No More General Tso's? A Threat to 'Knowledge Recombination'

Immigrants bring with them innovations from their homelands, knowledge that local inventors often build upon, says Prithwiraj Choudhury. Examples: turmeric medicine, double-entry bookkeeping, and American Chinese food.
by Michael Blanding
Turmeric is prepared into medicine. Doucefleur

In the early 1990s, an Indian plastic surgeon at the University of Mississippi, S. K. Das, was about to amputate the leg of a patient because of a wound that wouldn’t heal. Colleague Hari P. Cohly stopped him. A researcher in immunology from India, Cohly was skilled in Ayurveda, the ancient Indian practice of herbal medicine, and suggested they try using turmeric to heal the wound instead.

The spice worked, and Das and Cohly later conducted Western medical trials to prove that turmeric was an effective treatment for wounds. Together, they filed the first patent for the medical use of the herb.

“This knowledge had been locked in their home country for decades, if not centuries, and now these skilled ethnic migrants were able to transfer the knowledge here,” says Prithwiraj “Raj” Choudhury, assistant professor in the Technology and Operations Management Unit at Harvard Business School.

Perhaps even more surprising was what came next. Seeing the success, American researchers began investigating the use of turmeric as well, combining it with other chemicals to file their own unique patents. “There are now some 27-odd turmeric patents that have been filed in the US patent system,” Choudhury says. “If you look at the way knowledge gets subsequently recombined, then that knowledge production becomes much bigger than the initial transfer.”

Choudhury has long been investigating the role ethnic migrants play within the American workforce. In the current squabble over how many skilled immigrants to let into the United States on H1-B visas, academics have fiercely debated whether they create jobs by bringing new skills to companies or whether they take jobs away by displacing American workers.

“It’s only when ethnic migrants who harbor this knowledge transfer that it moves from one geography to another”

A migrant from India himself, Choudhury argues that debate is too narrow. “Instead of seeing migrants through the lens of whether they create jobs or not, we should view migrants as carriers of knowledge—knowledge that could be further recombined by locals,” he says. “If H1-B was scrapped, or Europe stopped admitting skilled migrants, the knowledge production of the global economy would suffer.”

One vivid example of this innovation transfer is General Tso’s chicken. Many Chinese immigrants to the United States in the 1800s, blocked from more traditional work, opened restaurants. But they had to cater to the American palate by adding sweetness to their dishes and combining Chinese and American ingredients into innovations like chop suey and the aforementioned General Tso’s. The American Chinese food industry was born.

Choudhury explores this phenomenon in a paper forthcoming in the Strategic Management Journal, The Ethnic Migrant Inventor Effect: Codification and Recombination of Knowledge Across Borders, written with HBS doctoral student Do Yoon Kim.

Throughout history, Choudhury contends, dozens of migrants have unlocked new technologies in countries where they’ve migrated.

  • When French Huguenots fled religious persecution in the 18th century, they came to Brandenburg-Prussia, where they brought secrets of cloth dyeing and silk production that transformed the local economy.
  • When Indian immigrants came to South Africa in the late 19th century, they brought with them accounting practices such as “double-entry” bookkeeping, which changed the country’s financial system.
  • More recently, hundreds of Soviet mathematicians came to the United States in the 1990s after the fall of the Soviet Union; far ahead in fields like partial differential equations and symplectic topology, they helped their American counterparts solve otherwise intractable problems.

“Knowledge that is locked in geographic regions doesn’t transfer globally instantaneously,” Choudhury says. “It’s only when ethnic migrants who harbor this knowledge transfer that it moves from one geography to another.”

Innovation transferred

There are two reasons for that delay, he says. The first is uncertainty for all involved about whether knowledge from one region will work in a new context. The second is that knowledge carried by migrants is often tacit in nature, consisting of specialized techniques that haven’t necessarily been written down. This knowledge needs to be transferred by the experts who hold it.

Which brings us back to turmeric. Seeking to quantify the extent of knowledge transfer by migrants, Choudhury came across a perfect natural experiment to explore. With the Y2K bubble in the early 2000s, the quota of H1-B visas first increased and then fell sharply, initially expanding and then limiting immigration by skilled workers into the United States. That shock not only affected high-tech firms; the limits also impacted pharmaceutical companies, which had been flying in scientists from China and India who, like Cohly, brought knowledge of native herbs.

Some pharma companies were able to earn exemptions from the H1-B limits. Choudhury and Kim used this discrepancy to create a database of some 758 herbal medicine patents filed between 1977 and 2013, then compared the firms that received exemptions versus those that didn’t during the 1999 to 2003 shock. In the period when H1-B quotas were increased, they found a 4.5 percent increase in herbal patents by firms that were exempted from the cap.

Continuing to follow the companies after the shock, they learned these firms were more likely to continue pursuing patents that recombined herbal remedies with other technologies. Surprisingly, first-generation migrants weren’t the ones doing the recombining; they were sticking with their original technologies. Instead, Choudhury and Kim found that mixed teams of migrants and natives or solely native teams were taking the new technologies and making something new.

Digging deeper into social networks, they found that the native teams most likely to file recombination patents were ones with social ties to ethnic migrants, which helped US firms speed the development of new technologies.

“Knowledge transfers through social networks,” Choudhury says. “If a firm created mechanisms where the locals would work with migrants, then that knowledge would transfer and potentially be combined by the locals.”

A question of policy

On a broader level, the findings argue for more open immigration policies for skilled workers. Even if such a policy didn’t immediately translate to jobs, it could translate to the development of new technologies that could spur job creation in the long run.

“If the world just erects walls and doesn’t let skilled migrants come in, there will be the loss of the first-order effect, which is the knowledge being transferred from their geography to here,” Choudhury says. “But the bigger loss would be the recombination loss.”

Over time, that could lead to less innovation in the United States, and—as migrants go to other countries—a potential loss of global competitiveness.

“It’s not only a loss to the firms, it’s also a loss to local workers, who will lose out if they do not imbibe this knowledge from migrants,” Choudhury says. “I would turn the debate on its head and say, it’s not about destroying or creating jobs, it’s about participating in or losing out on creating knowledge.”

Michael Blanding is a writer who lives in the Boston area.

About the Author

Michael Blanding is a writer based in Brookline, Massachusetts

Post A Comment

In order to be published, comments must be on-topic and civil in tone, with no name calling or personal attacks. Your comment may be edited for clarity and length.