The Yale Law Journal

March 2014

Pretrial Detention and the Right to Be Monitored

Samuel R. Wiseman

abstract. Although detention for dangerousness has received far more attention in recent years, a significant number of non-dangerous but impecunious defendants are jailed to ensure their presence at trial due to continued, widespread reliance on a money bail system.

This Essay develops two related claims. First, in the near term, electronic monitoring will present a superior alternative to money bail for addressing flight risk. In contrast to previous proposals for reducing pretrial detention rates, electronic monitoring has the potential to reduce both fugitive rates (by allowing the defendant to be easily located) and government expenditures (by reducing the number of defendants detained at state expense).

Second, despite the potential benefits to defendants and governments, electronic monitoring is not likely to be adopted by legislative or executive action. The best prospect for meaningful change is the Eighth Amendment’s prohibition of excessive bail. To achieve this goal, however, the courts will, for the first time, have to develop a meaningful jurisprudence of excessiveness to test the fit between the government’s pretrial goals and the means employed to accomplish them. This Essay begins this inquiry, arguing that the text, purpose, and history of the Amendment all support the requirement that the chosen means be, at minimum, not substantially more burdensome than necessary. Under this standard, a money bail system that leads to widespread detention without a corresponding increase in performance or savings cannot survive in the face of a less restrictive technological alternative.

author. Assistant Professor of Law, Florida State University College of Law. Many thanks to Professors Laura Appleman, Shima Baradaran, Shawn Bayern, Curtis Bridgeman, Dan Coenen, Adam Feibelman, Sam Halabi, Timothy Holbrook, Eric Kades, Jay Kesten, Ronald Krotoszynski, David Landau, Jake Linford, Wayne Logan, Dan Markel, David Markell, Murat Mungan, Garrick Pursley, Mark Seidenfeld, Christopher Slobogin, Mark Spottswood, Franita Tolson, Hannah Wiseman, and Saul Zipkin, and to participants at the Southeast Law School Junior-Senior Faculty Workshop for their valuable comments.


Innocent or not, roughly half a million people in the United States are in jail awaiting the resolution of the charges against them at any given time.1 Some of these defendants are dangerous, but a significant number are charged with nonviolent offenses and simply cannot afford relatively modest bonds imposed to assure their presence at future court appearances; roughly thirty percent of state court defendants assigned bonds of less than $5,000 are detained.2 They cannot work during the often considerable time that they spend in jail3—leaving any children and other dependents to fend for themselves—and their jobs may not be waiting for them when they get out.4Apart from the often devastating impact of pretrial detention on defendants and their families, the Attorney General has estimated that the annual cost to taxpayers is nine billion dollars.5 Nor is the bail system outstandingly effective: roughly fifteen percent of defendants released on commercial bonds fail to make at least one court appearance.6 Responding to these problems, both the Conference of Chief Judges and the Conference of State Court Administrators have recently called for the use of more accurate pretrial assessments of dangerousness and flight risk, and for the release of non-dangerous defendants.7

Although rising detention rates and shrinking governmental budgets have recently brought these issues wider attention, their basic contours have not changed for decades.8 With the rapid advance of computing technology, however, the available solutions have changed a great deal. Increasingly sophisticated remote monitoring devices have the potential to sharply reduce the need for flight-based pretrial detention. In a world in which scientists can monitor and recapture wolves,9 snakes,10 and even manatees11 in the wild, and AT&T Wireless offers family-member tracking for $10/month,12 the question of finding other ways of ensuring a non-dangerous defendant’s presence at trial is one not of ability, but of will—albeit a difficult one. By reducing jail populations, these technologies can lower overall costs—it costs at least four times as much to jail a defendant as it does to monitor him13—and, while invasive, are vastly preferable to a jail cell for most defendants.

Among bail reform advocates, however, monitoring has relatively few vocal proponents, perhaps due to understandable, but (I will argue) probably overstated, privacy concerns. Indeed, the problems addressed in this Essay are in some respects the reverse of the usual concerns about criminal justice technology. The rapid advance of technology has been accompanied by a corresponding increase in legal scholarship concerned about its effect on the relationship between government and society. And not without reason, of course: increasingly efficient, inexpensive, and nearly invisible methods of surveillance and control have the potential to radically alter that relationship, and law and lawyers are appropriately concerned with preventing an Orwellian disaster.14 Largely omitted, however, from these larger debates is discussion of where new government technology would be most beneficial.15 Although placing GPS monitors on the free population would impose enormous privacy costs, for those whom the government is already allowed to imprison in pursuit of its goals, technology that allows those goals to be achieved less obtrusively is a nearly unalloyed good.

More critically, as previous generations of bail reformers found, motivating governments to act for the benefit of the poor and unpopular against the will of the commercial bail industry is an arduous task; at the legislative level it is often nearly impossible.16 However, for the many detainees who are jailed not for dangerousness but for flight risk, there are meaningful constitutional and statutory grounds for future judicial intervention. A right to pretrial monitoring fits squarely within existing state and federal statutes requiring courts to impose the least restrictive conditions of release.17 More fundamentally, although the Bail Clause has recently been somewhat neglected by both courts and scholars, this is precisely the type of problem it was meant to address: monetary conditions resulting in detention are excessive when equally effective, and cost effective, alternatives for reducing flight risk are available. That is the argument I advance here: (1) that poor, non-dangerous criminal defendants are a discrete constituency in need of judicial protection because they are effectively locked out of the political process (an argument vividly illustrated by their current treatment under the Bail Reform Act), and because historic efforts at reform have repeatedly failed; (2) that, while further studies are needed, electronic pretrial monitoring is, or will soon be, a less expensive, less burdensome, and judicially administrable alternative to money bail for ensuring appearance at trial; and (3) that the emergence of this alternative necessitates a new jurisprudence of excessiveness under the Eighth Amendment prohibition of “[e]xcessive bail.”

The Supreme Court, although it has defined the narrow substantive bounds of the Excessive Bail Clause, has failed to clarify how close the fit must be between the reasons for restricting pretrial liberty and the burdens imposed. Nor has the Court explained whether the analysis must consider reasonable alternatives not provided by the legislature—thus leaving in question the definition of “excessiveness” itself. As the number of pretrial detainees continues to rise and monitoring technology improves, these questions are likely to become vital, and as an initial step towards resolving them I suggest that a standard resembling, at a minimum, intermediate scrutiny is warranted. Under this approach, maintaining a money bail system that consistently results in detention for poverty is substantially more burdensome than necessary if an equally effective and efficient option that does not rely on detention exists.

Part I of the Essay documents the serious problems with the current system and the limitations of the most commonly proposed alternatives. Part II provides an overview of existing monitoring technologies and their implementation and presents the argument that the benefits of replacing detention for flight risk with the use of pretrial monitoring far outweigh the costs. It concludes that the usual objections to government monitoringthe intrusion on individual privacy and the threat of surveillance extending to new segments of society18have relatively little force in the pretrial context, where detention currently all but extinguishes privacy interests and the number of criminal defendants is largely independent of the means of preventing flight. Part III lays out the doctrinal avenues for achieving a right to pretrial monitoring, defining the clearest and most likely path toward the right. In addition to a strong case under current—but amendable—state and federal statutes, there are powerful doctrinal, textual, and historic reasons to conclude that the constitutional prohibition against excessive bail includes a right, for many non-dangerous defendants, to have the option of electronic monitoring in lieu of imprisonment.

The argument is more modest than it might initially seem—it would be an incremental step in line with existing Eighth Amendment doctrine and would require only that, once proven cost-effective relative to the current system, certain forms of monitoring be available to defendants who would otherwise be detained for risk of flight, not for dangerousness.19 Recognizing that there is an important debate about the government institutions best suited to respond to technological change, however, Part IV explains why courts are a proper venue for reform. The long, mostly sad history of bail reform efforts suggests that between a perhaps-justified legislative fear of being seen as soft on crime and the existence of an entrenched public- and private-sector lobby opposed to change, prospects for legislative adoption of a monitoring alternative are likely to be dim in many jurisdictions.

I. flight risk, pretrial detention, and the need for alternatives to money bail

Historically, the U.S. system of bail and associated pretrial detention was employed solely to prevent pretrial flight,20 but increasingly, the many individuals awaiting trial in jail are detained because a judge has deemed them potentially dangerous.21 Although this type of detention raises serious constitutional concerns, the liberty22 and privacy burdens placed on this subset of detainees seem, to some extent, intuitively reasonable; if evidence suggests that individuals could jeopardize the safety of their community while they awaited trial, their detention might be merited.23 In light of the recent rise in officially sanctioned detention for dangerousness, however, the modern literature addressing the problems with bail tends to focus on this group of pretrial detainees, highlighting the problems with predicting dangerousness, the expansive judicial discretion allowed within this predictive process, and suggesting better constitutional protections.24 Nonetheless, the Department of Justice estimates that non-dangerous defendants make up approximately two-thirds of the 500,000 defendants held pretrial in jails at any given time.25

These individuals are the product of a long tradition of money bail in the United States.26 Since the founding of this country, judges have required individuals to post some form of collateral27 in order to incentivize them to appear at a trial that they strongly wish to avoid—a process that could ultimately lead to their conviction and imprisonment. This system of money bail is an archaic institution, a holdover from times when there were few police officers and jails and when fleeing across a county or state line was more likely to be an effectivemeans of avoiding trial;28 requiring an individual or family members to post something of value was a necessary and reasonable means of preventing flight.

