy insurance, the individual mandate was a tax, not a penalty (and not punishment for the failure to buy insurance). It was therefore constitutional.¹⁹⁸

The second issue the Court addressed was Medicaid expansion, which the Court again decided in terms of the acceptability of the Act’s “penalties.” Congress revised the existing Medicaid program, which is a federal-state partnership, to cover individuals within 138% of the poverty line, and to bring Medicaid coverage in line with the coverage health insurance policies offered on the exchanges.¹⁹⁹ Congress then gave the states a choice: accept federal funding in accordance with the new expanded Medicaid program or forego federal Medicaid funding.²⁰⁰ The majority in Sebelius objected that the “choice” was too coercive, effectively mandating state participation in the program.²⁰¹ It reasoned that while Congress could condition state eligibility for federal funding under a new program, it could not “penalize States that choose not to participate in that new program by taking away their existing Medicaid funding,”²⁰² describing the “inducement” in the Act as “a gun to the head.”²⁰³ Justice Ginsburg’s dissent objected that Congress was, as it had done in the past, just requiring states to comply with “conditions” imposed by Congress to receive Medicaid funding.²⁰⁴

The parallels between Sebelius and Wyman v. James are striking. The requirement that the states adopt Medicaid expansion in order to participate in the Medicaid program could have been described, as Justice Ginsburg wrote, as a condition for participation in a federally funded program. The Sebelius Court disagreed, finding that it penalized the states for the failure to agree to the program’s terms.²⁰⁵ The Court effectively treated the state’s existing funds as an entitlement the federal government could not threaten to take away in order to obtain the performance it sought. In Wyman, because welfare was not an entitlement, a welfare recipient’s failure to consent to intrusive home visits was not considered a penalty at all; it was labelled as “a condition of eligibility” to the continued receipt of benefits.²⁰⁶ The label—condition or penalty—resolved each case without engaging the substantive issue of whether the conditions themselves were reasonable or justified.

In Sebelius, the result cloaks the real issues underlying Medicaid expansion—skepticism about whether the poor merit medical benefits²⁰⁷ and opposition to the state role in meeting such needs.²⁰⁸ Indeed, the federal government picked up 100% of the initial costs associated with implementing the program, and 90% thereafter so that the financial burden on the states was relatively minimal²⁰⁹—and less than the state share of the pre-ACA Medicaid program²¹⁰ and arguably much less of a burden on the states than asking a welfare recipient to consent to frequent, unannounced, and intrusive home visits (or the uninsured to go without health care).²¹¹ What Sebelius did not address is why states opposed Medicaid expansion, given the substantial financial incentives in the ACA for the states to do so. Most commentators attribute the opposition to the states’ ideological opposition to government provision of health insurance, if not outright hostility to the poor people in their states.²¹² Some states continue to resist Medicaid expansion, despite widespread public support for it.²¹³ In effect, the Court, in the name of federalism, authorized the states to act with impunity in frustrating Congressional efforts to ensure accessible health insurance at the expense of people in their states who qualified for the benefits.

In a later ACA case,²¹⁴ the Supreme Court also used the concept of punishment to address the employer mandate, which gave businesses the choice of providing health insurance that met federal standards for their employees or contributing to the exchanges so that employees could purchase their own insurance.²¹⁵ Hobby Lobby, a closely held, for-profit corporation, provided health insurance for its employees, but refused to comply with federal requirements to cover certain forms of contraception, including the morning after pill, because, according to the company, they acted as an abortifacient.²¹⁶ In a 5-4 decision, the Court held that requiring a company to cover certain mandated health care benefits, such as the pills in question, violated the Religious Freedom Reformation Act.²¹⁷ The Court gave little regard to women’s loss of access to the contraceptives, holding that the federal government, if it chose, could provide them through “less restrictive means.”²¹⁸ In short, the Court held that it would be an unjustifiable penalty to compel corporate owners to comply with the terms of a neutral government program that benefitted their employees, if those terms conflicted with the owners’ religious beliefs.²¹⁹

The employer mandate was essentially the same as the individual mandate—it gave those affected, whether individuals or employers, a choice: meet the ACA requirements (individuals by purchasing insurance that met federal standards or employers by providing such insurance) or pay the mandated sums to the federal treasury, in each case less than the cost of the insurance. With respect to the individual mandate, the Court concluded that the payment was a tax on those without insurance and not a penalty because the federal government had not (and could not) compel the purchase of insurance. In the case of the employer mandate, the Court concluded that the required payments were, in effect, a penalty for Hobby Lobby’s desire to act on its religious beliefs, rather than a condition for participation in a program providing federal subsidies.

To be sure, the two cases do not arise under identical bodies of law. Sebelius addressed two distinct legal issues: Congressional power to enact the individual mandate under the Commerce Clause and the taxing power, and the limits of Congressional power under a federal system to incentivize state participation in a federal program. Hobby Lobby was decided in accordance with a third body of law, determining the religious rights of for-profit corporations. Yet, in each case, the Court’s framing of the law as punishment or not—that is, whether the intricate provisions of the ACA acted as sanctions designed to compel specific behavior—determined the outcome. And, in each of these cases, the Court upheld moral hierarchies: protection of religious employers at the expense of employees denied access to federal contraception benefits, protection of states disapproving of health care subsidies at the expense of their citizens who would benefit from such subsidies, and limits on the power of the federal government vis-à-vis other actors, including the states and privately-held businesses. ²²⁰

IV. RETURNING MORALITY TO THE PUBLIC SQUARE

In focusing on punishment, the Supreme Court oversaw a revolution in sexual mores without directly engaging the issue of what values should govern in the public square. The Court has also strengthened a neoliberal regime by simultaneously holding that imposing conditions on program beneficiaries does not constitute punishment while imposing conditions that require coverage constitutes a constitutionally unacceptable “penalty.” In relatively few of these cases did the Court, particularly in its majority opinions, directly engage the underlying values clash. The exception has come in the discussion of LGBT rights—and increasingly in the Court’s opinions on abortion. These exchanges pull back the curtain on the role of punishment in Supreme Court jurisprudence. In these cases, the argument for the losing parties, embraced by the dissents, maintain that punishment is the point—the necessary component to affirming the “right values” in the public square. In response, the Court, in a way it did not do so in the earlier cases, directly addresses the relationship between the status of those affected by punishment and the values they express by engaging in the prohibited activity.

A. LGBTQ+ Rights and the “Homosexual Agenda”

One of the clearest clash of values prior to Dobbs occurred in the Supreme Court’s decision in Lawrence v. Texas.²²¹ In Bowers v. Hardwick,²²² the Court had considered whether there was a fundamental right to engage in same-sex sodomy, a formulation that the Court repeated in Dobbs. In both cases, the Court referred to the long history of criminalizing the conduct at issue,²²³ with those arguing for the constitutionality of such criminal penalties maintaining that the history of punishment reflected disapproval of the underlying conduct and provided evidence of the continuing legitimacy of such sanctions.²²⁴

Lawrence, which involved a criminal prosecution for same-sex sodomy, directly involved the issue of punishment. The two men in the case were arrested in a private residence when the police arrived to investigate a purported weapons disturbance.²²⁵ In his opinion for the majority, Justice Kennedy’s opinion had two levels of analysis. Like the Griswold line of cases, it affirmed a right to privacy, observing that “[t]he statutes do seek to control a personal relationship that, whether or not entitled to formal recognition in the law, is within the liberty of persons to choose without being punished as criminals.”²²⁶ The majority opinion then emphasized that the Texas statute being enforced in the case was not just about prohibiting a “particular sexual act”;²²⁷ it involved intimate conduct as part of “a personal bond that is more enduring.”²²⁸ The opinion thus concluded that such punishment was not just constitutionally impermissible²²⁹ but that the behavior at issue had societal value.²³⁰

Justice O’Connor, in her concurrence in Lawrence, did not go as far as the majority. Instead, in a manner reminiscent of the earlier cases on contraception, she limited her analysis to a punishment lens, finding that Texas could not claim a legitimate interest. She thus rejected out of hand the asserted state interest in the case, which she described as nothing more than the “moral disapproval of an excluded group.”²³¹ For O’Connor, the impermissibility of the punishment—and its discriminatory character—were enough to strike down the statute without necessarily requiring an affirmation of the value of same-sex intimacy.

Writing in dissent, Justice Scalia made clear that he thought that moral disapproval of same-sex sexuality was exactly what the case should have been about. He denounced what he called the “homosexual agenda,” which he defined as “the agenda promoted by some homosexual activists directed at eliminating the moral opprobrium that has traditionally attached to homosexual conduct.”²³² He cast his dissent explicitly in terms of maintaining a moral hierarchy based on that opprobrium.²³³

The opinions in Lawrence thus frame, perhaps better than any of the other cases, the permissibility of punishment and the Court’s use of a punishment lens. They involve a clash between the ability to affirm moral values in the public square versus the preservation of private homes from the intrusion of the state. They also involve the use of the declaration of values to define those to be “protected,” in Scalia’s words, from those to be “excluded,” in O’Connor’s terms, thus reaffirming societal hierarchies between the groups. And they involve the permissibility of the imposition of criminal sanction to reinforce moral opprobrium, even when the behavior at issue is consensual conduct between two adults. The Lawrence Court’s 6-3 majority unequivocally rejected the propriety of punishment used to harden the lines between the protected and the excluded—and in the majority opinion, if not O’Connor’s concurrence,²³⁴ embraced an alternative view of the purpose of sexual conduct as an expression of commitment to a partner, not just as a means to procreation.

In Obergefell v. Hodges,²³⁵ the case upholding the right to marriage equality, the majority went even further in embracing same-sex relationships as an expression of family values²³⁶ while the dissents reaffirmed the need to channel sexuality into marriage—and to punish those who fell outside of such precepts. ²³⁷ Kennedy wrote that there “is dignity in the bond between two men or two women who seek to marry and in their autonomy to make such profound choices.”²³⁸

The majority opinion added that the right to marry is not just about the couples’ relationship to each other, but also about their children. “Without the recognition, stability, and predictability marriage offers,” Kennedy wrote, “their children suffer the stigma of knowing their families are somehow lesser . . . . The marriage laws at issue here thus harm and humiliate the children of same-sex couples.”²³⁹ The opinion thus saw denial of the ability to marry as a punishment imposed not only on the couple but on their children. ²⁴⁰ It accordingly equated the limitation of marriage to different-sex couples with imposition of a stigma on those raising families outside the institution.²⁴¹

The Obergefell majority did take sides in the culture wars—in recognizing the dignity and moral worth of same-sex relationships. In basing the decision on the changed nature of marriage, the Supreme Court acknowledged that marriage reflected a new moral sensibility: one that made autonomous choice, not religious or societal duty, the foundation of the marital relationship.²⁴² The Court accordingly went beyond the rejection of the punishment (while noting “the harm and humiliation” involved in the refusal to recognize same-sex families) to confer public recognition and moral worth on LGBT families.

The four justices who dissented rejected both the premise that marriage had changed and that the Supreme Court should acknowledge that change. Chief Justice Roberts’s dissent explained that “for the good of children and society, sexual relations that can lead to procreation should occur only between a man and a woman committed to a lasting bond.”²⁴³

This reasoning is the same as the reasoning that justified the vilification of nonmarital sexuality a half century ago. In accordance with this reasoning, heterosexual sex, not just procreation, needs to be channeled into marriage and marriage needs to be about a moral command to avoid nonmarital sexuality.²⁴⁴ Punishment, whether material or symbolic, is the necessary complement to this reasoning.²⁴⁵

Alito’s dissent made explicit his objection to overturning traditional moral hierarchies. He wrote: “I assume that those who cling to old beliefs will be able to whisper their thoughts in the recesses of their homes, but if they repeat those views in public, they will risk being labeled as bigots and treated as such by governments, employers, and schools.”²⁴⁶ In short, Alito’s concern lay directly with the ability to uphold the preferred values in the public square and fear that those who did so would now be the ones receiving punishment. And while he acknowledged that family understandings and behavior could change over time, he simply treated data such as the 40% nonmarital birth rate as further reason states could chose to double-down on traditional moral understandings²⁴⁷—drawing clear distinctions between preferred groups and those subject to moral condemnation even when a substantial or even majority of the public did not share such views.

Alito’s opinion accepted the right of moral traditionalists to insist on the primacy of heterosexual marriage and to punish those who create families or engage in sexual intimacy outside of marriage. He saw the majority, in contrast, as embracing same-sex families as entitled to equal moral worth and such views as necessarily punishing those who disagree as bigots. Moreover, he treated evidence of changing norms, such as the increase in nonmarital births, as evidence of a threat to the traditional moral order and therefore as additional reason for punishment. Framed in such terms, the legal question becomes one of power and authority to uphold the preferred views and, in Alito’s terms, punishment cannot be separated from the underlying values.

B. Abortion Revisited

With respect to abortion, however, neither the Court’s efforts to sidestep the morality of the underlying conduct nor its efforts to address the issues directly have yet succeeded. In the years after Roe, abortion became a political marker in part because the issue offers little opportunity for compromise.²⁴⁸ While the Court largely succeeded in making contraception more available without directly embracing the sexual revolution, the Court’s efforts to sidestep the moral issues underlying the abortion issue satisfied no one. Roe satisfied neither those who saw reproductive rights as essential for gender equality nor those who believe the status of the fetus is not an issue that could be “bracketed.” ²⁴⁹ These divisions, unlike those underlying recognition of LGBT relationships, have increased over time.

In Planned Parenthood v. Casey,²⁵⁰ the Court nonetheless tried to tamp down the divisions by directly engaging the values conflicts. Decided in the early 1990s, Casey²⁵¹ had been widely expected to reverse Roe outright.²⁵² Instead Casey preserved the core of the right to abortion, while permitting the states to impose new restrictions, such as waiting periods and parental consent provisions.²⁵³ Justice O’Connor’s plurality opinion was the only significant abortion decision for the Court written by a woman. She observed that the earlier decisions in Griswold, Eisenstadt, and Carey “support the reasoning in Roe relating to the woman’s liberty because they involve personal decisions concerning not only the meaning of procreation but also human responsibility and respect for it.”²⁵⁴ Casey, alone in the Supreme Court’s reproductive rights decisions, made women’s relationship to the growing fetus central to the decision.²⁵⁵ It succeeded, however, only in delaying the day of reckoning over Roe itself.

Dobbs v. Jackson Women’s Health Organization²⁵⁶ is radically at odds with previous decisions that have used the concept of punishment to distract attention from inflammatory subjects. It is also at odds with the conception of judicial statesmanship, through which courts legitimate the judicial system while also recognizing social change and creating community in the midst of conflicting values’ clashes.²⁵⁷ Although Justice Alito claimed otherwise, the decision is designed to inflame and, in doing so, it is likely to empower state officials who wish to exercise their authority to punish—in order to affirm the moral superiority of their position, to reaffirm their values in the public square, to impose dominance over outgroups, and to restore a sense of hierarchical order that validates their position in society. The opinion itself invites such a response.

First, it goes out of its way to say not just that opposing views, but Roe itself were never legitimate. Alito’s majority opinion declares that “Roe was egregiously wrong from the start. Its reasoning was exceptionally weak, and the decision has had damaging consequences.”²⁵⁸

Second, it dismisses women’s interest in their bodily integrity as of no consequence, suggesting that those interests are amply protected through existing laws.²⁵⁹

Third, while the opinion claims not to base the decision on recognition of a fetus as a human being from the moment of conception forward, it clearly views state actions based on such views as a legitimate basis for legislative action and declares that the fact that abortion serves to “destroy a ‘potential life’ ” justifies the Court’s treatment of Roe as precedent entitled to less deference than other Supreme Court precedents.²⁶⁰

Fourth, unlike other Supreme Court decisions announcing a major change in governing law (with all deliberate speed),²⁶¹ the Court provides no guidance for the states and no timelines for implementation. It simply overturns Roe and leaves the states—and the pregnant—on their own in the face of a rapidly shifting and still uncertain legal landscape.

The majority opinion thus has the hallmarks of an act of vengeance righting a wrong, rather than serving to provide judicial guidance in the face of contentious issues. It seeks to restore the moral hierarchy associated with the forces that see abortion as necessarily impermissible. It affirms states’ right to ban abortion without addressing the impact on the rights of states who wish to ensure its continuing availability. And in not only issuing the Dobbs’ decision, but in failing to restrain the states’ earlier vigilante laws,²⁶² the Court’s current stance suggests that the states will be free to treat abortion as murder and punish those who provide abortions, those who seek abortions, and those who aid and abet those involved with abortions in any way.

V. THE FUTURE OF ABORTION PUNISHMENT

Abortion has become a flash point for political division because it falls on the fault lines of cultural polarization and political realignment.²⁶³ After Dobbs, the factors that drive political divisions are likely to overlap with the factors driving calls to punish those seeking and providing abortions.

In analyzing and moving forward on these issues, it is first critical to understand the sources of the call for punitive measures and then to consider whether a focus on punishment can also provide a strategy for defusing the conflict.²⁶⁴ Without such a strategy, this Article concludes, the likely result is a replication of the conditions that preceded Roe: pregnancy as the punishment for sex, aggravating the existing class and regional bifurcation in unintended births; a high-profile fight between elite actors on the boundaries of post-Dobbs public morality; and selective enforcement that disproportionately penalizes poor and minority women. As an alternative, this Article proposes that using the punishment lens analysis can serve as a means to de-escalate the coming legal wars over abortion.

A. Values Polarization and Abortion Punishment

The analysis of the factors underlying the calls for punishment start with the factors driving political polarization. Political theorists link partisan polarization to a sorting between the parties based on cultural values. They describe those with conservative values orientations as favoring in-group unity and strong leadership, and having “a desire for clear, unbending moral and behavioral codes,” that include an emphasis on the importance of punishing anyone who strays from the code, “a fondness for systematization,” as well as “a willingness to tolerate inequality (opposition to redistributive policies).”²⁶⁵

Those with a liberal values orientation, in contrast, tend to be more tolerant to outsiders, to consider context rather strict rules adherence when it comes to determining appropriate behavior.²⁶⁶ They also demonstrate more empathy and less interest in strict punishment for violations of moral and behavioral rules and greater intolerance of inequality.²⁶⁷

Attitudes toward abortion both reflect and contribute to the partisan polarization.²⁶⁸ Abortion attitudes have become more partisan over time, and psychologist Drew Westen describes this outcome as a matter of intentional political strategy.²⁶⁹ Such a strategy was designed to attract people who see abortion in rigid moral terms to the Republican party in the 1990s,²⁷⁰ and as that happened, self-identified Republicans became more opposed to abortion.²⁷¹ Stances on abortion accordingly became a political marker.

Public opinion polls today confirm the high degree of partisan polarization on abortion. While 61% of all Americans believe that abortion should be legal in all or most cases, 60% of Republicans—and 72% of those who identify as “conservative Republicans”—believe that abortion should be illegal in all or most cases.²⁷² In contrast, 80% of Democrats and 90% of “liberal Democrats” believe that abortions should be legal in all or most cases.²⁷³ Public opinion polls indicate that support for the imposition of criminal sanctions closely tracks abortion views generally.²⁷⁴

These attitudes correspond to the purposes and pitfalls of punishment. All groups seek affirmation of their values, but the values to be expressed are not parallel in their relationship to the imposition of punishment. Abortion rights advocates seek to preserve a right to privacy free from government intrusion through the democratic process, including referenda as well as litigation. To the extent they wish to exact punishment for taking away abortion rights, they have suggested defeating anti-choice politicians at the ballot box,²⁷⁵ impeaching Supreme Court justices for perjury about their willingness to follow precedent,²⁷⁶ and requesting ethics investigations.²⁷⁷ We could also imagine more aggressive efforts to counter the efforts of anti-abortion activists who attempt to interfere with abortion in states where abortion remains legal.²⁷⁸ Some of the most important actions pro-choice states have taken, however, is greater support to assist those coming from out-of-state, protecting their own health care workers, and ensuring access to medication abortion. ²⁷⁹ The symbolism involves a greater and more visible state embrace of a right of abortion access.

