Inequality and Punishment

Abstract

After decades of steady expansion, state prison populations declined in recent years for the first time since 1972. Though the size of the decrease was small, it masks substantial state heterogeneity. This article investigates variation in state-level incarceration rates from 1980 through 2013, examining the factors associated with the rise and decline in prison populations. We find evidence for four key stories in explaining the prison decline: crime, budgets, politics, and inequality. Many of these relationships are consistent across decades, including the role of racial composition, violent crime, and Republican political dominance. In contrast, states’ fiscal capacity and economic inequality became more important after 2000. This research emphasizes the importance of examining changes over time in the correlates of incarceration growth and decline and represents the first effort to systematically understand the recent reversal in the trajectory of incarceration practices in the United States.

Keywords

mass incarceration prisons decarceration inequality

A wealth of scholarship now documents the spectacular rise in incarceration in the United States over the past four decades. During this time, incarceration rates increased more than fivefold, leaving the United States with the highest rate of incarceration in the world (Lacey 2008). The consequences of this dramatic rise in the prison population have yet to receive a full accounting, but a large body of research now points to the serious implications of mass incarceration for social, economic, and political inequality (National Research Council 2014). Individuals who have been incarcerated are less likely to stay connected to their families (Lopoo and Western 2005), face increased barriers to finding work (Pager 2007), and are often prohibited from political participation (Manza and Uggen 2006). In addition, mass incarceration further disadvantages vulnerable communities (Clear 2007) and children growing up with imprisoned parents (Wakefield and Wildeman 2013). With roughly one in ten men (and fully one in three black men) expected to serve time in prison during their lifetime (Bureau of Justice Statistics 2003), the criminal justice system has become a major stratifying institution in contemporary America.

And yet, even as scholars continue to grapple with the causes and consequences of mass incarceration, a new pattern has crept into view that potentially shifts the discussion in important ways. In 2010, the United States began to downsize its total prison population for the first time since 1972 (Guerino, Harrison, and Sabol 2011). Though the size of the decrease has been small—state prison populations declined by just over 3 percent between 2009 and 2013—it marks a notable reversal in a four-decade-long trend. Moreover, the recent national decreases mask substantial heterogeneity at the state level over a longer time span, with some states having reduced their incarceration rates by 20 percent in the past decade (Greene and Mauer 2010). Even as scholars of incarceration continue to investigate the causes and consequences of dramatic increases in recent decades, the possible reversal of these trends also warrants serious consideration (Clear and Frost 2013).

This article investigates variation in state-level incarceration rates from 1980 through 2013, examining the range of social, economic, political, and criminological factors associated with the rise and decline in prison populations in each decade. As a starting point for our investigation, we look to those factors most implicated in the buildup of mass incarceration. Rather than simply assuming that the same factors underlying incarceration’s rise are also central to its decline, however, we directly test the influence of each set of variables. To do this, we model the determinants of annual state incarceration rates, using interaction terms with decade indicator variables (1980s, 1990s, 2000s, and 2010s) to test for variations in the influence of important factors across time. Our analyses focus on four key stories that may account for recent declines in incarceration: a crime story, a budget story, a political story, and an inequality story.

Overall, the results point to continuity in the effects of some factors, while others appear to shift in influence. For example, violent crime consistently predicts increases in incarceration across decades, as does the strength of the Republican Party among state leaders. By contrast, state-level fiscal capacity was not a driving force behind the growth in incarceration until the most recent recession, when we have started to see budgetary pressures play a role in the slowdown in prison populations. Finally, the results point to the significance of racial and economic inequality in the buildup of mass incarceration. States’ racial composition has been a consistent predictor of prison growth as far back as we can measure. By contrast, as economic inequality climbed in recent decades, we have seen an increasing association between this indicator and the growth in incarceration. Together, these results point to the complex array of factors associated with historical growth and recent declines in incarceration and point to some of the consequences of living in a high-inequality regime.

The rise of mass incarceration

For most of the twentieth century, the size of the state prison population remained fairly constant—a little over 100 inmates in state prisons per 100,000 residents—in line with many of our Western European counterparts. Beginning in the early 1970s, however, these numbers changed dramatically, quadrupling over the next four decades. The rate of increase slowed somewhat after the mid-1990s but continued to climb steadily, reaching a high of 447 state inmates per 100,000—or 1.4 million individuals—by the year 2008).1 Though prison expansion has been the dominant trend in the American criminal justice system over the past four decades, tremendous variation exists across states along this dimension. States like Maine, Massachusetts, New Hampshire, and Minnesota have incarceration rates very similar to those in the UK, Spain, Hungary, and Turkey (between 150 and 200 inmates per 100,000 residents). Others, by contrast, such as Louisiana, Mississippi, Oklahoma, and Texas boast rates of incarceration that well exceed any other country on record (Walmsley 2013).

State variation in the growth of imprisonment has been primarily attributed to decisions made by state policy-makers during this era of “tough-on-crime” politics (Western 2006; Raphael 2009). Between 1984 and 1998, the chances of receiving a prison sentence following arrest increased by more than 50 percent, and the average length of sentences increased by nearly 40 percent (Blumstein and Beck 1999). These increases continued into the 1990s and beyond, with sentence lengths increasing by more than 35 percent between 1990 and 2009 (Pew Center on the States 2012; though see Pfaff 2011). The lengthening of prison sentences is often explained as a result of legislation such as “truth in sentencing” and “three-strikes” laws, as well as the emergence of harsh mandatory minimum sentencing schemes (Spelman 2009). While all states adopted some form of mandatory sentencing legislation, differences in the nature and scope of state law resulted in significant variation in subsequent incarceration trajectories.

