The most notable event last week was Friday’s employment report. Updating the data and estimates I presented in Stalling Engines: The Outlook for U.S. Economic Growth, The current labor force now represents 62.8% of the U.S. working age population. That participation rate is below the record high of 67% at the peak of the 2000 economic cycle, but remains several percent above the post-war norm prior to 1980. Civilian employment currently stands at 153.2 million, representing 95.6% of the labor force (160.2 million). The unemployment rate captures the remaining 4.4% of the labor force. Based on demographic factors, the Bureau of Labor Statistics projects the civilian labor force to grow to 163.8 million by 2024. Provided that the unemployment rate holds at 4.4%, employment growth can be expected to contribute just 0.3% annually to GDP growth over the coming 7-year period.
Any additional economic expansion will have to come from productivity growth. Over the past decade, productivity growth has declined from a post-war average of 2% to a growth rate of just 1.2% annually, with growth of just 0.6% annually over the past 5 years. Accordingly, barring any increase in the unemployment rate, a continuation of existing productivity trends would produce real U.S. GDP growth over the coming 7-year period ranging between 0.9% and 1.5% annually. Less likely, though not impossible, would be a full reversal of the slowing productivity trend of recent decades, restoring pre-2000 averages. This would push the prospects for real GDP growth as high as 2.3% annually.
Notably, fully 1.4% of the 2.0% average annual real GDP growth observed since the beginning of 2010 has been driven by growth in civilian employment. As slack labor capacity has slowly been reduced, the unemployment rate has dropped from 10% to just 4.4%. That jig is up. As a result, the U.S. economy is likely to hit stall speed even in the absence of outright recession.
If one assumes an inflation rate of 2% and a steady rate of unemployment at 4.4%, nominal annual growth in GDP and corporate revenues would be expected to average between 2.9% and 3.5% annually in the coming years. Since this estimate rules out the possibility of recession, the resulting growth range would be a modest acceleration from the past decade, when S&P 500 revenues grew by only 1.6% annually, and would be roughly similar to the past 20 years, during which S&P 500 revenues have grown by 3.1% annually. The corresponding average annual growth rates for reported S&P 500 earnings, even including the impact of share buybacks, were 1.9% over the past decade and 4.7% over the past 20 years.
In recent decades, S&P 500 earnings growth has outpaced revenue growth because of an expansion in profit margins, which has been closely tied to the failure of real wage growth to keep pace with even the weak rate of productivity growth (or equivalently, growth in unit labor costs at a slower pace than growth in the GDP deflator). See This Time is Not Different, Because This Time is Always Different for more detail on these relationships. With the unemployment rate now compressed to just 4.4%, we’ve already observed a downward reversal in the corporate profit share of GDP. Just a modest normalization of corporate profit margins by this mechanism is likely to drive S&P 500 earnings growth below zero in the coming years, even if revenues continue to expand, and even assuming no recession whatsoever.
[Geek’s Note: Denoting wages W, prices P, output Q, and labor L, and de-trending to reflect the fact that real wage growth has historically lagged productivity growth by about 0.4% annually, we find that corporate profit margins reliably increase when real wages meaningfully lag productivity growth: %W-%P<%Q-%L-0.4%, or equivalently, when unit labor costs meaningfully lag inflation in the GDP deflator: %W+%L-%Q<%P-0.4%, or also equivalently, when the share of nominal wages to nominal GDP falls rapidly: (%W+%L)-(%Q+%P)<-0.4%. Since weak wage growth tends to be associated with weak household income and savings, coupled with government deficits to stabilize income, we also observe that high corporate profits tend to emerge as a mirror image of high combined deficits in household and government saving].
Put simply, neither the prospects for economic growth nor the prospects for earnings expansion are likely to benefit from the same dynamics that they enjoyed starting from a 10% unemployment rate in 2009. What “Main Street” perceives as a continued weak economy has far less to do with jobs per se than with the growing income disparities (produced by low real wage growth that has its mirror image in high profit margins), along with declining contributions to GDP growth from population growth and productivity. The arithmetic of GDP growth isn’t going to change, but with a tightening labor market, the skewed division of income between wages and profits has already quietly started to normalize.
