Did California Prop. 47 Cause State Crime Boost?

Crime reformers say no, but law enforcement ready to pin blame


As the national consumption of margarine increases, so, too, does the divorce rate in Maine and vice versa. Between 2000 and 2009, in fact, those two seemingly unrelated sets of data correlate with eerie precision. Over the same time period, the more Americans buy Japanese cars, the more people commit suicide by crashing their vehicles.

These correlations are irrefutable, yet ridiculous. Who could claim that eating margarine causes people to file for divorce? I gleaned these statistical absurdities from the website of the author of a book called Spurious Correlations. Its subtitle echoes a maxim common among statisticians: "Correlation does not equal causation."

That well-known phrase is worth remembering as Californians analyze the results of Proposition 47, a statewide initiative passed last year that has led to the release from prison and jails of 13,000 offenders. Mainly, it reduced the punishment for a handful of drug and property crimes by reclassifying them from felonies to misdemeanors.

Since the initiative's passage, crime rates have gone up in California, especially in the state's 10 largest cities. Law-enforcement officials, including San Diego Police Chief Shelley Zimmerman, have been blaming the new voter-approved law for the crime increases. Certainly, there's nothing ridiculous about asserting that letting more offenders out on the street might lead to higher crime rates. But just because the correlation exists and the theory is plausible doesn't prove causation.

A November study by the Stanford Justice Advocacy Project looks at the data to date. "The correlation that is suggested by some law enforcement officials, frankly, does not square with the available data, and, certainly, the data that has been released by state agencies… indicates that those who've been released early under Proposition 47 are not responsible for the crimes being reported," argued the project's director Michael Romano.

The Stanford report points to a recidivism rate of 5 percent among those released under the new law, far below the state's typical 42 percent rate. The advocacy project helped draft the initiative, so it's hardly an unbiased source here. Then again, many of the police agencies blaming the proposition for the crime hikes had been strong opponents of it.

Another Proposition 47 backer, the American Civil Liberties Union of California, also looked at the existing data (gleaned from counties, probation departments, health agencies and sheriffs' departments across the state). It released a report last month that calls the measure a success based on reduced incarceration rates and predicted cost savings.

"The fact is that it's way too early to assess 2015 crime rates in California at all, let alone potential causes," according to the report. ACLU officials have, however, argued that some law-enforcement officials are resistant to the initiative's changes, which has resulted in spotty implementation.

The group proposed a variety of ideas to help boost its success. For instance, the ACLU calls for giving police "more options than to arrest," and pointed to cities in other states that let police officers divert petty drug offenders, prostitutes and other low-level offenders directly into treatment and mental-health programs rather than jail.

The Los Angeles Times recently asked in an editorial what the Los Angeles County sheriff's department and district attorney's office have been doing with the extra money from Proposition 47, given the reduction in arrests and incarceration rates: "It is in the nature of bureaucracy… For increases in workload to be accompanied by demands for more funding but decreases in workload to be absorbed without anyone offering up or even acknowledging their savings."

But if law-enforcement agencies do realize savings, that frees up money that can be used for the kind of programs the ACLU report detailed. The Association of Deputy District Attorneys disputes the notion of cost savings, by arguing that costs simply have been shifted around.

Proposition 47 isn't the only major reform that has softened California's tough approach toward crime and incarceration. The Brown administration implemented a "realignment" program that has moved prisoners from state penitentiaries to the county jails (and included early releases) to comply with a federal court order to reduce prison overcrowding. Voters also reformed California's toughest-in-the-nation Three Strikes law in 2012.

Crime rates have generally fallen since those two far-reaching criminal-justice-related reforms. The Public Policy Institute of California last year noted that state violent crime rates are at their lowest level since the 1960s and property crime rates are near record lows. Those latter rates spiked in 2012, but then continued their hard-to-explain free fall.

Of course, the correlation between falling crime rates and criminal-justice reform may be as specious as those previously mentioned oddball comparisons, or the premature declaration that Proposition 47 is the culprit in recent increases. When it comes to crime and recent policy changes, Californians might just need to wait for more data.

NEXT: Movie Review: Krampus

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  1. Who could claim that eating margarine causes people to file for divorce?

    If my wife stocked the fridge with margarine instead of butter, you bet I’d file for divorce. Oh, and I live in Maine BTW.

  2. One of these hypothesis tests is not like the others:

    H0: Consumption of margarine does not increase divorce rates in Maine.
    H1: Consumption of margarine increase divorce rates in Maine.

    H0: Driving a Japanese car does not increase the proclivity to commit suicide by car.
    H1: Driving a Japanese car increases the proclivity to commit suicide by car.

    H0: Early release of non-violent convicts does not increase crime rates.
    H1: Early release of non-violent convicts increases crime rates.

    Sure, correlation does not demonstrate causality. But, as the article admits, some correlations are clearly fatuous, and others are worth considering.

    If “The Stanford report points to a recidivism rate of 5 percent among those released” is true and if the “statewide initiative passed last year that has led to the release from prison and jails of 13,000 offenders” is true, it would seem that approximately 650 of those released early have re-offended. Assuming that they are not complete idiots, that probably translates to something like 10,000 crimes that would not otherwise have occurred.

