Segment 1 — The Cold Open

In March 1933, Americans queued around the block to yank their savings out of banks that might not open tomorrow. Half a century later, a money broker in New York could wire millions to a Texas savings-and-loan he had never seen, because the advertised rate was a quarter-point higher and every dollar was guaranteed by the United States. Deposit insurance was designed to kill the panic that destroyed thousands of banks. Instead, it taught depositors not to ask questions, and it handed failing thrifts a taxpayer-backed chip with which to gamble.

Segment 2 — The Good Intention

The Banking Act of 1933, signed by Franklin Roosevelt on June 16, created the Federal Deposit Insurance Corporation because the alternative had just been lived in public. Between 1929 and 1933 roughly nine thousand American banks suspended operations; in 1933 alone more than four thousand failed. Depositors who got to the window too late lost their money, and the rumor of a rumor was enough to finish the next bank down the street. Representative Henry B. Steagall of Alabama, whose rural constituents had watched local banks vanish, had been pushing a federal guarantee for years. Roosevelt himself had doubts. He worried that a government promise would invite careless banking and, eventually, a raid on the Treasury. The American Bankers Association fought the idea as a subsidy for weak institutions. They were not being frivolous. But in the spring of 1933 the political fact was simpler than the theory: people would not put money back into banks they could not trust, and the payment system of the United States had already stopped. Deposit insurance, initially $2,500 per account when the temporary plan took effect on January 1, 1934, was the price of getting deposits to return. A companion fund, the Federal Savings and Loan Insurance Corporation, followed in 1934 under the National Housing Act, wrapping the same logic around the thrifts that financed ordinary home loans. The intention was narrow and humane: stop the run, restore the checking account, let a solvent bank survive a false rumor.

Segment 3 — The Implementation

It worked, almost immediately, at the thing it was built to do. In 1933 more than four thousand banks suspended; in 1934 only sixty-one failed, and just nine of those were FDIC-insured. Deposits flowed back. Coverage was made permanent in the Banking Act of 1935 and raised to $5,000. For the next four decades the American banking system looked, from the sidewalk, almost boring. Failures were few. Premiums were flat — every insured bank paid the same rate regardless of how it lent — because the whole point was confidence, not a credit-rating agency with a government seal. The ceiling crept up with inflation and politics: $10,000 in 1950, $40,000 in 1974, then $100,000 in 1980 under the Depository Institutions Deregulation and Monetary Control Act. Skeptics had never fully disappeared. Economists and some bankers kept repeating the 1933 objection: if depositors do not care, someone else has to, and that someone is the supervisor, who is always late. Those warnings sounded academic while the 3-6-3 world of savings and loans still held — pay 3 percent on passbook savings, lend at 6 percent on thirty-year mortgages, and be on the golf course by 3. The insurance funds looked fat. The public remembered the breadlines, not the footnotes. Proponents could point to a genuine, measured victory: the contagious bank run, as Americans had known it, was gone.

Segment 4 — The Unintended Consequences

The victory had a mechanism, and the mechanism had a shadow. Once a deposit was insured, the depositor’s incentive to distinguish a careful thrift from a reckless one collapsed to a single number: the interest rate. The bank’s incentive shifted too. Losses beyond a thin slice of capital belonged to the insurance fund, which is to say, eventually, to the Treasury. For fifty years that moral hazard stayed mostly latent, because regulation boxed thrifts into home mortgages and because interest rates were tame. Then the 1970s broke the box. Inflation and Paul Volcker’s Federal Reserve pushed short-term rates toward 20 percent in 1981. Savings and loans were still stuffed with old fixed-rate mortgages yielding 6 or 8 percent, while they had to bid for deposits in a market Regulation Q could no longer contain. On a market-value basis, large parts of the industry were insolvent by the early 1980s — estimates of the hole ran well into the tens of billions, and some contemporary analyses put the industry’s aggregate net worth deeply negative. The straightforward response would have been to close the dead institutions, pay the depositors, and charge the loss to FSLIC. FSLIC did not have the money, Congress did not want the bill, and the industry had friends. So policy ran the other way. The Garn–St. Germain Depository Institutions Act, signed on October 15, 1982, let thrifts load up on commercial real estate, consumer loans, and other assets they had never been staffed to underwrite. California’s Nolan Act and Texas’s liberal state charters went further. Ronald Reagan called Garn–St. Germain the most important financial legislation in fifty years. It was meant as a life raft. For a thrift that was already broke, it was a lottery ticket bought with insured cash.