Recent, extensive changes in technology, such as the rise of Internet photos and enhanced police communication, have greatly decreased flight incentives, and technologies such as GPS monitoring also allow the police to easily monitor those individuals who still have an incentive to flee.29 Yet money bail still dominates the pretrial process in most states.30 This system typically employs both personal bonds, in which an individual, friends, or family members post the money or a percentage of the money with a court, and commercial bonds, in which a bondsman becomes responsible for the amount of the bond and charges the defendant a percentage of the bond amount as a fee;31 both types are exceedingly problematic. Money bail is increasingly not an alternative to pretrial detention but rather an enabler of the practice: as bail amounts are set higher, and as financial inequalities become wider in the United States, many individuals cannot pay and are thus detained while awaiting trial.32 Increased pretrial detention harms poor defendants and their families, leads to greater recidivism, and uses up scarce criminal justice resources. These pervasive problems, explored in further detail below, create a pressing need for an alternative to money bail and associated pretrial detention of non-dangerous defendants. Advancing monitoring technology will soon, if it does not already, provide this alternative.

A. The Burdens of Pretrial Detention

Non-dangerous individuals jailed to prevent flight suffer the same harms as those detained for safety reasons—the same harms suffered by convicted defendants.33 They are taken from their communities and physically barred from the outside world, restricted to limited visits by family members and attorneys.34 Their conversations are constantly monitored by guards and other inmates, their mail is searched, and they are subjected to frequent and invasive searches and pat-downs to ensure institutional security.35 To compound the gravity of the harm, these high liberty and privacy burdens are often prolonged; despite speedy trial requirements, many defendants awaiting trial are detained for months.36

Being jailed also has a variety of more quantifiable negative effects. It increases the likelihood that detainees will commit future crimes, substantially impacts the quality of their defense, and encourages plea bargains—all of which increase the likelihood that the detainee will be convicted, imprisoned, and subjected to prolonged deprivation of liberty, privacy, and other fundamental elements of human existence.

1. The Criminogenic and Plea-Inducing Effects of Pretrial Detention

Many inmates detained pretrial have been accused of low-level or non-violent crimes,37 yet they are jailed with convicted criminals and potentially dangerous defendants who await trial.38 Predictably, incarceration multiplies the chances that the accused will learn criminal behavior.39 Those accused of drug possession may develop new addictions, and non-violent criminals may quickly learn violence (if only to defend themselves at first40). As months pass and new defendants arrive, desperation may set in, leaving a potentially permanent mark and possibly lingering violent tendencies.41

Moreover, the current pretrial system produces false convictions in addition to training real criminals. In the mid-1960s, the Manhattan Bail Project led by the Vera Foundation concluded that “a person’s inability or unwillingness to post bail may result in more than a temporary deprivation of his liberty,”42 finding that those detained pretrial were more likely to be convicted and imprisoned than those released on bail, regardless of whether they had been previously charged or imprisoned.43 This trend has continued, leading some to conclude that “[t]he most glaring concern of the pretrial detainee is the large percentage of detainees who are eventually found guilty.”44 While this could simply suggest that judges assessing flight risk and dangerousness are also accurately predicting guilt, further research suggests several other likely contributors to this trend, which are troubling from an equality perspective—and, of course, with respect to defendants’ long-term liberty interests. One factor is the substantial difficulty faced by a pretrial detainee attempting to mount a successful defense from a jail cell.45 The defendant must recruit friends or family members to collect evidence and witnesses and will often have difficulty communicating with his attorney due to limited visiting hours.46 The difficulty of preparing an adequate defense makes the likelihood of success at trial much lower for pretrial detainees than for those who have secured release and have avoided the stigma of a prison cell.47

Faced with these high defense burdens, defendants jailed pretrial often accept plea bargains in lieu of persevering through trial. In some cases, the periods that defendants spend in jail awaiting trial is comparable to, or even greater than, their potential sentences,48 thus substantially incentivizing quick plea deals regardless of guilt or innocence. One empirical study found that of the federal pretrial detainees in 1987 and 1988, about eighty-five percent were criminally convicted, and that the majority of these convictions appeared to have “resulted from some form of plea bargaining.”49

2. Financial Harm to Defendants and Their Families

Even if detention does not lead to a conviction, it places significant financial costs on detainees and their families, who, in addition to suffering the stigma of having a loved one in jail, are also deprived of the detainee’s financial support.50 Many detainees lose their jobs even if jailed for a short time,51 and this deprivation can continue after the detainee’s release. Without income, the defendant and his family also may fall behind on payments and lose housing, transportation, and other basic necessities.52 More broadly, the removal of productive workers from the labor pool negatively affects the economy. As Attorney General Holder recently noted, nonviolent defendants “could be released . . . and allowed to pursue or maintain employment, and participate in educational opportunities and their normal family lives—without risk of endangering their fellow citizens or fleeing from justice.”53

3. The Tax Burden

High pretrial detention rates do not only impose high costs on defendants, but also on the public. During the recent economic downturn, the cost of money bail to society has been raised as a more practical rallying flag for reform.54 The American Bar Association notes that “the taxpayer implications of pretrial detention are significant given the expenses of operating detention facilities,” observing that “New York City spends approximately $45,000 annually to house a single pre-trial detainee.”55

Accurately assessing the exact costs of pretrial detention is itself difficult. Fixed costs of prisons must include the expense of housing both convicted criminals and pretrial detainees, and it is difficult to identify the point at which the number of pretrial detainees, in isolation, forces construction of a new facility. It is clear, though, that some states and counties have had to build new jails to accommodate burgeoning populations. The Baltimore City Detention Center, for example, in which ninety percent of women held are awaiting trial, is planning a new $181 million facility to accommodate more inmates.56The variable costs of pretrial detention are somewhat better known. Although they differ by jurisdiction, the costs of feeding, clothing, securitizing, and providing medical care for millions of pretrial defendants are high. Daily estimates range from $50 in Kentucky57 to $85 in Florida58 and $123 in New York.59 Additional estimates suggest that jail costs range from $84 million60 to $124 million61 or even $860 million62 annually.

B. The Problems with Money Bail

The broad liberty and privacy implications of pretrial detention are somewhat easy to pinpoint: individuals presumed innocent are deprived of most freedom of motion and interaction with family and community, they are subjected to deplorable privacy intrusions, and they face a higher likelihood of prolonged liberty deprivations. The specific harms of money bail are also deeply problematic, however, and have been explored and analyzed in-depth since the early twentieth century.63

1. Wealth Discrimination

Just as scholars have long noted the inequalities faced by defendants detained pretrial and those rich enough to be released, they have documented how money bail specifically contributes to this divide, observing in 1927, for example, that “[t]he amount of bail in a given case is determined arbitrarily and with little or no regard to the . . . financial ability of the accused.”64 A later survey of the Philadelphia bail system concluded:

One purpose for imposing a higher [bail] amount which would be consistent with the theory of bail would be that the increase in the defendant’s financial stake reduces the likelihood of non-appearance at his trial. In practice, however, higher bail usually means that appearance in court is being obtained by holding the defendant behind bars.65

A 1956 New York project observed similar inequalities, noting that even at a seemingly low bail amount of $500, “only three out of four defendants obtained pre-trial release, while at $7500 and above only one defendant in seven [was] able to post bail.”66 Yet for the felony prosecutions studied— 2,292 of which involved the setting of bail—69 cases were set at $7,500, and many more cases (732) were set at points “at which most defendants cannot post bail” ($2,500 and above).67

Unfortunately, these conditions have not meaningfully changed: in a study of data from 1990 to 2004, the Bureau of Justice Statistics noted a “direct relationship between the bail amount and the probability of release.”68 When bail was set at $100,000 or more, only one in ten defendants was released; at bail amounts between $10,000 and $24,999, approximately forty-five percent of defendants who received bail were released; and only when bail dropped to $5,000-$9,999 did more than fifty percent of defendants who received bail obtain release.69 Simply put, defendants without assets cannot obtain bail. As the American Bar Association concludes, “Detaining persons simply because they cannot afford bail is unwarranted . . . .”70 Despite the growing criticisms, judges in most states still do exactly that.71

2. Direct Financial Burdens on Defendants and Their Families

Even if a defendant is able to post bail and avoid pretrial detention, the result is often disastrous financially. Although bail is sometimes based on a defendant’s ability to pay, it is also largely determined by fixed bail schedules,72 which assign specific monetary amounts based on the charges lodged. This often forces indigent defendants and their families to spend money that otherwise would have covered basic necessities. The many challenges faced in this uphill battle of collecting adequate funds for bail are in some cases dramatic: in Wisconsin, for example, police allegedly confiscated the bail money that a family had managed to piece together through loans from friends and co-workers and various ATM visits because the police believed that the money collected was connected to the drug activity with which the defendant was charged.73

In sum, money bail and the high pretrial detention rates associated with this antiquated system impose high burdens on defendants, families, and society. This tragedy is not unavoidable, but rather has been perpetuated despite the availability of technology to meaningfully enhance equality, privacy, and liberty, and at a much lower cost than the current system.

