About this Issue

Mass incarceration has been in the news lately like never before. But how did we get into this mess? Was it our history? Our economic system? Our legal institutions? Comparisons with other countries may help us find an answer, and finding the correct answer will certainly be important if we want to reduce the rate of incarceration.

Joining us this month is Daniel D’Amico, a George Mason–trained economist who specializes in the political economy of the justice system. His thesis is provocative: D’Amico argues that mass incarceration is in part a product of the distinctive institutions found in common law countries. Is he right? Joining us this month to discuss are Mike Riggs, communications director for Families Against Mandatory Minimums; Professor of Criminology Susanne Karstedt of Leeds University; and Adam Gelb, director of the Pew Charitable Trusts’ Public Safety Performance Project.

Lead Essay

Why Nations Jail

Economist Deirdre McCloskey calls the massive increase of living standards that began 200 years ago “the Great Fact.”[1] When looking at the hockey stick–shaped graph of world GDP over time, its importance is obvious. Today’s material wealth is so much greater than the past’s, and it arrived so quickly, that it must play a central role in any theory of social change.

The U.S. imprisonment trend also looks like a hockey stick. Stable and modest growth occurred throughout the early twentieth century. Then in the 1970s, the line shot up, quintupling by the 2000s. Aside from dampening economic optimism, this trend evokes a similar visceral reaction to McCloskey’s. The change in U.S. inmate population has been so big and its accumulation so fast that it can’t be ignored.

David Garland coined the phrase “mass imprisonment” because he thought the phenomenon deserved a name all its own. Like Europe’s “great confinement” of the 17th century or Russia’s “gulag archipelago,” modern America’s mass incarceration appears unique via two defining features, its sheer size and its racially disparate application.[2] The term has become standard parlance, and such “American” features take center stage in most causal explanations. Popular accounts tend to focus on economic conditions; America’s individualist culture; its history of slavery, segregation, and racism; and conservative political preferences regarding prohibition and retribution. To fix incarceration, it’s implied that America must change its conscience and support controls against racial bias, better social programs for the poor, drug decriminalization, and less punitive policing.

Culture, racism, and the drug war obviously matter. But I’d like to challenge some of these presumed causes by simply asking how much they fully explain global patterns, and how much their implied reforms would substantially reshape the trends. I argue that popular narratives have largely overlooked some factors that substantially contribute to prison growth.

In economic theory, the “Great Fact” has received an appropriately international analysis.  Yet I worry that prison discourse is too focused on the United States, when in fact incarceration transcends the American experience. Most countries have experienced growth. Ecuador, Indonesia, Cambodia, Israel, Serbia, and Georgia don’t share much economic, partisan, or cultural American-ness, yet all doubled their prison populations in a decade, while Britain took three.[3] An analysis solely of U.S. idiosyncrasies can’t explain global or historic patterns. Academic research has developed empirical evidence and nuanced theories for why some nations imprison more than others, but media commentaries lag behind, and their oversimplifications carry practical consequences.

Is American-style capitalism, with its supposed inequality and market volatility, partially to blame for large inmate populations? Maybe somewhat. Economic factors like unemployment, welfare spending, and union power correlate with cross-country imprisonment rates,[4] but causation cannot be discerned. Many studies have investigated this popular view with ambiguous results.[5] In short, no data confirms a consistent relationship between more liberal market economies – or higher economic performance – and larger prison populations. In fact, more economically free countries host less homicide and yet have better crime reporting.[6]

Is racism the primary driver of American mass imprisonment? While U.S. history is undoubtedly plagued by the legacy of slavery, and significant bias still infects our system, these experiences don’t much explain why other nations host similarly disparate prison populations. Or, sometimes, worse. From the peasant classes of ancient Greece,[7] to Romani inmates in today’s Czech Republic, or to Turkish prisoners in Germany, in each context one finds a greater proportion of the socially disadvantaged than is found in their general society. In fact, in England, Canada, and Australia, the minority to white inmate ratios all outpace the United States.[8]

The advent of American mass incarceration also occurred alongside measurable racial progress. The relationship between race and imprisonment is not clearly causal from the former to the latter. The facts suggest more about the racial effects of imprisonment than they do how race drives growth in the prison population.

Is America’s vengeful culture responsible? Again maybe partially, but such features don’t explain why so many other nations have multiplied their prisons without similarly rugged individualists supporting Nixon- or Reagan-ite tough on crime policies. While measured U.S. opinions do seem more punitive than in the past,[9] less is known for cross-country comparisons. Most experimental evidence suggests vengeful preferences are common across identities,[10] and economist Naci Mocan reports the opposite cultural relationship, with poor, war-torn, and collectivist countries hosting more vengeance.[11]

The shared timing of American incarceration with drug prohibition seems too tightly linked to be coincidence. But prohibition is not a sufficient explanation for U.S. rates, let alone for global patterns. Here’s why.

First, when the inmates of asylums are included in imprisonment statistics, the trend line less conspicuously resembles a hockey stick.[12] Second, drug violations only make up about 17% of state inmates and represent about 20% of the growth since 1980, with current admission rates matching.[13] Without drug convictions, American incarcerations would have quadrupled rather than quintupled. Perhaps this underestimates the criminogenic effects of prohibition, but multiplicative growth seems to stand apart from drugs.

And again, if uniquely American factors cause mass incarceration, then why is there a global pattern?[14] The majority of popular explanations merely push the question back a stage. If American-style trends are to serve as general explanations for global imprisonment patterns, then other countries should share similar policy and cultural histories in proportion to their incarceration rates relative to the United States. What about the United States and other prison growth nations allowed for these phenomena to occur in the first place?

The common narratives are not wrong per se, but likely incomplete. Current research emphasizes a general relationship between social institutions and incarceration. Nations with similar political, economic, and cultural systems tend to host similar criminal justice procedures and similar prison populations.[15] Remaining debates largely surround how best to organize nations into institutional categories that make sense of the incarceration pattern. In short, what particular institutional types shape incarceration and how? Most tend to emphasize that social changes such as employment cycles, public opinions, or partisan trends shape class and racial tensions. Less appreciated is how the organizational patterns of institutions may relate to incarceration outcomes regardless of the particular interests or motivations of voters or policymakers.

