Hardening the System: Three Commonsense Measures to Help Keep Crime at Bay

After a long period of continuous violent-crime declines throughout the U.S.—spanning from the mid-1990s through the early 2010s—many American cities are now seeing significant increases in violence. Nationally, in 2015 and 2016, murders rose nearly 11% and 8%, respectively.[1] The national homicide rate declined slightly in 2017 and 2018, before ticking upward in 2019.[2] In 2020, the nation saw its largest single-year spike in homicides in at least 100 years—which was followed by another increase in murders in 2021, according to CDC data and FBI estimates.[3] In the last few years, a number of cities have seen murders hit an all-time high.[4] In addition to homicides, the risk of other types of violent victimizations rose significantly, as well.[5] While various analyses estimated a slight decline in homicides for the country in 2022,[6] many American cities still find themselves dealing with levels of violence far higher than they were a decade ago.

While violent crime—particularly murder—is the most serious due in large part to its social costs,[7] there have also been worrying increases in crimes such as retail theft,[8] carjacking,[9] and auto theft,[10] as well as in other visible signs of disorder in public spaces (from open-air drug use and public urination to illegal street racing and large-scale looting and riots).[11]

Although several contributing factors are likely, this general deterioration in public safety and order was unquestionably preceded and accompanied by a virtually unidirectional shift toward leniency and away from accountability in the policing, prosecutorial, and criminal-justice policy spaces. That shift is evidenced by, among other things, three major trends in enforcement:

• A 25% decline in the number of those imprisoned during 2011–22[12] 
• A 15% decline in the number of those held in jail during 2010–21[13] 
• A 26% decline in the number of arrests effected by law-enforcement officers during 2009–19[14] 

Notable contributing factors to the decline in enforcement include:

• A sharp uptick in public scrutiny and interventions—in the form of investigations and legal action taken by state attorneys general and the federal Department of Justice—against local law-enforcement agencies[15]
• The worsening of an ongoing police recruitment and retention crisis, particularly in large urban departments[16] 
• The electoral success of the so-called progressive prosecutor movement, which, by 2022, had won seats in 75 jurisdictions, representing more than 72 million U.S. residents[17] 
• Perhaps most important, the adoption of a slew of criminal-justice and policing reform measures at all levels of government[18] 

Those who are skeptical of the criminal-justice reform movement have devoted most of their efforts to arguing against the movement’s excesses and explaining why it would be unwise to enact certain measures.[19] Less effort has been devoted to the extremely important task of articulating a positive agenda for regaining what has been lost on the safety and order front.[20] This paper seeks to add to that positive agenda for safety by proposing three model policies that, if adopted, would help, directly and indirectly, stem the tide of rising crime and violence, primarily by maximizing the benefits that attend the incapacitation of serious criminals (especially repeat offenders) and by encouraging the collection and public reporting of data that can inform the public about the downside risks that are glossed over by decarceration and depolicing activists.

The three policies proposed here, which draw on policies proposed and adopted throughout the country in recent history:

• Modified “Three Strikes”—Creating a points system for various offenses as well as a points threshold that will trigger a mandatory minimum sentencing enhancement, in order to improve deterrence for those beneath the threshold and to maximize incapacitation for those who step over it
• “Truth in Sentencing”—Setting a floor for how much of a given sentence must be served before a convicted felon becomes eligible for initial release into community supervision, in order to maximize incapacitation for those who have been convicted of a serious offense and sentenced to a term of imprisonment
• “Data Transparency”—Identifies several types of crime-related data that jurisdictions will be encouraged to collect and report in a standardized manner to address the problem of making and evaluating policing and criminal-justice policies without the benefit of reliable, relevant data

These model policies should be viewed flexibly; policymakers should see them as starting points, feeling free to make changes that reflect the various concerns and idiosyncrasies specific to their respective jurisdictions.

The U.S. has a homicide problem. From an international perspective, the U.S. is, and has long been, an outlier on deadly violence, compared with western European democracies and our neighbors to the north. That problem has gotten significantly worse over the last decade.[21] But the problem does not affect all parts of American society equally. Homicide—arguably the most heinous (and socially costly)[22] of offenses—is extremely concentrated, geographically and demographically.[23] In other words, homicide victims in the U.S. tend to be relatively young black and Latino males living in small pockets of urban America with very elevated crime levels.