Beyond these specific policy levers, scholars have sought to understand the social, economic, and political contexts in which imprisonment rates have grown. In other words, why did some states enact the punitive legislation that rapidly expanded prison rolls while other states restrained growth? The results of this literature vary across studies depending on the nature of the data (cross-sectional or panel), model choices, and the range of variables included. Though by no means universal, the most consistent results point to the importance of crime rates, racial composition (especially percentage black), unemployment rates, partisan control of states’ legislatures and governorships, and state budget strength (Beckett and Western 2001; Greenberg and West 2001; Guetzkow 2011; Jacobs and Carmichael 2001; Smith 2004; Stucky, Heimer, and Lang 2007; Spelman 2009; Western 2006). We review this literature in more detail below as we describe the four central stories that guide our analysis.

A turning point in mass incarceration?

The literature on incarceration has been dominated by concerns over the growth of prison populations. This important line of work will be relevant for many years to come, as states continue to grapple with historic numbers of inmates and former inmates in their prisons and communities. And yet there is some reason to believe that the 40-year trend of prison expansion may at last have run its course (Clear and Frost 2013). In 2010, the national prison population declined for the first time since 1972, with half of states reporting a decline in the number of prison inmates (Guerino, Harrison, and Sabol 2011). It is likely no coincidence that these declines took place in the wake of a deep economic crisis, during which states faced record budget deficits (Aviram 2015). But there is reason to believe that recent reductions in incarceration have not been driven by the current economic crisis alone, with some states showing longer and deeper histories of decarceration.

How can we explain this important shift in the use of incarceration across states? What social, political, and economic factors may help us to understand state-level variation in recent declines? The literature on the rise of incarceration discussed above can offer some useful hypotheses for state-level correlates that may predict changes in incarceration. And yet it would be a mistake to simply assume that the same variables associated with the rise of incarceration are those driving recent declines (cf. Lieberson 1987, 86). It is thus important to empirically investigate whether similar social processes that led to the rise of incarceration in the 1980s and 1990s are also associated with the recent reversal in incarceration trends in the 2000s and/or whether a different set of factors appear to be associated with this decline.

Building on recent syntheses of historical research, Campbell, Vogel, and Williams (2015) pick up on a parallel question, examining whether there have been distinct “epochs” in the carceral buildup. Using state-level decennial data from 1970 to 2010, they find that the correlates of high imprisonment rates across states vary by decade. In a state-level fixed effects model, they find that violent crime had significant effects on incarceration rates in 1980 and 1990 (relative to 1970) and that Republican strength mattered in 1990 and 2000, but not in other decades. The interaction terms for 2010 incarceration rates showed few significant effects (other than percentage black), suggesting that the standard determinants of incarceration rates during the buildup poorly explain recent changes.

Our investigation furthers this nascent literature by considering four domains or “stories”—money, politics, crime, and inequality. Similar to Campbell, Vogel, and Williams (2015), we systematically test for changes in the nature and magnitude of effects over time. However, we model these changes in more detail, using annual—rather than decennial—data and extending the time series out to 2013, as well as providing descriptive visualization of recent trends. Finally, our analysis directly interrogates the role of rising inequality as a factor in the differential expansion and contraction of incarceration across time. Indeed, one of the consequences of “living in a high-inequality regime” may be a special form of punitive control that emerges with growing distance between the “haves” and the “have-nots.” Together, these analyses help us to make sense of the historical correlates of mass incarceration, as well as those factors associated with its recent signs of reversal.

Data and Methods

Following the bulk of literature on this topic, our primary dependent variable is state imprisonment rates, measured as the total number of prisoners in a state’s jurisdiction serving sentences of a year or more (including inmates housed in local jails, private facilities, or other states due to overcrowding) per 100,000 in the population, drawing on data published by the Bureau of Justice Statistics. We chose the state as the unit of analysis as most crime policy is set at the state level and thus most analyses, and data reporting, focus at this level of aggregation. In addition, we measure the current imprisonment rate, instead of the prison admission rate, as demographic, budgetary, and political shifts likely influence both admission and release trends.

In what follows, we provide descriptive evidence on the role of each stylized story in shaping recent reform trends. For each story, we have a primary indicator variable or set of variables. For the crime story, we rely on rates of violent index crimes as collected by the Uniform Crime Reports. These statistics measure the number of crimes reported to the police (per 100,000 inhabitants) for a series of specific offenses (murder and nonnegligent manslaughter, forcible rape, robbery, and aggravated assault). We also look at the drug arrest rate (reported by local jurisdictions to the FBI) to see if drugs represent a special crime category.² For the budget story, we examine data from the U.S. Census’ Survey of State and Government Finances, looking in particular at the ratio of all state revenue to state debt in a given year (logged to reduce skew). We also consider general state expenditures and the percentage of those expenditures devoted to police protection and welfare.³ For the political story, we rely on the State Partisan Balance Dataset (Klarner 2003). We create a composite term for Republican strength that interacts the presence of a Republic governor and percentage Republican in the state legislature. States with Democratic governors have a Republican strength value of zero, while states with Republican governors have a Republican strength value equal to the percentage of Republicans in the upper and lower legislatures. Finally, for the inequality story, we investigate both economic and racial inequality, with measures of state-level income inequality (gini coefficient) and racial composition (percent black) from the U.S. Census. For each variable, we examine whether starting levels and/or recent changes help to explain why some states downsized their prison populations.

