Stephen King: Be careful of reading too much into price changes

Supply and demand. When it comes to economics, you don't get much more fundamental than this. We know what they are. But we are a long way from knowing how to measure them. Those of you who have ploughed through your economics textbooks will have fond memories of demand and supply schedules and will remember that the equilibrium price of anything - pork bellies, cinema tickets, workers - would be found at the intersection of supply and demand. The problem, though, begins when the price changes. The price goes up. Demand has risen? Perhaps, but maybe supply has fallen. The textbooks could show you all the reasons for why prices might move. They were less helpful in telling you why, on any particular occasion, a given price would either rise or fall.

Here lies the problem. Much of what we think we know about the economic world comes from price movements. Yet price movements are, by their very nature, ambiguous. When prices change, we are forced to come up with a hypothesis - no matter how vague - for why the price change is occurring. The most obvious example is commodity prices. They are typically very volatile and people often see them as a lead indicator of broader economic activity. When commodity prices go up, people become more optimistic about economic growth because they see rising commodity prices as a sign that demand has gone up. But the oil price shocks of 1973 and 1979 can hardly be seen in the same way. Undoubtedly, commodity prices were higher but only because of a sudden restriction in supply.

What relevance does all this have for today? The simple answer is that markets have been subject to some extraordinary price moves over the past few months. Yet these price movements mean different things to different people. It's back to supply and demand. What's driving these price movements and how should policy makers react?

Oil prices are a good starting point. Before the war in Iraq, it seemed obvious to most people that a short conflict would lead to much lower oil prices. For the Western world, this would undoubtedly be a "good thing" - lower oil prices were the equivalent of a free tax cut, lowering costs to companies, leading to a terms of trade improvement for industrialised economies as a whole and, hence, kick-starting economic recovery. It's not surprising that everyone thought this; as my left-hand chart shows, the 1991 Gulf War led to a dramatic decline in oil prices that undoubtedly contributed to the end of recession - if not the beginning of decent recovery - in 1991 and beyond.

Not this time, though. Oil prices fell back a bit after the Americans took Baghdad but they're still very high. A $30 barrel was not what President George Bush had in mind when he was carrying out his cost-benefit analysis of an attack on Saddam Hussein. So how should we view this? A generous interpretation would simply be to say that the global economy is recovering nicely. With the war out of the way and business confidence heading upwards, the implied increase in global demand could quite reasonably be a catalyst for higher oil prices.

To my mind, though, this doesn't quite ring true. Oil prices today are more or less as high as they were in 1999 and 2000, at the peak of the global economic boom. On any measure, there is a lot more spare capacity sloshing around the global economy today than there was then. It would be reasonable, therefore, to think that oil prices should be a bit lower. That they're not suggests that other factors are at work. The US thought it could impose its will on the Middle East and, to date, it has failed. That failure could, in itself, justify higher oil prices. As a result, an oil price "peace dividend" has simply not materialised. Rather than a sign of recovery, I would regard higher oil prices today as a serious constraint on recovery.

Perhaps, though, this is too gloomy a conclusion. After all, oil prices are not the only price changes that we've seen. Economic indicators have improved, notably the business surveys. But the main source of excitement has been financial markets, particularly movements in bond and equity prices.

At a quick glance this would seem to be consistent with the recovery story. Since their low points, equity prices have risen a long way and bond prices have fallen a long way, taking bond yields up. Had these all occurred simultaneously, there would be a good case for saying that recovery was in the bag. Unfortunately, though, life is not that simple.

My right-hand chart shows why. It shows there have been two distinct phases in financial markets since May. The first phases I will label the "unconventional" period, when the US Federal Reserve seemed to be moving towards weird monetary policy options. Top of the list on unconventionality was the Fed's purchases of bonds. If this were going to happen, bond yields would fall. And if bond yields were going to be artificially low, there was a good case for thinking that equities should go up - which is exactly what happened.

The second phase I will label the "bond panic" period. For virtually all of this period, from the second half of June onwards, bond yields rose as prices fell. A lot of people will draw the reasonable conclusion that the further increase in bond yields strengthens the recovery story even more. As a cross-check on this story, one might have expected that equity prices would have risen at the same time.

The odd thing is that they haven't. Either it will be a profitless recovery - in other words, demand picks up but companies don't benefit, thereby leaving shareholders in the lurch - or recovery is not the only reason for higher bond yields. I am not suggesting that the economic data hasn't got better. Rather, I am arguing that other factors have conspired to push bond yields higher. The first of these, an argument I used in this column at the beginning of July, is that the Fed pulled the rug from under the bond market. By indicating that it was not ready to use unconventional measures, bonds no longer looked like such a good bet.

The second factor has been the reaction of markets themselves. The rise in bond yields has reduced the number of homeowners in the US who are prepared to refinance their mortgages: they are not going to do so if borrowing costs have gone up. For the mortgage providers, that means that pre-payment risk has gone down. Without going into the technicalities, that reduction in risk leads, directly or indirectly, to mortgage providers selling even more government bonds. It's a bit like the UK insurance companies being forced to sell equities at the bottom to maintain their solvency.

This process may have led to bond yields rising too far, in the same way that equities may have fallen too far at the height of the insurance company selling earlier this year. That, in turn, suggests caution in interpreting price moves. Unless we can be sure that we know precisely why a price move is happening - whether it's the demand or supply curve that's shifting, why the curve is shifting - we should be cautious about the message. After all, equity prices soared in the late 1990s, seemingly indicating a new age of economic wonder - and then look what happened.

Stephen King is managing director of economics at HSBC.

stephen.king@hsbcib.com