Valuation-related measures are often lousy instruments for timing the market. But that doesn’t mean they can’t hold interesting information about equity market returns. In the chart below, I have plotted the development of earnings per share (EPS) growth (blue bars) against the returns on some of the major equity indices (black diamonds) from the end of May 2007. May 2007 is used as a starting point as this marks the prelude to what will become the financial crisis of 2008. You could interpret this as the last point that everything was supposed to be ‘normal’.
Click to enlarge
What stands out immediately is that the differences in EPS growth can be huge, despite the ever more integrated world economy. At the one extreme, EPS of Chinese companies have on average increased 87% since May 2007, while on the other extreme current EPS of Brazilian companies have almost completely evaporated. I’ll be the first to admit that earnings growth of Chinese companies, or any emerging country for that matter, can be difficult to comprehend at times. Also, the chart is very sensitive to the moment the EPS numbers are calculated. But, even when this is taken into account it is fair to say there has been a massive gap between Chinese and Brazilian earnings.
If we turn to the equity returns of these two countries the differences are, again, remarkable. While Chinese companies are reporting earnings that are much higher than back in 2007, the Chinese equity market has lost close to half of its value. Brazilian stocks, however, managed to squeeze out a marginal positive return of 1%, which is pretty decent considering the enormous loss in earnings.
Although, the differences between China and Brazil are enormous, they are fairly easy to explain. It has everything to do with valuation. To keep it simple I did not take into account any possible differences in equity dilution, as this probably only has a minor effect. In May 2007 the P/E of the Chinese market was roughly 40, against a modest P/E of 13 for the Brazilian market. With hindsight it is not daring to state that China went into the financial crises hugely overvalued while Brazilian stocks were more reasonable priced. The same holds for the other outlier in the graph, Russia. Russian companies also experienced a relative strong EPS growth, but Russian stocks sank 14%. The story here is the same as for China. Even though Russian stocks are cheap measured by its absolute P/E, the multiple almost halved from May 2007.
However, this is not the most interesting part of the graph. It does not reveal any relationship between EPS growth and equity returns. But this changes if we look at the other countries in the sample. Data from the other countries show it is much harder to use multiple expansion or compression as the explanation for the (distorted) relationship between EPS growth and equity returns. Let’s take Spain and the US as an example. Since May 2007, EPS growth in the US has been roughly 17%, while the EPS growth of the Spanish equity market imploded by 90%. Meanwhile stocks have risen by 6% in the US, while Spanish stocks lost 46%.
So, what about the P/E multiple in this case? Well, it can’t explain the differences. In May 2007 the Spanish equity market traded at a P/E of 15. At the same time US equities were valued at a P/E of 18. Hence, US stocks were more expensive than their Spanish counterparts, not cheaper. This is the case most of the time, however, mainly because of differences in sector weights. But, if we take into account that both markets, US and Spain, are currently trading at about the same P/E as in 2007, which is also close to their long-term average, changes in P/E multiple should be ruled out as explanatory variable. Contrary to China and Brazil, this implies EPS growth and (long run) equity returns are positively correlated.
In fact, this is the main conclusion for many other countries in the sample, as well. Change in the P/E multiple is only of minor importance, which confirms the positive relationship between EPS growth and returns. Higher EPS growth is accompanied by higher stock returns, and vice versa. To make this more tangible let’s look at the same chart with the relationship between the EPS growth and equity returns, but this time it only includes countries for which changes in P/E have been benign during the data sample. The graph reveals a clear relationship between the two factors. Only Germany seems to be a little off, but in general countries with higher EPS growth experienced higher returns.
Click to enlarge
Valuation measures are often poor instruments for market timing. Nevertheless valuation-related factors, like EPS growth, do hold some interesting information in relation to equity returns. In the long run equity markets that exhibit stronger EPS growth tend to outperform. This relationship increases the importance of accurate EPS growth estimates for different markets.