But data from various jurisdictions tell us that homicide (and other violent-crime) perpetrators constitute an even narrower slice of the population. This is true both in the U.S. and abroad. For example, a study of convicted offenders in Sweden found that, during 1973–2004, approximately 1% of the country’s population was responsible for nearly two-thirds of its crime.[24] An analysis of youth violence in Boston during the mid-1990s found that just 61 gangs comprising 1,100–1,300 members constituted “less than 1% of the city’s youth between the ages of 14 and 24” but “were responsible for more than 60% of youth homicide” in that jurisdiction.[25] That study also revealed that most of those in this narrow slice of the population had an extensive prior criminal history, with 96 of the 125 known murderers studied having been charged with nearly 10 offenses each, prior to the homicide recorded.[26]

A more recent problem analysis of gun violence in Oakland, California, found that just 0.1% of the city’s population was responsible for the majority of its homicides.[27] On average, homicide suspects in that city had 10 prior arrests, with 84% of them having been incarcerated in the past.[28] And in Chicago, an analysis of gun violence in 2015 and 2016 found that shooting and murder suspects had nearly 12 prior arrests, on average.[29]

These statistics show that the most urgent of America’s crime problems is driven by small groups of prolific offenders, many of whom had active criminal-justice statuses (an open case, probation, or parole) at the time they took a life. What is to be done in light of this fact? One answer is to place meaningful limits on repeat offending by cracking down on those unwilling to heed the advice that Fletcher Reede (the main character in the 1997 film Liar Liar, starring Jim Carrey) gave his client Skull, after he was arrested for knocking over “another ATM.”[30] What follows is a proposal to do just that in the form of a modified “three-strikes” policy.

In essence, a three-strikes policy works by mandating a minimum sentencing enhancement for convicted felons with a certain number of prior convictions. Such policies are not new. The first such law was, according to the Prison Policy Initiative, adopted in 1993 in Washington State.[31] Within just a few years, 23 states and the federal government had adopted versions of the policy.[32] 

The main benefits of a three-strikes rule are twofold. First, by elongating the sentences of certain repeat offenders, the policy adds to the incapacitation benefits associated with their incarceration. The public-safety benefits of incapacitation are significant. Consider the main finding of a study of incapacitation done by UC-Berkeley researchers Rucker Johnson and Steven Raphael: For the period 1991–2004, “each additional prison-year served prevented approximately” eight index crimes.[33] That estimate is likely deflated by the fact that it is based on crime reports, because most crimes are not actually reported.[34] Older estimates of the direct incapacitation effects of incarceration on crime suggest that it might be even higher.[35] Estimates from other countries illustrate the effect, as well.[36]

The benefits of incapacitation are most pronounced for the most prolific offenders. A documented and extensive history of recent convictions, while perhaps an imperfect measure, provides a good indication that a given offender has a high chance of re-offending, which can be mitigated through incapacitation.

Second, a three-strikes regime enhances the potential deterrent effect of a penalty by substantially raising the costs of committing a future offense. A three-strikes policy is a more effective deterrent than, say, a marginal increase in sentencing severity, because it creates an expectation that a meaningful sentencing enhancement will be triggered by a future offense. Perhaps the best illustration of this comes from a study of the effect of California’s three-strikes policy on future offending, published in 2007 by economists Eric Helland and Alexander Tabarrok, who found that the law (which imposed substantial sentencing enhancements for third felony convictions) reduced the felony arrest rate for offenders with two strikes by 17%–20%.[37] Similarly, another study of California’s law, published in 2017 in the Atlantic Economic Journal, found that “the law had a significant deterrent effect on all crimes.”[38]

However, three-strikes regimes have not gone without criticism.[39] Some have argued, for example, that such a policy might incent criminals to use more violence to avoid the expected sanction, by using deadly force to resist an arrest or by threatening/killing witnesses. However, this criticism can be leveled against any policy that provides for lengthy sentences; and there does not seem to be any evidence suggesting that violent resistance or witness intimidation has become less likely in places that have lowered criminal penalties through various criminal-justice reform measures.

Perhaps the most notable of the critiques of strike regimes centers on examples of offenders whose third strike came after a long period of desistance or was the result of a relatively “minor” offense. The following proposal seeks to capture the incapacitation and deterrence benefits associated with earlier policies, while being responsive to these criticisms. Specifically, the model policy laid out herein features a built-in credit system that rewards periods of desistance by eliminating strikes gradually over time. It also features varied sentencing enhancements based on the nature of the third strike, such that an enhancement triggered by a nonviolent felony is substantially less than that triggered by a violent or firearm-related felony. The model includes modified strike counts with respect to convictions resulting from offenses committed by juveniles.

