Consequences and the Perils of Categorical Ambiguity

Collateral Consequences and the Perils of Categorical Ambiguity

ALEC C. EWALD

The United States is about a generation into an unprecedented experiment in mass incarceration. The pace of change has been remarkable: the U.S. now imprisons its citizens at five times the rate it did as recently as 1970, and at a far higher rate than most democracies. Our dramatically expanded penal system shapes American racial and class inequality, as incarceration percentages for blacks are eight times higher than those for whites, and the typical prisoner has not completed high school. On a human scale, the prison boom has effected “a historically significant transformation of the character of adult life” for millions of Americans.¹ More than 600,000 ex-offenders are released from prisons and jails in a typical year²—a figure approaching that of the entire population of the smallest U.S. states.

Policy-makers and academics are only now coming to grips with the distortions wrought upon some of our most fundamental social and political measurements by what Marie Gottschalk calls the new “carceral state.”³ For example, the census counts inmates as residents of the prison town, a practice that appears to skew apportionment and policy-making at the county and state level.⁴ Voter turnout figures need adjustment because the voting-age population on which those figures were historically based now includes so many people who are actually ineligible to vote under state law because of a criminal conviction.⁵ And unemployment rates in the U.S. and Europe cannot be directly compared, because incarceration removes so many people from the American labor pool.⁶ All told, our carceral institution now exerts “a salience and reach that are wholly unprecedented in American history as well as unparalleled in any other society.”⁷

Much of that reach today comes from a set of penalties rarely formally defined as punishments: “collateral” sanctions, restrictions called “indirect” consequences of a criminal conviction and that “attach to, but are legally separate from, the criminal sentence.”⁸ As many as 16 million Americans have been convicted of a felony,⁹ and thus may have their ability to work, serve on a jury, vote, own a firearm, join the military, seek federal student loans, receive food stamps, and live in public housing curtailed or eliminated altogether. Scattered throughout state and federal law, rarely contained in criminal codes, and flowing from “complicated intersections of state, quasi-state, and private associations,”¹⁰ collateral sanctions expand the penal institution in both place and time—beyond the prison walls (and probation offices) and beyond the sentence as well. Indeed, these restrictions sometimes place heavier, longer-lasting burdens on an offender than does the formal sentence.¹¹ Some collateral consequences are quite old, but state restrictions increased in number between 1986 and 1996; both the American Bar Association (ABA) and the National Conference of Commissioners on Uniform State Laws (NCCUSL) conclude that federal and state collateral sanctions have expanded in “variety and severity” in the last twenty years.¹²

Yet despite their wide impact and recent growth, collateral consequences remain in a cloudy legal gray area, exerting severe and long-standing punitive effects but not legally defined as punishment and absent from even excellent new texts on sentencing law.¹³ A fundamental ambiguity persists within American collateral-consequences policy: whether such sanctions are punitive, regulatory, or some combination of the two. Most restrictions are enforced not by the criminal-justice system proper, but by executive-branch administrative agencies, state and local bureaucracies, or professional boards, often years and sometimes decades after sentencing. Thus, many jurists and some commentators conclude that collateral sanctions are fundamentally regulatory: meant to ration scarce resources or ensure that only certain citizens are eligible for a given profession, for example. In the words of one nineteenth-century commentator, “[P]ublic protection and not individual punishment” is their aim.¹⁴

The line between criminal punishment and civil regulation has never been sharp or easily drawn. But in the last few decades, “[T]he distinction between criminal and civil law seems to be collapsing across a broad front,” as Mary M. Cheh wrote in 1991.¹⁵ Drawing the line has “become impossible,” wrote Susan R. Klein in 1999. Governments collect “civil” forfeitures and confiscations freely, and numerous “punitive civil sanctions” imposed in noncriminal proceedings blur the line between civil remedy and public punishment.¹⁶ What Katherine Beckett and Steve Herbert call “architectural modes of exclusion” spreading through U.S. cities, such as new uses of trespass authority and off-limits orders, also “rest upon a complex mixture of civil, administrative, and criminal legal authority.”¹⁷

