Cumulative impact: Why prison sentences have increased*

• Date: 01 February 2019
• DOI: http://doi.org/10.1111/1745-9125.12197
• Published date: 01 February 2019
• Author: Ryan D. King

Received: 15 February 2017 Revised: 25 September 2018 Accepted: 26 September 2018

DOI: 10.1111/1745-9125.12197

ARTICLE

Cumulative impact: Why prison sentences have

increased*

Ryan D. King

Department of Sociology, Criminal Justice

Research Center, The Ohio State University

Correspondence

RyanD. King, Department of Sociology, 238

TownshendHall, The Ohio State University,

1885Neil Avenue Mall, Columbus, OH 43202-

1222.

Email:king.2065@osu.edu

Fundinginformation

OhioState University

∗It hank SunnyBloomberg, Shawn Bushway,

AlexFraga, Richard Frase, Stefan Ivanov,and

MikeVuolo for assistance with this research.

Iam also g rateful to Dr.Janet Lauritsen and

multipleanonymous reviewers for their helpful

suggestions.The Department of Sociology at

TheOhio State University funded parts of this

research.The Minnesota Sentencing Guidelines

Commissiong raciouslyshared the data for this

article, but the author is solely responsible for

allclaims and any errors.

Abstract

Why has the probability of going to prison after a felony

conviction increased since the early 1980s? Social scien-

tists often try to answer this question through macro-level

research that is aimed at examining correlations between

prison admissions and crime rates or sociopolitical charac-

teristics of states. That type of macro-level inquiry, how-

ever, does not allowfor a close examination of how charac-

teristics of offenders changed over time, and whether such

changes are consequential for understanding trends in the

use of imprisonment. In the current study, I take a differ-

ent approach—one in which case-level data are observed

over a lengthy time span—to investigate why the likelihood

of going to prison for a given crime persistently increased

for several decades. The results of analyses of more than

350,000 felony cases sentenced in Minnesota during a

33-year period show that the probability of a defendant

receiving a prison sentence increased from 1981 to 2013, as

would be expected. The primary reason for the rising prob-

ability of imprisonment was the significant increase in the

average offender's criminal record, which more than dou-

bled during the observation period.

KEYWORDS

criminal records, imprisonment rates, sentencing

1INTRODUCTION

The rise of mass incarceration in the United States since the 1970s is among the most widely studied

topics in criminology, and scholars have proposed several explanations for how and why the United

Criminology. 2019;57:157–180. wileyonlinelibrary.com/journal/crim © 2018 American Society of Criminology 157

158 KING

States became the world's top jailer. Some of this scholarshiphas been focused on the proximate causes

of imprisonment, such as changes in criminal law and sentencing policies (Raphael & Stoll, 2013).

Other scholars have looked at long-run societal changes that set the stage for rising imprisonment

rates, such as the changing political climate (Campbell & Schoenfeld, 2013; Murakawa, 2014), cultural

changes that penetrated deep into the American psyche (Garland, 2001), legacies of racial animus

(Alexander, 2010), and public demand for stiffer punishments (Enns, 2014). As indicated in these

studies, scholars have not settled on a singular cause of U.S. penal exceptionalism. They do agree,

however, that the prison population did not rise simply because more people were committing crimes.

Rather, imprisonment rates swelled during the past four decades because the chance of going to prison,

if convicted, increased substantially (Pfaff, 2017; Raphael & Stoll, 2013: 42).

But why is the probability of going to prison after a conviction higher today than in decades past?

Most researchers have tried to answer this question by way of macro-level analyses of changes in

imprisonment rates across states or countries. Yet research in this mold does not allow for a close

inspection of characteristics of the offending population, which seems to have changed during the

mass incarceration era (see Porter, Bushway, Tsao, & Smith, 2016, on the age distribution). In the

current study, thus, I examine whether the chance of going to prison has increased in recent decades

because convicted defendants changed in an important way: They now have lengthier criminal records

than they did in the past.

Research findings showt hat the proportion of the U.S. adult population with a prior felonyconviction

has more than doubled since the late 1970s (Shannon et al., 2017; Uggen, Manza, & Thompson, 2006)

and that the average criminal record of those entering criminal courts has increased at a similar pace

(Reaves, 2013: 8–9). These changes are intuitive given the expansion of the U.S. justice system, yet

it remains an empirical question as to whether judges and prosecutors (via charging and plea offers)

are sending more people to prison because the offenders in the 2000s have more extensive criminal

histories than did those appearing before them 20 and 30 years ago. The findings from decades of

criminal sentencing studies show that the most powerful predictors of whether a defendant receives a

prison sentence are the severity of the crime and the criminal history of the defendant (Spohn, 2000;

Steffensmeier, Ulmer, & Kramer,1998: 775; Zatz, 2000). No research to date, however, has been aimed

at examining sentencing data over a period of 30 or more years to assess how much the probability of

receiving a prison sentence has changed, and whether a primary culprit for explaining that change is

the thicker RAP sheets of defendants.

In the current research, I examine precisely this issue, and in doing so, I attempt to connect four

dots. First, the historically high crime rate during the 1980s and into the early 1990s combined

with more aggressive prosecution (Pfaff, 2017) to produce a high volume of felony convictions. Sec-

ond, these convictions became part of offenders’ permanent records, and hence, the average crimi-

nal record of sentenced offenders grew over time. Third, the results of sentencing research indicate

that defendants with lengthier criminal records are more likely to be imprisoned than are those with

no criminal history. Fourth, it follows from these patterns that the probability of receiving a prison

sentenced increased over time because the average criminal record of defendants increased since

the 1980s.

I test this hypothesis by analyzing more than 350,000 felonycases sentenced during a 33-year period

from 1981 to 2013 in Minnesota. Similar to macro-level work on imprisonment, these data allow for

an investigation of trends in punishment over a multidecade period. Akin to case-level research, the

data include information on characteristics of defendants, such as the severity of their crimes, racial

demographics, age, and offenders’ criminal histories. Before describing these data and the analysis in

detail, I first put the research question in context bydiscussing three perspectives on why the probability

of imprisonment may have increased.