The punishment desired by those opposed to abortion, by contrast, has two components. The first involves the expressive function of law and the declaration that abortion is wrong.²⁸⁰ The declaration reaffirms the moral hierarchy that elevates those who oppose abortion entirely; empirical studies indicate that when abortion is perceived as a “moral wrong” that produces outrage in those who oppose it; they dehumanize the women (and their partners) who seek abortions.²⁸¹ Expressing this moral opposition even has a “shaming effect” on those who require abortions because of significant health issues.²⁸² It also justifies subjecting those who seek therapeutic abortions to intrusive review of their doctor’s medical determinations or requiring those experiencing rape or incest to face onerous proof requirements, retraumatizing victims of sexual assault. Yet, the symbolic effect can occur with limited punishment, prosecuting only occasional cases that involve public defiance of the new abortion bans.

This dehumanization and shame, in turn, empowers those who would pursue the second component: waging a war to root out the practice. The National Right to Life Committee has proposed sweeping measures, for example, that would not only criminalize abortion itself, but treat it as a “criminal enterprise” that needs to be eliminated using “RICO-style laws” that would reach anyone providing any type of support to someone seeking an abortion. ²⁸³ These provisions target not only medical personnel but those providing abortion information.²⁸⁴ Others propose empowering not only state prosecutors but individual citizens to conduct surveillance on those visiting out-of-state abortion clinics, accessing internet websites providing abortion information, or even monitoring the pregnant (and their friends and family) more generally.²⁸⁵ These activities, particularly when carried out by private “vigilantes,” combine opposition to abortion with a moral crusade. While some laws immunize the pregnant from prosecution, existing laws in many states have already been used to prosecute women experiencing miscarriages for “feticide”²⁸⁶ and more draconian laws have been proposed that provide for prosecution for crimes based on an abortion.²⁸⁷ Even without new laws, the Attorney General of Alabama, for example, threatened to prosecute those crossing state lines to terminate their pregnancies or using abortion pills as child chemical endangerment, even if the patients legally obtain the pills within Alabama.²⁸⁸

Finally, prosecutions, particularly if they are brought against those who seek abortions, are likely to enforce gender, race and class hierarchies. As anti-abortion fervor has mounted, some states over the last decade have increased criminal investigations of various types of pregnancy loss, including not just self-induced abortions but also miscarriages, stillbirths, and any form of infanticide.²⁸⁹ These cases overwhelmingly target “pregnant people who are poor, young, have substance abuse issues or live in areas with limited health services.”²⁹⁰ Advocates fear the reversal of Roe will fuel more such cases and particularly harm women of color, already disproportionately overpoliced and prosecuted on pregnancy-related issues.²⁹¹ Farah Diaz-Tello, an attorney who works on reproductive health rights commented, “It’s this vicious cycle where lack of access, . . . increased scrutiny and stigma around abortion, as it becomes further restricted or criminalized, leads to more criminalization.”²⁹² And the fact that the individuals are poor, minority group members, substance abusers, or otherwise lack full control of their lives contributes to the willingness of others to impose moral condemnation on their behavior.

Dobbs will only make this worse.

B. Punishment in the Courts

Striking down Roe invited the states to adopt abortion bans that, in criminalizing abortion, also prescribe punishment. The courts have historically policed the limits of criminal punishment, requiring, for example, that criminal laws provide clear notice as to what acts are proscribed, that those accused enjoy appropriate procedural protections, and punishments are proportionate to the offence. This Article has gone beyond these traditional concerns to address how the Supreme Court uses a punishment lens to accomplish broader objectives, particularly in the face of irreconcilable and intrinsically divisive issues, and issues that may threaten judicial legitimacy. Abortion certainly qualifies as divisive, and Dobbs has already raised serious concerns about judicial legitimacy.

Indeed, in the years since Roe, anti-abortion activists have made the fetus the issue—with the impact on the person forced to give birth disappearing from view.²⁹³ When the fetus becomes the subject of concern, consensual sex—with no victims other than public mortality—is beside the point. When prosecutors act to prosecute abortions, they are passing moral judgment on the permissibility of the abortion itself and often imposing significant penalties.²⁹⁴

Two arenas in particular, however, offer the Court an opportunity to tamp down the Dobbs-inspired conflicts.

First, if the Supreme Court seeks to deflect the outrage over Dobbs, the simplest way would be to take seriously its own statement that all it has to do is to return the issue to the states.²⁹⁵ Taking that seriously requires protecting the rights of states that wish to secure access to abortion—and protecting, as Justice Kavanaugh suggested in his concurrence, the constitutional right to travel.²⁹⁶ The most basic question involving the right to travel is whether citizens of one state can travel to another state, return to their home state, and be punished for their out-of-state conduct. Existing precedent from the Roe era suggests that such conduct is constitutionally protected and other limits on state jurisdiction ordinarily preclude punishment for out-of-state acts.²⁹⁷ Affirming the constitutional right to travel should also mean that states cannot burden exercise of the right to travel, by punishing, for example, those within the state who assist the traveler in leaving the state or acts that a pregnant person takes within the home state, such as researching out of state options, packing one’s bags, or driving to the state line for the purpose of accessing abortion in another state, just as the Court concluded in Hobby Lobby that forcing an employee to choose between an ACA compliant health plan or a monetary contribution to ACA funding constitution a burden on religious freedom.²⁹⁸ The Court should also strike down punishment that creates obstacles to First Amendment rights of expression, ²⁹⁹ such as penalizing websites or advice to individuals that contain accurate information about abortion and out-of-state availability.³⁰⁰ The Court could also recognize that states encouraging private citizens to track those accessing out-of-state abortion clinics, websites, menstrual periods or other personal information either serves no legitimate state purpose to the extent it is intended to penalize the right to travel or, like searching the marital bedroom for contraceptives, is so intrusive as to be constitutionally suspect. ³⁰¹ Striking down punishment that burdens the right to travel could simultaneously affirm state abortion bans and still protect its availability in the states that permit it.³⁰²

The second arena where a punishment lens could be effective in defusing abortion controversies involves women’s right to medical treatment to protect their health. Statutes banning abortion pose a dilemma for doctors; they report that they fear retaliation for performing abortion-like procedures—even when the fetus is dead or the health threat to the patient is significant.³⁰³ In these cases, the risks are asymmetrical: the doctor faces punishment for “doing the right thing” and little in the way of negative consequences for not acting, even if the patient dies as a result.³⁰⁴ Uncertainty itself thus imposes punishment—and serves the purposes of those who would root out abortion (with inevitable spillover effects to abortion-like procedures). Yet, criminal prosecutions of the doctor in these cases, while risky and expensive for the doctor personally, could bring the criminal justice system into disrepute. For those seeking to ensure abortion access, the question therefore should be how to bring the issue of punishing doctors—and the corresponding ability of the pregnant to receive abortions necessary to protect their health—into public focus. Test cases on enforceability of abortion bans in circumstances threatening the life of the mother might bring greater clarity.³⁰⁵ Such suits could also focus attention on the health threat that punishment poses to pregnant patients. Heavy-handed interventions into newborn care, in which governors sought to prolong the lives of children born with substantial birth defects, helped to discredit the interventions.³⁰⁶ The same approach might work in the context of pregnancy care. Justice Blackmun’s initial draft opinion in Roe sought to focus on the issue of professional judgment.³⁰⁷ Partisan differences on abortion are smaller (and overall support for punishment is substantially less) when the mother’s health is at risk.³⁰⁸ Striking down abortion laws that do not clearly immunize doctor’s decisions about medically therapeutic abortions is a first step; recognizing that the pregnant have a right to abortions necessary to protect their health is an important second step.

In cases of rape and incest, the effort ought to go further to highlight the callous treatment of such victims. Governor Greg Abbott declared, in response to questions about precluding abortion for the victims of involuntary sexual activity, that “Texas will work tirelessly to eliminate all rapists from the streets of Texas . . . .”³⁰⁹ In short, the Governor tried to deflect claims of punishment of one type (forcing the victims of rape to carry the rapist’s child to term) by talking about another type of punishment—that imposed on rapists. The veracity of the claim is not the issue, particularly because Texas has one of the highest rape rates in the country and Abbott had done little to combat it.³¹⁰ As with abortions necessary to protect the lives of the pregnant, partisan differences narrow considerably on cases of rape and incest and the failure to provide such exceptions underscores the punitive nature of the restrictions. ³¹¹

Finally, cases in which patients are prosecuted ought to be used to highlight the cruelty associated with abortion restrictions in the United States.³¹² Restricting access to abortion is in fact just one more form of punishment of the marginalized, with the same groups that support abortion restrictions also opposing more generous provisions to the poor.³¹³ White evangelical Protestants, for example, the religious group most opposed to abortion,³¹⁴ is also one of the groups most likely to respond that aid to the poor does more harm than good.³¹⁵ And the same groups have become more likely to oppose immigration and efforts to promote racial equality³¹⁶ and to favor imposition of preferred values through authoritarian means.³¹⁷ The cruelty of abortion bans is a large part of what motivated the decision in Roe. With abortion opponents calling for draconian enforcement measures,³¹⁸ it should be a factor in mobilizing the opposition to post-Dobbs enforcement of abortion restrictions.³¹⁹

CONCLUSION

Focusing on punishment will not resolve intractable values disputes; it simply changes the subject. Changing the subject, however, does offer a tactic for diffusing intractable disputes—or a long-term strategy for reframing what is at stake. In either case, it makes visible the consequences of public actions, such as abortion bans, on those affected by them in ways that can serve to underscore their cruelty. The public wants its core values expressed and respected in the public square; in cohesive societies the values are consensus based, and punishment reinforces them. The urge to punish, when embedded in group conflict, inflames divisions (threatening violence or civil war); channeling it effectively is central to the rule of law. Understanding this dynamic gives the Court tools (and a motive) to construct an offramp: it also allows states to decide their own approaches to abortion while protecting the pathways out of the states that ban it, and ensures that doctors can save the lives of their patients.

96 S. Cal. L. Rev. 1101

Download

June Carbone*

* Robina Chair in Law, Science and Technology, University of Minnesota Law School.

Naomi Cahn†

† Justice Anthony M. Kennedy Distinguished Professor of Law, Nancy L. Buc ’69 Research Professor in Democracy and Equity, University of Virginia School of Law. Thanks to workshop participants at the University of Minnesota Law School Squaretable for comments and to Sam Turco for research assistance, and to Katherine Bake, Mary Anne Case and John Q. Barrett for comments on an early draft.

After “McCleskey”

March 2024April 2024Robert L. Tsai*

In the 1987 decision, McCleskey v. Kemp, the Supreme Court rejected a black death row inmate’s argument that significant racial disparities in the administration of Georgia’s capital punishment laws violated the Fourteenth Amendment’s Equal Protection Clause. In brushing aside the most sophisticated empirical study of a state’s capital practices to date, that ruling seemingly slammed the door on structural inequality claims against the criminal justice system. Most accounts of the case end after noting the ruling’s incompatibility with more robust theories of equality and meditating on the deep sense of demoralization felt by social justice advocates. One might be forgiven for assuming that defense lawyers abandoned structural inequality claims and the use of quantitative evidence in capital cases altogether.

But that would be wrong and incomplete. For the first time, this Article recounts an unusual chapter of the fallout from the McCleskey litigation, focusing on the litigation and social activism in the wake of that decision. It draws on interviews with anti-death penalty lawyers working for or allied with the Southern Center for Human Rights in Georgia, including Stephen Bright, Ruth Friedman, Bryan Stevenson, and Clive Stafford Smith. It is also based on archival research into their case files. Drawing from these resources, this Article shows how a subset of cause lawyers in the late 1980’s and early 90’s had a remarkable reaction to that demoralizing ruling: they engaged in a distinctive form of “rebellious localism.” Instead of forsaking structural equality claims, they doubled down on them. Rather than make peace with what they believed to be an unjust ruling, they sought to subvert it. They also scrambled to formulate reliable quantitative evidence of intentional discrimination. Instead of accepting existing racial disparities in the criminal justice system, they went after prosecutors and state court judges to expose how racial minorities and poor people wound up on death row more often than their white, wealthier counterparts.

Understanding this untold episode of legal history teaches us about the limits of judicial control over constitutional lawmaking, the unanticipated consequences of trying to insulate the legal order from accountability, and the possibilities for keeping clients alive and earning pro-equality victories when political conditions are inhospitable. For those who pay attention, there are lessons that might humble the most ideologically committed judges and inspire reformers who confront challenging legal circumstances.

INTRODUCTION

We lost McCleskey because these folks don’t understand racial discrimination . . . . They’re trying to see something from too high an altitude. You’ve got to get closer to the ground.

—Bryan Stevenson¹

Stephen B. Bright² and his staff at the Southern Prisoners’ Defense Committee,³ later renamed the Southern Center for Human Rights (“SCHR”), let out a collective sigh upon learning about the U.S. Supreme Court’s ruling in McCleskey v. Kemp.⁴ For months, a team of experts led by the NAACP Legal Defense and Educational Fund had gathered and analyzed data on death penalty proceedings throughout the State of Georgia. David Baldus, a University of Iowa law professor, had run sophisticated regression analyses on over 2,000 murders since the reinstatement of capital punishment in 1976. After taking account of 230 nonracial variables, racial disparities remained. The Baldus study revealed, among other things, that defendants accused of killing white victims were 4.3 times more likely to receive a death sentence than defendants charged with killing black victims.⁵

McCleskey’s attorneys offered two theories of what was happening during the administration of capital punishment in the state, either of which would violate the Equal Protection Clause. First, they claimed that state officials were valuing white lives at the expense of black lives. Prosecutors did this by seeking the death penalty when white victims were killed but not when black victims were killed. Second, they argued that black defendants were being consistently disfavored when compared with similarly situated white defendants.⁶ The NAACP saw McCleskey’s case as a vessel for shutting down the death penalty for good, a reprise of its successful cause lawyering in Furman v. Georgia⁷ that demonstrated that capital punishment was being inflicted in arbitrary fashion.

Writing for a 5-4 Court, Justice Lewis Powell’s opinion dashed those hopes. The Court rejected McCleskey’s equality claims across the board, finding no violation of the Fourteenth Amendment. Without identifying any flaws in the study, the majority nevertheless seized the opportunity to make constitutional policies that tried to insulate the criminal justice system from future racial discrimination claims based on statistical evidence. “At most,” Justice Powell wrote, “the Baldus study indicates a discrepancy that appears to correlate with race.”⁸

Most accounts of McCleskey disparage the ruling for turning a blind eye to racism in the criminal justice system, giving plenty of attention to Justice Brennan’s dissenting view that the majority opinion exhibited “a fear of too much justice.”⁹ Anthony Lewis, writing in the pages of the New York Times, called the Court’s decision “cynical,” creating a legal standard that “would be almost impossible” to meet.¹⁰ Others said that the decision brought to a close the big-case phase of legal liberalism when advocates looked to the Supreme Court as the vehicle for social change.¹¹ For instance, Hugo Bedau said that the ruling “effectively ends the strategy in anti-death-penalty litigation that has dominated the thinking of civil-rights and civil-liberties activists for the past generation.”¹² It may be “that in the present climate of opinion no abolitionist strategy can make much headway,” he lamented.¹³

The Supreme Court’s backing away from Furman meant there was only room for incremental judicial solutions to problems with death penalty administration rather than large-scale court-led reforms. Centrists and conservatives joined forces to permit executions so long as state law guided a jury’s discretion as to whether to impose a death sentence and a defendant was allowed the opportunity to introduce all relevant evidence bearing on the sentencing decision.¹⁴ And in the wake of McCleskey, defendants were left with the possibility of mounting extraordinarily difficult equality-based challenges grounded in evidence of malfeasance by specific bad actors in a defendant’s case. That hardly seemed worth it to try from the perspective of the average person charged with a serious crime, represented by an overworked lawyer and no access to the black box of decision making by prosecutors or jurors. “Smoking gun” evidence was the stuff of movies, not the daily reality in criminal courts across the nation. For these and other reasons, the McCleskey decision exacerbated the crisis of legal liberalism, which for decades had depended on courts to correct structural problems in society. But those prospects dimmed as the necessary conditions for legal liberalism collapsed. The appointment of conservative jurists, a major shift within the Democratic Party to accommodate resistance to the legacy of legal liberalism, and mounting doctrine hostile to a progressive vision of legal change emerged as structural obstacles to abolitionists.¹⁵

Drawing on interviews with key players as well as archival research, this Article shows how a group of abolitionists based in the Deep South tried to exploit the McCleskey decision. They did so by subverting the Court’s intentions and assumptions, challenging a host of local practices and allegedly racist state actors. These advocates created favorable outcomes in subsequent capital cases, and then portrayed their discoveries in ways that shook the McCleskey Court’s tidy presentation of the legal system as well-functioning, rational, structurally sound, and non-racist.¹⁶

Part I revisits the McCleskey decision, emphasizing what the Justices tried to accomplish as well as the law of unintended consequences that often characterizes constitutional politics. Judges write opinions to decide constitutional questions, believing they will be able to persuade people beyond the specific litigants before them, but in reality, judges have little control over how others will receive their rulings or what they will do with them. In this instance, a subset of abolitionists responded by engaging in what I call “rebellious localism,” a strategy of marshaling available resources and deploying them at county and city officials in the service of transformative ends.¹⁷ They made these adjustments to public law advocacy so they could continue raising racial bias claims without giving up on their structural critiques. Besides trying to save their clients’ lives, their goal was to undermine the presumption of fairness and neutrality contained in the Supreme Court’s rulings. Chipping away at the parsimonious vision of racial equality in case after case might then lay a foundation for the eventual repudiation of McCleskey.

Typically, advocates who want to erode a Supreme Court precedent criticize it across the board to try to deny social acceptance of its vision of the law. In this instance, however, death penalty abolitionists denounced the ruling outside of the legal system but tried to exploit assumptions contained in the ruling for a broader vision of equality in their own cases. This adaptation was necessitated by not only the perception of hostility toward the ideal of equal justice, but also the asymmetrical obstacles encountered by defendants who raise fact-intensive constitutional issues. This newfound approach contained three key facets: intensifying issues, subverting doctrine, and scaling downward.

Part II focuses on specific cases in the wake of McCleskey in which SCHR lawyers or their allies escalated their tactics in capital cases. These cause lawyers aggressively asked for hearings (often getting them), put prosecutors and judges under oath, and moved to recuse prosecutors who engaged in misconduct or judges who tolerated racist trial practices. They pioneered creative strategies for gaining access to state records and quantitative evidence to challenge everything from jury venire practices to peremptory strikes by prosecutors. Rebellious localism in this context not only kept their clients alive and raised the costs of litigation for the state, but it also yielded valuable precedent for future cases.

Using two case studies, Part III shows how the strategy of rebellious localism paid off by closely examining how abolitionists’ use of these strategies unfolded in Horton v. Zant and State v. Brooks.¹⁸ Special attention is paid to the synergies between litigation and out-of-court advocacy, as well as the management of tensions between the primary goal of harm reduction and secondary goal of legal reform.

