Sears Put a Stockbroker Inside the Store in 1981 — and Sold It Back Out by 1993

In the early 1980s a shopper could walk into a Sears store for a set of wrenches and, a few aisles over, open a brokerage account, buy homeowner's insurance, and start looking at houses — all under one roof, all sold by the same company that made the wrenches. Sears called it the financial supermarket. The press called it 'socks and stocks.'⁴ It was one of the boldest reinventions ever attempted by a great American retailer, and within a decade Sears would spend three years quietly taking it apart, piece by piece.

The story usually told is a cross-subsidy tale: Sears built a finance empire to bankroll a fading store business, then dismantled it in one clean move in 1992. Both halves of that are wrong. Sears never sold these deals as finance-carries-retail. And it did not unwind them in 1992. What actually happened is stranger and more instructive — a company that bought two thriving businesses, pressed them against its own customer base expecting magic, got nothing, and set both free to prosper elsewhere.

What Sears actually promised investors in 1981: the pitch was cross-selling to 36 million households, not a plan for finance to fund the store

In October 1981, over the space of a few days, Sears announced two acquisitions that had nothing to do with retail. It agreed to buy Coldwell Banker, the largest independent U.S. real estate brokerage, and Dean Witter Reynolds, Wall Street's fifth-largest brokerage, for $607 million.¹² Chairman Edward Telling did not describe this as a war chest to defend the stores. He described it as diversification riding the retail base — the deal would help Sears 'become the premier provider of consumer financial services' for its 36 million regular customer families, and Coldwell Banker's 'premier position' would 'facilitate entry by Sears into the attractive brokerage business.'²¹ Sears already owned Allstate Insurance and a savings and loan.³ The thesis was simple and seductive: a company that already had 25 million credit-card customers walking through 859 stores could sell those same people a mortgage, a mutual fund, and a policy.³

So the cross-subsidy label is a retrofit. Nobody at Sears in 1981 said finance would carry retail. They said the two would sell to each other. That distinction matters, because it changes the question you ask about the failure. A subsidy fails when the profitable side stops being profitable. A cross-sell fails when the customers you assumed would buy both simply don't.

The shopper who bought socks would not buy stocks: the synergy was assumed, and four years in the numbers still refused to show up

The mechanism that was supposed to make this work was physical proximity: put Allstate, Coldwell Banker, and Dean Witter together in a Sears Financial Network booth inside the store, and let the retail crowd flow past. It is a tidy theory. It ran straight into a fact retailers underestimate — traffic is not intent. The person pushing a cart toward the tool aisle is not, in that moment, a person shopping for a brokerage account, and the shared roof did nothing to convert one into the other. Dean Witter's offices located inside Sears stores did less business than its freestanding ones.⁸ Four years into the strategy, the combined Sears Financial Centers co-located in 312 stores were still unprofitable overall.⁸ The synergy that justified $600 million never arrived; by the mid-1980s the whole effort had, in the words of one business history, 'achieved lackluster results.'⁸

The one clean break was actually three years of retreat: the 1992 announcement was a statement of intent; the surgery happened in 1993 and finished in 1995

On September 29-30, 1992, Sears announced it would sell off Dean Witter and the Discover Card, sell most of Coldwell Banker, and 'refocus' on retailing to slash corporate debt.⁴ That announcement is where the popular timeline stops — 'Sears unwound it in 1992.' But an announcement is a press release, not a transaction. The actual dismantling took years. Dean Witter Discover did not become an independent public company until March 1, 1993.⁵ The Sears Mortgage Banking Group went to PNC Bank for $328 million in 1993, and Coldwell Banker Residential was sold to The Fremont Group and its own executives in May 1993.⁶ Allstate — the oldest financial holding of all — was not spun off to shareholders until 1995.⁷ The 'clean break' was three separate operations spread across three years, each one peeling a business off the retailer it was supposedly enriching.

And here is the twist that undoes the subsidy narrative entirely: the businesses did better once they left. Dean Witter Discover started life as an independent company at $27 a share at spinoff — with 20% sold to the public and the rest distributed tax-free to Sears shareholders — and traded around $60 by 1996.⁷ If finance had truly been carrying a dead-weight retailer, cutting that retailer loose would not have doubled its stock. The retailer wasn't a burden the finance arms were bearing. The retailer was a distraction they were happier without.

Wasn't the idea sound and the execution just early?: the one-stop financial shop arrived eventually — but never through a department-store aisle

The fair objection is that Sears was simply ahead of its time. The one-stop financial relationship — banking, brokerage, insurance under one brand — is now ordinary. So maybe the vision was right and 1981 was just too early. There's something to it, but it misses where the idea actually landed. The integrated financial supermarket did arrive; it just never arrived through a department-store aisle. It arrived through banks and brokerages that already had the customer's financial attention, not their tool-shopping attention. That is the specific error Sears made, and the numbers named it in real time: the in-store offices underperformed the freestanding ones.⁸ Proximity to a shopper is not proximity to a saver. Sears had confused sharing a building with sharing a relationship — and the wrenches and the mutual funds turned out to live in different parts of the customer's head.

Sears spent $607 million and a decade discovering that a customer and a customer are not the same thing. The people who trusted Sears to sell them a lawnmower did not, in any profitable number, want it to manage their retirement — and the retreat that took from 1992 to 1995 was less a rescue of a subsidized store than a divorce that made everyone richer. The businesses weren't propping up the retailer. They were waiting to be let out of it. The real lesson of socks and stocks isn't that finance can carry retail. It's that owning the customer's foot traffic is not the same as owning the customer, and the difference is worth exactly $600 million to find out the hard way.