Skip to main content

The Changing Racial Dynamics of Women’s Incarceration

February 27, 2013
This report highlights factors contributing to the dramatic shift in the racial composition of women’s prison population between 2000 to 2009, and provides recommendations to address racial disparities in the criminal justice system.

In the first decade of the 21st century the United States began to experience a shift in the 30-year buildup to a world record prison system.  Although the decade ended with an increased number of people in prison, the rate of growth overall was considerably below that of previous decades and since 2008 the overall number of people in state prisons has declined slightly each year.

Scholars are beginning to analyze the relative contributions of changes in crime rates, criminal justice policies, economics, and demographics to the slowing growth rate of the prison system, but one area that has gone largely unexplored is the impact of such changes on racial disparities in imprisonment. As is well known, black/white disparities in the use of incarceration have been profound for quite some time. Since the 1980s a series of analyses have documented these trends at the national level as well as examining variation in disparity among the states.1)See for example: Blumstein, A. (1993). Racial disproportionality of U.S. Prison populations revisited. University of Colorado Law Review. 64; Mauer, M. & King, R.S. (2007). Uneven justice: State rates of incarceration by race and ethnicity. Washington, DC: The Sentencing Project; Tonry, M.H. (2011). Punishing race: A continuing American dilemma. New York: Oxford University Press.

As prison populations fluctuate, though, the relative rate of incarceration among racial groups may or may not reflect prevailing patterns. Further, as the prospect of a declining prison population has now become a distinct possibility for the next decade, it will become increasingly important to monitor whether reduced incarceration is experienced in similar ways across racial/ethnic groups.

This report first describes trends in incarceration for the first decade of the century, and contrasts this with patterns of the previous decade. We then assess the extent of change in the race and gender dynamics of incarceration over the past decade, and suggest factors which may be contributing to these trends. The data in this report document the following key findings:

  • Racial/ethnic disparities in U.S. incarceration remain substantial – In 2009,2)National data in this report are for 2010, but data for the analysis of prison populations by race and ethnicity are for 2000 and 2009. This is because the Bureau of Justice Statistics reports that “the data source used to estimate race and Hispanic origin changed in 2010.” African Americans and Latinos constituted more than 60% of imprisoned offenders. African American males were incarcerated in state and federal prisons at 6.4 times the rate of non-Hispanic white males, and Hispanic males at 2.4 times the rate of non-Hispanic whites.3)Guerino, P., Harrison, P.M. & Sabol, W.J. (2012). Prisoners in 2010 (NCJ 236096). Washington, DC: Bureau of Justice Statistics. Data on race/ethnicity and gender from this publication should be interpreted as general estimates of the race/ethnicity and gender proportions within the prison population. BJS is currently employing a revised methodology for estimating race/ethnicity for these and future years.
  • Declining rate of incarceration for African Americans – From 2000 to 2009 the rate of incarceration in state and federal prisons declined 9.8% for black men and 30.7% for black women.
  • Rates of incarceration for whites and Latinos generally rising – Incarceration rates for white men and women rose 8.5% and 47.1%, respectively from 2000 to 2009. For Hispanics the men’s rate declined by 2.2% while the women’s rate rose by 23.3%.
  • Dramatic shift in racial disparities among women – In 2000 black women were incarcerated in state and federal prisons at six times the rate of white women. By 2009 that ratio had declined by 53%, to 2.8:1. This shift was a result of both declining incarceration of African American women and rising incarceration of white women. The disparity between Hispanic and non- Hispanic white women declined by 16.7% during this period.

Similar trends can be seen among men, but at a lesser scale, with a decline of 16.9% in the black/white incarceration ratio over the decade. The disparity between Hispanic and non-Hispanic white men declined by 11.1%.

Slowing Growth in Incarceration

Beginning in the early 1970s the United States embarked on an unprecedented escalation of the use of imprisonment. The 2.2 million people behind bars in prisons and jails in 2010 represented more than a 500% increase from the level of 1972.4)The Sentencing Project. Trends in U.S. Corrections. During the highest growth years in the 1980s, state prison populations increased by as much as 12% in a single year.

Nationally, the rate of prison growth of earlier decades slowed considerably in the first decade of this century. The total number of individuals in state prisons rose by an average of 1.1% per year and in federal prisons by an average of 3.3% between 2000 and 2010,5)Carson, E.A. & Sabol, W.J. (2012). Prisoners in 2011 (NCJ 239808). Washington, DC: Bureau of Justice Statistics compared to 5.6% and 8.6% respectively in the 1990s.6)Beck, A.J. & Harrison, P.M. (2001). Prisoners in 2000 (NCJ 188207). Washington, DC: Bureau of Justice Statistics. Prisoner counts in this report reflect persons under federal or state jurisdiction.

