The Winner Raised Prices

In 1972, Warren Buffett paid $25 million for a candy company in San Francisco. See's Candies sold chocolate at $1.85 a pound. He raised the price. Every year. Five percent, steady, for more than fifty years. By 2024, a pound of See's cost around $24. Volume barely moved. The company generated over $2 billion in cumulative pre-tax earnings on $32 million of reinvested capital.¹

Charlie Munger put it simply: "We just discovered that we could raise prices 10% a year and no one cared."²

In 2010, Buffett told the Financial Crisis Inquiry Commission that pricing power was the single most important quality in a business. "If you've got the power to raise prices without losing business to a competitor, you've got a very good business. And if you have to have a prayer session before raising the price by a tenth of a cent, then you've got a terrible business."³

It is one of the most repeated lines in investing. The candy company is the proof. Fifty years of raising prices and nobody left.

But the same man who said this also owns GEICO and Nebraska Furniture Mart. Two companies whose entire existence is built on one principle: make it cheaper. His partner, Charlie Munger, sat on Costco's board for twenty-six years and never sold a share.⁴

Pricing power is real. See's Candy proves it. But it is not the only kind of power. And the people who quote Buffett on pricing almost never mention what he and Munger did with the rest of their money.

The reward

In 1995, the average American cable bill was $22.⁵

Cable operators held regional monopolies. One provider per territory. No competitive alternative. The customers had nowhere to go. So the operators did what monopolists do. They raised prices.

By 2000, the bill was $30. By 2005, $43. By 2010, $54. By 2019, $96. By 2024, the average cable bill including fees reached $147.⁶

Cable prices rose nearly fifty percent faster than inflation over three decades. The bill increased more than six times. The product did not improve six times. The channels did not multiply six times. The picture did not get six times clearer. The price went up because it could.⁷

In November 2010, a company in Los Gatos offered a streaming plan for $7.99 a month.⁸

Netflix did not build better cable infrastructure. It did not lobby Congress for regional rights. It offered a cheaper way to watch things on a screen. Cable had 105 million subscribers at its peak in 2010. By 2024, that number was 68.7 million. One-third of the customer base, gone in fourteen years.⁹

The cable companies had the power to raise prices. They used it. For thirty years, the power was real. The replacement was also real. It just took longer to arrive.

The sequence runs the same way each time. The visible result is more revenue. The invisible consequence is a signal: margins high enough to attract someone willing to undercut them. The timeline is the only variable. Cable's was thirty years.

The extreme case

In 2015, Turing Pharmaceuticals acquired the rights to Daraprim, a sixty-two-year-old drug used to treat parasitic infections in immunocompromised patients. The previous price was $13.50 a pill. Martin Shkreli, Turing's CEO, raised it to $750. Overnight. A 5,456 percent increase.¹⁰

The consequences arrived in layers. Public outrage. Congressional hearings. An FTC antitrust investigation. A criminal conviction. A federal judge ordering $64.6 million in disgorgement and a lifetime ban from the pharmaceutical industry. The Supreme Court declined his appeal.¹¹

But the price did not return to $13.50.

When the FDA approved the first generic version in 2020, five years after the hike, the generic manufacturers set their wholesale price at approximately $300 a pill. Not $13.50. Not the $1 the drug had cost a decade earlier. Three hundred dollars. The competitors did not race to the bottom. They settled at a level the inflated price had made possible.¹²

Shkreli is the extreme case. But the mechanism is not extreme. Raise the price high enough and you create a new floor that even the correction does not erase.

The ones who lowered

There is another version of this story. A version where the company wins, earns the right to raise prices, and does the opposite.

Ford

In 1909, Henry Ford released the Model T Runabout for $825. It was not the cheapest car on the market. It was the car Ford intended to make cheaper.¹³

The assembly line arrived in 1913. Each improvement in production efficiency became a price cut. Ford did not keep the savings. He passed them on.

The price fell 68 percent. Production rose nearly two hundred-fold. Fifteen million Model Ts were built before the line shut down in 1927.¹⁴

The logic was simple. A cheaper car meant more buyers. More buyers meant more volume. More volume meant lower unit costs. Lower costs meant a cheaper car. The cycle fed itself. Every price cut widened the moat, because every price cut made it harder for a competitor to match the cost structure that produced it.

Costco

Jim Sinegal cofounded Costco in 1983. He had spent years working for Sol Price, who in 1967 wrote a memo to the employees of his discount chain, FedMart. The memo contained one sentence that an investor named Nick Sleep would later frame on his office wall:¹⁵

Let us concentrate on how cheap we can bring things to the people, rather than how much the traffic will bear, and when the race is over Fed-Mart will be there. Sol Price, FedMart internal memo, 1967

Sinegal built Costco around that sentence.

