In America’s strange legal immigration system, every country receives the exact same quota for green cards—7 percent of the number issued—regardless of how populous it is. When immigrants—mainly Indians, Chinese, Filipinos, and Mexicans—hit these “per-country limits,” nationals of other countries may pass them in line. This creates massive wait times for some immigrants, while cutting the waits for everyone else.

In 2018, for example, employer-sponsored immigrants with bachelor’s or master’s degrees waited more than three years for a green card if they were born in China, and about a decade if they were born in India, while those from other countries waited less than a year. Going forward, the Indian wait will stretch on for decades. The system is unfair, and for that reason alone, Congress should end it.

But the per-country limits are also economically senseless. They prioritize the right birthplace over the right skills. In the employer-sponsored categories, businesses could decide to pay Indian or Chinese applicants much more than other immigrants, yet Indian or Chinese employees would still suffer the same pointless discrimination. Discriminating based on nationality, rather than skills, undercuts the productivity of the United States and lowers the average wage of new immigrants to the United States.

To see if this was happening, I reviewed the data on approved labor certifications submitted by employers in the EB2 and EB3 employer-sponsored immigrant classifications to the Department of Labor (DOL). These labor certification applications contain the wage offered to the immigrant as well as their birthplace. Employers with approved labor certifications may petition for a green card on behalf of their workers, but the worker may only apply for a green card once a visa number is available under the quota. The Department of Labor validates the information provided on the labor certification in order to deal with concerns that immigrants are taking jobs from U.S. workers.

I used the latest DOL wage data from fiscal year 2018 to produce the estimates in Figure 1. To produce the weighted average wage with the country cap, I weighted the wages for each nationality by the number of their nationals admitted under the country caps in the EB2 and EB3 employer-sponsored categories. The average wage without the country cap is the average of the approved labor certification wages in 2018.[*] The weighted average wage with the per-country limits was $95,534, while the wage without it would be $107,126. The per-country limits depress the average wage for new employer-sponsored immigrants by $11,592.

In other words, the per-country limits strongly discriminate against higher-paid immigrants. Figure 2 shows the average offered wage for immigrants from India, China, and the rest of the world. The wages were $118,071 for Indian immigrants, $111,172 for Chinese immigrants, and $90,422 for the rest of the world. Indian and Chinese applicants have wage offers that are $27,649 and $20,750, respectively, more than other applicants. Yet Chinese and Indian immigrants must wait much longer than immigrants from the rest of the world.

The per-country limits strongly discriminate against higher-paid immigrants. Immigrants who are offered higher wages actually wait longer under the U.S. legal system than other immigrants. That said, all employer-sponsored immigrants command much higher wages than the average income for all Americans (about $48,000).

Indian and Chinese immigrants are also more likely to be offered positions that require more experience and skills than other employer-sponsored immigrants. The Department of Labor categorizes jobs into five different “zones,” with Zone 5 commanding the most skills and experience. The average job zone was 4.1 for a position offered to an Indian immigrant, 4.0 for China, while all other immigrants were offered jobs with an average job zone of just 3.7.

EB2-EB3 employer-sponsored immigrants from India and China do not just happen to settle in higher wage but also higher cost areas than other EB immigrants. Out of the top 30 states for labor certifications in FY 2018—where 97 percent of them were filed—the average wage for Chinese and Indian EB2-EB3 immigrants sponsored by employers was higher in every state except for one (see Table 2). Moreover, it was higher in each of the top nine states for labor certifications where three quarters were filed. This explanation doesn’t fit the evidence.

Naturally, at least some of these differences come from forcing Indians and Chinese applicants to wait longer. They certainly do obtain higher wages and more experience while they wait longer for green cards. But whatever the reason for this difference, it makes no economic sense to continue to use country of birth as a factor in determining who receives a green card first.

The United States needs immigrants of all different skill and wage levels, but this diversity should emerge naturally from the free market, not from government attempting to micromanage America’s ethnic ancestry. According to a new study, the arbitrary delays are encouraging Chinese and Indian immigrants to leave the United States and take their talents elsewhere. Congress should repeal the per-country limits, and after that, it should revise or eliminate the arbitrary quotas on employer-sponsored immigrants, which have not been updated in nearly 30 years. The market—not government bureaucrats—should determine who will benefit the United States the most economically.

Updated to include state wage information on 10/13/2018

[*] Notes on methodology: Approved labor certifications include expired ones because they may still have been used to obtain a green card. At the high end of the wage distribution, there were some erroneous entries where wages were listed as hourly, weekly, or monthly when they should have been listed as yearly. As a data integrity measure, I excluded all labor certifications with listed wages of more than $1 million annually as well as anyone making more than $500,000 annually with a job zone of less than 5. This excluded about 30 people of a population of nearly 110,000. The offered wage of immigrants was annualized and, if necessary, was determined by taking the midpoint in any salary or wage range provided by the employer. Job zones were obtained by comparing Standard Occupation Classifications in DOL data to the relevant job zones. EB2-EB3 employer-sponsored immigrants include all EB2-EB3 immigrants except for those who receive “national interest waivers,” but these immigrants do not need a sponsoring employer and so do not have any wage offer. No data exist on their wages after they receive a green card and find jobs in the United States, but NIWs are a much smaller population, and there is no reason to believe that there are systematic differences in the nationalities in the NIW population compared to the labor certification population, so—for both these reasons—even if it were possible to include their post-green card wages, it would make little difference to the relative results between India-China and the rest of the world.