A Brief History of Doom

by Richard Vague

  Ever since 1882, when Thomas Edison first lit the homes of Manhattan customers through his Pearl Street Power Station, U.S. utility companies had been racing frantically to capture market share. Edison’s most important successor was his former personal secretary, Samuel Insull, who became one of the most powerful executives in the nation. Utilities were the hottest listed stocks—the era’s go-go stocks. At the 1929 peak, utilities were 18 percent of the New York Stock Exchange’s value, its largest single component. These stocks averaged prices fifty-seven times the earnings per share, with at least one more than ninety times the earnings per share, far above conventional levels, especially for such capital-intensive stocks.15

  Insull became the powerful chairman of Chicago’s Commonwealth Edison utility and was a master at building and buying utilities. The companies he controlled were collectively the largest supplier of power in the country. He built Commonwealth Edison into a company with $500 million in sales, using only $27 million in equity. His companies acquired a network of gas, light, power, and transit companies over thirty-two states, serving ten million people. In the late 1920s, he sought to further consolidate control of these companies using more debt, with the stock of these subsidiaries as collateral.16

  Across the industry, these highly leveraged utilities were ripe for investment speculation. “Holding company” ownership was increasingly popular: a shell company would buy operating utility companies, often with debt at both levels, sometimes referred to as “double leverage.” This was an ominous trend, given lofty stock valuations. Actually, there were holding companies that owned holding companies, creating a third layer of debt. In such a three-story house of cards, if 40 percent leverage were employed at each level, the ultimate owner could control a million-dollar company with only a $64,000 investment. This meant greater upside opportunity and greater downside risk. Acquisition debt was often more than 60 percent and, in some cases, more than 90 percent of the purchase price. This holding company activity was sufficiently widespread to attract a Federal Trade Commission investigation in 1928.17

  Merger and acquisition activity across the entire private sector reached an all-time high in 1928 and 1929, largely financed with debt and surpassing the 1890s levels that had inspired early 1900s antitrust enforcement. Insull’s biographer says of the era’s lenders:

  During the expansionist fever of the late twenties, [bankers] began throwing money at everyone who seemed prosperous; . . . At a party, the new president of the Continental Bank sidled up to [Insull’s son] and . . . said, “Say, I want you to know that if you fellows ever want to borrow more than the legal limit, all you have to do is organize a new corporation, and we’ll be happy to lend you another $21,000,000.” . . . Insull’s bookkeeper said, “The bankers would call us up . . . [and ask] isn’t there something you could use maybe $10,000,000 for?” . . . This situation had the impact of three stiff drinks on an empty stomach.18

  This passage could just as easily have been applied to the era’s real estate lenders.

  Hubris and a sense of invincibility among lenders typically run rampant in the years before a financial crisis. In the 1920s, the short-term boost to economic growth was so sweeping that then-presidential nominee Herbert Hoover boasted in 1928 that Americans were “nearer to the final triumph of poverty than ever before in the history of any” country; “the poor-house,” he claimed “is vanishing from among us.”19 Automobile executive Myron Forbes said, “There will be no interruption of our permanent prosperity,” and famed Yale economist Irving Fisher submitted that “stock prices have reached what looks like a permanently high plateau.”20

  In the legendary stock market spike during the first nine months of 1929, utilities gained 48 percent; industrials, 20 percent; and railroads, 19 percent. The Roaring Twenties had ushered in so much financial gain and optimism that Americans learned to buy stock on margin, and broker loans quadrupled from $1.6 billion in 1923 to $6.4 billion in 1928. These debt-fueled stock purchases helped power the Dow Jones Industrial Average from 94 in 1923 to 362 in 1929.

  Broker loans are made to brokers to fund margin accounts for their clients. A significant amount of the stocks purchased by broker customers during the good times were bought with margin loans of 90 percent, meaning that a customer could buy $1,000 of stock with only $100, and owe the remaining $900. However, if that stock went down to $800, the customer would immediately have to pay another $180 to get the loan back to 90 percent of the now-diminished value of the stock: this is referred to as a margin call. If, however, that customer did not have that amount of cash—and many did not in those speculative times—the stockbroker would have to sell the entire position since no fraction of that $800 could be sold to reduce the loan to less than the current value of the stock. The math is ruthless, as the world would soon realize.