C. Not Worth the Cost: The Ineffectiveness of Money Bail

From a bird’s-eye view of the U.S. system of money bail and associated pretrial detention for flight, one might assume that the high burdens imposed by this system are justified by its effectiveness—or perhaps a lack of feasible alternatives.74 In fact, however, neither effectiveness nor a lack of alternatives justifies this costly system.

A non-negligible percentage of defendants flee despite having posted large bonds. In the seventy-five largest counties in the country, twenty-one to twenty-four percent of state court felony defendants who were released on bail or personal recognizance between 1990 and 2004 failed to appear at trial.75 Twenty-five percent of the defendants who failed to appear had been released on surety bond;76 of all defendants released on surety bond during this time, there was an eighteen percent failure to appear rate.77 While this failure rate was lower than that of defendants on emergency release (forty-five percent of defendants released on an emergency basis failed to appear) and unsecured bonds (thirty percent of those released under this type of bond failed to appear),78 it shows that bail bonds of any type do not perfectly achieve their goal of ensuring a defendant’s presence at trial; the system tolerates a relatively high level of failure as compared to the alternative of jailing all individuals, which would guarantee nearly perfect appearance rates.

This failure is likely due to flight incentives that remain despite technological advances in tracking and monitoring defendants. Although there is no longer as high a likelihood of avoiding conviction by escaping across a state or county line, as the police will eventually detect and track down the defendant,79 the temptation remains. The common use of fixed bail schedules80 contributes to the problem: in addition to placing unfair burdens on indigent defendants charged with pricey crimes, it leaves rich defendants charged with the same crime in a relatively easy financial condition. A crime with a fixed bail rate of $50,000 is expensive for a poor man, in other words, but relatively cheap for someone with adequate funds. The wealthier individual may not think twice about absconding and forever forfeiting these funds, particularly if the alternative—a long jail sentence—has a particularly high cost.

In commercial bail states, individuals may also be incentivized to flee despite low chances of success because their bondsmen have insufficient incentives to monitor them. In the majority of states that allow commercial sureties, the system relies largely on private entities to track down individuals and ensure their appearance at trial. The defendant pays the bondsman a percentage of the bail set as a fee, often along with additional collateral, and the bondsman posts the bail.81 Depending on the size of the collateral, even defendants of reasonable means may have relatively little incentive to stay in a jurisdiction.82 Bondsmen, in turn, will only worry about funds that they have put down to the extent they think that the court will collect it upon the defendant’s failure to appear. Yet many courts have been lax about declaring bonds forfeited when defendants flee, thus allowing bondsmen to keep the money and reducing private incentives to monitor defendants.83

D. Traditional Alternatives to Money Bail

Unfortunately, the non-technological alternatives to money bail, which are often proposed as options to reduce the inequality, liberty, and privacy burdens associated with bail and pretrial detention, are similarly ineffective. Studies suggest that release on personal recognizancea promise by the defendant that he will appearis, unsurprisingly, less effective in securing presence at trial.84 Pretrial supervised release programs, which couple release without bond with different combinations of, inter alia, required check-ins, travel and curfew restrictions, monitoring, and/or treatment and classes, can be highly effective, and for many bail reform advocates they have been the preferred option.85 But because these programs typically rely on humans for supervision, job training, drug treatment courses, and sometimes for their monitoring, they can also be expensive86 and limited in the number of defendants that they can accept. Persuading policymakers to increase funding, often in the face of opposition from the bond industry, has in many cases been difficult or impossible, resulting in the current, bad situation. An inexpensive, effective alternative to money bail does—or likely soon will—exist, however, in the form of technological monitoring.

II. electronic monitoring as an alternative

Increasingly advanced technologies are able to closely monitor pretrial defendants’ locations while granting them far greater freedom—and with it the opportunity to continue working, consult with attorneys, and spend time with their families.87 Indeed, in recent years U.S. and international jurisdictions have deployed monitoring technologies both pretrial and post-trial for thousands of defendants.88 While these monitoring programs, described in greater detail below, represent a promising start, electronic monitoring has yet to meaningfully supplant pretrial detention for flight risk. There are numerous obstacles to that goal, including practical concerns as well as the likelihood of entrenched opposition from the bail industry. With this in mind, after describing existing technologies and programs, this Part addresses likely concerns about monitoring’s effectiveness, costs, and impact upon liberty, privacy, and equality. These issues will have legal significance, because, as developed in Part III, both the statutory and constitutional arguments for a right to monitoring depend on demonstrating that monitoring is at least as effective and inexpensive as money bail.89

A. Technologies and Implementation

Defendants and offenders in the United States and Europe have been electronically monitored since the 1980s, and monitoring has since spread to a limited number of other countries.90 As early as 1983, one judge required an offender in New Mexico to be confined to his home and monitored with an electronic bracelet that sent signals to his home,91 and in 1985, Palm Beach County deployed one of the first electronic monitoring programs using radio beepers—for convicted, not pretrial, defendants.92 Indeed, although it is counterintuitive, as monitoring seems better suited for locating fugitives than controlling their behavior,93 electronic monitoring is more widely used in sentencing.94

Nonetheless, electronic monitoring has a long history of pretrial use. In the late 1980s, Marion County, Indiana, ran an experimental program of pretrial home detention and electronic monitoring for those who could not afford bail or meet release on personal recognizance conditions. Discussing the benefits of the Marion County program, Indiana University professors note that “awaiting trial at home is less restrictive than confinement in jail” and that the program allowed “offenders to maintain employment and ties to their families.”95 And in 1991, Federal Pretrial Services began a national, pretrial home confinement programming using electronic monitoring.96 Monitoring was introduced in Europe around the same time.97

Current electronic monitoring technologies take several forms. In Europe, the most common monitoring systems use radio devices combined with home curfews.98 In continuous-signal curfewed monitoring systems,99 individuals wear a tag on their ankle, which sends a signal to a receiver attached to the individual’s phone.100 The individual is typically confined to the home during certain hours, and a 24-hour monitoring center, using data from the receiver, can track when the individual is at home and whether the equipment has been tampered with.101 Other monitoring does not rely on confinement to the home but rather requires periodic check-ins through “voice verification” or another means of proving location.102 In the United States, Federal Pretrial Services uses both radio and GPS tracking devices to enforce home confinement and other conditions of supervised release,103 along with frequent, required interactions with supervising officers.104 Cook County, Illinois, has used electronic monitoring—a radio signal and home monitoring unit—for more than 250,000 non-violent defendants since 1989, some of whom were released in the pretrial context.105

House-arrest models, however, are both more restrictive and, when not combined with real-time monitoring, likely less effective106 than the active tracking107 of individuals using GPS satellite technology, which has become more common in recent years.108 Mesa, Arizona, for example, releases and electronically monitors certain defendants pretrial using GPS satellite tracking devices.109 And Strafford County, New Hampshire, tracks certain defendants on pretrial release (as well as sentenced offenders in community supervision) using GPS systems that allow “officials to know within 10 meters where a person has been throughout the day.”110 Private firms have also begun to offer GPS monitoring to help wealthy defendants avoid pretrial detention.111

Although active monitoring can be limited by the availability of the cellular telephone networks through which the device transmits location data,112 it appears to be the best current option for both defendants and governments: its accuracy deters flight and allows fugitives to be readily located, and it is much less restrictive than a curfew requirement. Indeed, at least for relatively low-risk defendants, it potentially need only be actively (as opposed to periodically) monitored once a defendant has failed to appear for trial. And other technologies may emerge in the near future. The advent of phones capable of mobile videoconferencing and Google Glass,113 for example, suggests that live audio-video monitoring may be a possibility in the future, presumably for the highest flight risk defendants.

B. Effectiveness

One concern about the use of monitoring technology in lieu of pretrial detention for failure to post bond is purely practical: that it will never be totally effective at eliminating failures to appear.114 Of course, the effectiveness of any given monitoring program at reducing flight risk is an empirical question, and while, as discussed below, existing technology shows promise, no conclusive empirical evidence of effectiveness currently exists (and with respect to future innovations, obviously cannot). The sparse empirical studies addressing the cost and effectiveness of monitoring have, as a result of the predominance of post-conviction monitoring, focused largely on that context.115 In the United States, there are few studies of the effectiveness of monitoring pretrial, and the limited research available tends to involve small sample sizes.116 Federal courts have generally been positive in their assessment of the Federal Pretrial Services location-based implementation of monitoring,117 although, as noted above, it is typically combined with a high degree of supervision.118

The European literature is slightly richer, although still inconclusive. In a pilot study conducted in England between 1998 and 1999, judges imposed conditional bail with monitored curfews119 on a select group of defendants—in some cases, directly in lieu of pretrial detention.120 Of the 173 individuals who received monitoring curfews, researchers collected data on 118 individuals, eleven of whom absconded121—a failure to appear rate “lower than national and local figures” for other forms of bail.122 Other European studies suggest potentially positive results, although, again, not producing any firm empirical conclusions. In Portugal, very early results of a small pilot of a bail curfew and electronic monitoring program showed “no relevant non-compliances, nor revoked orders” in 2002, from a total of 39 participants.123 In Scotland, however, a study of the country’s bail monitoring pilot showed more compliance problems; in 31 of the 63 monitored bail orders completed, defendants were accused of breaching bail conditions or committing new offenses.124

Studies of post-trial monitoring in Europe and the United States also suggest potential success in terms of individuals completing their programs without recidivating.125 These statistics are not easily compared with the ability of monitoring to prevent flight, however—the core purpose of the monitoring proposed here.

