Commerce Secretary Wilbur Ross, at the request of the Justice Department, has instructed the U.S. Census Bureau to include a question on the 2020 decennial census asking whether the respondent is a U.S. citizen.
Such a request should be relatively uncontroversial, since census takers have been asking that question on one survey or another since the very first census. But these days, even the uncontroversial is controversial.
Most of the pushback is coming from the left, especially politicians and rent-seeking groups that thrive on redistributing taxpayer dollars. But such a question could be very helpful for those who work in public policy — for example, in counting the uninsured.
The Census Bureau included a citizenship question through 1950, stopping in 1960, as it sought ways to increase response rates. But the question was included on what is called the "long form" census in 1970, which went to fewer households. In 2005, the citizenship question was added to the American Community Survey, an annual survey of a very small percentage of households.
Because the Census Bureau's decennial census has not included a citizenship question for decades, analysts and elected officials do not know how many of the respondents are (1) citizens, (2) aliens in the U.S. legally, or (3) undocumented and in the U.S. illegally.
Thus, when the Census Bureau releases its annual survey of health coverage and the uninsured, it simply ignores how many of the uninsured are here illegally.
During healthcare reform debates in the past, some of us pointed out that perhaps 25 percent of the uninsured were undocumented aliens and so unlikely to be covered by health insurance reform efforts — and sure enough, Obamacare excluded illegals from receiving health insurance subsidies.
Even today, of the roughly 27.6 million (2016) uninsured, perhaps 8 million or so — a guestimate — are illegals ineligible for taxpayer subsidies. Very few of them are going to spend their own money, especially given the high cost of Obamacare coverage. They will simply remain uninsured.
And yet those pushing for some type of big-government solution to the uninsured — including those who backed President Barack Obama's Affordable Care Act — use the larger uninsured number to make the problem look bigger than it is, or at least bigger than any likely legislative solution would address.
To be sure, the Census Bureau's annual American Community Survey does ask a citizenship question. But while more frequent, those surveys are limited samples, about 3.5 million out of roughly 126 million households. The sample numbers are then extrapolated for the country as a whole. But even then, the bureau doesn't include an estimate of the uninsured who are in the U.S. illegally.
The result is that estimating the number of uninsured who are illegal has mostly been a guessing game. But the issue is not limited to the uninsured. The federal government funds a number of programs where taxpayer money supports illegal immigrants, either directly or indirectly. Knowing how many undocumented people are receiving those funds could help inform policy decisions.
Those opposing a citizenship question claim that the U.S. Constitution requires the government to count everyone who resides in the country, legally or not.
Ironically, these are mostly the same people who long ago abandoned the notion of a literal interpretation of the Constitution for what they call a "living Constitution," defined as "one that evolves, changes over time, and adapts to new circumstances, without being formally amended."
While a citizenship question might discourage some participation, the Census Bureau has increasingly used various methods, statistical and otherwise, to fill in the gaps. Some form of a citizenship question has been around for two centuries. Stressing it once again would help take a little of the guesswork out of many of our public policy challenges.