Ativo Capital

Rigorous Thinking

Financial and economic commentary reflecting Ativo’s world view:

Thomas Piketty and the Long-Term Investment Outlook

Wednesday, August 6, 2014


Thomas Piketty’s Capital has been one of the best selling books by an economist, the timing of the publication was propitious, since there is a lot of angst that the distribution of income and wealth has become more unequal. His central argument is that the share of income and wealth accruing to the top one percent, the top 0.1 percent, and the top 0.01 percent has been increasing because the rate of return on capital is higher than the rate of growth of GDP. He implicitly assumed that the rich reinvest most of their income from capital. That’s a factual error–the rich consume a lot of their investment income. And charities including hospitals, colleges and universities, churches and synagogues own perhaps ten percent of American wealth and spend most of their investment income.

The rich get wealthier when stock prices increase and poorer when stock prices decline. The beanstalk can’t grow to heaven and stock prices cannot grow more rapidly than GDP until the end of time. Stock prices vary widely around corporate earnings–and in the long run corporate earnings grow at the same rate as GDP. In the nineteenth century corporate earnings increased more rapidly as the corporate sector became a larger part of the economy.

Piketty made the same mistake that Jeremy Siegel a finance professor at the Wharton School of the University of Pennsylvania made fifteen or so years ago in Stocks for the Long Run when he claimed that the real rate of return on stocks has been seven percent a year for more than two hundred years.

Assume you won the lottery in 1789, when George Washington took the oath of office for the Presidency at the Federal Hall on Wall Street, and that you used all the prize money to buy stocks, and then used all of the dividend income to buy more stocks. If Siegel’s proposition that the real rate of return on your portfolio of stocks had been seven percent for the last 225 years, your annual income today from stocks would be about as large as U.S. GDP. Siegel assumed that all of the income from stocks could be reinvested without causing the rate of return on stocks to decline. His assumption violates the fallacy of composition; although an individual might be able to reinvest all of his dividend income without affecting the real rate of return on stocks, the community of investors cannot.

If you won the lottery today, what rate of return could you expect to earn on stocks, bonds, gold, and real estate? Fifteen years ago, as the Dow was approaching 6,000, three books with similar titles –“Dow 36,000”, “Dow 40,000”, and “Dow 100,000”–were published. (I am not making this up). These titles are remindful of the GDP forecast of $22,500 billion that one of my former colleagues made, he got the number right and the year wrong.

Consider the “Dow 100,000”. If the price-earnings ratio for stocks is 20, the Dow would reach 100,000 when the sum of the annual earnings of the thirty firms in the index was $5,000. And if the profit share of GDP is eight percent, than U.S GDP would be $120,000 billion. Thus far it’s glorified arithmetic, but when will GDP be seven or eight times larger than it now is?

If you won the lottery prize of $1 million today, should you buy stocks, bonds, gold, or real estate? The historical long run real rates of return have been six percent on stocks, three percent on bonds, one percent on bills, and zero on gold. If you believe that prices revert to the mean, then buy these assets when prices seem low relative to the long run averages and sell them when prices seem high relative to the averages.

The real rate of return on stocks is the sum of the dividend rate and the real rate of growth of the economy on assumption that the profit share of GDP is constant. (This statement is not quite accurate, since the universe of S&P500 stocks and the universe of GDP are not fully congruent. A significant part of the profits of U.S. firms are from production in foreign countries, and a non-trivial part of the corporate profits that are earned in the United States accrue to foreign firms–Shell, Daimler Benz, Sony, Toyota, Philips, Benetton, Nestle, Royal Bank of Canada, etc). If the U.S. real rate of GDP growth is three percent and the dividend rate is three percent, the real rate of return on stocks would be six percent.

The flacks for Wall Street fail to distinguish between the real rate of return on stocks, and the real rate of return to “outsiders” or the public shareholders. Corporate management–the “insiders”– pays itself stock options, which drive a wedge–a non-trivial wedge–between the real rate of return on stocks to both insiders and outsiders and the real rate of return to public shareholders. This wedge approaches one percent a year. (One percentage point seems like a small number, but it’s sixteen percent of the total return–a point that Piketty might have made.)

If you won the lottery today, should you use the prize money to buy bonds? The Federal Reserve’s four Quantitative Easing programs were a generous gift to bondholders and to those who refinanced home mortgages. (Bondholders profited from the increase in the price of bonds, but savers paid because interest rates on bonds were so low.) Interest rates will increase as the Fed takes away this gift, and the decline in bond prices in the next several years could be larger than the interest income on the bonds.

The current price of U.S. Treasury bonds is too high to provide a real rate of return of three percent, since the inflation rate is about 1.5 percent. The interest rate on the ten year bond is 2.46 percent, and the interest rate on thirty year bond is 3.27 percent.