We then assess whether the descriptive findings hold up in a multivariate time-series model that estimates the effects of these variables on states’ imprisonment rates between 1980 and 2013. As with prior research, we lag the independent variables by one year to better account for causal ordering between state-level trends and imprisonment rates.⁴ To control for secular time trends, we include a (centered) term for year.⁵ Since we are interested in exploring determinants across states, we include a random effect for states to control for time-invariant characteristics. In addition, we cluster the error terms by state (i.e., robust standard errors), further ensuring against spurious results due to correlation between state-years. To address asymmetrical forms of causations (Lieberson 1987), we interact decade (1990s, 2000s, and 2010s) dummy variables with the key outcomes of interest to test whether the effects of these state characteristics vary from the buildup to the scaling back of prison populations (with the 1980s as the omitted reference category). For parsimony, in the presented results, we include decade interactions for the key stories only when they show statistically significant differences over time. In addition to the variables of direct interest, we also incorporate a range of control variables, including: percentage Hispanic, percentage foreign-born, urbanicity, unemployment rate, and total state population. More detailed information on these variables can be found in Appendix A.

These data contain some missing information. The most commonly missing values are for the variables only available for census years (percentage urban and percentage foreign-born) and the drug arrest rate. Finally, the political partisanship of legislators variable is missing in all years for Nebraska. Rather than use list-wise deletion or linear interpolation—which can lead to biases and inefficiencies—we estimate missing values through the Amelia II multiple imputation program which allows us to model the uncertainty around missing values (Honaker and King 2010; Honaker, King, and Blackwell 2011). The analyses were then conducted on the multiply imputed data (with forty imputations).

Results

For each of the four stories we investigate as possible explanations of incarceration decline, we begin with basic bivariate associations in the most recent years as a way of identifying potentially important dynamics and to carefully observe trends across states in a transparent format. We then compare these results to those generated by our multivariate models, looking across decades. The full results for the state-level random effects model are listed in Table 1.

Table 1

State Random Effects Model Predicting Incarceration Rates, 1980–2013

	B	SE
Key stories
Violent crime rate	0.088**	(0.037)
Drug arrest rate	0.034	(0.032)
× 1990s decade	0.088***	(0.031)
× 2000s decade	0.042	(0.049)
× 2010s decade	0.039	(0.075)
Fiscal capacity	−8.430	(11.32)
× 1990s decade	−6.169	(10.58)
× 2000s decade	24.98*	(13.92)
× 2010s decade	31.65*	(16.16)
Expenditures per cap (log)	56.25*	(32.32)
Spending on welfare (%)	−0.395	(2.052)
Spending on police (%)	−68.00***	(17.33)
Republican strength	0.366***	(0.096)
Income inequality (gini)	−9.137***	(2.868)
× 1990s decade	9.544***	(3.085)
× 2000s decade	12.38***	(3.870)
× 2010s decade	13.73***	(4.474)
Percentage black	5.710**	(2.309)
Control variables
Percentage Hispanic	−2.829	(2.304)
Percentage foreign-born	−0.214	(1.146)
Percentage urban	−0.248	(0.398)
Unemployment rate	−4.940***	(1.417)
State population	0.007**	(0.003)
Year (centered)	7.219***	(1.159)
Intercept	575.9***	(138.0)
Decade dummy 1990s	−400.7***	(125.4)
Decade dummy 2000s	−532.9***	(166.6)
Decade dummy 2010s	−633.5***	(202.8)

NOTE: The final number of observations was 1,650 (50 states over 33 years). Predictor variables were lagged by one year. Missing data were estimated through multiple imputation (forty imputations). Standard errors are clustered by state.

p < .10. **p < .05. ***p < .01.

We begin with an account of recent trends in imprisonment rates. Nationally, the peak of total state prison populations was hit in December 2009 (at 1,366,000 prisoners) and declined in 2010, 2011, and 2012. In 2013, the national state prisoners population slightly increased—causing some observers to worry that the prison decline was already over. Between 2010 and 2013, over half of states (thirty) saw declines in their imprisonment rates, with historically tough-on-crime Texas declining by 7 percent and California by 20 percent. Taking a longer view, between 2000 and 2013, sixteen states showed declines in their imprisonment rates, with New York and New Jersey both declining by over 25 percent. This suggests that recent declines in imprisonment rates have a longer history than just the most recent fiscal crisis. Indeed, states’ average across-decade change in imprisonment rates dropped precipitously from a mean increase of 131 (prisoners per 100,000 residents) in 1990 to 2000 to just 28 in 2000 to 2010.