One can understand this missing factor of organizational influence by thinking comparatively about American drug prohibition. Legalized narcotics are extremely rare. You can’t walk down the streets of Germany, Sweden, or England and sell heroin without getting arrested. Most countries prohibit drugs, but only America launches an ominously but fittingly titled “war on drugs.” It is not so much that we prohibit drugs, but rather how we finance and manage that prohibition, which sets us apart. I believe that America’s drug war, American criminal justice services more generally in recent decades, and those criminal justice systems that have behaved similarly, are all united by how much more power they afford to the national as opposed to local levels in criminal justice decisionmaking. And the result is mass incarceration.

Theory shows that more hierarchical organizations commit more errors of overidentification.[16] Increases in criminal legislation, arrest rates, convictions, and sentence lengths would all seem to be relevant manifestations. Similarly, many public choice scholars have noticed that by concentrating perceived deterrent benefits while dispersing costs, democratic politics rewards the expansive spending, employment, and voter appeasement accomplished through criminalization and prison growth.

Cross-country empirics also support the idea that organizational factors are at play. Nicola Lacey[17] has observed the organization of electoral processes correlate with incarceration rates. Nations with winner-take-all elections host greater political incentives to appease punitive biases. They also have more difficulty coordinating voting blocks for reform. In contrast, plurality systems foster more inclusion of minority interests.

Similarly, my recent paper coauthored with Claudia Williamson shows nations founded in the British common law tradition rather than civil law host larger incarceration rates. While the common law is typically more decentralized, we suspect criminal justice systems were historically founded and subsequently organized more hierarchically relative to other common law social sectors.[18] Furthermore this concentration of national power was exaggerated in the twentieth century.[19]

Just as market economies require decentralized networks of individual actors for technological innovation to occur, so too do legal norms and public policies likely benefit from competitive-like arrangements. When criminal justice authority is held at the local level, as is more common around the world and throughout history, authorities are better attuned to the local demands and concerns of citizens. Yet in our system, the typical response to perceived local biases in criminal justice is to further empower federal oversight. 

When local police forces multiply rapidly and are equipped with military style technologies thanks to federal financing, there are often unintended consequences. Though motivated by concerns for the disadvantaged, there is no practical guarantee that national power won’t also have structural biases. Yes, local cops, judges, and juries can be racist, but coping with better-funded riot police and swat teams is nonetheless a burden on poor minorities. From Ferguson to Baltimore we seem to be stuck with the worst of both worlds: racially biased local cops and a militarized national response.

Further, changes in social morality, public opinion, or traditional policy initiatives do little to reform the organizational patterns of institutions. Politics doesn’t typically aim at, nor is it capable of, fundamentally reshaping the political or legal process.

That’s a problem, because measured correlations between organizational factors and prison populations seem big. In our study, we initially merely sought to give descriptive magnitudes for relevant variables associated with incarceration.  A standard deviation decrease in an obvious factor like homicides (about 12.5 per 100,000 citizens) was associated with approximately 48 fewer inmates per 100,000 citizens, whereas French civil law countries hosted approximately 76 fewer inmates per 100,000 citizens than common law nations. The organizational features of institutions don’t just matter, they seem to matter a lot.

Second, institutional organization shapes incentives and thus may serve a more foundational role than standard causes. Suppose mass imprisonment occurred not because Americans were so morally outraged about drug use or fearfully retributivist regarding violence, but rather because policymakers and citizens were structurally encouraged to embrace legislation, enforcement, and spending. Legalizing particular drugs won’t change this general incentive structure; it will only divert it. With similar incentives in place, future increased crime rates or headline cases could reignite the imprisonment cycle, albeit under some other form of crime.

Policies that ignore structural influences may thus fall short. Worse, by expending real resources or by enhancing national authority they may add insult to injury. More accurate theories of incarceration may be needed to produce more effective reforms.



[1] D. McCloskey 2011. Bourgeois Dignity: Why Economics Can’t Explain the Modern World. Chicago: University of Chicago Press.

[2] D. Garland (ed) 2001. Mass Imprisonment: Social Causes and Consequences. London: Sage Publications.

[3] R. Walmsley 2010. “Trends in World Prison Population,” in Harrendorf, Heiskanen, Malby (eds) International Statistics on Crime and Justice. Washington, DC: United Nations Office on Drugs and Crime, notes prison growth in 68% of 144 nations from 1997-2007.

[4] T. Chiricos, M. Delone 1992. “Labor Surplus and Punishment: A Review and Assessment of Theory and Evidence,” Social Problems 39, 421-46; B. Western, K. Beckett 1999. “How Unregulated is the U.S. Labor Market? The Penal System as a Labor Market Institution,” American Journal of Sociology 104, 1030-60.

[5] J. Neapolitan 2001. “An Examination of Cross-National Variation in Punitiveness,” International Journal of Offender Therapy and Comparative Criminology. 45: 691-710, finds “[l]ittle to no support… for civilization theory, inequality, or unemployment variables.” J. Sutton 2004. “The Political Economy of Imprisonment in Affluent Western Democracies, 1960-1990,” American Sociological Review 69, 170-189, finds high labor force participation positively and significantly impacts incarceration rates.

[6] E. Stringham, J. Levendis 2010. “Chapter 6: The Relationship between Economic Freedom and Homicide,” in Gwartny, Hall, Lawson (eds) Economic Freedom of the World: Annual Report. Vancouver: Fraser Institute; R. Soares 2004. “Crime Reporting as a Measure of Institutional Development,” Economic Development and Cultural Change 52: 851-71.

[7] D. Allen 1999. The World of Prometheus: The Politics of Punishing in Democratic Athens. Princeton: Princeton University Press; D. D’Amico 2010. “The Prison in Economics: Private and Public Incarceration in Ancient Greece,” Public Choice. 145(3-4): 461-82.

[8] M. Tonry 1994. “Racial Disproportion in U.S. Prisons,” British Journal of Criminology. 34(5): 97-115; M. Mauer 2003. “Comparative International Rates of Incarceration: An Examination of Causes and Trends: Presented to the US Commission on Civil Rights,” Washington, DC: Sentencing Project.

