Why it’s unwise to make long-term labour market forecasts

Pundits make long-term forecasts about global labour market trends at their own peril, writes UNSW Business School’s Frederik Anseel

Despite the economic turmoil, companies continue to look for talent. And because that search does not yield much, they invest heavily in training and education to develop their own talent. Excellent news, but not what you’d expect from listening to forecasts about an impending recession.

Predictions about the labour market skate on the same thin ice as economic forecasts. For the sake of convenience, I will call pundits, labour economists and work psychologists (like myself) “forecasters” here. Forecasters tend to linearly extrapolate past patterns to the future and isolate one changing environmental factor to support their prediction. A logical prediction would sound like this: “The economic shock from a pandemic will suppress global demand (not supply), and many will lose their jobs as a result.” That prediction turned out to be far from correct.

Another prediction was that technology (specifically artificial intelligence) would eliminate nearly half of the jobs. The idea was that there is a fixed amount of labour that remains constant worldwide. If technology replaces a whole series of human tasks, it is inevitable that a ‘package’ of labour disappears, and thus jobs disappear.

Read more: What’s really behind the sputtering supply of talent?

Voluntarist policymakers were already making wild plans. Because there would simply be no more work for everyone, they would introduce a universal basic income. Surprise – almost 10 years after the Oxford study made its famous automation prediction, we globally see some of the lowest unemployment rates in history! Countless new jobs are being created by new technology; ‘packages’ of labour are added all the time. There is no such thing as a constant “package” of labour.

Again you see the same linear thinking: pundits think of past labour, isolate one change in the environment (technology) and make their prediction based on that one change. Those confronted with their failed predictions often defend their misjudgement as follows: “Well, I would have been right, but no-one could foresee an epidemic and the resulting intervention of central banks.” That is exactly my point. As Nobel Prize winner Niels Bohr quipped, “Prediction is very difficult, especially if it’s about the future.”

The human brain has a hard time considering multiple changes at once (let alone the interaction between those changes). And so we limit ourselves to the most striking factor at that time. But there is never just one influencing factor. At any given time, dozens of important factors influence the labour market. How can an individual forecaster estimate how changes in inflation, energy, climate, demographics, war, technology and migration will affect each other? As a forecaster, you are doomed to fail in the long term. Master investor Warren Buffett guards against such macro predictions: “Predictions tell you more about the forecaster than about the future.”

Frederik Anseel UNSW Business School-min.jpeg
UNSW Business School’s Frederik Anseel says predictions about long-term labour market forecasts are a fool’s errand given the future is inherently unknowable. Photo: supplied

What skills will become important?

A good example is predictions about which skills will become most important in the labour market. Let’s play a game and travel 40 years back in time to predict which skills would become most important in the decades to come. Digital skills? Data analysis? Math? Coding and programming? Financial literacy? Wrong. Social skills. The number of jobs requiring social skills increased by 12 per cent between 1980 and 2012, during the Silicon Valley boom. The number of jobs with a mathematical basis (such as STEM professions) and few social demands decreased by more than 3 per cent. Wages rose most in jobs that required both strong social and math skills. The figures are based on an American study, but a Swedish study finds a similar pattern.

What does that mean for the future? The main labour market forecast for the next 40 years is extreme labour shortage due to declining birth rates. Have you seen the statistics? With a projected population decline of 50 per cent by the end of the century, China’s demographic pyramid looks frighteningly fragile. The impact of this on the labour market is difficult to estimate because we have never experienced such a demographic contraction.

Subscribe to BusinessThink for the latest research, analysis and insights from UNSW Business School

A shrinking population is also why countries like Portugal, the United States, Singapore, Australia, Canada and the United Kingdom have launched ambitious, targeted migration plans in recent months. The battle for talent is perhaps even more raging between countries than between companies. But does it actually come to such a shortage? I no longer dare to make predictions about global patterns. Things that have never happened before happen every day. History is explainable but not predictable. Our long-term future is fundamentally unknowable.

Frederik Anseel is Professor of Management and Senior Deputy Dean (Research & Enterprise) at UNSW Business School. His research focuses on the motivational micro-foundations of how people contribute to organisational success. For more information, please contact Prof. Anseel directly. A version of this post was first published in De Tijd.