For those who engage in the politics of repudiation, the ultimate objective is to eventually persuade an apex court to overrule a despised precedent, or water it down with exceptions, or to stop using that case to justify policies.¹⁹ Abolitionists and racial justice advocates have not yet achieved that, but other kinds of progress are worth noting, such as the rejection of McCleskeyby state legislators and state courts. The conclusion considers how to evaluate success in ongoing efforts to resist a hated decision and render it infamous in the public mind. It also considers what this episode teaches us about judicially driven efforts to insulate the criminal justice system, as well as prospects for rebellious localism in the future.

I. IN THE SHADOW OF AN INFAMOUS RULING

A. A Difference of Opinion at the Grassroots

Although discretion lies in the hands of several different actors within the criminal justice system, a great deal of responsibility for the disparities probably could be laid at the feet of prosecutors. Formally, in exercising executive power, they act as gate keepers: capital punishment can never be imposed unless a prosecutor files notice to seek it.²⁰ In practice, most prosecutors never seek the death penalty, while a handful opt for it every chance they get. When they do pursue death as a punishment, they also sometimes change their minds.

Some district attorneys operate under strict policies, while others do not even consult an office policy beyond what the state law says.²¹ For instance, in Fulton County, where Warren McCleskey was tried for shooting a white police officer in 1978, the District Attorney at the time “had no written or oral policies or guidelines to determine whether a capital case would be plea-bargained or brought to trial.”²² The absence of consistent policies raised the specter of unbridled discretion, exacerbating any racially biased decisions by individual actors.

At the time, the McCleskey litigation split the abolitionist community along generational lines, but no steps were taken to interfere with an effort driven by advocates who had been in the trenches the longest. While the older guard—represented by the brilliant Anthony Amsterdam—retained their faith that the Justices would once again use their Article III authority to enact broad changes to criminal policy like structural reform cases in the mold of Brown v. Board,²³ the younger advocates were wary of what the increasingly conservative Supreme Court might do with the case.²⁴ Some of them turned out to possess the unique skillset for the next phase of grinding, a case-by-case form of grassroots advocacy necessitated by the high court’s rightward tilt.

Stephen Bright had taken over the reins of SCHR after running D.C. Law Students in Court. From the start, Bright had his doubts about whether the McCleskey litigation would pay off in getting judges to declare the death penalty unconstitutional. “I don’t think it’s going to work like that,” he told Bryan Stevenson, then just two years out of law school and a staff attorney at SCHR.²⁵ Bright did not believe that the Court would allow capital punishment to resume nationwide in Gregg v. Georgia,²⁶ only to shut it back down so soon based on a broad equality rationale. He knew that Justice Lewis Powell, a centrist, had already objected to abolition “by judicial fiat” and expressed hostility to systemic inequality claims in Furman, saying that the penalty’s disproportionate impact on the poor and racial minorities was “tragic,” but ultimately irremediable, and that past intentional discrimination “is no justification for holding today that capital punishment is invalid in all cases.”²⁷ In fact, Justice Powell optimistically believed that the “discriminatory imposition of capital punishment is far less likely today than in the past.”²⁸

Neither Bright nor Stevenson played an active role in the McCleskey litigation. They feared a bad outcome, even though they agreed that discrimination remained a serious problem. In fact, Bright had even told Baldus at the time, “I see even more racism than what you show!”²⁹

On April 22, 1987, by a 5-4 vote, a bitterly divided Court in McCleskey refused to enforce the principle of equality in an opinion by Justice Powell, who had been on the losing side in Furman. In doing so, the Justices made it harder for defendants nationwide to prove racial discrimination within the criminal justice system, especially based on statistical evidence alone. Setting the constitutional bar high, the Justices were not satisfied by a showing of alarming racial disparities; they simply would not act unless someone could demonstrate exactly who was responsible for the inequities in a complex system with many moving parts. When a criminal law was race neutral, the mere risk of racially discriminatory enforcement was insufficient to make out an equal protection violation.

To justify this position, Justice Powell stated that “disparities in sentencing are an inevitable part of our criminal justice system” due to the discretionary roles afforded to prosecutors, judges, and jurors.³⁰ In fact, his interpretation of the Equal Protection Clause treated discrimination as the price of mercy. A jury can decline to impose the death penalty, Justice Powell wrote, but “[w]hereas decisions against a defendant’s interest may be reversed by the trial judge or on appeal, these discretionary exercises of leniency are final and unreviewable.”³¹ From this point on, in order to protect a prosecutor’s “traditionally ‘wide discretion’ ” in making charging decisions, “we would demand exceptionally clear proof before we would infer that the discretion has been abused.”³²

This amounted to a major retreat from precedents like Furman, where the Justices had worried about irrationality and racial discrimination infecting the administration of criminal laws, and even Gregg, where they had allowed the death penalty to be resumed but warned that “the penalty of death is different in kind from any other punishment” and could not be inflicted when there existed “a substantial risk” of depriving someone’s rights.³³ The McCleskey Court worried that the defendant’s equality argument would open the door to consideration of other disparities in the legal system—perhaps even leading to a kind of affirmative action for death sentences.

Justice Powell in particular had become convinced that a win for McCleskey would “throw[] into serious question the principles that underlie our entire criminal justice system.”³⁴ He heartily endorsed the view of prosecutors, who insisted that the defendant’s racial equality claim strikes at “the heart of the judicial system.”³⁵ Absurdly, Powell feared a kind affirmative action that flowed naturally from accepting statistics as evidence of intentional discrimination: “What if one accepts the study as reflecting sound statistical analysis? Would this require that no blacks be sentenced to death where victim was white?”³⁶ He also wished to deter further attacks on the criminal justice system through the use of statistics, troubled that accepting the form of proof represented in the Baldus Study would “invit[e] a system of ‘statistical jurisprudence’—unprecedented in civilized history.”³⁷ There seemed to him “no limiting principle to judgments in criminal cases based solely on statistics.”³⁸ Years later, Powell would express regret for his role in pushing his colleagues to this outcome. But far too late—the damage had been done.³⁹

B. Rebellious Localism

Constitutional actors must calibrate their strategies to deal with shifting political and cultural conditions. Arguments and approaches that work in one historical moment may not work in a different moment. How would social actors in the field of action react at a moment of crisis for legal liberalism? For decades, liberals had turned to the courts as a vehicle for reforming society’s key institutions. The fact that the criminal justice system was among the least affected by the Civil Rights movement did not dampen legal liberals’ ardor for court-centered projects. To the contrary, that fact simply reminded them of how much further judges needed to go. And yet the conservative revolution made most visible through Nixon’s war-on-crime politics, though propelled by elites of both major parties, put legal liberals on their heels.⁴⁰ Public officials elected after complaining of the Warren Court’s excesses secured the appointment of judges hostile to the further expansion of many rights—especially those of criminal defendants.

On a Court reshaped by Republican presidents, Justices Powell, Rehnquist, Scalia, and O’Connor were most skeptical of the Warren Court’s legacy and its approach to legal liberalism. They would work hardest to protect the state’s crime-fighting prerogative by making it more difficult for defendants to bring constitutional challenges and insulating the decisions of police officers, prosecutors, and judges from civil lawsuits. In response to changing institutional conditions, would activists abandon legal liberalism or find some way to rekindle their commitment to the enterprise?

Among racial justice advocates, the immediate reaction to McCleskey was disenchantment. Anthony Amsterdam, who argued Furman in the Supreme Court, called McCleskey “the Dred Scott decision of our time,” one that declares that “African-American life has no value which white men are bound to respect.”⁴¹ Bryan Stevenson felt shattered and never fully escaped the feeling of demoralization. “Most of us were just devastated, most of us were just unbelievably heartbroken,” he said years later.⁴² “I thought about Brown the day McCleskey was decided. I couldn’t make sense of how the United States Supreme Court . . . could be talking about the inevitability of racial bias in the administration of the death penalty.”⁴³

Stevenson denounced McCleskey for enabling unequal application of capital punishment, but in truth, the extinguishment of large-scale equality claims was part of a broader pattern of judicial entrenchment of War-on-Crime policies that fanned out in different directions. These policies included rulings that preserved police discretion on the streets and created doctrines restricting the availability of habeas corpus.⁴⁴ Eventually, McCleskey would form a part of a losing pattern as litigation aimed at making the U.S. Supreme Court’s death penalty jurisprudence more equitable or humane went sideways, reflected in such cases as Stanford v. Kentucky, which refused to stop the execution of juvenile offenders, and Penry v. Lynaugh, a case that permitted the execution of a man with the mental age of a 6-year-old child.⁴⁵ “By the end of the 80’s, all of that stuff had failed,” Stevenson summed up.⁴⁶

Bright similarly believed McCleskey to be “an everlasting blight on the Supreme Court and a badge of shame for the state of Georgia.”⁴⁷ He put McCleskey in the company of not only Dred Scott, but also Plessy v. Ferguson, the notorious ruling that upheld a Louisiana law requiring “separate but equal” passenger coaches and emboldened segregationists for generations.⁴⁸ Thus, for many, disaffection was as predictable as an overall reduction of system-wide legal challenges. That’s where most analyses of McCleskey have stopped.⁴⁹

But for some of the most committed abolitionists in the Deep South, disenchantment about the rule of law did not lead to the wholesale abandonment of equality claims, much less structural ones. As their work necessarily became grittier and more pragmatic—a form of legal trench warfare—these cause lawyers not only continued to emphasize racial inequality claims, but they also broadened their critiques to encompass poverty, a related ground that did some of the work of centering race but also had the benefit of moving to different terrain that might pick up new segments of the community.

In this sense, the Supreme Court’s ruling produced the first of many ironic effects: deepening the resolve of activists to redouble their efforts to prove systemic racism. Opponents of McCleskey would have to outmaneuver their counterparts, ratchet up their tactics against carefully curated targets, delay the worst harms arising from unjust policies, and persuade lower court judges to reshape doctrine to their preferred contours.⁵⁰ Outside the courthouses, they would shame local public servants and citizens into seeing their clients as victims of unequal justice and do something about it.

Of course, rebellious localism in the capital context was not the same as the strategy might appear in other contexts. It remained court-focused because the short-term goal of harm reduction in this high-stakes context—namely, keeping a client alive for as long as possible—remained paramount. But as practiced by SCHR lawyers, the method retained a structural critique as well as a commitment to long-term transformation, with an eye towards grassroots mobilization and construction of a counter-vision of equal justice. All of this was directed at contesting the perception that the Constitution no longer served the interests of the poor or racial minorities.

I call this aspect of their reaction to unfavorable law “issue intensification.” Instead of giving up on equality claims in this domain, which would have been a perfectly natural response to an increasingly hostile political and cultural climate, some social actors instead doubled down on equality. In certain situations where anti-death penalty lawyers might have passed over an equality claim, they lodged one anyway to preserve the issue and in the hope that more evidence to support it might materialize years later. Harsh procedural changes already pushed advocates toward such an adaptation in strategy as they tried to prolong litigation to save lives, but McCleskey reinforced their darkening expectation that problems of proof and procedure would be held against an individual on appeal and during collateral review.

Besides manipulating doctrine on substance and procedure in ways that disfavored progressive liberal legal projects, high-profile judges also advocated legislative reform to close the courthouse doors further, citing abusive litigation. A year after the McCleskey decision, Justice Powell agreed to serve on an ad hoc committee established by Chief Justice Rehnquist to propose restrictions on habeas corpus petitions by death-row inmates. Justice Powell’s leading role in McCleskey, followed by his subsequent call for additional restrictions on the Great Writ, deepened abolitionist suspicions that the Supreme Court was biased against racial minorities and wanted to speed up executions regardless of the costs. The growing judicial-legislative initiative, which would attract the support of War-on-Crime state attorney generals, heightened their awareness that every possible trick would be used to prevent structural injustice from being seen or addressed.⁵¹ As a result, advocates would at times raise more allegations of discrimination than they likely would have before McCleskey. Although they rejected nearly every word of the McCleskey ruling in their hearts, they managed any sense of dissonance by citing the case in court and treating it as good law, while denouncing it in other domains.

Bright was a realist, as well as a close and creative reader of judicial decisions. After “pulling the decision apart,” he urged fellow abolitionists to take the Justices at their word.⁵² The key was Justice Powell’s insistence that a defendant had to prove that decisionmakers in his own case acted with a discriminatory purpose. Bright told Stevenson and other allies: “There’s language in here we can do something with.”⁵³ The best response, he said, was to identify every single situation where a judge or prosecutor might have acted out of bias and “to litigate the hell out of it.”⁵⁴

Bright’s move to turn a shield fashioned for prosecutors into a sword to be wielded by defendants should be considered a form of “doctrinal subversion.” Instead of acceding to the Court’s plan to insulate the criminal justice system, SCHR and its allies redeployed the precedent to go after some of the system’s most prominent guardians: prosecutors and judges. This amounted to ignoring the system-insulating motives in McCleskey itself, and seeing just how far they could take the Court’s insistence that a defendant could still advance a claim if he offered evidence “that would support an inference that racial considerations played a part in his sentence.”⁵⁵

These abolitionists even called their filings “McCleskey motions” to reclaim that precedent on behalf of a more robust vision of equality. The Supreme Court’s ruling in McCleskey was well on its way to becoming an infamous decision in the legal academy and among the public interest lawyers at large, but in these Southern courtrooms, Bright and others treated

the decision as if it represented good law in demanding discovery and hearings.

One of the things McCleskey questioned was “whether any consistent policy can be derived by studying the decisions of prosecutors” across several different counties since so many variables were involved in charging decisions.⁵⁶ In other words, Justice Powell thought the inference of intentional discrimination was greatly weakened—he said the disparities were not “stark” enough—because of the sheer spatial expanse from which the data was derived as well as the large number of human actors making decisions based on “innumerable factors.”⁵⁷

Bright proposed a solution to this problem: identifying situations where a particular prosecutor, judge, or local jurisdiction had a reputation for engaging in discriminatory behavior. I call this aspect of SCHR’s effort to make equality claims more palatable to skeptical judges the scaling down of relevant constitutional theories and the targets of enforcement. By focusing on specific bad actors and simplifying the causal stories of discrimination, advocates hoped they would not trigger the slippery slope problem Justice Powell found so daunting.⁵⁸ The goal was to go hard after unethical prosecutors and “hanging judges” (judges elected based on their record as tough-on-crime prosecutors) who appeared to treat black defendants, black jurors, or black victims more harshly. So long as the evidence was compelling, offering a judicial remedy could be isolated to particular actors or local jurisdictions and therefore would not bring the entire justice system to its knees, as Justice Powell feared.

All three adjustments—intensifying their emphasis on issues of equality, subverting doctrine, and scaling down—comprised a response to negative developments in constitutional law. Together, these tactical choices helped resuscitate belief in legal liberalism at a time that Mike Seidman has incisively described as “an environment marked by liberal collapse and conservative hibernation,”⁵⁹ before even more muscular forms of judicial conservatism appeared.

These shifts in strategy entailed making tradeoffs: how to continue engaging in advocacy that might produce measurable gains through harm reduction as well as doctrinal improvement, while managing scarce resources in a time of cultural fatigue over racial equality claims. By making these adjustments, Bright and his allies developed a distinctive form of rebellious localism. This major shift in strategy can be understood as an example of what Mark Tushnet has termed “defensive crouch liberalism”⁶⁰—adaptations by progressives to the fact that federal courts became dominated by conservatives. But these adjustments lacked the “nervous” form of advocacy Tushnet suggests is emblematic of liberalism seen in so many quarters for fear of retaliation by conservatives.⁶¹ Rather, presuming a national conservative backlash was already underway, the trick became finding moments, spaces, and places far away from national elites where conceptions of rights could be pushed openly and fearlessly. In doing so, SCHR and its allies began to construct a counter-narrative that post-racial America had not yet arrived, and that the criminal justice system had never been successfully reconstructed.

C. Post-McCleskey Surprises

A fruitful point of entry into the historical record is to consider these strategic adaptations in the wake of seemingly unjust Supreme Court precedent as a species of unintended consequences. Sociologist Robert Merton has called the unanticipated consequences of purposive action “those elements which would not have occurred had the action not taken place”; such effects “result from the interplay of the action and the objective situation.”⁶² Most of the time we act “not on the basis of scientific knowledge, but opinion and estimate,” Merton writes. This is especially true when it comes to law as a species of social activity: judges create doctrine hoping that certain consequences will happen. They tinker with doctrine, often overestimating their own ability to shape the behavior of others. Even so, this is unavoidable: judges make constitutional law predicting that relevant social actors will internalize their pronouncements and behave in certain ways, but in truth judges have wholly inadequate information about other social actors’ motivations, commitments, and resources.

Merton identifies the situation where “[p]ublic predictions of future social developments” can themselves “become a new element in the concrete situation” and thereby “change the initial course of developments.”⁶³ Just as Marx’s prediction of progressive concentration of wealth and increased misery of the masses helped lead to the rise of organized labor, thereby slowing developments predicted by Marx, so too it might be said that popular awareness of the Supreme Court’s desire to insulate the criminal justice system from fundamental challenge and thereby wish a post-racist society into being itself became a factor in confounding those original expectations. As Merton points out, “[t]his contingency may often account for social movements developing in utterly unanticipated directions.”⁶⁴ A case like McCleskey can become so imbued with social meaning that it becomes part of a broader struggle over legal principles, thereby facilitating surprising shifts in organizational behavior and to the law itself.

Consider several unintended consequences. First, Justice Powell did not expect that his decision would be used as Exhibit A in teaching a generation of lawyers and activists about poor constitutional reasoning and the problem of judicial callousness. Within the criminal defense bar, and among the smaller, tight-knit community of cause lawyers, McCleskey became a rallying cry to reject the Court’s seeming command to abandon efforts to prove systemic racism. The ruling itself began to play a role in identity formation: more criminal defense lawyers began to see themselves as cause lawyers taking on a thoroughly unjust legal system—at least part of the time.

In fact, the ruling spawned political efforts to repudiate the ruling’s tolerance of racial inequities in the criminal justice system. Legislation was introduced in Congress to permit defendants to introduce statistical evidence to demonstrate that the death penalty is enforced in a discriminatory manner. Efforts to undermine McCleskey’s logic also emerged at the state level. Although the Racial Justice Act has never been enacted by the U.S. Senate (it has passed in the House), grassroots activism led to the enactment of state analogues in North Carolina, Kentucky, and—in the most far-reaching form—California.⁶⁵

Second, the Court underestimated the possibility of a differentiated response to the ruling due to the variety of criminal defense lawyers. It’s one thing to expect court-appointed lawyers heavily dependent on judicial favor or already overburdened public defenders to shy away from making equality claims after a doctrinal test is made harder to satisfy. It’s quite another to expect abolitionist organizations to renounce equality claims without a fight. Groups with the capacity to draw upon out-of-jurisdiction manpower and donations could still choose to invest in the heavy lifting necessary to pursue allegations disfavored by the Supreme Court and do so in venues beyond that institution’s capacity to control.

Nor did the Justices foresee that success in a single case could be leveraged successfully to throw other death sentences, and even some convictions, into serious doubt—and even spawn more structural equality claims. When a judge found a prosecutor or judge engaged in unconstitutional actions, especially a violation of the principle of equality, SCHR lawyers exploited that decision for all it was worth, arguing in other cases involving that person or the entire office was hopelessly tainted by racism.

To these advocates, McCleskey had ripped the mask off an unjust system. If they continued to behave like business as usual and gave such figures the benefit of the doubt, they would be complicit in perpetuating cruelty and inequality. Instead of propping up that illusion of neutrality, they vowed to break norms of civility and insist that judges and prosecutors step aside—another unanticipated byproduct of the ruling.

