Sometimes the study of economics – which has gone on for at least 250 years – can take a wrong turn. Many economists would like to believe their disciple is more advanced than ever, but in the most important economics book of 2020 two leading British economists argue that, in its efforts to become more “rigorous”, it’s gone seriously astray.
The book is Radical Uncertainty: Decision-making for an unknowable future, by Professor John Kay of Oxford University and Professor Mervyn King, a former governor of the Bank of England.
The great push in economics since World War II has been to make the subject more rigorous and scientific by expressing its arguments and reasoning in mathematical equations rather than words and diagrams.
The physical sciences have long been highly mathematical. Economists are sometimes accused of trying to distinguish their discipline from the other social sciences by making it more like physics.
Economics is now so dominated by maths it’s almost become a branch of applied mathematics. Sometimes I think that newly minted economics lecturers know more about maths than they do about the economy.
Kay and King don’t object to the greater use of maths (and I think economists have done well in using advanced statistical techniques to go beyond finding mere correlations to identifying causal relationships).
But the authors do argue that, in their efforts to make conventional economic theory more amenable to mathematical reasoning, economists have added some further simplifying assumptions about the way people and businesses and economic policymakers are assumed to behave which take economic theory even further away from reality.
They note that when, in 2004, the scientists at NASA launched a rocket to orbit around Mercury, they calculated that it would travel 4.9 billion miles and enter the orbit in March 2011. They got it exactly right.
Why? Because the equations of planetary motion have been well understood since the 17th century. Because those equations describing the way the planets move are “stationary” – meaning they haven’t changed in millions of years. And because nothing that humans do or believe has any effect on the way the planets move.
Then there’s probability theory. You know that, in games of chance, the probability of throwing five heads in a row with an unbiased coin, or the probability that the next card you’re dealt is the ace of spades can be exactly calculated.
In 1921, Professor Frank Knight of Chicago University famously argued that a distinction should be drawn between “risk” and “uncertainty”. Risk applied to cases where the probability of something happening could be calculated with precision. Uncertainty applied to the far more common cases where no one could say with any certainty what would happen.
Kay and King argue that economics took a wrong turn when Knight’s successor at Chicago, a chap called Milton Friedman, announced this was a false distinction. As far as he was concerned, it could safely be assumed that you could attach a probability to each possible outcome and then multiply these together to get the “expected outcome”.
So economists were able to get on with reducing everything to equations and using them to make their predictions about what would happen in the economy.
The authors charge that, rather than facing up to all the uncertainty surrounding the economic decisions humans make, economics has fallen into the trap of using a couple of convenient but unwarranted assumptions to make economics more like a physical science and like a game of chance where the probability of things happening can be calculated accurately.
There’s a big element of self-delusion in this. If you accuse an economist of thinking they know what the future holds, they’ll vehemently deny it. No one could be so silly. But the truth is they go on analysing economic behaviour and making predictions in ways that implicitly assume it is possible to know the future.
Kay and King make three points in their book. First, the world of economics, business and finance is “non-stationary” – it’s not governed by unchanging scientific laws. “Different individuals and groups will make different assessments and arrive at different decisions, and often there will be no objectively right answer, either before or after the event,” they say.
Why not? Because we so often have to make decisions while not knowing all there is to know about the choices and consequences we face in the world right now, let alone what will happen in the future.
Second, the uncertainty that surrounds us means people cannot and do not “optimise”. Economics assumes that individuals seek to maximise their satisfaction or “utility”, businesses maximise shareholder value and public policymakers maximise social welfare – each within the various “constraints” they face.
But, in reality, no one makes decisions the way economic textbooks say they do. Economists know this, but have convinced themselves they can still make accurate predictions by assuming people behave “as if” they were following the textbook. That is, people do it unconsciously and so behave “rationally”.
Kay and King argue that people don’t behave rationally in the narrow way economists use that word to mean, but neither do they behave irrationally. Rather, people behave rationally in the common meaning of the word: they do the best they can with the limited information available.
Third, the authors say humans are social animals, which means communication with other people plays an important role in the way people make decisions. We develop our thinking by forming stories (“narratives”) which we use to convince others and to debate which way we should jump. We’ve built a market economy of extraordinary complexity by developing networks of trust, cooperation and coordination.
We live in a world that abounds in “radical” uncertainty – having to make decisions without all the information we need. Rather than imagining they can understand and predict how people behave by doing mathematical calculations, economists need to understand how humans press on with life and business despite the uncertainty - and usually don’t do too badly.