Of the many policy arenas now merging punishment and regulation, post-sentence collateral sanctions are among the most consequential. This is not to gainsay the human, social, and philosophical importance of other civil/criminal policies: new forms of “socio-spatial exclusion” represent a troubling modernist recapitulation of banishment,¹⁸ while civil forfeiture and civil confinement (particularly the confinement of sex offenders) impose enormous costs on those affected and raise grave ethical questions. Similarly, the law and practice of deportation (now imposed on nearly 2 million people a year in the U.S.) also blends punitive and regulatory features. But in the number of people it affects, the range of activities it covers, and its complex, fascinating legal structure, the expanding collateral sanctions regime represents a major development in American political life. The late political philosopher Judith Shklar concluded that at the core of American citizenship are the ability to work and the right to vote.¹⁹ Millions of Americans now lack the ability to participate in those activities, and many others live in a darkened status Nora V. Demleitner calls “internal exile.”²⁰ No longer incarcerated, they walk through society free of any formal criminal supervision but deprived by piecemeal state and federal rules of core capacities of citizenship.

This chapter explores collateral sanctions’ awkward straddle of punitive and regulatory aims. In showing that these restrictions do not fit clearly into either category, I hope to demonstrate what I believe are the real dangers of that ambiguity. The harmful consequences of our massive, murky, ill-defined collateral-sanctions regime extend well beyond those directly affected, rendering citizens unable to judge the efficacy of such restrictions and undermining core commitments of the American political order. While I do conclude that many such restrictions seem quite unlikely to fulfill their purported objectives, my core claim is not that collateral sanctions are necessarily bad policy—but instead that our confusion over their character and purpose actually keeps us from knowing how to judge them in the first place.

To begin, I offer the detailed account of current U.S. collateral-consequence policies necessary to enable close inquiry into the themes most interesting to us here: how collateral sanctions combine punitive and regulatory aspects, feature both centralized and dispersed characteristics, and alter the content of American citizenship. The second section explains the ambiguity surrounding collateral consequences’ blend of regulatory and punitive elements. While U.S. courts have defined almost all such restrictions as civil and regulatory, commentators are virtually unanimous in emphasizing their punitive effects. Yet collateral sanctions do not fit comfortably into any of the frameworks by which we understand criminal justice or regulatory policies. Several core concerns of the criminological literature, such as the contemporary desire to denigrate and stigmatize offenders, the move toward “actuarial justice,” and the pervasive desire to reduce costs, do capture important elements of American collateral sanctions policy. Similarly, the “new regulatory” account, in particular its emphasis on “de-centered” policy in post-welfare state industrial societies, links up well to some elements of collateral consequences today, but runs afoul of others.

We need to look closely at contemporary American collateral consequences in order to appreciate these restrictions’ quantity and reach, their diverse, complex legal structures, and their varying mechanisms of implementation.²¹ The outline that follows moves from center to periphery, as it were, beginning with those consequences automatically imposed on all offenders by federal law and moving out through greater layers of delegation, discretion, and local authority. Though this tally may appear lengthy, in fact a truly comprehensive assessment would consume this entire volume: a compilation published in 2009 by the American Bar Association lists over one hundred different federal statutes.²²

First, a few terminological clarifications. Some authors employ the term “collateral consequences” to capture the effects of incarceration on the life course, employment prospects, and well-being of the offender, as well as on families, communities, and socioeconomic groups.²³ Tonry and Petersilia, for example, identify laws restricting political and economic behavior as only one of six different kinds of “collateral effects of imprisonment.”²⁴ Without minimizing the importance of such familial and social effects, I focus on public policies: ways “government intervenes to perpetuate the second-class citizenship of ex-offenders.”²⁵ (As will become clear, even drawing this line turns out to be difficult.) Meanwhile, despite the common emphasis on felony offenders, several of the most serious collateral consequences—including deportation, eviction, temporary loss of custody of one’s children, and job suspension—are routinely imposed not only on misdemeanants but also on people merely arrested or charged. Eviction from private or public housing in New York City, for example, may follow the mere issuance of a search warrant, if the relevant District Attorney believes narcotics activity is present.²⁶ And while I sometimes refer to “former prisoners,” the full slate of collateral sanctions follows anyone convicted of a felony, even those who never see the inside of a cell. Finally, implicit in virtually all areas of collateral-sanctions law is that these restrictions extend after the sentence proper has been completed.

