Steven Raphael is Professor of Public Policy at the University of California, Berkeley. Michael A. Stoll is Professor of Public Policy at the University of California, Los Angeles. They are the authors of Why Are So Many Americans in Prison?, a new book published by the Russell Sage Foundation that analyzes the shocking expansion of America’s prison system and illustrates the pressing need to rethink mass incarceration in this country.

Attorney General Eric Holder’s announcement last month that federal prosecutors will no longer seek stiff mandatory minimum sentences for low-level drug offenders marks a momentous shift in U.S. sentencing practices. After three decades of state and federal policy makers ratcheting up the severity of punishment for felony offenders, this is perhaps the most high profile example of policymakers across the country reevaluating the merits of these unusually harsh sentences. This is a welcome shift in policy that hopefully will be followed by state legislators across the country.

By all measures our national incarceration rate is extremely high. We have the highest incarceration rate in the world and a very high rate relative to our past. Since the 1970s, an era when our incarceration rate did not differ appreciably from those of Western European countries, our incarceration rate has increased nearly five-fold. Why are so many Americans in prison? We have recently completed a book providing a thorough empirical analysis of the last thirty years of U.S. corrections policy and arrive at a very simple conclusion. So many Americans are in prison because we have collectively chosen through our policy choices to put them there. In particular, sentencing policy changes emphasizing mandatory minimum sentences, state laws requiring that prison inmates serve minimum fractions of their sentences, and other such policy changes have driven the explosive growth in the U.S. prison population. Very little growth can be explained by changes in crime, as crime rates are currently at historic lows. On the other hand, nearly all incarceration growth can be explained by a higher likelihood that those apprehended for a crime are sentenced to prison, as well as longer sentences for those we incarcerate.

Politicians have escalated “get tough on crime” policies because they have led the public to believe that more incarceration means less crime. There is a growing body of research by criminologists, economists, sociologists and others on the effectiveness of prison in reducing crime. In theory the use of prison reduces crime through incapacitating the criminally active and deterring potential offenders. It is the case that on average, prison does incapacitate and lowers crime as a result. However, the degree to which it serves this purposes varies considerably from inmate to inmate. Moreover, as our incarceration rate has increased to very high levels, the marginal benefits of incarceration have declined drastically, as we are increasingly incarcerating less serious offenders. In other words, when we incarcerate less serious offenders or offenders well into their old age when they are much less likely to reoffend, the bang-per-buck spent on prisons in terms of crime control is quite low. This is precisely what has happened in the U.S.

Moreover, there are other arrows in our crime-combatting quiver that yield better returns on the dollar. For example, innovative efforts in Hawaii to monitor drug use among probationers with swift yet modest sanctions for violations has greatly reduced drug abuse, absconding from probation officers, and subsequent prison admissions. A growing body of research demonstrates substantial crime-deterrent effects associated with expanding police staffing levels. There is also ample evidence showing that preventing high risk youth from dropping out of school substantially lowers the risk of incarceration as an adult. In the presence of alternative and more cost-effective strategies for crime control, we could redeploy our crime-fighting resources and efforts in a manner that reserves prison for those who pose the greatest risk to society and employs other policy levers to deter less serious offenders and prevent offending in the first place.

To be sure, the changes announced by Attorney General Holder will make only a modest reduction in the U.S. prison population. Federal prisons hold only a small share of U.S. prison inmates (roughly 14 percent). However, the change could have potentially large effects on the size of the federal prison population. In our book, we estimate that returning federal drug sentencing practices to those of the early 1980s would eliminate almost half of the growth in the federal prison population that have occurred since.

Comparable changes in drug sentencing would have more modest effects on state prison populations. We estimate that returning state drug sentencing practices to those of years past would eliminate only one-fifth of state incarceration growth since the early 1980s. This is due largely to the fact that much of the growth in state prison populations (roughly half) is driven by tougher sentences for those who commit violent crime.

Despite these likely modest impacts, the Attorney General’s announced change provides an important example of how sentencing practices can be made more rational, efficient, and fair. Hopefully this is a bellwether event that will usher in a new era of smarter use of prisons.