Chart of the month: Was the pandemic uniquely hard to anticipate?

One of my favourite papers supporting our mantra’s argument is from the IMF in 2018, which runs through the poor track record in anticipating recessions from both the official and private sectors from 1992-2014. In the words of the authors, “both are equally good at missing recessions”.

We often hear that the pandemic was an inherently and abnormally unpredictable economic shock. We thought we could test that statement by extending the IMF’s research to cover the 2015-2022 period. We extend the original study by looking at the forecasting track record in the IMF’s April and October ‘World Economic Outlook’ across the same 63 emerging and developed economies, and then present summary statistics for the whole sample.

Our chart of the month shows the proportion of recessions correctly anticipated by the IMF. In the 30 years covered by the extended study, the IMF correctly anticipated 12 out of 236 recessions for a hit ratio of 5% in the April before they started, rising to 12% by October. They were quite good at recognising recessions that were already unfolding, but that’s not really much of a forecast.

I’m not disputing that the pandemic was hard to predict. I just question whether it was uniquely hard to predict.  Recessions during the pre-pandemic era were, seemingly, not much easier to anticipate. The IMF is the world’s premier international economic body, with a staff of over 3000. This track record should be humbling for the rest of us.

Macroeconomists do an incredibly valuable job analysing economic structures, calibrating the impact of shocks, and assessing the consequences of policy choices. Forecasting is just one of their outputs, but often the highest profile.

When it comes to investment strategy it’s all about the risks relative to what is priced at the time. Anyone tempted to think they’ve discovered the silver bullet to forecasting the business cycle, should think hard about this track record.

Prepare, don’t predict.