A 2018 New York Times article reported an incident at the East Mississippi Correctional Facility, run by the Management & Training Corporation, a private company under government contract. In the graphic images caught by surveillance cameras, prisoners with sticks freely circulating in the facility chase and violently beat another inmate who collapses on the floor. In the footage, there are no signs of guards to stop the aggression.
The cause of the mayhem, according to the article: “Mississippi pays the company just $26 a day – or about $9,500 a year – for each minimum-security inmate. That is far less than the $15,000 a year neighboring Alabama spends per inmate and only 13 percent of what New York, which spends more than any other state, pays per inmate.”
To beef up profits, the newspaper claims, the private company economizes on security, medical support, and other essential services within the prison. Six days after his inauguration as President of the United States, Joe Biden decided to end federal contracts with private prison operators.
In a research project led by Sandro Cabral, with whom I collaborated jointly with Paulo Furquim de Azevedo, we saw a different picture. In the late 1990s, the governor of Paraná, a southern Brazilian state, decided to build new prison facilities whose management would be outsourced to private operators.
Gathering data over nine years (2001–2009), we found that, compared to their public counterparts, the privately operated prisons had fewer deaths and escapes, similar levels of medical appointments per inmate, and higher speed of internal legal services. We personally visited some of these facilities, walked through cell corridors, and even had lunch with some inmates in a semi-open incarceration regime.
Despite these positive outcomes, a rival politician who was later elected governor decided to terminate the private contracts for ideological reasons (in his campaign, he strongly opposed privatizations). Although ill advised, this change in policy allowed us to compare not only distinct facilities but also the same units that were previously private and then became public.
How can we explain these diverging outcomes? They are not simply due to differences in country-level jurisdiction and policies. Unfortunately, the Paraná experience was not successfully replicated in other Brazilian states. In 2017, 56 prisoners were slaughtered at the privately run Anísio Jobim Penitentiary Complex in the northern state of Amazonas. In the words of a criminology specialist, “we can say with certainty that the process of privatization and commercialization of prison management … played a central role in the massacre that occurred.”1
To understand such distinct trajectories, we should examine how privatization affects two critical performance outcomes. The first is related to what the literature has generally referred to as quality: whether inmates are humanly treated and reintegrated to society (prisons), whether students effectively learn and develop valuable skills (education), or whether clinical services promote effective prevention and treatment (health). The other key outcome is inclusion: who should benefit, and who actually does benefit, from privatization, especially among vulnerable social groups. I will discuss each of the two outcomes in turn.
Crucially, quality comes with cost. The ratio of quality to costs is an indicator of service performance taking into account its required expenses – which in this book I refer to as effectiveness. A hypothetical example, in education: a system of public schools generates, on average, a score of 60 (out of 100) in standardized student tests, where each student costs $1,500 per year. Thus, the schools deliver 40 points in learning tests for each thousand dollars invested in the system. This is a measure of their effectiveness.
Are private operators more or less effective than governments? Unfortunately, there is no simple answer. We should analyze the incentives to deliver superior quality and lower costs. The New York Times article suggests that private companies will tend to reduce costs, but that this will lead to quality deterioration. The final effect on effectiveness is ambiguous.
This cost–quality trade-off is the basis of a seminal academic contribution by economists Oliver Hart, Andrei Shleifer, and Robert Vishny. They argued that prisons involve quality attributes that are difficult to monitor and enforce; and, unlike customers comparing alternative shops, prisoners are not free to switch their service provider in case of mistreatment. In the view of these authors, their work “suggests that a plausible theoretical case can be made against prison privatization.”2
Our study of the privately operated prisons in Paraná, however, led to a different conclusion. A central feature in our case was the combination of private management and on-site public supervision. Prison wardens were well-paid, handpicked public bureaucrats with a mandate to enforce quality. Potential corruption between public supervisors and private operators was checked by the presence of media scrutiny and civil society organizations. And the prison wardens feared losing their jobs (and extra compensation) in case of reported quality deterioration and public pressure.
Notice that the outcomes of privatization crucially depended on the existence of strong public action to enforce high standards of quality and prevent service deterioration. For this reason, some even question whether these cases are correctly labeled as “privatization.” In many contexts, privatization is best characterized as active public–private collaboration. Along these lines, public policy scholar John Donahue considers privatization a way to engage “private energies to improve the performance of tasks that would remain in some sense public.”3
Effectiveness, however, is only part of the story. The other key performance dimension is inclusion. Going back to my previous example in education, consider two distinct education systems, one private and another public. The first system, as before, delivers 40 points in learning tests for each thousand dollars. The second system is less effective: the same amount of money yields 20 points. However, the latter is more inclusive: 90 percent of the families with severe income constraints are enrolled in schools, compared to 40 percent in the first system.
In some instances, private firms may care about potential inclusion and may even see it as an opportunity. For instance, a substantial body of work in business strategy has argued that large populations at the bottom of the pyramid (say, living in slums) are poorly served, allowing companies to tap into these large markets with more efficient operations. There is also an emerging trend of investors seeking for-profit companies that concurrently pursue socio-environmental benefits.
