Kenneth Rogoff may have been joking when he said that the conventional wisdom of Davos participants is always wrong, but even so it provides an opportunity to consider whether this is indeed the case, and if so why. It’s why we devoted a session to it this year.

First, the facts. It’s true that neither a Donald Trump victory nor an UK exit from the EU were regarded back in January with much degree of certainty. But participants weren’t alone in calling the seismic political events of 2016 incorrectly.

One of the reasons, argues Molly Crocket, Associate Professor of Experiential Psychology at the University of Oxford, is that forecasters tend to have a bias towards the material world over emotion when making forecasts. “People often underestimate how powerful a motive expressing one’s world view is: predictions often instead focus on economic self-interest because this is more tangible and easier to measure.”

The answer, says Philip Tetlock, a Professor at the Wharton School at University of Pennsylvania, is to move towards more evidence-based forecasting in the future. But don’t hold your breath: “Any experts would be very hard-pressed to beat a simple extrapolation algorithm. Predicting change is the great challenge and this is where most systems fail,” says Tetlock.

If predicting change is so difficult, some of the forecasts by the Forum or its constituents might not be so wide of the mark after all. Christine Lagarde of the International Monetary Fund reminded a TV audience today that she warned the world of the dangers of inequality back in 2013. Even further back, Professor Klaus Schwab, Founder and Executive Chairman of the World Economic Forum, wrote in 1996 of a mounting backlash against globalization, while in 2012 the Forum’s Global Risks Report identified severe income disparity as the most likely risk to face the world in the next 10 years.

A year later, we called out another trend that was well beyond the horizon: massive digital misinformation – or digital wildfires as we called them at the time – warning of the challenge presented by the misuse of an open and easily accessible information system.

Tetlock is optimistic that, over time, systems will be built that are better able to predict change. Even so, both he and Crockett agree that the best way to prevent the anger that contributed to political change is to tackle its causes: feelings of inequality, anxiety and of being ignored.

Watch the Forecasting Failure session at Davos 2017 here