An Act to Place Meaningful Limits on Repeat Offending

Section I.—Sentencing Enhancements

1. Generally. In determining the appropriate sentence for a defendant, the sentencing judge shall, effective as of the date of enactment of this statute, determine whether the defendant is required to be sentenced as a three-strikes offender, as defined in Section II. 1) a), and sentence said defendant in accordance with Subsections 2)–6) of this section.
2. Strike Counts. In determining whether the defendant is required to be sentenced as a three-strikes offender, as defined in Section II. 1) a), the court shall determine the number of strikes accrued by the defendant according to the following scheme:
   1. Any conviction for a strike-eligible misdemeanor, set forth in Section II. 1) b) i), shall count as one-quarter (¼) strike.
   2. Any conviction for a strike-eligible nonviolent felony, set forth in Section II. 1) b) ii), shall count as one-half (½) strike.
   3. Any conviction for a strike-eligible firearm-related felony, set forth in Section II. 1) b) iii), shall count as three-quarter (¾) strike.
   4. Any conviction for a strike-eligible violent felony, set forth in Section II. 1) b) iv), shall count as one (1) strike.
   5. A defendant shall not accrue any strikes for any misdemeanor convictions for offenses committed while the defendant was a juvenile.
   6. Any convictions for strike-eligible nonviolent or firearm-related felony offenses in cases in which the defendant was a juvenile shall result in the accrual of only one-quarter (¼) strike.
   7. Any convictions for strike-eligible violent felony offenses in cases in which the defendant was a juvenile shall result in the accrual of only one-half (½) strike.
3. Desistance Credits. In determining whether a defendant is a three-strikes offender, as defined in Section II. 1) a), the sentencing judge shall reduce the number of strikes that the defendant possesses:
   1. by one-half (½) strike, if the defendant has not been convicted of a strike-eligible criminal offense for a consecutive period of at least three (3) years;
   2. by one (1) strike, if the defendant has not been convicted of a strike-eligible criminal offense for a consecutive period of at least five (5) years; or
   3. by two (2) strikes, if the defendant has not been convicted of a strike-eligible criminal offense for a consecutive period of at least 10 years.
   4. No desistance credit shall be given:
      1. to any defendant based on time the defendant was confined within a federal, state, county, or city correctional facility.
4. Enhanced Sentences. A three-strikes offender, as defined in Section II. 1) a), shall:
   1. have a consecutive sentence of 10 years imprisonment added to the sentence for the offenses of which the defendant was convicted if the most serious underlying offense in the instant case is a nonviolent felony set forth in Section II. 1) b) ii).
   2. have a consecutive sentence of 15 years imprisonment added to the sentence for the offenses of which the defendant was convicted if the most serious underlying strike-eligible offense in the instant case is a firearm-related felony set forth in Section II. 1) b) iii).
   3. have a consecutive sentence of life imprisonment added to the sentence for the offenses of which the defendant was convicted if:
      1. the most serious underlying offense in the instant case is a violent felony set forth in Section II. 1) b) iv);
      2. the defendant has been convicted of strike-eligible offenses on at least two prior occasions; and
      3. at least three of the defendant’s strikes stem from convictions for strike-eligible felony offenses.
5. Prohibition on Misdemeanor-Triggered Enhancements. A defendant who reaches or surpasses the three-strikes threshold due to an instant conviction of a strike-eligible misdemeanor shall not be eligible for a sentencing enhancement under Section I until said defendant is subsequently convicted of a felony, such that no sentencing enhancement shall be applied to sentences for misdemeanor convictions.