Three important federal collateral sanctions operate automatically and indefinitely on anyone with a felony conviction. By statute, federal grand and petit juries are closed to anyone convicted of a felony in federal or state court, unless and until that person has received a state or presidential pardon.²⁷ The same is also true of firearms rights—but those convicted of some business-practice felonies are exempted, and the state of the law is quite complicated. For example, the Bureau of Alcohol, Tobacco, and Firearms is authorized by statute to restore federal firearms privileges, but since 1992 it has not been allowed to spend any money on such restorations.²⁸ A third disqualification operating automatically and indefinitely on all felons is ability to serve in the military, and military pensions may also be forfeited by the incarcerated; however, unlike bans on jury service and firearms ownership, ineligibility for military service may be lifted by the secretary of a given branch of the armed forces.²⁹ This has happened with some frequency as the all-volunteer army fights wars in Iraq and Afghanistan.

While about a dozen different statutes disqualify people convicted of certain crimes indefinitely and automatically from the right to be elected to federal office, this sanction does not apply to all felons.³⁰ The same is true of many federal occupational-license statutes, which restrict access to certain professions for people guilty of specified offenses. Webb Hubbell, the former Arkansas Supreme Court Chief Justice convicted of tax and mail fraud after the Whitewater investigation, found that having been convicted of a federal crime (and particularly one involving deception), “almost any job requiring a federal license” was now off-limits, including working in the merchant marine, doing export business, or being a locomotive engineer. Not only was he barred from serving as an administrator of any federally insured bank, but he could not be an employee of such an institution; he could not work in the administration of any labor organization for thirteen years after the conviction. Multiple agencies, including the Commodities Futures Trading Commission, the Securities and Exchange Commission, and Health and Human Services, could refuse to allow him to work in any partly federally funded facility—”I couldn’t even sweep the floors of a nursing home,” as Hubbell put it.³¹

Such federal occupational and licensing bans vary, coming in at least four varieties of imposition, specificity, and release. First, some operate automatically and exclude from certain jobs—including some government jobs—everyone convicted of certain offenses unless and until pardoned. The professions most explicitly restricted by federal law include finance, insurance, union work, health care, child care, and transportation work.³² Second, other exclusions are also automatic, but may be lifted by an executive agency even if the person in question has not received a pardon. A third type of disqualification is authorized by federal law, but does not automatically accompany a conviction and must be imposed by the sentencing court. A person convicted of bribing a public official, or the official accepting such a bribe, may be disqualified from federal office by the sentencing court. More generally, a sentencing court may impose occupational restrictions as a condition of probation or other release. Both federal statutory law and sentencing guidelines allow judges considerable discretion to decide which restrictions are “reasonably necessary to protect the public,” and how long such restrictions are to last in a given case. Finally, a few Statutes do not themselves ban former felons from occupations, but explicitly authorize federal agencies to consider such offenses when making licensure decisions. Some disqualifications in each category appear to be indefinite, while others may be for only a set period.³³

Some peer nations, including Canada, Australia, and the United Kingdom, have incorporated protections for former offenders into antidiscrimination and human-rights laws.³⁴ For example, under the UK’s Rehabilitation of Offenders Act, a conviction is “spent” after a certain period of time, and may no longer be grounds for refusing a person employment.³⁵ By contrast, the U.S. has no federal statutory prohibition on private-employer discrimination against people with former convictions. The Civil Rights Act of 1964 prohibits hiring discrimination on the basis of race or color, and the Equal Employment Opportunity Commission (EEOC) has issued guidelines stating that where hiring choices shaped by former-offender status disproportionately impact racial minorities, they may violate Title VII of the act. However, this remedy is available only to racial minorities and is difficult to enforce.³⁶