Third, in constructing McCleskey, Powell believed he was protecting faith in the rule of law by shutting down all but the rarest system-wide constitutional challenges. Instead, the ruling would end up inspiring a fresh round of even more intensive equality-based challenges at the local level. Worried about the need to document all the ways that race and poverty shaped capital outcomes and educate fellow citizens as well as jurists, some advocates filed more pre-trial motions than they ever did before, challenging everything from jury selection processes to prosecutorial charging decisions. During post-collateral proceedings and retrials, SCHR lawyers also demanded transparency, seeking access to a prosecutor’s notes to bolster their claim that charging decisions were racially discriminatory or that prosecutors had tried to manufacture all-white juries.

A subset of cause lawyers thus engaged in more high-risk, high-reward moves to target bad actors involved in administering the justice system. Individual judges and prosecutors were swept up in these efforts to expose the inner machinery of the death penalty system, which blackened not only their reputations, but also that of state attorney generals who chose to defend local practices. That unintended development arguably undermined faith in the rule of law far more than endorsing McCleskey’s challenge would have done, especially if the Court could have envisioned a path to a narrow remedy in that dispute.⁶⁶

Fourth, with the exception of Justice Scalia, who objected privately to any suggestion that the Baldus Study was inherently unsound (he was simply more forthright in his willingness to tolerate persistent inequities),⁶⁷ the Justices were uneasy with quantitative analysis. The majority hoped to dissuade structural advocacy that relied upon these kinds of statistics. Justice Powell warned his colleagues that if the Baldus study were accepted as sufficient proof of an equality violation, it would “invit[e] a system of ‘statistical jurisprudence’—unprecedented in civilized history.”⁶⁸ The irony, of course, is that by expressing such a distaste for numbers and a strong preference for direct evidence of racial hostility, the McCleskey decision may have encouraged advocates to become more aggressive and comprehensive in introducing evidence of actions, sentiments, and associations that at times could actually be less reliable in demonstrating racial bias. Simpler evidence is not always better evidence. In retrospect, it may have been suboptimal to funnel such complicated evidence of structural bias into state courts, which might have less expertise with such evidence.

Acquiring direct as well as quantitative evidence of bias, and then arguing about its merits openly in court, would also inevitably prolong trial proceedings and threaten to turn them into communal spectacles. Such consequences, too, would not have been optimal from the standpoint of the Supreme Court, which had increasingly prized efficiency and finality in capital cases. The justices’ failure to consider this possibility was anticipated by Merton, who describes “the ‘imperious immediacy of interest’ . . . where the actor’s paramount concern with the foreseen immediate consequences excludes the consideration of further or other consequences of the same act.”⁶⁹

In short, Justice Powell wanted fewer structural equality claims, not more, and reduced conflicts over race, rather than more intense ones. He hoped to dissuade use of statistical proof to undermine criminal judgments and promote belief in the rule of law. As we shall also see, the McCleskey decision certainly did not put an end to a battle of experts. It did not even

stem allegations of racism in death penalty cases. It just shifted the theater of conflict.

II. SHAMING THE SYSTEM

Litigating in the shadow of McCleskey “was a time of transition,” recalled Stevenson; “[t]hinking more creatively about how to expose these problems became a higher priority.”⁷⁰ It’s important to see that rebellious localism was not just about adjusting litigation strategy; it also became about grounding these new courtroom tactics in grassroots mobilization. The McCleskey decision felt like a historical break to these activists. They saw it as the moment they realized “a new model would be needed.”⁷¹ Others observed a decisive rhetorical shift in Bright, who from that moment on “talked more proactively about issues of race, and class, and power.”⁷²

Instead of giving up on equality claims, Bright encouraged his staff to double down on proving inequality. McCleskey required a person to prove that specific individuals “in his case acted with discriminatory purpose,” and that’s exactly what they would do.⁷³ With Bright leading the charge, they vowed to go county by county, prosecutor by prosecutor, and judge by judge if necessary to ensure their clients got a fair trial. But they would not let go of their sense that the problem of inequality was not that of a few individuals, but entire offices and even jurisdictions.

“No one was prepared before McCleskey to go into some place like Swainsboro, Georgia, and use the language of intentional discrimination in a proactive way for an individual client,” said Stevenson. “But that’s what we did after McCleskey.”⁷⁴

A. Expanding the Scope and Targets of Equality Claims

McCleskey had emboldened death penalty enthusiasts. For instance, Georgia Attorney General Michael Bowers confidently stated, “I do not see any racial discrimination in motive or effect in the imposition of the death penalty.”⁷⁵ Bright believed that most members of the Supreme Court simply had no idea who was involved in keeping the machinery of death humming.⁷⁶ As a result, his staff had to build a persuasive record of bias in each case they handled. This meant two things: (1) intensifying challenges to suspected bad actors and getting more deeply into the background of prosecutors and judges to discern their racial attitudes; and (2) continuing to make structural inequality claims by experimenting with creative and aggressive ways to bring such claims in capital cases. Publicly expressed sentiments and actions, office practices, family histories, and even personal associations all became fair game.

In the view of SCHR lawyers, citizens and jurors became cogs in a killing machine that policymakers tried to make more efficient; the accused, dehumanized through the legal process. Through rebellious localism, Bright and his staff vowed to not only show the inherent brutality of this form of punishment, but also record and dramatize its racial effects in each case in which they played a role. They believed that citizens would be more hesitant to impose it in specific cases, and that the community as a whole would eventually turn away from the practice once stories of unfair and unequal administration spread.

Scaling down thus also meant bringing novel legal challenges to older forms of racial power that many people had just accepted long ago as part of the fabric of social life in the Deep South: Confederate battle flags in official state flags,⁷⁷ a judge’s membership in an exclusive club or voting record as a former legislator, a relative’s membership in the Ku Klux Klan or participation in a lynching. Their goal was to show just how much death penalty trials were permeated by racial bias in ways that went well beyond what the McCleskey Court had assumed.

Moving to recuse judges and prosecutors with a reputation for racial discrimination became part of the arsenal. SCHR lawyers also began filing motion after motion demanding that judges disclose how often they used the “N” word, whether they had ever hired African Americans as clerks and staff, and whether they sent their kids to segregated schools.

When a judge or prosecutor would resist making such disclosures, as expected, insisting that “you can’t put me on trial here,” Bright converted the McCleskey decision from a shield into a sword.⁷⁸ “Yes, I can,” he told skeptical trial judges.⁷⁹ “McCleskey v. Kemp says I have to show whether you are prepared to give fair and just treatment to my minority client.” ⁸⁰ This was “the kind of litigation most people ran from,” Stevenson explained, “but Steve embraced it.”⁸¹

More so than in its early years, SCHR treated this new era of “hand-to-hand combat” as a chance to get local communities involved. Before McCleskey, they worked mostly with the criminal defense bar and a smattering of cause lawyers. But afterward, “the connections with the traditional civil rights community started to take shape.”⁸² When SCHR lawyers went into a community to defend a person against the death penalty, they would bring a famous civil rights leader such as C.T. Vivian, who had been close to Martin Luther King, Jr., or Dr. Joseph Lowery, president of the Southern Christian Leadership Conference.

By spending time in these local communities, they met families who had experienced first-hand neglect and mistreatment by police and later, prosecutors and judges. “It became apparent that civil rights were going to have to be at the heart of what we do,” Stevenson explained.⁸³ For instance, during one pre-trial hearing, SCHR lawyers put on black witnesses who had been victims of violent crime but had not heard much from the district attorney’s office. Such evidence, in Bright’s view, showed that “black lives are not valued in the same way as white people’s lives,” and bolstered the defense’s empirical evidence indicating that the prosecutor’s office sought the death penalty far less frequently when similarly situated defendants had allegedly killed black people. Discriminatory charging decisions, they argued, denied one such black defendant on trial for killing a white person, William Brooks, equal protection of the law. In fleshing out this claim, the defense sharply contrasted how a District Attorney’s office would maintain close relationships with white victims of murder—even sometimes giving families a voice before major decisions in a case were made—with the frosty relationships with black victims. This amounted to not only evidence of official bias, but also what Monica Bell has called “legal estrangement” on the part of black residents.⁸⁴

Additionally, because black citizens were not responding to jury summonses at the same rate as white citizens, SCHR began holding events at black churches where a prominent civil rights leader would give a rousing speech “creating a consciousness that there was an obligation to serve.”⁸⁵ Stevenson or Bright would warm up the crowd by talking about a case they had in that county. Then a famous civil rights figure like C.T. Vivian would take the stage and tell stories of how he was beaten when he marched but had gotten back up for their right to vote and to serve on a jury. “Now when you get summoned for jury service, I want you to walk to the courthouse like we walked to the courthouse in 1963, like we walked across the bridge. Don’t wait until it’s your son sitting in a courtroom filled with white people and an all-white jury.”⁸⁶

Community organizing went hand-in-glove with litigation strategy. In the retrial of William Brooks, for instance, Bright filed a creative motion to deal with the fact that the poor response rate by black citizens to jury summonses had made it easier for prosecutors to manufacture an all-white jury during his first trial. In trying to deal with the pervasive problem of legal estrangement in their own case, they pointed out that “only eight of 160 summoned jurors were black” and that “a representative venire would have contained 50 black citizens.”⁸⁷ They told the judge that “to protect the constitutional rights of the accused,” he could not be passive but instead “should direct the service of summons to jurors be accomplished in a way that ensures a fair cross-section of the community actually appears for jury service.”⁸⁸

Bright and his staff asked the judge presiding over the retrial to provide specific relief: (1) that a “special venire of at least three hundred person be drawn with the initial selection process taking place in open court”; (2) that the Clerk issue summonses to those jurors selected and commanding them to appear “without making mention of the name of the case to be tried”; (3) that personal service be made upon any jurors who fail to respond; (4) that the clerk, sheriff, and other court personnel be directed not to disclose the case or name of the defendant prior to the jurors appearing in the courtroom; and (5) that any and all other steps be taken to ensure the venire represent a fair cross-section of the community.⁸⁹ These motions sent a powerful message to the prosecutor, the judge, and local citizens that trials of black defendants by all-white juries would be perceived as illegitimate even though the Supreme Court has never held that a defendant has a constitutional right to a diverse jury.⁹⁰

To ensure that poorer citizens would be able to participate, Bright also filed pre-trial motions asking that jurors be paid “their current wage and to compensate primary caregivers for day care costs.”⁹¹ They argued that “[d]aily wage earners and primary caregivers for young children are cognizable groups for Sixth Amendment purposes” and that their underrepresentation in Brooks’ trial and others would violate the Constitution.⁹²

Black citizens began appearing for hearings in SCHR’s cases, especially when they sought to recuse judges or prosecutors or get the state flag, which contained the Confederate flag, removed from the courtroom for the trial. Instead of an empty courtroom, the place would be packed with black people concerned about the quality of justice, who transformed a mundane legal proceeding into a communal event.

That happened for the retrial of Brooks. In a letter dated September 5, 1990, Bright and George Kendall discussed several themes that Dr. Joseph Lowery might work into an upcoming community presentation in Columbus.⁹³ The first theme was to explain that “discrimination is worse in the criminal justice system than anywhere else, that everyone has looked the other way for too long, and that it is urgent that something be done about it.”⁹⁴ Second, citizens could “learn more about discrimination in the justice system by going to our hearing in the William Anthony Brooks case on discrimination.”⁹⁵ Third, concerned citizens could sign postcards SCHR prepared to urge members of Congress to enact the Racial Justice Act. They could also organize to deny the current district attorney a judgeship “if he continues his practice of discrimination in seeking the death penalty.”⁹⁶ As Bright explained to Dr. Lowery: “The last two District Attorneys both are judges now and both used the death penalty to get there.”⁹⁷ According to Bright and Kendall, “[b]oth Whisnant and Smith were famous for lynch-mob type closing arguments at the sentencing phase where they would ask for the death penalty.”⁹⁸

Most critically, Bright and his staff came up with imaginative ways of developing their own evidence against the worst prosecutors and complicit trial judges. They did so, for example, by taking advantage of a state’s “sunshine laws” to inspect state records. In Georgia, those records extended to a prosecutor’s files once an appeal had been decided. SCHR lawyers also began putting historians, statisticians, and other experts on the stand during pre-trial hearings to document the state of local justice. They would litigate pre-trial motions—including any McCleskey motions—vigorously.

“We really focused hard on this,” Stevenson recalled. It was a concerted “reaction to McCleskey and the new world we were in. We had to more carefully articulate the nature of racial bias.”⁹⁹

And so they did—by cataloging and exposing inequality and unfairness wherever they could, in every setting in which an issue about the quality of justice could be credibly raised. To Stevenson, this major shift in strategy was “really exciting because we all knew the truth about how these people thought. We had experienced it.”¹⁰⁰ He could see that it was also “empowering” for Bright because “you observe bias time and time again, and the conventions of the law require you to stay silent. To now be able to use the law in service of saying something was very, very energizing.”¹⁰¹

Paradoxically, SCHR was well placed to engage in rebellious localism precisely because they were not part of the local power structure. They were based in the region, but not insiders. The organization’s funding did not come only from court-appointed cases. The overall strategy thus played to SCHR’s strengths. Staff could do the things that court-appointed lawyers, whose livelihood depended on good relations with local judges and prosecutors, simply could not do and for self-interested reasons, would not do. Bright also believed that requiring attorneys to take criminal cases for such low pay meant that lawyers often did not care about their non-paying clients. How many times had a court-appointed lawyer not bothered to make a constitutional claim, investigate something a client said or contact a possible witness, or lodge an objection simply because doing so would drag out the case or annoy members of the local power structure?

This point was not about comparative or absolute resources as much as it was about relative independence. This strategy was deployed in the early decades of the era of mass incarceration, when SCHR had only a shoestring budget and relied heavily on small donations, a few grants, and fee awards from prisoner’s rights lawsuits. Yet, freed from the financial and political conditions that structured how everyday criminal cases were handled, the organization found itself in a position to demand more of the legal system. Bright and his staff would still insist upon being paid by the state for criminal defense work just like local lawyers who had to eke out a living by taking court-appointed cases, even if they ended up working longer hours that would go unreimbursed, because the principle of equality for indigent representation mattered. At times, they would even file separate lawsuits to lift existing restrictions on how much court-appointed defense lawyers could earn, believing that fee caps led to less thorough investigations and less competent representation of the poor.¹⁰²

During these years, Bright began developing his critique that people were sentenced to death not so much for the crimes they commit but rather for having an inadequate lawyer.¹⁰³ He realized that the absence of a public defender system in many places made unfairness a built-in feature of the justice system. The already asymmetrical criminal process left the quality of counsel for the poor completely controlled by local judges. This often led to shoddy representation by court-appointed lawyers who often wanted to close cases as quickly as possible and get back to paid work. Funds for experts were hard to come by; many lawyers representing the poor never bothered to ask.

Rebellious localism yielded other benefits beyond spotlighting racial inequities. A certain synergy arose between legal challenges to the practices of certain prosecutors and judges on behalf of clients facing the death penalty and SCHR’s civil challenges brought against local indigent defense systems, plea bargaining practices, and bail rules. More than once, focusing on a particular judge’s practices alerted the organization and others to pervasive problems that could then be challenged separately. Eventually, rebellious localism acquired a civil dimension to supplement its criminal law focus, as the Center added a variety of civil litigation strategies to try to halt death penalty proceedings or gain access to evidence that might bear on their claims. In showcasing the various deficiencies of the justice system, rebellious localism also helped revive the public defender movement within the state.¹⁰⁴ Overall, the approach proved highly effective in attracting media coverage of the nested problems within the criminal justice system.

B. Unrigging Juries

Because “that’s where we had the best law,”¹⁰⁵ that is, precedent vindicating the importance of the right to a jury of one’s peers stretching back to the post-Reconstruction Supreme Court case, Strauder v. West Virginia,¹⁰⁶ one area of intensive focus became about challenging a jury’s composition. Strauder had struck down a state law that barred black men from serving on juries. Over a hundred years later, it was still the case in many death penalty jurisdictions that “there were no black prosecutors, no black decision makers, and the only opportunity for a black person to play a role was on the jury,” Stevenson observed.¹⁰⁷ A good illustration was Columbus, Georgia, whose 10-person prosecutor’s office were all white men until the mid 1980’s, when the first black prosecutor in that county’s history was hired.¹⁰⁸ But, of course, in case after case Bright and others noticed “elaborate efforts being made to exclude black people.” Through their cases, they decided to “elevate how exclusionary the juries had been.”¹⁰⁹

In Amadeo v. Zant, Bright’s first case before the U.S. Supreme Court, Justice Thurgood Marshall invalidated Amadeo’s death sentence because of Putnam County’s scheme to systematically underrepresent black people and women in its jury pools.¹¹⁰ The Court singled out District Attorney Joe Briley’s memo to jury commissioners instructing them on how to permit just enough women and blacks into the pool without making it look too suspicious. A lawyer bringing an unrelated civil rights lawsuit against the county had discovered the “smoking gun” evidence. Based on this new evidence, a federal district judge had granted a writ of habeas corpus, but the Eleventh Circuit reversed, saying the claim was time-barred. Justice Marshall’s decision disagreed with the Eleventh Circuit’s handling of the procedural default issue and restored the federal judge’s order granting relief.

When the case was returned to Putnam County, Briley promptly had Amadeo reindicted for murder. A secondary legal battle then ensued when the trial judge refused to appoint Bright and his staff as attorneys of record until the Georgia Supreme Court ordered him to do so. Instead, the judge had tried to stick Amadeo with a different set of lawyers who knew nothing about the case.

SCHR had to battle the trial judge, the district attorney, and the state attorney general’s office to vindicate Amadeo’s right to the counsel of his choosing. Ultimately, the Georgia Supreme Court issued a landmark decision reversing the trial judge for interfering with Amadeo’s “relationship of trust and confidence with prior counsel.”¹¹¹

Back on the case, Bright and William Warner filed a slew of pre-trial motions. One motion sought to quash the indictment on the ground that the grand and traverse jury master lists “do not reflect a fair cross-section of the community.”¹¹² Referring to Amadeo’s earlier win in the U.S. Supreme Court, they insisted that he could not be tried unless additional “precautions be taken to assure that this time,” his rights were protected.¹¹³ Besides pointing out that the county’s past jury selection methods excluded women and black people, they added a new Equal Protection claim based on class. “The selection system utilized in this county is not class-neutral,” they argued, “and does not assure the inclusion of persons who are not registered to vote, persons who have recently moved to the county, and low income persons.”¹¹⁴ Because Amadeo was not a resident of Putnam County but an indigent “outsider” charged with serious crimes against “a highly respected member of the community,” they objected to a system that generated juries that lacked important “experiences, opinions, and perspectives.”¹¹⁵

In addition, Amadeo’s team filed an ex parte motion seeking funds for an expert to help them determine whether the county’s jury pools were, in fact, underrepresenting “recent migrants, unemployed persons, daily wage earners, [and] persons who are not registered to vote.”¹¹⁶

In an even more daring effort, Amadeo’s attorneys moved to disqualify all judges from the Ocmulgee Judicial District, made up of eight counties, including Putnam, based on a violation of the Voting Rights Act. They argued that the judges of the Ocmulgee Judicial District were elected as a result of unconstitutional procedures. The use of at-large election procedures for superior court judges had never produced a single black judge in the circuit, they noted, even though the population of the counties were between 37.42% and 78.24% black.¹¹⁷

This last motion revealed SCHR’s resolve in thinking structurally. Amadeo’s team demanded an evidentiary hearing so they could show “total or seriously disproportionate exclusion of black citizens from the Georgia judicial system, disproportionate impact of the decision made in the election and judicial systems of Georgia, the opportunity for discrimination, as well as other circumstantial and direct evidence of intent to discriminate as may be available.”¹¹⁸

On a couple of occasions, Bright put Stevenson on the stand as an expert witness to testify as to the difference between comparative disparities and absolute disparities. Stevenson had received a degree in public policy at the Harvard Kennedy School to go along with his law degree from Harvard. That made him conversant in quantitative analysis when the SCHR brought legal challenges to the underrepresentation of black jurors.

