A Plague of Prisons: The Epidemiology of Mass Incarceration in America.

• Author: Pfaff, John F.
• Position: Book review

A PLAGUE OF PRISONS: THE EPIDEMIOLOGY OF MASS INCARCERATION IN AMERICA. By Ernest Drucker. New York and London: The New Press. 2011. Pp. xiv, 189. $26.95.

INTRODUCTION

The incarceration rate in the United States has undergone an unprecedented surge since the 1970s. Between 1925 and 1975, the U.S. incarceration rate hovered around 100 per 100,000. (1) Since then, that rate soared to 504 in 2009, dropping only slightly to 500 in 2010. (2) In absolute numbers, the U.S. prison population grew from 241,000 in 1975 to 1.55 million in 2010. (3) Not just exceptional by historical standards, this boom is unparalleled globally: the United States has the highest incarceration rate in the world. Despite having just 5 percent of the world's population, it houses nearly 25 percent of the world's prisoners. (4)

It is not surprising that academics, journalists, and policymakers have attempted to explain the causes of this growth. What is surprising, however, is the general weakness of such explanations. The formal empirical papers that tackle the issue, for example, all suffer from severe methodological shortcomings that fundamentally undermine their results. (5) Many of the common explanations--that prison growth is due to the war on drugs, to parole and probation violations, to longer sentences--are often asserted with little rigorous empirical support. And as 1 have pointed out before, and will continue to do so here, this conventional wisdom is frequently wrong. (6) A recent entry in this discussion is Ernest Drucker's A Plague of Pi4sons: The Epidemiology of Mass Incarceration in America. (7) An epidemiologist by training, Drucker attempts to use epidemiology's tools to shed new light on the complex causal roots of today's mass incarceration problem. While he largely fails in his efforts, he does so in a very useful way: the mistakes he makes are ones that permeate this literature, so that identifying and correcting them serves the broader goal of setting the record straight(er) about the causes and effects of prison growth.

Drucker makes four key points in his book. First, using New York State as a case study, he argues that the war on drugs has been the primary "pump" for prison growth. (8) Second, using the "years of life" metric that epidemiologists employ to quantify the losses from an epidemic, he claims that the prison "epidemic" is on the scale of the AIDS epidemic and other major disasters. (9) Much of the rest of the book is devoted to examining how the prison "epidemic" propagates itself through new arrests tied to the war on drugs (Chapter Seven), the recidivism-enhancing de jure and de facto restrictions that inmates face upon release (Chapter Eight), and the intergenerational transmission of incarceration risk (Chapter Nine). Drucker concludes with some recommendations on how to rein in our current reliance on imprisonment (Chapter Ten).

In this Review, I focus on four major substantive errors that run not just throughout Drucker's book but throughout the prison-growth literature more generally. As I discuss in Part I, Drucker overemphasizes the war on drugs. While Drucker sees it as the major source of growth, drug incarcerations account for only 21 percent of prison growth between 1980 and 2009. Of that growth, 67 percent comes from locking up more violent offenders (51 percent) and property offenders (16 percent). (10) That said, the war on drugs may play an important, albeit indirect, role (via its impact on prior convictions), but one that most commentators, including Drucker, have generally overlooked. And as a result, he makes policy suggestions that will not reduce prison growth and misses those that might.

The second error, discussed in Part II, is linked to the first: by overemphasizing the war on drugs, Drucker underemphasizes the importance of the rise in violent and property crime. Between 1960 and 1991--the heart of the boom in prison populations--violent crime rose by 371 percent and property crime by 198 percent. (11) Such surging crime rates surely played an important role in pushing up prison populations, but Drucker never seriously considers the relationship between incarceration and serious crime.

The third substantive error, discussed in Part III, is that Drucker treats the criminal justice system like a monolithic entity. The criminal justice "system" in the United States, however, is not a "system" at all, but rather a chaotic swirl of local, county, state, and (less frequently) federal actors, all with different constituencies and incentives. Not accounting for these jurisdictional struggles leads Drucker to overemphasize relatively minor explanations and overlook major ones.