  Figure 1.4. Broker Loans and the Dow Jones Industrial Average (DJIA), 1918–1938

  Margin loans exploded to 5 percent of total stock market value compared with 2 percent in the 2000s. Note in Figure 1.4 that when margin loans went up, so did stocks. Unfortunately, the converse is also true. Congress enacted a regulation to limit margin lending to 50 percent of a stock’s value in the aftermath of the Depression.

  The key period for overall private debt growth ran from 1923 to 1928, when total private loans skyrocketed by $40 billion—from $116 billion, or 137 percent of GDP, to $156 billion, or 161 percent of GDP—all in a $90 billion economy. U.S. government debt actually declined from $22 billion to $18 billion as the robustness led to temporarily high earnings and higher tax payments. The majority of this $40 billion increase came from just four categories: $13 billion in residential mortgages, an estimated $5 billion in commercial real estate debt, an estimated $5 billion in utility debt, and $5 billion in broker loans.

  Households borrowed for a lot more than housing. They borrowed for cars, appliances, land, and farms. The 1920s inaugurated an era of consumer installment debt, especially for automobiles such as the popular Model T, which were beginning to surpass trains as the preferred form of transportation.21 But even though some commentators have attributed much importance to them in this era, installment loans were still a very small part of the total debt picture, reaching only $2.9 billion.22 This figure included $1.4 billion in car loans. Overall, nonmortgage consumer debt was flat as a percentage of GDP during the 1920s and declined rapidly after 1932.

  Images of failed farmers in the 1930s Dust Bowl are indelibly part of the story of our Great Depression, but farm mortgage debt was not a part of the 1920s credit boom. It actually declined during the decade. But that masked another runaway lending story: farm mortgages had tripled to $10 billion between 1910 and 1920, a greater percentage increase than in any other sector. Farmers had been emboldened by rising agricultural commodity prices in that era, especially during global shortages caused by World War I. Eastern investors, inexperienced in farming, drove farm lending, often purchasing vast swaths of prairie sight unseen. They plowed under millions of acres of native prairie grass and planted corn, soybeans, and mostly wheat, and enjoyed several years of bumper crops. These debt-financed land purchases caused agricultural land prices to rise, as ever. However, after 1919, agricultural commodity prices plunged with restored, postwar European farm production, and the now heavily indebted U.S. farmers wrestled through the 1920s in a trap of high residual debt and low prices.

  After the 1929 crash, both farm mortgage totals and agriculture commodity prices collapsed as loans were called, and the farm industry was decimated. The initial impact on GDP was less than from the real estate sector because agriculture made up less than 10 percent of the economy; however, it provided 25 percent of all U.S. employment, and farm unemployment heavily affected overall unemployment numbers. As for the newer farms, absentee investors withdrew, leaving them unplanted. Without native grass, and now without a crop (and before the Civilian Conservation Corps’ contouring program), these lands contributed significantly to the Dust Bowl of the 1930s, one o
f America’s worst ecological calamities.

  Behind every financial crisis sit bankers and nonbank lenders who made a lot of bad loans. Banks, savings-and-loan associations, life insurance companies, bond houses, and other institutions competed vigorously for new loans, and a colorful array of lending executives grabbed the financial headlines. Prominent among them were S. W. Straus, inventor of the mortgage real estate bond business, and Charles E. (“Sunshine Charley”) Mitchell, chairman of National City Bank (now Citibank). Straus moralized against America’s wastefulness and overindulgence, even as his bond house dominated its peers when it came to financing new real estate–backed bonds that used lowered lending standards.23 Mitchell’s real estate lenders were among the most aggressive in the banking industry, searching high and low for new opportunities to lend and using generous credit standards. His salesmen buttressed the enterprise by selling millions of shares in his bank, shares that plunged in 1929.24

  Because lending growth was so extraordinary, banks led the way in the bull market of the 1920s and, though largely over the counter, had an even higher sector market capitalization than utility stocks. The Bank of New York and National City Bank were each worth more than $1 billion and stood among the top ten stocks in the country by market cap—greater than Chrysler, Standard Oil of New York, and Sears.