“The way the law is, we’d have to show an absolute disparity of over 10% before a court will see that as significant,” Stevenson said.¹¹⁹ “We’d routinely go to these counties where the county was 30% black, the jury pool would be 20%, if the county was 20% black, the jury pool would be 10%”¹²⁰ In any given jurisdiction, the disparity was not big enough for a defendant to mount a successful challenge, and yet it kept the absolute number of black citizens in the jury pool low enough that it was not hard for a prosecutor to eliminate the few black jurors using peremptory strikes.

Preserving the legal issue would also make the verdict of an all-white jury more vulnerable to reversal on appeal. Indeed, this frenzy of pre-trial activity created a general awareness among trial judges and prosecutors that legal errors might unsettle a conviction. This sensitivity improved advocates’ ability to “increase the numbers a little bit.”¹²¹ And that’s what they always wanted: a fighting chance in front of a mixed-race jury.

They also litigated jury demographic issues hard in Michael Berryhill’s case in Bartow County. Stevenson recalled that the jury commissioner had testified “they couldn’t get enough women on the jury because they had to be ‘upright and intelligent citizens,’ and we don’t have any of those or any black people.”¹²² However, the Georgia Supreme Court ruled on appeal that an 11% disparity between women in the county and in the jury pool was not “significant” enough absent evidence of “purposeful discrimination or systematic exclusion.”¹²³

They did not give up and introduced the same evidence as part of Berryhill’s federal habeas proceedings. This time, the Eleventh Circuit found that including only “intelligent and upright” women to the master jury lists was “highly subjective,” leading to severe underrepresentation.¹²⁴ A new trial was ordered because Berryhill had been denied a jury drawn from a fair cross section of the community.¹²⁵

C. Documenting Race-Based Jury Strikes

SCHR invested more and more resources into uncovering a related problem: proving that some Georgia prosecutors systematically used peremptory challenges to exclude black citizens from jury pools. Bright and his staff noticed that overzealous prosecutors would try to eliminate all, or nearly all black citizens from juries, even well into the 1980s and 1990s—especially when they wanted a death sentence against a black defendant. The way juries were composed, and the way prosecutors were using their largely unregulated strikes against black citizens in capital cases, seemed to be pillars holding up a system of unequal justice.

SCHR would eventually pile up victories on this front, not only in federal but state courts. Two of Bright’s later wins in the U.S. Supreme Court—Snyder v. Louisiana¹²⁶ and Foster v. Chatman¹²⁷—would involve reversals of death penalty convictions tainted by a prosecutor’s race-based peremptory strikes.

For now, though, they still had to deal with shifting legal standards. Until the Supreme Court made it easier to prove racial bias in peremptory challenges in Batson v. Kentucky,¹²⁸ the governing case was Swain v. Alabama,¹²⁹ which had established a standard that almost no defendant could overcome. In Swain, the justices rejected an Equal Protection claim even though no black citizen had served on a jury in a criminal trial for at least the prior 15 years. The only way for a defendant to prevail was to present proof that “the prosecutor in a county, in case after case, whatever the circumstances, whatever the crime and whoever the defendant or the victim may be, is responsible for the removal of Negroes who have been selected as qualified jurors by the jury commissioners.”¹³⁰

The battle over peremptory strikes played out in Georgia’s courts. Bright and Stevenson had gotten involved in another case that raised the jury composition issues—that of James Ford.¹³¹ Ford, a black man, was sentenced to die by a jury that contained only one black citizen. The prosecution used nine out of ten peremptory strikes to remove black jurors. At trial, the prosecutor had asked the trial judge whether he needed to explain his troubling pattern of strikes, but the judge said no, Swain did not require that, and simply overruled Ford’s objection. Swain itself had confirmed that “[t]he essential nature of the peremptory challenge is that it is one exercised without a reason stated, without inquiry and without being subject to the court’s control.”¹³²

The wrinkle presented in Ford’s case was a highly technical one: whether a defendant had lodged an objection in a timely fashion so that he could later raise the issue on appeal. In general, a failure to object barred a person from complaining about a trial-related error later. What made this case complicated was when the substantive law itself was in flux. What should be expected of lawyers during such a period of transition?

In Ford’s case, his lawyers had preemptively raised the equality issue by filing a pre-trial motion “to restrict racial use of peremptory challenges.”¹³³ They alleged that the county prosecutor had “over a long period of time” removed black jurors when a case involved a white victim.¹³⁴ After losing the motion, and receiving a death sentence, Ford again raised the issue, insisting that his Sixth Amendment rights had been violated.

While Ford’s cert petition was pending, the Supreme Court handed down Batson. If a defendant objected contemporaneously and could make out a strong enough showing of discriminatory conduct, a prosecutor would now have to give race-neutral reasons for exercising each peremptory challenge, and the judge had a duty to make a finding on the record as to whether to accept the explanations.¹³⁵ The new constitutional rule was far from perfect, for it was still hard to prove a nefarious motive on the fly, but it represented an improvement upon the earlier protocol.

The U.S. Supreme Court vacated Ford’s conviction and instructed the Georgia Supreme Court to review what happened under the new Batson test. But that court again upheld Ford’s death sentence, without conducting an evidentiary hearing or otherwise demanding that the prosecutor give explanations for his strikes as Batson required. In a Kafkaesque ruling, the Georgia Supreme Court simply ruled that Ford properly raised the Swain issue, which he had lost, but not Batson, which he was barred from raising now on appeal, deeming it too late.

Bright convinced Charles Ogletree of Harvard Law School to plead Ford’s cause before the U.S. Supreme Court on the return engagement. At oral argument, Ogletree hammered the inherent unfairness of the Georgia Supreme Court splitting hairs: “It is only after this Court’s remand, and without hearing from the parties, that the Georgia Supreme Court for the first time says that counsel has failed to make a record of this to survive a Batson claim.”¹³⁶ There were “no clear rules in existence” in the state courts governing when such an objection should be raised and “[t]o the extent that they were, counsel complied with them.”¹³⁷

Ford’s ordeal illustrated another facet of the legal system that could undermine the Constitution’s assurance of equality: the role of appellate judges. Whether out of indifference to rights or in defiance of a Supreme Court ruling, a state supreme court could exploit federalism to frustrate enforcement of constitutional guarantees. It could do so by applying state law broadly or by giving federal law a cramped reading. When that happened, a state court was basically daring a federal court to do something about it.

Indeed, one of the most egregious acts of state court defiance took place in 1955. That controversy also involved the Georgia Supreme Court asserting its prerogative to execute a black man despite an order by the U.S. Supreme Court to reconsider his constitutional rights.¹³⁸ Like Ford, the case also raised the problem of jury manipulation. Aubry Williams, condemned for killing a white liquor store clerk, appealed his case to the U.S. Supreme Court. Williams was tried by a Fulton County jury at a time when judges monitored the races of jurors by marking citizens’ names on different colored pieces of paper. In another case out of Fulton County called Avery v. Georgia,¹³⁹ decided just two months after Williams’ trial, the U.S. Supreme Court held that the county’s practice violated the Fourteenth Amendment. The problem for Williams was that his lawyers did not object at the time of his trial. His lawyers raised the issue for the first time in an extraordinary motion for new trial, but the state supreme court deemed the issue waived.¹⁴⁰ Under the “adequate and independent state ground” doctrine, a state’s ruling that an issue had been procedurally forfeited meant that a federal court had no jurisdiction to review it.¹⁴¹

When Williams sought certiorari, the Justices struggled with the possibility of letting a man die because of a racist jury selection procedure. On the other hand, rules were rules and federalism mattered. In an opinion by Justice Frankfurter, the Court vacated Williams’ conviction and invited the members of the Georgia Supreme Court to reconsider, appealing to their consciences that surely they would not “allow this man to go to his death as the result of a conviction secured from a jury which the State admits was unconstitutionally impaneled.”¹⁴²

Williams’s fate, as Ford’s would be a generation later, was trapped between evolving law and warring courts. In an angry opinion, the Chief Justice of the Georgia Supreme Court refused to give Williams a new trial and said that the U.S. Supreme Court lacked jurisdiction.¹⁴³ When Williams filed an emergency appeal, the Supreme Court backed down and refused to hear the matter a second time. That cleared the way for his execution.¹⁴⁴

Political scientist Del Dickson, who has studied that episode closely, believes that William’s rights were sacrificed on the altar of the Warren Court’s project to desegregate schools.¹⁴⁵ The Court was busy battling resistance to Brownacross the country and did not want to open another front on race matters. Forty years later, when the same court applied state rules to frustrate consideration of a constitutional right in Ford, the Supreme Court did not look the other way.

On February 19, 1991, Justice Souter wrote a unanimous opinion that repelled the Georgia court’s seeming defiance of Supreme Court precedent.¹⁴⁶ Ford’s attorneys had clearly raised a Swain claim. But in trying to draw a difference between these two cases, “the State assumes a distinction between the holdings in those two cases that does not exist,” he explained. “Both Swain and Batson recognized that a purposeful exclusion of members of the defendant’s race from the jury selected to try him would work a denial of equal protection.”¹⁴⁷

Georgia’s refusal to entertain Ford’s claim under the new standard amounted to a perverse form of logic, Justice Souter explained, because “Batson did not change the nature of the violation recognized in Swain, but merely the quantum of proof necessary to substantiate a particular claim.”¹⁴⁸ As a result, raising the peremptory strike issue was sufficient to preserve it, and the state court’s application of a new procedural rule announced two years later was not an “adequate and independent state ground” to prevent federal review.

To Bright, the lesson from handling Ford’s appeal was that they would have to battle appellate judges as much as prosecutors to ensure that basic rights would be enforced. Even so, their intensification of trial tactics proved to be worth it. Ford’s lawyers had earned him the right to a hearing. Once a hearing was finally held in state court, the original prosecutor was forced to testify as to his reasons for striking so many of the eligible black jurors. The trial judge summarily accepted those reasons, but on appeal, the Georgia Supreme Court reversed, building on Gamble v. State, another SCHR case and the very first time a Batson violation had been found by the state supreme court.¹⁴⁹

Chief Justice Norman Fletcher ruled that the district attorney’s explanations offered after the fact were not sufficiently “concrete” and “race-neutral” so as to overcome the “strong” pattern of discrimination in Ford’s case.¹⁵⁰ As he pointed out, “of the 42 persons on the panel from which the trial jury was chosen, ten, or 24%, were black.”¹⁵¹ In other words, the prosecutor “exercised 90% of his strikes to strike 90% of the blacks from the venire, while exercising 10% of his strikes to exclude a mere 3% of the whites on the venire.”¹⁵² Quoting Ford’s brief, Justice Fletcher said that “ ‘it does not require a statistician … to recognize’ the very high probability that this racial disparity did not occur strictly as a matter of chance.”¹⁵³ Against the weight of this pattern, the prosecutor’s explanations “fall short of proving that this disparity was the incidental result of neutral selection procedures.”¹⁵⁴

After losing in the U.S. Supreme Court and seeing the Georgia Supreme Court’s about-face, prosecutors folded and offered Ford a life sentence. Indeed, perhaps more than any other constitutional issue, a prosecutor’s use of peremptory strikes to prevent black citizens from participating in criminal trials had the potential to bring together judicial liberals and conservatives.¹⁵⁵ Not only did the rationale appeal to those who wished to preserve the democratic legitimacy of criminal judgments, it also appealed to those who might be open to upsetting a single criminal conviction without necessarily embracing structural critiques. For conservatives, deciding a case on this ground meant protecting the integrity of the legal system and dealing with “bad apples.”¹⁵⁶ Of course, judges could not completely control how advocates would characterize the violations that were discovered, and abolitionists tended to declare such misconduct evidence of structural flaws in the justice system.

D. Holding Prosecutors Accountable

After prevailing in the Georgia Supreme Court and being reinstated as counsel for Amadeo’s retrial, Bright and his co-counsel filed a giant pile of motions. The most important one sought to recuse Briley from leading the state’s case a second time.¹⁵⁷ This, too, became a part of SCHR’s arsenal whenever there was reason that a prosecutor had been involved in an equality violation. Rarely had there ever been any kind of consequences for such constitutional wrongdoing, whether professional, financial, or otherwise. Bright and his staff pitched their effort to block constitutional violators from (ab)using their power against the same defendants as a modest demand. But embedded in the move was a radical logic that state discretion was founded upon trust and good faith—notions that were fundamentally breached when a prosecutor acts in unequal and antidemocratic fashion by suppressing minority representation on juries.

Amadeo’s team also arranged for legal ethics experts to file a brief in support of the recusal motion. The signatories came from each of the state’s accredited law schools, including Mercer Law School, where Briley had gotten his J.D. In their brief, these experts argued that Briley’s “misconduct is shocking to the conscience” and violated his duty to ensure that “all citizens” in the community “stand on an equal footing before the law.”¹⁵⁸ Friends of the court cited Blackstone for the proposition that “special vigilance is required to keep our system of trial by jury ‘sacred and inviolate, not only from all open attacks, . . . but also from all secret machinations, which may sap and undermine it.’”¹⁵⁹ Invoking ethical standards created by the National District Attorneys Association as well as ABA Standards on the Prosecution Function, they observed that “the duty of the prosecutor is to seek justice, not merely to convict.”¹⁶⁰

Amici thus supported SCHR’s position that qualification of Briley was “required to protect the integrity of the adversary system, to assure public confidence in the administration of justice, to assure that the past misconduct does not interfere” with Amadeo’s retrial.¹⁶¹ The prosecutor’s misconduct presented “that most rare case,” they argued, “where there is a factual determination upheld unanimously by the United States Supreme Court that a prosecuting attorney intentionally tampered with the jury pools in an effort to subvert the constitutional rights both of the black jurors, and of Tony Amadeo, the litigant.”¹⁶² Additionally, the manipulation of jury demographics required legal expertise as part of the “intent to deceive” so that “a prima facie case of discrimination could not be established.”¹⁶³ Such misconduct also violated bar standards that prohibit “illegal professional conduct involving moral turpitude,” “dishonesty, fraud, deceit, or misrepresentation,” or behavior “prejudicial to the administration of justice.”¹⁶⁴

On the day of the hearing, witnesses were prepared to testify that Briley’s continued participation in Amadeo’s case would taint the outcome.¹⁶⁵ Bright had assembled Tom Johnson, county attorney for Hennipen County, Minnesota, Stephen H. Sachs, the former Attorney General of Maryland, and Professor Bennett L. Gershman, a leading expert on prosecutorial ethics. In the glare of such tactics, the state blinked. Rather than go through with a full evidentiary hearing, Briley offered a plea deal. He would withdraw the death penalty if Amadeo pleaded guilty to murder and promised not to seek parole for at least 25 years.¹⁶⁶

SCHR would continue to try to recuse prosecutors whenever they had proof someone was involved in a Fourteenth Amendment violation, particularly on the basis of race or poverty. And in the retrial of Brooks, they stepped up this tactic by enlarging their recusal motion against an entire District Attorney’s office, attacking its death penalty charging decisions and use of peremptory strikes over time as racially discriminatory.

E. Bouncing Judges

The first time Bright and his staff went hard after a trial judge believed to be “a hanging judge” was in the case of George Dungee, a Black man who had an IQ of 65-69. Dungee was one of three men who escaped from a Maryland prison and went on a crime spree. Seminole County juries tried Carl Isaacs, Wayne Coleman, and Dungee separately and sentenced them to death for the horrific slaying of a family.¹⁶⁷

Hanging judges were those known as tough-on-crime figures who might look the other way as prosecutors cut corners, including by manufacturing all-white juries in capital cases. Typically, they were former prosecutors themselves, elected or appointed after securing death sentences for notorious crimes. Having ridden to glory for their pursuit of Old Testament-style justice, they often looked with favor on other prosecutors who wanted a similar career arc.¹⁶⁸ Walter C. McMillan had a reputation as a hanging judge.

After Dungee’s conviction was overturned due to the trial judge’s failure to deal with prejudicial pre-trial publicity, McMillan was assigned to oversee the retrial.¹⁶⁹ In 1987, when Dungee’s case returned to Seminole County, Bright and his team moved to prevent Judge McMillan from presiding over the case, alleging “invid[i]ous discrimination against poor people and black people.”¹⁷⁰ Although the issues were “somewhat painful” and “delicate,” it was necessary to probe the judge’s beliefs and past actions. Because this was “a racially sensitive” death penalty trial, they argued that it was “inappropriate for him to sit on this case.”¹⁷¹ Bright alleged that “George Dungee was called a ‘n*gger’ in Open Court in his last trial” and that nothing had been done about it.¹⁷²

Dungee’s lawyers also pointed to Judge McMillan’s membership in an all-white country club, which they insisted violated the canons of judicial conduct, as well as his past political support for Ernest Vandiver, who ran for governor in 1959 on a platform of “No, Not One”—meaning not one black child in white schools, which they believed showed the judge was prejudiced. As Bright also noted during the recusal hearing, McMillan had never appointed a non-white jury commissioner for the counties he oversaw, “not a single time,” in over twenty years on the bench. These commissioners exploited discretionary state law to underrepresent black people on juries, despite the fact that two counties within his jurisdiction were majority-black.¹⁷³ As a result, Bright argued, Judge McMillan was “very much a part of that system of white dominance.”¹⁷⁴ Indeed, to bolster structural arguments of bias, Bright insisted that the judge had helped perpetrate “systematic [mis]treatment of indigent defendants” and black people, and not merely isolated acts of discrimination.¹⁷⁵

During the hearing, Bright called Judge McMillan himself to the stand and asked about his use of racial epithets—which the judge acknowledged under oath.¹⁷⁶ Questioned about his views about Brown v. Board, Judge McMillan testified that “he [has] no quarrel with it” now, but “don’t recall what I thought then,” though “it was controversial.”¹⁷⁷ On the stand, he acknowledged that he supported Vandiver for governor, and that the candidate “was espousing a racial segregation in the schools.”¹⁷⁸ Dungee’s lawyers also alleged that McMillan had withdrawn his daughter from public school out of opposition to integration, but could not get him to admit this in court.

