If the U.S. economy is so bad, why has the stock market been so good? A commonplace answer is that the Fed pushed stocks too high by pushing bond yields too low. In fact, that is why many worry that ...
There are only two ways the Fed's bond-buying spree could have pushed stock prices up, and neither is consistent with reality. One way the Fed might have raised stocks would be through improving fundamentals — namely, earnings. Unsurprisingly, stocks hit a new high after earnings hit a new high. Earnings per share for the S&P 500 last peaked near $22 in mid-2007, and then fell below zero at the end of 2008 before rebounding to about $25 by the end of March. But it would be problematic to give the Fed much credit for the cyclical rebound in earnings, considering the weakness of the recovery.
Earnings per share can rise because of stock buybacks or cost-cutting, after all, which have nothing to do with monetary or fiscal "stimulus." Cheap credit encourages cost-cutting by investing in labor-saving machinery, but that aggravates unemployment. Lower interest rates surely boosted profits of leveraged corporations. But the flatter yield curve (e.g., "operation twist") squeezed bank margins, while near-zero returns reduced the interest income of cash-rich tech firms. Lower interest rates likewise helped indebted consumers, but had the opposite effect on the incomes of seniors and other prudent savers.
In short, the net effects of quantitative easing on corporate earnings appear ambiguous at best. Yet the only other way Fed policies could have lifted stock prices would be by raising the ratio of stock prices to earnings — the price-earnings multiple. Before 2009, it would have been perfectly reasonable to expect any Fed-induced drop in bond yields to be reflected in a higher p-e multiple. Stock prices mirror the discounted present value of future earnings, which usually leads investors to bid up stock prices whenever longer-term interest rates fall. This is why it makes no sense to worry that the recent p-e ratio of about 18 on the S&P 500 is slightly higher than the long-term average of 15 without taking into account that yields of 1.8% on 10-year Treasury bonds are immensely lower than the average of 5.5%.
Let's consider how the 10-year Treasury yields compares with the S&P 500's earnings-price ratio, which is sometimes called the "earnings yield." This earnings-price (or e-p) ratio is based on 12-month trailing earnings divided by the index price, which is simply the inverse of the price-earnings ratio. Turns out that the earnings-price ratio tracked the yield on 10-year Treasury notes quite closely until 2009 when quantitative easing began.
Since then, the relationship broke down completely. The e-p ratio of 6.2 in 2012 was actually higher than it had been in any year since the bond market crash of 1994-95. More recently, an e-p ratio of about 5.5 (equivalent to a p-e ratio of 18) has heretofore been seen only when 10-year bond yields were at least 6%.
If the e-p ratio had instead fallen as much as bond yields have, then it would now be no higher than two and the p-e ratio would be above 50. Fed efforts to push down the bond yield clearly had no impact on the earnings yield, and therefore no impact on p-e multiples.
What accounts for the unprecedented divergence of stock multiples from bond yields during the Fed's experiment with quantitative easing? One possibility might be that trailing earnings artificially raise the e-p ratio during cyclical downturns, when past earnings far exceed expected earnings. But that cannot explain why the e-p ratio has been so high for more than four years.
A better explanation for the gap between stubbornly high e-p ratios and falling bond yields is that the market does not believe the low yields on long-term Treasuries are sustainable. That is, the stock market does not like bonds. That explanation requires assuming that marginal stock investors are a different clientele than bond investors, for tax or other reasons. But that assumption is clearly true in at least one respect — the Fed is buying bonds, not stocks.
Whatever the explanation, the dramatic unlinking of multiples from bond yields since 2009 strongly suggests the Fed's bond buying program did not raise stock prices through this channel. Fed policy might nevertheless have lifted stock prices by raising earnings rather than multiples, but that conjecture is difficult to reconcile with the weak pace of economic growth. It is facile to claim the economy would have grown even more slowly if monetary and/or fiscal wizards had not done whatever they did. Such ad hoc apologetics would make all policies untestable — a matter of faith rather than fact.
On the basis of facts, it is highly unlikely that the Fed deserves credit for the stock market's cyclical recovery, or for favorable effects of higher stock prices on household wealth and economic growth.