Sunday, December 27, 2009
A record-breaking year - for all the wrong reasons
Posted by David Smith at 09:00 AM
Category: David Smith's other articles

crystal.jpg

One of the effects of the credit crunch and recession has been to make us even more sceptical of economic forecasts. Model-based forecasts, or even those based on hunch and guesswork, suffer from a fundamental flaw. Humans are not blessed with the gift of being able to foretell the future. The human element even in model-based forecasting is very important.

What forecasters should be blessed with, and one hopes they will be when the lessons of the past year or two have been absorbed, is better ways of telling which way the wind is blowing.

Until the credit crunch, most economists would have struggled to explain what a subprime mortgage-backed security was, let alone its significance. The shadow banking system, which turned out to be enormously important, was kept firmly in the shadows.

In future, economists will be looking harder for clues in the credit markets and will no longer take the banking system for granted. Whether any of that will make them better forecasters remains to be seen.

That said, this recession could have taken several paths. A year ago the debate was over whether the financial hurricane unleashed by the collapse of Lehman Brothers the previous September had blown itself out.

The final three months of 2008 were terrible. Could it get any worse? The answer was yes, and by enough to push most forecasts badly off course.

Britain's gross domestic product (GDP) had been expected to fall by progressively smaller quarterly amounts. As it was, the GDP fall in the first quarter alone, 2.5%, not only equalled or exceeded most people's forecasts for the full year but was bigger than any previous annual fall in the post-war period. This, as I noted last week, was the "falling off a cliff" moment for the economy.

Things did get progressively better as the year progressed, with declines of 0.7% in the second quarter and 0.2% in the third. But the first quarter did it, scuppering pretty well everybody's forecast.

At the start of the year, no forecaster predicted a bigger GDP decline than 3.2%. The outturn, according to the Treasury, was a decline of 4.75%.

I have therefore had to be a little more generous with my scoring system widening the bands as the Bank of England has done for its famous fan charts in assessing the record of economists on growth.

Inflation and interest rates were less tricky. It was clear a year ago that the Bank of England was heading into a period of very low interest rates, even though we had never before seen a Bank rate below 2%. It was also evident that inflation would be low. Those who got this one wrong thought low inflation would tip into outright deflation. That happened, briefly, for the retail prices index, but not more generally.

One interesting development, right at the death, has been the improvement in Britain's balance of payments which, taken together with the announcement of a saving ratio of 8.6% in the third quarter, suggests the economy is starting to rebalance.

Last week's downward revisions to the current account deficit, however, took the Treasury by surprise. In the pre-budget report only a couple of weeks ago it estimated that the 2009 current account deficit would be 31 billion. Now with the deficit for the first three quarters revised down to only 11.1 billion, it looks like ending up at only half that. I still think we are heading back towards a current account surplus before too long.

The smaller than expected rise in unemployment has been covered here on a number of occasions. Forecasters were not that gloomy a year ago, expecting an end-year claimant count of just above 1.8m on average. They became extremely gloomy around March, however, after a couple of huge monthly increases. As it was, the year ended with the count edging lower.

So who got it most right? As I say, I have had to widen the bands greatly for GDP because nobody came close to getting it right. But on the basis of my scoring system, congratulations are due to Peter Warburton of Economic Perspectives and the team at Hermes, the fund manager.

Warburton, whose forecasts were too pessimistic during the good times, has fared much better during the past two difficult years. Hermes, one hopes, has been making a lot of money in the markets on the back of its forecasts.

Also worth noting is that the Ernst & Young Item Club, which uses the Treasury's model, did better than the Treasury itself, though the Treasury, like other official organisations, handicaps itself by not making predictions for all the items that are in my table. In 2010 we will at least have a Treasury unemployment prediction to include.

My forecasts were for a 2.5% GDP decline, 1% inflation and Bank rate, 1.75m unemployment and a current account deficit of under 20 billion. That was worth seven points, the same as the winners but, as I always say, on the basis of my own scoring system.

A year ago, after being constantly told by readers that they could do better than the so-called professionals (and much better than me), I launched an experiment by inviting people to submit their own forecasts.

Many readers did so, so much so that in recent days I have felt a bit like an exam marker. So how did the Class of 2009 do?

One immediate observation is that amateurs, if that is what they were, were much less constrained than the pros. Not for them any clustering around the consensus. So we had plenty of predictions of GDP falls of 5% or much more.

The trouble was, and this was reflected in the scoring, if people were gloomy about one thing they tended to be gloomy about everything. So predictions of enormous falls in GDP tended to go alongside forecasts of 10% inflation and interest rates, unemployment running into many millions and current account deficits of a couple of hundred billion.

One useful thing about economic models, which may be their only useful thing, is that they pull forecasts back towards some kind of consistency. So, except in very rare circumstances, the deeper the recession, the lower are inflation and interest rates.

Despite this, many readers did well and there were plenty of threes, fours and fives, as well as zeroes, among the scores. As I was approaching the end of my marking, however, I was ready to conclude that the best amateurs were not quite good enough to beat the best professionals. Three excellent scores of six were achieved by Edward Reeve, Barry Thomas and Andy Turney but nobody was able to top seven.

Then I came across Clive Watkin's entry. He predicted a 4% drop in GDP, 1.5% inflation, a 30 billion current account deficit, unemployment of 1.4m and a 1% Bank rate. That was worth nine points. He also revealed that his family of four plus cat have been running their own forecasting competition for some years, though he did not say if the cat ever won.

Anyway, the Watkin forecast puts the professionals to shame, and some prizes will be on their way to him. As a result of his success, this is an experiment we clearly have to repeat, so please send in your entries.

From The Sunday Times, December 27 2009

The table to accompany this article is in the print edition of The Sunday Times.