Not Taking Crime Seriously California’s Prop 47 Exacerbated Crime and Drug Abuse

In November 2014, California voters approved a criminal justice reform measure, Proposition 47 (“Prop 47”), with almost 60% support.[1] Ten years later, California voters are now considering rolling back some of its soft-on-crime policies. Prop 47 identified six “petty” crimes—grand theft, larceny, personal drug use, forgery, and two types of check fraud—and reclassified them. It downgraded these crimes, including thefts with property values under $950 and illegal drug possession for personal use, from felonies to misdemeanors.

This paper presents a data-based argument on how Prop 47 shifted dynamics in both offender behavior and prosecutorial decision-making that damaged public safety and public health. Representative data from Riverside, one of California’s largest counties, suggest that Prop 47 increased re-offending, including serious felony re-offending, detention times, failures to appear in court, warrants issued on offenders, case dismissals in conjunction with plea deals, and the persistence levels of drug and theft offenders. Additional data collected from both Riverside and San Bernardino law-enforcement agencies show a significant drop in sentencing and in arrests, due partly to the diminished incentive for businesses to promptly report thefts.

These shifts have also resulted in fewer defendants participating in in-custody drug treatment programs or other mandatory, supervised services because the incentives for doing so (avoiding prosecution and significant sentences) have evaporated. And, as California business owners can attest, reducing the cost of repeatedly committing theft removes the incentive for offenders to change their behavior. This has fueled increases in organized retail theft and fencing rings. Prop 47 also strained the resources of counties, by increasing the number of defendants sentenced to serve in overcrowded jails rather than prison.

On November 5, Californians will vote on whether to allow felony charges and increased sentences for certain drug and theft crimes. This opportunity comes as many Californians are frustrated with measures that businesses are taking to prevent out-of-control retail theft, from store closures to locking up all products behind barricades.[2] If passed, Proposition 36 would again allow felony charges for possessing certain drugs and for thefts under $950 for defendants with two prior drug or theft convictions.[3] This paper does not necessarily endorse Prop 36, but the fact that it threatens Prop 47 may suggest that California residents are dissatisfied with the original reform’s crime consequences. Readjusting down the felony threshold would be one step in the right direction toward correcting some of the harms delineated below.

Understanding the impact of Prop 47 has important implications for jurisdictions nationwide. California was not alone in passing legislation in recent years that aimed at lessening consequences for “low-level” offending. For example, Chicago raised the threshold for felony shoplifting, and New York State made hundreds of crimes no longer eligible for bail setting.[4] As in California, the stated intentions of these reforms were to replace criminal justice responses with social services and to reduce incarceration levels, but the result has been more re-offending and less participation in effective treatment programs.

In the early 2000s, California’s state prison system faced an overcrowding crisis, augmented by policies of the preceding decade aimed at reducing high crime rates by, among other things, keeping offenders off the streets. By 2011, the resulting increase in incarceration rates had stretched the population of a prison system designed to house 85,000 inmates to approximately 156,000.[5]

In response to this intense overcrowding, several legal challenges culminated in a 2009 federal court order requiring California to cut its prison population down to 137.5% of design capacity. This decision was upheld in 2011 by the U.S. Supreme Court (Brown v. Plata),[6] with a ruling that the state prison conditions violated the Eighth Amendment, which prohibits cruel and unusual punishment. As a result, the California Department of Corrections and Rehabilitation (CDCR) was tasked with reducing the inmate population.

To comply, California implemented reforms such as Assembly Bill 109 in 2011,[7] “realigning” nonserious, nonviolent, and nonsexual offenders from state prisons to county jails and probation systems. Proposition 36,[8] enacted in 2012, amended the three-strikes law to reduce life sentences for certain nonviolent offenses.

Unfortunately, by relieving overcrowding in prisons, AB 109 exacerbated it in jails in counties where it was already a problem, such as Riverside.[9] Indeed, jail overcrowding has plagued some California counties for decades, as these were responsible for housing all pretrial, in-custody defendants, as well as all defendants sentenced to jail as part of a misdemeanor sentence.

Riverside County’s Sheriff’s Office, which runs its jail, operates under a permanent federal injunction, requiring it to release inmates once it has reached its maximum bed capacity.[10] Indeed, Riverside County jails were already running near or at capacity before AB 109 took effect, so they were forced to start regularly releasing people from custody. The cumulative result has been the release of many inmates—not for considerations of justice but merely due to facility limitations.