Section II.—Definitions

1. Terms Defined. When applying this Act, the following terms shall have the following meanings, which shall not be overridden by the definitions of similar terms set forth in other sections of the laws of [insert state]:
   1. Three-Strikes Offender. A defendant shall be deemed a three-strikes offender if:
      1. the instant strike-eligible offense of which the defendant is convicted constitutes a felony set out in Subsection b) ii)–iv; and
      2. the defendant has, inclusive of the conviction for the instant offense and accounting for desistance credits, accrued three (3) or more strikes through prior convictions.
      3. A defendant may not be deemed a three-strikes offender if all the defendant’s strikes are the result of convictions in one or more cases stemming from a single episode of criminal action. If all the defendant’s strikes are the result of convictions stemming from two episodes of criminal action, each episode must include at least one (1) strike-eligible felony conviction for the defendant to be deemed a three-strikes offender.
   2. Strike-Eligible Offenses. The following offenses are strike-eligible in accordance with the scheme set forth in Section I 2) a)–d):
      1. Strike-Eligible Misdemeanors. The following misdemeanor offenses shall be deemed strike-eligible:
         (1) [Insert offenses here]
      2. Strike-Eligible Nonviolent Felonies. The following nonviolent felony offenses shall be deemed strike-eligible:
         (1) [Insert offenses here]
      3. Strike-Eligible Firearm-Related Felonies. The following firearm-related felony offenses shall be deemed strike-eligible:
         (1) [Insert offenses here]
      4. Strike-Eligible Violent Felonies. The following violent felony offenses shall be deemed strike-eligible:
         (1) [Insert offenses here]
   3. Single Episode of Criminal Action. An event or continuous series of events related to the furtherance of a single criminal act, excluding the defendant’s violent resistance of any law-enforcement officer or citizens acting lawfully in defense of themselves or another that proximately causes death or serious bodily harm to the officer, citizen, or any other person.

Section III.—Severability

1. A decision of any [insert state] or federal court ruling that any provision, section, subsection, clause, or application of this Act is unlawful under the constitutions or laws of the state of [insert state] or of the United States shall not affect the remaining provisions, sections, subsections, causes, or applications of this Act, which can continue to be enforced without the component(s) of this Act deemed unlawful.

The modified three-strikes policy proposed above is meant to address the problem of serious, repeat offenders not receiving sufficiently long sentences. But that alone is insufficient because too often, even when criminals are sentenced to prison, they are not held for very long before being released back into communities. This is true despite evidence showing that, in approximately eight out of 10 cases, those released will go on to re-offend.

When we look at the data available on who goes to prison in the U.S.—particularly in the state system, which accounts for approximately 87% of those imprisoned[40]—what we see is that the vast majority of those who enter and leave prison have extensive criminal histories. For example, a Bureau of Justice Statistics (BJS) report of a cohort of offenders entering state prisons in 2009 and 2014 found that the median prisoner had nine prior arrests and four prior convictions.[41] More than half those prisoners had criminal histories of at least 10 years.[42] BJS reports of recidivism among cohorts of prisoners leaving prison in 2005, 2008, and 2012 showed that the median prisoner released in those years had eight or nine prior arrests and three to five prior convictions.[43] The mean values for these measures were even higher, with prior arrest averages of 10.6–12.1 and prior conviction averages of 4.9–5.8.[44]

Some have suggested that these data are the result of the “criminogenic” effect of incarceration—that is, the degree to which incarceration drives post-release criminality. But the empirical support for this claim is limited, largely because it is rooted in analyses of offenders who pose different risks to the public’s safety from those of the median prisoner.[45]

The recidivism reports recently published by BJS establish a troubling reality: the overwhelming majority of those released from state prisons will be rearrested, and the typical releasee will be rearrested five times over 10 years.[46] Despite this grim reality—this sure-to-be-realized risk—a recent BJS report shows that state prisoners do not actually serve much time before they are paroled. Among state prisoners released in 2018, the median releasee served just 1.3 years (16 months) prior to his initial release.[47] On average, that year’s releasees served 2.7 years.[48] That same report showed that more than 40% of releasees served less than a year in state prison, while nearly 20% served less than six months; on average, state prisoners served less than half (44%) of their maximum sentence prior to being released on parole.[49] The 2018 figures are not, unfortunately, anomalous. BJS’s report on recidivism for 2016 produced nearly identical findings across these measures.[50]

The risks associated with the early release of prisoners are illustrated by data on the representation of parolees among those charged with, suspected in, or convicted of, serious violent crimes. An earlier-referenced problem analysis of gun violence in Oakland showed that 31.3% of homicide suspects were on probation or parole at the time of the homicide in question.[51] Reporting out of Baltimore in 2012 and 2019 shows that 25%–40% of homicide suspects were on probation or parole at the time of the offense in question.[52] A BJS report that analyzed a sample of approximately 9,000 offenders convicted of a violent felony during 1990–2002 in one of the 75 largest urban counties in the U.S. found that 7% of the individuals convicted were on parole at the time of the crime, while 18% were on probation.[53] Moreover, a Manhattan Institute report on New York State’s recent parole reform found that those with an active supervision status (probation or parole) were significantly more likely to be rearrested after post-arraignment release than their counterparts who were not under state supervision for a previous offense.[54]