Further study—particularly of the use of active GPS tracking in place of pretrial detention—will be essential to convincing wary judges and legislators. But the potential of advanced tracking technology to reduce flight risk and aid in fugitive recovery appears enormous. Anecdotally, this intuition is supported by the use of GPS monitors by bondsmen themselves,126 as well as by recent high-profile examples of GPS monitoring as an alternative or addition to bail, including for arms dealers, gangsters, and financial fraudsters.127 A judge initially ordered Bernie Madoff, for example, to wear a GPS monitoring ankle bracelet in addition to paying $10 million in bail and remaining on nightly house arrest.128 Dominique Strauss-Kahn was similarly granted bail and assigned a GPS electronic ankle bracelet—along with house arrest, armed guards, and “24-hour video monitoring of every door”—at Strauss-Kahn’s expense,129 leading Slate magazine to the conclusion advanced here: “Most defendants don’t run the International Monetary Fund. They don’t have citizenship in non-extraditing countries or standing arrangements to board any Air France flight. They don’t need guards to keep them from escaping justice. They just need an ankle monitor.”130

Nonetheless, it might be argued that no amount of high-tech monitoring will ever be as effective at ensuring a defendant’s presence at trial as detaining the defendant (which is, of course, nearly 100% effective). As Blackstone observed, defendants facing the most serious penalties could not be bailed because of their great incentive to flee: “in . . . offenses of a capital nature, no bail can be a security equivalent to the actual custody of the person. For what is there that a man may not be induced to forfeit, to save his own life?”131 This will almost certainly be true of monitoring as well. No matter how ingenious the technology, it is likely that highly motivated defendants will find a way to defeat it, perhaps by damaging or removing the tracking device or by blocking its signal.132 Technology, then, cannot completely eliminate pretrial detention for flight risk; at most, by being more effective than money bail, it could narrow the class of defendants considered too great of a flight risk to release (most of whom, under contemporary practice, would be detained for dangerousness anyway). But this is not a particularly serious objection: the principal beneficiaries of replacing money bail with monitoring are not those who, facing serious charges, have too much at stake to be released, but those who, facing less serious charges, simply have too little to stake. These concerns, moreover, can also be addressed by imposing higher penalties for failing to appear while monitored or for tampering with a monitoring device.133

There will likely be missteps, in the form of malfunctioning technology and fugitive defendants, along the way to widespread deployment as an alternative to pretrial detention.134 But, in the near term, it has the potential to effectively replace unmeetable monetary requirements for non-dangerous defendants. Technology might not be able to completely eliminate detention for flight risk, but it should be able to eliminate detention for poverty.

C. Cost

As the American Bar Association and other organizations have begun to emphasize the expense of pretrial detention,135 the practical benefits of the technological alternative have become even more compelling. Increasingly computerized, they do not require the staff, medical programs, and vast security controls of pretrial detention. Monitoring programs appear to generate significant savings if used in place of pretrial detention, although some of the savings may be lost if convicted defendants are not given time-served credit for time-monitored, and thus eventually spend the same amount of time incarcerated.136

Pretrial services programs that combine technology with relatively inexpensive monitoring have substantially reduced the financial cost of preventing flight. Miami-Dade County cut costs from approximately $20,000 per pretrial defendant to $432 annually for released, monitored defendants,137 and the Southern District of Iowa saved $1.7 million over one fiscal year by releasing 15% more defendants.138 Federal active monitoring of pretrial defendants in the 1990s cost approximately $2.77 to $9.04 daily,139 compared to daily costs of pretrial detention ranging, according to some estimates, from $50 to $123.140Other estimates suggest that electronic monitoring programs “[o]n average . . . cost between five and twenty-five dollars per day.”141 And approximately one out of three offenders who were electronically monitored in Florida between 2001 and 2007 would have otherwise been jailed at six times the cost, according to one study, which concluded that monitoring was a “cost-effective method of dealing with offenders.”142 Similarly, results from Europe also suggest that monitoring can be far less expensive than other options if implemented properly—ensuring that monitoring is implemented in lieu of jail, thus offsetting costs.143

As GPS, live audiovisual monitoring, and other technologies become more common outside of the criminal world for ease of navigation and of sharing life experiences with friends and family, costs likely will continue to decline, while effectiveness will rise. Governments need not operate the programs themselves: already, multiple competing private providers exist (and one could imagine bond agents, some of whom already use tracking devices, becoming monitoring agents).144 The cost-effectiveness of a monitoring program, of course, will depend on the details. A minimalist system, designed to track the location only of those who have already failed to appear and giving at least partial time-served credit, which is clearly more desirable from a privacy perspective, will also be more cost-effective than a more intrusive program145 without credit. And, for better or worse, it is likely that monitoring programs will shift pretrial flight prevention costs to defendants; some defendants in pretrial release programs already pay for the cost of their own monitoring.146If electronic monitoring is implemented on a broader scale, more legislatures will try to recoup the costs of monitoring from indigent defendants as they have done with counsel147 and jail costs.148 As others have noted, this is deeply problematic,149 but it is still preferable to detention (which defendants might also have to pay for).

Empirically, the cost savings of monitoring in lieu of detention require further detailed investigation. The goal here is not to suggest that monitoring is completely effective or costless, but rather that the available data suggest that it can be at least as cheap and effective as money bail.

D. Privacy and Net-Widening

Another, more fundamental set of reservations centers on privacy. The degree to which a monitored defendant’s privacy is invaded depends on the technology employed—a device that transmits location data only on the day of a court appearance is less invasive than one that transmits constantly, and both are far less invasive than a device that transmits audio and video. But even the most limited version is a serious intrusion, and privacy concerns almost certainly explain why monitoring technologies have not so far been widely heralded by academics and criminal justice advocacy groups as a solution to the serious and seemingly intractable problems with money bail and pretrial detention described above.150

Focusing solely on defendants who would otherwise be detained for failure to post bond, privacy objections have little purchase. Even the most thorough observation—even if it causes defendants to carefully monitor and restrict their behavior in order to limit the government’s knowledge of their lives—would for most defendants almost certainly be preferable to imprisonment. Agence France-Presse, for example, described Strauss-Kahn’s ankle bracelet as a “symbol of shame for the beleaguered global finance titan,151 but even a high-profile figure like Strauss-Kahn apparently preferred shame (and constant surveillance) to imprisonment. In one study of those subject to home curfew and monitoring, the most common complaints voiced included “[n]ot being able to go to the store when you want” and “[n]ot being able to go out to eat when you want,” followed by “[h]aving to wear a visible monitor.”152 These are significant deprivations, but, unsurprisingly, “most electronically monitored offenders prefer house arrest to jail.”153 A fortiori, a less intrusive, curfew-less monitoring regime would also be preferable to jail.

This calculus holds even if, leaving aside the tremendous increases in liberty and physical and psychological well-being, privacy is used as the sole criterion. Being in jail, after all, involves not only near-constant surveillance by guards, but also by fellow inmates. And in the absence of other realistic options for systemic reform, the perfect must not be the enemy of the good.

Not surprisingly, then, opposition to the use of monitoring technology has largely focused not on those already subject to a high level of government surveillance, but on the risk that technology will allow the government to surveil more people154: ever cheaper and more powerful monitoring equipment lessens resource constraints, and the physically unobtrusive nature of the monitors themselves lessens political and constitutional opposition.155 These net-widening concerns are slippery slope arguments—the use of monitoring in a given context may not be bad in itself, but it will lead to the use of monitoring in other, more objectionable contexts. And as Frederick Schauer observed, slippery slope arguments are empirical arguments.156 Thus, the question is whether the benefits of replacing pretrial detention for failure to post bond will outweigh the harms from any resulting increase in the number of people monitored. These net-widening concerns will be considered in two parts, first across society at large, and then among pretrial defendants.

In many cases, of course, there is good reason to be skeptical of the expanded use of surveillance and control technology in criminal justice. Although the precise causes of our astoundingly large prison population are disputed,157 the costs of physical imprisonment impose at least a weak restraint on our desire to incarcerate. Technology eases that restraint, and since political support for crime control measures remains strong, and the Supreme Court has taken an extremely limited view of the privacy rights of those convicted of a crime,158 this allows government to extend its control over a far wider group. Thus, for example, as Wayne Logan159 and Erin Murphy160 have noted, custodial “treatment” of dangerous sex offenders is quite expensive, but monitoring is not,161 allowing non-dangerous offenders to be swept into the net. For the many who, quite reasonably, think that government already exerts too much control, particularly over vulnerable groups, concern over advancing technology is generally justified.