It was in this context, with lingering concerns about overcrowding, that California voters approved Prop 47. The intent was to reduce the state’s prison population and reallocate resources toward community-based services. To this end, Prop 47 established the Safe Neighborhoods and Schools Fund, directing the savings from reduced prison costs into programs aimed at mental health and drug treatment, homelessness prevention, K–12 educational programs, and support for crime victims. By reclassifying several nonviolent felonies as misdemeanors, Prop 47 changed the trajectory of defendants from prison to jail—and from jail to the streets, due to overcrowding.

Prop 47’s changes applied retroactively, allowing individuals serving sentences for these reclassified crimes to petition for resentencing. But the impact was even broader. Under AB 109, some defendants who had prior convictions for certain crimes were still required to serve their felony sentences in prison, even if the new crime was affected by the law. But if that crime was one that Prop 47 then reduced to a misdemeanor, anyone convicted of that offense would serve his custody time in county jail instead of state prison.

Arguably, Prop 47 was unnecessary for achieving many reduced sentences. Pre–Prop 47, for a defendant with a prior theft-related crime who committed a new crime in which (1) the amount taken was less than $950, or (2) was a commercial burglary,[11] prosecutors had discretion to charge the new incident as either a misdemeanor or a felony.

Pre–Prop 47 discretion over charging felonies also provided the safeguard that people convicted of a felony are subject to closer supervision after release than those convicted of a misdemeanor. Individuals with a history of theft tend to pose more of a threat than those who commit theft once. But Prop 47 took away prosecutors’ ability to charge those repeat offenders with more serious felony crimes, instead treating all their crimes as misdemeanors, where the maximum sentence is 364 days in jail for some crimes, and only 180 days for theft.

Prop 47 was a significant shift in California’s approach to criminal justice, investing in the popular narrative that slashing incarceration while increasing spending on services would provide better outcomes for both defendants and public safety. Several previous studies, however, found that property crimes increased under Prop 47. One 2018 report found that Prop 47 may have contributed to a larceny theft increase of approximately 9% in California, compared with other states.[12] Another study from that same year found a moderate increase in larceny and motor vehicle thefts.[13] And a 2019 analysis found that property crime increased 5%–7%.[14]

While other analyses support the innocuousness of Prop 47, this report’s findings indicate its harm to public safety.[15] As we will see, by removing some of the incentives and oversight that actually motivate at-risk individuals to correct their behavior, Prop 47 has had detrimental effects.

Despite the positive intentions behind Prop 47—reducing stringent punishments and investing in services—evidence from Riverside County, one of the state’s most populous counties, points to many negative outcomes.

Method

The Riverside County District Attorney’s (RCDA) office constructed nine data analyses to compare the period immediately before the passage of Prop 47 with the period soon after its implementation. These analyses are listed below. To ensure an appropriate comparison of defendants, the results compare individuals who are of the same gender and age and share various aspects of their criminal history, including, most importantly, the top charge associated with their case. For Figures 6ff., this was achieved using a method called Nearest Neighbor Matching (NNM), which identifies individuals who are identical across all these features.

Critically, the analysis includes only individuals who are charged with an offense that was affected by Prop 47. By narrowing the analysis to individuals who are comparable along all these margins, the set of defendants affected by Prop 47 can be compared with a nearly identical set of defendants who were prosecuted in the period before the passage of Prop 47. This approach better allows comparisons of like-to-like in measuring the impact of Prop 47 on criminal justice and public safety outcomes.

Data Set

The data used here include administrative records from RCDA from November 2013 to November 2015. The quantitative analysis itself was contracted internally by RCDA, and this report utilizes that work, obtained directly from RCDA. The data comprehensively encompass information at the case-defendant level, i.e., each observation in the data used in this analysis separately examines each defendant associated with a specific case—even if multiple defendants are associated with a case.

The data include information about which charges were referred to the prosecutor’s office by law enforcement, the decision to file charges and the specific charges that were filed, the disposition of the case, and the dates associated with each of these stages of the prosecution of a case. The period between the filing and disposition date determines the case length.