In addition to the three-strikes policy proposed above, lawmakers should consider limiting the discretion of parole boards by setting a floor on the portion of a sentence that an inmate must serve before he is eligible for release. This measure would minimize the risk that repeat offenders pose by maximizing the incapacitation benefits associated with their convictions and sentences.

Once again, this is not a new idea. Truth-in-sentencing (TIS) laws date back to 1984, when Washington State enacted its own. The now-infamous but then-popular 1994 Crime Bill included grant provisions meant to encourage other states to adopt their own; and, in less than five years, 27 states and Washington, D.C., had laws mandating that offenders serve a minimum share of the sentences handed down to them.[55] These laws likely contributed to the buildup of the country’s prison population—which, in turn, contributed significantly to the decline in crime over that decade.[56] While few analyses have specifically disaggregated the effects on crime of TIS versus those of incarceration generally, one study published in the Journal of Law and Economics attempted to do so, finding that TIS had significant crime-reduction effects at the county level.[57]

By lengthening the period of time that offenders must serve in prison, TIS extends the protection that communities enjoy when likely-to-recidivate inmates are incapacitated.[58] The following model provides a framework through which interested jurisdictions could do just that.

An Act to Mandate the Service of Criminal Sentences

Section I.—Mandated Sentence Service

1. Tier-One Offender Sentence Service. No offender sentenced to a term of imprisonment for a Tier-One Offense, as defined in Section II, herein, committed on or after [the effective date of this Act], shall be eligible for release from prison, whether on parole, probation, house arrest, or a halfway house, prior to serving 85% of the sentence received as a result of his conviction within the confines of a [insert state] correctional facility, notwithstanding any sentence credits that the offender has earned for good behavior, program completion, educational attainment, or for any other reason.
2. Tier-Two Offender Sentence Service. No offender sentenced to a term of imprisonment for a Tier-Two Offense, as defined in Section II, herein, committed on or after [the effective date of this Act], shall be eligible for release from prison, whether on parole, probation, house arrest, or a halfway house, prior to serving 75% of the sentence received as a result of his conviction within the confines of a [insert state] correctional facility, notwithstanding any sentence credits that the offender has earned for good behavior, program completion, education attainment, or for any other reason.
3. Exceptions. An inmate serving a sentence of imprisonment for a Tier-One or Tier-Two Offense, as defined in Section II, may, upon prosecutorial motion and subsequent judicial approval, earn credit toward his parole eligibility, or initial release as a reward for cooperation with, or assistance in, a criminal investigation leading to the felony conviction of another in the state of [insert state].
   1. Any credits earned under this subsection shall not exceed 10% of the inmate’s maximum sentence.
   2. This exception shall not make eligible for release, parole eligibility, or a stay of execution any offender sentenced to life imprisonment or death.
4. Early Release Credits for Other Offenders. Notwithstanding the provisions of this Act to the contrary, offenders convicted and sentenced to a term of incarceration for an offense other than a Tier-One or Tier-Two Offense shall retain their ability to earn and redeem credits toward their early release or parole eligibility.
5. Tier-One and Tier-Two Offender Credit System. Offenders convicted of, and sentenced to a term of imprisonment for, a Tier-One or Tier-Two Offense shall earn good time credits for good behavior, program completion, and educational attainment redeemable for privileges and accommodations, including, but not limited to, reduced security classification, preference for limited work and educational program slots, and time out of cell, in accordance with rules and regulations that the [insert state] Department of Corrections is hereby authorized to promulgate pursuant to this Act.
   1. The [insert state] Department of Corrections shall, within 120 days of [insert the effective date of this Act], disseminate to the wardens of each [insert state] prison facility a complete set of guidelines setting out the manner in which good time credits can be accrued, lost, and redeemed.
   2. The wardens of each [insert state] prison facility shall, within 60 days of the receipt of the guidelines disseminated pursuant to Subsection 4) a), begin keeping a running log of the total credits earned, lost, and redeemed by [insert state] prison inmates serving time for a Tier-One or Tier-Two Offense, setting out the date and manner in which credits were earned, lost, or redeemed.
      1. The balance of a Tier-One or Tier-Two Offender’s credits shall be reported to said offender, either verbally, or in writing, every 60 days and, when practicable, upon request, submitted in writing by a Tier-One or Tier-Two Offender.