Although the prospect of the government requiring all citizens to wear ankle monitors or, worse, a pair of Google glasses that upload everything we see and hear to the Internet, is indeed a frightening one, using monitoring in lieu of detention is unlikely to move us meaningfully closer to that nightmare scenario. Crucially, the expanded use of monitoring technology in the pretrial context does not have the same potential to directly increase the number of people subject to the power of the state–that is, the number of pretrial defendants.162 This is because the costs of pretrial supervision of non-dangerous defendants, whether in the form of traditional pretrial services programs, monitoring, or detention, although sizeable, simply do not play a meaningful role in determining the number of prosecutions. Relative to the costs of maintaining police departments, prosecutors’ offices, and prisons (or their future technological alternatives), they are a drop in the bucket. The same may be said with respect to the political costs of maintaining a system in which the potentially innocent are thrown in jail with convicts for lack of funds. Thus, even a drastic reduction in the resource and political costs of pretrial supervision would be unlikely to cause a measurable increase in the number of criminal defendants. To be sure, as discussed below, it may affect the amount or form of that supervision, but the size of the population potentially subject to it will continue to be determined by larger factors, such as policing levels and policies (and thus arrest rates) and incentives for charging rates within prosecutors’ offices.

Nonetheless, there remains the worry that broader use of monitoring in the pretrial context will lead to its expanded use in other contexts, including probation and parole. As the supply of monitoring technologies increases, competitive production expands, and manufacturers learn new and cheaper production methods, the cost of monitoring is likely to shrink, making it generally more appealing. More subtly, it could increase familiarity with, and acceptance of, monitoring technology among both law enforcement and the public. Although the rapid proliferation of location-tracking cell phones and services arguably poses a far greater threat in this regard than pretrial monitoring,163 this concern cannot be easily dismissed. Ultimately, while future proposed uses of monitoring technology should be carefully scrutinized, this necessarily somewhat vague threat should not prevent its use to help the very real people currently in jail.

Turning to the narrower class of pretrial defendants eligible for release, the risk is that expanding the use of monitoring as an alternative to detention will lead to the increased use of monitoring on defendants who would previously have been released on bail, personal recognizance, or other less restrictive conditions.164 And, to some extent, it likely would: once a monitoring infrastructure is in place, the marginal cost of adding to the monitored population is likely to be relatively low, and if monitoring is more effective at producing presence at trial than the alternatives, policymakers will have an incentive to use it. As discussed above, expanded use will lead to greater economies of scale. Moreover, increasing the demand for monitoring technology may lead to a corresponding increase in the financial ability of its producers to lobby governments for further expansion.165 These fears appear to have been partially realized in England and Wales.166 Nonetheless, the potential harms should not be overstated. Many defendants are likely to be monitored in the future regardless of whether technology is used as a replacement for flight-risk detention. Indeed, electronic monitoring by pretrial services departments is increasingly imposed as an additional condition of release, while private bondsmen have begun exploiting monitoring as a means of protecting their investment.167 (As discussed in Part IV below, this is an unsurprising outcome given the interests of both bondsmen and technology purveyors in maximizing their profits.) To the extent that wider monitoring of non-dangerous pretrial defendants is probable in any case, net-widening concerns diminish. There are some ways to combat net-widening within the class of pretrial defendants. Perhaps most significantly, granting time-served credit, whether in full or part, for the monitoring period would both acknowledge the very real privacy cost to the defendant and likely reduce the incentive to use monitoring in place of non-incarcerative options. Maintaining a money bail option for those able and willing to pay for it could help as well—indeed, this is advisable from both a privacy and political economy perspective. Finally, the Scottish experience suggests some cause for optimism about the ability of legislation to control net-widening. Judges in the small trial program there were instructed to consider monitoring only as an alternative to detention, and this appears to have been effective.168

On the whole, then, Orwellian fears about monitoring—however well justified elsewhere—are not as strong in the context of its use as a substitute for pretrial detention for failure to post bond. From the perspective of the defendant who would otherwise sit in jail, the privacy and liberty gains are immense. Larger segments of society are unlikely to be snagged by the criminal justice system as a result. And while some released defendants may be monitored who would otherwise not be if monitoring were to replace flight-risk detention, the liberty and privacy costs must be weighed against the benefits to those who would otherwise not be released. Similarly, if monitoring decreases the marginal cost of arrests by reducing jail costs, arrests may increase—a benefit if more murderers are caught, but, in the eyes of many, a cost if more low-level drug offenders are arrested.169 The exact balance of this tradeoff is difficult to predict, and it depends, inter alia, on the form of monitoring employed—the more invasive it is, the lower the benefit to the newly freed and the greater the harm to the newly monitored. As discussed below, the doctrinal bases for courts to limit the extent of flight-risk monitoring exist, and for all but the most intrusive technologies, the result is likely to be a net gain of liberty and privacy.

E. Inequality

Finally, there are concerns about continued inequality if monitoring is used in lieu of commercial bail. These are, in a way, the opposite of the net-widening objection discussed just above: to the extent that unmonitored release on bail remains an option for those who can afford it, the economic discrimination of the current system is maintained. So far, in fact, the advent of GPS tracking, combined with older and more expensive forms of monitoring, has in some cases worsened this discrimination, as rich-and-high-flight-risk defendants have avoided detention by a combination of electronic monitoring and expensive private guards.170 But if it is true, as argued here, that electronic monitoring is a major improvement over imprisonment, then the gap between rich and poor will be narrowed significantly by using it in place of imprisonment for failure to post bond. In the absence of better alternatives, opposing the expanded use of monitoring on equality grounds would seem perverse, but equality concerns might justify a call for universal monitoring of pretrial defendants. Such a proposal pits liberty (for the wealthier) against economic equality in an unusually stark way, and while the resolution of the moral question may be in some doubt, the practical question is not: courts as well as legislatures are unlikely to curtail the rights of moneyed defendants in the name of equality.171

Considered as a whole, the objections to the replacement of pretrial detention for flight risk with electronic monitoring pale in comparison to the arguments in its favor—the tremendous gains in liberty, privacy, fairness, and equality for those released. Because pretrial detainees are subjected to an extremely high and extremely burdensome level of government control, a less repugnant method of control offered by technology is a boon. And because the size of the group subject to this control is governed almost entirely by factors unrelated to the financial and political costs of exercising it, lowering those costs will not cause the government to extend its grasp much further. There is something unsavory about a government electronically monitoring its citizens, but in this case, it is more savory than a government imprisoning them for lack of funds; it is an evil, but a lesser one.

Nonetheless, as explored further below, the political process cannot be trusted to widely embrace electronic monitoring as a replacement for detention.172 For nearly a hundred years, the existence of a powerful commercial bondsmen lobby173—and a lack of interest in the plight of poor defendants—has hampered reform efforts. The judiciary, then, is the most promising locus of reform. Part III will show that there is a solid basis for both a constitutional and statutory right to monitoring for defendants detained for lack of funds, and Part IV will demonstrate that despite a vigorous, ongoing debate over the courts’ role in responding to technological change, they should not hesitate to implement those rights.

III. the right to be monitored

Over the years, numerous solutions have been proposed to the broken bail system,174 and some, including the expanded use of personal recognizance bonds and pretrial services programs, have been implemented on a small scale—often against the opposition of the commercial bail industry—with varying degrees of success. Recently, New York’s Chief Judge proposed bail reform legislation that he hoped will make the bond industry “basically irrelevant.”175 He may be able to achieve this, but history suggests that, if so, it will be through his judicial decisions, not his rhetoric. As has long been recognized, judges can sometimes succeed in protecting the rights of unpopular groups where legislatures do not,176and in the bail context, state and federal constitutions and statutes form a perhaps surprisingly solid foundation for a right to monitoring. As discussed below, the statutory arguments are somewhat more straightforward, but, of course, much more subject to change.

A. An Eighth Amendment Right to Be Monitored

Although pretrial detention for failure to post bond clearly implicates other constitutional concerns,177 the Excessive Bail Clause is the provision that speaks directly to pretrial detention, and the strongest case for a right to monitoring rests on it. This is true even though it has not generated a great deal of judicial or academic interest in recent years. In a previous generation, a vigorous debate focused on whether the Clause created a right to a bail calculation based solely on flight risk.178 Interest waned, however, after the Supreme Court answered this question in the negative in United States v. Salerno,179 upholding detention for dangerousness and indicating that neither the Bail Clause nor Substantive Due Process meaningfully limits the permissible justifications for pretrial detention.180

But Salerno did not completely empty the clause of content: regardless of what ends are permissible, “excessive”-ness clearly implies an inquiry into the relationship between those ends and the means employed to achieve them. As one court in the Southern District of New York put it, although the Excessive Bail Clause contains no “absolute ‘least restrictive conditions’ requirement . . . it must preclude bail conditions that are more onerous than necessary” and “result in deprivation of the defendant’s liberty.”181 This must surely be broadly correct, but it also reflects the lack of a clear standard. Distinct from identifying the types of threats restrictions on pretrial liberty may be used to counter or the methods used to quantify that threat, however, the question of how close a fit the Clause requires between that threat and the response remains uncertain. In other words, the Court has told us what restrictions on pretrial liberty must be measured against, but not how they are to be measured—how closely the means must fit the ends.

