It's all about control

Tony Blair admits Labour didn't fully understand complex financial sector source

Quite. Unfortunately his Government behaved as though it did understand the financial sector. Even more unfortunately, governments everywhere behave as though they understand everything: climate change, agriculture, education, housing, health, transport, crime, war and all the rest. Government so often doles out subsidies, regulates or deregulates, or otherwise tries to control complex systems that it does not, and cannot, understand. Some relationships between cause and effect are easy to understand and manage: that between paying for primary education and literacy, for example, or between organising rubbish collection and human health. But in so many other areas government is well out of its depth.

Social Policy Bonds would remove the need for government to identify in advance the extremely obscure relationships between cause and effect that are ever more a feature of our complex society. Instead of "picking winners" such as the financial sector and lavishing implicit subsidies on it, government could instead target outcomes that are meaningful to ordinary citizens - basic health, literacy, housing and health outcomes, say - and let others do the endless, and endlessly difficult, task of identifying the important links between activities and outcomes. A Social Policy Bond regime would enable government to do something that it's actually quite good at: articulating society's wishes and raising the revenue to achieve them. But the actual achievement is something that, as is becoming clearer, is best left to people who are paid according to their effectiveness and efficiency, rather than how well they can convince the politicians that they are doing something useful.

An alternative to doing nothing

Bill McKibben summarises progress on reducing greenhouse gas emissions: 

Last month the world's nations, meeting in Rio for the 20th-anniversary reprise of a massive 1992 environmental summit, accomplished nothing. Global warming's terrifying new math, Bill McKibben, 'Rolling Stone', dated 2 August

 And again: 

everyone at the Rio conference renewed their ritual calls for serious international action .... The charade will continue in November, when the next Conference of the Parties (COP) of the U.N. Framework Convention on Climate Change convenes in Qatar. This will be COP 18 – COP 1 was held in Berlin in 1995, and since then the process has accomplished essentially nothing.

No surprises here. Mr McKibben writes eloquently of the financial disincentives for governments and energy companies to do anything to cut greenhouse gas emissions. It's one policy area where there's insufficient scientific proof to persuade political parties to take actions that adversely affect the people who make contributions to their election campaigns. More precisely: the relationship between imposing a carbon tax (say) and benefiting human, animal and plant life is too indirect to persuade policymakers that it's in their interests to do anything beyond paying thousands of bureaucrats to fly round the world drafting meaningless documents. Our current policymaking system lets governments off the hook if they can point to uncertainties in the scientific relationship between cause and effect: in this case, between cutting greenhouse gas emissions and an improved environment. Sure, politicians might privately be quite convinced by the evidence but, if it's not obvious to everyone, they are unlikely to take action that brings long-term benefits to society while imposing upfront costs on themselves and their supporters.

Here's another suggestion. If we want to do something about cleaning up the environment, let's pay people to clean up the environment. If we want to tackle climate change, let's (after defining exactly what we want to achieve) tackle climate change. We need to target outcomes, rather than wait decades for government to agree on what's required and then make policy that might, or might not, do something to achieve it. What is important is not how our urgent social and environmental problems are solved, but that they are solved. Social Policy Bonds are one way of getting government to focus on what we as a society actually want to achieve and out of the business of identifying genuinely (or not) difficult scientific and social relationships. The first is something that democratic governments can actually do quite well. The second is something that they do badly, whether because coming to a consensus about most scientific and social relationships is unachievable, or because of their incompetence or self interest.

Climate change is quite possibly such a huge and urgent challenge that we can't wait for politicians to be persuaded that they have more to gain by dealing with it than by pretending to deal with it. If the world's governments collectively issued Climate Stability Bonds it would be motivated bondholders, not motley interest groups, who would decide what needs doing and how best to go about doing it. Governments - and taxpayers - could relax in the knowledge that if, in fact, the doubters are right, and there's no climate change problem, then the way the market works would ensure that costs to society are minimised. And, if there is a problem, investors in the bonds would have incentives to explore and implement the most cost-effective ways of solving it. They would do all this in ways that take account of our rapidly expanding scientific knowledge - something that no government can do. By targeting outcomes, rather than the supposed ways of achieving it, a bond regime would be more efficient than any other policy mix. Perhaps even more important, a bond regime would compel government to do its duty and address big, urgent problems, rather than, as now, get away with pleading uncertainty and doing nothing.

The state doing what it does best

Tony Judt:

We have freed ourselves of the mid-twentieth century assumption—never universal but certainly widespread—that the state is likely to be the best solution to any given problem. We now need to liberate ourselves from the opposite notion: that the state is—by definition and always—the worst possible option. quoted by Zadie Smith in The North West London Blues, 'New York Review of Books', June 2012

The state does do some things very well. Democratic states are good at articulating society's social and environmental goals, and at raising the revenue necessary to achieve them. Where it often fails is in achieving these goals efficiently. Partly this is because government employees are rarely paid according to how successful or efficient they are in achieving social outcomes. Social Policy Bonds could transcend this limitation. They would allow government to contract out the achievement of society's goals to the private sector, while still defining these goals, targeting them, and paying for them. The state would be relinquishing some of its power, particularly its often-abused power of patronage, in issuing Social Policy Bonds, but it would be doing so in the service of better outcomes for all. 

A bond regime could also encourage the efficient targeting of global problems; something that is rarely done effectively under the current system. We could look to governments or supra-national bodies to target such universally desired goals as an end to war, the mitigation of natural or man-made disasters, including climate change. It is probably governments - and only governments - that could raise the funds necessary to solve such huge and urgent problems. They are trying, but largely failing, to achieve our goals deploying the tools currently available. But, if they collectively backed Social Policy Bonds targeting our global problems, they might be much more successful.

"Seeing Like a State"

I haven't read Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed, by James C Scott, published in 1999. (There is a good review of it here.) The description on the Amazon page sums it up: 

Centrally managed social plans misfire, Scott argues, when they impose schematic visions that do violence to complex interdependencies that are not -- and cannot -- be fully understood. Further, the success of designs for social organization depends upon the recognition that local, practical knowledge is as important as formal, epistemic knowledge. The author builds a persuasive case against "development theory" and imperialistic state planning that disregards the values, desires, and objections of its subjects. He identifies and discusses four conditions common to all planning disasters: administrative ordering of nature and society by the state; a "high-modernist ideology" that places confidence in the ability of science to improve every aspect of human life; a willingness to use authoritarian state power to effect large-scale interventions; and a prostrate civil society that cannot effectively resist such plans.

I've argued similarly in the past. Our current way of solving social and environmental problems is, in essence, centrally managed. The result is something like a policy monoculture, and the results are predictably lamentable. But the important distinction to make is that between centrally planned outcomes and centrally managed ways of achieving them. We all want to see reduced poverty, the ending of violent political conflict, and universal literacy, for examples. Government does a good job at articulating our wishes in these and other areas. But centrally planning the ways of achieving these goals just will not work. We need diverse, adaptive solutions; ones that take into account circumstances that vary with time and space. Central planning can't do that and the results of its failure are widespread and tragic. 

Which is why I advocate Social Policy Bonds. Under a bond regime we would set goals and contract out their achievement to people motivated to investigate and implement the only the most efficient projects. These projects would adapt to changing circumstances, and be sensitive to local conditions. Under a bond regime, the complex interdependencies that Scott writes, which cannot be understood by government, need not be understood by government. Instead, via an automatic system of cascading incentives, Social Policy Bonds would encourage diverse, adaptive initiatives that would contribute to achieving our large-scale - even global - goals.