The substantial growth of the 1990s is best seen in states such as Texas, in which the number of persons under state prison jurisdiction tripled, as well as in ten states in which the population more than doubled (Appendix Table 1). Overall, the median change in the number of people in state prison systems was 71% for the 1990s. In contrast, from 2000 to 2010 the median increase declined to 21%, with only three states experiencing more than a 50% rise in their populations. Further, five states – Delaware, Maryland, Michigan, New Jersey, and New York – had fewer people in their prisons at the end of the decade and 47 states experienced at least one year of decline.

While reduced crime rates are likely to have contributed to these declines, deliberate changes in policy and practice are also playing a role in these reductions in some states.7)Greene, J. & Mauer, M. (2010). Downscaling prisons: Lessons from four states. Washington, DC: Justice Strategies and The Sentencing Project.  These include initiatives such as treatment diversion programs, sentence reduction incentives for participation in prison programming, enhanced reentry support, and reduced technical violations of parole.

Of particular note are trends of the most recent years in which prison populations at the state level have essentially stabilized after nearly four decades of increase. As a result of the fiscal crisis and the requirement in almost all states that the government maintain a balanced budget, governors of both major parties have been carefully scrutinizing corrections policies and population options. This is often being done with a goal of achieving short-term population reductions and consequent cost savings.8)Mauer, M. (2010). Sentencing reform amid mass incarcerations: Guarded optimism. Criminal Justice, 26. American Bar Association.

In a number of states, for example, executive orders have moved up consideration of parole release by 60 to 90 days, thereby achieving short-term population reductions. Other states have enacted changes to parole revocation procedures, with a goal of reducing the scale of technical violations to prison. In states such as Kansas and Michigan, this has resulted in substantial reductions in some years, as parole systems developed graduated sanctions for responding to violations and greater oversight of decision making.

In states in which substantial reductions have been achieved there has also been a growing trend of prison closings, the major means by which significant cost savings may be achieved. During 2011 and 2012, 17 states either closed or considered closing prison facilities, with a total capacity reduction of more than 28,000 beds.9)Porter, N. (2012). On the chopping block: State prison closings. Washington, DC: The Sentencing Project.

While this trend of population reduction has drawn much attention, it is important to note that the decline in state prison populations in the past several years is disproportionately influenced by reductions in the prison population in California. Between 2009 and 2010, for example, the reduction in California constituted 57% of the total state prison decline of nearly 11,000 inmates nationally.10)Guerino, P., Harrison, P.M. & Sabol, W.J. (2012). Prisoners in 2010 (NCJ 236096). Washington, DC: Bureau of Justice Statistics. Data on race/ethnicity and gender from this publication should be interpreted as general estimates of the race/ethnicity and gender proportions within the prison population. BJS is currently employing a revised methodology for estimating race/ethnicity for these and future years.

California is likely to be increasingly influential in the national picture as a result of recent developments. In the U.S. Supreme Court’s Brown v. Plata decision in 2011, the court found that overcrowding in the state’s prison system produced unconstitutional conditions of health care. Its ruling required the state to reduce its prison population by more than 30,000 persons over two years. In response, the state adopted “realignment” legislation which took effect in October 2011, and is largely directed at having lower-level offenders serve their sentences either in local jails or under local probation supervision. Consequently, it is expected that there will be continued reductions in the state prison system over the next several years.

Race and Gender Disparity in Incarceration

As prison populations appear to be stabilizing nationally, and some states are reducing their populations, it is instructive to explore whether these changes are reflected in changes in disparities in imprisonment. We begin by looking at overall changes in imprisonment by race and gender in Figures 1 and 2 below. Overall, the rate of increase of women continued to outpace that of men, as it has for several decades. From 2000 to 2009 the number of women incarcerated in state or federal prisons rose by 21.6%, compared to a 15.6% increase for men.

Looking at the racial/ethnic composition of prison population changes, we see that the direction of change is similar for both men and women, but much more pronounced among women. Overall, the numbers of African Americans in prison declined during this period, while whites and Hispanics experienced increases. Substantial changes for 2000-2009 can be seen in the rise of 48.4% in the number of white women incarcerated and the decline of 24.6% among African American women. Although the number of Hispanic women rose by 75% during this period, the increase among Hispanics in the general population resulted in a rate increase below that of white women, as will be seen below.