    1. Of course, Prop 47 could have effects other than this. If it created the perception that the penalties for criminal behavior have been reduced (i.e., the criminal’s “cost of production”), it makes crime more profitable, thereby encourage new market entrants and more crime. On the plus side, if by releasing the 13,000 offenders early, the State is able to prevent release of even more dangerous violent offenders due to overcrowding, Prop 47 may be protecting society from more vicious crimes.

      Unfortunately, there is no way to definitively calculate either of these two indirect consequences of Prop 47. However, the decline in violent crime rates cited by the article is encouraging.

      It would be far, far better for California to reduce the number of crimes by decriminalizing vices such as narcotics, prostitution, and gambling. That way many of these unfortunate souls would avoid prison altogether.

      1. I voted for this law assuming it was mainly about drug crimes.

        Only after it passed, a Facebook friend who is very apolitical but hates thieves with the passion of a thousand suns (especially bike thieves) posted that Prop 47 also affected property crimes and how this would come back to bite us all in the rear.

        Yep. I was rolled.

        If you make stealing anything under $800 a nothingburger, guess what will happen?

        A lot of petty theft.

        The release of prisoners may not matter as much as thieves figuring out they won’t have many consequences to their crimes.

        Also, on a related note, supermarkets like Safeway now do not even attempt to apprehend shoplifters due to the legal risk. They don’t call the police either. Some people have figured this out and now blatantly steal.…..530612.php

        I understand why Safeway allows this: they save on lawsuits and they can pass the cost onto consumers.

        But for “society” this may lead to a much higher cost of a breakdown of cultural rules about theft.

        1. “These offenses include shoplifting, writing bad checks, most drug possession cases, receiving stolen property ? and forgery, and theft of money or goods worth less than $950.”

          From the comments on the article.

          Also, police may simply not pursue these cases because “its only a misdemeanor.” That said, the police don’t really do much against theft as it is. If we end the war on drugs, and we can’t fire any law enforcement, we should demand a re-purposement to focus on solving smaller crimes that don’t excite cops but anger citizens.

        2. and that was my second thought – will those who were not previously in jail believe the consequences for crime, especially petty crime, no longer a deterrent?

    2. well reasoned comment, and it is a subject for further research

      If early release is tied to an increase in crime, I’d look to see how much crime is being committed by those released.

  3. “The Stanford report points to a recidivism rate of 5%”
    I call bullsh*t!
    That percentage is unheard of in any study I’ve ever read.

    1. Th study defines recidivism as a return to prison. They are not considering any of the persons who weren’t in prison to begin with but who nevertheless benefited from Prop 47. Additionally because the new definition of petty theft can’t send you to prison unless you steal a certain amount or have a very narrow range of prior convictions, there will be substantially fewer people returning to prison. In other words, according to the Stanford project you are not a recidivist because the statute won’t let you be one if you steal less than $950. Even then there is no guarantee you will be sentenced to prison by the court. There has been a substantial reset of criminal penalties in day-to-day practice because of realignment and prop 47.

      Finally, the 5% only refers to those who are caught, prosecuted, convicted, sentenced; it does not capture all true recidivists.

  4. the 5% recidivism rate is a faulty number since those released and recommitting crimes are not being returned to jail if their crime is below a new high level, like stealing guns for example, the 5% only accounts for those who have increased their level of crime above the $950.00 cut off. bad numbers from people with a purpose behind those numbers

    1. furthermore cops are actually telling people to not bother to file since nothing will happen. this has happened to me and to a friend who robbed of a lot of work tools. So yes the new law has increased crime because the criminals know nothing will happen

      1. correction “a friend who was robbed of a lot of work tools”

      2. Petty theft has been a thing that no one cared about for years. Years ago when I worked retail I was told to accept all returns simply because it was cheaper to give them their money or exchange than it was to bother a manager. The same with people running out the door. The store was out $15 for whatever they could carry, so if you went and apprehended them, the store would pay more for your effort than whatever the person got away with.

        Years later, a construction company I worked for did the same thing. People would break into pickup tool boxes and loot them. The company would just file a police report and replace the contents (to get stolen 6 months later). Doing more was too expensive.

        Petty theft from the police side is the same way in terms of inaction, but for a different reason: if they generate a report it makes their “unsolved crime” statistics go up and makes them look bad. They tend to tell you to not file but usually will anyway if you press them to do so for a report number.

        From an individual cost:benefit analysis, I can see that line of thinking. But from an aggregate stand point, having no repercussions or not punitively punishing a petty thief for them labor to bring them in seems to encourage them.

  5. H0: Decriminalizing petty theft will not increase petty theft
    H1: Decriminalizing petty theft will increase petty theft

  6. Property crime set a *record* in San Fran. 25000 crimes or SEVENTY PER DAY.

    Typical loss is $2000 per citizen. That’s a $50M/year cost burden on tax payers *in addition* to the stress of being robbed over and over.

    But yeah… I’m sure that’s just correlation. And the best part is the “$100M in savings!!!!” at *best* goes to “programs” to provide even more help for professional smash and grab thieves and at *worst* goes nowhere at all.

Please to post comments

Comments are closed.