JESSEL’S PARADOX
I was first exposed to this bull market paradox as a young teenager when George Jessel, a well-known comedian of the time, told the story of a skeptical investor who decides reluctantly to invest in stocks. He started by buying one hundred shares of a rarely traded fly-by-night company. And surprise, the prices moved from $10 per share to $11 per share. Encouraged that he had become a wise investor, he bought more. Finally, when he had bid the price up to $30 per share, he decided to cash in. He called his broker to sell out his position. The broker hesitated and then responded: “To whom?” I call it Jessel’s paradox.
In an extension of Jessel’s paradox, each “skeptical” buyer gradually becomes a committed bull. The cumulative process of conversion of bears to bulls propels prices ever higher, driven in part by herd behavior. In the simple case, at the market top everyone has turned into a believer and is fully committed. There are no unconverted skeptics left to buy from the first new seller.
I am not sure whether my subsequent insights years later added much to the Jessel story. But as a measure of how far the appetite for “Jessel” risk taking beyond the securitized mortgage market had gone in the mid-2000s, long-sacrosanct debt covenants were eased as a classic euphoric global bubble took hold.24 By 2007, yield spreads in debt markets overall had narrowed to a point where there was little room for further underpricing of risk. A broad measure of credit risk, the yield spread between “junk” bonds rated CCC or lower and ten-year U.S. Treasury notes, fell to an exceptionally low level in the spring of 2007 (Exhibit 3.7). Almost all market participants of my acquaintance were aware of the growing risks but were also cognizant that risk often remained underpriced for years.
Financial firms were thus fearful that should they retrench too soon, they would almost surely lose market share, perhaps irretrievably.25 Their fears were given expression in Citigroup Chairman and CEO Charles Prince’s now-famous remark in 2007, just before the onset of the crisis, that “when the music stops, in terms of liquidity, things will be complicated. But as long as the music is playing, you’ve got to get up and dance. We’re still dancing.”26
The financial firms accepted the risk that they would be unable to anticipate the onset of crisis in time to retrench. However, they believed that those risks were limited because when a crisis was clearly at hand, the seemingly insatiable demand for their array of exotic financial products would dissipate sufficiently slowly to enable them to sell almost all of their portfolios without loss. They were mistaken. They failed to recognize that market liquidity is largely a function of the degree of risk aversion of investors, clearly the most dominant animal spirit that drives financial markets. Leading up to the onset of crisis, the lessening in the intensity of risk aversion among investors had produced increasingly narrow bid-asked credit spreads, backed by heavy volumes (the measure of liquidity) creating the illusion of being able to sell almost anything. But when fear-induced market retrenchment set in, that “liquidity” disappeared literally overnight as buyers pulled back and offers hit progressively lower bids. In fact, in many markets, at the height of the crisis of 2008, bids virtually disappeared.
As I noted in Chapter 2, some bubbles burst without severe economic consequences—the dot-com boom and the rapid run-up of stock prices in the spring of 1987, for example. Others burst with severe deflationary consequences. The more pernicious class of bubbles, as Carmen Reinhart and Kenneth Rogoff demonstrate, appears to be a function of the degree of leverage in the financial sector, particularly when the maturity of debt is less than the maturity of the assets it funds.27
Even given the excesses of the GSEs, had the share of financial assets funded by equity been significantly higher in September 2008, arguably the deflation of asset prices would not have fostered a default contagion, if at all, much beyond that of the dot-com boom.
WHY DID THE BOOM REACH SUCH HEIGHTS?
Aside from the excesses of Fannie, Freddie, and of much of the financial sector, why did the 2007 bubble reach century-rare euphoria? The answer, I believe, lies with the dot-com and 1987 bubbles, which, as I noted in Chapter 2, burst with very little footprint on U.S. and global GDP. These two episodes led many in the economics profession and many a sophisticated investor to believe that future contractions would also prove no worse than a typical postwar recession.
