When economists’ numbers don’t add up

Illustration by Suneesh K.

With oil prices at a steep $80 a barrel, up from the sub $20 last year, there is a case for hauling up all those pundits who confidently predicted the end of the oil cycle at that time. On March 8, 2020, oil prices plunged 24 per cent to post their worst day in nearly three decades after OPEC, the worldwide consortium of big oil producers, failed to reach a deal on production cysts in the light of demand shrinkage following the Covid-19 lockdowns. This sparked a price war that ultimately saw prices fall to multi-year lows, even briefly entering negative territory.

What followed was a series of obituaries on the demise of big oil with prices headed deep South and forecasted to stay there. Instead a surge in demand as economies emerge from Covid-19, along with severe supply shortages, has led to crude prices climbing to a three-year high.

That means, several forecasters who last year confidently predicted “the end of oil”, have been proved wrong. And not for the first time.

Forecasts are the bread and butter of economists, the kind of thing they do in their sleep. Of course, in hindsight many recent forecasts by learned economists do appear to have been made while they were grabbing a few precious winks. In 1992, for instance, MIT professor and well-known economist Lester Thurow in his book Head to Head, confidently predicted that China “will not have a big impact on the world economy in the first half of the 21st century.” China is today the world’s second-largest economy with a 15 per cent share of world trade.

The 2009 financial crisis, as well as the US housing bubble that triggered it, also seemed to have escaped the predictions of many eminent forecasters. On the eve of the biggest financial meltdown since the Great Depression, they were still predicting a period of boom.

No wonder that Paul Samuelson, who knew a thing or two about economics having been the first American to win the Nobel Prize in Economic Sciences, famously remarked: “Wall Street indexes predicted nine of the past five recessions!’

To be fair, forecasting isn’t the exclusive domain of economists, though that’s how they would like it to be. Of late, virologists, bureaucrats and even politicians have joined the game. One of them famously predicted the end of the pandemic in India just before its worst and most deadly phase was about to begin, while another announced with equal aplomb that a deadlier third phase would begin by a certain date in July this year. Thankfully, three months later that phase seems to have stayed away.

Elsewhere, the cerebral minds at Goldman Sachs went to work before the last soccer World Cup in Russia. Using artificial intelligence and machine learning, the bank’s team ran two lakh models, mining team data and individual player attributes to project specific match scores. With that they simulated a million variations of the tournament and calculated the probability of winning for each team.

Its eventual prediction: quarter-finals between France and Spain, Brazil and Belgium, Portugal and Argentina, and Germany and England, with France and Brazil, and Portugal and Germany facing off in the semi-finals. At the end, Germany and Brazil would be in the finals and Brazil would lift its sixth title.

As it turned out, they got four of the eight quarter finalists, three of the four semi finalists and of course both finalists and the winner wrong. A bit more of intelligence and less of artificial intelligence might have told them that a team like Germany was in precipitous decline.

Back to oil prices and the forecasts. World oil and gas supermajor BP, in its Energy Outlook 2020 indicated that oil demand might never return to 2019 levels. A year later, OPEC just raised its forecast for global oil demand for 2022 to 100.8 million barrels a day, higher than 2019's demand level of 100.3 million barrels.

Oh, and one last thing: Bank of America's gurus are now saying crude oil could hit $100 per barrel! You don’t want to bet on that.