How is excessiveness to be measured? Crucially, for a right to monitoring, is it to be calculated solely using the internal, actuarial logic of the money bail system, or should it be measured on the basis of all viable alternatives? The courts have not yet provided a clear answer to these questions.

A richer jurisprudence of excessiveness is needed, and towards that goal the balance of this Section will argue the text, history, and case law of the Excessive Bail Clause suggest that at least an intermediate level of scrutiny should be applied to the fit between the legislature’s goals for its system of pretrial release and the means employed to achieve them. Applying this standard, which requires that the means chosen not be “substantially broader than necessary to achieve [the government’s] interest,”182 a bond requirement resulting in detention is clearly excessive if monitoring could serve the state’s goals equally well (and equally efficiently).

The key to this constitutional argument, as well as the statutory arguments below, is establishing monitoring’s superior effectiveness. In light of the strong liberty, privacy, and equality infringements of jailing pretrial defendants, the standard of proof should be somewhat low, perhaps requiring proof that it is “more likely than not” that jail is a substantially overbroad means of preventing flight. Here, the relevant reference point is the appearance rate of defendants released on financial conditions against that for otherwise similarly situated monitored defendants who would have been detained for lack of bail—if the latter rate (and the cost of monitoring relative to imprisoning183) is as low or lower, a concern over flight risk cannot justify imprisonment. The effectiveness of pretrial detention at producing the defendant at trial cannot be the point of comparison, because bail, not detention, is the explicitly preferred option, and its continued use establishes beyond doubt that Congress does not require 100% success. Nonetheless, producing such a record is a serious hurdle, and without it, defenders of the bail system would be able to argue that the same lack of resources that prevents the poor from obtaining release makes them less likely to fear the repercussions of failing to appear for trial. The best hope for the creation of such a record may lie in one of the four states that have banned commercial bail184—or, perhaps, from entrepreneurial bondsmen. The limited success of systems in Europe, as discussed in Part II, will also provide an important foundation.

1. Current Eighth Amendment Doctrine

There is surprisingly little precedent interpreting the Excessive Bail Clause, and the methodology for determining excessiveness remains somewhat uncertain. Nonetheless, a heightened scrutiny approach is consistent with the case law.

The leading case remains United States v. Salerno, in which the Court considered a challenge to the Bail Reform Act of 1984. The petitioner argued that the Act, which allows defendants to be detained out of concern that they would commit a crime if released pending trial—a concern largely unrelated to flight—was unconstitutional. The Court rejected, 6-3, this argument in an opinion authored by Chief Justice Rehnquist. After reviewing the roots of the Excessive Bail Clause in the English Bill of Rights and the Court’s scanty and somewhat contradictory precedents,185 the Court declined even to decide whether the Clause placed any limits at all on Congress, as opposed to the judiciary:

[W]e need not decide today whether the Excessive Bail Clause speaks at all to Congress’ power to define the classes of criminal arrestees who shall be admitted to bail. For even if we were to conclude that the Eighth Amendment imposes some substantive limitations on the National Legislature’s powers in this area, we would still hold that the Bail Reform Act is valid. . . . The only arguable substantive limitation of the Bail Clause is that the Government’s proposed conditions of release or detention not be “excessive” in light of the perceived evil.186

The Excessive Bail Clause, so interpreted, provides a rather narrow window through which to challenge conditions of pretrial release—or a lack thereof.187 Indeed, as I have argued elsewhere, the Clause has been of so little importance that the question of its incorporation against the states appears to have been decided in a footnote—in a case about an unrelated constitutional provision.188 Nonetheless, assuming that the “only arguable substantive limitation” imposed by the Excessive Bail Clause—the prohibition on excessive means of achieving desired bail purposes—does in fact exist, the continued detention of impecunious defendants despite the existence of cost-effective alternatives to ensuring their presence at trial conflicts with even this rather stunted conception of pretrial liberty. And this limit should exist: applying the Clause only to the judiciary would be highly anomalous.189 The Framers, along with later constitutional theorists,190 understood that legislative majorities can be as much of a threat to individual rights as the executive, and certainly the judicial, branch.191 And indeed, the lower courts have consistently assumed or held that the Salerno limit applies.192 The question, then, is what the contours of that limit are. Although, as discussed below, the Court has left them rather vague, nothing in the case law precludes the requirement of a close fit between means and ends.

In Stack v. Boyle, the Court considered an Eighth Amendment challenge brought by twelve petitioners charged with conspiring to violate the Smith Act, which criminalized, inter alia, advocating the violent overthrow of the United States government or organizing or joining a group that did so.193 Bail had been set for each petitioner at an amount greater than was typical for crimes with similar maximum punishment.

The Court began its analysis by noting that “[b]ail set at a figure higher than an amount reasonably calculated to fulfill [its] purpose is excessive,” and that “the fixing of bail for any individual defendant must be based upon standards relevant to the purpose.”194 It then looked to the “traditional standards as expressed in the Federal Rules of Criminal Procedure,” which included the offense charged, the weight of the evidence, and the defendant’s financial ability and character.195 The government had put forth no evidence, but asked the Court to affirm on the theory that “each petitioner is a pawn in a conspiracy and will, in obedience to a superior, flee the jurisdiction.”196 The Court rejected this argument, holding that “[t]o infer from the fact of indictment alone a need for bail in an unusually high amount is an arbitrary act,” and that in the absence of evidence the bail set violated “statutory and constitutional standards.”197

Stack, then, is more concerned with how flight risk must be proven than with matching flight risk to the government’s response, but there are some indications that the Court was engaging in something more than a rational basis review. “Reasonably” and “arbitrary” can both suggest a range of levels of scrutiny, and arguably foreclose strict scrutiny.198 At the same time, the Court’s reasoning seems inconsistent with a rational basis approach. In a time in which the American Communist Party was viewed as “a highly disciplined organization, adept at infiltration into strategic positions,” dedicated to the “overthrow of the Government by force and violence”and possessing the “slavish[]” obedience of its members,199 it hardly seems irrational for the Court to have thought that those who were charged with Smith Act violations categorically posed a greater risk of flight.200

In Salerno, the excessiveness analysis was more cursory. The Court stated that “to determine whether the Government’s response is excessive, we must compare that response against the interest the Government seeks to protect . . . . [B]ail must be set by a court at a sum designed to ensure that goal, and no more,”201 and added little more before concluding that detention was a permissible means of preventing serious pretrial crime.

The Salerno Court did, however, acknowledge that “the individual’s strong interest in liberty” is of a “fundamental nature,”202 and it would be highly anomalous not to subject a deprivation of such a right to meaningful review under a constitutional provision designed to protect it. And importantly, the Court has not looked to its rather permissive Cruel and Unusual Punishment proportionality jurisprudence to determine excessiveness under the Bail Clause as it has under the Excessive Fines Clause.203 This is wholly appropriate, because bail is not meant as punishment—which legislatures have a great amount of discretion to tailor204—but rather process. And the courts and other parts of the Constitution typically give far less leeway to legislatures in limiting criminal procedural rights.

Thus, although there is nothing definitive either way, requiring, at the minimum, a substantial fit between the ends served by a system of pretrial release and the means employed is consistent with current doctrine.205 Nor is there anything to foreclose looking beyond the internal logic of the money bail system; due to the historical dearth of highly effective, efficient alternatives, the issue has not yet arisen.