The need for large-bank capital buffers appeared increasingly less pressing during the period of great moderation, the years 1983 to 2007, when cyclical volatility shrunk perceptibly. The banking regulations adopted internationally under Basel I did induce a modest increase in capital requirements well before the crisis arose. But the debates in Basel that I attended over the pending global capital accord—so-called Basel II—were largely focused on whether to keep bank capital requirements unchanged or to reduce them. Leverage accordingly ballooned.
As I discussed in the Introduction, outsized leverage and financial intermediation in general collapsed on September 15, 2008, engendering possibly the greatest financial crisis ever. Five months later, with the carnage of global economic activity still under way, I reviewed the forces at play during the previous months before the Economic Club of New York. I quote extensively from that speech because it conveys the contemporaneous fear and yet the emerging critical paths to recovery as the selling climax in global equity prices was only weeks away.
We tend to think of fluctuations in stock prices in terms of “paper” profits and losses somehow not connected to the real world. But the evaporation of the value of those “paper claims” can have a profoundly deflationary effect on global economic activity. . . .
Of course, it is not simple to disentangle the complex sequence of cause and effect between change in the market value of assets and economic activity. If stock prices were wholly reflective of changes in economic variables, movements in asset prices could be modeled as endogenous and given little attention. But they are not. A significant part of stock price dynamics is driven by the innate human propensity to intermittently swing between euphoria and fear, which, while heavily influenced by real economic events, nonetheless has a partial life of its own. And in my experience, such episodes are often not mere forecasts of future business activity but are an important cause of that activity.
Stock prices are governed through most of the business cycle by profit expectations and economic activity. They appear to become increasingly independent of that activity at turning points. That is the meaning of being a leading indicator, the conclusion of most business cycle analysts.
When we look back on this period, I very much suspect that the force that will be seen to have been most instrumental to global economic recovery will be a partial reversal of the $35 trillion global loss in corporate equity values that has so devastated financial intermediation. A recovery of the equity market driven largely by a receding of fear may well be a seminal turning point of the current crisis.
The key issue, of course, is when. Certainly by any historical measure, world stock prices are cheap. But as history also counsels, they could get a lot cheaper before they turn. What is undeniable is that stock market prices today are being suppressed by a degree of fear not experienced since the early twentieth century (1907 and 1932 come to mind). But history tells us that there is a limit to how deep, and for how long, fear can paralyze market participants. The pace of economic deterioration cannot persist indefinitely.28
Shortly thereafter, the market found its bottom and began its ascent. The role of equity markets in shaping the dynamics of the broader economy is the subject of the next chapter.
FOUR
STOCK PRICES AND EQUITY STIMULUS
It was one of those episodes of childhood you never forget: my thirty-four-year-old stockbroker father trying to explain the intricacies of stock market forecasting to his increasingly befuddled ten-year-old son. As I remember it, he was very patiently trying to instruct me about how certain patterns of stock pricing behavior foretold future movements of the Dow Jones Indu
strial Average. I couldn’t wait for the tutorial to end so I could get back to perusing numbers I cared about: the batting averages of my favorite baseball heroes—the stalwarts of the New York Yankees. Many years later I still cared about batting averages, but my father’s passion finally stirred my interest. I now appreciate the challenge of short-term investing as I have grown to recognize the chart-pricing signals my father used in the 1930s—“head and shoulders,” “descending channel,” and “trading breakouts”—in many of the postwar stock market chartist services that, to this day, attract a large audience.
Over my more than six decades of intensive interest, I have encountered very few consistently successful stock price forecasting techniques. Many appeared to work for a time but subsequently failed. But are there any guideposts that at least raise the probability of investment success? Short-term investing is complex, and for guidance I defer to chartists (my father would be pleased) who argue with some validity that market momentum ebbs and flows in definable patterns. Longer-term strategies, especially “buy and hold,” however, are much more promising because stocks have risen from generation to generation almost without exception.1
The S&P 500 composite stock price index has averaged nominal annual increases of nearly 7 percent from the end of World War II through 2007, and almost never failed to log gains over any ten-year time frame. Moreover, prices exhibited no tendency, even during our most dramatic post‒World War II bear market (2008), to fall back to even the stock price highs of 1929, and during the stock market depths of early 2009, prices were still higher than during the early years of the dot-com boom. If bear markets never fully retrace all bull markets (which they do not), it necessarily means an ever increasing part of capital gains becomes quasi-permanent.