Various federal benefits are automatically withdrawn from people with criminal convictions under diverse statutes, most of them enacted since 1988. Even misdemeanor drug offenders lose access to federal student loans, grants, or work assistance (with increasingly lengthy periods of ineligibility for additional offenses). Despite clear evidence that they reduced recidivism, college prison programs were restricted by Congress, in large part by curtailed access to Pell Grants—drug offenders were stripped of eligibility in 1988, those sentenced to life in 1992, and all state and federal prisoners in 1994.³⁷ Later statutes made drug offenders ineligible to participate in the National Community Service Trust Program or to receive the Hope Scholarship Tax Credit.³⁸

One set of federal collateral sanctions is established with unusual clarity in American law. Established under the Anti-Drug Abuse Act of 1988, the Denial of Federal Benefits Program is run by the Bureau of Justice Assistance (BJA) within the Department of Justice. The program empowers state and federal courts to deny selected benefits—including loss of access to grants, contracts, financial aid, and licenses—to people convicted of narcotics possession or more serious drug offenses.³⁹ Unlike the restrictions summarized above, here disqualification is squarely described as a “sentencing option for federal and state courts.” Indeed, the bureau specifically identifies its character and purposes as penal: “The denial of federal benefits is a sentence pronounced by a state or federal judge as a result of a conviction for trafficking or possession of drugs,” explains a BJA document. Denial of “most taxpayer-supported economic benefits,” the DJA notes, is an “intermediate step” that helps to “enhance a drug conviction’s impact” and also “close [s] the gap between incarceration and probation.”⁴⁰ In addition to its retributive goals, the program aims at preventing crime, the DJA explains, particularly because it “alerts casual drug users to the fact that. . .”

as students, they can lose their student loans; as broadcasters, they can lose their Federal Communications Commission licenses; as physicians, they can lose their authority to prescribe medicine; as pilots they can lose their Federal Aviation Administration licenses; as business owners, they can lose their Small Business Administration loans or the right to contract with the Federal Government; and as researchers, they can lose medical, engineering, scientific, and academic grants.⁴¹

Judges do not have unlimited discretion in imposing these restrictions: housing, welfare, Social Security, and veterans’ benefits cannot be withdrawn under this statute. Moreover, time limits are placed on disqualifications: one year for a first possession offense, up to five years for second offenses; those convicted of trafficking crimes may be declared ineligible for five- and ten-year periods for first and second convictions but are rendered permanently ineligible by a third conviction.⁴²

In addition to its direct statement of penal aims, the program also differs from most other collateral sanctions regimes in that it maintains its own database. Where most such restrictions rely on some combination of disclosure, general-use criminal databases, and ad hoc research, the BJA’s “Denial of Federal Benefits Program Clearinghouse” compiles and maintains a list of convicted individuals and which benefits are denied each of them. The database held about seven thousand cases as of 2001.⁴³ As the BJA explains, not just “federal agencies and lending institutions” but also “nonfederal persons” access the database as needed—indeed, federal agencies as well as organizations in charge of certain federally funded awards or benefits “are required to consult the Debarment List” prior to allocating funds.⁴⁴

A final federal sanction is deportation of noncitizens, a particularly complex and weighty policy—and one whose doctrinal status changed in 2010 due to a landmark Sixth Amendment decision by the U.S. Supreme Court. As the Court has acknowledged, deportation is “at times the equivalent of banishment or exile.”⁴⁵ Several federal statutes list crimes that subject aliens and permanent residents to mandatory deportation, including narcotic offenses likely to bring only relatively minor sentences for citizens.⁴⁶ Legal permanent residents convicted twice of crimes of “moral turpitude” may be removed from the country regardless of how long they have lived in the U.S., and regardless of whether those offenses are felonies or misdemeanors; the Immigration and Naturalization Service (INS) may deport any alien convicted of an aggravated felony at any time.⁴⁷ Remarkably, more than 900,000 people were deported from the U.S. because of their criminal status from 2000 to 2009.⁴⁸ Formally, deportation is sometimes imposed at the discretion of the INS, immigration judges, and the Board of Immigration Appeals, rather than automatically. Until the Supreme Court’s decision in Padilla v. Kentucky (2010), that fact had been key to defining deportation as an “indirect” rather than “direct” sanction.