  Lenders relaxed their credit standards, and the more people who qualified for home loans, the higher the ratio of buyers to sellers and the higher home prices went. Rapid lending growth in the 1920s directly affected housing prices. In the mid-1920s, housing prices in Manhattan rose 32 percent.25 The 1920s was easily one of the greatest real estate booms in American history. Between 1923 and 1929, sale price per square foot for office buildings in Manhattan increased by a sizzling 70 percent. For a while, it was all a jubilant self-fulfilling prophecy.26

  Construction peaked in 1926 but remained strong through 1928. As early as 1925, however, signs of overbuilding were clear. At their 1926 convention, the president of the National Convention of Building Owners and Managers expressed concern about overproduction, denouncing speculative builders for borrowing “the full cost of construction, regardless of return,” and selling the buildings “at a profit” before moving on to erect additional structures.27

  Until the Federal Reserve raised rates in 1928, two years after the construction peak, it was not acting to curb loan growth. If anything, Anna J. Schwartz states, “The Fed may also have induced more risk taking by providing banks near the brink of failure with loans from the discount window, contravening the rule that a central bank should lend only to illiquid, not insolvent, banks. In 1925, the Federal Reserve estimated that 80% of the 259 national banks that had failed since 1920 were ‘habitual borrowers.’ These banks were provided with long-term credit. A survey in August 1925 found that 593 member banks had been borrowing for a year or more and 293 had been borrowing since 1920.”28

  With this, banks were both encouraged to lend more and to be less than diligent when it came to building their own independent deposit and funding sources. Overbuilding was so pronounced that a correction was inevitable, and overbuilding impairs the success of all projects, even the earliest, originally sound ones.

  There was an abundance of fraud in this era, too—advertising worthless land as wonderful, pumping and dumping stocks, committing insider trading, and selling stock in fraudulent companies. Such names as Ivar Kreuger and Charles Ponzi led the way with fraudulent investments and illicit sales.29 Fraud is an element in many financial crises but not the cause. In the 1920s it was a symptomatic sideshow—rampant because it was harder to detect in the madness and because so many people were easy targets owing to their greed and fear of missing out on the easy wealth they saw around them.

  Figure 1.5 provides a sense of the relative size of certain key spending and balance amounts during this lending period. Nominal GDP grew a strong $12 billion in this five-year period, after growing a mere $3 billion in the previous, troubled four-year period. But private debt grew by a staggering $40 billion, including $9 billion of growth in bank loans.

  Notice how lopsided those figures are. Private debt does not normally outgrow GDP by this large an amount. It’s direct evidence that too many of these loans were not yielding proper economic returns and that they were bad loans that would not get fully repaid. Economists refer to this relationship as the credit intensity ratio, and when private debt growth exceeds GDP by this wide a margin, it’s usually a very bad sign.

  Bank deposits and broad money, which are close to the same thing, were both growing rapidly. Since loans are one of the key means for creating an increase in deposits (when you get a loan, the bank puts the proceeds into your deposit account), in this case it is reasonable to assume that most of that deposit increase is a function of the loan increase.

  Growth in net exports was a minor factor in this period, as was the growth in the holdings of gold. The total gold in the United States during this period was about $3 billion to $4 billion, in an economy with a $90 billion GDP—and $160 billion in loans. The changes in the total U.S. gold stock during this period were only a few hundred million dollars in any year, an amount dwarfed by the changes in private debt. Gold holdings by the Treasury and the Fed barely budged—while private loans, again, grew by $40 billion.

  Figure 1.5. Change in Key National Financial Indicators, 1923–1928

  In the same period, government debt declined.

  Figure 1.5 vividly illustrates the driver of financial crisis: runaway lending. As noted earlier, in a larger economy, a rough guideline is that when a private debt–to–GDP ratio grows by 15 to 20 percentage points in a five-year window, especially when it reaches or exceeds an overall ratio of roughly 150 percent to GDP, a financial crisis is likely. In the United States, by 1928, both of these thresholds had been met, and the proverbial horse was out of the barn.

