Tim Harford's splendid book Adapt

In 2011, Tim Harford wrote a whole book called Adapt on how it is important for key decision makers to be humble.  He points out that we live an a very complex and uncertain world, and because of this, it is difficult for experts to know in advance what ideas will work best.  In his view, the world is full of surprises and the best way to take advantage of this is to explicitly adopt a strategy of trial and error, or what he calls variation and selection, to try a lot of things, expect a certain amount of failure, and then take what you learn and create an even more successful system or idea.  He goes on to provide a number of examples where expert opinion fell short, and it was important for organizations to adapt to new insights.  He describes in fascinating detail how the War in Iraq, the Battle of Britain in WW II, implementing development programs in poor countries, and fighting climate change all benefited from this process of variation and selection.

I found this book especially delightful because it fits nicely with my own philosophy that it is important to carefully manage what I call epistemological risk whenever designing a new public policy.  Many people already know about evidence based policy, where policy makers deliberately use the latest research to improve their policy design process.  Managing epistemological risk takes this one step further, where you admit that you might be wrong about something, and develop a policy strategy that allows you to learn what works and what does not as quickly as possible in a way that minimizes the damage to society if you did get something wrong.  The idea is to do a lot of small scale trials and state level experiments and then dramatically scale up the successful ones once you learn what works best.  Waiting for an experimental trial can slow progress in the short run, but can speed up progress in the long run by reducing the number of policy mistakes that might take decades to fix, and to increase our rate of  learning about what works and what does not.

Even if it is very tempting to make change as quickly as possible on the issue you care about most, government as a whole has to deal with hundreds of different policy problems and improving the process by carefully managing epistemological risk has the potential to dramatically improve outcomes across the entire portfolio of issues.  Communist Russia showed us a top down approach has its limits, though it did achieve impressive levels of economic growth at the beginning, and America's system of markets that adopts this bottom up approach of variation and selection demonstrates that this strategy is vastly more successful over the long run.  Incorporating that into our political and policy strategies now could potentially be a very effective technique for improving the well being of our society for many decades into the future, and Tim Harford's book Adapt makes a very strong argument in its support.