2. Text, History, and Purpose

Requiring at least a substantial relationship between pretrial burdens and their justification also finds support in the text of the Eighth Amendment, which mandates that “[e]xcessive bail shall not be required . . . .” From the beginning, this deceptively simple phrase has seemed more noble sentiment than source of concrete rights. As the speaker of the only recorded comment on the Bail Clause during the congressional debate over the adoption of the Bill of Rights put it: “The clause seems to express a great deal of humanity, on which account I have no objection to it; but as it seems to have no meaning in it, I do not think it necessary. What is meant by the terms excessive bail? Who are to be the judges?”206

Nonetheless, we can put some meat on the bare bones of the Clause. Looking to the original public meaning207 of constitutional provisions, as the Court has done with respect to the Excessive Fines Clause,208 “excess” meant, according to Samuel Johnson’s dictionary, “[m]ore than enough; superfluity,” and excessive “[b]eyond the common proportion of quantity or bulk.”209 And when the Fourteenth Amendment, which applied the Eighth Amendment to the states, was adopted, “excessive” meant “[b]eyond the established laws of morality and religion, or beyond the bounds of justice, fitness, propriety, expedience, or utility,” with the Excessive Bail Clause specifically cited as an example of this usage.210Thus, in contrast to Fourth Amendment reasonableness,211 excessiveness clearly embodies a requirement of meaningful proportionality.212 And unless it is to be a constitutional nullity, it must require more than a merely rational relationship between means and ends—that much, of course, is required by the Due Process Clause. Moreover, constitutional proportionality must necessarily take into account changing real-world conditions to avoid obsolescence. We would not, for example, judge the excessiveness of a fine (or bail setting) using eighteenth-century standards; inflation would render that approach absurd. Similarly, whether an amount is “excessive” under the Bail Clause should not be judged in terms of the money bail system of the past, but in light of modern alternatives. And as argued below, this rich conception of excessiveness is essential to fulfilling the purpose of the Clause. As Jamal Greene has noted, the Framers’ intent remains relevant to courts dealing with constitutional questions,213 and purpose more broadly is important to theorists of many stripes.214 The Founders transferred the phrase “[e]xcessive bail shall not be required,” with slight modification, from state charters, which had borrowed the phrase from the English Bill of Rights and other English laws.215 The English Bill of Rights was a direct response to the Crown’s practice of imprisoning political dissidents indefinitely without cause and denying them bail. While the Petition of Right in 1627 and the later Habeas Corpus Act provided some protections, judges were still able to effectively deny bail by setting it at a high monetary amount—a practice that still occurs in American courts today, although not for such egregiously political reasons as those of the 1600s.216 The Excessive Bail Clause of the English Bill of Rights therefore directly addressed the problem, outlined in a House of Commons report, of the “‘illegal and a high breach of the liberty of the subject’ [in] the refusing of ‘sufficient’ bail.”217

The history of the Excessive Bail Clause in England and its transfer to the United States suggests a direct response to the sovereign’s use of political power—and his influence on judges—to wrongly and indefinitely imprison individuals to punish or coerce them.218 The right to monitoring is consistent with this core purpose of the Excessive Bail Clause: keeping people locked up increases the likelihood of a guilty plea,219 weakening the core protection against arbitrary or discriminatory prosecution—the jury trial. Indeed, as Professor Sanford Levinson has observed, the federal-state relationship is not the sole guarantee against a tyrannical national government; the Bill of Rights also intends to enlist citizens in this mission by protecting “popular liberty against state depredation.”220 Incarceration is the most intrusive and thorough means of depriving individuals of liberty, yet we currently jail thousands of citizens not yet proven guilty of any crime. We prevent these individuals from forming a good defense,221 thus raising their chances of being convicted and incarcerated, and this liberty deprivation extends to families and society as a whole. The right to remain free before trial—albeit within a somewhat intrusive monitoring regime—seems to rest comfortably alongside, or even above, other rights often cited as necessary to free citizenship. This, too, strongly suggests that the required fit between the legislature’s ends and the chosen means must not be overly loose: a rational basis analysis, under which almost anything is permissible, would do little or nothing to prevent these abuses.

And while an amendment that specifically contemplates the existence of bail might at first glance seem an unlikely source of a right to an alternative, this is not an especially serious objection. In the same way that even a low bail setting is excessive if the traditional alternative of release on personal recognizance would equally satisfy the purposes of the bail setting, any bail requirement aimed at reducing flight risk and resulting in detention is excessive if monitoring would achieve that goal as efficiently and effectively. More fundamentally, it is clear that the setting of non-excessive bail is not an end in itself, but rather a means toward the goal of greater pretrial liberty.222 Its purpose was not to enshrine the money bail system, but to ensure pretrial liberty (using the best option available at the time) and to prevent tyrannical use of pretrial detention. As technology makes achieving society’s goals with respect to pretrial defendants easier, so should our measure of what is excessive.223 And indeed, the Excessive Bail Clause has not been thought to apply only to financial conditions—it has been used to successfully challenge non-monetary conditions.224

3. Applying Intermediate Scrutiny

At least an intermediate level of scrutiny is consistent with precedent and warranted by the text and purpose of the Eighth Amendment. Applying this standard, the question is whether the use of money bail to ensure the defendant’s presence at trial is excessive under Salerno—that is, substantially broader than necessary to achieve the governmental interests at stake—in the face of a record, like that discussed above, proving the existence of an equally effective and inexpensive monitoring alternative. Given the large percentage of defendants detained for failure to post relatively modest bonds in many jurisdictions,225 the answer appears to be very likely yes—rendering those jurisdictions susceptible to Eighth Amendment (and/or state constitutional) challenges to bail settings from detained, non-dangerous defendants not posing an extreme flight risk.226 Courts would, in all likelihood, give these jurisdictions reasonable time to deploy a monitoring regime, but eventually they would be forced to release these defendants, monitored or not.227

At this point, however, an acknowledgement of the limits of the proposal’s ability to eliminate wealth inequality in pretrial outcomes or eliminate net-widening concerns is appropriate. To be clear, under the analysis set out above a jurisdiction could very reasonably conclude, even once the efficiency of monitoring is clearly established, that it prefers to maintain a money bail option because it is less intrusive for those who can afford it. And it is very likely true that for some of those who can afford it, even a large amount of money would not seem excessive if the alternative is some form of electronic monitoring. Nothing in the foregoing argument would prevent them from doing so. A right to the alternative would simply be an option to select monitoring in lieu of pretrial release. But it would forbid a jurisdiction from jailing a defendant because he could not pay this “luxury tax.” This outcome might be unsatisfying to those concerned with economic discrimination, as it would perpetuate, to a lesser degree, the disparity of pretrial treatment based on financial resources. And beyond principle, ensuring that white-collar defendants were subject to the same conditions as the less affluent would help limit the invasiveness of those conditions. But although these concerns bear careful consideration, similar arguments have received short shrift under the Equal Protection Clause, and it is difficult to find a toehold for them under the Eighth Amendment.228

However, for those concerned that even monitoring is not a carefully delineated approach to achieving pretrial presence and that defendants who would have simply been released may now be swept into the monitoring net, Salerno and Stack provide the doctrinal foundation for limiting the extent of monitoring, at least to a degreethe amount and nature of monitoring would have to be justified with reference to their effect on flight risk. In light of the intrusive nature of many current conditions of pretrial release, this might not be a particularly strong protection, but it would likely prevent extreme forms of monitoring, such as twenty-four-hour surveillance, that the government might seek in order to gather evidence or prevent crime.

B. Statutory Approaches

For federal detainees and detainees in states with similar bail statutes,229 another path to monitoring in lieu of detention for flight risk lies in the Bail Reform Act’s requirement that a judge impose “the least restrictive further condition, or combination of conditions, that such judicial officer determines will reasonably assure the appearance of the person as required and the safety of any other person and the community.”230 Electronic monitoring is clearly less restrictive than a monetary requirement resulting in detention; therefore, once a sufficient record of a technological alternative’s efficiency and effectiveness exists, it should be required by the Bail Reform Act and similar statutes. At least one court has taken a preliminary step in this direction. In Karpouzis v. Government of the Virgin Islands,231 the Appellate Division of the District Court of the Virgin Islands applied the least restrictive means requirement of a territory law to reverse a $2 million bail requirement when the “trial court gave no explanation why house arrest with electronic monitoring, a responsible third-party custodian, and a very significantly reduced monetary bail in the range proffered by appellant would not adequately ensure appellant’s presence at trial.”232

Karpouzis was limited, of course, to the individual circumstances of the case. For meaningful reform, a broader approach is necessary. If monitoring’s superiority in reducing flight can be established, then it should always be required under these statutes as an alternative to an unmeetable bail setting. Further, since judges (and defendants themselves, who very often lack counsel at the bail stage233) often cannot know in advance exactly what a defendant can afford,234 and since a meetable bond will often be more restrictive than electronic monitoring, courts should be required to offer monitoring as an alternative when setting bail. In this way, a portion of the least restrictive means analysis is given over to the party with the greatest abilityand incentiveto get it right.

The Bail Reform Act applies only in federal court, leaving only constitutional arguments for detainees in states without analogous statutory language. Indeed, even in jurisdictions with similar legislation, judges may be more comfortable imposing the adoption or expansion of monitoring on constitutional grounds. Constitutional grounds, moreover, have the added advantage of being further removed from the reach of legislaturesand special interests.235

IV. the case for judicial intervention

Having made the normative, constitutional, and statutory case for a right to monitoring, it is necessary to consider one last set of objections. In light of an ongoing, important debate over whether courts or legislatures are the proper bodies to define the limits on governments’ use of technology,236 it is worth explaining why judicial intervention is necessary. Broadly speaking, on one side, the unpopularity of criminal defendants and the existence of a potent law enforcement lobby may lead legislatures to systematically undervalue defendants’ rights, thus necessitating increased judicial scrutiny of encroachments on liberty and privacy.237 On the other hand, because society as a whole benefits from, and is burdened by, increased police power, democratically elected legislatures may accurately reflect the preferred balance.238

But regardless of the resolution of the debate over institutions best suited to regulate the government’s use of technology that affects society as a whole, it is clear as a matter of theory and history that legislatures cannot be relied upon to appropriately weigh the interests of the unpopular and politically weak, particularly when those interests conflict with those of a concentrated lobby.239 If it is true that GPS, or future forms of monitoring technology, will soon, if it does not already, offer a fairer, more effective, and more efficient alternative to money bail for producing defendants for trial, it would be natural to assume that, over time, the political processindeed, mere prudencemight lead to its widespread adoption. And in some jurisdictions it probably will be adopted. But the adoption of beneficial technology depends on the priorities of those in chargefootball teams get iPads before philosophy departments. And modern political theory, amply illustrated by a long history of stunted bail reforms, suggests that in many jurisdictions, pretrial monitoring will not be adopted without court involvement. Because of this, the courts, despite some limitations, should be a driver of reform.