Despite a somewhat mixed hearing record, a surprising thing did take place that altered Dungee’s fortunes. Bright’s aggressive tactics baited the judge into committing a misstep. Judge McMillan became so enraged at his impartiality being questioned that he retained his own lawyer for the hearing on the motion to recuse (heard by a different judge). Through this lawyer, he tried to intervene in the proceeding, monitor what was going on, and influence its outcome by conferring with attorneys for the state. McMillan claimed that he was “a participant” to the proceedings because “it’s my duty and responsibility to defend the propriety of my office.”¹⁷⁹

That gambit ultimately failed when the Georgia Supreme Court decided the interlocutory appeal on the recusal motion in Dungee’s favor. Without ruling on the ultimate issue of bias, and without mentioning “race” or “discrimination” a single time in the opinion, the justices seized on McMillan’s aggressive efforts to protect his own reputation as evidence that his “impartiality might reasonably be questioned.”¹⁸⁰

Once Judge McMillan was off the case, a deal was reached to take the death penalty off the table. Dungee pleaded guilty to three consecutive life terms.¹⁸¹

SCHR pulled the same move in the retrial of Willie Gamble. Judge McMillan had presided over Gamble’s original murder trial, where prosecutors used all ten of their peremptory strikes against black jurors to create an all-white jury. That jury sentenced Gamble to die, but the Georgia Supreme Court, citing Batson, reversed.¹⁸²

Bright, Bryan Stevenson, and Clive Stafford Smith¹⁸³ did not want McMillan anywhere near the retrial. They moved to recuse him from overseeing Gamble’s case too, arguing that his previous convictions were reversed “due to violations of his constitutional rights, resulting from intentional, flagrant discrimination by the district attorney, which was routinely approved by Judge McMillan.”¹⁸⁴ In the very first paragraph of their motion, they invoked McCleskey as a command that “the person on trial must show racial discrimination in the locality and the court in which he was tried.”¹⁸⁵ Judge McMillan was just such a “decision maker” within the meaning of that case, they wrote, whose “racially discriminatory acts and encouragement and support of the racially discriminatory acts” of others already violated Gamble’s rights and would likely do so again.¹⁸⁶ The unequal imposition of the death penalty “is the result of discrimination by the District Attorney in deciding in which cases he will seek the death penalty,” they insisted.¹⁸⁷

Gamble’s lawyers turned up the rhetoric further, calling Judge McMillan a “rubber stamp racist,”¹⁸⁸ and recited a litany of facts documented earlier in the Dungee case. This time, they also argued that Judge McMillan had demonstrated lingering “antipathy” toward Bright’s staff because of “his public humiliation in the Dungee case” at the hands of SCHR lawyers.¹⁸⁹ Judge McMillan initially refused to appoint Stafford Smith to Gamble’s case for the retrial, and instead came up with a plan to solicit bids to handle the defense and appoint the lowest bidder. SCHR accused McMillan of creating “a scheme to sell off justice to the lowest bidder,” and called it further proof of his bias.¹⁹⁰ The judge had even called up two lawyers and begged them to take the case, which Bright believed was an effort to “preclude Mr. Gamble from raising the issues of judicial racism at the retrial” and block him from “moving to recuse Judge McMillan for bias.”¹⁹¹

Remarkably, Gamble’s defense team also researched Rule 11 sanctions to see how far they might go in federal court before they landed in hot water. According to Stafford Smith, what prompted this concern was that the team had drafted a civil RICO lawsuit against the judge, where violations of civil rights laws would serve as the predicate offenses. In the end, they wisely opted not to file it.¹⁹²

At all events, the full-court press via state court proceedings turned out to be enough. Judge McMillan eventually stepped off the Gamble case. The prosecutor, seeing that Judge McMillan would not oversee the retrial, eventually resolved the case with a plea deal guaranteeing a life sentence.

III. MCCLESKEY AS A SWORD

Although they publicly sought to deny McCleskey moral legitimacy and constitutional validity outside of the courtroom, inside the courtroom SCHR lawyers deployed the decision as a sword. They did so in several ways: (1) to justify their efforts to build a record of discrimination by the local power structure on the basis of race and poverty of the defendant and social worth of the victim; (2) to expand their structural critique of unequal justice to entire criminal legal systems on a county and/or judicial circuit basis; (3) to piggyback on those bias claims and request evidentiary hearings and documents so as to counteract the position of information disadvantage experienced by criminal defendants; and (4) to justify recusal of any judges and prosecutors found responsible for unconstitutional or unethical behavior.

Uncertain of their chances of success or when their window of opportunity would close, SCHR lawyers pushed their “McCleskey motions” as hard and as far as they could to maximize the probability of judicial consideration of their motion, pressure the state’s lawyers to rethink their choice to pursue a death sentence, and to draw attention to issues of race and poverty.

A. Horton v. Zant

After Amadeo’s case, SCHR lawyers learned to build on findings of prosecutorial misconduct to go after DA Joe Briley’s other cases. They decided to represent Jimmy Lee Horton, a black man under a death sentence who was also prosecuted by Briley. This time, Bright and his staff found an ingenious way to go after Briley’s use of peremptory challenges to remove prospective black jurors.¹⁹³

Horton and his co-defendant, Pless Brown, Jr., had been arrested for burglarizing an apartment and killing the companion of the homeowner during a shootout when the two came home and surprised them. The dead man turned out to be Don Thompson, the District Attorney of Macon Judicial District, Georgia.

Joe Briley was brought down specially from Okmulgee Judicial District to secure death sentences against the perpetrators. Brown, who wound up with a mixed-race jury, received a life sentence. By contrast, Horton was sentenced to die by an all-white jury. His court-appointed lawyers did no investigation and presented no mitigating evidence during the sentencing phase. In closing remarks to the jury, one of his attorneys stated that his own client was “a worthless, despicable human being” and praised the prosecutor’s argument.¹⁹⁴

Because Briley had such a long track record of aggressively seeking death sentences and scoffed at charges that capital punishment practices were racist, SCHR brainstormed ways to dig more deeply into his jury strikes over the course of his career. Their success in doing so created a template for action in subsequent cases.¹⁹⁵

1. A Mini-McCleskey Hearing

Judge Wilbur Owens of the Northern District of Georgia was assigned to review Horton’s habeas petition—the same judge who had granted relief in Amadeo’s case based on the jury-rigging violation. Judge Owens scheduled an evidentiary hearing on Horton’s petition and ordered the state to engage in “expeditious discovery.”¹⁹⁶ At that time, before enactment of the AEDPA,¹⁹⁷ it was still possible to get evidentiary hearings on habeas petitions. Only later would federal judges increasingly deny claims based on the papers alone.

On October 18, 1989, Bright handled the live hearing.¹⁹⁸ He called Briley to the stand to authenticate the jury-rigging memo. The memo, which had led to Amadeo’s conviction being overturned in the U.S. Supreme Court for jury manipulation,¹⁹⁹ now became evidence in Horton’s case, as Bright argued that the document ought to be treated as proof that the prosecutor had a history of racially discriminatory behavior. Bright wanted the judge to infer that Horton’s case, too, was tainted by Briley’s participation.

McCleskey v. Kemp hung over every facet of this pre-trial hearing. “The issue is who is the person behind this discrimination?” Bright reminded.²⁰⁰ He was concerned about Justice Powell’s insistence in McCleskey that legal theories of inequality must be traced to specific perpetrators. If the proof of discrimination was not in the record, if the judge did not find that someone in particular was responsible for purposeful discrimination that occurred in his client’s case, then Horton could not benefit even if there were racial disparities.

The prosecutor’s past misconduct was also useful for attacking his credibililty in other settings. “Your Honor, Mr. Briley in case after case was resisting jury challenges, taking advantage of the fact that people weren’t prepared for jury challenges,” Bright argued.²⁰¹ “[A]ll of that flies in the face of this testimony that in this one particular case he was advising the judge that he ought to put the full number of blacks and women in the pool.”²⁰²

“You’re not going to win or lose based upon that one incident,” Judge Owens said. “The pattern of years, all these cases” was what mattered.²⁰³

Bright turned to a study compiled by SCHR staff on Briley’s juror strikes over his career. In the months leading up to the hearing, the staff hunted down old case files, trying to reconstruct the twists and turns of jury selection in the cases tried by Briley across eight different counties since 1974. Three of the counties did not retain the records, but they successfully charted 25 capital and 159 non-capital cases of Briley’s. They then had to cross-reference that information with county voter registration lists to identify each juror’s race.²⁰⁴

Displaying the document in his hand, Bright asked Briley: “In 1977 you also tried the Walker case in Jasper Country . . . And in that case you used seven of your peremptory strikes against seven black people; is that correct?”

“If that’s what the record shows,” Briley shrugged.

“You’re not able to recall any reasons why you struck those people?”

“No sir,” Briley responded.

Bright said to Briley: “[I]t is your testimony that the fact that all seven happened to be black people is just coincidental?”

“Yes sir.”

“In the Finney case in Jones County you used seven of your strikes against black people?” Bright inquired. “Are you able to recall any of the reasons for that?”

“No sir,” Briley testified.²⁰⁵ His strategy as a witness was to deny any memory of why he rejected so many black citizens as prospective jurors and hope that the silence in the record would be enough to satisfy tolerant judges.

After leading him through the exhibit and getting Briley to admit to the disturbing pattern of decisions to excuse black jurors from his trials over many years, Bright got Briley to admit that he used nine out of ten strikes against black jurors in Brown’s case. And in Mr. Horton’s case, he also used nine out of ten strikes against black jurors.

On cross-examination, the state’s lawyer, Paula Smith, asked Briley to describe the process of jury selection in Georgia. He explained that “[y]ou’ve got to have all twelve jurors to vote for the death penalty because if one of them votes against the death penalty . . . [i]t is a life sentence. So it takes a perfect score to get a death penalty.”²⁰⁶ But his response seemed only to underscore the connection between the jury’s demographics and his prediction of trial outcomes rather than rebut the strong pattern of apparent discrimination to achieve a desirable jury composition.

On redirect, Bright tried to underline the racial patterns. “You used ninety-six of your hundred and three strikes, ninety-four percent of your strikes, against black people. Do you know why these reasons you have tend to fall so heavily upon black people?” Briley could not provide an explanation.²⁰⁷

When Briley had a case with a black defendant and black victim, he exercised seventy-three of eighty-six jury strikes against black jurors. Yet when it came to cases involving white defendants, the prosecutor did not exhibit the same enthusiasm for excluding black jurors. In such cases, Bright pointed out, “you only used a third of your strikes against blacks.”

Next, Horton’s lawyers called Gary Liberson, a PhD in mathematical statistics to the stand.²⁰⁸ Liberson testified that he examined Briley’s juror strikes over many years on a county-by-county basis to see whether they were “consistent with the ratio or percentage of blacks in the population for that particular county.” What was the probability that Briley’s decisions reflected “just the random selection of jurors” who “just happened to be black?” He answered that “there was about one chance in a hundred thousand that that would have been so.”²⁰⁹

As to Briley’s comparatively less vigorous use of juror strikes in cases with white defendants, it was also roughly “one chance in a hundred thousand” that the pattern was a random one. On cross-examination, the state’s attorney got Liberson to admit that “the most your analysis would reveal is that the numbers do not show what one would expect if the prosecutor struck at random.” She tried to suggest that any number of subjective factors not considered by the expert could explain the disparity. In response, Liberson said that even if there were some other set of components that were strongly associated with race, “they would dampen the results” but not “make the results non-significant.”²¹⁰

The state’s lawyers called to the stand Joseph Katz, the very statistician Georgia had used in the McCleskey case to rebut the Baldus study. Katz disputed Liberson’s conclusions but did not offer any alternative study or explanation, saying “some things don’t easily reduce to numbers.” Under cross-examination, he conceded that Liberson’s methods were “appropriate” and that he did not try to replicate Liberson’s findings. He also acknowledged the probability of Briley’s pattern of strikes being random was “very low.” In fact, he agreed that Liberson’s finding of racial disparity to be “statistically significant” because “[i]f you were a black defendant Mr. Briley was more likely” to strike black jurors.²¹¹

After all the witnesses and exhibits were entered into the record, Bright addressed Judge Owens. What the evidence showed was that “Mr. Briley has a remarkable history of striking black people from juries.” In capital cases “[N]inety percent of the time he used his strikes against black jurors.” The “only time he changes,” Bright argued, “is when he has white people on trial.”²¹² Summing up, Bright urged Judge Owens to find that when he picked his client’s jury, he “deprived members of Mr. Horton’s race of participation in the judicial process.”²¹³

During her remarks, Smith acknowledged that Horton had shown that the state “has struck a higher proportion of black jurors than white jurors,” but said that still did not violate the Constitution. The petitioner’s burden under Swain was to prove “historic systemic exclusion,” she insisted.²¹⁴

On April 12, 1990, Judge Owens denied the writ. On the most promising claim, Judge Owens concluded that Horton did “not show that the prosecuting attorney has historically and systematically excluded blacks from serving on trial juries.” He made quick work of Horton’s remaining claims, leaving his conviction undisturbed.²¹⁵

2. The Eleventh Circuit Validates Rebellious Localism

Horton’s Eleventh Circuit brief put the jury issue front and center. “Swain does not require a defendant who has been victimized by the prosecutor’s racially motivated abuse of peremptory challenges to show . . . that the prosecutor had always successfully obtained all-white juries in the past,” his legal team argued. Instead, “the central question is purposeful discrimination.” And on that issue, the actual “history of Mr. Briley’s practices reveals his intent and strips his anti-black peremptory challenges at Mr. Horton’s trial of any presumption of propriety.”²¹⁶

On September 3, 1991, the Eleventh Circuit issued a unanimous ruling reversing Judge Owens.²¹⁷ Judge Frank Johnson first took up the issue of peremptory challenges. Although the Supreme Court in Batson made it easier for a defendant to challenge jury strikes, Horton could not rely on that decision because his trial took place before that case was decided. Instead, he was bound by Swain, in which a black man was convicted of rape by an all-white jury after a prosecutor used his peremptory challenges to remove all the black citizens from the pool.²¹⁸ Swain also showed that there had never been a single black person allowed to serve on a criminal trial in Talladega County, Alabama since 1950, even though the county was 26% black.²¹⁹ Yet the fact that the jury panels were consistently underrepresented at 10–15% was deemed by the Swain Court not enough to make out a violation of the Equal Protection Clause. No one is entitled “to demand a proportionate number of his race on the jury which tries him,” Justice Byron White wrote.²²⁰ A defendant is entitled only to a trial free of intentional discrimination.

Justice White ruled in Swain that the fact that in a single trial “all Negroes were removed from the jury” was not enough to overcome the presumption that a prosecutor is doing so “to obtain a fair and impartial jury.” The only way to strip a public servant of the presumption of good faith was to present proof that “the prosecutor in a county, in case after case, whatever the circumstances, whatever the crime and whoever the defendant or the victim may be, is responsible for the removal of Negroes who have been selected as qualified jurors by the jury commissioners.”²²¹

Over the years, Eleventh Circuit applied Swain, already a demanding standard, in exacting fashion. As Judge Johnson pointed out in Horton’s case, the court had in the past rejected claims even when a prosecutor struck every black juror in six trials during a single week.²²² Luckily, Horton’s lawyers had met that high burden. “Mr. Briley’s hardball tactics clearly do not comport with the prosecutor’s obligation to ‘do justice,’” Judge Johnson admonished. It was the first time the Eleventh Circuit had ever found a Swain violation in a Georgia case.²²³

Citing Amadeo, Judge Johnson also emphasized Briley’s role as “author of a now infamous memo designed to underrepresent blacks, women and all individuals 18-24 years old” in Putnam County’s juries. As for Briley’s new tale “that he should not be morally culpable” because a now-deceased judge had instructed him to manipulate the jury pool, “even if we were to assume Mr. Briley’s version of the story was true, we cannot condone his behavior.” The “mere fact that a judge orders a prosecutor to engage in unconstitutional discrimination on the basis of race and sex does not make the behavior right.” He added: “A prosecutor has a duty to ‘do justice’ and . . . not engage in the subsequent cover up.”²²⁴

After pointing to evidence of Briley’s past misconduct, the Eleventh Circuit rejected the state’s argument that Swain’s strong presumption of regularity could not be defeated absent smoking-gun evidence in the prosecutor’s notes or damning testimony. The panel then turned to the pattern of Briley’s jury strikes over time uncovered by SCHR investigators. Judge Johnson embraced the weight of their findings: “[B]etween 1974 and 1981, Mr. Briley exercised 1,580 peremptory strikes” across three counties, 70% of them against black jurors. In capital cases that percentage ticked even higher: 79% to remove black jurors without explanation (184/234 strikes). When a murder trial involved a black person charged with killing a white person, Briley exercised 96 out of 103 strikes against black jurors. The panel was impressed with Dr. Liberson, twice citing his findings that the likelihood Mr. Briley exercised his peremptory challenges for race neutral reasons “was less than 1 chance in 100,000.”²²⁵ The panel made no mention of the state’s expert, rejecting his testimony implicitly.

Judge Johnson dismissed the state’s explanation for these racial patterns: “courts cannot make a blind leap of faith that there exists some set of legitimate factors which correlates one-to-one with the race of the venire members.” But even a prosecutor who acts with a “mixed motive”—where race is one among other factors in striking black jurors—is violating the Constitution.²²⁶ “From Horton’s statistics,” the judges said, “we cannot help but conclude that race was a very significant factor in Mr. Briley’s decision-making process.”²²⁷ Racial discrimination was believable in this context in part because, as the panel observed, “in our society interracial crime is treated differently than other crimes.”²²⁸

A prosecutor’s past violation of the Equal Protection Clause in one case finally had actual consequences in subsequent cases. Moreover, the mini-McCleskey strategy had worked. Despite McCleskey, or ironically perhaps because of it, the Eleventh Circuit embraced statistical evidence presented on a smaller scale. It’s not obvious that the statistical evidence was more scientifically sound than the Baldus study, but the bite-sized sample, involving one person in power across time, appealed to judges’ desire for simpler stories about causation and evidence about racial discrimination that could be attributed to specific individuals. Doubling down on equality, subverting doctrine, and scaling down cause lawyers’ targets had paid off.

This was a resounding legal victory that exposed prosecutorial misconduct, racial inequality, and the subpar quality of justice for poor Southerners. The media called it “yet another rebuke” of Briley “for keeping blacks off juries.”²²⁹

Faced with the prospect of having to retry the case after all these years, Briley agreed to a plea deal that assured a life sentence for Horton. Success in Amadeo had created legal precedent on trial fairness and racial equality, and that precedent, in turn, had been successfully leveraged for Horton. Additionally, the Eleventh Circuit’s surprising ruling now gave cause lawyers a useful weapon. In theory, Putnam County’s jury-rigging practices could be leveraged to force a second look at other cases—if the issue had been preserved or else new evidence can overcome procedural obstacles.

B. State v. Brooks

William Anthony Brooks, an African American man, was charged with the abduction, rape, and murder of a white woman, Carol Jeannine Galloway. The victim’s mother saw her climb into Brooks’ car in front of her home, and her body was later discovered behind an elementary school. An all-white jury in Columbus, Georgia sentenced Brooks to die in the electric chair after Mullins Whisnant, Pullen’s predecessor, told jurors that Georgia’s crime rate had risen since 1964, the date of the last execution, and that “[w]e didn’t have this kind of murder . . . when we had capital punishment.”²³⁰

At sentencing, Whisnant invoked “the war on crime” as a reason to deal with Brooks harshly and told jurors that it would be more expensive for taxpayers to imprison Brooks than kill him. Whisnant urged jurors to think of themselves as “soldiers” in the battle against a crime wave: “[Y]ou can tell William Brooks, and you can tell every other criminal like him, that if you come to Columbus and Muscogee County, and you commit a crime . . . you are going to get the electric chair.”²³¹

The Eleventh Circuit found these tactics disturbing. A panel called out the prosecutor for inviting the jurors to “forego an individualized consideration of Brooks’ case and instead choose execution merely because he was part of the broad ‘criminal element’ terrorizing American society.”²³² Bright and Kendall, who took on Brooks’ appeal, emphasized Whisnant’s improper political rhetoric inside the courtroom, but the Eleventh Circuit ultimately found these remarks amounted to “harmless error” because of the overwhelming evidence of guilt. Nonetheless, the panel overturned Brooks’ conviction for a different reason: the trial court mistakenly instructed jury members that they could presume every murder to be “malicious.” Doing so improperly shifted the state’s burden of proof to the defendant.²³³

Brooks had been tried in 1977 before an all-white jury and sentenced to death even though black people made up 30% of the community. Bright believed that race had played a significant role in Brooks’ case—and not just because of Whisnant’s conduct. He felt that John Henry Land, the judge who presided over Brooks’ trial, gave him two uncaring lawyers who did little to protect him from the state. Land’s father, Aaron Brewster Land, had also participated in two lynchings in the area when John Henry was just a boy. First elected in 1964, Judge Land himself publicly acknowledged long-held segregationist views. He had even belonged to the States’ Rights Council of Georgia, Inc., organized to preserve “the traditional establishment of segregation in both public and private places” and oppose Brown v. Board, which the group derided as “the Thurgood Marshall Plan,”²³⁴ though Land later renounced his past support for segregation.