A crime story

Intuitively, we think of levels of punishment in a society as primarily determined by the corresponding rate of crime. Yet scholars of criminal justice largely agree that changes in policy, rather than changes in crime, have been the primary drivers of incarceration over the past three decades, particularly as imprisonment rates continued to climb in the 1990s and 2000s despite steep declines in crime (Blumstein and Beck 1999; Western 2006). Drug enforcement, for example, a major driver of the prison boom, was largely disconnected from broader patterns of criminal activity, and even from drug use and abuse (Tonry 1995). At the same time, there does appear to be some evidence that more punitive policies developed in states with a recent or longer-term histories of violent crime (Greenberg and West 2001; Jacobs and Carmichael 2001; Western 2006).

In considering the role of crime in recent trends, we considered both levels and changes. While we find little relationship between states’ level of violent crime in 2010 and subsequent imprisonment rate trends, we find that states with greater declines in violent crime between 2010 and 2012 were more likely to see decarceration between 2010 and 2013 (.26, p < .1 ), as depicted in Figure 1. Similarly, states experiencing greater reductions in violent crime between 2000 and 2012 were more likely to reduce their prison populations between 2000 and 2013 (.31, p < .05).

Figure 1

Violent Crime Rates and Imprisonment Trends

In the multivariate model, we find a positive, statistically significant effect (at the .05 level) for violent crime rates on imprisonment rates (see Table 1). No significant decade interactions were found, so they are omitted from the final model. While the effect of violent crime remains unchanged, its level has seen a dramatic reduction, with violent crime rates today lower than they were at the start of the data series. The scale of the effect is such that one standard deviation increase decrease (or 129 crimes per 100,000) in states’ 2012 violent crime rate is associated with a decrease in the following year’s imprisonment rate of 11 prisoners per 100,000. While this effect may seem small, the average change in states’ imprisonment rates in 2013 was -0.3 prisoners, with a standard deviation of 16 prisoners per 100,000. These findings also suggest that despite fears about decarceration’s effect on crime, there is no evidence that declining prison populations are associated with rising crime rates—and in fact the descriptive results suggest the opposite effect.⁶ Thus, reductions in imprisonment appear to be entirely compatible with continued improvements in public safety.

In addition to index crime rates, we also consider a number of additional factors related to the crime context in a given state. Drug crimes, for example, are not reflected in the index crime rate, despite the fact that the “war on drugs” played a significant role in the expansion of the criminal justice system (Tonry 1995). Indeed, drug arrests show a highly significant and positive relationship to incarceration rates in the 1990s, the decade with the most rapid expansion of drug prosecution. In contrast, there is no significant effect for the 1980s (baseline), nor are the interaction terms for the 2000s or 2010s statistically significant. This suggests that the buildup of drug arrests in the 1990s was a strong determinant of states’ imprisonment rates but grew less important in the 2000s and 2010s as the drug war fervor began to fade.

A budget story

Funding for corrections has been one of the fastest-growing state budget categories in the past decade, second only to Medicaid. Since the early 1970s the proportion of state funding dedicated to corrections has doubled; by 2008, states spent nearly $50 billion on prisons, probation, and parole (Pew 2008). The recession of 2008 triggered a flurry of interest in corrections spending as states struggled to close record budget deficits (Brown 2013; Vera Institute of Justice 2010). Between 2009 and 2010, more than half of states reported a decline in prison expenditures (Scott-Hayward 2009; Subramanian and Tublitz 2012) and at least forty states cut back some types of correctional expenditures (e.g., on labor costs, programs, or food services) (Porter 2012). By 2011, the average change in corrections budgets across states was negative, with some states decreasing corrections spending up to nearly 10 percent (Vera Institute of Justice 2010).

We assess state economic well-being using a measure of fiscal capacity, which we calculate as the ratio of state revenue to debt (logged to reduce skew).⁷ We find that states with lower fiscal capacity in 2010 were significantly more likely to reduce their incarceration rates between 2010 and 2013 (.25, p < .10). Another way to frame this question is to look across time, particularly at the financial downturn. States with the greatest declines in fiscal capacity between 2000 and 2009 were more likely to reduce their incarceration rates in the subsequent three years (.31, p < .05), as displayed in Figure 2.

Figure 2

Fiscal Capacity and Imprisonment Trends

Our multivariate models tell a consistent story. We find no significant effects for fiscal capacity in the 1980s or 1990s. The 2000s interaction term is positive and marginally significant, and the 2010s interaction term is larger still and just barely past the .05 significance level. States with a one standard deviation drop in fiscal capacity in 2012 are expected to see a decline in the imprisonment rate of 11 prisoners (per 100,000) in the following year, the same magnitude as the effect of violent crime. Together, these results suggest that the buildup of mass incarceration was achieved with little attention to its financial implications—with states facing a series of budget crises while continuing to build. Yet more recent budgetary pressures have led to greater rethinking of punitive criminal justice policy.

Our multivariate models also include several other spending measures. After controlling for fiscal capacity, we find a positive (but only marginally significant) effect for states’ total general expenditures per capita, confirming that prison populations are tied to state spending. We find no link to the percentage of states’ general expenditures devoted to welfare spending and imprisonment rates (but see Beckett and Western 2001). We do, however, find a strong negative relationship between state spending on police and imprisonment rates (although note that police forces also rely on local budgets, which are not included). A one standard deviation increase in the percent of general spending devoted to police in 2012 (or an increase of 0.3 percentage points) is associated with a decrease of 21 (per 100,000) in the following year’s imprisonment rates. This suggests that states can control both crime and correctional populations through targeted spending on police. Indeed, targeted policing using “hot spots” and other response tactics have been credited for significant crime declines in major urban areas like New York City, although they have also been critiqued for increasing collateral consequences (Weisburd and Eck 2004).

