The dramatic rise of incarceration and the precipitous fall in crime have shaped the landscape of American criminal justice over the last two decades. Both have been unprecedented. Many believe that the explosion in incarceration created the crime drop. In fact, the enormous growth in imprisonment only had a limited impact. And, for the past thirteen years, it has passed the point of diminishing returns, making no effective difference. We now know that we can reduce our prison populations and simultaneously reduce crime.
This has profound implications for criminal justice policy: We lock up millions of people in an effort to fight crime. But this is not working.
The link between rising incarceration and falling crime seems logical. Draconian penalties and a startling expansion in prison capacity were advertised as measures that would bring down crime. That’s what happened, right?
Not so fast. There is wide agreement that we do not yet fully know what caused crime to drop. Theories abound, from an aging population to growing police forces to reducing lead in the air. A jumble of data and theories makes it hard to sort out this big, if happy, mystery. And it has been especially difficult to pin down the role of growing incarceration.
So incarceration skyrocketed and crime was in free fall. But conflating simple correlation with causation in this case is a costly mistake. A report from the Brennan Center for Justice at NYU School of Law, called What Caused the Crime Decline? finds that increasing incarceration is not the answer. As Nobel laureate economist Joseph Stiglitz writes in the foreword, “This prodigious rate of incarceration is not only inhumane, it is economic folly.”
Our team of economic and criminal justice researchers spent the last 20 months testing fourteen popular theories for the crime decline. We delved deep into over 30 years of data collected from all 50 states and the 50 largest cities. The results are sharply etched: We do not know with precision what caused the crime decline, but the growth in incarceration played only a minor role, and now has a negligible impact.
The Crime Decline
The drop in crime stands as one of the more fascinating and remarkable social phenomena of our time. For decades, crime soared. Cities were viewed as unlivable. Politicians competed to run the most lurid campaign ads and sponsor the most punitive laws. Racially tinged “wedge issues” marked American politics from Richard Nixon’s “law and order” campaign of 1968 to the “Willie Horton” ads credited with helping George H.W. Bush win the 1988 election.
But over the past 25 years, the tide of crime and violence seemed to simply recede. Crime is about half of what it was at its peak in 1991. Violent crime plummeted 51 percent. Property crime fell 43 percent. Homicides are down 54 percent. In 1985, there were 1,384 murders in New York City. Last year there were 333. The country is an undeniably safer place. Growing urban populations are one positive consequence.
During that same period, we saw the birth of mass incarceration in the United States. Since 1990, incarceration nearly doubled, adding 1.1 million people behind bars. Today, our nation has 5 percent of the world’s population and 25 percent of the world’s prison population. The United States is the world’s most prodigious incarcerator.
The Role of Incarceration
What do the numbers say? Did this explosion in incarceration cause the crime decline?
It turns out that increased incarceration had a much more limited effect on crime than popularly thought. We find that this growth in incarceration was responsible for approximately 5 percent of the drop in crime in the 1990s. (This could vary from 0 to 10 percent.) Since then, however, increases in incarceration have had essentially zero effect on crime. The positive returns are gone. That means the colossal number of Americans cycling in and out of prisons and jails over the last 13 years was not responsible for any meaningful fraction of the drop in crime.
The figure below shows our main result: increased incarceration’s effectiveness since 1980. This is measured as the change in the crime rate expected to result from a 1 percent increase in imprisonment—what economists call an “elasticity.” During the 1980s and 1990s, as incarceration climbed, its effectiveness waned. Its effectiveness currently dwells in the basement. Today, a 1 percent increase in incarceration would lead to a microscopic 0.02 percent decline in crime. This is statistically indistinguishable from having no effect at all.
Increased incarceration accounted for about 6 percent of the property crime decline in the 1990s, and 1 percent of that drop in the 2000s. The growth of incarceration had no observable effect on violent crime in the 1990s or 2000s. This last finding may initially seem surprising. But given that we are sending more and more low-level and non-violent offenders to prison (who may never have been prone to violent crime), the finding makes sense. Sending a non-violent offender to prison will not necessarily have an effect on violent crime.
