Why Census Is Right to Ask for Racial and Ethnic Data

Peter Skerry
Peter Skerry Former Brookings Expert, Professor of Political Science - Boston College

April 16, 2000

CLAREMONT—As Americans complete their census questionnaires, it is evident that the objectionable questions are those about race and ethnicity. As a colleague complained about the short form, which is sent out to almost 85% of households, “All the U.S. government wants to know about me is what color I am.” Already surfeited with racial controversy and hostile to race-conscious policies, many Americans are clearly unhappy to find one-quarter of the census short form devoted to race and ethnicity.

Does it make sense to ask these questions when without them the census might obtain higher levels of cooperation? Why do we require the Census Bureau to address head-on what most other government agencies can avoid or at least finesse: inquiring about the racial and ethnic identity of each and every American?

To be fair, the 2000 census is dominated by race and ethnicity because the Census Bureau eliminated almost half the questions that appeared on the 1990 short form, leaving the race and Hispanic-origin questions more prominent than ever.

Yet, not only whites are unhappy with the bureau’s efforts in this regard. African Americans, in particular, cannot forget “the first undercount,” when the 1790 census, as prescribed by the Constitution, counted slaves as three-fifths of a person. Today, minorities, as well as nonminorities, are distrustful of the uses to which census data might be put, despite repeated assurances from the Census Bureau that all responses are strictly confidential. The recently renewed controversy over the bureau’s involvement in the wartime internment of mainland U.S. Japanese further fuels distrust among minorities, which is clearly one reason why disproportionate numbers of them do not get counted in the first place.

The bureau is in an awkward position when justifying its gathering of racial and ethnic data. One possible rationale is that such information has always been collected, starting back in 1790. But, again, this history hardly reassures minorities.

A more frequently heard rationale is that the Census Bureau is merely collecting the data necessary to administer the laws of the land. But this is not very satisfying to the many Americans who object to the race-conscious orientation of many of these laws. Never mind the irony that the census itself functions unlike affirmative action. Think about it. If a student identifies herself as a minority on a college application, she improves her chances of admission. Yet, identifying as a minority on the census form offers no such direct benefits, which is one reason why there is a minority undercount.

The bureau’s favored rationale is to emphasize the self-interest of all Americans in obtaining a complete count to maximize their communities’ federal funds and legislative representation. But this is pretty weak stuff. It is obviously unclear to many Americans that they will benefit individually from filling out the forms. As the affirmative-action example highlights, many minorities do not believe so, or at least they don’t believe any such benefits to their group or community outweigh the risks to themselves individually from cooperating with the census. Besides, the more the Census Bureau stresses self-interest to minorities, the more disaffection this generates among nonminorities upset about race-conscious policies.

Is there a solid reason to ask Americans about their racial and ethnic identities on the census? Yes. But first we need to be more realistic about what to expect from census data. Minority advocates and their allies need to recognize that while the minority undercount is real, its consequences—fiscal as well as electoral—have been grossly exaggerated. After all, not being counted in the census is regrettable and arguably has some impact on minority legislative districts, but it hardly deprives minorities of their real source of power: the franchise.

On the other hand, Americans disaffected with race-conscious policies need to understand that it is not the census that is causing dissension over race in the United States. The census is merely the messenger. If Americans are upset with affirmative action and similar policies, they should, as they have, attack these policies, not the census.

We have all gotten caught up in the notion that demography is destiny. But census numbers are just not as important as what policymakers do with them: whether drawing district lines, constructing funding formulas or setting affirmative-action goals. As Rep. Barney Frank (D-Mass.) once put it: “How many members of the National Rifle Assn. are there? I don’t know. I don’t think my colleagues know. What’s important, politically, is not how many there are, but what you do about it. The extent to which you mobilize enormously outweighs the numbers.”

Those who think the bureau’s multiracial option on the 2000 form is evidence that the entire scheme of counting by race is about to collapse of its own complexity are deceiving themselves. Race and ethnicity are real, if frequently misunderstood, forces in American life. Whatever the future of the multiracial option, race is not about to disappear from U.S. politics.

Finally, those most disaffected with the national discourse on race need to realize that data gathered by the census are their best weapon in arguing against policies they find objectionable. If, for example, minority advocates downplay racial progress, census data will help make the case against them. To be sure, those advocates will also rely on the data to make a case.

That’s the point. Racial and ethnic data from the census allow each of us, individually or as members of groups or organizations, to make self-interested or partisan arguments. This is not always a pretty sight. But, over time, census data used in this way allow us to assess racial progress—or lack of it. The alternatives to the collection of racial and ethnic data by the census are ignorance, malice, paranoia, opportunism and prejudice, of which there is already an abundance across the political spectrum.