Of course, one limitation of this analysis of Prop 47’s effects is that it is restricted to data from Riverside County. Nevertheless, Riverside is the fourth most populous of California’s 58 counties, with roughly 2.5 million residents. Therefore, these findings paint a powerful data picture—and suggest that similar studies should be done throughout California. The geographical limitations of the data mean that some outcomes will be systematically undercounted. For example, this analysis will examine the number of “trailing referrals”(new charges) that a defendant incurs while his initial case is being prosecuted. If, however, the defendant has new charges filed in another county during the time that his case is being prosecuted in Riverside, this would not be captured in the data. Similarly, the analysis will examine various measures of recidivism. Once again, if a defendant re-offends in another county, this information will not be captured in the measure of recidivism utilized in this analysis.

Additionally, as discussed below, when retailers doubt that contacting law enforcement will result in prosecution or other consequences, they sometimes alter their policies, reporting fewer retail thefts than are committed.[16] Thus, this report should be viewed as a best attempt to identify a set of comparable defendants to analyze the impact of Prop 47 on defendant outcomes and prosecutor decisions. As a recent Manhattan Institute brief that focused on shoplifting rates and causes described, many factors confound forming a complete picture;[17] but the data presented here indicate that Prop 47 has had a significant and detrimental effect.

Overview: Comparing 2014, 2016, and 2023 Public Safety Metrics

Table 1 provides a snapshot of the Riverside County public safety metrics presented throughout this report before the passage of Prop 47 (2014 column), shortly after the passage of Prop 47 (2016 column), and currently (2023 column). The rates listed in the table represent the proportion of cases in each offense category out of all relevant cases that year. As depicted below, the negative impact of Prop 47 has been persistent or become worse. For example, nearly every measure of recidivism increased soon after the passage of Prop 47, and has continued increasing. (As a reminder, “trailing referrals” are new charges for individuals with pending cases.)

Table 1

Change in Measure of Public Safety Produced, 2014, 2016, and 2023

Measure	2014 (Before Passage of Prop 47)	2016 (Immediately Following Passage of Prop 47)	2023
Trailing cases—misdemeanors	3%	14%	19%
Trailing cases—felonies	20%	10%	13%
Chronic offenses—theft	8%	13%	16%
Chronic offenses—drug	19%	17%	28%
Recidivism—referred	47%	47%	51%
Recidivism—convicted	39%	39%	36%
Recidivism involving felony—referred	31%	30%	34%
Recidivism involving felony—convicted	26%	25%	27%
Recidivism involving misdemeanor—referred	36%	38%	40%
Recidivism involving misdemeanor—convicted	27%	28%	23%
Recidivism involving serious felony—referred	4%	5%	6%
Recidivism involving serious felony—convicted	3%	3%	4%
Recidivism involving violent felony—referred	5%	4%	6%
Recidivism involving violent felony—convicted	4%	3%	4%
Recidivism involving victim—referred	14%	17%	21%
Recidivism involving victim—convicted	12%	14%	14%
Case length (no. of days)	192	220	309
Any failure to appear	26%	28%	30%
Warrant issued within 30 days	16%	19%	25%
Dismissal in lieu of plea	2%	3%	6%

In addition to the worsening public safety stats, case proceedings metrics also deteriorated. The average case length increased by nearly 15% soon after the passage of Prop 47—and has continued to grow. By 2023, the case length had increased by 61% relative to the period immediately before the passage of Prop 47. This is likely explained, in part, by a 56% increase from 2014 to 2023 in the probability that a case is in warrant status—i.e., a warrant has been issued for a defendant who failed to appear in court—within 30 days of filing. More concerning is the fact that RCDA has dismissed 200% more cases in conjunction with plea agreements (from 2% in 2014 to 6% in 2023).

Next, this paper will examine nine of these data points—and one additional point demonstrating decreased arrests for Prop 47 crimes—in greater detail.

1. Drop in Felony Cases

Following Prop 47, the total number of felony cases decreased by nearly 30% while misdemeanor cases rose marginally by about 3.5%. This was to be expected, since many offenses that were formerly felonies were reclassified as misdemeanors. However, the scale of this shift is noteworthy in the change it represents within the dynamics of the criminal justice system.

