By Laura Tyson and Lenny Mendonca/Berkeley

The US has 5% of the world’s population and 25% of the world’s prison population – about 2.2mn people, five times as many as in 1980. One out of every 100 American adults is incarcerated – the highest per capita rate in the world, 5-10 times higher than in Western Europe or other democracies. The social and economic toll is similarly high.
The boom in America’s prison population in recent decades is the result of ramped up punitive crime-prevention measures, including tougher drug penalties and mandatory minimum sentences, backed up by growing numbers of police and other law-enforcement officials. Beyond the financial costs of larger police forces and increased pressure on the judicial system is $60bn a year in spending on state and federal prisons, up from $12bn 20 years ago.
And then there are the huge costs for those imprisoned (many for non-violent crimes) and for their families and communities – costs that fall disproportionately on the poor, the uneducated, African-Americans and Latinos, and the mentally ill.
Perhaps the worst part is that the expected benefits of America’s “get tough” approach have failed to materialise. Indeed, there is only a modest correlation between higher incarceration rates and lower crime rates.
Moreover, the recidivism rate is shockingly high: according to a recent US Department of Justice report, more than one-third of released prisoners were rearrested within six months, and more than two-thirds were rearrested within three years. In order to reduce the size of the prison population, the recidivism rate must drop.
Of course, this is a complex problem. Released inmates face huge barriers to employment, housing, health care, and education. At least half face the added challenge of mental health and addiction issues. Yet very few have support networks to navigate re-entry to society.
Clearly, a new approach is needed – one that capitalises on the comparative advantages of the private sector, state authorities and the federal government.
The federal government should rely on “progressive federalism” to catalyse and fund state-level programmes to combat recidivism. Many of these programmes can be based on pay-for-performance contracts, such as social impact bonds, in which the federal and state governments share risk with private-sector actors.
Social impact bonds, or SIBs, require that private investors and other non-government actors cover most or all of the upfront cost of a pilot project, to be reimbursed by the contracting government agency only if independent evaluators conclude that the project achieves its goals and saves taxpayers money. The world’s first SIB, launched in the United Kingdom in 2010, focused on reducing recidivism rates among 3,000 prisoners at Her Majesty’s Prison Peterborough in Cambridgeshire, and showed promising results.
The first American SIB, aimed at reducing recidivism among juvenile inmates at New York City’s Rikers Island Correctional Facility, was not so effective. Indeed, last month, independent evaluators concluded that the program had not met New York City’s targets.
But, far from being a failure, the Rikers Island project validated the pay-for-performance approach. New York City did not have to foot the bill for the failed effort, but officials gained valuable knowledge about what does and does not work. Based on this experience, another US firm is developing a SIB for a different recidivism project run by New York State.
In fact, many state-level pilot projects – some based on pay-for-success deals – to reduce recidivism are underway across the country. New York State and Massachusetts have launched statewide pay-for-success experiments. Minnesota and Texas have run pilot projects with promising results. Georgia, using a $6mn grant from the Department of Justice, is financing 15 separate pilot programmes that range from job-training and housing-support services for released prisoners to faith-based “in-reach” programmes for those still in prison.
These state projects collectively amount to a form of crowdsourcing – an effective method of testing a wide variety of innovative ideas, which can then be scaled up if they prove successful. In this sense, state and local governments are serving as “laboratories of democracy.”
This is where progressive federalism comes in. The federal government can work to sharpen and reinforce state efforts by providing funding and encouraging best practices, while avoiding imposing any ideology on the projects. The federally assisted pilot projects to combat recidivism in Georgia, for example, reflect both “liberal” and “conservative” strategies, supported by Democrats and Republicans, respectively.
The social sector also has an important role to play in combating recidivism. Advocacy groups like The Coalition for Public Safety are changing the national dialogue. There is an entire sub-industry of nonprofits like Defy Ventures working to help former inmates build new careers and even become entrepreneurs. Others programmes, such as Measures for Justice, are creating the data needed to gauge results, while non-profits like Pew Charitable Trusts, California Forward, and Third Sector Ventures are helping counties and cities find economically effective alternatives to expensive and ineffective prison expansion.
Democrats want an active government that solves tough social problems, whereas Republicans want private-sector investment and innovation to do the job. But both parties like the idea of testing rival strategies in the real world, as evidenced by bipartisan support in Congress for a new federal fund to support pay-for-performance projects on a wide range of social problems, including health care, child care, and job training. The combination of pay-for-performance contracts and progressive federalism seems to meet both sides’ requirements.
The Violent Crime Control and Law Enforcement Act of 1994 – which provided funding for states to put more police on the beat, impose tougher prison sentences, and build more prisons – was an example of how the federal government can encourage action by state and local authorities.
A new federal crime bill, incorporating pay-for-success contracts, could encourage states to take a smarter approach to crime, reducing mandatory prison sentences and investing in effective anti-recidivism programmes.
Such an approach would reduce incarceration and recidivism rates – and drastically cut the onerous social, economic, and moral costs of imprisonment.- Project Syndicate

- Laura Tyson, a former chair of the US president’s Council of Economic Advisers, is a professor at the Haas School of Business at the University of California, Berkeley, and a senior adviser at the Rock Creek Group. Lenny Mendonca is a former director of McKinsey & Company.