George Stigler, one of the many great economists based at the University of Chicago, won a Nobel Prize many years ago for demonstrating the unfortunate fact that regulation, however well intentioned, tends over time to produce the opposite effects to that which its proponents intend. For further proof of this timeless theory, it is necessary to look no further than what is happening currently to the government bond markets, where conventional thinking is daily being turned on its head.
Time was when government bonds were rated highly for their capital guarantees, but feared for their exposure to the ravages of inflation. In an era when government spending routinely accounts for 35-40 per cent of national income, standard financial theory suggests that investors should normally pay more, the longer the term of a bond. For a so-called long bond, with a maturity date 25-30 years ahead, it is natural to assume that the risk that inflation will take away some of the real value of the principal when it is repaid increases.
Longer-term bonds traditionally, therefore, command a higher yield than shorter-term issues, in order to compensate for that increased risk. The result is a rising yield curve, with yields that are higher, the further out in time the bond has to be held. The exception to this normal state of affairs comes when short-term interest rates are being raised to prevent an economy overheating, at which point you can expect a temporary inversion of the yield curve.
A more straightforward way of expressing the same idea is to say that only an extreme optimist will lend money to a government at a rate that does not include a decent premium over the current and expected inflation rate. As a rule of thumb, long-term government bonds are generally held to be a risky buy, unless you are getting at least a 2.5-3 per cent premium over the expected inflation rate.
Yet, look around you at what is happening today in financial markets and the normal rules of thumb are being forgotten and ignored. While short-term interest rates have been rising on both sides of the Atlantic, long-term bond yields have been continuing to fall. The price of long bonds today is about 4.5 per cent in the United States and 4 per cent in the UK. To make any sense of this behaviour, you have to believe that inflation will remain at about 1 per cent per annum over the entire course of the next quarter of a century.
Yet, although the drive to eliminate inflation has been the great story of the past 25 years in all Western economies, you will not find many professional investors who think that it can now be safely ignored. Some experts still worry about the prospect of deflation, which could justify buying bonds at today's levels, but they remain a distinct minority.
The main reason behind the bizarre developments in the bond markets has been heavy buying of long-dated bonds by life companies and pension funds. These institutions, which are paid to look after the long-term savings of millions of ordinary investors in this country, have been steadily baling out of the stock market (which has been rising since its bear market nadir in 2003) and instead piling into bonds, as well as - tentatively and belatedly - hedge funds and other types of assets, such as commodities.
This is a complete reversal of their behaviour during the great bull markets of the 1980s and 1990s, when the typical pension fund had 70 per cent or more in equities and little money in bonds. (Life companies have always held larger proportions of their assets in bonds than pension funds, in order to match their longer-term liabilities, but the same trend is evident with them as well.) The proportion of equities held by the average pension funds has fallen from 66 to 45 per cent and that of life companies from 52 to 35 per cent since 1999.
There is no secret about why these purchases are taking place. The main drivers have been the bear market in shares and the efforts by regulators to require pension funds and life companies to match their future assets and liabilities more carefully. That this is so is evident from the fact that the most downward pressure on long-term bond yields has been felt in countries such as the UK and the Netherlands, which have pension funds most closely tied to equity market performance.
Yet, historical evidence suggests that the most likely return from a bond over time is related to its starting yield. A regression analysis of past experience suggests that an investor buying a US long-dated government bond can expect to make a real return of little over 1 per cent. For UK gilts, according to calculations by Barclays Capital, published this week, the expected return from long-term bonds is actually negative at these price levels.
In the words of Barclays Capital's Tim Bond: "From a total return perspective, the purchase of long-dated bonds at yields in the vicinity of 4-4.5 per cent represents risk reduction in the same way as bungee jumping without a cord lowers risk. The outcome is certainly more predictable, but not necessarily more benign."
Of course, we don't know what will happen in the next 25 years, and in the short term, it is possible that long bond yields could fall even further. Deflation is not impossible. But if you believe there is a link between demographic changes and investment yields, as many do, a more likely outcome is that yields generally (dividends and bonds) will have to rise over the next 10-20 years. (See chart.)
In that case, historians will record that, not for the first time, the investment institutions have been doing the wrong thing at the wrong time.
Their excuse will be that their own governments made them do it, but their clients will pay for it nonetheless.Reuse content