Figure 1 depicts the total number of felony or misdemeanor cases filed between November 2013 and October 2014 (before Prop 47) and December 2014 and between November 2015 (after Prop 47). Cases are identified as a felony or a misdemeanor based on the “top charge” or most severe offense filed against a case-defendant. If the top charge is a felony, the case is categorized as a felony even if additional misdemeanor charges are filed. I.e., a case is categorized as a misdemeanor case only if the top charge is a misdemeanor.

As depicted in the left two bars, there were 26,426 felony cases in the year before Prop 47, but only 18,533 cases in the year after.

2. Drop in Felony Defendants

Following Prop 47, the total number of unique defendants decreased by more than 23% for felons, while only dropping 2.3% for misdemeanants (Figure 2). Again, Prop 47 reclassified many criminal activities from felonies to misdemeanors—but the size of the shift is noteworthy.

Before the passage of Prop 47, there were a little more than 19,000 unique felons prosecuted by RCDA. After the passage of Prop 47, the number fell to just under 15,000 felons. Misdemeanor defendants decreased to 760 fewer unique misdemeanants.[18]

3. Decreased Cases per Felon; Increased Cases per Misdemeanant

The number of cases per defendant stayed relatively stable after the passage of Prop 47. However, there is a near exact offset in the reduction of cases per felon by the increase in cases per misdemeanant. The shift is significant because it indicates that offenses were more reclassified than ameliorated: crime is not going down; it is being renamed.

The number of cases per felon decreased by 31.6%, likely because of the reclassification of felonies to misdemeanors. Correspondingly, the number of cases per misdemeanant increased by over 47%.

Combining data from Figures 1 and 2 displays an important finding about the changes in the number of cases per unique defendant before and after the passage of Prop 47. Figure 3 displays the average number of cases for all defendants, felons, and misdemeanants.

Again, the total number of cases per defendant changed very little after the passage of Prop 47, resulting in approximately 2% fewer cases per defendant. But the number of cases per felon decreased by 31.6%, while cases per misdemeanant increased by over 47%. Thus, there appears to be a nearly identical transfer of cases from felonies to misdemeanors post–Prop 47.

4. Increase in Misdemeanor Trailing Referrals and Cases Going to Warrant

One important metric of public safety is the number of new cases filed against a defendant while awaiting the disposition of his initial case. The new cases are called “trailing referrals.” The following figures display two sets of trailing referrals broken apart by felony and misdemeanor case referrals from law enforcement.

Figure 4 shows that after the passage of Prop 47 at the end of 2014, the number of trailing felony case referrals—i.e., new felony charges filed against defendants with pending cases—decreased by about 400 cases per quarter. Conversely, trailing misdemeanor cases increased by about 800 cases. Thus, it appears that trailing referrals increased by approximately 5% after the passage of Prop 47.

Figure 5 focuses on cases in which a warrant has been issued because a defendant failed to appear for his court case. Cases that have gone to warrant make up approximately 20% of all cases. After the passage of Prop 47, the number of trailing referrals for felony cases for which warrants were issued remains virtually unchanged. Problematically, the number of misdemeanor cases that have led to a warrant being issued increased by over 60%.

In other words, after Prop 47, misdemeanants are committing many new misdemeanors while their cases are pending—and they are less and less likely to show up to court for these cases. This speaks to a decreased fear of repercussions for not coming to court.

5. Surging Case Dismissals in Conjunction with Plea Deals

As noted, many defendants acquire new cases while their initial case is ongoing. A common practice in prosecutors’ offices is to consolidate several cases to develop a plea agreement for all of them. The result is that the defendant will have a plea agreement on just one case and all the rest dismissed. Figure 6 (unlike the previous figures) utilizes Nearest Neighbor Matching techniques to examine differences in the number of dismissed cases in conjunction with a plea agreement for observationally similar defendants.

Before the passage of Prop 47, defendants who would be affected by Prop 47 had 1.5 cases, on average, dismissed in conjunction with a plea agreement. After Prop 47, this number rose to nearly four cases dismissed, along with the offer of a plea agreement. This constituted a 166% increase in cases dismissed in conjunction with a plea agreement, suggesting that Prop 47 is not deterring defendants from allegedly committing more crimes.

