There is a standard joke among economists that equity markets have predicted about 10 of the last 5 recessions. As the joke acknowledges, equity prices embody predictions of future earnings and this implies that they also embody predictions of economic growth rates.
So, what is the rate of economic growth implied by current equity prices?
A good way to think about this is to consider why there is a difference between the current average dividend yield (annual dividends per share as a percentage of the current share price) and the real bond yield (bond yield minus expected inflation rate). This difference is required to cover two elements: the equity risk premium and the expected future rate of growth in dividends. If it is reasonable to assume that the expected rate of growth in dividends will be equal to the rate of economic growth over the longer term, the market’s expected rate of economic growth is given by:
y = (r – p) + x – d
where: y = expected real GDP growth rate;
(r – p) = real long term bond yield;
x = the equity risk premium; and
d = dividend yield.
So, it is a simple matter to calculate y if we know r, p, x and d. Unfortunately, however, there are a couple of thorny issues that need to be considered regarding appropriate numbers to use for the real bond yield and the equity risk premium.
When I last looked at this question (about five years ago) I decided that it would be more appropriate to use a long term average real bond yield than a current real bond yield. If the current bond yield is used, the results seem to become unduly sensitive to current monetary policy settings. In my calculations for Australia I used a real bond yield of 4.5 percent.
What rate of equity risk premium is appropriate? The equity risk premium is one of the few topics for which it could actually be reasonable to claim that if you laid all economists end to end, they still would not reach a conclusion. To cut a long story very short, I used the average equity risk premium implied by the relationship between GDP growth rates, average real bond yields and average dividend yields in Australia over the previous 20 years. This implied an equity risk premium of about 3.3 percent. (I am prepared to make available an unpublished paper discussing the methodology to anyone requesting it by email.)
When I did the arithmetic with the dividend yield prevailing in August 2003 (4.3 percent), I came to the conclusion that the expected real GDP growth rate for Australia implied by then current equity prices was 3.5 percent per annum. Since this was only marginally above the average growth rate for the previous 20 years, it did not seem to me to be unduly optimistic.
When I do this arithmetic now, with the current average dividend yield (6.6 percent on 18 November, 2008), it suggests that the expected real GDP growth rate for Australia implied by current equity prices is 1.2 percent per annum. That seems to me to imply that current share prices in Australia embody an unduly pessimistic view of longer term economic growth prospects.
There is a rumour going around among former work colleagues that when I was living off my earnings as an economic consultant I was heard to say, more than once, that free economic advice was not worth much. That rumour is true, but I have since changed my opinion. There is no truth at all in the rumour that I have been heard expressing the view that there are three kinds of economists: those who can count and those who can’t. I tried to say that once, but I ended up saying that I didn’t know whether I should be considered to be in the first or second category.