Stephen King: The economic news you may have missed

We learnt last week that Charles and Camilla are going to get hitched, that the population of Edmonton is obsessed with flat-pack furniture and that B&Q's do-it-yourself offerings now extend to extremely quick round-the-world sailing expeditions. We also learnt how much money the Government and the Bank of England lost following Black Wednesday. And we learnt, sadly, that one of the greatest playwrights of the 20th century had passed away.

Against all this, it's not that surprising that a sudden reduction in global long-term interest rates didn't receive a lot of attention. Admittedly, the reduction didn't last very long. In the middle of the week, US 10-year Treasury yields dropped momentarily below 4 per cent before spiking back up towards 4.1 per cent. Nevertheless, the fact that Treasury yields managed to fall below 4 per cent for any time at all is remarkable in itself.

To see why, it's worth thinking about economic developments over the past 12 months. US growth has come in roughly in line with expectations but inflation has been higher, helped along by higher oil prices. The dollar has been weaker than expected and the current account deficit bigger than expected. Despite President Bush's plans, there are no clear signs that fiscal consolidation is about to take hold. The Federal Reserve has raised short-term interest rates. All reasons, one might think, for the markets to demand higher bond yields.

Yet higher bond yields simply haven't materialised. A year ago, when yields were not much different from where they are today, markets were convinced that yields should go a lot higher. After all, recovery was coming through, inflation was edging higher and, most importantly, the Federal Reserve was planning to raise interest rates. Back then, it was difficult to find anyone who wanted to buy a bond. To most people, it looked as though bond markets were a one-way bet: prices simply had to fall and yields simply had to rise.

The reason for thinking this was simple. The consensus view was - and is - that bond yields should, in some way, reflect expectations about nominal GDP growth - in other words, how fast an economy expands and at what inflationary cost. Given that the US economy is seen to have a "trend" growth rate of 3 to 3.5 per cent and given that the majority of forecasters believe inflation is likely to hover around 2 per cent over the long term, it's not difficult to see why people thought that bond yields should be moving up to over 5 per cent last year.

Economists may be good at simple arithmetic but they often have a bit more difficulty in making the more meaningful sums add up. Although it might seem reasonable to think that bond yields should reflect nominal rates of economic growth, reality is rather different. The left-hand chart shows that, for the most part, the relationship between US bond yields and nominal GDP growth has been rather tenuous. In the 1950s and 1960s, bond yields were persistently lower than nominal GDP growth and more so in the early 1970s. Then, in the 1980s and 1990s, yields were persistently higher than nominal GDP growth.

Another way to show the change in the relationship is to remove inflation from the equation and look at real interest rates relative to real GDP. As you'd expect, the picture is the same: very low real interest rates relative to the rate of real economic expansion in the 1950s and 1960s, lower still in the early 1970s and then very high real interest rates in the 1980s and 1990s.

The biggest single factor influencing the relationship between growth and interest rates is inflation or, more accurately, inflationary expectations. The bond market's failure to anticipate higher inflation in the early 1970s meant that nominal yields were slow to react to the inflationary threat, thereby leaving real yields unusually low. Equally, the belief that inflation could be lurking just around the corner, a view shared by many a scarred investor in the 1980s and early 1990s, kept real interest rates unusually high: feared inflation was persistently higher than realised inflation.

Yet these observations aren't enough to explain why bond yields can be persistently below the nominal growth rate of the economy. It's easy enough to explain the mistakes of the 1970s but it's a lot more difficult to explain the attractions of government bonds in the 1950s and 1960s. Yet some explanation is required because, in some ways, economic conditions today don't look very different from those experienced in the years before Charles and Camilla had first set eyes upon each other. Back then, bond yields remained remarkably low, seemingly on an extended structural basis.

One reason might be that, in a world of strong growth and low inflation - a world of productivity improvements, of economic catch-up, of greater integration of trade and capital flows - the incentive for governments to resort to the inflationary printing press is remarkably low. If real growth is strong, tax revenues will also be strong and, as a result, governments have no reason to think about using inflation as a default mechanism. In this kind of world, government bonds become a remarkably safe investment.

The catch with this argument, of course, is that if real growth is strong, perhaps real interest rates should be relatively high. I wonder, though, whether this necessarily has to apply to government securities. For companies that want to raise capital, there's a strong case for a higher cost of capital because, in a world of rapid growth, the hurdle rate for investment projects should be higher. But for governments, the argument isn't quite so obvious. Government bonds offer an alternative to company profits, a promise to pay creditors back through future tax receipts. And because tax receipts are dependent on the performance of the economy as a whole rather than of one, often volatile, sector, and because a government has a unique, monopoly power to collect those receipts, there's a strong case to be made that the cost of borrowing to governments should be lower than that for companies - and, if that's true, the cost of borrowing to governments should be lower than the average cost of borrowing for the economy as a whole.

Another important reason is the role of international capital flows. Given that anyone can own a US Treasury bond, the overall cost to a government of borrowing these days will depend on global savings and investment patterns, not national patterns. And, over the past 15 years, one country stands out as a low-cost provider of capital to foreign governments. Japan's deflationary stagnation initially drove Japanese bond yields down to remarkably low levels but other bond markets may now be following suit as Japanese investors diversify.

And then there's demographics. As populations age, so their risk tolerances change. When the baby boomers were in their 30s, capital gains were the thing that mattered. Now, heading into their 60s, they should be thinking more about getting a stable stream of income from their assets to live on in retirement. And bonds - government bonds, corporate bonds - fulfil that role rather better than the majority of equities. Even if the baby boomers don't see it in quite those terms, regulators and governments probably will, forcing higher demand for bonds and thus pushing their yields lower. Something for the new Duchess of Cornwall to discuss with Charles, perhaps, as they review the Duchy's investment portfolio.

Stephen King is managing director of economics at HSBC

stephen.king@hsbcib.com