Figure 1. Number of male inmates, by race, 2000-2009
Source: BJS Prisoners in 2010
Figure 2. Number of female inmates, by race, 2000-2009
Source: BJS Prisoners in 2010

Rates of incarceration vary notably by state. Imprisonment rates in 2010, for example, ranged from a high of 867 per 100,000 people in the population in Louisiana to a low of 148 per 100,000 in Maine. It is also the case that racial disparities in incarceration vary broadly. A study of incarcerated populations in 2005, for example, found a range of black/white disparity from a high of 13.6 to 1 to a low of 1.9 to 1.11)Mauer, M. & King, R.S. (2007). Uneven justice: State rates of incarceration by race and ethnicity. Washington, DC: The Sentencing Project.

At the national level, the first decade of the century produced substantial changes in the racial disparity in incarceration. Although African Americans were still incarcerated at dramatically higher rates than whites at the end of the decade, the degree of that differential changed noticeably (Table 1). Between 2000 and 2009, the incarceration rates for both black men and women declined, while rising for both white men and women. Hispanic men’s rates declined modestly, while Hispanic women’s rates rose substantially.

Changes during the decade were most pronounced among women, with black women experiencing a decline of 30.7% in their rate of incarceration, white women a 47.1% rise, and Hispanic women a 23.3% rise. In the following sections we explore possible explanations for these changes for women in particular.

Table 1. Change in incarceration rates (state and federal prison) by race, ethnicity, and gender, 2000 to 200912)Beginning in 2004, the Bureau of Justice Statistics changed its racial categorizations, adding a new category of “two or more races.” That year, 2.9% of respondents identified with that description. It is likely that many of these individuals previously would have been characterized as black, and thus, this would account for a portion of the reduction in the black prison population that year. For the period of 2000-2009 as a whole, though, this change does not substantially affect the analysis described in this paper. For example, incarceration rates for both black men and women continued to decline in the two years following that change, as well as for most, though not all, years in this period.
Group Incarceration rate
(per 100,000)
% change in rate Ratio black/Hispanic to
white rate
% change in ratio
2000 2009 2000 2009
Black men 3,457 -3,119 – 9.8%  7.7:1  6.4:1  -16.9%
White men 449 487 + 8.5%
Hispanic men 1,220 1,193 -2.2%  2.7:1 2.4:1  -11.1%
Black women 205 142 – 30.7% 6.0:1 2.8:1  – 53.3%
White women 34 50 + 47.1%
White women 60 74 +23.3%  1.8:1  1.5:1  -16.7%
Source: BJS Prisoners in 2010

Changing Racial Composition of Women’s Incarceration

Women have long represented a modest share of the overall prison population. In 1980, about 13,000 women were incarcerated in federal and state prisons combined representing 4% of the total prison population.13)Greenfield, L.A. & Minor-Harper, S. (1991). Women in prison. Washington, DC: Bureau of Justice Statistics.  Since that time, the rate of growth of women in prison has exceeded the rate of increase for men, rising 646% from 1980 to 2010, compared to a 419% increase for men. As a result, in 2010 there were 112,000 women in state and federal prison and 205,000 women overall in prison or jail; women now constitute 7% of the prison population.14)Guerino, P., Harrison, P.M. & Sabol, W.J. (2012). Prisoners in 2010 (NCJ 236096). Washington, DC: Bureau of Justice Statistics. Data on race/ethnicity and gender from this publication should be interpreted as general estimates of the race/ethnicity and gender proportions within the prison population. BJS is currently employing a revised methodology for estimating race/ethnicity for these and future years.

As was the case with men, a substantial portion of the overall increase in women in prison was produced by “get tough” initiatives focused on harsher sentencing policies and lengthening time served in prison. Women were particularly affected by the policies of the “war on drugs.” Since women have always represented a small share of persons committing violent crimes, their numbers in prison would not have grown as dramatically had it not been for changes in drug enforcement policies and practices. As law enforcement increased targeting of drug law violators and as sentences for drug offenses became more severe, drug offenders came to represent a rapidly growing share of the incarcerated population, with the proportion of women in prison for drug crimes exceeding that of men. In 1986, 12% of women in state prison were serving time for a drug offense compared to 8% of men. Over time, these proportions increased, and as of 2009, 25.7% of women in prison were serving time for drug offenses, as were 17.2% of men.