[9] P. Enns 2014. “The Public’s Increasing Punitiveness and Its Influence on Mass Incarceration in the United States,” American Journal of Political Science. 58(4): 857-72.

[10] D. Houser, E. Xiao, K. McCabe, V. Smith 2008. “When Punishment Fails: Research on Sanctions, Intentions and Non-Cooperation,” Games and Economic Behavior. 62(2): 509-32; F. Guala 2012. “Reciprocity: Weak or Strong? What Punishment Experiments Do and Do Not Demonstrate,” Behavioral and Brain Sciences. 35(1): 1-15.

[11] N. Mocan 2013. “Vengeance,” Review of Economics and Statistics. 95(3): 969-82.

[12] B. Harcourt 2011. “An Institutionalization Effect: The Impact of Mental Hospitalization and Imprisonment on Homicide in the United States, 1934-2001,” Journal of Legal Studies. 40(1): 39-83.

[13] J. Pfaff 2015. “The War on Drugs and Prison Growth: Limited Importance, Limited Legislative Options,” Harvard Journal on Legislation. 52: 173-220.

[14] When statistical techniques are used the U.S. is not detected as an outlier. D. D’Amico, C. Williamson 2014. “Do Legal Origins Affect Cross-Country Incarceration Rates?” Journal of Comparative Economics, forthcoming. 

[15] M. Cavadino, J. Dignan 2006. Penal Systems: A Comparative Approach. Los Angeles: Sage.

[16] R. Sah, J. Stiglitz, (1986). “The Architecture of Economic Systems: Hierarchies and Polyarchies,” The American Economic Review. 76(4): 716-27.

[17] N. Lacey 2008. The Prisoners’ Dilemma: Political Economy and Punishment in Contemporary Democracies. Cambridge: Cambridge University Press.

[18] H. Berman 1983. Law and Revolution: The Formation of the Western Legal Tradition. Cambridge: Harvard University Press.

[19] W. Stuntz 2011. The Collapse of American Criminal Justice. Cambridge: Harvard University Press.

Response Essays

Better Question: How Do We Unjail?

I can’t speak to Dan’s research about the carceral legacy of British common law versus civil law (other than to say it seems plausible!), but I can, as an advocate for sentencing reform who works in D.C., vouch for his claim that concentrated benefits and dispersed costs are just as distortive in criminal justice policy as they are elsewhere in society, particularly since safety is one of only a handful of public goods every U.S. politician, regardless of party or ideology, feels obligated to deliver to his or her constituents. As they should, both as a matter of substance (i.e., it’s their job), and politics (i.e., it helps them keep their jobs). Americans will throw a politician out if it appears he or she has left them vulnerable to crime; they almost never throw a politician out for being overly punitive on their behalf.

Because public safety is such an important public good, the question of how mass incarceration came to be is less pressing to me (as Dan’s opening essay argues, we have many leads, but no one answer) than the more immediate question of how we can maximize the efficiency of the criminal justice system such that it delivers public safety while minimizing the cost as measured both directly in tax dollars and indirectly in the liberty of our errant neighbors.

To that end, I propose what I concede might be a very unlibertarian thought experiment: For a moment, imagine that our historically high incarceration rate – more than 700 per 100,000 – is not bad per se. After all, when we talk about the U.S. incarceration rate, it is usually in the context of our own national history and comparative politics. We say our current incarceration rate is “bad” because it is higher than it used to be, and it is doubly bad because it is higher than the incarceration rates of both the free countries of Western Europe and the oppressive regimes of China, Russia, Iran, Cuba, and Venezuela (and many others). But none of those comparisons tells us exactly what our rate should be, or whether the rates in those countries are prima facie better or worse than ours.

In this exercise, it makes more sense to approach incarceration by asking how we can incarcerate as much as we need to, but neither more nor less. I have never written this previously, but federal law actually provides a pretty good guide. Specifically, 18 U.S. Code § 3553 – the beginning of it, anyway – titled, “Factors to be considered when imposing a sentence,” says criminal sentences should serve the following goals:

reflect the seriousness of the offense, to promote respect for the law, and to provide just punishment for the offense;

afford adequate deterrence to criminal conduct;

protect the public from further crimes of the defendant; and

provide the defendant with needed educational or vocational training, medical care, or other correctional treatment in the most effective manner;

History suggests it’s unlikely that Congress has given us the perfect framework here. In fact, these four ideas actually date back to the Enlightenment, when philosophers like Cesare Beccaria proposed that criminal punishment serve a purpose beyond terrifying the poor into perpetual obedience.

So, while Congress has adopted a centuries-old framework by which we can determine “good” sentencing practices, our current habits and policies defy all four of these goals. For instance: a sentence of life without parole for drug “trafficking” – often a dysphemism for distribution of illegal drugs above an arbitrary weight threshold – is too often far more severe than the crime warrants; it does not deter drug trafficking; it protects no one if the offender was nonviolent (under all but the most preposterous theories of what constitutes a “violent crime”); and it provides no incentive for either a correctional facility to rehabilitate an offender or for the offender to seek rehabilitation.

What we know, then, is one simple, undeniable truth: incarceration often fails to serve any of the traditional purposes of criminal sentencing, and those purposes can often be me met by sanctions other than prison. The recognition of that fact alone should be sufficient for policymakers and the general public to question our reflexive reliance on incarceration as the country’s dominant method of punishment and crime control.

If only it were that simple.

The major challenge to ending unnecessary incarceration is not proving to legislators that “mass incarceration” is inherently bad (legislators who oppose reform claim confidently, and not unreasonably, that people who obey the law deserve to be safe more than people who break the law deserve to be free). Nor is the major challenge proving to policymakers that our prison population is too large (a claim that makes sense only in comparison to the historical record and the experiences of other countries, but doesn’t – and can’t – tell us how large our incarcerated population should be). The major challenge is demonstrating to legislators what decades of evidence shows conclusively: that safety and justice are not at odds, but are complementary to one another.

D’Amico’s essay – and his larger research program – offer valuable insight into how we got here. And that matters, both theoretically and for practical purposes. But for the 2 million people in America’s jails and prisons, and their family members, figuring out how to get where we need to be is far more valuable information.

