Last Sunday a friend suggested in a conversation that the recruitment of organ donors is badly in need of a policy fix. In the US alone, more than 120.000 people are on a list waiting for an organ transplant, and it is said that of those 18 die each day. Practically everybody loses because of this lack of donated kidneys, hearts, lungs and the more than 50 organs or tissue parts that each potential donor doesn’t provide. Non-donated organs are buried in the ground or cremated. The issue is in need of a fix.
What is a policy fix, and how does it differ from good old politics? Politics in part is about what and how to take from Peter to give to Paul. A successful policy fix, by contrast, should make practically everybody better off, minus some profiteers. In donating his kidneys, Peter has limited reason (and, if the doctors don’t fudge it, no sense) to be aggrieved. Unless you are a nihilist, you will agree that Paul alive is a happier result than Peter and Paul dead.
At their best, think tanks are Pareto optimizers. They seek to improve everybody’s lives. Given the societal stakes, proof that practically everybody can be better off should be rigorous. As gaps crossed with the best intentions can turn into abysses, we need a reliable map to get from here to there. Such maps, as all accurate representation, are the product of painstaking measurement and triangulation.
Defining think tanks as Pareto optimizers highlights their distinct role: they should be fixers, optimizers and gapclosers. Their legitimacy rests on the systemic and replicable examination which we call research. In proposing inclusive solutions, they case the risks and rewards. The emphasis on getting from here to there distinguish think tanks from scholarly endeavours. Academic research already does its job when it clarifies why Peter does not donate. Think tank and policy research seeks arrangements that let Paul live, and ideally Mary as well.
Why call them Pareto optimizers? Vilfredo Pareto, an Italian economist, defined optimality as a state “in which it is impossible to make any one individual better off without making at least one individual worse off”. In its idealized form policy can be a process of creating, as you move towards the Pareto optimum. Once you have reached it or got close, politics is a negotiation on how to share. The focus on inclusive optimization sets think tanks apart from political parties or partisan pressure groups, which have more redistributive traits.
This definition is more suited to pinpoint an ideal type than to drawing categorical boundaries. Yet one of its attractions is that it helps to define excellence. Next to outer trappings of success such as funds raised, think tanks truly excel if they expand our sense of what is possible by discovering and promoting new ways of making lives better. Orienting themselves toward that excellence can help think tanks make their own zero-sum decisions on where to focus. Saying that think tanks need to show that we can do better summarizes what to expect from their research. Demanding that they be able to chart the risks helps them scale their ambition. For example, if institutions do not have the experience to propose solutions or weigh risks, measuring the actual situation can be a sensible first step. If you cannot be a gapcloser you can contribute as a gapminder.
Within that ideal of Pareto optimization, it is fine for researchers to start from different values and premises. The one requirement is that they be transparent about where they are coming from. Thus, the market-oriented Cato and American Enterprise Institutes have both extensively argued that organ donors should be rewarded with incentives. The Center for American Progress has suggested that donor recruitment be more inclusive. RAND publishes technical papers, evaluating existing approaches with cost-benefit analyses. The New America Foundation mentions organ donation in the context of broader “choice architecture”. Most of these contributions also consider the ethical trade-offs which would be involved in various ways of getting the likes of Peter to donate, and for Peter’s family to be comfortable with that decision. Premises being transparent, policymakers and the engaged public can make up their own minds about how they want policy challenges to be fixed.
Getting back to seeing think tanks as optimizers, and applying this to policy research in developing contexts, two potential implications stand out. First, in most developing contexts there arguably is even more optimization to do, since state institutions often struggle to implement effectively, and expertise is transient. Such optimization need not take the shape of a grand plan, but perhaps more realistically can be part of Popperian piecemeal engineering, or what now is called iterative adaptation. Figuratively speaking, rather than one five year plan, five parallel one-year plans, with a thorough evaluation of what worked and how to scale it to the next stage. Secondly, in my experience one asset that is in short supply in developing contexts is credibility. Thus consistent quality is particularly important, and credibility is an end, not just a means.
Academia is not a role model. Academic research that gets its basic data wrong is wasteful but can sink into benign irrelevance. By contrast, bad policy research, in claiming to guide, is misleading. Becoming systematic about quality control thus is an essential step towards contributing to the optimization that think tanks can excel at. Especially in developing contexts such an improvement in quality helps the entire policy research community, and people like Paul.
One year ago, Enrique asked me how I would define a think tank. It has taken me a while to give one answer. Now what do I need to optimise?