In addition, the advent of mandatory sentencing policies for many drug offenses at times imposed a particularly harsh burden on women offenders, with one aspect of this sometimes described as the “girlfriend” problem. That is, since the only means of avoiding a mandatory penalty is generally to cooperate with the prosecution by providing information on higher-ups in the drug trade, women who have a partner who is a drug seller may be aiding that seller, but have relatively little information to trade in exchange for a more lenient sentence. In contrast, the “boyfriend” drug seller is likely to be in a better position to offer information, and so may receive less prison time for his offense than does the less culpable woman.

In addition to the factors influencing the overall rise of women in prison, by the year 2000 a variety of other factors had coalesced to produce severe racial/ethnic disparities in women’s incarceration rates. These factors included differential involvement in certain crimes; law enforcement practices, particularly related to the drug war, targeting black neighborhoods; and, more limited access to treatment and alternatives to incarceration for low-income women. Thus, in 2000 black women were incarcerated at a rate six times that of white women, and Hispanic women at a rate 1.8 times that of white women.15)Guerino, P., Harrison, P.M. & Sabol, W.J. (2012). Prisoners in 2010 (NCJ 236096). Washington, DC: Bureau of Justice Statistics. Data on race/ethnicity and gender from this publication should be interpreted as general estimates of the race/ethnicity and gender proportions within the prison population. BJS is currently employing a revised methodology for estimating race/ethnicity for these and future years.  Over the next decade, as previously noted, these dynamics changed in fairly dramatic fashion.

Analyzing Changes in Racial Disparity Among Women in Prison

With the exception of the relatively modest number of women held in federal prisons (about 12,000 in 2010), changes in women’s national incarceration rates represent the combined effects of changes in population, crime, and criminal justice policies and practices among the 50 states. There is not necessarily a single explanation for the racial/ethnic differences in incarceration rate changes. Rather, varying combinations of circumstances relating to involvement in crime, criminal justice responses, and other factors likely affected white, black, and Hispanic women differently, with those factors varying among states as well. Assessing the relative contribution of these factors is complicated by the fact that many key data sources track changes by either race/ethnicity or gender, but not both. Nevertheless, available data provide a starting point for examining sources of changes in the past decade.

Changes in Offending

One possible explanation for the relative distinction in changing black and white rates of women’s incarceration is that black women may have become less likely to be involved in criminal activity and white women more so. Aside from studies of self-reported involvement in crime, which are not done on a consistent basis over time, the best proxy for criminal behavior is arrest data compiled by the FBI and published in its annual Uniform Crime Reports (UCR). These data have two limitations for the current purpose. First, the ethnicity of arrestees is not documented, so all Hispanic arrestees are categorized only by race, with the vast majority classified as white. This contrasts with the incarceration data used in this report, in which the Bureau of Justice Statistics treats white, black, and Hispanic as mutually exclusive categories. Second, UCR data are available by race or gender, but not both combined.

Changes in total arrest rates show considerable differences by race for the period 2000 – 2009 (Table 2). First, the decline in the rate of arrest for violent crimes for blacks was twice that of whites (including a share of Hispanics in both categories), which is particularly noteworthy given that this category of crimes is most likely to result in a prison term. Second, black arrest rates declined in all three offense categories (violent crimes, property crimes, and drug offenses), while white arrest rates declined for violent crimes, but rose for both property and drug offenses (modestly). Overall, changes in involvement in crime appear to be contributing to changes in incarceration rates of African Americans overall and it is likely, though not certain, that this is true for black women as well.

Table 2. Percent change in rates of arrest by race, 2000-2009
Race Violent Property Drugs
White -11.1% 16.6% 2.2%
Black -22.2% -11.5% -11.8%
Source: FBI Uniform Crime Reports

Prison Populations by Offense

Another means of gleaning insight into the influence of changes in offending on women’s incarceration rates is to look at changes over time in the offenses for which women in prison are serving time. We look first at national estimates broken out by gender and race/ethnicity separately (Table 3), and then at data specific to white, black, and Hispanic females from a subset of states.

