Skip to main content

Skilled Workers Flee from Polluted Cities, Hampering Economic Growth

In China, highly educated people are more likely to move away from areas with poor air quality. Reducing pollution could substantially increase GDP there and in other countries, according to a new study co-authored by Yale SOM’s Mushfiq Mobarak.

A cyclist in Beijing on a day of heavy pollution in December 2015.

Beijing on a day of heavy pollution in December 2015. Photo: Lintao Zhang/Getty Images.

In 2014, China launched a “war on pollution,” cutting production in polluting industries and converting coal heating to natural gas. In the years since, the level of dangerous particulates has fallen, although pollution remains high. Research has shown that the health effects of pollution make workers less productive, which means that cleaner air can boost a country’s economy. A new study of pollution and migration in China offers an additional argument for reducing pollution, suggesting that it has a deeper negative effect on economic growth than previously recognized—and that cleaning up cities could have a substantial economic upside.

Mushfiq Mobarak, a professor of economics at Yale SOM, and his co-authors wondered if poor air quality prompted workers to move away from the areas where they would be most productive, dampening economic growth.

“Does that mean that pollution imposes some other cost on society that we hadn’t been thinking about?” he asks. If so, “then reducing pollution will have even larger gains than what we had thought.”

“Pollution is twice as bad for the economy as what we thought,” according to Prof. Mushfiq Mobarak, which means “the benefits of pollution control are actually twice as large as previously believed.”

To find out, Mobarak, with Gaurav Khanna of the University of California, San Diego, Wenquan Liang of Jinan University, and Ran Song of Harvard University, analyzed 18 years of pollution data and migration records in China. They found that, indeed, people tended to flee from polluted cities—but this trend was much more pronounced among skilled, educated employees. Those individuals were leaving places where there was already a scarcity of skilled workers, and relocating to places where they were more skilled people, making their presence a little less useful—causing aggregate productivity to drop. Further, since they leave behind a shortage of skilled labor to manage and train unskilled colleagues, productivity drops further in the places they escape.

The negative economic effect of pollution-driven migration was roughly on par with the effect that pollution has on workers’ health, the team estimates. In other words, “pollution is twice as bad for the economy as what we thought,” Mobarak says. On the plus side, that means improving air quality could substantially increase GDP. “The benefits of pollution control are actually twice as large as previously believed,” he says.

These results also help us understand a long-standing puzzle in economics. Within a country, some areas are more productive and offer higher wages. One might expect that people would continue moving from less to more productive cities until wages even out across the country. But that doesn’t happen. Workers remain stuck in places where they’re not the most productive.

People may stay put partly because moving is expensive, housing is scarce in the other city, or they consider it risky to enter a new job market. But air quality also could play a role; workers might not want to move to a smoggy area.

To investigate, Mobarak and his co-authors examined air quality data in China from 1998 to 2015. They also obtained information about the migration of skilled workers—those with at least one college degree—and unskilled workers.

A clear pattern emerged: skilled workers tended to leave more polluted places. “It’s educated people, richer people, who are much more likely to move,” Mobarak says. “There’s asymmetry in the response.”

Why would that be the case? First, “they care more about pollution,” he says. Wealthier people can afford to care, whereas a poor family may need to prioritize the highest possible wages. Second, China’s hukou policy restricts the public services that people can receive when they move to a new city; getting these benefits is often easier for highly educated than for low-skilled workers.

The result held up when the team conducted more stringent tests to check whether the moves were caused by the pollution and not some other local factor. They compared migration in cities that were downwind of coal-fired power plants—and thus more polluted—to those that were a similar distance from plants but upwind. The researchers also looked at the strength and frequency of thermal inversions, a meteorological event that traps pollution, which creates some random variation in the air quality experienced by city residents.

Skilled workers were more likely to leave both the downwind cities and those with more thermal inversions. For instance, 10% higher pollution due to being downwind of a coal-fired plant was linked to a 1.4 percentage point rise in the fraction of educated workers who moved away, but only a 0.61 percentage point increase for unskilled workers.

Next, the team created a mathematical model of how pollution affected productivity, which allowed them to estimate the economic effects of improving air quality.

For instance, the researchers found that reducing pollution in three major growth corridors in China which offer many economic opportunities for skilled workers—a goal that the government has set in its 12th “Five-Year Plan”—would bump up China’s GDP per worker by 3.6%. But educated workers would reap most of the benefits of that pollution control. To improve equity, migration restrictions for unskilled people would need to be relaxed.

The researchers estimate that 14 percent of the difference in wages between the very productive and polluted city Tianjin and the comparatively cleaner Chongqing is due to pollution.

The team also uses their model to examine how much of the wage gap between pairs of cities in China is due to educated workers’ aversion to pollution. They estimate that 14% of the difference in wages between the very productive and polluted city Tianjin and the comparatively cleaner Chongqing is due to pollution, and that halving the level of pollution in Beijing would raise incomes by 12%. For countries interested in generating economic growth by encouraging citizens to move to the places where they would be most productive, Mobarak says, those figures give a sense of how much effort should be directed toward improving air quality.

Companies that want to encourage their talent to move to offices in cities where they can generate more revenues and profits also need to know about the impact of pollution, Mobarak notes. “It is no surprise that many Chinese companies and multi-nationals now offer ‘pollution hardship allowances’ or ‘pollution wage premiums’ to encourage employees to take up positions in economically attractive markets like Beijing,” he says.

Although air quality is generally much better in the United States, places like California, where wildfire smoke could drive people away, may also need to grapple with this issue. Many tech companies are establishing offices in other states such as Texas and Colorado; as pollution in the Bay Area worsens, high-skilled workers may find it less palatable to stay there.

When companies are deciding where to locate their offices, “it’s becoming more and more important to pick cleaner cities,” Mobarak says.

Department: Research