Section II.—Definitions

1. Terms Defined. When applying this Act, the following terms shall have the following meanings, which shall not be overridden by the definitions of similar terms set forth in other sections of the laws of [insert state]:
   1. Tier-One Offender. Any person convicted of, and sentenced to a term of imprisonment for, any one of the following offenses committed on or after [insert effective date of this Act]:
      1. [List Tier-One Offenses (e.g., arson, murder, manslaughter, aggravated assault, robbery, rape/sexual assault) here]
   2. Tier-Two Offender. Any person convicted of, and sentenced to a term of imprisonment for, any one of the following offenses committed on or after [insert effective date of this Act]:
      1. [List Tier-Two Offenses (e.g., burglary, grand larceny, grand larceny auto, unlawful firearm possession) here]

The strength of the case for a tougher approach to policing and criminal-justice policy ultimately rests on reliable empirical evidence. For example, the policies proposed here thus far are justified by the data showing that serious crime is driven disproportionately by a small network of high-rate offenders who have been given several bites at the proverbial apple—i.e., offenders who could, under a tougher regime, have been incarcerated, but were not. Unfortunately, many outstanding questions about the U.S. criminal-justice system still cannot be answered definitively because many jurisdictions in the U.S. do not regularly and systematically publish important criminal justice–related data. For example, what shares of robberies, homicides, sexual assaults, and thefts are committed by parolees, pretrial releasees, or probationers? On average, how many prior arrests or convictions do the typical homicide offenders have? How have those numbers changed over time, if at all? How are cases of serious crime resolved? How often do criminal convictions result in serious penalties?

These are all questions that many jurisdictions across the country leave unanswered for most periods.

Not only would publishing such data better inform the media and the public—which is essentially forced to base its perceptions of crime on limited and general governmental reports, news coverage, and word of mouth (the last two of which can be highly skewed)—but it would also give policymakers and analysts more to work with when seeking to develop sound policy.

Perhaps just as important, it would give traditional law-enforcement institutions (police, prosecutors, and correctional authorities) an indispensable tool with which to defend themselves against attacks from those who seek to seek to neuter, if not outright abolish, these institutions.

However, there is some downside risk to any proposal requiring or encouraging the collection, organization, and uniform reporting of such data. One of those risks is that the imposition of new compliance requirements on resource-constrained agencies may lead to an inefficient use of limited human resources. Therefore, the best way to get better data is to fund its collection, organization, and reporting, while remaining cognizant of the fact that smaller agencies—particularly in rural and exurban jurisdictions—may not have the capacity to gather such data.

Some members of Congress are already thinking about this, as evidenced by a recently proposed bill sponsored by Rep. Nicole Malliotakis (R-NY-11).[59] In addition to Rep. Malliotakis’s bill, Congress should consider launching a grant program aimed at facilitating the acquisition of the technology (particularly database management systems) that will allow jurisdictions to streamline the collection, organization, and reporting of a variety of information on various outcomes in criminal-justice systems throughout the country.

The following proposal draws heavily on the conceptual framework and language of the bill introduced by Rep. Malliotakis, which, if adopted, would certainly be a step in the right direction.