What followed was not a mystery of markets so much as a predictable use of other people’s guarantees. Insolvent “zombie” thrifts paid above-market rates to attract deposits, then threw the money at acquisition-development-construction loans, raw land, and, in the more adventurous shops, junk bonds. Deposit brokers in New York and elsewhere sliced large sums into $100,000 certificates of deposit — the insurance ceiling — and shopped the nation for the highest yield. No one had to inspect the loan file in Dallas or Irvine. The government had already inspected the ceiling. Vernon Savings and Loan in Texas, run by Don Dixon, became a grim mascot: a French villa, a fleet of aircraft, a $5,500 teapot, and, when federal conservators finally walked in, a loan book that was almost entirely delinquent. Lincoln Savings, under Charles Keating, grew from a sleepy Orange County thrift into a vehicle for junk bonds and desert real estate; it also sold uninsured bonds to thousands of mostly elderly customers in its branches, many of whom believed they were as safe as the insured CDs advertised a few feet away. Lincoln’s failure cost the government on the order of $3 billion. Across Texas, California, Florida, and the Southwest, similar books piled up. Accounting forbearance and “regulatory accounting principles” let institutions count supervisory goodwill and other fictions as capital, so the zombies could keep growing. Edward Kane and other economists described the result as a put option: heads, the owners and managers won; tails, FSLIC paid. By 1987 the FSLIC fund itself was insolvent. In 1988 and 1989 the failures came in waves. From the mid-1980s through the mid-1990s, more than a thousand thrifts were closed or resolved. The Resolution Trust Corporation, created in 1989, would eventually take over hundreds of them and dispose of hundreds of billions of dollars in assets, much of it distressed real estate dumped into a weak market. Second-order damage spread through construction employment, local tax bases, and a political scandal — the Keating Five — that showed how the same guarantee that calmed depositors could be used to buy time from senators. Third-order, the cleanup itself tightened credit as surviving institutions and the RTC retrenched, adding weight to the early-1990s downturn. The FDIC later put the taxpayer cost of the savings-and-loan crisis at roughly $124 billion; broader tallies that include industry assessments run higher, commonly around $150 billion. The runs of 1933 had been stopped. The bill arrived, with interest, in 1989.

Segment 5 — The Aftermath

Congress did not repeal deposit insurance. It tried to price the gamble back into the system. The Financial Institutions Reform, Recovery, and Enforcement Act of 1989, signed by George H. W. Bush, abolished FSLIC, shifted thrift insurance to the FDIC, created the RTC, raised capital expectations, and restricted some of the wilder asset powers. The Federal Deposit Insurance Corporation Improvement Act of 1991 went further: risk-based premiums, so riskier banks would pay more; prompt corrective action, a ladder of mandatory constraints as capital declined; least-cost resolution rules; and tighter limits on brokered deposits for weak institutions. Those tools were a confession that the 1933 design had been incomplete. They were also imperfect. Premiums still lag the risk; capital ratios can be gamed; prompt corrective action is only as honest as the accounting. The 2008 crisis revealed that the largest banks carried an implicit guarantee beyond the statutory cap — too big to fail — and coverage was raised to $250,000, first as emergency, then as law. In March 2023, when Silicon Valley Bank and Signature Bank collapsed, uninsured depositors were protected after all, on the theory that a run on the uninsured would become a run on the system. The safety net widened again, for reasons that would have sounded familiar to Steagall: panic is expensive, and guarantees are fast. The countermeasures remain. So does the original bargain.

Segment 6 — The Lesson

A guarantee that removes fear also removes watchfulness, and the watchfulness has to live somewhere — in premiums that bite, in capital that is real, in supervisors who close the dead instead of hoping they earn their way out. Incentive structures find their loopholes, especially when the loophole is legal and the loss is socialized. Forbearance is not mercy; it is a loan of public money at a hidden interest rate, and the borrower who is already insolvent will not use it conservatively. Deposit insurance is still one of the great stabilizing inventions of the twentieth century. The question it leaves, every time a new class of liabilities looks run-prone, is not whether to calm the crowd. It is who, exactly, is supposed to stay afraid.

💬 Reply to this email — Patrick reads every one.

Share: X · LinkedIn · WhatsApp

Forwarded this email? Subscribe here — it's free.