Coronavirus, Ray Dalio and forecasting in an age of uncertainty

Gillian Tett at the Financial Times: “Predictive models only get you so far. We also need to maintain our peripheral vision…

What is interesting to ponder is what this episode reveals about the nature of forecasting — and our modern attitudes towards time. As anthropologists often point out, the way we think about time is a defining feature of the post-enlightenment world. During much of human history, the future was viewed as a vague and terrifyingly unknowable blur marked by constant bargaining with deities (to ward off disaster) or cyclical seasonal rhythms (of the sort that underscore Buddhist cognitive maps).

In modern, post-enlightenment western cultures, however, a linear vision of time emerged that presumes the past can be extrapolated into the future, with a sense of progression, not just cyclicality.

In the 20th century, this gave birth to the risk management and finance professions, as Peter Bernstein wrote two decades ago in his brilliant book Against the Gods: the Remarkable Story of Risk.

By the turn of the century, innovations such as computing and the internet were turbocharging the forecasting business to an extraordinary degree, as Margaret Heffernan notes in her excellent (and very timely) new book Uncharted. “Human discomfort with uncertainty . . . has fuelled an industry that enriches itself by terrorising us with uncertainty and taunting us with certainty,” she writes.

However, as Heffernan stresses, while the forecasting business has made its “experts” very rich, it is also based on a fallacy: the idea that the future can be neatly extrapolated from the past.

Moreover, the apparent success of some pundits in predicting events (such as the 2008 crash) makes them so overconfident that they get locked into particularly rigid models. “The harder economists try to identify sure-fire methods of predicting markets, the more such insight eludes them,” she writes. Is there a solution? Heffernan’s answer is to embrace uncertainty, build resilience, use “narrative” (or qualitative) analyses instead of rigid models and to respect the wisdom of diverse views to avoid tunnel vision….(More)”.