The fourth error, discussed in Part IV, is perhaps more procedural. In the second (and more successful) half of the book, Drucker turns his attention to the challenges faced by those released from, or at risk of going to, prison. Prison inmates are drawn disproportionately from communities filled with people living socially fragile lives, and it is inarguable that incarceration exacerbates these frailties. Yet Drucker fails to provide essential baseline comparisons, making it impossible to understand the marginal contribution of incarceration. For example, he states that couples with children are much less likely to get married if one partner has been to prison (p. 143), but he does not adequately address how much of the disparity comes from incarceration, and such an approach implicitly attributes all the observed social harm to incarceration. Without a better understanding of prison's marginal contribution to these social ills, policymakers may overemphasize providing services to prisoners rather than to the community more broadly, or they may underestimate the extent to which the problem is one that affects nonprisoners as well. Drucker's approach may even go so far as to confuse matters of causation, attributing to incarceration broader social ills that may, in fact, drive that incarceration (or the criminal conduct that fuels it).

A common theme ties many of these failings together: Drucker misidentitles the source of the "infection" he wishes to understand. Prison is not the real disease--crime is the disease, and prison is but one potential treatment. But any treatment can harm the patient when misapplied or used improperly, perhaps even leaving the patient more vulnerable to future "illness." But by looking at prison as the disease--as the cancer, rather than the chemotherapy--Drucker overstates the importance of the war on drugs, overlooks the importance of violent and property crime, and overestimates the collateral costs of incarceration.

1. JUST SAY No TO OVEREMPHASIZING THE WAR ON DRUGS

   In 1980, 580,900 people were arrested for drug violations, comprising 5.5 percent of all 10.5 million arrests that year; in 2009, 1.66 million were arrested on drug charges, making up 12.2 percent of that year's 13.7 million arrests. (12) In total, over 38.7 million drug arrests were made between 1980 and 2009 (or 9.5 percent of the total 407 million arrests). In fact, more people are arrested each year on drug charges than are incarcerated for all crimes. These numbers certainly seem to imply that the war on drugs must have played a major role in prison growth. Not surprisingly, many people have made just this argument. (13)

   Drucker unambiguously plants himself in this camp:

   What occurred in New York State to explain [its] surge of incarceration? All signs point to a new set of drug policies, drug laws, and drug enforcement strategies--the Rockefeller drug laws [("RDLs")] of 1973. New York's epidemic of incarceration, which continues to this day, began the year that [the RDLs] came into effect .... In an example of what would ultimately happen across the nation, [the RDLs] proved to be the "pump" responsible for the state's epidemic of mass incarceration. (pp. 50-51) Drucker's core argument is justified by two figures, which I recreate (in improved form) as Figures 1A and 1B below. (14) Figure 1A plots the number of prisoners serving time in New York prisons for drug charges between 1965 and 2008 (with a few missing years due to unavailable data), and Figure 1B the share of New York prisoners serving time for drug charges. Perhaps when viewed quickly, Figures 1A and 1B seem to support Drucker's claim that drug arrests contributed significantly to prison growth, since they show that both the absolute and relative numbers of drug offenders rose sharply during the 1980s; however, they do not support his statement that prison growth in New York "has a well-defined starting point, May 1973, when the new drug laws were put into place" (pp. 52-53), given the decade-long gap between passage of the RDLs and the uptick in drug incarcerations.

   Ultimately, though, Drucker is wrong in at least two important ways-one broad and the other narrow. Broadly, rising drug incarcerations simply did not "pump" up prison growth, either nationally or even in New York. Narrowly, to the extent that drug incarcerations rose faster in New York relative to the rest of the country, it is a mistake to link it simply to the RDLs; other, more serious crime trends played material roles.

   Start with the broad theory. Figure 2, which plots the annual percentage of all state prisoners serving time for drug offenses, indicates that drug incarcerations have not been the primary engine of prison growth. While the share rises strikingly from 6.4 percent in 1980 to almost 22 percent by 1990, nondrug offenders always constitute at least 78 percent of all prisoners, even in 1990, when the share of drug offenders peaked (at 21.8 percent). In 2009 drug offenders comprised only 17.8 percent of all state prisoners. (15)

   Given these relatively low percentages, why are people so quick to blame drug incarcerations for prison growth? Table 1 suggests a possible answer.

   The percentage change in the number of drug prisoners dwarfs that of violent and...