However, the promise of this movement proved more complicated than initially thought. Take the case of microcredit. Despite some positive experiences (such as Muhammad Yunus’s Grameen Bank), microloans do not automatically improve the lives of poor communities and may in some cases even cause harm. Entrepreneurs in vulnerable areas often lack basic skills to choose where to invest, craft productive operations, and manage their financial cash flow. They may simply end up with a pile of debt and with even fewer opportunities for individual progress.4
The objective of this book is therefore to promote a conceptually and empirically informed discussion of when and under which conditions private firms can outperform the state (via their state-owned companies and government units) in the execution of effective and inclusive activities.
Contrary to what we often find in public debates, I do not have any intrinsic preference for private or public ownership, for a simple reason: theory and evidence show that they both matter depending on a host of contextual conditions and the relative weight that each society gives to effectiveness and inclusion.
The right question, therefore, is not whether privatization is better than public management, but when and in which conditions. Furthermore, even in cases where private firms are relevant, we have the key regulatory role of governments as well as complex forms of hybrid public–private interaction – as in our case of the Paraná prisons as well as other instances in education (e.g., charter schools), urban systems (e.g., public support for affordable housing), technology development (e.g., publicly funded private research), and many other activities.
In the next chapters, I develop my argument as follows. I start with a review of the academic thinking on the role of private and public organizations in generating social benefits – moving from an analysis of their relative ability to provide public goods to a more encompassing comparative analysis of the various factors that affect the merits and costs of alternative arrangements.
I then explain in greater detail the concepts of effectiveness and inclusion and how their relative emphasis will depend on societal preferences. Considering these performance dimensions, I offer a decision-making framework describing conditions that will favor the adoption of alternative ways to deliver key services: generally speaking, privately managed activities, public–private collaborations, and public (state-owned) organizations, in their multiple forms and varieties.
I compare these forms based on their relative ability to generate effectiveness and inclusion, as well as with respect to their perceived legitimacy. In several countries, privatization suffers strong opposition and is not seen as a legitimate alternative. In part, societal concerns derive from the aforementioned possibility that private operators might neglect relevant service attributes and exclude vulnerable communities. Privatization is not an easy task – ignoring societal expectations and rushing to implement changes without proper analysis and deliberation are recipes for failure.
Next, I argue that crafting successful privatization programs depends on capable governments (which does not mean big ones). Importantly, as the previous example of the Brazilian prisons shows, having good governments is not a general, country-level attribute. Even in countries whose institutions are flawed or underdeveloped, we often see public units at subnational levels acting as pockets of government capabilities: mobilizing resources, committing to well-specified policy objectives, and transparently engaging private operators.
Therefore, a central argument of this book is that privatization depends on good governments that not only set performance standards in dimensions that may not be prioritized by private firms but also guarantee that the whole process is diligently crafted and monitored. In other words, privatization is not a way to get rid of bad governments. Rather, capable private firms and capable governments are complementary.
Yet capable governments experiment not only with outright privatization but also with multiple forms of delivery, including hybrid public–private collaborations and even improved state-owned operations. Ironically, although this is a book about privatization, an implication is that capable governments make the decision to privatize less relevant. Instead of pursuing single solutions, good governments create and nurture multiple paths of improvement – which may include privately managed activities but also well-run public operations.
I then delve into several ramifications of my main argument. Although private actors are generally depicted as profit maximizers with little concern for social impact or inclusion, I show how this assumption can be relaxed with the emergence of private owners with socially oriented preferences and/or who are subject to contractual incentives – such as when governments and nonprofits compensate private investors based on contractual targets reflecting improved social outcomes.
However, thus far, private firms still need to show whether they can truly accept and deal with the inherent financial trade-offs in activities with critical service attributes or targeting highly vulnerable beneficiaries. Some of the claimed social investments of private investors are based on poor metrics and small-scale projects that fail to demonstrate solid improvements.
In addition, although it is often assumed that private firms will already have the required skills to generate impact, in many contexts governments actively sponsor new private capabilities. For instance, responding to the COVID-19 pandemic, state units collaborated with private firms to develop vaccines and provide essential services. Government capabilities are again critical, as poorly targeted policies may end up wasting public resources with unproductive firms that might request continuous support even when their initiatives fail.
Finally, I present a sequence of steps to successfully design, implement, and monitor privatization processes. In cases where proposals to privatize services suffer strong opposition and become unfeasible, I examine how various reform initiatives can lead to better and more effective public organizations, which may also interact with and complement the services of private firms.
The final picture that emerges from this discussion is, to quote philosopher Isaiah Berlin, plural.5 Over time, societies may learn to propose and build on diverse experiences. They do not try to find “the” best option but live with alternatives. They may err, but they soon find new possible paths of improvement. Instead of pursuing simple formulas, they embrace diversity. Privatization, in this sense, is not a final goal, but an opportunity for informed choice and public deliberation.
