The comments by American Enterprise Institute (AEI) economist Nicholas Eberstadt may add a new layer to a fierce political battle over what questions the government should include on its decennial survey of U.S. residents.
In prepared remarks, Eberstadt said the United States has failed to keep good records on people with criminal records, leaving a gaping hole in the country’s economic statistics. His argument centers on his earlier research, showing that men with arrest records are more likely to be unemployed than those who have not had any trouble with the law.
Although it is too late to include a question on this issue on the 2020 Census, doing so on future surveys could help the federal government to formulate policies to get ex-convicts back into the labor force, he said.
“It is an enormous blind spot and, given the realities of life in our country today, a critical and inexplicable statistical oversight,” Eberstadt will say in a Joint Economic Committee hearing about the economic impacts of the 2020 Census, and how businesses use the underlying data.
Eberstadt’s 2016 book, “Men Without Work: America’s Invisible Crisis,” examined why men of prime employment age were not working as much as they did in prior generations, and what effects that has on society.
In his statement on Wednesday, Eberstadt will cite Justice Department statistics showing that 110 million Americans had an arrest record in 2016, more than twice as many as in 1997.
The Trump administration wants to ask respondents to next year’s survey whether they are citizens, a proposal that has drawn a line between Republicans, who generally support the question, and Democrats, immigrant advocates and demographers who say it will stop some people taking part.
That could result in undercounts, they say, which could deprive some communities of funds and political representation because the Census determines how the federal government distributes some $900 billion in aid, as well as seats in Congress.
Harvard researchers predict the citizenship question would lead to an undercount of more than 4 million Hispanic residents.
The issue has been tied up in litigation since Commerce Secretary Wilbur Ross first announced plans to include it in March 2018. The U.S. Supreme Court is expected to issue a decision on the matter soon.
The Census Bureau tries to balance a desire to obtain as much useful information as possible with the reality that fewer questions lead to better response rates, and people are less likely to participate if controversial questions are added.
Adding a new question to the Census usually requires extensive testing and consultation with lawmakers.
Eberstadt is one of four panelists scheduled to speak at Wednesday’s hearing. Others include Andrew Reamer, a research professor at George Washington University; Howard Fienberg, a lobbyist for marketing research and data analytics with the Insights Association; and Mallory Bateman, coordinator of the State Data Center at The University of Utah.
Prepared remarks from other panelists emphasized the importance of Census data for businesses and the broader economy.
Reporting by Lauren Tara LaCapra; Editing by Sonya Hepinstall