Table 3. Percent change in most serious offense of sentenced state prisoners, 2000-2009.
Offense Total Male Female White Black Hispanic
Violent 23.0% 22.2% 40.0% 24.7% 16.6% 34.9%
Property 9.5% 6.5% 44.3% 21.3% -8.8% 21.1%
Drug -3.5% -3.6% -2.8% 26.6% -15.8% -4.6%
Sources: BJS Prisoners in 2001,16)Harrison, P.M. & Beck, A.J. (2002). Prisoners in 2001 (NCJ 195189). Washington, DC: Bureau of Justice Statistics. Prisoners in 2010

Between 2000 and 2009 the proportional increases in women’s incarceration for violent and property offenses were substantially greater than for men. Given that the overall number of African Americans incarcerated for a property offense declined and that the black increase in violent offenses was substantially less than for whites and Latinos, this suggests that the substantial increase in violent and property offense incarceration for women was likely influenced primarily by changes among whites and Hispanics. Conversely, the modest decline in women’s drug offense incarceration was likely influenced by the substantial overall decline among African Americans, particularly in contrast to the overall white increase.

This interpretation is further supported by data from a subset of 12 states for which data on the most serious offenses committed are available for black and white women.17)The states represented by these data are California, Florida, Georgia, Missouri, Mississippi, New Jersey, New York, Oklahoma, South Carolina, Tennessee, Texas, and Washington. They are ones for which the Bureau of Justice Statistics has data by gender and race/ethnicity for the years of 2000 and 2009. These data are not reported by BJS in their Prisoners series because the methods individual states use to categorize individuals by race/ethnicity are inconsistent across states and may be based on administrative data rather than self-reported data. To overcome this limitation, BJS derives and reports only national estimates that have been adjusted to account for bias in the classification of race/ethnicity in administrative data. Therefore, these figures should be viewed as providing a general overview of the racial/ethnic composition of these prison populations, rather than a precise accounting. While these states are reasonably diverse geographically and the mix of offenses for which women are incarcerated is close to the national proportions, we cannot say to what extent their prison populations are entirely reflective of national trends.

As seen in Table 4 below, the decline of 1,813 black women in these 12 states is overwhelmingly a function of reduced drug incarceration (103.3% of the total change), with changes in the number incarcerated for violent offenses (13.3% increase) and property offenses (10% decline) accounting for more modest amounts of the total difference. For white women, almost half of the increase in incarceration is a function of an increased number of property offenders, with the remainder of the increase divided between incarceration for violent and drug offenses.

Table 4. Change in most serious offense of female inmates by race in twelve states, 2000-2009
Race Offense 2000 2009 Difference % of Total Difference
White Violent 5,165 6,992 1,827 27.4%
Property 4,571 7,588 3,017 45.2%
Drug 4,866 6,697 1,831 27.4%
TOTAL 14,602 21,277 6,675  
Black Violent 6,332 6,574 242 13.3%
Property 4,570 4,388 -182 -10.0%
Drug 6,387 4,514 -1,873 -103.3%
TOTAL 17,289 15,476 -1,813  

In addition to drug offenses, the racial dynamics of women’s incarceration for property offenses is also quite distinct for white and black women and, again, parallels the previously reported changes in arrest rates. Note here, though, that some proportion of property crime is “drug-related”; for example, persons committing theft in order to obtain money to buy drugs. So, any overall changes in rates of drug use by race may influence involvement in property crime. It is also possible that changes in relative rates of property crime commission may be related to the differential impact of the economic changes of the past decade, by race and/or geography.

Some observers have suggested that changes in women’s incarceration by race are driven by increased arrests for prescription drug offenses or changes in the drug of choice among those who engage in such practices. Because of the rise in the use of methamphetamine, a drug disproportionately used by whites and Latinos, in certain regions of the country white women may now be more likely to come under criminal justice supervision. At a national level, there is no breakdown of drug offenders in prison by the type of drug that they used or sold. But there is some modest, and mixed, evidence on this at the state level. For example, the Minnesota Department of Corrections reported that nearly a third of the increase of 3,000 prisoners (male and female combined) from 2000-2007 was due to a rise in the number of methamphetamine offenders, although there has been a decline since then.18)Cook, M. (2013). Minnesota’s prison population remains among lowest in U.S. Twin Cities Daily Planet. January 23.  In Iowa, though, state officials report that legislation restricting the sale of pseudoephedrine contributed to a decline in meth labs and the number of prison admissions for related offenses dropping by half from 2005 to 2008.19)Join Together. (2009). Iowa Meth Legislation Leads to Decline in Prison Population. January 27.

