The Nifty Fifty: One-Decision Stocks and the 1973–74 Crash

On 11 January 1973 the S&P 500 closed at 120.24 and the Dow Jones Industrial Average at 1,051.70, the highest either index had ever reached. That evening Richard Nixon went on television to announce Phase III of his economic programme, lifting most of the mandatory wage and price controls imposed in August 1971. Wall Street had spent fourteen months buying on the proposition that American inflation had been brought under administrative control.

Neither index saw those levels again for a long time. The S&P did not close above 120.24 until the summer of 1980. The Dow did not close above 1,051.70 until November 1982, nearly ten years later.

What fell hardest in the twenty-one months that followed were not the speculative issues. They were the fifty or so large, profitable, household-name companies that institutional money managers had spent the preceding four years deciding were the only stocks worth owning.

Stocks You Only Had to Buy Once

Nobody ever published an official roster. The label came out of the trade press around 1972, and two lists circulated behind it. Morgan Guaranty Trust, the largest bank trust department in the country, maintained one; Kidder Peabody maintained another. They overlapped heavily without matching, which is part of why arguments about the group's subsequent performance have never quite resolved.

Individually the names were ordinary: Coca-Cola, IBM, Xerox, General Electric, Procter & Gamble, Eastman Kodak, Johnson & Johnson, McDonald's, Walt Disney, Polaroid, Avon Products, Merck, Pfizer, Philip Morris, Texas Instruments, Sears Roebuck, International Flavors & Fragrances, Schlumberger, Digital Equipment. Between them they spanned pharmaceuticals, consumer products, photography, computing and oilfield services. What united them was a record of earnings growth that had not broken in a decade or more.

Out of that record came the phrase that defined the period. These were one-decision stocks: you decided to buy, and the second decision, the one about when to sell, was held to be unnecessary. A company compounding earnings at fifteen per cent a year indefinitely would eventually grow into any price. Selling it meant paying capital gains tax and then facing the problem of what to do with the proceeds.

The idea had an institutional logic that made sense from inside a trust department. A bank managing pension assets for a corporate client could be dismissed for owning something that blew up. It could not easily be dismissed for owning Coca-Cola. Concentration in unimpeachable names was a career-risk trade before it was an investment thesis, and by 1972 an enormous quantity of money had made it.

After the Gunslingers

Every element of that consensus was a reaction to what came before it. Wall Street in the 1960s had belonged to the performance managers — Gerald Tsai at the Manhattan Fund, Fred Carr at the Enterprise Fund — who traded conglomerates, franchisers and new issues at high turnover and posted the returns to prove it. John Brooks, who chronicled that market as it happened, described a decade in which fund performance was reported monthly like a batting average and portfolio managers became public figures (Brooks, 1973).

That market ended in the bear market of 1969–70, which took the S&P down roughly 36 per cent and destroyed several of the go-go funds outright. The collapse of the Penn Central Transportation Company in June 1970 froze the commercial paper market and forced the Federal Reserve into the discount window in a way it had not been used since the 1930s. Anyone who had been buying speculative paper had a bad two years.

Growth investing survived the episode by moving upmarket. The appetite for compounding earnings did not go away; it relocated from second-tier conglomerates into the largest and most established growth companies in the country, where it seemed to carry none of the risk. Financial journalists began calling the result a two-tier market, because the fifty and everything else had stopped moving together.

The Multiples

By the end of 1972 the prices attached to that logic had reached levels with no American precedent outside 1929. Jeremy Siegel, working from the Morgan Guaranty list, calculated the group's average price/earnings multiple in December 1972 at 41.9 against 18.9 for the S&P 500 as a whole (Siegel, 1998).

Multiples of that size encode a specific and checkable claim. At ninety times earnings a company must grow earnings very fast for a very long time before the buyer's money comes back, and the arithmetic leaves no room for an interruption. Polaroid at 94.8 times was priced as though the SX-70 instant camera, launched that October at $180 and losing money on every unit, would carry the company for a generation.

Twenty-One Months

Nothing dramatic happened in the first quarter of 1973. Prices simply stopped going up and began a slow grind that gave holders every opportunity to persuade themselves it was a pause.

Underneath, the monetary order was coming apart. The suspension of dollar convertibility in August 1971 had been patched over by the Smithsonian Agreement, and the patch failed. On 12 February 1973 the Treasury devalued the dollar a second time, by ten per cent. By March the major currencies were floating and the fixed-rate system built at Bretton Woods had effectively ceased to exist. Phase III controls were reimposed in June as a sixty-day freeze, which is not what a market wants to hear about an inflation that was supposed to be beaten.

Then came the shock nobody's earnings model contained. The Yom Kippur War in October 1973 and the embargo that followed took the posted price of Arabian light crude from $2.90 a barrel to $11.65 by January 1974. Consumer price inflation, 6.2 per cent in 1973, reached 11.0 per cent in 1974. The federal funds rate was pushed to roughly 13 per cent by July 1974.