A politics story

Much of the story of the rise of mass incarceration focuses on dramatic changes in the politics of punishment—or how crime and punishment are framed in the public arena (Beckett 1997; Simon 2007). Prior to the 1970s, crime and criminal justice policy had largely been regarded as a local problem, appropriately managed through decentralized police units, courts, and local governments. Starting in the mid- to late 1960s, and particularly Nixon’s “war on crime,” punishment began to feature prominently in state and national campaigns. While “tough on crime” rhetoric has been popular among Republicans and Democrats alike, some research has shown a relationship between punitive crime policies and Republican strength and the public’s conservativism (Guetzkow 2011; Jacobs and Carmichael 2001; Smith 2004; Yates and Fording 2005). The politics of imprisonment may be important to the story of recent reductions as well, as leaders on both sides of the aisle begin to propose criminal justice reform.

In the multivariate models, we collapse control of the governor’s office and the partisan composition of the legislature identified as Republican into one indicator of Republican strength (Jacobs and Carmichael 2001). For the descriptives, we focus on the association between incarceration and the percentage of a state’s legislature controlled by Republicans as this measure is easier to display visually. While the 2000 and 2010 levels of Republican control show little relationship to trajectories of incarceration, changes in the percentage Republican between 2010 and 2012 are positively correlated with incarceration rate changes from 2000 to 2013 (.36, p < .05), as depicted in Figure 3. States with more modest increases in the percentage of legislators identified as Republicans in this period were more likely to show incarceration reductions. Likewise, changes in Republican control between 2000 and 2012 are associated with changes in incarceration rates from 2000 to 2013 (.29, p < .05).

Figure 3

Republican Legislators and Imprisonment Trends

Multivariate models confirm the import of Republican state leaders. The results show a significant relationship between Republican strength and states’ imprisonment rates with no significant decade interaction terms. This relationship suggests that in states with a Republican governor, a one standard deviation increase in the percentage of the legislature controlled by Republicans in 2012 (or 17 percentage points) is associated with an increase of 6 (per 100,000) in the imprisonment rate in the following year. In contrast to commentators and empirical findings of a waning partisan influence (Campbell, Vogel, and Williams 2015), we find that party politics continue to matter for imprisonment rates. Though criminal justice reform represents an increasingly bipartisan priority, our results suggest that Democratic-controlled states have to date been more effective at translating that rhetoric into measurable reductions in prison populations. Whether Republicans will soon catch up (as reforms in conservative states such as Texas suggest) remains to be seen.

An inequality story

The rise of income inequality has been one of the most spectacular social changes in the United States since the 1970s. Alongside the growth of incarceration, we have seen a steady widening of the gap between those at the top and bottom of the income distribution and a growing concentration of wealth among the top 1 percent (Piketty 2014). In part crystallized by the Occupy Wall Street movement, concerns about rising inequality have been growing, and scholars argue that such gaps between the rich and poor produce less forgiving and more punitive social policy (Wilkinson and Pickett 2009; though see Saunders 2010; Deaton 2003). This line of research also calls attention to the centrality of racial disparities in understanding inequality (again highlighted by recent protests, particularly those around police violence in Ferguson and elsewhere). Particularly in discussions of the criminal justice system, the racial context cannot be overstated (Alexander 2010).

We consider two key variables related to inequality: income inequality among households (gini coefficient) and racial composition (percentage of states’ resident population identified in the census as black, including Hispanics).⁸ We find a cross-sectional relationship between states’ gini coefficients and incarceration rates in 2012, with high imprisonment rates concentrated in those states with the greatest levels of inequality (the correlation is .23, with a p-value just above the p < .1 cut-point; however, if New Mexico is excluded, the correlation increases to .30 and is significant at the .05 level). We find a very strong bivariate relationship between percent black and the imprisonment rate in 2012 (.58, p < .001). Yet in part because these two variables tend to change slowly, bivariate correlations for changes in inequality and incarceration rates produce few significant results.

The multivariate context, by contrast, illustrates the effects of both variables more clearly. In terms of economic inequality, the results show a negative and significant effect for states’ gini coefficients in the 1980s, suggesting counterintuitively that early trajectories of rising incarceration were concentrated in states with low levels of inequality. By the 1990s, however, the direction of this relationship had flipped, and for the remaining time series we see a strong positive relationship between inequality and imprisonment. By the 2010s, the coefficient translates into an increase of 12 inmates per 100,000 residents for every one standard deviation increase in the 2012 gini. The multivariate results also reveal that percentage black is an important and significant correlate of imprisonment rates, and its effect does not vary significantly across decades. The coefficient estimates that a one standard deviation increase in the percentage black in 2012 (10 percentage points) is associated with an increase of 55 (out of 100,000) in the imprisonment rate.

Together, these results suggest that both race and class inequalities are crucial for explaining differences across states. While these characteristics change very slowly over time, making it more difficult to detect significant associations in the descriptive data over time, the multivariate results point to the persistent race and class dimensions shaping trends in the American criminal justice system. Furthermore, the impact of income inequality has strengthened over time—a particularly problematic development given the increases in states’ gini coefficients over the decades. As economic inequality has crept up in the postrecession period, it remains to be seen how patterns of incarceration may be affected moving forward.