A recent federal case demonstrates how stepped-up methamphetamine enforcement combined with harsh sentencing policies can contribute to escalating prison populations for these offenses. Lori Newhouse, a 32-year old Iowa woman with a lengthy history of substance abuse, was convicted of manufacturing or attempting to manufacture 5 grams or more of pure methamphetamine in 2012. Federal District Court Judge Mark Bennett described her as a “pill smurfer” who purchased legal cold remedies that she supplied to methamphetamine producers in exchange for the drug. Under federal mandatory minimum penalties she was subject to a 10-year sentence, but based on two previous drug convictions in 2002 her sentencing guideline range would be enhanced to 21-27 years due to federal Career Offender sentencing provisions. The prior convictions arose out of a police raid of her motel room, following which she pled guilty to two separate charges of possession with intent to deliver methamphetamine and psilocybin mushrooms. Had the prior cases been considered as only one offense, Newhouse would not have been subject to the enhanced penalties. Judge Bennett rejected the Career Offender guidelines based on a policy disagreement and taking into account Newhouse’s substantial assistance to prosecutors, imposed the lowest sentence available to him of 8 years in prison, which he described as “still exceptionally long.”20)U.S. v. Newhouse, No. CR11-3030-MWB (N.D. Iowa Jan. 30, 2013).

Even to the extent that methamphetamine or prescription drug offenses may be contributing to rising numbers of white women in prison, though, this does not provide insight into why fewer black females are now incarcerated for a drug offense. In this regard, overall changes in drug sentencing policy in some states may provide much of the explanation. Because the rate of incarceration of black women in some states has been so heavily influenced by drug sentencing in the past, any amelioration of such policies is like to disproportionately benefit African American women.

In New York, for example, which has led the nation in its prison population decline, the racial dynamics of drug incarceration and decline have been quite stark. First, the decline of 1,002 women incarcerated for a drug offense from 2000 to 2009 explains virtually the entire decline of women in prison during that period (with an increase of property offenders and a slight decrease of violent offenders).21)Data provided by New York State Department of Corrections and Community Supervision, February 13, 2013.  In addition, the entire (99.4%) female drug offender decline was comprised of African American and Hispanic women, representing a decline of about 70% for each group. Thus, with such dramatic racial/ethnic disparities in drug incarceration in the state, it was virtually inevitable that a substantial reduction in overall drug offense incarceration would result in fewer women of color in prison.

Changing Socioeconomics

While exploring criminal justice data provides an indication of the proximate causes of the changing racial dynamics of women’s incarceration, recent research on life expectancy suggests that explanations for these changes may reflect broader structural socioeconomic shifts. This analysis may be particularly relevant for understanding the rising number of white females in prison.

Research by the MacArthur Foundation Research Network on an Aging Society sheds light on the situation of less educated white women in particular.22)Olshansky, S.J., et al. (2012). Differences in life expectancy due to race and educational differences are widening, and many may not catch up. Health Affairs, 31, 1803-1813. The Research Network’s analysis explored changes in life expectancy at birth between 1990 and 2008 for racial/ethnic/gender groups by level of education. The findings show that women still have greater life expectancy than men and that white males and females outlive black males and females at virtually every age range. Although Hispanics are found to have the highest life expectancy of any group at birth, the researchers caution that the relative mix of the immigrant and US-born populations exerts a strong effect on these figures, and is likely to change in coming years as this mix changes as well.

The influence of educational attainment on life expectancy suggests the most relevant effects for the purposes of the incarceration changes we have documented. As the authors note, “Education and its socioeconomic status correlates of income and wealth have powerful associations with duration of life for both sexes and races, at all ages.” Changes in this regard have been most profound among whites, and especially women, whose life expectancy at birth declined by more than five years from 1990 to 2008 for those with less than a high school education. As a result, despite advances in health care, this group of white women now has a life expectancy at birth that is only equivalent to that of all women in the country born in 1964. In contrast, black women with similar education levels experienced modestly increased longevity during this period, which now slightly outpaces that of white women in this group.

The effect of socioeconomic disadvantage on life prospects suggests possible connections with criminal justice involvement as well. Many of the socioeconomic factors that contribute to lower life expectancy are also associated with high rates of criminal behavior and incarceration. These include unhealthy lifestyles (especially substance abuse and tobacco use), limited access to healthcare, limited ability to cope with stress or manage chronic illness, poverty, and high levels of unemployment. As the authors of the study note, education provides beneficial effects “through the adoption of healthier lifestyles, better ability to cope with stress, and more effective management of chronic diseases,” as well as increasing access to better-paying jobs and higher incomes. Other data demonstrate that during this period the share of working-age adults without a high school diploma who lacked health care coverage rose from 35% in 1993 to 43% in 2006,23)Tavernise, S. (2012). Reversing trend, life span shrinks for some whites. The New York Times. September 21. and that smoking rates for women (both black and white) who have not completed high school have been rising, even as they have declined for men at this educational level. Thus, the cumulative social disadvantage experienced by certain groups, and for less-educated white women in particular, may contribute to greater likelihood of involvement in substance abuse and crime.

