• 05 Nov 2012
  • Research & Ideas

What Wall Street Doesn’t Understand About International Trade

Firms that correlate their international trading activity with the local ethnic community significantly outperform those that don't, according to new research by Lauren H. Cohen, Christopher J. Malloy, and Umit G. Gurun.
by Carmen Nobel

Making the best international trading decisions may be as easy as taking a stroll around the local neighborhood.

A recent research paper states that it's possible to predict whether a US firm will trade with any given country by studying the ethnic makeup of the nearby community, according to new research. What's more, firms that correlate their international trading activity with the local ethnic community significantly outperform those that don't—a fact that has escaped notice of financial analysts.

The findings could help Wall Street make better earnings performance forecasts, according to the authors of Channels of Influence, by Harvard Business School Associate Professors Lauren H. Cohen and Christopher J. Malloy, and Umit G. Gurun, an associate professor at the University of Texas at Dallas.

“When nearly half of anything a firm does is X, it's important to understand what's going on with X.”

Global sales are critical to the success of most big businesses. In fact, for companies in the S&P 500, some 46 percent of sales came from outside the United States in 2011.

"When nearly half of anything a firm does is X, it's important to understand what's going on with X," says Cohen.

The research team reasoned that firms would most likely export products and services to countries where they had strong informational ties—and that people with personal knowledge of those countries could help firms to form those bonds.

"The hypothesis was that the surrounding ethnicity would help them to translate and transfer information about potential international trade decisions," Cohen explains. "We thought maybe the people who have ethnic links back to these countries—immigrants or other people who have a former relationship with the country—can help firms to decide whether it makes sense to do business with that country. What part of that country is the best trade partner? Who are the best contacts?"

Analyzing Data

Of course, immigrants most likely to aid trading decisions are those who work at the firm. But because firms generally don't divulge information about their staff's ethnic makeup, the research team had to look at other related factors. For starters, Cohen, Malloy, and Gurun had access to ethnicity data of US metropolitan areas (from the Census Bureau and the American Communities Project at Brown University). They also were able to determine the nationality of firms' corporate board members, using data from a private research company specializing in biographical information on company officials. Looking at the data, the team found a large correlation between the ethnic makeup of a firm's board and the ethnic makeup of the surrounding neighborhood. "So if you have tons of Vietnamese people in the neighborhood, you also have a lot of Vietnamese people on the board," Cohen says.

To determine which companies were trading with which countries, the researchers looked at data from the Journal of Commerce's Port Import Export Reporting Service (PIERS), which collects and parses shipping information from US Customs and Border Protection. "From there we simply tested: If you have a lot of one country's ethnic makeup around you, are you more likely to import or export from that country?" Cohen explains.

The researchers found that the propensity of a firm to trade with any given country increased by more than 60 percent if the neighborhood surrounding headquarters had a high percentage of immigrants from that country. What's more, "the same effects happen with boards," Cohen says. "If you have a lot of Chinese board members, you import and export more from China. If you have a lot of board members who are Vietnamese, you trade a lot with Vietnam."

Correlation Versus Causation

The team knew, though, that the correlative data did not necessarily prove the causal impact of ethnic ties on international trade decisions. "You could imagine lots of reasons for this correlation," Cohen says. "In California, for instance, you may have more Japanese immigrants because it's the closest point of entry with Japan. And you may also do more trade with Japan because it's the closest place for you to ship. So it's possible for there to be cases in which there's a correlation, and yet one thing does not cause the other."

Establishing scientific proof required a situation in which the team could exogenously change the ethnic population of metropolitan areas near firms. The legality and feasibility of such an experiment seemed unlikely from a human rights perspective.

"We can link the causal effect from the immigrants to the firm trade decisions, even 60 years later.”

However, the researchers decided to focus on a specific period in US history in 1942, following the Japanese attack on Pearl Harbor, when the United States forced more than a 100,000 Japanese Americans to relocate from their homes on the Pacific coast to internment camps in other parts of the country. Not knowing how long the internment would last, many of the internees hurriedly sold their houses and assets before leaving. And so, when they were freed a few years later, many no longer had homes. Others tried to return to the West Coast, only to find that they faced hostility and violence from their neighbors. As a result, after they were freed a few years later, many internees ended up resettling in the regions surrounding the internment camps—including Arizona, Arkansas, Idaho, Wyoming, and Utah. Thus, the Japanese American populations in these areas grew significantly and suddenly.

"The Japanese population in Arkansas in 1940 was literally 3 people," Cohen says. "With the internment camps, the government increased that population by almost 18,000. For sure, this was a huge exogenous shock."

The researchers then looked solely at the international trading activity of firms located near the internment camps that were exogenously shocked with the increased Japanese population. They found that these firms traded significantly more with Japan, thus establishing the causal link between the exogenous population change and trade decisions. When additionally examining only those firms formed before 1946 (when the internment camps were evacuated), they found similarly large impacts on trade with Japan. "We can link the causal effect from the immigrants to the firm trade decisions, even 60 years later," Cohen says.

Analysts Take Note

Having determined a causal effect, the team went on to look at the financial results of companies having trades correlated with the ethnic population, comparing them against firms that did not employ this trading strategy. They found that the former generally outperformed the latter in risk-adjusted returns by at least 5 to 7 percent. "The 'strategic traders'—those that trade in accordance with their ethnic population—have much higher sales, much higher profitability, and much higher stock returns than the 'non-strategic traders,' " Cohen says.

Historically, Wall Street has failed to consider the local ethnic population trading strategy when assessing the value of a firm. In fact, the researchers found that analysts are significantly less accurate in their earnings forecasts on "strategic" trading firms than on non-strategic firms.

"With nearly 50 percent of sales being driven by overseas sales—and the surrounding population being a big driver of that activity—understanding this is crucial to understanding the value of a firm," Cohen says.

About the Author

Carmen Nobel is the senior editor of HBS Working Knowledge.