In Padilla, the Court ruled that the Sixth Amendment requires defense counsel to apprise their noncitizen clients of the possible deportation consequences of a conviction. Noting dramatic changes in recent decades to federal immigration law, the Court observed that deportation is now “virtually inevitable” and “nearly an automatic result for a broad class of noncitizen offenders.” Those facing deportation once had avenues short of Presidential pardon by which to win relief, but judges, the INS, and the attorney general have seen their relief authority severely restricted by a series of Congressional statutes.

Calling deportation a “harsh” consequence and a “drastic measure,” the Padilla Court found that its automatic nature now gives deportation such a “close connection to the criminal process” that lawyers fail their Sixth Amendment duties if they fail to inform noncitizen clients—particularly those engaging in plea bargaining—of their jeopardy. Notably, the Court added deportation advice to counsel’s Sixth Amendment duties while calling this sanction “uniquely difficult to classify as either a direct or a collateral consequence.”⁴⁹

Thus, a wide range of rights and benefits are withdrawn under direct authority of national law, some automatically and others at the discretion of judges and bureaucrats. Some explicitly identify regulatory aims and a few openly pursue punitive goals, but most do not articulate their purposes. Many restrictions in national law, meanwhile, operate in a very different way, one mediated by the American federal system.

Federal law grants states a measure of flexibility in how they impose restrictions in at least four major areas: availability of driver’s licenses; provision of public assistance and food stamps; access to public housing; and parental rights. Under a 1992 law, states must either remove driver’s license privileges for at least six months from a person convicted of any drug offense, or enact both legislative and gubernatorial statements opposing this rule. Failure to do one or the other would result in a 10 percent reduction in certain federal highway funds.⁵⁰ In essence, here the federal statute erects a default policy, requiring states to actively refuse to strip driving privileges from drug offenders. States have responded in varying ways: as of 2003, twenty-seven had opted out of the suspension requirement, seventeen employed a six-month penalty for first offenders, four put in place a restriction lasting longer, and two states employed shorter bans.⁵¹

A similar structure—federal default policy of exclusion, with an “opt-out” avenue—exists with regard to cash assistance and food stamps. Under the Personal Responsibility and Work Opportunity Reconciliation Act of 1996, anyone convicted of a drug felony is permanently barred from these benefits. But states are allowed to opt out of that exclusion, either entirely or in part, and in the years since have erected widely varying policies.⁵² Twelve states have eliminated the TANF/food stamp ban entirely; twenty-three employ the ban, but limit it in significant ways, imposing it only on more serious offenders or facilitating restoration via treatment, testing, or the passage of time, for example; and fifteen states adopted the federal ban without modification.⁵³

Housing policies feature a different mix, with local agencies authorized to wield a good deal of discretion. Federally funded housing authorities are required by federal law to evict and permanently exclude anyone convicted of certain sex offenses and narcotics crimes. Other statutes, however, enable housing authorities to determine not only which crimes bring eviction and exclusion, but for how long, as well as whether arrests not leading to conviction merit exclusion; notably, an offender’s family members may also be evicted, even if they commit no crime.⁵⁴ Federal law includes classic regulatory language: a public housing agency may “establish standards that prohibit admission” of anyone the owner thinks “may interfere with the health, safety, or right to peaceful enjoyment of the premises by other residents.”⁵⁵

Parenting rights are also restricted by federal law. The Adoption and Safe Families Act of 1997 requires states to perform background checks on potential adoptive and foster parents in order to receive Social Security payments, and recommends that states bar certain classes of offenders from being foster or adoptive parents permanently and drug offenders for five years. Until a 2006 statute removed this flexibility, states were allowed to follow those rules or opt out and create their own policies; thirty-five weighed each applicant’s offense individually, while fifteen barred all people with felony records from adoptive or foster parenting.⁵⁶ (Many states consider a felony conviction evidence of a parent’s “unfitness” to supervise or care for their children, with state policies varying both in terms of the length of any such suspension and which offenses it applies to.)⁵⁷

