Hamish McRae: Professionals failed to call this recession, so watch the risk-takers to spot the recovery

How can the world economy stage a decent recovery when hardly anyone is prepared to take on risk?

How can the world economy stage a decent recovery when hardly anyone is prepared to take on risk?

One of the disturbing features of the markets in recent weeks has been the loss of self-confidence of the business and investment communities. Businesses have continued to slash investment, or only invest in areas that are associated with labour savings and so bring immediate rewards – as companies that make capital goods, such as Marconi and ABB, have discovered to their cost. More alarming, though, has been the evidence that perfectly sound companies that would like to increase investment are prevented from doing do by their financial situation. They cannot sensibly do a rights issue because of their share price; they cannot go to the bond market because of widening spreads; and the banks are so worried about existing bad debts that they are chary of increasing their lending.

In one sense this is classic bottom-of-the-cycle stuff. What gives it a particular sting is the sense that the professionals have been humiliated in their ability to gauge risk. In previous cycles there were more excuses: the impact of the two oil shocks in the 1970s and early 1980s, and as far as the UK was concerned the distortions of German reunification and ERM membership in the early 1990s.

This time there have been fewer excuses. The professionals ought to have seen trouble on the horizon and largely failed to do so. The graphs show two areas where there have been massive failures and a third where a further failure may be looming.

The most obvious failure has been one of calling the recession. Get that wrong and you get just about everything else wrong. The left graph shows consensus forecasts for growth this year made each month from the beginning of January 2001.

The present forecasts are probably a bit too optimistic and are still being revised down. So even three-quarters of the way through a year, the professionals still cannot get the numbers right. But the shots made in the first half of last year have turned out to be seriously embarrassing. In the case of Japan it was not a question of getting the big number wrong: they even got the sign wrong, for an expectation of plus two has become minus one. Gulp. The US is down one percentage point, but with a dip in the first half of this year and then a recovery of confidence in the second half. Expectations have been gradually downgraded, which seems to me to be an acceptable level of error, unlike the more serious downgrading of growth for the eurozone.

When you look at this performance, you have to ask whether this exercise is worth doing at all. I think the answer is yes, just, but forecasters need to be more honest about the levels of uncertainty. The best model here is the Bank of England's practice of giving a range with levels of uncertainty attributed to each outcome. Thus you can see the extreme cases – the ones that have only a 5 per cent probability or less – on either end of the scale. Recent history has taught us that you have to consider the extreme possibilities as well as the mainstream ones.

Now turn to credit risk. This is something that is meat and drink to any lending institution so it is helpful to focus on the risk that markets apply to lending institutions themselves. The middle graph shows how the spread has changed this year for the 10-year bonds of two European banks over a government bond of similar maturity.

At the beginning of the year the debt of the large German bank, Commerzbank, yielded 1 per cent more. In other words, investors got a 1 per cent premium for the perceived added risk of holding the bank's debt instead of having government debt. Holders of Lloyds TSB got only half a percentage point extra, suggesting that people thought Lloyds a somewhat safer bet. Now look at the change in the last month. Lloyds is still pretty much where it was, but Commerzbank would have to pay a 3 per cent interest rate premium over and above government bonds were it to try to raise more capital. Looked at the other way round, holders of Lloyds' debt will have done OK, while holders of Commerzbank will have suffered a serious loss of capital.

I happen to think that this mis-prices the risk. Commerzbank is not going to go bust, and if it were to need a government rescue, I am 99 per cent confident that the bonds would be repaid in full. But this does raise two questions. The obvious one is: why did people not see back in June that Commerzbank might be in serious trouble? The less obvious one is: why are there so few investors who are prepared to take the risk of holding this debt, given that the bank is not going to fold?

Now look at the third graph. This shows consumer inflation in the US over the past six years, plus the GDP deflator. Most people focus on the former measure, the green line, which shows that price rises have varied between 3.5 per cent and 1 per cent over this period. There is nothing wrong with consumer prices, for those are what most of us are interested in. But the GDP deflator is a better measure partly because it takes into account price changes in the output of the entire economy and partly because it does not include distortions such as changes in land prices, which simply redistribute wealth between different parties.

I had been aware that the GDP deflator had been running below consumer inflation in the latter 1990s but had not realised until Andrew Smithers, the City economist, pointed it out, that it had dipped below zero this year. America has already experienced deflation, albeit briefly.

The legitimate question, surely, is whether the risks of deflation are being properly considered. What might the world be like were there to be a long period, say a generation, where global price levels gradually declined? This happened during much of the 19th century. Indeed, looked at historically, price stability is a more normal condition than inflation. Some of us have been banging on about this possibility for several years but only now are the professionals starting to acknowledge the risk.

The analysis of risk has become extremely sophisticated and the tools used by the financial services industry – the mathematical models and the like – are very helpful as guides. But markets also need a sense of history. People peering into the future should bolt down what they really know for sure – and be frank about what they can never know. Knowing what you don't know is more important than knowing what you do. Once people recognise that, the risk-takers will return. The rewards for taking on risk are greatest at this stage of the cycle, precisely because it is the time that other people shun it.

This is not a call to hit the roulette tables. Rather it is to give a lead indicator of recovery. One signal that recovery is secure will be known risk-takers – mainly private individuals – buying companies that others have given up on. Leicester City?