Systemic complexity and criticality in economic forecasting

It is, once again, the season of the year when not only are we preparing for Christmas holidays and starting to think about new year resolutions, but economic forecasters are also offering their outlooks for the upcoming year. However, the last two years should have convinced even the most stubborn hedgehog that there is far less predictability, let alone certainty, around us than we like to believe. In particular, problems resulting from “system complexity” are, in our view, not sufficiently appreciated by forecasters and the recipients of these forecasts, alike. The critical assumptions, nota bene assumptions not predictions, driving – to a large extent – GDP and inflation forecasts for the next one or two years, are the future development of the COVID-19 pandemic and the – hoped for – gradual easing of supply bottlenecks, both almost textbook examples of system complexity. So are, probably, the Philips curve models used to forecast inflation. Let’s face it, believing in inflation forecasts with exact numbers, even behind the decimal point, for several years out, is little different to believing in Santa Claus. [more]