Laboratories of Incarceration

Diagnosing how the U.S. incarceration rate ballooned to 1 in every 100 adults, and prescribing remedies to shrink it, are both frustrated by one central fact: The “system” is actually 51 different systems, each almost entirely independent and driven by its own legal, political, cultural, sociological, and organizational factors. The federal system plays an outsize role because it is the largest and gets the most national media attention. But the 50 states house nearly 90 percent of prisoners, to say nothing of hundreds of thousands of inmates in local jails, and—as the laboratories of incarceration—they must be the main subject of analysis.

Each state has its own criminal code that controls how offenders are sentenced and its own budget that determines the availability of alternatives to incarceration. The variability among states on these and other critical elements of the criminal justice system is breathtaking. In one state, driving with a suspended license is one of the top ten crimes for which people are admitted to prison; in another state, it doesn’t lead to a prison term at all. In some states, more than half of prison admissions are for failing drug tests, missing appointments with probation officers, or other technical violations of the conditions of community supervision. In others, violators are held accountable in day reporting centers or similar noncustodial programs, and almost no one goes back to a state prison cell for rule infractions.

These variations in sentencing and post-release supervision give rise to vast differences in rates of incarceration. At the end of 2014, Louisiana had the highest portion of its population behind bars, with 1,072 per 100,000 adult residents in prison, while Maine was at the opposite end of the scale, with 189 per 100,000 adults. Even geographic proximity is no guarantee of similar rates: North Carolina and South Carolina’s rates were 465 and 552 per 100,000, respectively, while South Dakota at 558 per 100,000 doubled North Dakota’s 278.

Differences in crime rates account for only a small share of the inconsistencies between these pairs of states. The primary explanation lies in the myriad policies, programs, and practices that are set at the state level and interpreted by local prosecutors, defense attorneys, judges, corrections officials, and parole board members. The adoption, enhancement, or repeal of mandatory minimum sentences, for example, is highly dependent on the strength of the political forces on both sides of the debate. If a mandatory sentencing law is in place, local attitudes toward the policy, availability of time on court dockets, and opportunities for alternative programs are major determinants of the extent to which a mandatory prison term will be enforced—or set aside as part of a plea negotiation.

This is not to say there are no national policies or norms that influence how the punishment pendulum swings. There’s little debate that fears surrounding urban unrest in the 1960s and 70s helped cause the rapid growth of U.S. incarceration that Mr. D’Amico describes. The trend was accelerated by the dismantling of large and outdated psychiatric facilities without a commensurate expansion of community services, leaving jails and prisons to house and feed thousands of the mentally ill. In the 1980s and 90s, new federal laws and sentencing guidelines substantially raised the likelihood and length of prison terms for federal offenders and offered billions in grants to states that required violent offenders to remain behind bars for at least 85 percent of their court-ordered sentences.

The new federal statutes transformed the Bureau of Prisons from a sleepy sliver of the Justice Department to a behemoth that chews up nearly one quarter of the department’s budget. But even without the eightfold increase in the federal prison population since 1980, the U.S. incarceration rate would lead the rest of the world many times over. And on the state level, responses to the federal incentives and the “get tough” rhetoric of political campaigns were far from uniform. Between 1995—the first year federal prison-building incentives became available—and 2005, prison populations in five states grew less than 10 percent while jumping by more than 100 percent in four others. The remaining states were spread out evenly in between.

If national phenomena explain only a small part of the explosion in the incarceration rate, they explain only a little more of the recent decline. But from the data we do know this: After reaching a peak of 1 in 100 adults behind bars in 2007–08, the overall rate, which includes local jail inmates as well as state and federal prisoners, has dropped steadily each year since, falling to 1 in 110 adults at the close of 2013. During those five years, 30 states cut their imprisonment rates, while rates in 15 states continued to grow and five were flat.

Many of the states that stopped or reversed their prison growth did so after undertaking an extensive analysis of what was causing the higher incarceration rates in the first place. The reasons then became targets for reform. For many policymakers involved in the reform efforts, one of the surprising drivers of the growth in the prison population was the high rates of incarceration for supervision violators. Specifically, there was a fundamental misalignment of authority and responsibility between state and local governments. In many states, the state government runs and pays for prisons, while local government runs and pays for probation. So if a probationer is causing trouble, local officials can relieve themselves of the cost of supervision—and any potential political fallout—by revoking community supervision and sending the offender to state prison. This dynamic isn’t about punitive attitudes, puritanical culture, or racial bias. It springs straight from the chapters of a textbook on levels of government, separation of powers, and bureaucracy.

Searching for overarching explanations for why the “land of the free” locks up more of its citizens than any other nation is an important exercise. But the size of prison populations the world over is determined by only two factors: the number of people who come into the prison system and how long they stay. In the United States, those factors are determined largely by state policy choices, behind each of which is a unique and tangled story.

Great facts and new facts: The end of U.S. mass imprisonment?

Why do nations jail? It’s a rich man’s folly, says Jan van Dijk from Tilburg University in the Netherlands. He demonstrates this with a neat graph, which shows a straight line between the imprisonment rates in relation to homicides and the GDP of a large global sample of countries: The richer countries are, the more they use imprisonment when meting out punishment to citizens. Similar results are obtained by researchers who use the United Nations Human Development Index: the more developed nations, which also have a higher GDP, have higher imprisonment rates than poor and less developed countries. Jan van Dijk has a point even from a historical perspective; prison became the dominant tool of crime control in the nineteenth century when European states became more affluent and could count on a steady stream of (tax) income. Obviously, imprisonment is a tool of criminal justice not affordable to all and sundry.