This is detrimental because offenders are averaging more crimes before their first is resolved. Now, a prosecutor handling a defendant’s case will frequently see him commit four or more new cases while the first is ongoing. Overburdened prosecutors are forced to bundle all these trailing cases together, offer a plea deal on just one, and dismiss all the rest. This removes some of the deterrence against committing new offenses, since, on average, defendants are more likely to see all their additional trailing cases dismissed. This also represents more crime victims along the way. And prosecutors’ conviction rates will appear artificially low because they are bundling more cases together, while still offering a plea deal on one and dismissing the others. Nor can prosecutors bundle together the value of, for instance, multiple theft cases, in order to achieve a higher charge or greater sentence.

In other words, chronic offending has increased under these dynamics, while culpability has decreased.

6. Surge in Cases Out on Warrant and Failures to Appear

When a judge must decide whether to release a defendant on bail or on his own recognizance, the judge is often concerned with whether the defendant poses a threat to the community and the likelihood that he will return to court for future court proceedings. When a defendant fails to appear for a hearing in his case, a warrant is issued for the defendant.

Cases for which warrants were issued significantly increased after the passage of Prop 47. The increase ranges from a 62.5% increase in the likelihood that a case is out on warrant for 30 days to a 200% increase in the likelihood that a case is out on warrant and is never recalled.

Figure 7 breaks the data into five categories: (1) cases that have been out on warrant for longer than 30 days; (2) cases that have been out on warrant for longer than 90 days; (3) cases that have been out on warrant for longer than 180 days; (4) cases that have been out on warrant for longer than 365 days; and (5) cases that have never been recalled. While the empirical differences in the likelihood that a case is out on warrant differs across these five groups, the changes are generally parallel.

Specifically, after Prop 47, there were large increases in the number of cases out on warrant. For example, pre–Prop 47, approximately 4% of cases were out on warrant for more than a year, but the share of cases out on warrant for longer than a year jumped to over 10% post–Prop 47.

These increases represent a less efficient and less effective criminal justice system, in which delinquent defendants prolong their cases as they fail to show up to court. This is also related to the dynamic depicted in Figure 6: failures to appear in court are driving longer case times. This, in turn, gives defendants more time to commit new, trailing offenses while they are dodging court for earlier cases. The net result is that, on average, more cases get bundled up and dismissed with each plea deal.

7. Increased Length of Time to Prosecute

As noted above, the disposition of cases changed drastically after the passage of Prop 47. As described, more cases are now referred to as misdemeanors, defendants are more likely to have cases dismissed in conjunction with a plea agreement, and cases are significantly more likely to have warrants issued on defendants.

These changes can have a significant impact on the length of time required for a case to reach a final disposition. Figure 8 displays the average case length before and after Prop 47 for matched defendants of all case types—and then separately for felony and misdemeanor cases. The measure of the case length is calculated as the length of time between the filing date and the date of the first disposition of charges in a case.

Before Prop 47, the average case length across all case types was approximately 160 days. After the passage of Prop 47, the average case length increased a shocking 74%, to nearly 300 days. Examining the case length by the severity of the case reveals that misdemeanor cases experienced a larger increase in case length after the passage of Prop 47, perhaps because many of these new time-consuming misdemeanor cases would have once been time-consuming felony cases.

8. More Drug and Theft Offenders Are Becoming Chronic Offenders

Another concern, post–Prop 47, is that more drug and theft offenders are becoming chronic offenders—despite the targeted rehabilitation programs that accompanied the softer criminal justice response. Figure 9 examines changes in the persistence of drug and theft offenders after the passage of Prop 47. Prior to the law, nearly 60% of all individuals with a drug offense were chronic drug offenders, defined here as having had three or more such offenses referred to RCDA. This definition is especially useful because one component of Prop 36, the ballot initiative under consideration of California voters in this election, would change the sentencing associated specifically with those drug and theft offenders who have three or more offenses.[19]

After the passage of Prop 47, the share of drug offenders who were chronic drug offenders increased by 10.3%. Similarly, the share of theft offenders who were chronic theft offenders increased by 6.3%. Notably, of these chronic offenders, roughly 20% were both chronic drug and theft offenders before and after Prop 47 was enacted. This is not surprising, since these two offenses are often linked. E.g., chronic shoplifters might steal to support drug addictions, while chronic drug abuse might fuel shoplifting.

This detrimental dynamic is one reason that more rigorous criminal justice consequences often deter offenders from developing into chronic offenders better than the alternatives. With the softening of prosecutorial and carceral responses under Prop 47, offenders move toward a cycle of more offending rather than toward an off-ramp to a better lifestyle.