Numerous sanctions exist in national law, then, some implemented with considerable discretion and variation by federal agencies and state governments. But state law proper contains even more restrictions on the rights, benefits, and privileges available to former offenders—hundreds in some states, according to scholars who have sought to tally them.⁵⁸ These include core political activities such as jury service. State courts handle the vast majority of American criminal and civil trials, and most states join the federal government in preventing former felons from serving on juries. As of 2003, thirty-one states barred felons from juries.⁵⁹ Meanwhile, the policy of barring felons from voting—”felony disenfranchisement”—has received a great deal of attention in recent years, and these laws too vary across state lines. The U.S. Supreme Court has said all state criminal disenfranchisement laws are constitutional, excepting only those explicitly motivated by racial bias.⁶⁰ As of 2010, two states allow incarcerated felons to vote; in thirteen, anyone who is not incarcerated may vote, but felons in prison cannot; in twenty-four, anyone incarcerated as well as some people on parole and/or probation may not vote; and in eleven states, at least some people no longer under sentence are barred from voting, either during a waiting period or indefinitely.⁶¹ State notification and restoration rules amplify the variation.

State employment restrictions can be placed into three general categories. First is public employment—six states bar all felons from government jobs permanently, while one does so for a three-year period and six others exclude only those convicted of certain offenses.⁶² Second, many states preclude felons from working in specific occupations. Some of these positions involve vulnerable populations, such as children or the elderly, but most do not. Third, states regulate occupational licensing requirements, and many of these standards either explicitly reject those with previous convictions or require that an applicant possess “good moral character,” a standard former offenders are often said to fail. By one tally, something like six thousand occupations are licensed in at least one state; jobs closed to a person with a felony conviction typically include pumping septic tanks, tending bar, mending teeth, and cutting hair.⁶³ The result is that only someone able to hire an attorney or other expert to “guide you through a panoply of bureaucratic hoops” is likely to win a state license.⁶⁴

The legal landscape regarding employment is complex. Rather than a black -or-white world in which former offenders are completely barred from scores or hundreds of professions, the reality is that a given offender may face restrictions in a given professional field, which may be waivable after a certain period, depending on various discretionary decisions and the details of state law and administrative policies. Fourteen states prohibit employment discrimination against qualified applicants solely because of criminal history; two-thirds of states have enacted some legislation requiring that a conviction be “reasonably” or even “directly” related to the work in question—for example, in order to justify termination or refusal to hire. In many states, however, that restriction applies only to public employers and licensing agencies.⁶⁵ But the most comprehensive analysis of these laws concludes that while they do “look good on paper,” their force is severely limited by the existence of a wide array of separate licensing requirements and employment restrictions and, particularly, by the absence of enforcement mechanisms.⁶⁶

The question of licensure brings us to the expanding use of criminal background checks by employers. Ample evidence shows that a criminal record “presents a major barrier to employment.”⁶⁷ Private employers’ decisions might seem to fall outside the ambit of collateral sanctions proper, if collateral sanctions are defined as consisting only of explicit governmental exclusions.⁶⁸ But in fact, consideration of a criminal record by private employers is an activity fully visible to and regulated by public policy. In effect, various state policies either tacitly or explicitly authorize the rejection of a job applicant because of past criminal activity.

First, over the last two decades states have increasingly funded and facilitated the on-line availability of criminal history information, either by maintaining databases themselves or selling data to private companies which charge a fee for searches. States then need to decide how much arrest and conviction information will be available—to whom, for how long, and with what removal or closure options after the sentence, for example. As of 2006, fourteen states allowed unrestricted access to all criminal records, while others limited such access in various ways, such as by only providing records of those currently serving sentences.⁶⁹ As many as 71 million Americans have arrest records on file in various state databases.⁷⁰ Naturally, many of those arrests never resulted in a conviction. But in thirty-six states, public and private employers and licensing agencies may inquire about and consider arrests that never led to a conviction in making hiring decisions.⁷¹ (In effect, not conviction or even indictment but arrest may legally be treated as proof of culpability and bad character.) Other states have put in place several different levels of protection for people with criminal histories, ranging from sealing of arrest records not leading to conviction to clearing and expungement of convictions.⁷²

