An independent professional's take on the latest news and trends in global financial markets

The truth about bonds

There was more interesting research out today from Dimson, Marsh and Staunton, the three London Business School academics who produce the annual Global Investment Returns Yearbook in conjunction with the Credit Suisse Research Institute. This has become established as the world’s most comprehensive database of asset class returns, covering as it does equities, bonds, cash, inflation and currencies for 19 countries from 1900 to the present day.

I will be publishing a more detailed review of this interesting study on the Independent Investor website shortly, but five points are worth highlighting here today. The main focus of the research this year is on Government bonds and the related issue of investors’ search for yield:

  • While investors look to fixed income as protection against the kind of severe losses (drawdown) which we all know characterise the equity market, history provides no assurance that this is a safe assumption. In fact, the LBS professors say, “historically bond market drawdowns have been larger and/or longer than for equities” (as anyone who remembers the period before 1980 will know from experience). US investors who bought bonds in December 1940 experienced a 67% loss in real terms on their investment over the next forty years and it was not until September 1991 – more than half a century later – that they finally became better off in real terms than they were at the outset. For UK investors who bought bonds at the peak in October 1946 their investment remained underwater for 47 years, thanks to the devestating losses produced by inflation.
  • Bonds have proven their value as diversifiers in a portfolio, but the degree of correlation between bond and equity returns (the measure of how effective they are in providing the benefit of diversification) varies sharply over time.  Broadly speaking, although there are periods when it is sharply negative, as it was nearly everywhere in the world in the first ten years of this century, in the long run the correlation is mildly positive, meaning both asset classes move in the same direction.
  • When inflation rises, as it is doing at the moment, so too does the correlation between the behaviour of the two asset classes. It is during the period of negative correlations that bond-equity diversification pays off the most, but those benefits diminish in periods when inflation is rising more than the markets expect (which was, broadly speaking, the story of the late 1960s and 1970s).
  • The performance of the bond markets since the early 1980s, when policymakers led by Paul Volcker finally decided to crack down on inflation, has moved from being very good to becoming quite exceptional for the durability of its positive returns. For the first five years after 1982, US government bonds produced record annualised real returns of 14% per annum, and over the whole period to 2008, a quite remarkable sustained annualised return of 7.7% per annum. Given how low bond yields have fallen, this rate of return will not be repeated any time soon (though the LBS trio cautiously stop short of describing it as a bubble, as I would tend to do).
  • Investors with perfect foresight about the rate of inflation one year ahead could make a handsome return from picking the bonds of countries with the lowest rates of inflation and avoiding those with the highest rates of price increases. The former, the LBS study of long term market performance clearly shows, produce significantly higher rates of return over time.

Sounds simple to capitalise on that? Unfortunately as with so many observable moneymaking opportunities in the investment world, it ain’t that easy. The reason is that forecasting inflation is no easier than forecasting future economic growth (just ask the Bank of England!). What happens in practice is that investors, quite logically, mark down the prices of those countries with the highest rates of known inflation to adjust for the higher risk involved. As a result it is the bonds of countries with the highest rates of inflation that actually produce the better real rates of return over time.

Written by Jonathan Davis

February 28, 2011 at 5:11 PM