Why do stock prices rise so persistently? The bottom line is that over the long run, stock price-to-earnings ratios, and hence their inverse, earnings-to-stock price ratios, have no discernible trend going back to at least 1890 (Exhibit 4.1). The equity yield peaked in 1949 at 16 percent and ranges between 5 percent and 10 percent for most years. The reason for this relative historical stability is that these ratios are tied to interest rates that reflect our inbred stable time preference (see Chapter 1).2 Corporate earnings are tied to gross corporate product, or more generally, GDP. But GDP over the long run can be reasonably proxied by the product of the civilian labor force, productivity, and, since 1933, inflation—all of which have persistently grown. Thus, earnings per share and stock prices rise with nominal GDP. Implicit in these relationships is that common stocks, in addition to being an investment in corporate performance, act as a hedge against inflation as well. Since 1921, for example, nominal stock prices have risen by 6.0 percent annually, the product of a 2.6 percent annual inflation rate (in prices of core personal consumption expenditures) and a 3.3 percent annual rate of increase in inflation-adjusted earnings.
The equity risk premium, calculated as the earnings yield less the real riskless long-term interest rate, is a measure of the compensation spread that investors require to hold equity rather than riskless bonds (Exhibit 4.2). It is often used to assess the degree of risk perceived in holding equities. These data recommend an investment strategy of concentrating purchases when equity premiums are in the upper bound of their range (that is, when stocks become “cheap” relative to their earnings and to bonds). Such mechanical trading strategies, if rigorously followed, displace “intuition” or “gut” trading, which demonstrably is biased by fear and hence is rarely as successful as buy and hold. Even experienced investors and/or speculators have difficulty in overcoming our ingrained aversion to risk, which is a far more emotionally gripping propensity than its obverse, euphoria.3
But not all professional traders are undermined by the biases of stock fund investors. There are a handful of professional investors who have had considerable success in amassing large fortunes by, for example, shorting the British pound sterling, American subprime mortgage securities, or some exotic derivatives. It is not a trading strategy for the faint of heart.
FEAR AND RISK AVERSION ARE DOMINANT PROPENSITIES
But suppressing fear is easier said than done. It is perhaps no accident that professionally managed equity funds have difficulty doing as well as “unemotional” index funds that avoid the biases of human judgment.4 Fear, even when introspectively recognized, can undermine rational investment behavior. I recall many seasoned investors confessing to me that they sold near the bottom of the record one-day stock price decline of October 19, 1987, even though they understood that, given history, it was the wrong strategy. The physical pain they felt in seeing their net worth evaporate led them to seek relief by disengaging from the market.
THE UPSIDE
The momentum of euphoria-driven stock price booms, egged on by herd behavior, appears to have been remarkably similar through the past century, though at an increasingly dampened pace. In the five years preceding the peak of the most recent boom (October 2007), for example, stock prices rose at a 12 percent annual rate. In the five years preceding the peak of the dot-com boom (2000), stock prices rose at a 22 percent annual rate. The five-year boom that preceded the peak of 1987 produced an average rise of 24 percent, and the boom prior to the 1929 crash, an average annual gain of 28 percent. The pace of the rise in bull markets in earlier decades appears more erratic, but on average, somewhat smaller than the pace of the past eight decades.
This is probably as close as we can get to measuring the pace of the systematic buildup of euphoric herd-driven behavior. In less liquid markets we should expect the degree of tenacity and effectiveness of herd behavior to be smaller, and hence should anticipate a slower rate of euphoric growth. However, to my surprise, the average annual rate of home price increase in the five years leading up to the peak of the housing price bubble, at 12 percent, matched the concurrent pace of advances in equity prices.