In most global comparisons of imprisonment the United States is a visible outlier. It not only stands out among its peers, the rich and developed countries, with its “historically unprecedented and internationally unique” experiment in mass incarceration housing 25% of the world’s prisoners. Imprisonment in the United States also by far exceeds imprisonment in the poorest and cruellest dictatorships in the world, even if we concede that these have other ways of dealing with citizens and their crimes and provided their data are credible. The unprecedented growth and unparalleled size of the U.S. prison population between 1970s and the first decade of the 21st century attracted an enormous amount of attention among criminologists. Was this just another case of American exceptionalism, or was the United States in the vanguard of a development that would become infectious globally? Most criminologists bought into the latter perspective. As a “hegemonic punitive worldview” took hold of them they saw “no escape from a punitive future,” neither for the United States nor elsewhere. [1]

This was the start of the search for the one magic-bullet variable that could explain why nations generally jail, and why the United States does so extraordinarily and massively. The emerging candidate was broadly labelled “neo-liberalism.” Neo-liberalism’s manifestations of deregulating the economy and downsizing the welfare state were at the core of most accounts of penal systems and the punitive turn across the globe, in the United States, Latin America, Australia, New Zealand, and Europe with the widely accepted argument that “the rich get richer and the poor get prison.”[2] Another master narrative was based on the observation that nations with majority rules in their electoral processes have higher imprisonment rates. In both cases, the empirical foundations were shaky: a small number of highly selective country cases were used, imprisonment rates differed much less than the theory would have required, and evidence to the contrary was ignored, as e.g. the fact that neither Reagan as governor of California or Margaret Thatcher as Prime Minister of the United Kingdom had overseen increases in imprisonment during their time in office.

And the search goes on. Do civil law systems have in-built restraints, as Daniel D’Amico proposes? A global sample would include a large group of well-off European countries and another group of developing Latin American countries both with low imprisonment rates but for very different reasons. The European countries in particular stand out with a regional regime of oversight and care for prisoners, which considerably increases costs per prisoner, and in most of them prisoners have the right to vote (notable exceptions are the United Kingdom and a few post-Soviet states). It is however worthwhile to have a close-up on civil law systems. They have indeed in-built restraints and moderating elements which insulate and distance criminal justice from direct democratic accountability and community pressure that has ratcheted up U.S. imprisonment rates. American criminologist Frank Zimring argues that such “professionalization of punishment” combines several “leniency vectors” that keep imprisonment low.[3] It is a defining feature of civil law systems that decisions are made by professionals and according to professional standards, and criminal justice officials are not directly elected. In contrast, communities and the electorate seem to have been a driving force behind skyrocketing imprisonment in the United States. During the past decades they were in ardent opposition to “big government” criminal policies that were seen as ignoring community demands and demonstrating undue sympathy for and “inappropriate leniency” toward the offender. [4] Voters in California supported criminal justice policies that lay the ground for ever increasing imprisonment in that state;[5] they finally turned around in 2012, when they prudently decided that they would not spend a tax increase on ever more prisons and prisoners. D’Amico’s own findings definitely rule out the local level as a panacea against high imprisonment rates.

Criminologists have long known that factors that work when comparing countries, communities, and neighbourhoods do not explain changes over time. Thus the factors that explain why nations jail might fail when it comes to explaining the “hockey stick” shaped trend of U.S. mass imprisonment. The year 2009 provides a reality check for all theorizing on mass imprisonment in the United States: It was the year when imprisonment growth in the United States came to a halt, and when the numbers started to decrease after nearly four decades of relentless growth, a downward trend that has not stopped up to date. California led the U-turn way with more than 4,000 prisoners less, and half of the states have followed suit.[6] As a robust signal of an end to the trend of mass imprisonment, prisons were closed in California, Nebraska, and New York. In 2012 28,500 prison places were either closed or under consideration for closing with estimated savings of nearly $340 million in the first year alone.[7]

2009 was the year after the financial crisis, and this is no coincidence. The 2008 financial crisis was related to a severe fiscal crisis which was recognized in California and other states long before; New York repealed legislation that had extended prison sentences across a range of offences and offenders as early as 2003. Fiscal and cost arguments played an important role in these changes and continue to do so.[8] It comes as a surprise for many criminologists that the same neo-liberals and fiscal hawks who had been blamed for “mass imprisonment” were now in the driving seat for penal reform. Prominent Republican Newt Gingrich now saw an “urgent need to address the astronomical growth in the prison population, with its huge costs in dollars and lost human potential,” and concluded that “conservatives must lead the way to fix it.”[9] “Right on Crime,” a conservative think tank, explains the turn in conservative penal policies: “How is it ‘conservative’ to spend vast amounts of taxpayer money on a strategy without asking whether it is providing taxpayers with the best public safety return on their investment?”[10] “[Conservatives] have changed their minds about what prison means. Prisons increasingly stand for big government waste …”. [11]

Elizabeth K. Brown tested two sets of factors that were associated with states’ legal reforms that reduced “reliance on incarceration”: budget pressures and the presence of Republicans in the state legislature and the governor’s office. Budget pressures increased the inclination of the state legislature to pass such legal reforms. In contrast, the amount of Republican seats reduced the efforts and success of such reforms.[12] As it turns out bipartisanship is decisive: As neither of the two parties wants to be seen as “soft on crime,” there is a huge incentive for bipartisan agreement, which is furthered by a strong presence of Democrats in state legislatures. Change is initiated and gains momentum from “what strong, ideologically defined partisan activists and politicians come to believe is their own, authentically conservative or liberal position.” Information on the failure of mass imprisonment besides its tremendous costs is not seen as threat but rather as a confirmation of political and faith-based identities and beliefs.[13] When conservative activists and politicians enter unexpected alliances, and readjust their positions, and when bipartisan politics are increasingly important, it is time for criminologists to re-think their theories of U.S. mass imprisonment.

If two trends look like hockey sticks there might be some common ground which kicks them off. Economist Deirdre McCloskey identifies the values of human dignity and liberty, and the revaluation of the status of the middle classes, as decisive drivers of the unprecedented growth of GDP across the globe.[14] My own research shows that values like liberal individualism and egalitarianism are not related to how many are sent to prison, but significantly to the treatment of prisoners.[15] Both might be nonetheless related: more care for the dignity and liberty of those who failed might make nations more cautious in the use of imprisonment, not the least because it increases the costs of imprisonment. In June 2015 a group of criminal justice practitioners, politicians, and criminologists, including a democratic governor and chief prosecutor from the United States visited Germany to see “how Germany does prison”. What they brought home was the need to “fundamentally rethink values” with an emphasis on the protection of and respect for the human dignity of prisoners, which not only is part of the German Basic (Constitutional) Rights, but also animates the Eighth Amendment of the U.S. Constitution that bans cruel and unusual punishment.[16] A discourse on human dignity and “second chances” for prisoners has started across the political spectrum; these represent fundamental American values, which need to be regained in criminal justice, and thus lead a revaluation of its citizens in prison. Many European countries, including Germany, have considerably lower recidivism rates then the United States. Spending on prisoners rather than on warehousing them seems a better way to deliver the goods.

