FIGURE 15.1. “Housing Bubble” Google Search Queries, 2004–19
Internet searches shot up just before the world financial crisis of 2007–9; news media response was partly delayed. Source: Google Trends.
In 2005, during the housing boom that preceded the 2007–9 financial crisis, Web searches for housing bubble increased dramatically. The curve, shown in Figure 15.1, resembles the Ebola epidemic curve (see Figure 3.1). Something very contagious was clearly happening then. Some tried to capitalize on the boom, not just by flipping homes but also by promoting the boom. Enthusiasm for real estate investments infected a significant portion of the population. In 2005, Trump founded a business school, Trump University, saying, “I can turn anyone into a successful real estate investor, including you.” Trump’s timing was bad—the Economist ran a cover story on June 18, 2005, about the prospect of a bursting housing bubble.21 Trump University went out of business right after the world financial crisis, in 2010, amidst cries of fraud and deceit.
The Housing Market Today
Since 2003, I have collaborated with my late colleague Karl Case and now with Anne Kinsella Thompson to conduct an annual survey of recent homebuyers in four US cities. The survey is conducted under the auspices of the Yale School of Management. One of our questions is “In deciding to buy your property, did you think of the purchase as an investment? 1. Not at all; 2. In part; 3. It was a major consideration.” The percentage who answered, “It was a major consideration” peaked at 49% in 2004. The percentage choosing that answer fell to 32% in 2010, just after the world financial crisis, and by 2016 it had risen to 42%.
The survey also asks about the general level of conversation about the housing market. Specifically, we ask, “In conversations with friends and associates over the last few months, conditions in the housing market were discussed (circle the one which best applies): 1. Frequently; 2. Sometimes; 3. Seldom; 4. Never.” The percentage who answered, “Frequently” reached a high of 43% in 2005, the end of the 1997–2005 boom. By 2012, the percentage choosing “Frequently” reached a bottom of 28%, significantly below the number during the boom periods. The likely interpretation is that the contagion rate for housing market narratives had decreased, and that indeed the decline in home prices could be viewed as the end of an epidemic.
What were the narratives in spring 2005? ProQuest finds 246 stories with the phrase housing bubble from March to May 2005, before the cover stories in the Economist and other places. One of these stories included a statement from Alan Greenspan, who said that he saw “a little froth” and an “unsustainable underlying pattern” in the housing market. This statement was then compared with his “irrational exuberance” speech about the stock market in December 1996. Between 2005 and 2007, there were 169 news stories with both Greenspan and froth in them. It was a colorful, quotable story featuring an economic celebrity. It contributed to a colorful, and quotable, constellation of narratives, among them narratives with the power to change economic behavior and to bring on a financial crisis.
We turn in the next chapter from real estate to the stock market, to chart another powerful narrative, putting the stock market at the center of the economy. We shall see some similarities between the narratives, both contagious in the context of perceived grand opportunities for investors, both intertwined with stories of investor greed and foolishness.
Chapter 16
Stock Market Bubbles
Narratives about stock market bubbles are stories about excitement and risk taking, and about relatively wealthy people who buy and sell securities. Like the real estate narratives discussed in chapter 15, narratives about stock market bubbles are driven by social comparison. Because they are fueled by psychology, and because stock prices are related to general confidence, these narratives also relate to the confidence and panic narratives presented in chapter 10.1 But the stock market is different from the economy as a whole. Therefore, the narratives that create and sustain stock market bubbles constitute another distinct constellation of narratives, with a different path and different sources of contagion.
A Narrative Is Born
The word crash quickly became associated with the one-day stock market drop on October 28, 1929, along with a slightly smaller drop on October 29, 1929, and it became inextricably linked to the Great Depression that followed. Crash calls to mind reckless or drunk drivers or race cars pushing their limits, and the crash narrative typically implies that a period of exceptional boom, of crazy optimism and maybe even reckless and immoral behavior, preceded the crash. The narrative of human folly expressed in a stock market boom followed by a horrendous stock market crash is still very much with us today.
The atmosphere of speculation in the 1920s was unsurprisingly associated with a technological advancement: the Trans-Lux Movie Ticker (also called the ticker projector). First mentioned in the news in 1925, and proliferating after that in brokerages, clubs, and bars, the ticker projector was invented amidst the public excitement about the stock market. The projector showed the latest trades in the stock market on a screen large enough to be seen by a substantial audience. Watching the information displayed by the projector was like watching a movie, or, as we would say today, like watching a large flat-screen television. A crowd could gather at one of the tickers, thus encouraging the contagion of stock market stories. According to an Associated Press account in 1928, the movie ticker brought in “wild trading”:
This has whetted the speculative appetite of thousands and created many new ones, the thrill of seeing one’s stock quoted at advancing prices on a heavy turn-over being akin to that of the race track devotee who sees the horse on which he has placed his bet come thundering down the home stretch in advance of the field.2
The persistence of this narrative helps explain the public fascination in subsequent decades, and even today, with domestic stock price indexes, which the news media display constantly. People widely believe that the stock market is a fundamental indicator of the economy’s vitality.