An Act to Provide for Better Crime Data

Section I.—Reporting Requirements

1. Parameters of Annual Reports. On an annual basis, through December 31, 2050, on the first day of the third month of the fiscal year, each chief executive of a [district, state, city, or county prosecutor’s] office shall submit to the state attorney general, the governor of this state, and to the leaders of the [insert state legislative chambers] a report containing the following data pertaining to the previous calendar year:
   1. The total number of cases referred to the office for prosecution of a qualifying offense, disaggregating nonviolent felonies, violent felonies, and misdemeanors;
   2. The number of cases set out in Subsection a) that the office declined to prosecute, disaggregating nonviolent felonies, violent felonies, and misdemeanors;
   3. The number of cases set out in Subsection a) resulting in a conviction, disaggregating nonviolent felonies, violent felonies, and misdemeanor convictions;
   4. The number of total cases resolved via plea bargain and trial;
   5. The number of cases involving qualifying offenses initiated against a defendant who, at the time of the commission of the offense:
      1. Had at least one previous arrest for a qualifying offense arising out of separate conduct;
      2. Had at least one previous conviction for a qualifying offense arising out of separate conduct;
      3. Had at least one open criminal case, disaggregating qualifying and nonqualifying offenses, arising out of separate conduct;
      4. Was serving a term of probation; and
      5. Was on parole.
   6. The average and median sentences—disaggregating sentences of probation, jail, and imprisonment—given to defendants convicted of qualifying offenses, further disaggregating cases resolved via plea bargain from those resolved via trial; and
   7. The number of defendants charged with a qualifying offense who, at their arraignment or pretrial release hearing, were:
      1. Released on their own recognizance;
      2. Required to post bond or bail;
      3. Remanded to pretrial detention; and
      4. Booked into county or city jail as a result of not posting bail or bond at their arraignment or pretrial release hearing.
2. Systematic and Uniform Standards. The state attorney general shall, within 90 days of [insert date of enactment of this Act], define and communicate to each chief executive of a [district, state, city, or county prosecutor’s] office the uniform manner in which the required data shall be defined, organized, and reported pursuant to Section I, including the form that such reports shall take and the process by which such reports shall be delivered to the attorney general, governor, and leaders of the [insert legislative chambers].
3. Cause of Action. Any elected member of the [insert state] legislature shall have the right to assert an institutional injury and seek redress in [insert state civil court] for the case of nonenforcement of this Act by an elected member of the state executive branch. No money damages shall be awarded to an injured party under this section. Redress shall be limited to injunctive relief in the form of a judicial order for the attorney general to carry out his or her duties under this Act. The failure of the attorney general to comply with a judicial order made pursuant to this Act shall result in a finding of contempt.
4. Funding. [set out the total budget and manner of disbursement of said funds to the various offices tasked with collecting, organizing, and reporting the data set in Section I. 1) for the purpose of funding the same].

Section II.—Definitions

1. Terms Defined. When applying this Act, the following terms shall have the following meanings:
   1. “Qualifying offense,” as used in Section I of this Act, shall mean any of the following offenses:
      1. Murder, as defined in [insert applicable state code here]
      2. Rape, as defined in [insert applicable state code here]
      3. Robbery, as defined in [insert applicable state code here]
      4. Aggravated Assault, as defined in [insert applicable state code here]
      5. Burglary, as defined in [insert applicable state code here]
      6. Larceny, as defined in [insert applicable state code here]
      7. Motor Vehicle Theft, as defined in [insert applicable state code here]

There are no silver bullets for the crime problem in the United States. But policymakers who wish to find effective solutions should be encouraged by the fact that, in recent history, the U.S. was able to reduce serious violence to an almost astounding degree—a feat that was achieved without eradicating poverty, reducing economic inequality, or solving any of the other seemingly intractable social problems often pointed to as “root causes” of crime. In the process, there were valuable lessons about the potential effects of selectively incapacitating the repeat, high-risk offenders driving serious crime. These lessons can and should be reapplied by the policymakers of today—not blindly, but with care and an eye toward addressing the legitimate concerns about specific ways in which this goal was achieved, as well as with the understanding that we should take the necessary steps to facilitate the ongoing evaluation of the various approaches taken by those seeking to control crime.

Getting serious crime under control is going to require a multifaceted approach because crime is a multifaceted problem. The proposals outlined here are not a complete policy program but are major components of a multifaceted strategy that recognizes the value of what the criminological literature refers to as social control.[60] Moreover, for jurisdictions that already have some version of these policies, these proposals should be viewed as potential alternatives that inform debates about whether it is worth amending existing laws to make them more effective. It is worth reiterating that these are only starting points for policy development; lawmakers may, through their own fact-finding and political processes, decide that certain provisions—such as the length of the sentencing enhancements or weights attached to offenses in the three-strikes model—should be different from what is set forth above.

The author would like to thank Austin Severns for his invaluable assistance with this project, as well as the other Manhattan Institute scholars who provided thoughtful feedback on earlier drafts. While the assistance received was crucial to the crafting of these model policies and accompanying analysis, any errors should be held exclusively against the author.

Please see Endnotes in PDF

Photo: gorodenkoff/iStock