With the yield on the 10-year Treasury at historic lows, and corporate earnings at historic highs, there's a disconnect between the earnings yield on stocks and the 10-year. According to the Fed Model, the S&P 500 should go up as Treasury yields go down, but that's not happening. Using historical data, this article explores the paradox, describes the relationship as it appears in the data, and goes on to discuss the implications for the path of the S&P index.
Charting the Relationship
Robert Shiller developed the Cyclically Adjusted P/E ratio, abbreviated CAPE. It's a P/E for the S&P 500, where the earnings are the average of the past ten years, adjusted for inflation. The data, which was used for his book, Irrational Exuberance, is available on his website. The inverse of the CAPE ratio is the implied yield. That is, with CAPE at 20, the implied yield is 5%. Using data from Jan 1987 to the present, the chart plots the CAPE implied yield against the yield on the 10-year US Treasury.
The regression is a 2nd order polynomial, or quadratic equation, and has a correlation of .80. SA contributor Jeff Miller did two articles on this issue, and that's where I got the idea for the type of equation. We had a very productive telephone discussion, which was helpful to me in clarifying my thinking and refining the presentation.
What Drives It?
Forward earnings is logically the most important variable affecting share prices. As such, the use of 10-year backward-looking information is questionable. However, to the extent that expectations for future earnings differ from a long term average, they reflect either optimism or pessimism. Current estimates for S&P earnings for the nest twelve months, at 109.12, are much higher than the 10-year inflation-adjusted historical average of 59.01, and reflective of considerable optimism.
Thinking along these lines, our nicely fitted regression quantifies normal skepticism about future earnings, with the 10-year Treasury yield as a reference point.
There are three periods of time to be looked at separately. The color-coded periods were included in the data analysis that developed the formula, but have been flagged for consideration of what factors were in effect that affected market levels and perception.
From October 2008 to August 2009, the markets did not believe that future earnings would have any resemblance to past averages. Skepticism is a weak word; panic is more accurate. The data points lie well above the regression line.
From April 1998 to June 2001, considerable credence was given to the proposition that future earnings would be substantially higher than historical averages. The data points lie below the line.
At the present moment, we have a new outlier, shown with a star. It lies well below the line. To the extent that Treasury yields are driven by fear, the current level of the S&P 500 is actually reflective of considerable credence given to the idea that forward earnings will be well above historical averages.
The Fed Model
The "Fed model" is a theory of equity valuation that has found broad application in the investment community. The model compares the stock market’s earnings yield (E/P) to the yield on long-term government bonds. In its strongest form the Fed model states that bond and stock market are in equilibrium, and fairly valued, when the one-year forward-looking earnings yield equals the 10-year Treasury note yield (Y10):
Using 109.12 estimated forward earnings, and 2% for Y10, the equilibrium value for the S&P is 5,456. Adding a 5% risk premium, the equilibrium value would be 1,558. In discussing the present level of the index, the Fed Model and the CAPE model derived above can be compared, as is done in the next chart.
Implications for the S&P 500
Using the relationships presented above, it's possible to develop a historically based normal level for the S&P 500, as a function of the 10-year Treasury yield. Here's the chart, which also includes an equilibrium level using the Fed model and a 5% risk premium:
The dual mandate calls for monetary policy to be directed with the aim of fostering full employment and controlling inflation. Nevertheless, a substantial number of market participants believe that monetary policy is directed at increasing share prices by maintaining low interest rates. If such an objective is actually being pursued, the data developed cast doubt on its possible achievement.
Given that the historical relationships described here are counter-intuitive and perhaps illogical, it's possible they won't hold true going forward. If the Fed keeps long term interest rates artificially low, there is no reason for investors to stick around. They can flee to the equities markets and get decent returns, based on earnings.
To the extent that European money has sought refuge in treasuries, a credible resolution of the sovereign debt crisis would cause both Treasury yields and share prices to rise. An uncontrolled systemic event arising from the crisis would likely push the S&P down to something in the 900 area.