Summary

Overall, these patterns offer some support for each of the four stories we investigate, with some persistent effects and others newly emerging. The “crime story” was supported, with violent crime rates positively correlated with imprisonment rates across the decades. The “budget story” also revealed that states struggling the most with fiscal capacity were the most likely to reduce prison populations after 2010. This suggests that the rhetoric around financial pressures was not simply empty talk but, rather, an important contributor to recent downward trends. In contrast, we expected to find a waning influence of partisan control; but instead the results suggest that Republican legislators and governors continue to be associated with higher imprisonment rates. Whether this will change as we move further into the 2010s remains to be seen. Finally, working against the progressive trends associated with reductions in violent crime and shrinking state budgets, we find persistent racial dynamics and growing economic inequality may slow the pace of reform. Despite these barriers, the 2010s indeed represented a real turning point for U.S. incarceration rates. Setting the stage for that decline—two decades of falling crime rates and the worst economic recession of a generation—was a fairly profound reconfiguring which is likely to be felt well into the future.

Conclusions

Recent trends in states’ prison populations offer a glimpse into a possible future of crime and punishment. The large declines over the past decade in some states suggest that the recent national downturn is not just a momentary response to the current financial crisis but, rather, a potential sign of declines for many years to come. As the politics of crime and punishment continue to shift, we may continue to see declines even as states’ budget health increase. Whereas earlier debates on crime policy equated alternatives to incarceration as being “soft on crime,” recent reforms show that reductions in incarceration have appeared alongside continued crime drops. To the extent that reductions in prison populations can be achieved without a corresponding increase in crime bodes well for continued public support of alternative (and less punitive) approaches.

If states plan decarceration wisely, this trend has the potential to become one of the most equality-enhancing institutional shifts we have witnessed in many years. For each of the “collateral consequences” of mass imprisonment, we can outline parallel gains for reductions in imprisonment rates. For example, fewer inmates serving prison sentences will alleviate the stigmatizing labor market consequences of imprisonment (Western 2006; Pager 2007), make voter rolls more democratically representative (Lerman and Weaver 2014), and reunite families and communities that otherwise would have faced the separation of imprisonment (Comfort 2008; Clear 2007). This could in turn help to better protect the next generation of children growing up in the families most affected by criminal justice control (Wakefield and Wildeman 2013). Lastly, to the extent that decarceration reduces racial disparities and/or the experience of imprisonment among black families, it has the potential to reduce racial inequalities for children and adults by reducing the number of “missing men” and de-normalizing the prison experience (Alexander 2010).

At the same time, it is important to recognize that how reductions in imprisonment unfold will have deep ramifications for its consequences (Gottschalk 2014). As mentioned above, states are working to reduce prison populations through a variety of mechanisms, some of which affect the inflow of inmates—including sentencing reform and improvements to probation and parole practices that reduce the rate of revocation to prison—and others that affect the outflow of inmates—including shorter presumptive sentences and increasing opportunities for early release. Reforms that reduce imprisonment rates simply via a transfer to community corrections (probation and parole) seem particularly limited. Individuals with a criminal record, regardless of their prison experience, will still face discrimination and legal barriers, including limitations on employment, housing, and voting (Western 2006; Pager 2007; Petersilia 2003). The arena where this may make a substantial difference is in family and community-level effects, since individuals on probation and parole are generally not removed from the community.

In addition, without the proper resources and practices in place for probation and parole, a transfer of prisoners to community supervision (or no supervision at all) could increase revocation rates and drive up correctional costs (Phelps 2013; Weisberg and Petersilia 2010). This could in turn spike crime and create a new wave of reform backlash, particularly as states’ budgets recover (Gottschalk 2014). It is worth remembering that the individuals being sent to prison bring with them a host of problems, including high rates of substance abuse, mental health problems, and poor earnings prospects, all of which create challenges for reentry. Recognizing these concerns, a number of states are now participating in “justice reinvestment,” which aims to reduce spending on corrections and to reinvest in disadvantaged communities (often through community corrections). While the implementation of these initiatives has met with mixed evaluation (Austin et al. 2013; LaVigne et al. 2014), there is good evidence to suggest that states can make probation and parole “smarter” in ways that can reduce overall correctional populations and costs while maintaining a strong commitment to public safety (Jacobson 2005; Pew 2009).

In addition, there is growing support for various decriminalization and diversion options, particularly for drug offenders. These policies work by downgrading criminal charges (allowing more individuals to avoid felony-level convictions) and/or allowing for diversion and record expunging once alternative sanctions are complete (Subramanian and Moreno 2014). Reforms that reduce the number of individuals targeted with the stigma of a conviction record may have the greatest effects on reducing inequalities. However, it is worth noting that in many cases, states are simply expanding the reach of misdemeanor punishment, rather than truly decriminalizing, and even this low-level involvement with the criminal justice system can have negative effects on individuals’ life outcomes (Natapoff 2014).

Thus, it remains to be seen whether and to what extent recent declines will persist into the future and how their implementation will shape patterns of inequality. The massive changes in the criminal justice system over the past four decades have been among the most significant transformations in the stratification system during this time; recent signs of a reversal could signal a similarly historic shift. Indeed, given the substantial impact of incarceration, particularly in the lives of the poor, changes in the scale of incarceration will likely represent one of the most significant institutional shifts affecting inequality in the years to come.