9. Increased Recidivism Overall for Serious Crimes and Crimes Involving Victims

Recidivism rates are an important metric for evaluating criminal justice policies. As a measure of recidivism, this section analyzes whether a defendant is referred and whether he is convicted on a subsequent charge within three years. Figure 10 examines the proportion of case-defendants who recidivate and who are referred or convicted of a serious felony, including a violent felony, or a crime involving a victim.[20]

These data show a reduction in violent felony referrals and convictions after the passage of Prop 47. However, there were even larger increases in serious felonies overall and in cases involving victims. Cases involving serious felony referrals and convictions both rose after the passage of Prop 47. Serious felony referrals increased by nearly 23%, while convictions increased by 20%. Arrests involving a victim increased 42%, with nearly one in four arrests involving a victim. Similarly, convictions involving a victim increased by nearly 36%, such that nearly one in five case convictions after Prop 47 involved a victim.

Additional reporting from a July 2024 report issued by the auditor of the State of California found that the number of individuals convicted four or more times rose 12% following Prop 47.[21] Using data from the Riverside County Sheriff’s Office and the police departments of Riverside and San Bernardino, the auditor also found that theft convictions among those who re-offended four or more times increased by 14%. The report suggests that reduced sentences may be driving this increase in individuals committing multiple offenses in general and multiple theft offenses in particular. Indeed, the auditor found that the average sentence length for Prop 47 offenses decreased by more than 40% after the measure’s implementation—as recidivism rose.

10. Decreased Arrests for Prop 47 Offenses

An important aspect of Prop 47’s impact on prosecutorial decisions and offender behavior is the changes that these have caused in police enforcement.

According to the July 2024 report from the auditor of the State of California, the Riverside County Sheriff ’s Office and the police departments of Riverside and San Bernardino arrested 30%–67% fewer individuals for Prop 47 offenses in the period following the implementation of the law than prior.

The auditor analyzed the number of arrests, including citations and bookings, for Prop 47 offenses and non–Prop 47 offenses during 2009–22 for Riverside Police and Riverside County Sheriff’s Office; and 2012–22 for the neighboring San Bernardino Police. The report found that arrests for Prop 47 offenses declined at all three agencies during 2014–22. In 2014, for example, Riverside County Sheriff’s Office made 8,960 arrests for Prop 47 offenses; but in 2022, it made only 6,228 such arrests, a 30% drop. This was a significantly greater drop than a congruent, milder decline in arrests for non–Prop 47 offenses.

The auditor’s report cites one law-enforcement executive’s explanation that police put their energies where they would be the most effective, and Prop 47 “had a cooling effect on agencies’ devotion of scarce resources to Proposition 47 offenses.” With many arrestees more likely to end up offending again and within a shorter time frame, police become less motivated to make arrests.

One of the animating aims of Prop 47, according to its advocates, was the rehabilitation of individuals who are experiencing drug or mental health issues. Indeed, the language on the ballot noted that one objective of Prop 47 was to apply the “savings to mental health and drug treatment programs.”[22] The argument, popular across many American cities, is that these issues are better ameliorated by keeping sufferers out of the criminal justice system and in voluntary care.

However, this misses the critical interplay between defendant behavior and the leverage that a robust criminal justice system can provide for moving offenders into supervised treatment and away from both temptation and a vulnerable public.

Statistics from the Substance Abuse and Mental Health Services Administration (SAMHSA), which collects substance abuse treatment admissions rates in its Treatment Episode Data Set (TEDS), suggest that abuse has gotten only worse under Prop 47.[23] While admissions rates sometimes remain level or decline, even during periods of increased drug abuse in the population, the Riverside County area saw increased admissions post–Prop 47.

More specifically, TEDS includes data for the Inland Empire, a region primarily centered in Riverside and San Bernardino Counties. In 2014, it recorded 15,378 drug admissions; in 2016, following Prop 47’s passage, there were 19,296 drug admissions. This marks a nearly 16% increase in drug admissions under the law.[24]

Drug overdose rates serve as another important measure of public health safety. According to the Riverside University Health System, between 2011 and 2015, spanning the introduction of Prop 47, the number of overdose deaths increased from 10 deaths per 100,000 people to 15.2 overdose deaths.[25] This represents an increase of 52% from 2011 to 2015 and an increase of 8.5% just from 2014 to 2015. One qualification of these data is that this period coincided with the spread of fentanyl use. However, opioid overdose deaths constituted 50% of all overdose deaths in 2011, and had fallen to 35% of all overdose deaths by 2015. This indicates that the increase in overdose deaths in this period was not likely driven by the opioid epidemic.