Among the delegates was a district attorney whose father had narrowly escaped death in the Bergen-Belsen concentration camp as a five-year-old. Seeing German prisons in 2015 he was confident that “countries can change.”[17] Let us not forget that it was the United States that set post-war Germany on this trajectory of change. The people of the United States and their politicians have an amazing and unparalleled capacity for change; half a century for the road from segregated buses to an African-American in the White House is a short time span. I am confident that in 50 years’ time mass imprisonment in the United States will be a distant memory.



[1] S. Listwan, et al 2008. “Cracks in the Penal Harm Movement: Evidence from the Field” Criminology and Public Policy 7: 423-465, at 423

[2] M. Cavadino, J. Dignan 2006: Penal Systems. A Comparative Approach. London: Sage; J. Reiman 2000. The Rich Get Richer and the Poor Get Prison. Oxford, New York: Routledge

[3] F. Zimring, D. Johnson 2006. “Public Opinion and the Governance of Punishment in Democratic Political Systems.” Annals of the American Academy of Social and Political Science, 605: 266 – 280, at 274  

[4] F. Zimring, G. Hawkins, S. Kamin 2001. Punishment and Democracy: Three strikes and you are out in California. Oxford: OUP, 231

[5] V. Barker 2009. The Politics of Imprisonment. Oxford, New York: OUP

[6] Pew Center on the States (2010): Prison Count 2010. Washington DC

[7] N. Porter 2012. On the Chopping Block 2012: State Prison Closings. The Sentencing Project, Washington DC

[8] J. Greene, M. Mauer 2010. Downscaling Prisons. The Sentencing Project. Washington DC

[9] Quoted in D. Dagan, S. Teles 2012. “The Conservative War on Prisons.” The Washington Monthly, November/ December 2012 http://www.washingtonmonthly.com/magazine

[11] Dagan, Teles

[12] E.K. Brown 2013. “Foreclosing on Incarceration? State Correctional Policy Enactments and the Great Recession.” Criminal Justice Policy Review 24: 317-337

[13] Dagan, Teles

[14] D. McCloskey 2011. Bourgeois dignity. Why Economics Can’t Explain the Modern World. Chicago: University of Chicago Press.

[15] The US is not included in the data on prison conditions as these are collected by the US State Department; S. Karstedt 2011. “Liberty, Equality and Justice: Democratic Culture and Punishment.” A. Crawford (ed.): International and Comparative Criminal Justice and Urban Governance: Convergence and Divergence in Global, National and Local Settings. 356-385.

Cambridge: CUP

[16] N. Turner, J. Travis 2015. “What we learned from German Prisons.” The New York Times, 7 August 2015 . http://www.nytimes.com/2015/08/07/opinion/what-we-learned-from-german-p… (31.8.2015)

[17] M. Chammah 2015. “How Germany does prison”. https://www.themarshallproject.org.2015/06/16/how-germany-does-prison. (31.8.2015)

The Conversation

The Many Factors Leading to Mass Incarceration

I first want to thank the other contributors for replying to my essay. I’ve now had the time to read each response and learned much in the process. I merely offer some brief comments regarding each in turn below.

Mike Riggs agrees with my observation that the criminal justice system, like many other forms of bureaucracy, has the tendency to concentrate benefits and disperse costs. Thus punitive outcomes are often unhinged from their practical intentions. The criminal justice system instead can foster waste, domination by private interests, rent-seeking, corruption, and the like. Mike proceeds to note correctly that such does not prove the U.S. system is excessive relative to other nations, or that other national levels are preferable to our own. I agree.

This context dependency is one of the major implications of my previous study co-authored with Claudia Williamson. Different nations with different institutional patterns face different costs and benefits to punishment via imprisonment. The strategies for social control that prove effective in one environment may not work for another. If other nations rely upon forms of social control other than prisons, it does not necessarily follow that the United States can efficiently reduce its prison population by simply substituting for these other social controls. Other social controls carry costs, and how those costs will manifest in the U.S. experience is uncertain.

First, common law countries have far fewer bureaucracies dedicated to providing social welfare services like housing, health care, and employment assistance relative to civil law nations. Insofar as it is these types of infrastructural bureaucracies that are leveraged as social controls, it is not known how much investments in similar bureaucratic structures will cost in the American context.

Substituting away from incarceration within a common law environment may require large upfront costs to instigate and establish comparable bureaucracies to provide social monitoring and enforcement in lieu of imprisonment. In addition to the direct costs of financing new bureaucratic infrastructures, such institutional forms can be argued to correlate with weaker economic performance and a greater potential for corruption, especially when implemented in nontraditional contexts like the demographically and economically diverse population of the American citizenry.

In other words, without denying the potential for failures from the current American prison system, some enduring legacy of larger incarceration rates may be inevitable or dare one say socially optimal for the United States or other common law countries relative to civil law counterparts.

Sadly we know a lot more about what does not sufficiently explain incarceration patterns than we know for certain about what does. We have good theory to suggest that some amount of greater levels are to be expected in common relative to civil law countries, but we don’t know precisely how much more imprisonment is necessarily warranted from these organizational differences. We lack objective criteria for discerning inefficiently excessive rates from an innocuously stable equilibrium from currently observed cross-national patterns.

Mike writes pointedly, “[w]hat we know, then, is one simple, undeniable truth: incarceration often fails to serve any of the traditional purposes of criminal sentencing, and those purposes can often be met by sanctions other than prison.” Again, unfortunately, we know less about what the relevant costs of implementing “other sanctions” in the United States entail and how they would compare to the status quo. We also don’t know what purposes or functions incarceration may be currently servicing apart from those explicitly declared in the rationale of legislation. I agree that asking “how do we unjail?” is perhaps better and more challenging than “why do nations jail?” but the former inquiry is largely unknowable without the latter.