THE DOWNSIDE
The inevitable contraction following all bull stock markets initiated and governed by fear has been far greater, more diverse, and less predictable than the rate of gain during bull markets. Fear, judging solely by the way markets react, is certainly a far more intense human propensity than euphoria.5
The degree of fear and euphoria is measured reasonably well by yield spreads, both with respect to credit risks and maturity. The longer the expected life of a prospective investment, the more uncertain the return and hence the greater the rate of discount applied to income from such assets. Disregarding credit risk, that discount rate should reflect the riskless yield of government bonds of the same maturity. And the difference between the yield on the thirty-year Treasury obligation and that on a five-year Treasury obligation6 indicates how severely discounting increases with maturity. The effect of heavy discounting of long-lived assets, primarily buildings, is discussed in Chapter 7.
The variance of risk on capital investment (see “From Gut to Reason,” page 85) can be proxied by the probability of default indicated by the yield spread between marginal junk bonds (S&P ten-year BB+ interest rates) and ten-year riskless U.S. Treasury notes. As can be seen in Exhibit 4.3, the spread slipped to 2 percent during the euphoric stock price and housing booms, only to soar to near 9 percent at the height of the post–Lehman crisis. Equally important, both fear and euphoria can also be measured by equity risk premiums.
THE LITTLE GUY
Fear plays an especially dominant role for investors with modest means. The risk-averse small investor confronts the bias of a gambler who has a small stake and knows if he loses it, he is out of business. His appetite for risk taking is limited. It is only when an investor’s stake is large that he, or she, can suffer substantial losses with relative equanimity. In addition, an investor with large accumulated capital has the resources to disregard periodic market crashes, and indeed usually employs such opportunities to build up his or her stock portfolio.
ASSET PRICES
Only after many years of involvement with Wall Street did I become aware of how sig
nificant a role stock prices, or more generally, asset values, play in overall economic performance. The economic collapse of 2008 reinforced what previous experience had clearly shown me: Stock prices are not merely a leading indicator of business activity but a major contributor to changes in that activity.
. . . AND THE REAL ECONOMY
Stock price gyrations have a profound effect not only on financial markets and financing but on the real economy as well. Capital gains and losses are key factors in the ups and downs of the business cycle. Most pronounced is their effect on consumer spending. Over the past six decades, the market value of all stocks held by American households and nonprofit organizations directly or indirectly through equity holdings of pension and mutual funds, insurance companies, and other financial intermediaries has risen in value by nearly $20 trillion. The historical data strongly support the view that a rise in the market value of stocks held, for example, in 401k pension funds through contributions or quasi-permanent capital gains will induce households to spend part of their gains on personal consumption expenditures.7 The empirical results developed in the appendix suggest that approximately 2.1 cents of every dollar in the equity holdings of households (on average during a year), 3.0 cents of every dollar of market value of equity in owner-occupied homes, and 2.0 cents of assets accumulated otherwise are spent, during that year, on personal consumption expenditures (Exhibit 4.4). These data do not do much to clarify whether, as many analysts believe, a dollar in home equity gains has a greater effect on consumption expenditures than a dollar of stock market capital gains. The Federal Reserve has long argued that there is no significant difference. My result does indicate a higher propensity to spend out of home equity than stock equity—but the difference is only marginally significant. In fact, data for the years 1953 to 2012 indicate that, on average, 12 perccent of personal consumption expenditures were determined by increases in net worth. The remaining 88 percent were determined by the level of disposable personal income and short-term savings rates (see Exhibit 4.5).8 Nonetheless, analysts, myself included (in this book), still measure the rate of savings of a household as a percent of income only. Before econometricians discovered these important relationships, I, and I assume others as well, assumed that only disposable income determined personal consumption expenditures.
The Map and the Territory Page 7