Employer background checks sit at the margin of collateral sanctions as I am defining the term here. Unlike other consequences of felony convictions, private-employer hiring decisions are not themselves state actions, yet they are behaviors regulated by government. As one critic writes, this aspect of the collateral sanctions “rule regime” represents “not merely the government stepping aside to allow private discrimination against ex-offenders, but rather active government construction and maintenance of excessive barriers blocking ex-offenders’ re-entry into productive society.”⁷³ Indeed, U.S. courts have agreed that some such barriers are excessive, invalidating them as violations even of the deferential rational-basis standard courts traditionally employ when non-fundamental rights of unprotected classes are limited. In these cases, courts have called for more limited policies that establish a closer connection between the nature of the offense and the public-safety interest in a particular profession.⁷⁴ But the strongest such rulings have related to flat bans on public employment of all former felons, leaving other restrictions in place.

A similar state of affairs exists with regard to private rental housing. Bars on access to federally funded housing for people with criminal convictions clearly qualify as collateral sanctions. But only about 5 percent of all rental households in the U.S. are in public housing.⁷⁵ Meanwhile, private landlords have dramatically expanded the use of criminal background checks to exclude potential tenants—sometimes after having undergone “landlord training programs” run by local police departments, which may grant official certification to property owners completing the training.⁷⁶ Thus governments endorse, shape, and support this restriction: a federal program allows local housing authorities to exclude former offenders, and private property owners are sometimes trained by police in how to avoid renting to people with criminal records.

Most collateral sanctions lack an expiration date, making it difficult for a former offender ever to remove the taint of criminal conviction: serve their time, pay their debt, and rejoin society as a citizen in full. States do offer some procedures by which a former offender may petition a court or other office to lift some burdens accompanying a state conviction, whether by sealing a record, expunging a conviction, awarding certificates of rehabilitation or certificates of relief, or granting a full pardon. But as former U.S. pardon attorney Margaret Colgate Love shows, each of these procedures offers only limited relief: “[T]here is not a single jurisdiction in the country whose criminal law incorporates a formal mechanism for recognizing rehabilitation that is routinely available to all offenders who can qualify.”⁷⁷ Differing bureaucratic means of addressing collateral sanctions, and different uses of common terms, can bring serious costs to former prisoners. Despite the common use of the term “ex-felon,” for example, there really is no such status: a person must assume she is a “felon” forever. So, for example, someone able to win a business license despite a long-ago conviction is still subject to numerous other restrictions and faces serious criminal charges if she abridges those limits.⁷⁸

A “Potential Hybrid”: Collateral Sanctions and the Criminal/Civil Divide

In the 1958 case Trop v. Dulles, the U.S. Supreme Court declared that disen-franchisement—an old and archetypal collateral sanction—aims to “designate a reasonable ground of eligibility for voting,” and so “is not a punishment but rather a non penal exercise of the power to regulate the franchise.”⁷⁹ Trop’s classification of the sanction has become something of an “interpretive fact,” to borrow Ronald Dworkin’s term: a conclusion for which judges need no evidence. As David Faigman points out, though, interpretive facts can be dangerous, resisting reexamination and threatening to “petrify” despite new data.⁸⁰ In the half-century since Trop, “real-world facts”⁸¹ have transformed the impact of collateral sanctions, but Trop’s view of collateral consequences remains the consensus among American courts (though not without exception). Because such restrictions are “administered largely outside of the criminal justice system,” most courts have treated them as “mere civil regulation.”⁸²

“[I]ndividuals are hurt,” as one nineteenth-century commentator summarized the judicial view, but “such hurting is only an incident to the purpose of the statute and is not, properly speaking, a punishment at all.”⁸³ The tension in that language illustrates the deep uncertainty surrounding collateral-sanctions’ straddle of the civil and criminal law. At the most general level, there is wide agreement that these policies combine both penal and regulatory elements, but beneath the surface lies sharp disagreement and a great deal of variation.