In my opening essay, I referred to Nicolai Lacey’s focus on voting systems along with my own research with Claudia Williamson on legal origins, both as examples of institutional theories highlighting the importance of organizational structures. Basic organizational theory suggests hierarchically structured systems tend to suffer more errors of overidentification than decentralized systems.


I suggest that some degree of the cross-country variation of incarceration rates can be understood in this light. Despite Adam Gelb’s correct observation that, “[e]ach state has its own criminal code that controls how offenders are sentenced and its own budget that determines the availability of alternatives to incarceration,” it seems that substantial changes in the balance of criminal justice decisionmaking between the federal and local levels can help explain a substantial facet of these global trends. In other words, even though the American system has historically been and remains today organized according to some significant degree of decentralized decisionmaking, the modern surge of mass imprisonment and a variety of other identified failures in the criminal justice system can be linked to these organizational patterns becoming on the margins less localized and more governed and shaped by federal level authority.

My suggestion would be to focus more upon the relative changes in institutional organization that have occurred in the American criminal justice system, rather than to focus upon its organizational pattern in absolute terms. I admit there is no clear or objective ability to identify what degree of decentralization is optimal for the American system. But it still seems reasonable and accurate to notice that the role of the federal government in criminal justice decisionmaking radically expanded in the twentieth century and thus displaced a significant level of local autonomy. In general, I think popular narratives placing primary blame upon Republican-dominated tough on crime campaigns should be substantially broadened to recognize that many of the seeds of today’s mass incarceration were in fact sown much earlier amidst the progressive era and the New Deal. Small townships previously independent from federal legislation or financing were newly equipped with more formally managed and financed resources than previously.

By the time the war on drugs was launched in the 1970s most states and local townships had ceded significant authority to the federal legislature and increased their dependence upon federal financing. Complying with federal policies was that much more a political and social inevitability. At the global level, nations similarly walking away from locally autonomous criminal justice sectors in favor of more nationally managed controls seem to share proportionate placements within the imprisonment spectrum.

Yes, the U.S. national prison population rate cannot be understood without some due attention to the amalgamation of different state incarceration processes that comprised it. But it is also reasonable and true to notice that a substantial majority of the states began to incarcerate similarly as their relevant decisions were more heavily shaped by federal incentives.

Case in point, much of the national trend is driven by states like California, Texas, Louisiana, and Florida. All of which as border states have a challenging time dealing with key national immigration and prohibition policies. Louisiana in particular as founded by the French civil tradition at first seems at odds with my general focus on legal origins, but it need not be perceived as such. Louisiana is further demonstration that effective reforms are not as simple as just benchmarking civil law tactics. Civil law strategies likely operate differently when placed in non-civil law contexts.

In a way, Louisiana is a tragically archetypal case taking the worst facets of both legal origin traditions. Despite rising to one of the nations most prosperous early states, thanks to key geographic location for trade and shipping, the civil law legacies of formal legal codes in the financial sector weakened Louisiana’s economic performance over time. The civil law legacy meant low wealth and job prospects and an oftentimes corrupt political system and infrastructural oversight. Louisiana citizens face an environment of failing public schools, high crime, and real monetary incentives for illegal participation in the drug trade. Coupled with the bias toward punitive severity fostered by the national common law, it is not surprising that if Louisiana were its own nation it would lead the world in imprisonment.

Last, I was fascinated by some of the references noted in Susanne Karstedt’s essay. In particular, she mentions Elizabeth Brown’s work finding a correlation between Republican representation and modern reform success. This is something I think various facets of the literature have alluded to without popular recognition for sometime. Thus I’m glad to see it mentioned here. Mainly, while the political left should be applauded for rightly expressing concerns over the conspicuously unequal effects of mass imprisonment, the right may require due recognition for having the more viable strategies of reform. In short, the United States imprisons so much because it can, and curtailing this national trend may require real limits upon how much or on what federal and local authorities are permitted to spend money on.

Though still, I’m hesitant to share Susanne’s optimistic belief that mass incarceration is en route to being an artifact of history. First, her prediction was shared by the majority of researchers and commentators in the 1970s, ironically preceding incarceration’s largest and fastest growth. What makes their forecasts so wrong then but more feasible today?

Susanne rightly notices 2009 and the financial crises had a halting effect on prison growth and such is consistent with the research suggesting the greatest reform potentials stem from fiscal limitations. States across the nation, with low to empty budgets and growing public attention to the failures of the drug war and police misconduct are less approving of spending on criminal justice than in the past. But as much of the behavioral science surrounding these trends suggest, such punitivity is at least a common feature of the human condition. While both political parties seem to be embracing at least the rhetoric of criminal justice reform, it remains to be seen how the political process will insulate itself in the long run from repeating the sorts of punitive escalation that led us here in the first place.

Criminology across States and Nations

Adam Gelb’s comment is spot-on. Cross-national comparisons nearly never acknowledge the significant differences between the states of the U.S.. Such differences actually mirror those in Europe, with a range of imprisonment rates and processes of both divergence and convergence.[1] Comparisons between U.S. states reveal huge differences. They differ widely in their imprisonment rates, and in their crime policies that drive imprisonment rates up or down. Only half of the states in fact adopted three-strikes laws, however those who did hopped on the band-wagon within the first two years, with some neglecting their own requirements of budgetary evaluation and prudence. As states climb down from high imprisonment rates some take the lead like California, Michigan and Texas. While half of the states started to reduce the numbers of prisoners, these numbers were still increasing in the other half. None of these states exceeded the number of 2,000 additional prisoners, which was half of California’s reduction. Were the rich states the first to increase incarceration in the 1970s? California was leading, presumably for a bundle of other reasons which had more to do with actors and mechanisms than with their budget. Texas in 2005 was the first state to shelve prison plans; they did this because they had strict rules against overcrowding in place at the time, which spoke to budgetary prudence rather than to a fiscal crisis or declining income. What about Louisiana? In 2009 the state was among those with the highest increase in prisoners, but in 2012 on the road to close more than 1,500 prison places. If there is any indication of path dependency in its criminal justice system it might be more illuminating to look at its history of lynching rather than searching for civil law remnants.