How Rising Incarceration’s Effect on Crime Waned
There is no question that some level of incarceration had some positive impact on bringing down crime. There are many habitual offenders and people committing serious, violent crimes who may need to be kept out of society. Criminologists call this the “incapacitation” effect: Removing someone from society prevents them from committing crimes.
But after a certain point, that positive impact ceases. The new people filling prisons do so without bringing down crime much. In other words, rising incarceration rates produce less of an effect on crime reduction. This is what economists call “diminishing returns.” It turns out that the criminal justice system offers a near perfect picture of this phenomenon.
As incarceration doubled from 1990 to today, it became less effective. At its relatively low levels twenty years ago, incarceration may indeed have had some effect on crime. The positive returns may not have yet diminished.
Incarceration rates have now risen so high that further increases in incarceration are ineffective. Due to the war on drugs and the influx of harsher sentencing laws in the 1980s and 1990s, an increasing proportion of the 1.1 million prisoners added since 1990 were imprisoned for low-level or non-violent crimes. Today, almost half of state prisoners are convicted of non-violent crimes. More than half of federal prisoners are serving time for drug offenses. The system is no longer prioritizing arresting, prosecuting, and incarcerating the most dangerous or habitual offenders. In this case, each additional prisoner will, on average, yield less in terms of crime reduction. We have incarcerated those we should not have. This is where the “more incarceration equals less crime” theory busts.
Even those who have argued for the effectiveness of incarceration acknowledge this possibility. University of Chicago economist and Freakonomics co-author Steven Levitt found in his 2004 study that incarceration was responsible for over a third of the 1990s drop in violent crime. He noted that, “Given the wide divergence in the frequency and severity of offending across criminals, sharply declining marginal benefits of incarceration are a possibility,” which, if present, could have affected his findings.
Decrease in Incarceration and Crime
Can the United States safely reduce its incarcerated population? After all, it would be too bad if reducing incarceration yielded a spike in crime.
Fortunately, there is a real-time experiment underway. For many reasons, including straitened budgets and a desire to diminish prison populations, many states have started to cut back on imprisonment. What happened? Interestingly, and encouragingly, crime did not explode. In fact, it dropped. In the last decade, 14 states saw declines in both incarceration and crime. New York reduced imprisonment by 26 percent, while seeing a 28 percent reduction in crime. Imprisonment and crime both decreased by more than 15 percent in California, Maryland, New Jersey, New York, and Texas. Eight states—Connecticut, Delaware, Massachusetts, Michigan, Nevada, North Carolina, South Carolina, and Utah—lowered their imprisonment rates by 2 to 15 percent while seeing more than a 15 percent decrease in crime.
This is all very significant. Incarceration is not just any government policy. Mass incarceration comes at an incredible cost. “A year in prison can cost more than a year at Harvard,” Stiglitz points out. Taxpayers spend $260 billion a year on criminal justice. And there will continue to be less and less to show for it, as more people are incarcerated.
There are significant human costs as well—to individuals, families, communities, and the country. Spending a dollar on prisons is not the same as spending it on public television or the military. Prisons result in an enormous waste of human capital. Instead of so many low-level offenders languishing behind bars, they could be earning wages and contributing to the economy. Incarceration is so concentrated in certain communities that it has disrupted the gender balance and marriage rates. The costs are intergenerational. There are 2.7 million minor children with a parent behind bars. More than 1 in 9 black children have a parent incarcerated.
Research also shows that incarceration can actually increase future crime. Criminologists call this the “criminogenic eﬀect” of prison. It is particularly powerful on low-level oﬀenders. Once individuals enter prison, they are surrounded by other prisoners who have often committed more serious and violent oﬀenses. Prison conditions also breed violent and anti-social behavior. Former prisoners often have trouble finding employment and reintegrating into society due to legal barriers, social stigma, and psychological scarring from prison. Approximately 600,000 prisoners reenter society each year. Those who can find employment earn 40 percent less than their peers, and 60 percent face long-term unemployment. Researchers estimate that the country’s poverty rate would have been more than 20 percent lower between 1980 and 2004 without mass incarceration.