His conduct as a proud “hanging judge,” particularly in high-profile criminal trials, suggested something different. The more that Bright learned about Judge Land’s handling of the first trial, the more he became convinced that Judge Land had helped carry out a “legal lynching” of Brooks.²³⁵ Over the objections of the jurors, Land had permitted cameras in the courtroom for the sensational cross-racial murder trial—a first in Georgia’s history.²³⁶ Just by reading the transcripts of jury selection, Bright could guess the race of the jurors. “I could tell by the way they treated those eight people which ones were the blacks because they would lead them into answering the questions in a way that got them excluded from the jury,” Bright recalled.²³⁷

During voir dire, prospective jurors who acknowledged hearing about the high-profile killing in the news were asked follow-up questions. Judge Land and the prosecutors would lead the white prospective jurors into saying that they could put aside detailed facts they already knew about the case and decide the case based on the evidence. With black jurors, they would lead them into saying they had already formed an opinion about the case, even when what they remembered from pre-trial coverage was vague. Judge Land’s supervision of jury selection in Brooks’ original trial left few black jurors in the pool, who were easily removed by prosecutors through peremptory challenges.²³⁸

1. Recusing All the Judges and Prosecutors of a Circuit

Bright and Kendall remained counsel for Brooks’ retrial, adding recent Yale Law School graduate Ruth Friedman²³⁹ as well as Columbus attorney and state senator Gary Parker. Because Columbus had the “dubious distinction” of being an outlier—having “condemned more people to death since 1973 than any other place in Georgia,”²⁴⁰ Bright felt that they would have to escalate their tactics to match county officials’ strong push for the death penalty. When the case returned to Muscogee County, they filed a 23-page motion to recuse Judge Land because of his “intentional racial discrimination” and requested that the case be assigned to a judge outside of the Chattahoochee Circuit.²⁴¹ Brooks’ lawyers added a claim about “bias[] against poor persons,” citing the judge’s own public comments calling indigent people in his courtroom who needed counsel “freeloaders.”²⁴²

Although he claimed it had nothing to do with the substance of the motion, Judge Land eventually stepped aside, saying that “the fact that I did try it before might give an appeals court something to hang their hat on.”²⁴³ On March 24, 1988, Bright then moved to recuse Judge Rufe Edwards McCombs, the next judge assigned to Brooks’ case, and renewed the defense request that the case be moved outside the circuit entirely.²⁴⁴ In that same motion, they insisted that none of the judges in the circuit could be impartial given that under McCleskey v. Kemp their client “must show the discriminatory intent of all key ‘decision makers’ ” and that the key figures “in this case are friends, colleagues and associates who are a part of the Court and judicial circuit” on which Judge McCombs and others serve.²⁴⁵ Judge Land, the chief judge of the circuit, “both participated in discrimination against Mr. Brooks and allowed it to take place before him in the previous prosecution of this case.”²⁴⁶ Furthermore, all of these judges regularly conferred with Judge Whisnant, who originally prosecuted the case against Brooks. Bright insisted “that as district attorney Mullins Whisnant discriminated against black people in his decisions to seek the death penalty and that he sought the death penalty against William Brooks as part of those discriminatory practices.”²⁴⁷ Now that Whisnant was a judge on the circuit, his colleagues could not be trusted to handle accusations of racial bias against him impartially. The trial should be moved or Brooks should be given an out-of-circuit judge.

The defense also signaled that they would attempt to show that the current District Attorney William Smith, who happened to be “a candidate to join Judge McCombs on the bench,” had merely “continued the discriminatory practices of his predecessor,” as assistant district attorneys “have sought to exploit these discriminatory practices to get the white vote in the current and past elections.”²⁴⁸ Whoever ultimately presided over the trial would also have to evaluate evidence that other officials and employees of Muscogee County “have conspired to deny black people full-fledged participation on jury commissions, on the master jury lists, on grand and traverse juries, and in positions of employment in the clerk’s office, the offices of the judges . . . resulting in discrimination against black defendants who come before the Court.”²⁴⁹ Bright argued that “[t]he appearance of a conflict of interest will be created” when Judge McCombs—or frankly any other judge who worked closely with Judge Land in the circuit—is called upon to determine Mr. Brooks’ McCleskey claim.”²⁵⁰

Brooks’ lawyers added another motion demanding a public hearing on their recusal motion, as well as discovery of any evidence in the possession of the District Attorney as well as any judge or court personnel.²⁵¹ Bright repeatedly deployed McCleskey offensively: “Where the state seeks to utilize society’s ultimate sanction, it has an overriding obligation to provide a fair forum. In the vernacular of McCleskey, its decisionmakers must stand beyond repute. Mr. Brooks has filed a motion claiming that Muscogee county decisionmakers fail this test.”²⁵²

The next step involved trying to hold the district attorney’s office responsible for past discrimination by blocking their participation in subsequent cases. Seeking maximum media coverage of their racial bias claims, Bright followed up with a motion to disqualify the entire Muscogee County District Attorney’s office.²⁵³ “The current occupant of that office is one whose political identity is based almost exclusively on the death penalty,” the motion stated. “The death penalty was the major issue in his campaign for District Attorney.”²⁵⁴ That public posture meant that prosecutors were violating the “due process and other fair trial rights” of the accused by failing to consider legal reasons for whether to seek the death penalty—that is, “the unique circumstances of the offender and the crime.”²⁵⁵ In fact, whether out of improper “personal or political” interests or “loyalty” to one part of the community over another, the district attorney “has refused even to consider a sentence of less than death in this case.”²⁵⁶

It was a long shot, but Judge Whisnant, who handled the case when he was a prosecutor, ultimately decided to recuse himself. Eventually, Brooks’ team got what they wanted: the case was assigned to Judge Hugh Lawson from the Oconee Judicial Circuit.²⁵⁷ That key switch in judicial personnel opened the possibility for the defense to ratchet up rebellious localism even further.

Bright and his staff immediately moved to quash the original indictment against Brooks from 1977, arguing that it was still tainted by the constitutional defect identified in the Amadeo litigation. The district attorney wisely chose not to contest that motion. Instead, he resubmitted the case to a grand jury in 1987 and received another indictment. Brooks’ lawyers then objected to the new grand jury proceeding as “fundamentally unfair” because District Attorney Doug Pullen gave a campaign-style speech during his presentation.²⁵⁸ This motion was denied.²⁵⁹

2. Structural Racism Claims

Before a new judge, Brooks’ team filed multiple motions to block the death penalty, making a variety of equality arguments. All told, Bright explained to Judge Lawson, “we come out with about fifteen issues approximately that are worthy of the Supreme Court of Georgia looking at.”²⁶⁰ In late September 1990, Judge Lawson wavered, asking Pullen whether he “feel[s] so absolutely certain about the State’s legal position on all of these issues that you are prepared to spend the time and the money to have the trial and run the risk that any of these could make the train jump the track when the trial is over?”²⁶¹

One defense motion sought “to bar the death penalty because of racial discrimination.”²⁶² A second sought to “bar the death penalty because of victim-impact evidence.” A third motion tried to stop capital punishment from being inflicted because it “is sought based on worth and status of victim.” A fourth asked for a preliminary ruling to “bar any prosecutorial misconduct . . . that has been sort of a hallmark of capital cases in the Chattahoochee Circuit.”²⁶³ This last motion did not just try to stop certain kinds of racist litigation practices during Brooks’ retrial, but also sought to bar all future capital prosecutions in the jurisdiction. According to Bright, the constitutional violations ran deep:

[Y]ou can take the first capital case that Judge Whisnant tried when he was district attorney and you can look at the closing argument in that case, which was basically the kind of argument you would make on the courthouse stairs at a lynch mob, not the kind you make in a courtroom to a jury. And you can look [at] an argument Mr. Pullen made the last time he argued a case and you’ll find all those elements. They go right through Whisnant’s arguments to Judge Smith’s arguments and to the arguments this present day, and we suggest that the Court has to take action to stop it.²⁶⁴

Bright alleged that the District Attorney’s office had a practice of racial discrimination that encompassed not just jury selection tactics but also abuse of charging discretion and racially inflammatory oral advocacy. He wanted Brooks’ trial to become a vehicle for documenting these overlapping practices that infected all capital cases handled by the office. “Racial discrimination in Muscogee County in its criminal justice system, and in the practices of the District Attorney’s office, is so deep-rooted and pervasive that it is impossible to ensure that such discrimination will not play a determinative and unconstitutional role in the outcome of Mr. Brooks’ trial,” they argued in their memorandum of law.²⁶⁵ “Thus, under McCleskey, this Court must prohibit the State from seeking a sentence of death against Mr. Brooks.”²⁶⁶ Quoting Justice Powell’s own words, they insisted that “any evidence which suggests that the ‘risk [of racial discrimination is] constitutionally acceptable,’ McCleskey . . . is valuable and probative” on the question of whether a prosecutor has abused his prosecutorial discretion.²⁶⁷ Recounting the history of racism in the state and county, the legal memorandum also invited judges to subvert the original McCleskey decision by linking Justice Brennan’s dissent in that case to other cases that documented discrimination against black residents of Georgia.

Once the proof was tendered and the patterns shown, Bright demanded an injunction against the prosecutor’s office and anyone who might be involved in carrying out a death sentence:

We will show that, first, when Judge Whisnant, then District Attorney Whisnant, was district attorney for a number of years here, his right-hand person was Bill Smith. During that time Mr. Pullen came into the office. Judge Whisnant, only three months after getting the death penalty for William Brooks in the first televised trial in Georgia history, went on to become a judge. Judge Smith took over as district attorney. Mr. Pullen moved up to being the chief assistant. . . . Judge Smith tried the Carlton Gary case, the silk-stocking stranglings. He went on to his judgeship. Mr. Pullen moved into the district attorney’s office . . . it’s been a history with a lot of racial discrimination.²⁶⁸

The decision whether to seek the death penalty has “been made primarily by three white men over the last 17 years: Judge Whisnant, Judge Smith, and Mr. Pullen . . . those three people have really decided which 24 people would be subject to capital cases or not.”²⁶⁹ During argument on the motion, Bright argued that their evidence of racial bias would be stronger and more direct than that laid out in McCleskey itself. “I want to make it clear that this is not an elaborate multiple-regression analysis that we’re going to be putting on.” Rather, “[i]t’s a very straight-forward” set of exhibits demonstrating the prosecutor office’s biased charging decisions.²⁷⁰

As for the motion seeking to replicate McCleskey’s findings on a smaller scale, Bright argued that the District Attorney’s office improperly considered race in deciding whether to seek death. He noted that “the three capital cases pending in this court,” including that of Brooks, “all involve black people charged with murders of white people.”²⁷¹

In advance of a public hearing on these motions, Parker and Rev. Lowery held an anti-death penalty event at the Fourth Street Baptist Church in Columbus. The two men told the citizens assembled that race, not the nature of the offense, was “the most significant criteria” for the death penalty in the area.²⁷² “It’s wrong to teach our children to be vindictive rather than preventive” Rev. Lowery told the crowd. “It’s wrong to teach . . . that white life is more valuable than black life.”²⁷³ After the event, the two men paid a visit to Brooks’ jail cell to highlight popular complaints of local injustice. Their explicit goal was to marry harm reduction in court and abolition outside the courtroom: “to make people (in Columbus) aware of the inequities and tremendous cost inherent in this system” of killing their own.²⁷⁴ The media marveled at SCHR’s strategy of rebellious localism, noting that it involved a “dramatic” turning of the tables to focus “on the prosecution’s motives” because of McCleskey v. Kemp.²⁷⁵

Pullen denied racism in the district attorney’s office and bristled at being called “nasty, nasty names in front of the courthouse.”²⁷⁶ In response to the motions, he demanded access to any raw data, any “information that went into this computer that led to this so-called study.”²⁷⁷ He lashed out at abolitionist lawyers, saying “they will do any deed, tell any lie, hurt any cause, nation or individual in order to achieve the end to capital punishment in this country.”²⁷⁸ Otherwise, Pullen said in court, “Let the games begin.”²⁷⁹

At one point, Bright and his staff questioned Pullen and two other former prosecutors, including Judge Whisnant, under oath. Asked about their charging decisions, the prosecutors recited exactly the same race-neutral criteria, that is, whether there were aggravating factors to justify a death sentence. Yet Brooks’ team tried to rebut their testimony as to their charging decisions by putting on relatives of black homicide victims who testified that the investigation and prosecution of their cases were not pursued with the same kind of diligence as crimes against white victims. While black citizens were the victims of 65% of homicides in the county, “almost all of the death penalty cases that have been tried have involved white victims, 20 out of 24.”²⁸⁰ These family members, “their voices quavering with emotion, described the district attorneys’ callous disregard for their cases.”²⁸¹ They said that the district attorney’s office often did not even stay in close contact with them before resolving cases. One African American woman learned that her daughter’s killer had already been sentenced when she called the police to see when the trial would be. Another witness, also black, testified that the district attorney’s office allowed her husband’s killer to plead guilty without consulting with her beforehand. By contrast, when Brooks offered to plead guilty and accept life without parole, Pullen consulted the victim’s family before deferring to them and refusing a deal.²⁸²

At one point, Pullen tried to stop the testimony from the family of black murder victims. He stood up and conceded that victims’ families “had been treated callously in the past” and that as the new district attorney, he was trying to make amends.²⁸³ But the judge permitted the defense to proceed.

During a remarkable courtroom exchange, Bright objected to the judge’s initial ruling that admissible evidence as to racism within the county’s legal system would be restricted “to the last 18 months”—only the period that Pullen had been district attorney.²⁸⁴ But Bright replied that so sharply limiting the historical proof would be tantamount to doing “what we have done for years in the South,” which is to “wear blinders to sort of avoid really focusing and dealing with the problem of race and racial discrimination in the society.”²⁸⁵ It would be “like saying that if there’s a change in leadership of the Ku Klux Klan we can’t assume that that organization is a racist organization because we can’t look back at the church bombings and all that because that was when J.B. Stoner headed up the organization but now someone new has taken over.”²⁸⁶ Instead, Bright advocated a holistic test: looking at “the totality of the facts to see to what extent the history influences the present.”²⁸⁷ The evidence introduced by SCHR sought to portray “a larger historical context of discrimination against African-Americans in Georgia and Muscogee County” and urged that the trial of Brooks “be viewed in the context of racial violence against black people in Muscogee county and in Columbus,” what Bright called “cradle-to-grave discrimination . . . that includes and is most pernicious, in fact, in the criminal justice system.”²⁸⁸

Judge Lawson shrewdly allowed Brooks’ defense a wide berth to introduce what they wished into the record, but not everything could be live testimony. At one point, Bright was permitted to proffer the testimony of a Columbus native on the history of local racial violence, although Judge Lawson ultimately decided the testimony was not directly relevant to the motions. As Bright explained, Bill Winn, who wrote for the Columbus Ledger-Inquirer, would explain “the lack of trust of the criminal justice system” among black citizens as a result of “the uninterrupted history of racial violence inside and outside the criminal justice system in this community.” He would have spoken about the 1912 lynching of a 14-year-old “on the front steps of the courthouse here in Columbus,” as well as the lynching of a white proponent of racial equality that “caused so much racial unrest that black people in the community threatened to burn Columbus down.”²⁸⁹

Bright then turned to specific evidence of historic and systematic underrepresentation of minorities on Muscogee County juries. Bright asked Judge Lawson to take judicial notice of several decisions: a 1966 Fifth Circuit ruling called Vanleeward v. Rutledge²⁹⁰ “that the jury pools in Muscogee county discriminated on the basis of race”;²⁹¹ as well as Peters v. Kiff,²⁹² a 1972 ruling in which the U.S. Supreme Court reversed a sentence from Muscogee County, agreeing that even white defendants are harmed when a county excludes black citizens on a systematic basis.

“An all-white country club is a terrible thing,” Bright argued, “but an all-white jury is worse because it denies people participation in the judicial process.” He hammered the democratic deficits uncovered through the legal process.²⁹³ Over half the capital cases in the county, including that of Brooks, “have been tried with all white juries.”²⁹⁴ Bright introduced “the entire record from the first trial”²⁹⁵ against Brooks and told Judge Lawson that what happened was “a legal lynching,” which he defined as a “perfunctory criminal trial[]” that accomplishes the “same result” as an extralegal act of justice “without the unseemly way of doing it by just simply taking a person out and hanging him up.”²⁹⁶ According to Bright, who laid out how one racist practice exacerbated another,

The outcome of this case was a foregone conclusion. All the publicity; . . . the all-white jury; the exclusion of jurors on the basis of race; the closing argument, which . . . appealed to passion and racial prejudice; . . . a perfunctory trial broadcast to the community; and Brooks was off to the electric chair; and, of course, Judge Whisnant was off to the bench.²⁹⁷

3. Race-Based Jury Strikes Over Time

Brooks’ team lost the bulk of their pre-trial motions, but Bright persuaded Judge Lawson to order discovery on the racial discrimination claims. This momentous turn of events gave them time to develop a record and the opportunity to show that since the death penalty was reinstated in Gregg, prosecutors for the Chattahoochee Judicial District had tried 27 capital cases, with 21 involving white victims. The sample size was not large, but it also was not paltry. Nineteen people wound up with a death sentence.

Even more damning were the patterns in the office’s juror strikes over the years. In capital cases, prosecutors exercised peremptory challenges 59 times (85.5%) against black jurors, and only 10 times (14.5%) against white jurors. SCHR proved that these tactics led to eight black defendants tried for capital crimes in front of all-white juries: Joseph Mulligan, Jerome Bowden, Jamie Lee Graves, Johnny Lee Gates, William Brooks, and William Lewis.²⁹⁸ All eight cases involved black defendants, with seven of the eight cases involving black defendants and white victims.

Ruth Friedman, Mary Eastland, and several investigators and interns spent the months leading up to Brooks’ retrial going through the case files of the District Attorney. “We went through every file, coding them” for every factor that could conceivably be relevant to understanding prosecutorial practices such as prior offenses, the number of victims, and so on.²⁹⁹ This amounted to some 10,000 pages of documents in the county’s files. Denying racial bias, Pullen had said he sought a death sentence only when aggravating factors existed, and so they went about testing whether this was true or whether, instead, the race of the defendant or victim could be isolated as a motivating factor in a prosecutor’s decisions.