Looking at differences between U.S. states rather than between the United States and other countries has opened up new perspectives for the penal comparationist. They now focus on actors in criminal justice, their decisions, and the institutional framework which sets incentives and equally constrains their decisions. Just like U.S. states adopt policies from each other less because they are neighbours but because they see themselves as “cultural peers,” so do European states. Regional and supranational systems like in Europe and the European Union and also in Latin America increasingly shape national criminal justice systems and imprisonment. For Europe this implies the abolishment of the death penalty in all member states, decisions by the European courts on the vote for prisoners, or a system of monitoring prisons and detention in member states.  European states also are part of a supranational regime of welfare policies and provision, and thus European civil law countries have particular combinations of welfare and penal policies in place, which are absent e.g. in civil law countries in Latin America. Nonetheless, European countries and citizens still widely differ in their welfare regimes, the principles that guide them, and the attitudes that support them, and it is often ignored that communist countries in Central and Eastern Europe had a particular authoritarian type of welfare provision from cradle to grave until 1990. Differences in imprisonment rates seem mostly to be independent from certain types of welfare regimes.[2]

In light of these regional and supranational developments federal systems like the one of the United States are of particular interest to track the processes of the exchange and migration of crime policies, as well as the impact of the federal and supranational levels. I find both perspectives necessary and valuable, the close-up on the U.S. states as well as the cross-border gaze. However for exploring the boom in imprisonment and current process of “de-jailing” a comparison between U.S. states will presumably produce the more important and in-depth insights. Therefore Tiffany Bergin from Kent State University and I have embarked on a project that aims to shed light on exactly this.

Imprisonment is the “back-end” of criminal justice, and finalizes the process of doing justice. It has of course always been related to the “front,” to the safety and security of citizens from crime and its deterrent impact on potential offenders. Consequently the question arises how much imprisonment at the back-end is needed to produce safety at the front end. This perspective is wrong for a number of reasons. In particular it neglects the fact that criminal justice is a process that starts with an encounter with police, and continues with a journey through prosecution and courts. In principle, it would be unreasonable to expect the highest impact or any impact at all from the final act of justice and ignoring all other stages and players. It cannot come therefore as a surprise that the relationship between crime (“front,” we need to measure this as violent crime, namely homicide rates) and imprisonment rates is weak and inconsistent in international comparisons. Consequently, no empirically derived equilibrium between imprisonment and (violent) crime is in sight (I gather that both Mike Riggs and Daniel D’Amico are looking for such a chimera), and as long as this is missing, principles rather than data should guide the use of prison for certain types of crimes and offenders, as it does in civil law countries.

In contrast, research evidence and recent experience in Ferguson and elsewhere have clearly demonstrated the importance of the police at the front end of the law, which sheds further doubt on the link between back-end and front. Tom Tyler from Yale Law School and his colleagues across many countries (including both civil and common law countries) show that people are more inclined to obey the law if they are treated with respect and “fairly” by the authorities. “Procedural justice” increases the trust of citizens in the justice system, and this does not only apply to police, but equally to courts, and finally in prisons, where it greatly improves the relationship between staff and prisoners, as Alison Liebling from the Institute of Criminology at Cambridge University found. Safety and security not only means safety for all citizens, but also from violent abuse at the hands of criminal justice and its representatives.

Civil law countries have highly bureaucratized systems of justice, and a number of them are notorious for their kafkaesque and arcane systems, which result in a slow-down and often standstill of justice. Italy is notorious for this, and as “justice delayed is justice denied,” the trust that its citizens have in justice is at the lowest level among European countries. In Latin American countries with often dysfunctional criminal justice systems, trust levels are even lower, and this significantly encourages vigilante and self-help violence as findings from the Latinobarometro show thus driving violent crime up. In sum, safety and security for citizens cannot and should not be simply related to incarceration rates, but we need to take the whole criminal justice process into account as each part contributes in its own way. 

What role for punitiveness among the citizenry? Those who assume that a certain level of punitiveness and time-invariable demand for vengeance is part of the human condition will either build safeguards around the criminal justice system (as in civil law countries) or they will aim at balancing such demands with the sanctions that are meted out. Here certainly our capacity for forgiveness and empathy is ignored which balances our thirst for vengeance. Further, from such a perspective the relationship between criminal law and popular beliefs looks like a one-way street: law is shaped by popular demands rather than vice versa. In fact there is considerable evidence to the contrary. Here is my favorite example:  When the Federal Republic of (West) Germany was founded in 1949, the death penalty was abolished in the constitution. At this time and throughout the 1950s the majority of Germans supported the death penalty, however support decreased continuously until the second half of the 1960s when at about two third of the population were in favour of abolition, and thus in agreement with their constitutional law. Both educated middle classes and working classes changed their opinion but in opposite directions. While the educated and middle classes had been in support of the death penalty they now opposed it, while a majority of blue collar and less educated strata of society, who had been against the death penalty after the war (they had suffered most from the extensive use of the death penalty and executions by the Nazi regime) now turned around and supported it. Ever since then there has been little change in the majority opposition to the death penalty.

Is optimism reasonable? In the 1970ies a famous article proclaimed that “nothing works” in criminal justice. This was actually the start of a long period of penal pessimism which saw the rise of incarceration in many countries, and encouraged the punitive worldview that took hold of practitioners and researchers. David Green argues that an era of “penal optimism” has dawned in the United States and across the religious and political spectrum.[3] These signs of hope and optimism include new initiatives like the Second Chance Act, and signals like the visits of the first sitting president and a presidential candidate to a prison. I do not think that I am unreasonably hopeful. 



[1] S. Karstedt 2015. “ Cultural peers and penal policies:  A configurational approach toward mapping penal landscapes.” Punishment & Society, DOI: 10.1177/1462474515590890

[2] S. Karstedt 2014. “Europe as a normative power: Cultural peers and penal policies.” Hungarian  Society of Criminology, Proceedings of Criminology, 73: Budapest

[3] D. Green 2013. “Penal Optimism and Second Chances: The Legacies of American Protestantism and the Prospects for Penal Reform.” Punishment & Society 15(2): 123-146.