Every one of those numbers attacks a high multiple directly. A stock priced on earnings twenty years out is a long-duration asset, and its present value falls further than a cyclical's when the discount rate doubles. Robert Shiller's long series shows the cyclically adjusted earnings multiple for the whole market falling from roughly 19 at the start of 1973 to below 9 by the end of 1974, one of the sharpest deratings in the record (Shiller, 2015).

Parts of the financial system began breaking in 1974. Bankhaus Herstatt was closed by German regulators on the afternoon of 26 June, stranding counterparties who had already paid out Deutsche Marks and were waiting on dollars that never came. Franklin National Bank, the twentieth-largest bank in the United States, was declared insolvent on 8 October, the largest American bank failure to that date. Nixon had resigned on 9 August. Gerald Ford launched his "Whip Inflation Now" campaign on 8 October, five days after the market had already bottomed.

The S&P 500 closed at 62.28 on 3 October 1974, down 48.2 per cent from the January 1973 peak. The Dow's own low came on 6 December at 577.60, a fall of 45.1 per cent. Consumer prices had risen more than a fifth across the same stretch, so the loss to a holder's purchasing power was closer to 57 per cent. British investors fared worse: the FT 30 fell about 73 per cent from its 1972 high to its low on 6 January 1975.

What Happened to the Fifty

The group that was supposed to be the safe tier was not.

Avon was the emblem. A door-to-door cosmetics company with no factories to speak of and a direct sales force of hundreds of thousands had been capitalised at more than the whole American steel industry; by the end of 1974 it traded at a single-digit multiple. Polaroid's SX-70 worked, sold, and never earned anything close to what the 1972 price required. Xerox's patents on plain-paper copying began expiring, Japanese competitors arrived, and the growth rate that justified 45.8 times earnings quietly became an ordinary one.

Burton Malkiel's account fixes the deflation with a second column of multiples: by 1980 the same names were changing hands in the high single digits and low teens — Avon near 9, McDonald's near 9, Disney near 11, Xerox near 8 (Malkiel, 2019). Earnings at most of these companies had kept growing. What collapsed was the price anyone would pay for them.

The Argument That Followed

Twenty-five years later Siegel reopened the question and produced an answer nobody expected. Taking the Morgan Guaranty fifty at their December 1972 prices and holding them through August 1998, he found the portfolio had returned almost exactly what the S&P 500 returned over the same period — a gap of a fraction of a percentage point. He then computed a "warranted" multiple for each stock, the price that would have delivered the market return with perfect foresight, and found the group's warranted average at about 40.6 against the 41.9 actually paid (Siegel, 1998).

On that reading the Nifty Fifty were not a bubble at all. They were, in aggregate, correctly priced by a market that then took twenty-five years to be proved right — and the aggregate concealed enormous dispersion. Philip Morris returned far more than its 1972 price implied. Polaroid and Emery Air Freight destroyed their holders permanently. Buying the group was not the same bet as buying any member of it.

Jeff Fesenmaier and Gary Smith reran the exercise on the Kidder Peabody list and reached a stronger conclusion still, finding that group ahead of the index rather than level with it (Fesenmaier and Smith, 2002). The dependence of the verdict on which of two unofficial lists you use is the most honest finding in the whole literature.

None of this was available to anyone in 1974, and none of it would have helped. A pension fund that had bought at 41.9 times earnings needed the money before 1998. Investment committees were being wound up, managers fired and mandates reassigned on the basis of results measured over quarters.

What the Wreckage Built

Two institutions came out of the period.

Congress supplied the first. The Employee Retirement Income Security Act was signed on 2 September 1974, five weeks before the bottom, and it imposed a fiduciary standard on pension trustees that made undiversified concentration in fifty stocks legally hazardous rather than merely unwise. It also required funding discipline, which pushed corporate plans toward measurable, benchmarked management.

The other was cheaper. John Bogle launched the First Index Investment Trust in August 1976, eighteen months after the bottom, into an industry that had just demonstrated in public what concentrated active management could cost. Its initial underwriting raised $11.3 million against a target of $150 million, and the trade press called it Bogle's Folly. The argument that made the fund eventually unavoidable — that the average active manager cannot beat the index after fees — had been made in academic journals for a decade. It took 1973–74 to make it viscerally legible to the people writing the cheques.

The pattern recurred at the end of the 1990s, when the internet bubble attached comparable multiples to companies with far less history behind them, and again after 2020 in a narrower set of large technology names. In each case the defence offered for the price was the same one Morgan Guaranty's trust officers made in 1972: the business is durable, the growth is real, and the multiple therefore does not matter. Twice out of three times the business was durable and the growth was real.

Polaroid filed for bankruptcy in October 2001. Avon was sold to Natura & Co in 2020 for about $2 billion, less in nominal dollars than its market value in 1972. Coca-Cola, bought at 46.4 times earnings at the top and never sold, made its holder rich.