ModelsThe Problem of Modeling

The lessons starting to emerge from the coronavirus crisis are predominantly not epidemiological but highly general aspects of public policy, Paul Collier writes in the Times Literary Supplement.: the over-reliance on expert modelling, the mismanagement of public services, and the contrasting pressures faced by libertarian, statist, and communitarian approaches to governance. This virus is one of many to have swept around the world. “Some, like Ebola, are deadly but, as a consequence, do not spread very quickly. Others, like SARS-CoV2 (which causes Covid-19) are easy to catch but mercifully seldom fatal. And one, yet to come, may be easy to catch and deadly,” Collier writes. “We need to learn from this one not only to limit its current damage but so as to be able to respond more effectively if such a future virus strikes.”

He adds:

The current epidemic is a classic application of what economists call “radical uncertainty” (most recently explored by John Kay and Mervyn King in their brilliant book of that title, which came out last month [Radical Uncertainty: Decision-Making Beyond the Numbers]): in a world that has inevitably become too complex to be adequately captured in models, a world of both “known unknowns” and “unknown unknowns”, the most sensible response to the question “what should we do?” is “I don’t know.” At the onset of this crisis, we could not put probabilities on which forms of social distancing would best limit its spread because we’d never done it before. We didn’t know how people would alter their behavior in response to the appeal to “save the NHS.” We didn’t even know whether reducing the spread was desirable: perhaps fewer deaths now would come at the cost of more next winter. And these were just the known unknowns. With a disruption as big as this, unknown unknowns are also lurking. We have no experience of the material and economic repercussions from shutdowns of this nature and their aftermath in a modern economy, and no meaningful way of assigning probabilities; nor of how people’s behavior will evolve. 

Collier, a professor of economics at Oxford, writes that a common refrain in economics is that “it takes a model to beat a model”, and “that attitude encapsulates the fundamental error of British policy. When we’re faced with events like this, all models will be wrong. They have their uses, but the lesson is to think beyond the models, not just within them.”