I question the path of monetary, fiscal and regulatory policy in the US. Specifically, I would be more optimistic if monetary policy were limited to providing a stable supply of money and a reasonable rate of interest to savers. Fiscal policy might include stimulus laser-focused on job skills, employability, and necessary investment in infrastructure, means of production, and new technology. Regulation could focus on outlawing naked CDS, restoring Glass-Steagall, and returning banks to their legitimate and necessary functions as financial intermediaries between those who work, save and invest, and those who use capital to create and produce value in the real economy.
Supply and Demand
The supply of AAA debt has been greatly reduced by the 2008-2009 financial crisis and its aftermath. Securitizations of RMBS and other questionable collateral, including synthetic assets, created an immense supply of bogus investment-grade bonds. That supply was called forth by demand, under relatively benign economic conditions.
The number of AAA rated corporations has declined.
The eurozone financial crisis is in the process of further reducing the supply of AAA debt. If the US can be downgraded to AA+, many sovereigns in Europe have debt to GDP as bad as, or worse than, this country. The ongoing debate about bailing out Greece and other PIIGS could be considered equivalent to the US debt ceiling circus as a precipitating political incident. Sovereign states inevitably prop up systemically important financial institutions when they come under attack, potentially straining their own solvency.
Europe no doubt has CDOs stuffed with synthetic collateral. As the situation plays out, the nature of the CDS contained therein will become more apparent. Many of them that were highly rated may not stand up to the test of time and circumstances, further reducing the supply of AAA rated debt.
The low yields on US Treasuries are driven in part by a long-term supply and demand imbalance. There hasn't been enough AAA debt to meet demand for some time. Temporary imbalances driven by fear exacerbate the situation.
The supply of shares in high-quality US large caps is also diminishing, as buybacks become an attractive use of excess capital. There are many companies that reliably pay a dividend in excess of 2%, increasing it annually. Some of them are also buying back meaningful amounts of their own shares, which may be viewed as an additional return on investment.
There is a growing movement in favor of dividend-paying stocks as a viable substitute for bonds in conservative income portfolios. "Core" type mutual funds, normally large holders, have seen capital outflows that are unlikely to continue indefinitely.
Large US corporations, trading at low P/E multiples and enjoying the ability to issue highly rated debt, will eventually accept the invitation to financial engineering. It's too easy: Borrow money and buy back shares.
It will be a while before securitization can step into the breach and create more AAA bonds from less highly rated collateral. Other forms of financial engineering may evolve in order to take advantage of the opportunities presented by the current discrepancies.
How it Might Play Out
I believe supply and demand considerations will keep the yield on AAA debt, to include the US 10-year, at relatively low levels. There wasn't enough of the genuine article back in 2005 through 2007, when confidence was much higher than it is today.
US equities of the stable earnings and cash flow type, and especially dividend payers, are in diminishing supply due to buybacks and consolidations. When demand returns, risk premiums will decrease or even go negative, and their price action will get closer to the Fed model than it has been in recent years.
The losses from the 2008-2009 financial crisis will recede into history. Investment methods that use backward-looking data are designed to keep these risks in mind. However, as a practical matter, somewhere between 3 years and 5 years out, investors start to look past historical losses. When this occurs, the CAPE ratio will increase.
A Greek default or other systemic events in Europe may defer the outcome, but are unlikely to prevent it. As mentioned earlier, a Greek default could push the S&P 500 down to 900 on a temporary basis.
The recent move by the Swiss National Bank to establish a peg for the Swiss franc precipitated a large downward move in that currency, severely straining its credibility as a safe haven.
All of this taken together suggests large cap US stocks, particlurly those with secure and increasing dividends, as an alternative safe haven, one which also has the potential of long term growth. Looking at dividend yields, Procter & Gamble (PG) yields 3.36%, and 3M (MMM) yields 2.77%. There are many others. Some P&C insurance companies hold high quality fixed income assets, and pay attractive dividends. Examples would be Travelers (TRV) at 3.36% and Chubb (CB) at 2.58%.
Additional disclosure: I am net long US equities.