Footnotes

Appendix A

NOTE:

Direct comments or queries to Michelle Phelps (phelps@umn.edu). Thanks to Tracy Snell from the Bureau of Justice Statistics for assistance in assembling the data for our criminal justice time series. We received helpful comments on an earlier version of this article from Rob Mare, Gerald Marwell, Pam Oliver, Marcy Carlson, and attendees at the Vera Institute of Justice’s Neil A. Weiner Research Speaker Series (October 2011) and the 2012 American Sociological Association meeting, where earlier drafts of the article were presented.

Notes

Michelle S. Phelps is an assistant professor in sociology at the University of Minnesota and is affiliated with the University of Minnesota Law School and Minnesota Population Center. Her research has been published in Law and Society Review, Law and Policy, and Theoretical Criminology.

Devah Pager is a professor of sociology and public policy at Harvard University. Her research focuses on racial inequality in labor markets and the criminal justice system. Her 2007 book, Marked: Race, Crime, and Finding Work in an Era of Mass Incarceration, examines the labor market barriers facing those recently released from prison. Pager was adopted into the Hauser clan in 1998 and has been forever since blessed by their wisdom, kindness, and dogged commitment to getting the science right.

References

Alexander

2010. The New Jim Crow—Mass incarceration in the age of colorblindness. New York, NY: The New Press.

Austin

James

Cadora

Eric

Clear

Todd R.

Dansky

Kara

Greene

Judith

Gupta

Vanita

Mauer

Marc

Porter

Nicole

Tucker

Susan

Young

Malcolm C.

2013. Ending mass incarceration: Charting a new justice reinvestment. Unpublished Paper.

Aviram

Hadar

. 2015. Cheap on crime: Recession-era politics and the transformation of American punishment. Berkeley, CA: University of California Press.

Beckett

Katherine

. 1997. Making crime pay: Law and order in contemporary American politics. New York, NY: Oxford University Press.

Beckett

Katherine

Western

Bruce

. 2001. Governing social marginality: Welfare, incarceration, and the transformation of state policy. Punishment and Society 3 (1): 43–59.

Blumstein

Alfred

Beck

Allen J.

1999. Population growth in U.S. prisons, 1980-1996. In Crime and justice: A review of the research, Vol. 26, Prisons, eds. Tonry

Petersilia

, 17–61. Chicago, IL: University of Chicago Press.

Brown

Elizabeth K.

2013. Foreclosing on incarceration? State correctional policy enactments and the Great Recession. Criminal Justice Policy Review 24 (3): 317–37.

Bureau of Justice Statistics. 2010. Sourcebook of criminal justice statistics, Table 6.13.2010. Washington, DC: National Institute of Justice.

Campbell

Michael C.

Vogel

Matt

Williams

Joshua

. 2015. Historical contingencies and the evolving importance of race, violent crime, and region in explaining mass incarceration in the United States. Criminology 53 (2): 180–203.

10.

Clear

Todd R.

2007. Imprisoning communities: How mass incarceration makes disadvantaged neighborhoods worse. Studies in Crime and Public Policy. Oxford: Oxford University Press.

11.

Clear

Todd

Frost

Natasha

. 2013. The punishment imperative: The rise and failure of mass incarceration in America. New York, NY: New York University Press.

12.

Comfort

Megan

. 2008. Doing time together: Love and family in the shadow of the prison. Chicago, IL: University of Chicago Press.

13.

Deaton

Angus

. 2003. Health, inequality, and economic development. Journal of Economic Literature 41 (1): 113–58.

14.

Gottschalk

Marie

. 2014. Caught: The prison state and the lockdown of American politics. Princeton, NJ: Princeton University Press.

15.

Greenberg

David F

West

Valerie

. 2001. State prison populations and their growth, 1971-1991. Criminology 39:615.

16.

Greene

Judith

Mauer

Marc

. 2010. Downscaling prisons: Lessons from four states. Washington, DC: The Sentencing Project.

17.

Guerino

Paul

Harrison

Paige M.

Sabol

William J.

2011. Prisoners in 2010. Washington, DC: Bureau of Justice Statistics.

18.

Guetzkow

Joshua

. 2011. Bars versus butter: The prison-welfare tradeoff and its political underpinnings. Working Paper.

19.

Honaker

James

King

Gary

. 2010. What to do about missing values in time-series cross-section data. American Journal of Political Science 54 (2): 561–81.

20.

Honaker

James

King

Gary

Blackwell

Matthew

. 2011. Amelia II: A program for missing data. Journal of Statistical Software 45 (7): 1–47.

21.

Jacobs

David

Carmichael

Jason T.

2001. The politics of punishment across time and space: A pooled time-series analysis of imprisonment rates. Social Forces 80 (1): 61–89.

22.

Jacobson

Michael

. 2005. Downsizing prisons: How to reduce crime and end mass incarceration. New York, NY: New York University Press.

23.

Klarner

Carl

. 2003. The measurement of the partisan balance of state government. State Politics and Policy Quarterly 3 (3): 309–19.

24.

Lacey

Nicola

. 2008. The prisoners’ dilemma: Political economy and punishment in contemporary democracies. Cambridge: Cambridge University Press.

25.

