Another century of mass incarceration?
If the U.S. doesn't make reducing the correctional population a policy priority, generations will continue to be burdened by mass incarceration.
by Joshua Aiken and Wendy Sawyer, December 30, 2016
It’s that time of the year again, when many of us like to take stock and commit to changes that will bring us closer to our goals. It’s not a bad time for policymakers to do the same.
Here at the Prison Policy Initiative, we often find it useful to look back at how our criminal justice system has evolved over time. Year-to-year fluctuations are less distracting from that perspective, so we can also identify the pivotal political changes that have had truly dramatic impacts, like the “tough on crime” policies that led to the current era of mass incarceration.
Looking forward, however, is harder when you are data-driven. It’s hard to tell how current policies will play out in the future, and the impact of policy changes is even harder to gauge. We have been warned about the dangers of extrapolating the results of past policy changes to predict the future, but this time of year encourages some imaginative thinking — so we extrapolated recent trends anyway. We hope that as new elected officials are preparing to take office, these rough projections will drive home the urgency of criminal justice reform.
To be taken with an extra-large grain of salt, here are our projections of how long it would take us to undo the aberrant buildup of correctional control in the U.S. since the 1970s, holding constant the current average yearly decline:
- It’ll be 2049 when the federal prison population returns to normal — 33 years from now.
- It’ll be 2122 when the state prison populations return to normal — 106 years from now.
- It’ll be 2084 when the local jail populations return to normal — 68 years from now.
- It’ll be 2098 when the population under probation and parole supervision returns to normal — 82 years from now.
(See how we chose the “normal” baseline in the methodology notes below.)
Policymakers can speed things up: for example, the federal prison population has declined drastically over the last five years, in large part due to policy changes. But, sadly, millions more Americans are going to be churned through jails, locked in prisons, and placed under correctional control unless lawmakers at all levels of government start taking criminal justice reform seriously. If the U.S. doesn’t make reducing the correctional population a policy priority, our children and even great-great-grandchildren are more than likely to be burdened by mass incarceration.
A few notes about methodology:
The late 1970s is an especially useful point of reference because it is right before the “war on crime” really took hold–incarceration rates had been relatively flat for decades and the number of people under correctional control had not yet exploded. In order to determine our baselines we used incarceration rates to ensure that 1975, 1977, and 1980 were representative of historical trends. We used 1975, 1977, and 1980 as our starting points, as those are the years the U.S. first started collecting yearly data on the two leading forms of correctional control: state prison populations (1975) and probation (1977). Before 1983, jail population data was recorded every ten years, so we used the count from 1980 as the closest match to the other baseline years. Federal prison population counts are available back to at least 1925; we used 1975 to match the state prison population count.
For the average yearly decline, we calculated the total decline in population from each correctional population’s peak to 2015, then divided by the number of years since the peak. To project future populations, we started with 2015’s population and subtracted the average decline for each subsequent year. This treats the annual population as an arithmetic sequence; we chose to make it as simple as possible.
These projections are for total counts, not rates, so they do not account for changes in population size. Extrapolating one set of data was dicey enough for us, without adding additional uncertainty about the size of the general population.
Sources:
The data for our projections came from the Bureau of Justice Statistics 2015 updates: see our posts on the incarcerated population and the population under probation and parole supervision.
For the federal prison population, we used 1975 as our baseline (24,131 people) and 2011 (197,050) as our peak value with an average yearly decline since the peak (4,591/year) to arrive at our projection of 2049.
For the state prison population, we used 1975 as our baseline (216,462) and 2009 (1,365,688) as our peak value with an average yearly decline since the peak (11,255/year) to arrive at our projection of 2122.
For the local jail population, we used 1980 as our baseline (163,994) and 2008 (785,533) as our peak value with an average yearly decline since the peak (8,190/year) to arrive at our projection of 2084.
For the population under probation and parole supervision, we used 1977 as our baseline (990,157) and 2007 (5,119,047) as our peak value with an average yearly decline of the probation population since the peak (62,896/year) and an average yearly increase of the parole population since the peak (18,184/year) to arrive at our projection of 2098.
Thinking of the U.S. criminal justice system as one system can be a mistake. There are 50 state systems, plus one in each of the approximately 3,000 counties and often separate systems in the approximately 30,000 municipalities. These differences matter a great deal. 
People in local jails unable to meet bail are concentrated at the lowest ends of the national income distribution, especially in comparison to non-incarcerated people. As this graph shows, 37% had no chance of being able to afford the typical amount of money bail ($10,000) since their annual income is less than the median bail amount. 
Since the 1980s, there has been a significant, nationwide move away from courts allowing non-financial forms of pretrial release (such as release on own recognizance) to money bail, although this does vary substantially depending on jurisdiction. This chart illustrates the possible paths from arrest to pretrial detention. Almost all defendants will have the opportunity to be released pretrial if they meet certain conditions, and only a very small number of defendants will be denied a bail bond, mainly because a court finds that individual to be dangerous or a flight risk. The only national data on pretrial detention that we are aware of comes from the Bureau of Justice Statistics’ 
The District Court locations were grouped by the per capita income of the towns and cities they each serve. Probation rates are higher in court locations that serve populations with lower incomes.
For most cities, murder is much lower now than during the 1980s or 1990s. The data for 1985-2014 comes from the FBI. The FBI has not published the 2015 full year data yet, so we relied on data from the Major Cities Chiefs Association, which historically is very similar if slightly higher than the FBI’s figures. For that reason, this graph may overstate the increase in murder in 2015 in these cities.
Since 1999, private prisons and local jails have housed roughly the same number of those serving state and federal prison sentences. But when those interested in justice reform talk about profiting off of mass incarceration, they almost always leave local jails out of the conversation.