Though most doctrine is of newer vintage, the U.S. Supreme Court has wrestled with the regulatory/penal conundrum at least since the Civil War. In the 1866 case Ex Parte Garland, for example, a bare majority of the Court rejected a statutory requirement that lawyers admitted to the Supreme Court bar take a loyalty oath excluding former Confederates and their allies. The Court concluded the oath requirement was actually a punishment—and an unconstitutional ex post facto one. “[E]xclusion from any of the professions or any of the ordinary avocations of life for past conduct,” wrote Justice Field, “can be regarded in no other light than as punishment for such conduct.” Dissenting, Justice Miller argued forcefully that taking an oath was no trial, and restriction of the bar no punishment—no more so than were the requirement that the President be native-born or that the franchise be restricted to whites.⁸⁴

In the century and a half since Garland, U.S. courts have erected a “formal-istic distinction” dividing direct and collateral sanctions.⁸⁵ Despite the Garland majority’s view, courts now define most collateral consequences as regulatory and preventive: meant “to protect society from the ex-offender’s corrupting influence,” and to “prevent the commission of future offenses by ex-offenders.”⁸⁶ Implementation aligns with purpose: when administration of a sanction by nonjudicial actors is required, a consequence is defined as collateral, indirect, and discretionary. “[W]here the consequence is contingent upon action taken by an individual or individuals other than the sentencing court—such as another governmental agency or the defendant himself,” as the Ninth Circuit put it in 2000, “the consequence is generally’collateral.’”⁸⁷ Another distinction concerns the difference between a disadvantage based on conduct and one following conviction, the focus of a prominent 1898 Supreme Court decision relating to New York’s policy of stripping the ability to practice medicine from anyone convicted of a felony.⁸⁸ Today, sex offender registration and confinement are probably the only commonly imposed sanctions explicitly premised on conduct rather than conviction.⁸⁹

A new federal statute builds into American law sharpened definitions of collateral consequences, employing a distinction first developed by the American Bar Association in 2003. Collateral “sanctions,” says the Court Security Act, are restrictions imposed automatically upon conviction, while “disqualifications” are those penalties a court, agency, or official is authorized but not required to impose.⁹⁰ This important distinction (one which, notably, does not resolve the penal/regulatory question) may gain wide currency, but this chapter employs the terms more or less interchangeably, clarifying details of implementation as needed.

Defining these burdens as civil regulations rather than punishments carries significant implications. If defined as punishment, a penalty imposed on people convicted before that restriction was enacted may violate the Ex Post Facto clause (as we saw in Garland). Moreover, a defendant must be made aware of all criminal penalties to which he is subject before he can plead guilty—meaning pleas could be invalidated if defendants were not fully informed of collateral sanctions. As a result of the 2010 Padilla decision, this is now the law with regard to deportation: if counsel do not inform noncitizens of the deportation consequences of a conviction, any subsequent plea deal can be thrown out. (Notably, however, even as it called deportation “sometimes the most important part” of the penalty levied on noncitizen offenders, the Padilla Court avoided an explicit reclassification of deportation as a direct and punitive consequence.) Since the numerous other collateral consequences faced by people with criminal convictions are not punishments, lawyers do not need to apprise defendants of the collateral consequences they face prior to a guilty plea,⁹¹ judges do not need to articulate them at sentencing (with occasional exceptions, as we have seen), and appellate courts need not inquire into their compatibility with core punitive requirements of “proportionality and desert.”⁹² In light of this, jenny Roberts concludes that courts’ separation of direct and collateral consequences has emerged from judicial perceptions of the needs of the plea-bargaining process. “By strictly circumscribing the category of direct consequences,” Roberts writes,

courts promote finality and efficiency in the plea bargain process. The fewer consequences that a defendant must be aware of prior to a guilty plea, the simpler and more efficient the plea process and the lesser the chance of a post-conviction attack upon the guilty plea based on a failure to warn.⁹³

Despite this judicial consensus, commentators from a variety of perspectives have concluded that collateral sanctions are “a legal burden constituting punishment.”⁹⁴ One authority concludes that imposing these restrictions “has become an increasingly central purpose of the modern criminal process.”⁹⁵ The National Conference of Commissioners on Uniform State Laws recently called collateral sanctions “what is really at stake, the real point of achieving a conviction.”⁹⁶ And the president of the National District Attorneys Association in 2001 described collateral consequences as “simply a new form of mandated sentences.”⁹⁷ “Virtually every felony conviction carries with it a life sentence,”⁹⁸ concludes one critical analysis.