A key feature of the modern world is the “emerging middle class”—the demographic group resulting from the rise into relative economic comfort of once-poor populations in developing nations. The phenomenon—particularly notable in Brazil, China, India, and Indonesia—has obvious interest for sellers of consumer products, such as foods, drinks, medicines, and household appliances. But few researchers have sought to understand the buying decisions made by “new consumers” with their first experience of disposable income.

Now, though, Alberto Salvo, an assistant professor of management and strategy at the Kellogg School of Management, and Alon Eizenberg, an assistant professor at Jerusalem’s Hebrew University, have begun to remedy that situation. Their vehicle: a study of soda choices by Brazilians new to the middle class. The result: a model, estimated from data, that reveals persistence in buying behavior—persistence that should cause concern among vendors of premium brands. If price-sensitive new consumers start out by purchasing cheaper, local brands—a.k.a. generics—they will continue to favor them rather than choosing premium brands, such as Coca-Cola, favored by the traditional middle class.

Social Economic Transformation
To reach that conclusion, the two researchers needed to advance general understanding of the emerging middle class. “Alon and I are studying markets where there’s been a huge social economic transformation over a very short period of time owing to an income shock,” Salvo explains. “We’re looking at how the demand for a particular type of product changes as you change the socioeconomic composition of the population and how producers respond.”

In doing so, Salvo and Eizenberg developed empirical tools to incorporate social mobility and the persistence of consumption habits into their study of the competitive interplay between high-priced, heavily advertised established brands and “value brands” that compete on price. “Many markets have premium brands that are experiencing very tough competition from the generic fringe,” Salvo says. “This is a new phenomenon and it’s hard to get reliable data on such markets. We got it.”

Brazil represented an appropriate location for the study as a result of its experience of economic transformation. “It came in the mid-1990s, after high chronic inflation came down to single-digit levels,” Salvo notes. “After this sharp shock you saw tens of millions of new consumers entering markets for soda and other packaged foods, cement and housing, TVs and refrigerators.” Soda proved an effective product for studying consumption because of high demand.

“The Brazilian market trails only the United States and Mexico by volume,” Salvo and Eizenberg write. “Following a successful economic stabilization plan in 1994, aggregate consumption of soda doubled by 1997, and continued to grow at an annual rate of about 10% through 1999. As is well documented, this growth was fueled by pronounced upward mobility among lower income households, who were no longer forced to pay an inflation tax.”

Mixing and Matching Data
To study the consumption trends, Salvo and Eizenberg examined consumers’ soda buying between December 1996 and March 2003. That process involved pioneering work in mixing and matching data from three organizations. Nielsen supplied regional data on the sales and prices of soft-drink brands. IBOPE, a private-sector demographic data company, provided information on the socioeconomic composition of urban Brazilian households. And details on purchases of various types of soda by different socioeconomic groups came from IBGE, a Brazilian government organization roughly equivalent to a combination of the United States Census Bureau and the Bureau of Labor Statistics.

To interpret the buying habits, the two researchers created an innovative classification of socioeconomic groups. Rather than using the traditional A through E groupings that categorize the richest to the poorest, they defined three classes: poor, established affluent, and newly affluent. “We assume that the As, Bs, Cs that we see in the data right before the income shock are existing established affluent households; Ds and Es we label as poor,” Salvo says. “But there is a mass of households moving from DE to ABC, which we define as newly affluent.”

When poor individuals, who have no disposable income, move up to the newly affluent category, their buying behavior may differ from that of the established affluent in two ways. “They’re perhaps more price-sensitive,” Salvo says. “And since they are new consumers, they haven’t yet formed habits—especially the habit to shop for premium brands.”

A “Persistence Mechanism”
To model that behavior, Salvo and Eizenberg matched the socioeconomic data with information on relative sales of the high-priced, heavily advertised premium brands such as Coca-Cola and the low-priced generics typically produced by small regional companies that routinely appeal to less affluent consumers. The model reveals what Salvo and Eizenberg call a “persistence mechanism” among newly affluent buyers. “It’s not brand loyalty we’re after,” Salvo explains. “It’s about staying with the type of brand you’ve been consuming in recent months. And when the newly affluent buy generic rather than premium brands, they might develop the habit of buying frugally.”

If that habit does develop, it can cause premium brands to lose significant market share to the generics. In Brazil, the share of generic soda brands, numbering in the hundreds, increased from 20 percent to 40 percent during the first three years of the study. Executives at the Coca-Cola Company, the leading seller of premium brands, responded by making the painful decision to reduce their prices by 20 percent.

That decision made sense according to the new model. “This [persistence] mechanism captures a world in which premium brands have to act quickly in the wake of an emerging middle class,” Salvo and Eizenberg write in their paper. “If they wait too long, a substantial mass of the ‘new middle class’ might be captivated by the generic habit. It may then prove to be much more difficult to convince these consumers to pay much higher prices for a highly advertised premium brand.”

National and Global Implications
Ironically, a decision to reduce prices can benefit more affluent groups in the population and the national economy as a whole. “To the extent that firms cannot discriminate between existing and new consumers, by cutting prices Coca-Cola wrote a check to their established consumers,” Salvo points out. “The innovation had impact on prices paid by the rich and hence moderated inflation.”

The study also has broad global implications. “While our application focuses on the very concrete example of the Brazilian soft drink market, we view the issues that are tackled in this work as likely characterizing many consumer goods markets in the developing world,” Salvo and Eizenberg assert in their paper. “Consumers in emerging markets are viewed as an engine of growth for multinational firms and for the global economy as a whole. Understanding the features of demand and the microeconomics of competition in such markets—in particular the tough match premium brands face from local ʻvalue players’—should be of great interest for policymakers and firms alike.”

Related reading on Kellogg Insight

Concrete Collusion: Economic data reveals little competition in Brazil’s cement industry

Reframing the Poverty Problem: Marketing can help address social ills