Costco caps its markup at 14 percent on third-party products and 15 percent on its own Kirkland Signature brand. A typical grocery chain runs 25 to 30 percent. Walmart runs about 24 percent. Costco runs 12.¹⁶

In 1985, Costco introduced a hot dog and soda combo for $1.50. It still costs $1.50. Forty-one years. Adjusted for inflation, the combo should cost $4.40. When CEO Craig Jelinek told Sinegal in 2013 that the company was losing money on the hot dog and suggested raising the price to $1.75, Sinegal said: "If you raise the effing hot dog, I will kill you. Figure it out."¹⁷

They figured it out. Costco switched from Hebrew National to its own Kirkland Signature hot dog. Cut the cost. Kept the price.

Sinegal told the Seattle Times: "Do you know how tempting it is to make another $7 on a pair of jeans? But once you do it, it's like taking heroin. You can't stop."¹⁸

Revenue in 2005 was $51 billion. In 2025, $275 billion. Costco stock has returned 21.56 percent annualized over the past decade, compared to 13.64 percent for the S&P 500. The membership renewal rate is 93 percent.¹⁹

The money is in the membership, not the margin. In 2024, Costco collected $4.83 billion in membership fees alone. The products are sold near cost to create the incentive to renew. The low price is the reason people come back. The renewals are the profit. Raise the product prices and you sever the wire that connects the two.²⁰

Amazon

Jeff Bezos described the mechanism in his 2001 shareholder letter: "Focus on cost improvement makes it possible for us to afford to lower prices, which drives growth. Growth spreads fixed costs across more sales, reducing cost per unit, which makes possible more price reductions. Please expect us to repeat this loop."²¹

They repeated the loop.

Amazon Web Services has cut prices 134 times since launching in 2006. Company revenue went from $8.5 billion in 2005 to $386 billion in 2020. A forty-one-fold increase in fifteen years.²²

"Your margin is my opportunity," Bezos told Fortune in 2012. Apple's operating margins that year were 31 percent. Amazon's were 2 percent. Amazon was not failing to capture margin. It was choosing to give it away.²³

In a passage that the investor Nick Sleep quoted at length, Bezos explained why the math always says to raise prices and why he overrode the math:

When we lower prices, we go against the math that we can do, which always says that the smart move is to raise prices. This is because our quantitative understanding of elasticity is short-term. We can estimate what a price reduction will do this week and this quarter. But we cannot numerically estimate the effect that consistently lowering prices will have on our business over five years or ten years. Jeff Bezos, via Nick Sleep's Nomad Investment Partnership Letter, December 2006

The spreadsheet says raise prices, and in the short term it is right. It cannot model a flywheel.

GEICO

When Buffett was twenty years old, he took a train to Washington, D.C. and knocked on the door of GEICO's offices on a Saturday morning. He had read that his professor, Benjamin Graham, was chairman of the board. A man named Lorimer Davidson spent four hours explaining the insurance business to him.²⁴

The explanation was simple. GEICO sold car insurance directly, without agents. No agents meant lower overhead. Lower overhead meant lower premiums. Lower premiums meant more customers. The cost advantage was structural.

In 1996, Buffett bought the rest of GEICO. The company held less than 3 percent of the U.S. auto insurance market. Buffett told them to spend on advertising. The budget went from $30 million a year to over $1.5 billion. The pitch was always the same: fifteen minutes could save you fifteen percent.²⁵

By 2021, GEICO held 14.4 percent of the market. From number seven to number two in twenty-five years. In 2022, GEICO cut its advertising spend to focus on profitability. Progressive passed it for the number two position. The cost advantage was still there. The spending that communicated it to customers was not.²⁶

"The difference between GEICO's costs and those of its competitors," Buffett wrote in 1986, "is a kind of moat that protects a valuable and much-sought-after business castle."²⁷

A moat built on being cheaper.

The flywheel

Nick Sleep ran the Nomad Investment Partnership from 2001 to 2014. He returned 921 percent to his investors. Then he closed the fund, returned everyone's money, and walked away from hundreds of millions in future fees.²⁸

Sleep had spent those years studying what he called "scale economics shared." The concept is simple. When a company grows and its costs drop, it faces a choice: keep the savings or pass them to customers. Most keep the savings. The ones that pass them on start a flywheel.

"There are two kinds of companies," Bezos said. "Those that work to raise prices and those that work to lower them."²⁹

Sleep's counter-intuitive rule, arrived at after watching a family activity center beat two competitors and become the only one left, was this: after you win, lower prices. Not raise them.

The risk with super-normal profitability is that the profits are an incentive for a new competitor: far better to earn less, but for a much longer time. Nick Sleep, Nomad Investment Partnership Letter, December 2012

Super-normal profits are a signal. To entrepreneurs, they read: there is money here that the incumbent is not defending. To regulators, they read: there is extraction here that voters will notice. To customers, they read: there is an alternative worth looking for. High margins do not protect a business. They advertise a business worth attacking.

Lowering prices after winning inverts all three. Fewer entrepreneurs try to enter, because the margins do not justify the risk. Fewer regulators intervene, because the prices do not justify the scrutiny. Fewer customers leave, because the price is already where they want it to be.

