Sears Built a Brilliant Financial Empire. It Just Forgot It Was Running a Store.

On September 17, 1985, a Sears employee from the Chicago area walked into a Sears store in Atlanta and spent $26.77. The amount is forgettable. The card is not: it was the first-ever Discover transaction, a test purchase for a product that would become one of America's four big card networks.⁵ Think about what that sentence contains. A retailer, inside its own store, ringing up a sale on a credit card its own brokerage subsidiary had just invented. By the mid-1980s Sears didn't just sell you a refrigerator. It could insure your car, sell you a house, manage your stock portfolio, and lend you the money to do all of it. The idea was that you'd handle your whole financial life between the lawn mowers and the Craftsman tools.

The official story is that Sears got greedy, wandered far from retail, and was punished for it - a cautionary tale about staying in your lane. Almost none of that survives contact with the numbers. The diversification made money. The acquisitions created shareholder wealth. The problem was never that Sears bought a brokerage. It was what stopped happening at the store while management fell in love with the brokerage.

The empire that started during a bridge game

Sears in financial services was not a 1980s fad. It was a half-century-old instinct. The whole thing started, of all places, on a commuter train in 1930, when insurance broker Carl L. Odell suggested to Sears CEO Robert E. Wood - his neighbor, mid-bridge game - that the company sell auto insurance by mail. The board approved it. Allstate, named after a Sears tire brand, was incorporated in April 1931 and sold policies straight out of the Sears catalog.¹ For decades it worked beautifully, which is the part that matters: Sears had real evidence that it could bolt a financial product onto a retail customer base and make it sing.

So in October 1981 Sears did the logical thing and pressed the bet. It agreed to buy Dean Witter Reynolds - then Wall Street's fifth-largest brokerage house - for $607 million, days after picking up the real-estate firm Coldwell Banker for about $179 million.³⁴ The pitch was clean and seductive: the 'Sears Financial Network,' where a family that trusted Sears for appliances would walk over to an in-store booth and trust it for stocks and mortgages too. The slogan of the era was 'socks and stocks.' Four years later, Dean Witter spun the magic up another notch and launched Discover, with no annual fee and a Cashback Bonus, tested at Sears stores and then announced to the country in a Super Bowl commercial.⁵

Here's the inconvenient part: the diversification worked

The lazy version of this story ends with the financial empire collapsing in flames. It didn't. When a Journal of Financial Economics study went back and measured what the 1981 acquisitions actually did for shareholders, the answer was the opposite of the legend: the deals created roughly $400 million in wealth at announcement. And when Sears finally unwound the whole thing, the divestiture handed shareholders a further $1.113 billion - more than $1.5 billion in cumulative gains across the life of the strategy.⁷ Discover became a national network. Allstate's eventual IPO was, at the time, described as the largest in U.S. history.² By the brutal scoreboard of 'did the assets create value,' Sears was right.

Which forces a sharper question. If buying the brokerage made money, and selling it made money, what exactly went wrong? The answer is hiding in the same study, in a phrase that does all the work: shareholders suffered a significant opportunity loss versus a portfolio of focused firms.⁷ Translated: Sears didn't lose money on the empire. It lost the company it could have been if it had spent those years minding the store instead. That's a different and far more dangerous failure - because it never shows up as a loss on any single deal. It shows up as the thing that didn't happen.

Synergy was the cover story. Attention was the real currency.

The 'Sears Financial Network' assumed a synergy that never arrived: that the customer buying a power drill would, in the same trip, buy mutual funds from the booth fifty feet away. People didn't. The anticipated cross-selling synergies simply did not materialize.⁷ Meanwhile, the new businesses weren't even cheap to run while everyone waited for the magic - Discover's operations alone bled a loss of $22 million in the fourth quarter of 1986 and another $25.8 million in the first quarter of 1987.⁶ But the dollar losses are a footnote. The real cost was where executive attention went, because that is the one resource a diversified company can never expand. Every board meeting spent on brokerage margins and card-network economics was a board meeting not spent on the discount retailers quietly building category-killer stores across the parking lot.

And the rot was already visible at the start. The most damning number isn't from the end of the story - it's from the beginning. In 1981, the very year Sears announced its grand diversification, its retail group's return on sales was already 36% worse than the retail industry median. Over the years Sears owned the financial firms, that gap didn't close. It widened, to 49% worse.⁸ The store was sick before the brokerage arrived, and the brokerage gave management a far more glamorous problem to solve than fixing it. By the time Sears walked away from the strategy twelve years later, it had permanently lost its crown as the world's largest retailer.⁸ You cannot out-diversify a core that is quietly dying - you can only buy yourself a distraction from watching it happen.

Wasn't it the focused investors who saw it clearly?

The fair objection is that this lets diversification off too easily. If the conglomerate was such a good idea, why did Sears tear it apart? And the honest answer is that it didn't decide to on its own. The September 1992 divestiture announcement came coincident with institutional investor activist pressure - the focused-portfolio crowd, looking at the same numbers, concluded Sears was worth more in pieces than as a financial-retail hybrid.⁷ They were right, and that seems to indict the whole adjacency strategy.

But look closer at what the activists actually proved. They didn't prove the businesses were bad - they proved the businesses were better off apart, run by people thinking about nothing else. Allstate, freed from Sears, became fully independent in 1995.² Discover became its own network. Each one thrived once it had an owner whose entire attention it commanded. That's not an argument against the assets. It's the clinching argument for the thesis: the value was real, and the conglomerate structure was the thing destroying it, because the structure forced one set of executives to do five jobs and they did all five worse than five focused teams would have. The market didn't reject diversification. It rejected divided attention.

Sears spent more than a decade and a fortune in lost ground proving a strategy lesson it could only learn the expensive way: that a portfolio of good businesses is not the same as a good business. Allstate and Discover were not the mistake. They were excellent companies that happened to be living inside a retailer that needed every ounce of its leadership's focus and gave that focus away. The diversification didn't bankrupt Sears. It did something quieter and worse - it kept the lights on, the share price respectable, and everyone looking the wrong way, while the only business that had to survive slowly stopped being worth saving. The empire made money. The store lost the future. Same company. Opposite math.