Although the proportion of the national population without a high school degree declined from 22% to 12% during this period, it may be that for many of those in this category, the disadvantages of this status are proving increasingly profound. The authors argue that “it is now reasonable to conclude that at least two Americas have formed, with notably different longevity prospects.”


The trends documented in this report are complex. After decades of steadily rising incarceration of African Americans, we have now seen a decade during which these rates have declined modestly for men and substantially for women. While these developments should not be taken to suggest that the era of mass incarceration of African Americans has ended by any means, it is nonetheless significant that there have been changes in this regard.

At the same time, we note that despite several years of modest decline in state prison populations, the overall rate of incarceration in the United States remains at five times the rate that prevailed in 1970. In addition, the declining ratio of incarceration between African American and white women is a function of two trends: fewer black women in prison, but growing numbers of white women incarcerated. Thus, the nearly 30-year trend of women’s incarceration increases outpacing that of men has not abated; rather, the racial dynamics of those changes have shifted.

This report has explored the factors which may be contributing to these trends, but is clearly just a first step in such an analysis. Changes in the number of African American women who are incarcerated could be a function of changing involvement in crime, changes in law enforcement or sentencing practices, socioeconomics, or some combination of these and other factors. Our examination of the data suggests that it is likely that reduced numbers of drug incarcerations explains a significant portion of the trend.

Conversely, the rising number of incarcerated white women may have several causes. As noted, increased prosecutions of methamphetamine offenses may be a contributing factor, but broader socioeconomic trends contributing to declining life expectancy for low-income white women suggest possible underlying conditions that may contribute to changes in offending rates as well.


In order to better understand these dynamics and to address racial disparities in the use of incarceration, policymakers should pursue the following options:

  • Conduct state-based analyses of changes in disparity – Research over the course of the past several decades has documented both substantial racial disparity in incarceration and broad variation in this regard among the states. Given the trends that we have documented in this report, it would be useful for policymakers in all states to conduct analyses of these trends and to identify causal factors of either rising or declining incarceration by race and gender. Similar analyses focused on analyzing state-specific racial disparities (though not for women specifically) have been conducted in recent years by scholars and state-appointed commissions in Minnesota24)Frase, R. (2012). What factors explain persistent racial disparities in Minnesota prison populations? Paper presented at the annual meeting of the American Society of Criminology 2012. and Wisconsin.25)Commission on Reducing Racial Disparities in the Wisconsin Justice System. (2008). Final report.
  • Establish statewide racial disparity task forces – The state of Connecticut in 2000 established the Commission on Racial and Ethnic Disparity in the Criminal Justice System, charged with developing recommendations to eliminate such disparities. In Illinois, the state legislature established the Illinois Disproportionate Justice Impact Study Commission in 2009 and charged the group with analyzing arrest and criminal processing data “in order to guide the Commission in recommending systemic changes that could mitigate or eliminate racial disproportionality in sentencing for drug crimes.”26)Lyons, T., Lurigio, A., Roque, L., & Rodriguez, P. (2013). Racial disproportionality in the criminal justice system for drug offenses: a state legislative response to the problem. Race and Justice. 3:83. Such task forces have the potential for focusing public and policymaker attention on the sources of the problem, along with changes in policy and practice that can ameliorate these outcomes.
  • Adopt proactive racial disparity assessments – Similar to fiscal impact statements, the goal of projecting the racial effects of proposed sentencing policies is to anticipate any unwarranted changes in racial disparity that might be produced through legislative change, and to provide lawmakers with the opportunity to meet the goals of public safety without exacerbating existing disparities. In 2008, the states of Connecticut and Iowa adopted legislation to establish a process of producing racial impact statements prior to adoption of new sentencing legislation, and in Minnesota, the Sentencing Guidelines Commission employs such a process internally.27)Mauer, M. (2009). Racial impact statements: Changing policies to address disparities. Criminal Justice. 23: 4. American Bar Association.
  • Provide technical assistance to aid jurisdictions in addressing disparities – Racial/ethnic disparities in incarceration result from a combination of factors, which include the use of discretion and allocation of resources by criminal justice practitioners. In recent years a number of projects have attempted to engage practitioners in analyzing and responding to disparities in local law enforcement or court processing. These have included such efforts as the Racial Justice Improvement Project of the American Bar Association28)Jones, C. (2012). Confronting race in the criminal justice system. Criminal Justice. 27: 2. American Bar Association. and engagement on juvenile justice reform by the Haywood Burns Institute, both oriented toward engaging practitioners and community leaders in efforts to reduce unwarranted disparities in the court system.29)See