The word crash was not commonly attached to stock market movements before 1929, and the new use of the word became a name for a different view of the economy, that economic growth depends heavily on the performance of the overall stock market, so that the stock price indexes are taken as oracles. The phrase boom and crash had been popular in the nineteenth century, but it was used most often to refer to cannons firing, storm waves beating upon the shore, or even Richard Wagner’s music. After 1929, boom and crash went viral and usually described the stock market.
Crash: The Breaking Point between Speculative Excess and Hopelessness
Economists still puzzle over the stock market crash of October 28, 1929, a date on which no sudden important news occurred other than the crash itself. Just as baffling, though less discussed, is the exponential growth of stock values over most of the decade of the 1920s that preceded it. The year 1929 saw the most dramatic upswing ever, with more than a fivefold increase between December 1920 and September 1929. By June 1932, the value of the market had fallen back down to below its December 1920 level.
Earnings per share also increased dramatically over the 1920s, but the puzzle is why the stock market responded so heavily to these earnings increases. It is more normal for the stock market to react hesitantly to such upswings in earnings, which are exceptionally volatile from year to year and could even fall to zero in a single year. But surely the stock market should not fall to zero because of one bad year. Nor, normally, should it rise to match earnings in one spectacular year.
The crash of 1929 is not best thought of as a one- or two-day event, though the narrative usually suggests that it was. The combined October 28–29, 1929, crash brought the Standard & Poor’s Composite Index down only 21%, a fraction of the decline over the next couple of years, and this drop was half reversed the next day, October 30, 1929. Overall, the closing S&P Composite Index dropped 86% from its peak close on September 7, 1929, to its trough close on June 1, 1932, over a period of less than three ye
ars. The October 1929 one-day drops are talked about most often, but much more noteworthy was the stock market’s irregular but relentless decline, day after day, month after month, despite the protestations of businessmen and politicians who said the economy was sound.
This narrative was especially powerful in its suddenness and severity, focusing public attention on a crash as never before in America. Certainly, the October 1929 one-day drops set records, and records always make for good news stories. In addition, there was something about the timing of this story that caused an immediate and lasting public reaction. In his 1955 intellectual history of the 1930s, Part of Our Time: Some Ruins and Monuments of the Thirties, Murray Kempton wrote:
And it is also hard to re-create that storm which passed over America in 1929, which conditioned the real history of the 1930s.… The image of the American dream was flawed and cracked; its critics had never sounded so persuasive.3
That storm was not fully unexpected. In October 1928, during the presidential election campaign and a year before the 1929 crash, Alexander Dana Noyes, financial editor of the New York Times, wrote:
An observant traveler, returning from a recent tour of the United States, remarked that conversation on the trains and in the hotel sitting-rooms, after directing itself in a perfunctory way to the political campaign, would always turn with real animation to the stock market. Another testifies that even the conversation of women which he happened to overhear, would sooner or later be absorbed in discussion of their favorite stocks. Something like this was observed in 1925, in 1920 and particularly in 1901.… In one respect, however, the present situation differs strikingly from all the others. On all these previous occasions sober financiers, perhaps believing that some entirely new economic force had upset accepted precedent, kept silence, hesitating to predict collapse of the speculation. In this present season, on the contrary, conservative opinion has frankly and emphatically expressed the unfavorable view. In a succession of utterances by individual financers [sic] and at bankers’ conferences, the prediction has been publicly made that the end of the speculative infatuation cannot be far off and that an inflated market is riding for a fall.4
Clearly, evidence of speculation was available to the public, which read about it in the news and talked about it on train cars. For example, in the year before its 1929 peak, the US stock market’s actual volatility was relatively low. But the implied volatility, reflecting interest rates and initial margin demanded by brokers on stock market margin loans, was exceptionally high, suggesting that the brokers who offered margin loans were worried about a big decline in the stock market.5
So the evidence of danger was there in 1929 before the market peak, but it was controversial and inconclusive. A high price-earnings ratio for the stock market can predict a higher risk of stock market declines, but it is not like a professional weather forecast that indicates a dangerous storm is coming in a matter of hours. Most people will heed that kind of storm warning. However, in 1929 a great many people did not heed the warning communicated by the high price-earnings ratio. After the crash, many of them must have remembered the warnings and wondered why they had not listened.
FIGURE 16.1. Frequency of Appearance of Stock Market Crash in Books, 1900–2008, and News, 1900–2019
This graph shows extreme short epidemics in 1929 and 1987 in news, with a long-lagged response in books. Sources: Google Ngrams, no smoothing, and author’s calculations from ProQuest News & Newspapers.
