Historically, two factors have made important contributions to stock market returns in the years following U.S. recessions. One of these that we review frequently is valuation. Very simply, depressed valuations have historically been predictably followed by above-average total returns over the following 7-10 year period (though not necessarily over very short periods of time), while elevated valuations have been predictably followed by below-average total returns.
Thus, when we look at the dividend yield of the S&P 500 at the end of U.S. recessions since 1940, we find that the average yield has been about 4.25% (the yield at the market's low was invariably even higher).
Presently, the dividend yield on the S&P 500 is about half that, at 2.14%, placing the S&P 500 price/dividend ratio at about double the level that is normally seen at the end of U.S. recessions (even presuming the recession is in fact ending, of which I remain doubtful). At the March low, the yield on the S&P 500 didn't even crack 3.65%. Similarly, the price-to-revenue ratio on the S&P 500 at the end of recessions has been about 40% lower than it is today, and has been lower still at the actual bear market trough. The same is true of valuations in relation to normalized earnings, even though the market looked reasonably cheap in March based on the ratio of the S&P 500 to 2007 peak earnings (which were driven by profit margins about 50% above the historical norm).
Stocks are currently overvalued, which – if the recession is indeed over – makes the present situation an outlier. Unfortunately, since valuations and subsequent returns go hand in hand, the likelihood is that the probable returns over the coming years will also be a disappointingly low outlier. In short, we should not assume, even if the recession is ending, that above average multi-year returns will follow.
That conclusion is also supported by another driver of market returns in the years following U.S. recessions: prospective GDP growth. Every quarter, the U.S. Department of Commerce releases an estimate of what is known as "potential GDP," as well as estimates of future potential GDP for the decade ahead. These estimates are based on the U.S. capital stock, projected labor force growth, population trends, productivity, and other variables. As the Commerce Department notes, potential GDP isn't a ceiling on output, but is instead a measure of maximum sustainable output.
The comparison between actual and potential GDP is frequently referred to as the "output gap." Generally, U.S. recessions have created a significant output gap, as the recent one has done. Combined with demographic factors like strong expected labor force growth, this output gap has resulted in above-average real GDP growth in the years following the recession.
The chart below shows the 10-year growth rates in actual and potential GDP since 1949 (the first year that data are available).
The blue line presents actual growth in real U.S. GDP in the decade following each point in time. This line ends a decade ago for obvious reasons. The red line presents the 10-year projected growth of "potential" real GDP. This line is much smoother, because the measure of potential GDP is not concerned with fluctuations in economic growth, only the amount of output that the economy is capable of producing at relatively full utilization of resources.
One of the things to notice immediately is that because of demographics and other factors, projected 10-year growth in potential GDP has never been lower. This is not based on credit conditions or other prevailing concerns related to the recent economic downturn. Rather, it is a structural feature of the U.S. economy here, and has important implications for the sort of economic growth we should expect in the decade ahead.
The green line is something of a hybrid of the two data series. Here, I've calculated the 10-year GDP growth that would result if the current level of GDP at any given time was to grow to the level of potential GDP projected for the following decade. This line takes the "output gap" into effect, since a depressed current level of GDP requires greater subsequent growth to achieve future potential GDP. Notice here that even given the decline we saw in GDP last year, the likely growth in GDP over the coming decade is well under 3% annually - a level that we have typically seen in periods of tight capacity (that were predictably followed by sub-par subsequent economic growth), not at the beginning stages of a recovery.
The situation is clearly better than it was at the 2007 economic peak, where probable 10-year economic growth dropped to the lowest level in the recorded data, but again, the likely growth rate is still below 3% annually over the next decade even given the economic slack we observe.
Aside from a gradual recovery of the "output gap" created by the current downturn, there is no structural reason to expect economic growth to be a major driver of investment returns in the years ahead. With valuations now elevated above historical norms, there is no reason to expect strong total returns on an investment basis either.
0 comments:
Post a Comment