A. Public and Private Interests in Pretrial Release

Commercial bail bonding is among the oldest forms of privatization in criminal justice. Because it has always relied on the government to create demand for its services by imposing financial conditions for pretrial release,240 the bond industry has long invested in government relations. Factors crucial to the industry’s profitability—the number and amount of bonds imposed—are determined by government policy. And it is a multi-billion dollar industry.241 This creates a classic public choice problem: the benefits of the bail system are concentrated in the industry, while the harms are borne by a politically weak group, criminal defendants. Indeed, bondsmen have opposed many efforts to change the traditional money bail system—including efforts to implement pretrial monitoring.242 They view pretrial release programs as direct threats to their business,243 and despite using monitoring technologies themselves to track defendants,244 they have argued that the technologies are ineffective and costly to defendants. In response to concerns about the affordability of bonds, one consultant writing for a bondsman blog suggests that family members or friends do not post bonds for an accused and leave him in jail because “the accused is unable to abide by any semblance of rules and regulations.”245 “More than anything,” he argues, “this is an indictment of our society’s deteriorating parental skills which seems to have trouble teaching responsibility, accountability, and discipline. Instead, a family would rather have their ‘deadbeat’ left in jail where they might learn a lesson or two as opposed to being released on bail.”246

The greatest harm from the system, moreover, is suffered by the least influential of criminal defendants, those too poor to make bail, while the relatively few defendants of significant means suffer little inconvenience. Compounding the problem, the costs of detaining defendants who cannot post bond are widely dispersed to taxpayers, while bondsmen and their political allies can claim that the private surety system is otherwise run without cost to the public fisc. And the employees of the jails housing pretrial detainees have their own lobbies to protect their continued employment.

All of this goes a long way towards explaining why money bail, a practice developed at a time when there were no professional police and few jails,247 the earliest possible trial might be months away,248 and the odds of an absconding defendant being recaptured were long, has survived in modern times despite vigorous reform efforts. Despite compelling arguments for the expanded use of personal recognizance and pretrial services programs, in many jurisdictions, detention for failure to post bond remains widespread. And indeed, despite the enormous budget pressures facing state and local governments in the wake of the 2008 financial crisis and subsequent economic downturn,249 moves to release a greater number of pretrial defendants have been only sporadic. This pattern is visible in the history of monitoring technology as well. Although providers of monitoring technology and services have their own lobbies to pursue their interests, the result has largely been the monitoring of those who would previously have been released anyway. In at least one documented case, the bond industry has successfully lobbied to kill a large-scale monitoring program.250 Thus, while monitoring technology will probably replace detention in some jurisdictions in the coming years, likely those with weak or absent bail lobbies or especially pressing budget problems, there is reason to believe that in many places the result of the political process will be the continued existence of money bail—and detention for flight risk.

B. An Imperfect Judicial Solution

Under these circumstances, it is clear that regardless of the resolution of the larger debate over the relative merits of courts and legislatures in responding to new technologies, the best hope for an actual monitoring right for flight-risk detainees lies in judicial intervention. In light of the near certainty of legislative inaction, the limitations of judicial responses to new technology lose their usual force; half a loaf is better than no loaf at all. The judiciary, particularly the elected judiciary, is of course not immune from political considerations, but its relative independence makes it a far more promising locus of reform. And although generalist judges typically have no particular scientific expertise, courts have long experience evaluating the efficacy and affordability of new technology. The clearest example of this may be in the torts context, which relies heavily upon proof of available technologies in ascertaining standards of care to which various actors are held.251 In the famous T.J. Hooper case, Judge Learned Hand declared that tugboat operators who could have but failed to use new weather information technologies negligently damaged barges and cargo.252 Despite the fact that the radio receiving technologies that would have warned of storms had not yet been “generally adopted” in the business—indeed, there was “no custom at all” to use the sets, at least according to Hand253—the court observed that in some cases, “a whole calling may have unduly lagged in the adoption of new and available devices,” thus inducing court action.254 Indeed, subsequent cases have confirmed that where affordable, state-of-the-art technology would make products safe but has not been widely adopted, products manufacturers can be liable for failure to install these products.255 And, of course, federal courts are experienced in assessing the effectiveness of monitoring provided by Federal Pretrial Services and private operators in the context of individual bail determinations.256

Nonetheless, courts are likely to be much more hesitant to base significant new constitutional or statutory law, as opposed to new tort liability (or individual release determinations within an established framework), on emerging technology. As Professor Kerr has observed in his influential work, judges are largely reliant on parties to bring them information about new technologies, and, due to stare decisis, the rules they announce cannot easily be changed.257 This, as the Supreme Court has acknowledged, makes them cautious when evaluating the effect of new technology on constitutional law.258 Moreover, though judges’ intimate knowledge of the bail system (with monitoring as its use becomes increasingly common) may help make them more comfortable with intervening, that same comfort level, and indeed their active role in setting bail, may cut against intervention. Adding to the case for judicial caution are the deep marks left by the history of the judiciary’s involvement in prison reform259 and the lasting backlash against it: in recent years, the courts have been extremely reluctant to require costly remedies even for the most severe constitutional violations.260 In light of that experience, judges will be extremely hesitant to be seen as requiring taxpayers to foot the bill for improvements in criminal defendants’ quality of life or as micromanaging detention policies.

Due to these factors, courts are likely to require extremely convincing proof of the effectiveness and efficiency of monitoring technology before mandating that it be offered as an alternative to detention. Even the relatively modest performance advantage of money bail over personal recognizance bonds and its arguable cost savings (to the state) over existing pretrial services programs currently insulate it from judicial attack. Despite some promising results from pretrial monitoring and growing experience with its use in other contexts, courts will be hesitant to consistently grant monitoring in lieu of detention to indigent, non-dangerous defendants until at least a few jurisdictions adopt monitoring more widely in lieu of detention and have established a well-documented record of failure-to-appear rates at least equal to those of traditional money bail systems for similarly situated defendants at equal or lesser cost.261 A large-scale study, funded by the National Institute of Justice, of the effectiveness of electronic monitoring at reducing recidivism among released felons in Florida showed that it reduced failure rates by over thirty percent compared with other forms of supervision.262 A similar study with respect to flight risk and money bail would be a critical data point. The four states that have abolished money bail are perhaps the most likely candidates to provide the necessary data, and they would be a logical starting place for reform efforts.

Further, any remedies that are eventually imposed are likely to be imposed gradually to allow jurisdictions to spread out fixed costs and gain familiarity with the technology. In the pretrial context, the judiciary will likely not be at the vanguard of reform. But delayed, gradual reform is better than its probable alternative, no reform. Even the relatively modest intervention contemplated here would bring significant benefits. In the absence of judicial action, thousands of criminal defendants will continue to be detained—and suffer all of the deleterious effects of detention—long after available technology would allow the government to achieve its goals at lower financial and human cost. No matter how sound the policy, of course, judicial intervention requires a judicial basis for decision, which this Essay has located primarily in the Eighth Amendment.


At any given time, thousands of criminal defendants around the country are imprisoned to ensure their presence at trial despite being eligible for release, simply because they lack the financial resources to make bail. Monitoring technology, which will only increase in effectiveness and decline in cost, is already in wide use, including as a condition of bail,263 and represents a far less burdensome means of achieving the government’s aims. Moreover, especially as compared to the potential gains from releasing those currently held for trial, net-widening concerns are fairly minor, as the means of ensuring trial presence appear to have relatively little effect on the size of the overall criminal defendant population, and within that population there are relatively few currently released without various types of conditions.264

In many cases governments will adopt less-intrusive, cost-effective technologies voluntarily, without judicial prompting. This has been the case with police use of advancing investigative techniques generally. But well-established theory and long and bitter experience strongly suggest that when adopting these technologies would aid the poor and unpopular at the expense of established lobbies, they will not be adopted through the political process. If they are not, courts must intervene. The Eighth Amendment prohibits the imposition of excessive bail, and if it is to fulfill its core purpose of preventing needless or discriminatory pretrial detention, it must require the use of technologies that prevent flight as efficiently and effectively as the bail system without its concomitant detention. Building the necessary evidentiary record will be a significant hurdle for advocates, and courts will have to overcome their reluctance to involve themselves in the mechanics of both pretrial justice and detention, but these challenges must be met. If they are not, the iniquitous result will be a twenty-first-century system of detection coupled with a nineteenth-century system of detention.