Deflation threat for the US stock market
26 June 2002
In the latest Investment Strategy Weekly published on 24 June 2002, Edward Yardeni, chief investment strategist at Prudential Financial, suggested that the S&P 500 is more than 15 percent undervalued and recommended an asset allocation of 30 percent bonds and 70 percent stocks for the moderately aggressive investor. That seems rather optimistic, and in direct contradiction with numerous other analysts who consider the US stock market overvalued.
Yardeni's valuation model for the S&P 500 assumes a fair-value forward P/E ratio that is equal to the reciprocal of the 10-year Treasury bond yield. With the yield at 4.8 percent at the time of his writing, the fair-value P/E ratio works out to be 20.8, higher than the market's forward P/E of about 18.
However, apart from the problem of using forecasted earnings which may not be accurate, there is a potentially serious theoretical deficiency in the above valuation model. The model inevitably results in a rise in the fair-value P/E when bond yields decline and a decline in the P/E when bond yields rise. Historically, this relationship has been true since the 1970s. However, the relationship did not hold in the 1950s and 1960s, when yields rose and P/Es also rose -- or at least, failed to decline -- at the same time.
In fact, there is theoretically no reason for the relationship to hold in general. Periods of high yields are often associated with high economic and earnings growth rates, which, in the absence of high yields, would actually justify high P/Es. Low yields often arise in conditions of economic distress or deflation, which, if not for the low yields, would justify low P/Es. In other words, bond yields and earnings growth rates often move together such that they tend to cancel each other in stock valuation calculations.
However, the relationship could theoretically hold in two opposite conditions: when inflation is accelerating and growth decelerating (thus justifying both high bond yields and low P/Es), and when inflation is decelerating and growth is accelerating (thus justifying both low bond yields and high P/Es). Such conditions, while not rare, are not quite the norm, but they happen to apply to the 1970s (accelerating inflation and recession), 1980s and 1990s (decelerating inflation or disinflation and economic boom).
Such conditions may no longer exist. Many economists believe that we are in a period of deflation. This deflation is being attributed to industrial overcapacity -- partly the result of an investment binge during the stock and high-tech bubble of the late 1990s -- and the impact of China's increasing participation in the global economy as a low-cost producer.
In such a condition, bond yields and interest rates in general will be low, but industrialised countries will see little growth to justify a high market P/E, and Yardeni's valuation model will fail.
In other words, the S&P 500 is not undervalued if we are entering a deflationary period.
The crux of the problem is determining the economic environment that we are entering, or more specifically, the risk of deflation. The evidence appears contradictory. The Conference Board's consumer confidence index for June fell to 106.4 from 110.3 in May, while data from the US Census Bureau show retail sales falling 0.9 percent in May. On the other hand, housing starts rose 11.6 percent in May, while the composite index of leading indicators was up 0.4 percent in that month.
Other indicators may not be as clear-cut as they might appear at first sight. "Industrial commodity prices are rising, suggesting that global industrial activity is recovering nicely," wrote Yardeni in the Investment Strategy Weekly. Unfortunately for US corporations and US stocks, the increase in global industrial activity may be occurring primarily in China and other low-cost regions, and may not benefit US corporations except those with significant manufacturing capability in such places.
The hazard of using Yardeni's valuation model in a deflationary environment can be seen in the case of Japan. Low interest rates there have been associated with deflation. As a result, economic and corporate earnings growth in Japan have been virtually non-existent since 1996. Coincidentally, using Yardeni's model, the Japanese market became undervalued at the end of 1996. It has stayed undervalued ever since, and the market has underperformed its more "overvalued" counterparts in other industrialised countries.
Disclaimer: The commentaries posted here represent the opinions of the author at the time of posting and should not be taken as investment advice. Readers who wish to take any investment action based on information obtained from this site should seek appropriate advice from a qualified financial adviser.