  As in the other crises examined in this book, overlending was the necessary and sufficient explanation for the boom and bust of the 1920s and 1930s.

  On to phase two. By the late 1920s, sky-high real estate values had started to fall—in the same way that housing values started to decline in 2006 well in advance of the 2008 crisis. Then, in October of 1929, the stock market collapsed, with stocks eventually dropping by more than 80 percent. The unprecedented reliance on margin debt ensured that the correction was sharp. From the end of September to the end of November, the value of utilities dropped 55 percent; industrials, 48 percent; and railroads, 32 percent. Without private debt and leverage, stocks would not have scaled these heights, and the downside in the fall of 1929 would have been far lower.

  A financial crisis often takes time to unfold and is often staggered over a period of several years. The Great Depression didn’t begin suddenly on October 29, 1929, when the market crashed. Overall construction activity had started to slow in 1927, and the decline had accelerated in 1929; steel production and automobile sales slowed in 1928; stocks crashed in 1929; and widespread bank failures did not begin until 1931. In December 1929, two months after the crash, industry titan and then U.S. Treasury secretary Andrew Mellon intoned, “I see nothing in the present situation that is either menacing or warrants pessimism. . . . I have every confidence that there will be a revival of activity in the spring.”30

  From 1929 and 1933, GDP contracted by an astonishing $48 billion, from $105 billion to $57 billion. The easiest way to think about GDP is as total spending for final goods and services in an economy. Economists express it as consumption plus investment plus government spending (plus net exports)—or C+I+G. That simply means that consumers and businesses and the government were collectively spending $105 billion in 1929 but only spending $57 billion in 1933. Large GDP drops of varying magnitudes had occurred in the nineteenth century, but nothing remotely comparable has happened to a developed economy since then, which makes it hard for us even to imagine. It would be as if U.S. GDP collapsed from $14.7 trillion to $7.3 trillion in 2009. Instead, U.S. GDP declined only by $3
00 billion, to $14.4 trillion. There is simply no comparison. Even adjusting for inflation, the differences are profound. But note that adjusting for inflation itself can be misleading because even though prices dropped, the amount of debt owed did not and debtholders had to pay for this high debt with now-diminished incomes.

  Figure 1.6 gives a sense of the relative size of declines in certain key spending and balance amounts during this period.

  Spending collapsed. Bank runs—customers withdrawing their deposits en masse—depleted those banks, forcing them to call loans, and that would contribute significantly to the spending collapse. Lending and GDP were inviolably lashed together in the boom time—the former feeding the latter—and they were just as inviolably joined in the contraction and decline.

  Banks are notorious for having bad loans on their books they either aren’t aware of or are slow to acknowledge and disclose, and this was true in the late 1920s. In fact, banks can readily defer the recognition of a given loan problem by a year or two simply by being slow to revise forecast assumptions or deferring reviews—not to mention modifying payment terms. Banks reported little in the way of credit quality problems, even when billions in problem loans already lurked on their books. As in most crises, these credit problems were recognized only well after the point of gross overcapacity had been reached.

  Figure 1.6. Change in Key National Financial Indicators, 1929–1933

  Since by 1928 total loans had skyrocketed by 34 percent in just five years, it is reasonable to estimate, based both on analysis of this period and comparison to other crises, that 15 percent to 20 percent of this loan growth, or $6 billion to $8 billion, was composed of problem loans, a large portion of which eventually became losses. That ratio of losses to new loans in an era of rampant loan growth is not unusual. Banks alone, which were responsible for about 25 percent of all private lending during this period, charged off $3.1 billion over the next several years on a capital base of $8 billion. (“Charging off” meant writing the loan off as a loss on the books of the bank.) Real estate bonds issued between 1920 and 1930 totaled $3.9 billion, and 80 percent of them were failing to meet their contracts in 1936—which likely meant that roughly another $2 billion in losses were already present. Other types of lending institutions had losses as well.