Don’t Cry for Coca-Cola

Peter Golder’s study of brand persistence shows that a recession may be a top brand’s best friend.

You're in the grocery store, about to reach for a six-pack of Coke when you remember how light your wallet is. Polar Cola, at a fraction of the price, beckons. Do you grab it and pinch pennies? Or reach for your Coke, economy be damned?


Tuck professor of marketing Peter Golder bets you'll go for "the real thing." In the first-ever study of brand persistence over varying lengths of time, Golder and his colleagues found that market share leaders are more likely to hold their leadership during recessions and periods of high inflation. And more likely to lose leadership when the economy is booming.

"Our findings are contrary to the conventional wisdom expressed frequently in Advertising Age and even The Wall Street Journal," says Golder. "People assume that since market leaders tend to be premium brands, consumers shift away from them to save money. When you have a natural theoretical argument, it takes on a life of its own."

The study of brand persistence—what causes one leading brand to last 80 years while another flames out after five—requires its own persistence. Golder and his colleagues spent hundreds of hours scouring business and advertising magazines to identify market leaders from 1921 to 2005 and correlated their rise and fall to GDP growth and inflation.

Among their insights is the discovery that, contrary to comments from branding experts like Jack Trout, top brands don't last forever. The study quantifies a halflife phenomenon and shows that while more than 95 percent of leading brands maintain their leadership from any given year to the next, over a couple of decades half of them will lose their top spots.

While the study didn't formally evaluate why leading brands fare well during a downturn, Golder offers several explanations, including stores cutting midtier brands to reduce inventory and consumers finding "comfort and familiarity in trusted brands. It's the brands in the middle that are being squeezed," he says. "The pain is being felt somewhere."

Golder's study is among the first to show definitively that not all categories are created equal. As expected, the researchers found a higher level of brand persistence for food, with which consumers identify personally and respond to with nostalgia, and a low level for clothing brands, which are subject to the whim of fashion.

But they also found slightly lower long-term persistence for personal care products such as razors and deodorants that had been thought to inspire greater loyalty. Golder surmises that consumers are more swayed by product improvements—an extra blade, a new scent—than previously thought. On the other hand, household supplies such as laundry detergent have higher persistence, despite consumers' lack of personal identification with them. Here Golder credits simple inertia: since consumers don't care much, they are less likely to shop around for something different.

The findings could help managers determine how to allocate resources based on economic conditions— for example, propping up mid-tier brands during a recession. "While the economic conditions that affect brand persistence are outside the control of managers," says Golder, "they still must consider those conditions when making the decisions they do control."

If Golder has his way, the study could also lead to bonuses for managers who beat the odds. "Once you know how conditions affect your portfolio of brands," he says, "you should be rewarding people who exceed expectations."

Golder P N, Irwin J R, Mitra D, “Will You Still Try Me, Will You Still Buy Me, When I’m 64? How EconomicConditions Affect Long-Term Brand Leadership Persistence,” forthcoming