"Scale economics shared is not a strategy," Sleep wrote. "It's a destination."³⁰

The paradox

Here is the problem with the pricing power quote.

Buffett is right. And he also owns GEICO and Nebraska Furniture Mart, where Rose Blumkin built the largest furniture store in America on five words: "Sell cheap, tell the truth, don't cheat nobody." Munger held Costco stock personally until the day he died. Never sold a share.³¹

Buffett once described what it was like to compete with her: "I'd rather wrestle grizzlies than compete with Mrs. B and her progeny. They buy brilliantly, they operate at expense ratios competitors don't even dream about, and they then pass on to their customers much of the savings."³²

That is not pricing power. That is pricing restraint as a weapon.

In his 2007 letter to shareholders, Buffett resolved the contradiction in a single sentence:

A truly great business must have an enduring "moat" that protects excellent returns on invested capital. Therefore a formidable barrier such as a company's being the low-cost producer (GEICO, Costco) or possessing a powerful world-wide brand (Coca-Cola, Gillette, American Express) is essential for sustained success. Warren Buffett, Berkshire Hathaway Annual Letter, 2007

Two moat types in the same portfolio, from the same investor, with opposite strategies.

See's Candy raising prices 5 percent a year for fifty years is a masterclass. Costco raising prices 5 percent a year would be suicide. It would destroy the membership incentive that produces $4.58 billion in annual fees. GEICO charging premium prices would eliminate the only reason customers switch from their existing insurer. Same investor, same holding company, opposite pricing decisions. Both correct.³³

The error is not raising prices. The error is assuming one strategy works everywhere.

See's Candy can raise prices because people are not comparison-shopping chocolate by the pound on Valentine's Day. The brand absorbs the increase. Coca-Cola can raise prices because 1.9 billion daily servings across two hundred countries makes a penny per serving worth $2 billion in pre-tax earnings. These are brand moats. The price increase does not invite replacement because the product is not being evaluated on price.

But cable was evaluated on price. And Daraprim was. And every commodity, every utility, every product where a customer compares cost before deciding is in the other category. The category where the winner who raises prices is writing an invitation.

The bill

The airline industry found a variation. After deregulation in 1978 drove real airfares down 45 percent and killed 162 carriers, the survivors learned not to raise the sticker price. They unbundled. In May 2008, American Airlines became the first major carrier to charge for a first checked bag. Within months, every competitor followed. By 2024, U.S. airlines collected $7.27 billion in baggage fees alone. The base fare stayed competitive. The total cost went up. The price increase wore a different costume, but it was still a price increase.³⁵

Southwest Airlines resisted for fifty-four years. No bag fees. No assigned seats. No change fees. Herb Kelleher built the airline in 1971 on a single principle: fly cheaper, fill the plane, do not charge for the things that make flying tolerable. The discipline produced forty-seven consecutive years of profitability.³⁶

In June 2024, the activist investor Elliott Management disclosed a $2 billion stake. Elliott's thesis was simple: Southwest was leaving $1 to $1.5 billion in annual revenue on the table by not charging bag fees. The margin gap between Southwest and its competitors was, to Elliott, not discipline. It was waste.

By October, Elliott had six board seats. In February 2025, Southwest announced it would charge for checked bags for the first time in its history. Thirty-five dollars for the first bag. Forty-five for the second. On January 27, 2026, it flew its last open-seating flight.

The model survived deregulation, September 11, the financial crisis, and a pandemic. It did not survive an activist fund that looked at fifty-four years of restraint and saw a number to be harvested.

Spirit Airlines, the purest expression of the unbundling model, went the other direction. It stripped the base fare to nothing and charged for everything else: bags, seats, carry-ons, water. It filed for bankruptcy in November 2024. Emerged in March 2025. Five months later, it filed again.³⁷

Sowell called this Stage One Thinking. Raise the price, and the visible result is more revenue. The replacement, the defection, the structural erosion, arrives later and lands on someone else's watch.³⁴

Even the price-lowerer is not immune forever. The question is not whether pressure arrives, but how many decades the flywheel runs before it does.

The winners who lowered prices did not avoid the bill. They delayed it by decades. Ford held the automobile market for nearly twenty years. Costco has compounded for forty-two. Amazon has run 134 price cuts over eighteen years and shows no sign of stopping. GEICO grew from 3 percent to 14 percent in twenty-five years before pulling back. Each one bought time by refusing to take the reward.

The winners who raised prices collected the reward immediately. And then spent the rest of their existence watching the replacement arrive.

The most repeated line in investing says pricing power is the single most important quality in evaluating a business. It does not say what to do with it.

See's Candy uses it to charge $24 a pound. Costco uses it to hold the markup at 14 percent and the hot dog at $1.50 for forty-one years. Both are correct. The moats are different. The discipline is the same. Both companies know exactly what they are.

The price is one more visible number that satisfies before the bill arrives.

The winner raised prices. The empire lowered them.