This lack of stability increases the odds that former prisoners will commit new crimes. The more people we put into prison who do not need to be there, the more this criminogenic effect increases. That is another plausible explanation for why our massive levels of incarceration are resulting in less crime control.
Our findings do not exist in a vacuum. A body of empirical research is slowly coalescing around the ineffectiveness of increased incarceration. Last year, the Hamilton Project issued a report calling incarceration a “classic case of diminishing returns,” based on findings from California and Italy. The National Research Council issued a hefty report last year, finding that crime was not the cause of mass incarceration. And, based on a summary of past research, the authors concluded that “the magnitude of the crime reduction [due to increased incarceration] remains highly uncertain and the evidence suggests it was unlikely to have been large.”
We go a few steps further to fully reveal the complex relationship between crime and incarceration. By using thirteen years of more recent data, gathered in the modern era of heavily elevated incarceration, combined with an empirical model that accounts for diminishing returns and controls for other variables, we are able to quantify the sharply declining benefits of overusing prison.
Other Factors Reducing Crime
But if it was not incarceration, then what did cause the crime decline?
There is no shortage of candidates. Every year, it seems, a new study advances a novel explanation. Levitt attributes about half the crime drop to the legalization of abortion. Amherst economist Jessica Reyes attributes about half the violent crime drop to the unleading of gasoline after the Clean Air Act. Berkeley law professor Franklin Zimring credits the police as the central cause. All three theories likely played some role.
Instead of a single, dominant cause, our research points to a vast web of factors, often complex, often interacting, and some unexpected. Of the theories we examined, we found the following factors had some effect on bringing down crime: a growth in income (5 to 10 percent), changes in alcohol consumption (5 to 10 percent), the aging population (0 to 5 percent), and decreased unemployment (0 to 3 percent). Policing also played a role, with increased numbers of police in the 1990s reducing crime (0 to 10 percent) and the introduction of CompStat having an even larger effect (5 to 15 percent).
But none is solely, or even largely, responsible for the crime drop. Unfortunately, we could not fully test a few theories, as the data did not exist at the detailed level we needed for our analysis. For those, we analyzed past research, finding that inflation and consumer confidence (individuals’ belief about the strength of the economy) probably had some effect on crime. The legalization of abortion and unleading of gasoline may also have played some role.
In aggregate, the fourteen factors we identified can explain some of the drop in crime in the 1990s. But even adding all of them together fails to explain the majority of the decrease.
A Sensible Way Forward
No one factor brought down crime. Today, incarceration has become the default option in the fight against crime. But more incarceration is not a silver bullet. It has, in fact, ceased to be effective in reducing crime—and the country is slowly awakening to that reality. Incarceration can be reduced while crime continues to decline. The research shows this and many states are watching it unfold.
Where do we go from here? As President Obama said it in his State of the Union last month, “Surely we can agree that it’s a good thing that for the first time in 40 years, crime and incarceration have come down together, and use that as a starting point for Democrats and Republicans, community leaders and law enforcement, to reform America’s criminal justice system so that it protects and serves all of us.” And indeed, reforming our criminal justice system is emerging as a bipartisan cause. Everyone from Jeb Bush to Hillary Clinton to the Koch Brothers to George Soros has made similar calls.
We should listen to them. There are bold, practical policy solutions starting to gain bipartisan support. Incarceration can be removed as a punishment for many non-violent, non-serious crimes. Violations of technical conditions of parole and probation should not lead to a return trip to prison. Sentence maximum and minimum lengths can be downscaled across the board. There is little reason to jail low-risk defendants who are simply waiting for their trials to begin. And, government funding streams can change to reward reducing incarceration.
Crime is expensive. We do well to fight it. But increasing incarceration is definitely not the answer.