As Figure 16.1 shows, the stock market crash narrative shot up with such strength in 1929 that it persists today, though more in books than in newspapers. The epidemic of stock market crash, which even today generally refers to 1929, seems to have begun weakly in 1926, several years before the actual crash of 1929, but it was not taken seriously. In newspapers, there were two fast epidemics, each peaking within a year, implying very strong short-run contagion. The first assumed massive proportions in 1929 with the record 12.8% one-day drop in the Dow Jones Industrial Average on October 28, 1929, and a further drop the next day. The second started on October 19, 1987, when the Dow experienced a 22.6% drop (almost double the percentage of the October 28, 1929, drop, though falling short of the two-day drop in 1929). Apart from the 1987 drop, no other stock price movement since 1929 has been widely called a crash. Why? As we’ve seen, newspapers are very focused on records, presumably because their readers are, and 1987 was the only record one-day drop after 1929. Folklore suggests that the stock market epidemic generated extremely high contagion in 1929. We know there was high contagion in the days before October 19, 1987, too. Stories involving the news media and investors brought to mind and amplified the story of the 1929 crash.6
The 1987 crash appears to be a flashbulb memory event (see chapter 7), like a sudden bombing attack, an automobile accident, or a declaration of war, and thus it is not easily forgotten. But after decades its story no longer seems to fit into any lively narrative constellation, and hence it is no longer virulent.
The 1929 Suicide Narrative
The October 28–29, 1929, crash was another flashbulb memory event, one that may have been stronger than the 1987 event. The 1929 flashbulb memory is magnified partly by the stories of death associated with the crash. That is, stories abounded of businesspeople committing suicide.
There is some question whether the crash really led to these suicides or whether writers learned that blaming business conditions for suicides just got a greater reaction from readers. In his best-selling 1955 book The Great Crash, 1929, John Kenneth Galbraith argued that there really weren’t many more suicides after the crash.7 But there really were many narratives about such suicides, with twenty-eight such stories in ProQuest News & Newspapers in November 1929 alone. The principle of psychology called the affect heuristic, discussed in chapter 6, predicts that such narratives make people temporarily more fearful about everything.8
The narrative of death at the time of the 1929 crash was reinforced by many stories of people who were financially “ruined” by the crash and therefore had no reason to continue living. Two months after the crash, a newspaper article in the Louisville Courier-Journal implored:
Don’t Shoot Yourself!
With amazement I read of men who kill themselves at 50. The stock-market crash has ruined them—but only financially.
Have they not the same brains that made the money for them?9
In 1970, Studs Terkel published Hard Times: An Oral History of the Great Depression, which was based on Terkel’s interviews with people who were of retirement age when Terkel was researching the book. The interviews reveal how the 1929 narrative had evolved in the interviewees’ memories after forty years. Suicide and 1929 came up frequently, along with embellishments and obvious exaggerations. One interviewee, Arthur A. Robertson, the chairman of the board of a substantial company when Terkel interviewed him, was thirty-one years old in 1929. Robertson said:
October 29, 1929, yeah. A frenzy. I must have gotten calls from a dozen and a half friends who were desperate. In each case, there was no sense in loaning them the money that they would give the broker. Tomorrow they’d be worse off than yesterday. Suicides, left and right, made a terrific impression on me, of course. People I knew. It was heartbreaking. One day you saw the prices at a hundred, the next day at $20, at $15. On Wall Street, the people walked like zombies.10
Knud Andersen, a painter and sculptor, recalled:
When the shock of losing what you had worked for comes, I found refuge in my art. To stew in a deplorable situation … where people were affected … some to suicide … I lost myself in my art. The pain that came with economic loss, I felt would pass. These things, like the eclipse of the sun.… People first observed it and committed suicide … not realizing that this would pass.11
Julia Walther, the wife of a businessman in 1929, said:
When the Crash came, the banks withdrew their support, stock held on margin was called in. Fred, unable to meet this in the falling market, lost everything he had. H
e was completely wiped out. Fred always laughingly said, “The only million dollars in my life I ever saw were those I lost.”
I felt the fever period was unreal. And the Depression was so real that it became unreal. There was a horror about it, with people jumping out of windows.12
The 1987 epidemic in Figure 16.1 looks far stronger than the 1929 epidemic. The 1987 epidemic draws much of its strength from memories of 1929. Suicides were attributed to the 1987 crash too, but these stories do not seem to have formed long-term memories, for a strong narrative did not develop and there was no reinforcing story of depression after 1987. A 50% margin requirement in force in 1987, but not in 1929, meant that in the United States many fewer people were “wiped out” or “ruined” by the 1987 crash than by the 1929 crash.
Moral Narratives about 1929
How did the 1929 crash narrative achieve such strength? Ideas about morality may have played a role. The 1920s had been a time not only of economic superabundance but also of chicanery, selfishness, and sexual liberation. Some critics viewed these aspects of the culture negatively, but they were unable to make a case against this putative immorality until the stock market crashed.
Sermons preached on the Sunday after the crash, November 3, 1929, talked about the crash, attributing it to moral and spiritual transgressions. The sermons helped frame day-of-judgment narratives about the Roaring Twenties. Google Ngrams shows that the term Roaring Twenties was rarely used in the 1920s. Use of the term, which sounds a bit judgmental, did not become common until the 1930s, when the broad moral story line in the Great Depression gradually morphed into a national revulsion against the excesses and pathological confidence of the 1920s. Purveyors of morality likened the one-day event on October 28, 1929, to a lightning bolt from heaven.
Narrative Economics Page 26