Often overlooked in the debate over immigration policy and “sanctuary cities” (and now states, with California imposing noncooperation “sanctuary” policies state-wide) is the real boost that such immigrants give to the political power of these jurisdictions.

The gross population, including illegal immigrants, is counted in determining representation in the House and state legislatures, rather than the citizen population. Therefore, the more undocumented residents a jurisdiction has, the more legislative seats it gets, both in Washington and in its state capitol.

Determining how many extra seats go to sanctuary cities (and their states) is difficult. The census used to ask about citizenship, but it no longer does. Last month, the Department of Justice asked the Census Bureau to return a question about citizenship status to the 2020 census. The DOJ’s reason was to gather data to better enforce the Voting Rights Act. However, beyond this perfectly legitimate purpose, we need the census to count citizens as well as total residents in order to better understand how illegal immigration is leading to malapportionment and distorting our political system.

The Census Bureau has attempted to measure the numbers of noncitizens in surveys which count a small fraction of Americans. The results are alarming. According to the resulting estimates, the proportion of noncitizen population in the states varies widely, ranging from 14 percent in California to less than 1 percent in West Virginia.

This has a direct impact on the states’ political power. California, the first sanctuary state, may have five or six more members of the House (and consequently Electoral College votes) because of its noncitizen population than it would if House seats and Electoral College votes were based on only citizen population. This gives the votes of eligible Californians significantly more weight than those of people in other states. No wonder California politicians favor illegal immigration.

This disparity can be even more pronounced within states. In my new book Fifty States, Not Six, I show how New York City has 10 or more seats in the 150 member state assembly (the lower house of the New York State Legislature) than it would if apportionment were based on the citizen population. This is true across the nation, where illegal immigrants tend to concentrate in urban areas, which generally are the areas which have declared themselves sanctuary cities.

Some believe that noncitizens are voting in large numbers. But even if this is not the case, it is undeniable that noncitizen residents are increasing the representation of sanctuary cities in Congress and state legislatures.

In our era of historically high immigration, this has a major impact on legislative apportionment because of their very uneven distribution. A constitutional amendment would be needed to correct this imbalance on the federal level. (Such an amendment is proposed in Fifty States, Not Six.) On the other hand, states arguably could apportion their legislatures based on citizen rather than gross population. However, states rely on the federal census to do their apportionment.

Therefore, in addition to supporting voting rights enforcement, the 2020 census should count the citizen population as well as the gross population, so that states wishing to apportion based on the citizen population would have the information to do so. In addition, the results of these parallel counts might illuminate the selfish motives of sanctuary cities, and build support for a constitutional solution under which our country is governed equally by all of its citizens.

James W. Lucas is an attorney in New York City. He is the author of Fifty States, Not Six: A bipartisan approach to reforming the Electoral College and assuring that every citizen’s vote counts and Are We the People? How We the People Can Take Charge of Our Constitution.

If you would like to write an op-ed for the Washington Examiner, please read our guidelines on submissions here.