Connect with the author
Crime rates in the United States today are about what they were in the late 1960s – yet the numbers of Americans sitting in jails and prisons are at an unprecedented high. Over the past 35 years the penal population has increased five-fold. The total inmate population now tops 2.3 million, about one percent of the adult population. The United States incarcerates a higher fraction of its population than any other country in the world, including Russia and South Africa.
Young black men are the ones most likely to be imprisoned, especially those without high school diplomas. On any given day in 2009, one in nine black men resided in jail or prison – including a remarkable 37 percent of young black male high school dropouts. By 2009, more than two-thirds of such men born between 1975 and 1979 had spent at least a year in prison. Researchers have been paying increased attention to the impact of high rates of black male imprisonment on families and communities. But a key issue has been overlooked – the omission of prisoners from official census counts and social statistics, which undermines the accuracy of our efforts to track the wellbeing of various groups of Americans.
The Prisoners Not Counted
Much of the data used to track social trends comes from recurrent surveys of people who live in regular households. Many federal surveys – such as the authoritative Current Population Survey taken each month – do not include individuals who are living in institutions, such as prisons and jails, military barracks, or long-term care facilities for the mentally ill. Such people are just left out. But this distorts the data – especially given the burgeoning growth of the prison and jail population over the past several decades, combined with the skew of imprisonment toward young black men without a high school diploma.
What difference does it make to our picture of the United States that government statistics track only people living in households? To find out, I have compared social indicators that do, and do not, include jail and prison inmates.
- In 2008, data from the Current Population Survey placed the high school dropout rate of young black men at 13.5 percent, but with prisoners included the dropout rate was actually 19 percent.
- Regular population data suggest that the gap between black and white high school graduation has declined over the past few decades, but with imprisoned black men included, the black dropout rate is 40 percent higher and the racial gap in high school completion has not narrowed since the early 1990s. The racial gap has remained about 11 percentage points for the last two decades.
- In 2008, 19 percent of young black men did not finish high school and black male dropouts were more likely to be in prison or jail than to be employed. Wages of young black men have seen little improvement, relative to the wages of whites, since the mid-1980s. The economic boom of the 1990s did not trickle down to young black male dropouts, as some scholars once thought. It is also highly likely that these men have similarly suffered in the recent economic downturn. We cannot tell for sure because black male dropouts are increasingly in prison or jail and are left out of official statistics.
Voting Trends are Also Misunderstood
Voter measures are also skewed by leaving out prisoners. Young African Americans were said to have turned out in record numbers to support Barack Obama for president in 2008 – and theorists have spun many hypotheses to explain why, according to available data, poorly educated blacks are turning out to vote at higher rates than their socioeconomic status would imply. But the idea of growing democratic engagement among young black men is an illusion, merely an artifact of surveys that are limited to individuals living in households.
- When incarceration rates are accounted for, only 20.4 percent of young black male dropouts voted in the 2008 election – nearly identical to the 20.7 percent that turned out to vote in the 1980 election that pitted Ronald Reagan against Jimmy Carter.
- Adjusting turnout rates to include inmates suggests that the primary explanation for unexpectedly high turnout rates among African American male dropouts is a phenomenon called “sample selection.” Excluding inmates from calculations of voter turnout removes the most unlikely voters, which artificially boosts the estimated turnout rate.
Invisibility is Not Inevitable
The invisibility of specific segments of the American population – and the overall inequalities we fail to correctly perceive when young black male dropouts are left out in household surveys – are not inevitable byproducts of prison growth. Invisibility is due to our methods for collecting official data, compounded by the failure of scholars and analysts, until now, to notice and correct for the people missed in official surveys.
The collection of official U.S. social information is grounded in the Constitutional mandate to enumerate the population every ten years for the purposes of congressional apportionment. Over the last half century, moreover, governments have relied on indicators about levels of education, employment, and income to inform policymaking and allocate federal money to states and localities. To do its job, social science also needs high quality, regularly collected data.
Unfortunately, counting only individuals living in households, while omitting people residing in institutions, has systematically skewed official social data – particularly in an era of skyrocketing incarceration that disproportionately removes young, less educated black men from households. By relying on skewed indicators, we have developed a statistical portrait of the early twentyfirst-century United States that overstates the progress of African Americans. To do valid social science and produce well-informed public policies, we must develop improved indicators more representative of the entire American population, no matter where people reside.