Just as with the Horton case, SCHR hired an expert to examine the data and try “to show that race was the constant.”³⁰⁰ But this time, they broadened their net to encompass all the prosecutors from the office. After examining the data, Dr. Michael Radelet testified that “those who kill whites are 5.9 times more likely than those who kill blacks[s]” to have the death penalty sought against them by the district attorney’s office.³⁰¹ But Pullen’s explanations, such as that some homicides are accompanied by an additional felony such as robbery, simply “does not explain this racial disparity.”³⁰² Neither did the degree of relationship between perpetrator and victim (even though more people who kill strangers have the death penalty sought against them).³⁰³ Other details supported Brooks’ motion: “given a female victim, those who kill whites are 5.1 times more likely to have the death penalty sought than those who kill a black female.”³⁰⁴ The pattern held when comparing black male victim versus white male victim: “those who kill white males are 11 and half times more likely to have the death penalty sought than those who kill black males.”³⁰⁵

As for the state’s jury strikes over time, Radelet testified that there was less than 1 in 1,000 chance that Pullen’s office was striking black jurors at such a high rate randomly. Pullen admitted being out of his depth, saying he was “not familiar with statistics.”³⁰⁶ During cross-examination of Radelet, Pullen tried to suggest that the expert was compromised because he appeared at defense lawyers’ press conference and handed out sheets with details of his research, but Radelet insisted that he was not morally opposed to the death penalty and merely had an “academic” interest in how it was enforced.³⁰⁷ Pullen strongly implied that Radelet did not include all factors that went into their charging decisions, such as the ages of perpetrators and victims, but Radelet replied that he took into account the most likely ones. When Pullen tried to get Radelet to say that the Baldus study had been rejected by judges, he retorted: “Not my reading of McCleskey. I think the Supreme Court made it quite clear that they accepted that study” and “[t]hat study has received some very prestigious awards.”³⁰⁸

At one point, Pullen lost his cool: “[Y]ou’re calling me and my two predecessors racists.”³⁰⁹ Parker objected, saying that “he has never said that the evidence indicates that he is a racist. He says the evidence raises a strong suggestion of racial discrimination in the application of the death penalty in this circuit.” Judge Lawson sustained the objection.³¹⁰

Pullen moved on to inquire why Radelet chose not to focus on the race of defendants, and Radelet explained that Bright asked him “to focus on race of victim.”³¹¹ This amounted to an effort to focus on one aspect of the Baldus study (racial disparities in cross-racial killings), and to try to replicate it on a more intimate scale. Radelet acknowledged that the raw numbers showed that of the 45 capital cases they were discussing, 23 defendants were white and 22 were black.³¹² Radelet pushed back, saying that “the lesson that we have learned from this research project and from many others done over the last decade is that if we look at defendant’s race alone, that creates a totally misleading picture of the effect of racial variables on the imposition of the death penalty.”³¹³ Other factors either could not be quantified or were not for purposes of Radelet’s study, such as strength of a case, criminal history, whether there was provocation, degrees of premeditation, and defendant’s state of mind.³¹⁴

After this hearing, the NAACP called a press conference to attract media attention to the hearing. The NAACP hailed SCHR’s aggressive approach in Brooks’ trial as “a model of new defense strategies” to satisfy the more stringent standard of proof for discrimination demanded by McCleskey.³¹⁵ Earl Shinholster, the NAACP’s southern regional director, said that the evidence of racial bias tendered during the hearing proved once again that the answer is to “abolish the death penalty.”³¹⁶

Ultimately, Judge Lawson refused to bar the death penalty at the pre-trial stage. It was always highly unlikely that a single elected judge was going to stop a capital trial on the basis of past discrimination by prosecutors, much less issue a blanket order preventing the entire office from seeking death sentences in the future. Nevertheless, rebellious localism served several objectives. First, SCHR’s holistic approach to advocacy disrupted the status quo, where a defendant in cross-racial killing was often briskly tried and sentenced to death. This more aggressive pre-trial posture litigating equality issues raised the costs of litigation for the government, forcing prosecutors to ask themselves repeatedly whether insisting upon a death sentence was worth it when the entire case might be more expeditiously and securely resolved through a plea deal. Not only would a district attorney’s office have to defend against several pre-trial motions and prepare for lengthy hearings, but they would also have to invest in experts and build a separate case in favor of death even if they secured a conviction. The costs would continue to escalate if other issues emerged, such as challenges to the possible method of execution, especially if a trial were delayed so a defendant could pursue an interlocutory appeal.

Second, such litigation tactics served an educative function by surfacing the often-subterranean racial dimensions of legal justice. They heightened public awareness of the case and led affected segments of the community to pay closer attention and even participate in the proceedings. While Bright and his staff never lost sight of the primary goal of harm reduction (here, saving their client’s life), rebellious localism ensured that constitutional issues of equality and fairness would be debated more broadly, whatever fate befell their client in the end.

Third, by hammering the issues of race and poverty in the administration of capital punishment so relentlessly, Brooks’ team increased the participants’ sensitivity to a greater possibility of a reversal on appeal or during a subsequent post-conviction proceeding. The new judge, no doubt wanting above all as clean a trial as possible on the second go-round, had given defense lawyers plenty of leeway to raise their concerns and preserve issues for appeal. Exploiting that desire helped them to overcome the usual obstacles to building a robust record of discrimination during a criminal trial. If their client was convicted, that record would be before an appellate body should judges be in the mood to engage in constitutional policymaking.

During jury selection, Pullen still used all his peremptory strikes against black jurors. Even so, aggressive defense lawyering led to a jury seated with nine black citizens and three white citizens—a far cry from what happened the first time around.³¹⁷

Fighting so hard for racial equality now gave Brooks’ team another chance to battle for human dignity in front of a mixed-race jury. Their change of venue motion was granted, and the trial took place in Morgan County. This time, the state’s case did not feature overtly racist appeals.³¹⁸ After Brooks was convicted of murder, Bright pleaded with the jury to spare his client from the electric chair. “We get down now to the ultimate issue here: Do we kill William Brooks?” Bright said to the jurors. “An issue about as stark and about as great a moral decision as a human being could ever be called upon to make.”³¹⁹ He proceeded to give a textbook closing argument weaving together themes of residual doubt and the possibility of redemption, as well as a deep dive into the aspects of Brooks’ troubled upbringing, including childhood abuse, that favored a merciful outcome.

After deliberating for less than an hour, the jury decided to impose a life sentence. Once again, rebellious localism had made the difference. This time, unlike in the Horton case, rebellious localism had yielded a life sentence without necessitating an appeal and collateral litigation.

Brooks’ ordeal illustrated another aspect of abolitionists’ harm reduction goals: using the legal process to keep a client alive long enough for political conditions to shift back in their direction. As Rachel Barkow has explained, “[p]opulist fears and impulses among the electorate create pressure on prosecutors to make ill-advised short-term decisions that end up compromising public safety”³²⁰ and violating rights. High-profile, cross-racial murder cases during the War on Crime years tempted prosecutors into breaking constitutional rules and engaging in unethical conduct. But as Bright’s strategies illustrated, extending a life beyond a particular historical moment might lead a different prosecutor to agree to a plea deal or, as in Brooks’ case, secure a jury with sufficient distance from highly racialized politics to consider mercy.³²¹

As for the cause lawyer’s goal of achieving an authoritative repudiation of McCleskey, that would take more time.³²² More lasting policy transformation, too, would require building on legal errors and structural injustice uncovered in SCHR’s cases. Whether the issue involved abolishing capital punishment, reforming how jury lists were constituted, reducing or eliminating peremptory strikes, or constraining how prosecutors made charging decisions—reform would require political action, not merely favorable judicial decisions.

Even so, what happened in Johnny Lee Gates’ case in 2018 should offer abolitionists hope that rebellious localism can do some good for long-term goals. Citing SCHR’s win in Foster v. Chatman, lawyers for Gates gained access to the prosecutors’ notes during his original 1977 murder trial. Gates, a black man who has always maintained his innocence, was convicted of rape and murder of a white woman by an all-white jury. The notes showed that prosecutors closely tracked the race of prospective jurors, marking white jurors with “W” and black jurors as “N.”³²³ Additionally, prosecutors described black jurors as “slow,” “old + ignorant,” “cocky,” “con artist,” “hostile,” and “fat.”³²⁴

Because the prosecutors were Doug Pullen and William Smith, from the same office that had prosecuted Brooks, SCHR lawyers Patrick Mulvaney and Katherine Moss introduced the study they had compiled based on that earlier case. They argued that Gates was entitled to a new trial based on the “newly discovered notes” plus “the pattern of strikes across cases establish systematic race discrimination.”³²⁵ From 1975 to 1979, the state brought seven capital cases against black defendants in Muscogee County and struck 41 black jurors—in 6 of those cases, every single black juror was eliminated to secure all-white juries. Pullen was involved in 5 of the 7 cases, striking 27 of 27 black jurors. Smith was involved in four of those cases. According to the affidavit of Dr. Michael Lacey, “the probability that they
exercised those strikes without considering race is 0.000000000000000000000000000004.”³²⁶

On January 10, 2019, a state court granted Gates’ extraordinary motion for a new trial on a separate ground: that DNA testing of evidence from the crime site did not contain Gates’ DNA. But in the stunning opening pages of his order, the judge also accepted the evidence of racial discrimination presented by SCHR, finding the evidence of bias “overwhelming.”³²⁷ At last, abolitionists could point to formal recognition of historical evidence of structural racism. Not only did the judge find that the prosecutor’s office had a practice of race-based peremptory challenges, but he also connected that jury selection strategy to outrageous rhetorical appeals made by prosecutors during capital trials. He noted that arguments in front of all-white juries were more explicitly racist and deplored the “racially charged arguments [that] spanned across multiple cases, including Gates’s case.”³²⁸

The Georgia Supreme Court upheld the lower court order granting relief based on the DNA evidence. Yet the justices, too, like the local judge, went out of their way to note the “troubling findings” of prosecutorial misconduct in capital cases “held in the Chattahoochee Judicial Circuit between 1975 and 1979.”³²⁹ Even if it was not the explicit rationale for relief, the structural racism claim—with its focus on highlighting intentional racism in particular counties and circuits—almost certainly shaped how these judges perceived the rest of the defendant’s arguments.

This victory mattered because in recent years, successful arguments to end capital punishment have not sounded in religious or other abstract rationales, such as the efficacy of deterrence, but rather concerns about actual innocence, unequal administration on the basis of race and poverty, and excessive costs. For instance, when the Washington State Supreme Court halted the death penalty on state constitutional grounds in 2018, it did so because capital punishment “is unequally applied—sometimes by where the crime took place, or the county of residence, or the available budgetary resources at any given point in time, or the race of the defendant.”³³⁰ With Gates, abolitionists acquired another crowbar with which to pry open the machinery of death.

CONCLUSION

Rebellious localism is a strategy of legal resistance that is most effective during tough political times, when legal doctrine is in the process of becoming more complex and less humanitarian in orientation, when public servants are becoming less interested in vindicating robust theories of equality and fairness and instead focused on maximizing other objectives. Its practitioners search for spaces within doctrine to subvert the ideological and policymaking intentions of judges and exploit opportunities to turn the tables on public servants who fall short of ideals. In part to correct for the waning of trust in legal institutions, they continue to appeal to broader, intuitive notions of what is just and fair given the immense stakes involved in capital cases.

When those stakes were the highest—a client’s life hung in the balance—and national political conditions were least favorable for maximalist progressive constitutional projects, Bright and his staff were willing to do what it took to implement this approach. For them, it meant not giving up on wider reform goals, such as abolishing the death penalty and improving the quality of indigent defense systems, but rather entailed shifting the emphasis to harm reduction and incremental advancement of structural claims. From the standpoint of harm reduction, success would be measured according to lives saved or extended. Capital cases were slowed down, executions were put off time and again. In Georgia, a state supreme court that began the post-Gregg era generally unwilling to review capital trials closely eventually became alarmed by the inequities on display and, as in cases like Ford and Gamble, started to send appalling trials back to be done over. On that score, rebellious localism accomplished more than one might have dreamed at the outset.

If doctrine is in part a method of empowering allies and weakening adversaries, then rebellious localism also advanced this agenda. In terms of advancing principles of racial equality and fundamental fairness in criminal trials and resisting the impoverished vision of justice exemplified in McCleskey, rebellious localism created fresh templates for action and legal precedents such as Horton and Gates to build upon—often, in situations far removed from the influence of the U.S. Supreme Court. These victories, which included pre-trial rulings forcing the disclosure of evidence in the hands of prosecutors or forcing public servants to testify, went the farthest. Equally useful were rulings that excoriated the underrepresentation of jury pools or practices that denied eligible black citizens a chance to participate in the criminal process. These rulings showed the lengths to which some public officials were willing to go to kill fellow citizens—mostly black people charged with serious crimes against white people. By insisting that the misconduct uncovered was not just evidence of isolated bad actors but rather proof of a system gone awry, SCHR’s work advanced abolitionist aims. At this broader level, rebellious localism shamed elites by exposing the inequitable aspects of the criminal justice system.

Nevertheless, rebellious localism was an extremely time-intensive endeavor. Given the unavoidability of court-centered and litigation-focused tactics, in-court progress would be limited by the nature of the adjudicative process unless out-of-court mobilization could capitalize on momentum created during legal proceedings. Putting local systems on trial entailed front-loading and multiplying colorable constitutional claims, as well as escalating legal rhetoric about fundamental values at stake. That meant consciously creating more work for themselves and their adversaries by converting the pre-trial process into an engine for public law-style litigation. It also meant that advocates could not serve quite as many clients as they might otherwise, for the maximalist model of representation meant fewer cases they would be able to add to their plate.

This technique involved destroying the possibility of maximally efficient capital trials and generating friction between the cause lawyers themselves and other legal actors who were repeat players. Another drawback, at least in court, was that abolitionists often chose to play to existing stock stories about racial hostility rather than delve more deeply into what was likely a more complicated truth. As Randall Kennedy puts it, cross-racial disparities often have to do more with “racially selective empathy rather than racially selective hostility” because white prosecutors, judges, and jurors “relate more fully to the suffering of white victims” who can be readily “imagined as family or friends.”³³¹ Some of this account emerged through pre-trial hearings, but not enough to inform specific proposals to reduce the distorting effects of selective empathy throughout the criminal justice system.

An advocacy group can only hope to pull off rebellious localism of this sort if it has sufficient ties to the community and is capable of tapping the people’s sense of history, their long-held grievances, and their own dreams of justice. It must also have the right personnel: people who would rather be feared as adversaries than loved, willing to battle every legal point.

SCHR could not save every client, nor did they win every motion. Yet the organization’s successes over time—and its staff and cooperating attorneys—enjoyed several legal victories after beating significant odds—suggest that a sharp theoretical dichotomy between radicalism and gradualism is too simplistic. Rather, historians and theorists should be looking for forms of gradualism that can co-exist with radical aims, striving to better understand when hybrid approaches to legal reform might work and when the tensions are too great to manage.

To be sure, SCHR did not accomplish these goals alone. Others were engaged in local battles against the death penalty systems in their states, exposing inequities in terms of race, wealth, and geography. Nevertheless, SCHR’s efforts contributed to the burgeoning discourse that capital punishment was unjust because it was administered unequally and unfairly. The organization’s work represents only one part of that broader story of abolition that continues to unfold.³³²

From the standpoint of judges making constitutional law at the highest levels, the lessons should be more sobering. Efforts to portray the law in terms that privilege order over justice are not only likely to fail, but could wind up horrifying the law’s subjects and causing counterproductive effects. Such projects will engender furious resistance, even when the winds appear to be blowing in the very direction that Justices might wish to go. An overweening desire for legal efficiency can in fact produce its opposite: a flurry of legal texts and orations that call into doubt the prospects for equal justice and the legitimacy of the very institutions that order-preserving jurists hope to bolster. Deep worries about whitewashing inequality may prompt accusations of racism that taint outcomes, practices, and judicial personnel—whether or not they are satisfactorily proven.

Concerns about the value of statistical evidence had played a role in Justice Powell’s desire to insulate the criminal justice system in McCleskey. But he had not done much, if anything, to call into question the integrity of the Baldus study, or to explain convincingly why that evidence was any less reliable than in other settings. His solution—to neither seriously question it but to suggest statistical evidence was less useful on a state-wide scale—did not end advocates’ efforts to quantify racism.

As SCHR’s experience demonstrates, it remained possible to use expert testimony and quantitative evidence to shed light on prosecutorial motives in charging decisions and peremptory strikes. It is true that their studies involved fewer state actors, but it is not obvious that the evidence was significantly more reliable than the Baldus study. While replicating similar cross-racial patterns, the local studies had a smaller sample size and considered fewer non-racial variables. The mini-McCleskey hearings did give prosecutors a chance to explain their apparently race-based conduct, something that Justice Powell went to some length to say did not happen with the Baldus study. The failure by prosecutors such as Briley and Pullen to rebut the evidence of racial discrimination was then held against them.

Another factor seems important: In SCHR’s cases, statistical evidence of bias was not the only evidence offered. Instead, such evidence of unequal administration of the law bolstered other evidence of discrimination by a particular bad actor or practices of a discrete jurisdiction. For this reason, the narratives of unconstitutional behavior were simpler, more familiar. They resonated with beliefs about the quality of local justice and pricked the consciences of people to reflect more deeply on how much more might still need to be done.

At the end of the day, the basic point remains: McCleskey did not end constitutional debate or solve problems of proof once and for all. To the contrary, the ruling opened a more intense and grueling chapter of the debate over the meaning of equal justice. On this question, as well as with the ultimate scope and significance of that precedent, elites would not have the last word.

96 S. Cal. L. Rev. 1031

Download

Robert L. Tsai*

* Professor of Law and Law Alumni Scholar, Boston University School of Law. I am grateful for the help of Mary Sidney Harbert in locating old case files at the Southern Center for Human Rights, and to Nathan Jordan, Archives Specialist for the National Archives. Earlier versions of this Article were presented at workshops at Boston University’s School of Law, University of Colorado School of Law, University of Connecticut Law School, Fordham Law School, New York Law School, and the University of Utah Law School, which brought me out to Salt Lake City as the 2022 Howard Rolapp Distinguished Visitor. Generous support from Dean Angela Onwuachi-Willig and Boston University School of Law allowed me to complete this project. I deeply appreciate the feedback on previous drafts from Bethany Berger, Stephen Bright, Debby Denno, Jonathan Feingold, Brandon Garrett, Erika George, Aya Gruber, Aziz Huq, Gary Lawson, Benjamin Levin, Justin Murray, Sachin Pandya, David Seipp, Matt Tokson, and Mark Tushnet. Invaluable research assistance was provided by Victoria Gallerani, Greg Margida, Colin Wagner, Allie Wainwright, and Catherine MacCarthy. Finally, thanks to Jessica Block, Diana Chung, and the staff of the Southern California Law Review for their fine editorial assistance.

Robert L. Tsai*

After “McCleskey”

Robert L. Tsai*

In the 1987 decision, McCleskey v. Kemp, the Supreme Court rejected a black death row inmate’s argument that significant racial

The Court’s Morality Play: The Punishment Lens, Sex, and Abortion

June Carbone* & Naomi Cahn†

This Article uncovers the hidden framework for the Supreme Court’s approach to public values, a framework that has shaped—and will

Red, White, and Blue—And Also Green: How Energy Policy Can Protect Both National Security and the Environment

David M. Schizer*

Too often, energy policy protects the environment while neglecting national security, or vice versa. Since each goal is critical, this

The Healthcure System: A Regional Accountable Care Model to Remedy Healthcare’s Pricing Problem

Christopher A.G. LoCascio*

INTRODUCTION The most sinister game show in American life commences every time a hospital provides care, draws up an eye-popping

Divided Agencies

Brian D. Feinstein* & Abby K. Wood†

Clashes between presidential appointees and civil servants are front-page news. Whether styled as a “deep state” hostile to its democratically

The Social Context of the Law: A Critical Analysis of Reliance Interests in the Department of Homeland Security v. Regents of the University of California

Raquel Muñiz,* Maria Lewis,† Grace Cavanaugh‡ & Melissa Woolsey††

In 2020, the U.S. Supreme Court ruled on the Department of Homeland Security v. Regents of the University of California case.