The initiatives proposed above are critical to state policymakers and citizens being able to identify and respond to the racial/ethnic dynamics of incarceration. This is also an important moment for such an undertaking, given that there is increasing momentum for reform of policies that have produced a world record prison system. With growing understanding of the consequences of these developments for all communities, continued progress for reform in the next decade may be possible.

Appendix Table 1. Number of people under the jurisdiction of state prisons 1990, 2000, 2010
State Number % Change
1990 2000 2010 1990-2000 2000-2010
AKa 2,622 4,173 5,285 59.2 26.6
AL 15,665 26,332 31,764 68.1 20.6
AR 6,766 11,915 16,204 76.1 36.0
AZb 14,261 26,510 40,130 85.9 51.4
CA 97,309 163,001 165,062 67.5 1.3
CO 7,671 16,833 22,815 119.4 35.5
CTa 10,500 18,355 19,321 74.8 5.3
DEa 3,471 6,921 6,794 99.4 -1.8
FL 44,387 71,319 104,306 60.7 46.3
GAb 22,345 44,232 49,164 98.0 11.2
HIa 2,533 5,035 5,891 98.8 17.0
IA 3,967 7,955 9,455 100.5 18.9
ID 1,961 5,535 7,431 182.3 34.3
IL 27,516 45,281 48,418 64.6 6.9
IN 12,736 20,125 28,028 58.0 39.3
KS 5,777 8,344 9,051 44.4 8.5
KY 9,023 14,919 20,544 65.3 37.7
LA 18,599 35,207 39,445 89.3 12.0
MA 8,273 10,722 11,312 29.6 5.5
MD 17,848 23,538 22,645 31.9 -3.8
ME 1,523 1,679 2,154 10.2 28.3
MI 34,267 47,718 44,113 39.3 -7.6
MN 3,176 6,238 9,796 96.4 57.0
MOb 14,943 27,519 30,614 84.2 11.2
MS 8,375 20,241 21,645 141.7 6.9
MT 1,425 3,105 3,716 117.9 19.7
NC 18,411 31,266 40,116 69.8 28.3
ND 483 1,076 1,486 122.8 38.1
NE 2,403 3,895 4,587 62.1 17.8
NH 1,342 2,257 2,731 68.2 21.0
NJ 21,128 29,784 25,007 41.0 -16.0
NM 3,187 5,342 6,659 67.6 24.7
NV 5,322 10,063 12,653 89.1 25.7
NY 54,895 70,199 56,461 27.9 -19.6
OH 31,822 45,833 51,712 44.0 12.8
OK 12,285 23,181 26,252 88.7 13.2
OR 6,492 10,580 14,014 63.0 32.5
PA 22,290 36,847 51,264 65.3 39.1
RIa 2,392 3,286 3,357 37.4 2.2
SC 17,319 21,778 23,578 25.7 8.3
SD 1,341 2,616 3,434 95.1 31.3
TN 10,388 22,166 27,451 113.4 23.8
TX 50,042 166,719 173,649 233.2 4.2
UT 2,496 5,637 6,807 125.8 20.8
VA 17,593 30,168 37,410 71.5 24.0
VTa 1,049 1,697 2,079 61.8 22.5
WA 7,995 14,915 18,235 86.6 22.3
WI 7,362 20,754 22,724 181.9 9.5
WV 1,565 3,856 6,681 146.4 73.3
WY 1,110 1,680 2,112 51.4 25.7
Median 71 21
Source: BJS National Prisoner Statistics Program
aPrisons and jails form one integrated system. Data include jail and prison populations
bPrison population based on custody counts

Footnotes   [ + ]

Related Posts
January 19, 2017

State Advances in Criminal Justice Reform, 2016

During 2016, reforms were adopted in at least 17 states targeted at reducing prison populations and addressing collateral consequences for people with criminal convictions.
The Obama Legacy: Chipping Away at Mass Incarceration
December 21, 2016

The Obama Legacy: Chipping Away at Mass Incarceration

Marc Mauer assesses the accomplishments of the Obama Administration and provides thoughts on criminal justice reform for the coming years.