As for the stock market, earnings-based measures of market valuation have been severely distorted by weakness in labor compensation in recent years, but the resulting amplification of profit margins is unlikely to persist in the face of a compressed unemployment rate. The historical record is very clear that long-term S&P 500 total returns are much better correlated with revenue-based valuation measures than earnings-based measures. That’s because variations in profit margins weaken the usefulness of earnings as “sufficient statistics” for the very, very long-term stream of cash flows that investors are actually buying when they invest in stocks (see Exhaustion Gaps and the Fear of Missing Out for a comparison of the relative reliability of various market valuation measures).
On the valuation measures most reliably correlated across history with actual subsequent S&P 500 total returns, the S&P 500 is currently 135% to 165% above levels that would be associated with historically normal S&P 500 total returns. Given the still-depressed level of interest rates, we can easily concede that valuations may not reach or break below these norms over the completion of the current market cycle. Still, every cycle in history, including those featuring similarly low interest rates, has brought valuations within 25% of those norms or below. That alone would imply a market decline on the order of 50% over the completion of the current cycle.
Notably, we do not at all require a market retreat of that magnitude as a prerequisite for establishing a constructive or aggressive market outlook. The strongest market return/risk profiles we identify are associated with a material retreat in valuations that is joined by an early improvement in our measures of market action, particularly as measured by uniformity across a broad range of market internals. That combination would prompt us toward a more favorable outlook, even in response to a market decline that ends well above historical norms.
At present, we would not risk a constructive or aggressive market outlook in the current hypervalued market, featuring the most extreme overvalued, overbought, overbullish syndromes we identify, and most importantly, coupled with still-divergent market internals on our measures. That said, an improvement in the uniformity of market internals, even here, would encourage us to suspend the immediacy of our negative market expectations and instead shift to a more neutral outlook. We’ll respond to changes in market conditions as they emerge.
Finally, I should note that while a century of history gives us very strong measures by which to estimate prospective returns in stocks and bonds (for Treasury bonds, prospective returns can be read directly from the yield to maturity), it’s also true that the beauty of those prospective returns is in the eye of the beholder. If investors are comfortable with a 2.4% annual return prospect on Treasury bonds over the coming decade, then from theirperspective, bonds can reasonably be considered “fairly valued.” In my own view, bond yields are likely to visit higher levels over the coming years, but we don’t view upside pressures as immediate, and in the face of low global rates and extreme valuations in other asset classes, our outlook on bonds is actually modestly constructive. My impression is that the main impact of inflation pressures would be to reduce real interest rates rather than drive nominal rates higher, which is an environment that would tend to favor precious metals as well.
By contrast, my view is that investors can only consider the S&P 500 to be “fairly valued” if they are comfortable with negative expected 10-year nominal returns of about -2% annually, and roughly zero expected total returns over the coming 12-year period, coupled with a reasonable prospect of losing half of their investment in the interim. For investors who are uncomfortable with those prospects, I continue to believe that a far stronger prospective return/risk profile is likely to emerge over the completion of the current market cycle, most notably at the point where a material retreat in valuations is joined by early improvement in our measures of market action. Those who know my history will recall that I identified similar shifts in real-time in October 1990, April 2003, and late-October 2008, each following bear market declines, though the latter shift was truncated by my insistence on stress-testing our methods against Depression-era data. The resulting challenges, and how we addressed them, are detailed in Being Wrong in an Interesting Way. It is terrible a mistake to confuse that narrative for a license to ignore the lessons of history in a wildly overvalued and speculative market that has already lost internal uniformity.
I fully expect strong and identifiable opportunities to accept market risk at a much better profile of expected return/risk. The notion that “this time is different” as a result of central bank intervention will likely prove to be a fatal conceit. I have no quarrel with disciplined buy-and-hold investors who recognize the cyclical risks inherent in a passive investment strategy, and the link between valuations and subsequent long-term returns. It’s the notion that central banks have removed the risk of major loss from the financial markets, when instead they have magnified it by vastly overextending this speculative episode, that will likely end in a ball of flames.
© Hussman Funds
© Hussman Funds
Read more commentaries by Hussman Funds