The Long-term, Real Return on Stocks is only 4-5%

Caltech Professor Bradford Cornell has written a great paper published in the Financial Analysts Journal called "Economic Growth and Equity Investing". He has performed a detailed, eloquent analysis that backs up my stock market valuation model (described here, updated many times here, and most recently conducted here).  He posits that long-term, real earnings and dividend growth is unlikely to exceed 2% (I use 1.7% for the S&P 500 based on the historical trend).  When combined with the dividend yield you are looking at a total real return of 4-5%.

Professor Cornell's paper is here: Download Economic Growth and Equity Investing I warn you, it is "wonky".

So if trend earnings growth is fixed, and the dividend yield is known at any given time, the variable in valuing the market is the assumption for future inflation.  (Projected Return = dividend yield + real trend earnings growth rate + long term inflation rate.)  Provided the long term inflation assumption is moderate, and thus interest rates are reasonably low, inflation and equity prices are positively correlated.  I explain here how sensitive the market multiple is to changes in inflation expectations.

The key to beating the market short term is determining whether the market assumption for long-term inflation expectations is too low or too high.  The rise in inflation expectations from 1.3% in March to 2.7% in January can explain virtually all of the rise in the stock market in that time.

I believe that investors should err on the side of assuming inflation will be lower than normal.  I know this conflicts with what alot of people believe given our aggressive monetary and fiscal stimulus.  The problem is, that monetary and fiscal stimulus is just offsetting aggressive deleveraging in the private market, particularly among consumers and banks.  Private deleveraging will continue until the housing market stabilizes and the banking system's leverage stabilizes.  The banking system will not stabilize until after it has digested the proposed financial system reforms.  Since the United States can not simply devalue its currency the way that smaller countries can, it can only deleverage via a period of belt-tightening.  Deleveraging and belt-tightening mean struggling with deflation.

In my previous article "These are not Unprecedented Times" I discuss the long wave pattern called the Kondratiev Cycle.  Google that term and you can learn all about it.  While I don't think it can be used as a market-timing system and I realize that each cycle is different, the pattern it describes provides a good framework for understanding what is going on.  Most economists have been building models with data that goes back to World War II and have left out a key part of the long cycle: the Kondratiev Winter.  When the Autumn-season leverage-driven asset inflation has run its course, a long period of debt and asset deflation sets in.  While stimulus may offset actual deflation, it will be difficult for inflation to be above average if the banking system is deleveraging and without a large currency devaluation.  I'm not saying it's impossible.  It's just highly unlikely.

The current market assumption for long-term inflation is 2.6%.  That is slightly above the Fed's implied target of 2.0% to 2.5%.  This with Federal deficits of 10+%, the Fed Funds rate of 0% and a large bout of "quantitative easing".  The foot can not be on the stimulus pedal much harder than it is.

I don't make short-term market calls.  Clearly, the stock market could continue to rise for a whole host of reasons.  The next recession, whenever it comes, will likely be deflationary (no more bullets are in the stimulus gun) and the stock market will be hammered anew.  Investors beware.

Q3 Market Update – A low return world

In the third quarter of 2009, we have seen some pretty significant market shifts relative to the second quarter. To refresh yourself on how my market valuation model works, please refer to this page.

Inflation and the Treasury Yield Curve

As I outlined in my May 25th post "Inflation is not a threat (yet)", I look at the treasury curve, the dollar and gold to take the market's pulse on inflation.

The TIPS spread, which is the difference between the nominal yield on bonds less the yield on the Treasury Inflation Protected Security (TIPS) has seen some interesting shifts. While the 10-year inflation rate has remained right around 1.7%, the 5-year inflation rate rose from 0.8% on May 25 to 1.7% at the end of Q2, to 2.2% at the end of Q3. The 30-year inflation rate, on the other hand, has fallen from 2.2% on May 25 to 2.0% at the end of Q3.

