Toward Refining the Criminology of Mass Incarceration: Group-Based Trajectories of U.S. States, 1977–2010

• DOI: 10.1177/0734016815627859
• Author: Elizabeth K. Brown
• Date: 01 March 2020
• Published date: 01 March 2020
• Subject Matter: Articles

Article

Toward Refining the Criminology

of Mass Incarceration:

Group-Based Trajectories of

U.S. States, 1977–2010

Elizabeth K. Brown

1

Abstract

The development of mass incarceration in the United States has occurred unevenly across American

states. Prior time series, fixed effect, and case study research have failed to fully illuminate the

determinants of incarceration rate change in states with varying patterns of growth. As a supplement

to previously utilized approaches, the present research uses group-based trajectory modeling to

consider patterns of incarceration rate growth across 48 U.S. states in relation to crime, political,

structural, and institutional variables. In order to account for periodicity, group-based trajectory

models of state incarceration rates are estimated separately for 1977–1990, 1990–2000, and 2000–

2010. Findings suggest that political and economic factors vary in their relationships to incarceration

growth over time and that, controlling for crime, the percentage of young Black males in state

populations was the most consistent predictor of incarceration rate growth, particularly among high

incarcerating states from 2000 to 2010. The implications of these findings for ‘‘the criminology of

mass incarceration’’ are considered.

Keywords

incarceration, states, punishment, mass incarceration, trajectories

Introduction

While the development of mass incarceration has been one of the most notable and dramatic trends

in U.S. criminal justice in the last 40 years and has given rise to the ‘‘criminology of mass incarcera-

tion’’ (Lynch, 2011), the dynamics of incarceration rate growth have not been experienced uni-

formly across states (Barker, 2009; Bosworth, 2010). Indeed, a disaggregated view of what has been

happening in states reveals a wide range of patterns in incarceration rates. Some states have seen

rapid increases since the 1970s with recent downturns, others have experienced relatively modest

1

Department of Sociology, University of Massachusetts Boston, Boston, MA, USA

Corresponding Author:

Elizabeth K. Brown, Department of Sociology, University of Massachusetts Boston, 100 Morrissey Boulevard, Boston, MA

02125, USA.

Email: elizabeth.brown@umb.edu

Criminal Justice Review

2020, Vol. 45(1) 45-63

ª2016 Georgia State University

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/0734016815627859

journals.sagepub.com/home/cjr

increases and a gradual leveling off, and still others have experienced consistent increases in

incarceration rates over time (Pfaff, 2008, p. 551; Stucky, Heimer, & Lang, 2005, p. 235).

1

Overall,

different patterns of incarceration rates across states have created increasing variation in rates over

time (the standard deviation of incarceration rates across states tripled from 53 in 1977 to 146 in

2010).

Empirical research on state-level incarceration has utilized time-series analysis (Jacobs & Car-

michael, 2001; Stemen & Rengifo, 2011), fixed effect regress ion models (Campbell, Vogel, &

Williams, 2015; Travis, Western, & Redburn, 2014), and case studies of particular states (Barker,

2009; Campbell & Schoenfeld, 2013; Lynch, 2009; Miller, 2008). Overall, prior time series analyses

have suggested that variables such as violent crime rates, drug arrests, and Republican political

power, among others, are positively related to growth in incarceration rates while variables such as

determinate sentencing policies and more generous welfare spending are negatively related to state

incarceration rates (Jacobs & Carmichael, 2001; Stemen, Rengifo, & Wilson, 2005; Stucky et al.,

2005). A recent fixed effects analysis of determinants of incarceration rates in 1970, 1980, 1990,

2000, and 2010, built upon recent case study research that inductively identified distinct periods in

the development of mass incarceration (Campbell & Schoenfeld, 2013), found evidence for peri-

odicity and that partisanship, size of the Black population, and violent crime rates were determinants

of state-level incarceration up until 2000, but that in 2010, only the size of the Black population

remains predictive of incarceration rates (Campbell et al., 2015).

Unfortunately, both time-series and fixed effect analyses produce results that aggregate states

together, obscuring potentially important differences in the determinants of incarceration rate

change across states with widely varying patterns of incarceration rates over time. For example,

Texas experienced relatively moderate and stable incarceration rates through the 1980s (at approx-

imately 200 per 100,000 persons). It was not until the 1990s that rates of incarceration began rising

dramatically after which time rates in Texas began slowly declining. Even with periodicity taken

into account, both time-series and fixed effect approaches obscure those patterns of incarceration

change over time.

Considering determinants of incarceration rates across groups of states aggregated according to

their variable patterns of incarceration rate change can illuminate forces that have contributed to

patterns of dramatic incarceration growth across some states while others have experienced lower

rates and growth. Ultimately, knowing about patterns of incarceration rate growth can inform both

grounded discussions of differential avenues to mass incarceration and opportunities for and barriers

to reducing reliance on incarceration. Using group-based trajectory modeling, this research con-

siders relationships between demographic, economic, political, and policy-specific factors and

incarceration rates for states grouped together according to their pattern of incarceration rate change

in three time periods identified by prior research (Campbell & Schoenfeld, 2013; Campbell et al.,

2015) as being developmentally distinct: from 1977 to 1990, from 1990 to 2000, and from 2000 to

2010.

Toward Refining ‘‘The Criminology of Mass Incarceration’’

The rise of incarceration rates across states beginning in the mid-1970s has been linked to sentencing

and correctional policy shifts assoc iated with politicization and racializ ation of crime rhetoric,

concern about rising crime rates, shifting social mores, and a backlash against perceived liberalism

both in the courts and in social life in the 1960s (Beckett, 1997; Clear & Frost, 2014; Garland, 2001).

Theoretically, research on state-level incarceration has drawn heavily on narratives about the broad

sociopolitical and cultural forces driving the rise of ‘‘mass incarceration’’ (Garland, 2001; Simon,

2007; Tonry, 2009). These narratives, taken together, have been described as the criminology of

mass incarceration (Lynch, 2011, p. 674).

46 Criminal Justice Review 45(1)