LaVigne

Nancy

Bieler

Samuel

Cramer

Lindsay

Helen

Kotonias

Cybele

Mayer

Debbie

McClure

Dave

Pacifici

Laura

Parks

Erika

Peterson

Bruce

Samuels

Julie

. 2014. Justice reinvestment initiative state assessment report. Washington, DC: The Urban Institute.

26.

Lerman

Amy E.

Weaver

Vesla M.

2014. Arresting citizenship: The democratic consequences of American crime control. Chicago, IL: University of Chicago Press.

27.

Lieberson

Stanley

. 1987. Making it count: The improvement of social research and theory. Berkeley, CA: University of California Press.

28.

Lopoo

Leonard

Western

Bruce

. 2005. Incarceration and the formation and stability of marital unions. Journal of Marriage and the Family 67:721–34.

29.

Manza

Jeff

Uggen

Christopher

. 2006. Locked out: Felon disenfranchisement and American democracy. Studies in Crime and Public Policy. Oxford: Oxford University Press.

30.

Natapoff

Alexandra

. 2014. Misdemeanor decriminalization. SSRN Scholarly Paper ID 2494414. Rochester, NY: Social Science Research Network.

31.

National Research Council. 2014. The growth of incarceration in the United States: Exploring causes and consequences. Washington, DC: National Academies Press.

32.

Pager

Devah

. 2007. Marked: Race, crime, and finding work in an era of mass incarceration. Chicago, IL: University of Chicago Press.

33.

Petersilia

Joan

. 2003. When prisoners come home: Parole and prisoner reentry. New York, NY: Oxford University Press.

34.

Pew Center on the States. 2008. One in 100: Behind bars in America in 2008. Washington, DC: Pew Center on the States.

35.

Pew Center on the States. 2009. One in 31: The long reach of American corrections. Washington, DC: Pew Center on the States.

36.

Pew Center on the States. 2012. Time served: The high cost, low return of longer prison terms. Washington, DC: Pew Center on the States.

37.

Pfaff

John F.

2011. The myths and realities of correctional severity: Evidence from the National Corrections Reporting Program on Sentencing Practices. American Law and Economics Review 13 (2): 491–531.

38.

Phelps

Michelle S.

2013. The paradox of probation: Community supervision in the age of mass incarceration. Law and Policy 35 (1–2): 51–80.

39.

Piketty

Thomas

. 2014. Capital in the twenty-first century. Boston, MA: Harvard University Press.

40.

Porter

Nicole D.

2012. The state of sentencing 2011: Developments in policy and practice. Washington, DC: Sentencing Project.

41.

Raphael

Steven

. 2009. Explaining the rise in U.S. incarceration rates. Criminology and Public Policy 8 (1): 87–95.

42.

Saunders

Peter

. 2010. Beware false prophets: Equality, the good society, and the spirit level. London: Policy Exchange.

43.

Scott-Hayward

Christine S.

2009. The fiscal crisis in corrections: Rethinking policies and practice. New York, NY: Vera Institute of Justice.

44.

Simon

Jonathan

. 2007. Governing through crime: How the war on crime transformed American democracy and created a culture of fear. Oxford: Oxford University Press.

45.

Smith

Kevin B.

2004. The politics of punishment: Evaluating political explanations of incarceration rates. Journal of Politics 66 (3): 925–38.

46.

Spelman

William

. 2009. Crime, cash, and limited options: Explaining the prison boom. Criminology and Public Policy 8 (1): 29–77.

47.

Stucky

Thomas D.

Heimer

Karen

Lang

Joseph B.

2007. A bigger piece of the pie? State corrections spending and the politics of social order. Journal of Research in Crime and Delinquency 44 (1): 91–123.

48.

Subramanian

Ram

Moreno

Rebecka

. 2014. Drug war détente? A review of state-level drug law reform, 2009-2013. New York, NY: Vera Institute of Justice.

49.

Subramanian

Ram

Tublitz

Rebecca

. 2012. Realigning justice resources: A review of population and spending shifts in prison and community corrections 2006-2010. New York, NY: Vera Institute of Justice.

50.

Tonry

Michael

. 1995. Malign neglect. New York, NY: Oxford University Press.

51.

Vera Institute of Justice. 2010. The continuing fiscal crisis in corrections: Setting a new course. New York, NY: Vera Institute of Justice.

52.

Wakefield

Sara

Wildeman

Christopher

. 2013. Children of the prison boom: Mass incarceration and the future of American inequality. Oxford: Oxford University Press.

53.

Walmsley

Roy

. 2013. World prison population List. 10th ed. London: International Centre for Prison Studies.

54.

Weisberg

Robert

Petersilia

Joan

. 2010. The dangers of pyrrhic victories against mass incarceration. Daedalus 139:124–33.

55.

Weisburd

David

Eck

John E.

2004. What can police do to reduce crime, disorder, and fear? The ANNALS of the American Academy of Political and Social Science 593:42–65.

56.

Western

Bruce

. 2006. Punishment and inequality in America. New York, NY: Russell Sage Foundation.

57.

Wilkinson

Richard

Pickett

Kate E.

2009. The spirit level: Why greater equality makes societies stronger. London: Bloomsbury Press.

58.

Yates

Jeff

Fording

Richard

. 2005. Politics and state punitiveness in black and white. Journal of Politics 67 (4): 1099–1121.