If the Fed is successful in achieving its target inflation rate of two percent (a misbegotten policy objective, and a topic for another letter), the prices of the ten year bond and of the thirty year bond will decline by non-trivial amounts.

Either the effort of the Federal Reserve to achieve an inflation rate of two percent will be disappointed or the owners of bonds will incur significant losses as bond prices decline. (Many of these owners of bonds are hedged, for example, they have annuity liabilities that are fixed)

Should you use your prize money to buy gold? The U.S. dollar price of gold now is $1300 and change. The U.S. dollar price of $35 in 1970 persisted because the U.S. Treasury was willing to sell gold to prevent the price from increasing; somehow the Washington establishment convinced itself that it that the world would end if the U.S. dollar price of gold were increased. Some analysts at the time suggested the price should be increased to $70 so the market price of gold would bear the same relationship to the U.S. price level as in 1940. But assume that the price had been set at $100 an ounce in 1970, and that in each subsequent year the U.S. dollar price of gold had increased as the U.S. price level had increased. Then the U.S. dollar price of gold today should be in the range of $500 to $600.

Many investors now seem apprehensive about purchasing U.S. stocks, in part because stock prices are twice as high as they were at the end of 2008; they’ve increased by fifteen percent a year. Stock prices declined by more than forty percent from December 2007 to December 2008. If December 2007 is the base date, then stock prices have increased by three percent a year.

The August issue on Money magazine has an interview with Brian Rogers, the chairman and chief investment officer of T. Rowe Price, the large mutual fund group. He said, “In the context of history, the S&P 500 trading at 16 times earnings is an expensive multiple”.

Rogers’ statement is fascinating, since a price-earnings ratio of 16 is about the long run average or a nickel or so above the long term average. (His statement also is puzzling, since Barron’s reports that the price-earnings ratio on the S&P500 as 20.) His cautiousness seems representative of a large number investors.

Whether stocks are cheap or expensive depends on the prospective rate of return from owning stocks and not on the price-earnings ratio. Stocks are expensive if the prospective rate of return is significantly below the historic average of six percent; in 1998, the dividend price ratio was 1.35 percent, the price-earnings ratio was 33, and the economy was at full employment with the prospect that it might grow at two or two and one-half percent a year. Stocks were then extremely expensive–and they became more expensive in the next fifteen months. In 1985, the dividend price ratio was 3.81 percent, the price earnings ratio was 14 and change, and there was a lot of excess capacity in the U.S. economy; stocks were cheap even though the price-earnings ratio was nearly at the long run average.

The shortcoming of the price-earnings ratio as a useful metric is that it provides no forward looking information. .

What is the likelihood that owners of corporate shares will achieve a real rate of return of six percent a year for the next three years or five years? The price earnings ratio on the Dow Jones industrial shares is 15, and the comparable ratio on the S&P500 is 20; the dividend price ratio is two percent on both indexes. Moreover corporate America has been distributing cash to the shareholders from share repurchases.

Neither of the price-earnings ratios seems exceptionally high relative to the long run average of each of these ratios. If the consumer price level increases by two percent a year and real U.S. GDP continues to grow by three percent a year and the price-earnings ratios were to remain about where they are, the real rate of return to shareholders would be north of six percent. Stock prices would increase by five percent in nominal terms and by three percent after adjusting for the increase in the inflation rate, dividends would remain at two percent, and stock re-purchases would add another one to two percent.

Stock re-purchases are a tax-efficient way for firms to distribute cash, since shareholders pay tax at the capital gains rate rather than at the ordinary income tax rate. No shareholder is obliged to sell during the buyback. Stock buybacks provide firms with flexible way to vary the amount of cash provided to shareholders. (The advantages of stock repurchases are so large that one wonders why any firm continues with cash dividends.)

Several factors could lead to an increase in the real rate of return on stocks above six percent. One is that growth rate might increase; the output gap may be sufficiently large so that the growth rate might increase by one percentage point for two or three years before the economy is a full employment. The profit share of GDP could increase, but this share now is exceptionally high. Stockholders might secure an even higher return if the price earnings ratio were to increase, but this increase would be temporary because eventually stock prices will revert to the mean.

Obviously any of several accidents could derail the anticipated return of six percent. A surge in interest rates as the Fed reduces its support in the bond market could lead investors to sell stocks and buy bonds. Any of a large number of political accidents could lead some shareholders to sell and run for cash. The U.S. growth rate could accelerate and the costs and prices would increase and profits might be squeezed, which often happens toward the end of a boom.

Often at the end of an economic expansion, investors sell stocks and buy bonds. Then bond prices are low because the Fed adopted a contractive monetary policy to dampen the increase in inflation rate. The anticipation is that bond prices will increase when the Fed relaxes its contractive policy. But that history is not relevant in this cycle, since the Fed is following a policy to inflate asset prices and increase spending.


  1. Am reading Thomas Piketty’s “Capital” and find that this review is excellent! – logical and even eloquent

Comments are closed.