Embedded in these numbers is the assumption that inflation surges from zero today to average 2.2% per year over the next 5 years. Inflation is then expected to slow to 1.3% per year from 2015 to 2019, before averaging 2.1% per year for the 20 years after that. This scenario is plausible. It implies a surge in economic activity after all the stimulus currently in the pipeline, before re-succumbing to the disinflationary undertow as the economic cycle turns down several years from now.

My other near-term inflation signals are also flashing yellow. Gold has traded to slightly higher than $1000 per ounce, and the dollar is very close to the bottom of its long term trading range. They aren't yet past the point where I'll start screaming that the Fed needs to hit the brakes, but they are right at the edge.

What does a flattening of the real yield curve mean?

In my May 25th post I discussed how the proper rate for overnight money is around 2.75% if inflation is averaging 2% per year. This would deliver a zero percent after-tax, after inflation rate of return, which is what you should earn for taking no risk. That would imply an equilibrium overnight TIPS spread of 0.75%. In my equilibrium model, I have assumed an upward-sloping real yield curve of 1% for the 2-year, 1.5% for the 5-year, 2% for the 10-year and 2.5% for the 30-year. The current TIPS real yield curve is 0.5% below my "equilibrium" along the curve from the 5-year on. The difference between the 0.75% overnight real yield and the higher yields further out is driven by uncertainty regarding future inflation volatility, which increases as the time horizon gets longer. If the TIPS curve has flattened, that implies that future inflation volatility assumptions have come down.

Much of the flattening move came in the last week after Fed Governor Kevin Warsh wrote an Opinion piece in the Wall Street Journal declaring that the Fed would be vigilant about removing stimulus if inflation became a threat. In other words, they wouldn't make the mistake they made earlier this decade, when they let inflation run and were too slow to remove monetary accommodation. If a flatter TIPS curve becomes a permanent feature of the financial markets, then asset yields would come down permanently and asset values would rise permanently.

What happens to my market equilibrium assumptions?

(As a reminder, I use the yield of various Vanguard bond mutual funds for my market rates of non-treasury bonds, my treasury yield curve information is from the PIMCO web site and my S&P 500 earnings estimates come from Standard and Poors)

The rally in TIPS is catching up to the rally in the bond market generally and allowing the intermediate and long ends of the treasury, muni and corporate bond markets to be in proximity to fair value, while the short end of the curve is still overvalued.

The equity market (S&P 500) is about 22% overvalued if you feel the proper return is 7% per year. 7.1% per year would imply a 5% equity risk premium in after-tax terms to the 30-year treasury bond. The market is currently pricing in an equity risk premium of about 4.5%, which is low by historical standards, but in line with the drop in TIPS term risk premiums.

What level of S&P 500 earnings am I using?

As a reminder, I am using the long-term trend for inflation-adjusted AS REPORTED earnings. The next twelve month trend earnings number I am using is $56.56, and assume it grows at its historical inflation-adjusted rate of 1.64% plus the market long-term inflation rate of 2%. In the press you often hear a higher number for earnings, which is the operating earnings number for the next twelve months. Operating earnings allows for companies to exclude the effects of all of the poor acquisitions and perma-restructurings they conduct. The long term trend in as reported earnings gives a much more accurate view of accrual of value to the equity holders. With the S&P at 1030, the price to trend forward earnings ratio is 18, well above the long term average of about 14.  The actual top-down, as-reported earnings number projected by S&P analysts for 2009 is only $39.35.

Conclusion

It's tough to have conviction about this market. It's good that the embedded volatility premium in the market has declined, but if it rises again (which it very well could) valuations of bonds and stocks could fall a great deal. Rising tax rates could hit the valuations of stocks and bonds pretty well, too. With inflation signals flashing yellow, the Fed could very well start pulling back stimulation soon. I have made a series of adjustments to my investment model to justify the shifts occurring in the markets. Either we've entered into a permanently low-return world, in which case these markets make sense, or we're rationalizing the effects of cheap liquidity and are in for a rude awakening sometime in the not-too-distant future.

Given a lot of unexciting choices, I like muni bonds, hedged with cash, a bit of gold and crossed fingers.

Disclosure: I am not a financial advisor. Seek investment advice from your own financial advisor.