Insurance Asset Management

With the benefit of hindsight, we can all see the equity valuation bubble that began to burst in March 2000. Without the benefit of hindsight, it is difficult to estimate the magnitude of overshoot we may achieve on the other side of the bubble, and when we might see a more consonant relationship of risk and return again. But the comparison of early 2000 and today shows that the ten-year periods ending in each year have one thing in common—endpoint bias. An analysis looking at a long-term history measured from one or the other point in time is overwhelmed by the magnitude of changes just prior to the point of measurement. It is no more rational to turn our backs on the rewards of investment risk now than it was to assume in 1999 that the risk of the stock market no longer existed.

A high risk premium implies, eventually, high returns to risk, as long as investments are purchased at a risk premium comparable to or greater than the premium at disposal. To put it in more familiar stock market terms, the lower the P/E at purchase (= higher risk premium), the higher the potential returns (we are speaking here of the market as a whole, not individual stocks). Economist Robert Shiller makes this very point graphically on page 11 of his book, "Irrational Exuberance", plotting a scatter diagram of each year's rolling ten-year return against the market P/E at which it was purchased.

Source: "Irrational Exuberance" by Robert Shiller, Princeton University Press, 2000, p.11. Scatter diagram of annualized 10-year returns against P/E ratios. Horizontal axis shows P/E ratio (real S&P Composite Stock Price Index / moving average of real S&P Composite earnings, over the preceding 10 years). Vertical axis shows geometric average real annual returns per year on investing in the S&P Composite Index in January of the year shown, reinvesting dividends, and selling 10 years later. Years are indicated as follows: the 19 has been dropped from the twentieth century years; the 18 has been dropped from the nineteenth-century years and an asterisk(*) has been added.

Many of the best ten year periods in the stock market began when the earnings yield was high (i.e. P/Es were low— 1919–21; 1947–50; 1980–86). P/E multiples cannot be described as low today, but given very low interest rates and what we view as an earnings trough, we find the stock market more reasonably priced than in 2000. Furthermore, with nominal yields at such low levels, the stock market will not have to deliver very high returns to outperform bonds over the coming years. In fact, this could even be achieved by earnings growth simply outpacing a slow continued multiple contraction.

The lion's share of a bond return comes from its yield. Therefore, as the yield premium for credit risk widens, the risk-bearer collects a higher running yield advantage over time. Thus the spread has to widen ever faster to negate the running yield. Ultimately, this has a gravitational effect on spreads. We view current credit premia in the bond market as unusually excessive and more likely to narrow over the next year or two. They need not narrow to their old lows to provide handsome returns.

In fact, as this article goes to press, we note that a dramatic reversal in the trend of widening spreads has occurred since October 9, 2002. This reversal has mirrored a similar reversal in the stock market.

While the growth in endogenous risk certainly gives rise to concern, the market has shown its ability to revert towards economic fundamentals when the skin of the bubble reaches its breaking point. Given the reasonably positive state of the economy, we think this is cause for some optimism.

Looking again at a chart from Part I of this series, we can expect the relationship between risk and reward to re-align itself in the future.

These are uncertain times, but we believe the fundamental principles of economics have not been repealed now any more than they were a few years ago. There is no reason to believe that the economy cannot continue to grow at a pace that will support growth and debt service capacity in excess of that priced into corporate bonds. In fact, our analysis here suggests that it is all but inevitable that these assets will become over-priced again at some point in the future. Excess returns will once again accrue to risk-takers, and these relationships will once again return to a more normal distribution. As to the timing of further reversion, we can only guess, but we remain confident in its arrival. We also admonish investors that it can be as sudden as the 15.2% rebound in stocks that has just taken place in October and November.

The world is never quite as wonderful or as bleak as it looks at one single point in time and today's risk is tomorrow's opportunity. Returns will accrue to those who do not, as Margaret Thatcher famously said, "go wobbly".