One explanation for this shift in overdose rates is that Prop 47 took away much of the incentive for defendants with substance abuse issues to sign up for in-custody treatment programs. Those programs require defendants to take steps like attending classes, going to Alcoholics Anonymous or Narcotics Anonymous meetings, checking in regularly with the judge, and more. The incentive to participate in these supervised programs was the threat of harsher sentences: a yearlong program was more appealing than a felony sentence and more time in custody.

Now that similar defendants receive only misdemeanor sentences, these programs are longer than the amount of time that they would serve in custody. Accordingly, program participation dropped, according to RCDA.[26] This trend makes it harder to help even defendants who want to seek treatment for their addiction to dangerous drugs like methamphetamine, cocaine, and fentanyl.

A decade since the passage of Prop 47, the data indicate that moving the threshold for what constitutes a felony has significantly altered the behavior of criminals and the decision-making of prosecutors and police. Rather than inspiring offenders to move away from criminality and toward voluntary treatment programs for substance abuse, Prop 47 appears to have increased chronic, serious offending, especially relating to theft and drugs. The law has also deprived defendants of the incentive to join involuntary treatment programs that could more reliably improve their lives and future choices.

For all these reasons, it is important to repair the damaging overreach of the law.

Prop 47 can be amended by a two-thirds vote of the members of each house of the California Legislature, along with the governor’s signature—but only if the amendments are consistent with, and further the intent of, the proposition. Alternatively, the legislature may also enact a proposal to amend Prop 47; however, the proposed amendment would not become effective until submitted to and approved by the voters. This latter process is required if the amendment is inconsistent with or does not further the proposition’s intent. Further, Californians may introduce ballot initiatives that amend or repeal Proposition 47.

One such step, ballot initiative 23-0017A1, Prop 36, is qualified for the statewide November 2024 general election.[27] If the voters adopt it, one consequence of this initiative would be to make a third conviction for petty theft and shoplifting a potential felony.

This measure would address some of the mechanisms that Prop 47 harmfully altered. However, it stops short of supporting the best outcomes for both public safety and for offenders who base their behavior on the deterrents and incentives created by criminal justice laws.

Rather, the full range of tools should be restored to the criminal justice system to punish serial thieves with more discretion over felony charges and sentencing. These responses were not simply punishments: they deterred chronic theft, kept citizens and businesses safe, and incentivized individuals who were stealing to feed a substance addiction to enter substantial treatment programs. A return to the pre–Prop 47 threshold for felony charges across crime categories, and/or expanding the recidivist provision across crime categories, would once again push at-risk offenders away from criminality and drug abuse, and offer them and the citizens of California better opportunities to thrive.

To this end, California will also need to identify better solutions for the overcrowding of jails and prisons that, in part, spurred the passage of Prop 47 and aggravated some of its negative outcomes. As challenging and costly as expanding facilities might be, current efforts have detrimental impacts on public safety and long-term overcrowding.

Finally, similar statistical analysis should be run in other jurisdictions where analogous measures have been passed. In New York, for instance, research has shown that statewide bail reform laws in 2020 have since increased re-offending among chronic recidivists.[28] Understanding the dynamics behind these negative outcomes for well-intentioned reforms will help all American jurisdictions pass laws that are truly progressive, helping advance the lives of citizens and defendants alike.

Hannah E. Meyers is a fellow and director of policing and public safety at the Manhattan Institute. She is an appointed member of the New York State Domestic Terrorism Task Force. Previously, Meyers managed corporate and private investigation teams for an international firm and directed research strategy for a counter-extremism NGO. She served for five years with the Intelligence Bureau of the New York City Police Department. Meyers holds an M.A. in international relations from Yale University—for which she performed research embedded with British law enforcement—and a B.A. in government from Dartmouth College.

Please see Endnotes in PDF

Photo by AGUSTIN PAULLIER/AFP via Getty Images