Looking for your keys under the lightning: social science in turbulent times

It is time to come out. I am writing a book. When I tell this to senior academics, they wish me "Good luck," and their eyes gain a mysterious, troubled look. Writing books is not what economists do.

Here is a self-justifying sketch of the relationship between social science and its historical setting.

Think of the possible states of society as a multi-dimensional space, each dimension corresponding to a variable: the growth rate, inequality, the price of corn, the prevalence of opiate addiction and so on. Different societies or parts of society – Kansas and New York, Denmark and Greece, or South London and North Oxford – are at different points in the space.

Nearby societies are at different points in a many-dimensional space

Some exogenous variables are like the weather: they change naturally, but we cannot change them. Others are policy variables, in the broadest sense that they can be affected by collective human choices. The optimal value of the policy variables depends in on the value of the exogenous variables. Social science involves working out how our welfare depends on the policy and exogenous variables.

At settled times in history, society appears static. Expected changes in the exogenous variables are small, and welfare appears high, suggesting that policy variables are about right. In this state, social science involves looking nearby in the social space – probably within the convex hull formed by different existing parts of our society. This gives us lots of data.

Economists talk about "looking for your keys under the street light". The idea is that your keys, if you have lost them, are no more likely to be lying under the street light than elsewhere; but it makes sense to look under the street light, rather than in the dark where you won't see them anyway.

In settled times, we look under the street light. Prestigious social science involves careful empirical work, leveraging the available data to recommend incremental policy changes. Think of Esther Duflo and the economist as an engineer, or, as Keynes once wistfully suggested, a modest profession like dentistry.

Social science in settled times: looking for small changes within the convex hull* of what is known
[* the shape shown is not quite convex, but eh.]

Other times in history are times of turbulence, of fast change and widespread dissatisfaction with the status quo. Expected changes in exogenous variables are large, or social welfare appears low, suggesting that we need large changes in policy variables.

In turbulent times, we need to explore remote points in the policy space, beyond the convex hull of what is known. Social science involves bold steps of imagination and deep theoretical examination of current and possible structures. Think of Marx and what Joan Robinson called his intellectual "seven-league boots", striding forward over trivial details. Or think of the public choice movement of the 1970s, which looked at the foundations of political constitutions. In fact, we need to "look for our keys under the lightning" – flashes of insight which illuminate the whole landscape and point to big possible changes.

Social science in turbulent times: looking for keys under the lightning.

Today we are living in turbulent times. The economy is changing fast as the centre of the world moves East. Politics, driven by voter dissatisfaction, is just as fast-moving and less predictable.

Of course, it is arrogant to assume that you can produce a bolt of intellectual lightning! But if the premium is on lightning, then it may be worth trying.

This is why I am ignoring my senior colleagues' troubled looks, and writing a book.