>Forecasting: Relationship of trust is permanently damaged

By Chris Giles – Financial Times – October 5 2009

No one, not even the most gloomy economic forecaster, predicted the global economy’s sudden collapse a year ago. Those that said the crisis would get nasty can feel vindicated, but even they did not foresee the global crisis of confidence that followed the collapse of Lehman Brothers.

Arguably, they could not, says Bart van Ark, chief economist of the Conference Board, the global business organisation. “What you cannot forecast is a shock,” he says.

But in a less precise sense, the crisis of the past year has been the most predicted shock in history. For years, economists and officials have warned that many aspects of the world’s economic system were unsustainable.

“Even if nobody could be sure exactly when the timebomb was going to explode, it was at least useful to know we were sitting on one,” says Andrew Smith, chief economist of accountancy firm KPMG.

Most thought a crisis would arise from a loss of confidence in the dollar, resulting from the huge and prolonged US trade deficit. Instead, something similar arose after it was found that the capital flows to the US had been frittered away by banks lending money to unsuitable sub-prime borrowers and keeping those debts in their off-balance-sheet vehicles.

But though the risk of an economic crisis was regularly highlighted, it was impossible to incorporate into economic models and was therefore put to one side – filed in the “interesting and scary, but not very likely” box. That meant economic forecasts have had a terrible recent record.

Just over a year ago, in summer 2008, the International Monetary Fund thought the prospects for both 2008 and 2009 had improved and were forecasting world economic growth of 3.9 per cent in 2009 – far from a recession. The forecast was a reflection of the prevailing consensus and fears of global inflation.

Then Lehman filed for bankruptcy and economists began a painful process of recognising the likelihood of recession in the advanced world, the spread of the pain to the entire global economy and that the economic crisis was even deeper than had been thought.

By November, the IMF had almost halved its 2009 forecast to 2.2 per cent. And by April this year the Fund was talking about a “slump”, with world output sliding in 2009 by 1.4 per cent, the first drop since the second world war.

By spring 2009, the consensus of private forecasters expected the worst economic performance in the world economy since the Great Depression and little recovery in 2010.

The ever-changing outlook has raised many questions for those making policy. And yet economists have never been so uncertain about the future. It depends crucially on the financial system and its interaction with confidence and trust among people around the world.

Willem Buiter, professor of European political economy at the London School of Economics, argues that it is a fallacy to think economic models, particularly those based on history, can hope to understand the fundamental relationships in a large and complex economy. “So what is the poor economist to do when confronted with the need to make predictions or counterfactual analyses? Fundamentally, you pray and you punt,” he says.

Economists and policymakers are doing just that, alongside telling everyone that their forecasts are dependent on specific events and highly uncertain. Users of forecasts now know they cannot simply take them on trust.

In future there is also likely to be a greater tendency for economists to use their own judgment rather than the outcome of many equations run through a large computer, says Mr van Ark. “Don’t just rely on your model,” he says, “and you want to combine the short-term outlook with a longer term perspective”.

Modesty among economists is also something that will be needed in future, a soft skill that comes with difficulty to many.

But with signs that the global economy is recovering and will grow next year, the world is already hanging on forecasters’ words, however unwise that might be.

Economists will still be stuck with the problem that the unpredictable will happen. And when it does, the vast majority will have much egg on their faces.

* * *

Há paralelos muito interessantes entre o que a economia está vivendo agora, e o cotidiano da meteorologia. Ambos vivem de prognósticos sobre o futuro.

Economistas, que nos Estados Unidos gostam de atuar na área de psicologia (e vice-versa – mas psicólogos, até onde eu sei, não fazem prognósticos públicos), agora sentem na pele os dilemas e pressões só sentidas por meteorologistas.

E, com relação ao uso da experiência e do bom senso como complemento (ou no lugar de) modelos matemáticos – vide parágrafos finais do texto acima -, quando um meteorologista faz isso, é criticado por todos os lados (vide livro de Gary Alan Fine, Authors of the Storm).