Nestled in the wooded hills lining the Hudson River lies the stately Blithewood, home of the Jerome Levy Economics Institute. In these serene sylvan environs, Joel Perlmann of Bard College and Mary Waters of Harvard University recently hosted a conference entitled "Multiraciality: How will the new Census data be used?" They and their guests spent two days discussing Census 2000 and the multiracial population, one of the most contentious issues of the past decade.

Throughout the 1990's, Washington D.C. buzzed with concern over how Census 2000 would classify the increasing number of Americans who identify with more than one race. Two positions defined the spectrum of the debate: a "multiracial" category, or no changes.

In 1997, the Office of Management and Budget announced a hybrid approach. While the race categories would remain essentially intact, people completing the race questions for Census 2000 would be able to check as many boxes as they liked. While this approach allowed multiracial people to describe their rich heritage, it created another thorny question: how would the government count people who checked multiple race boxes.

In March 2000, OMB announced that for civil rights enforcement, the federal government would maximize the number of minorities by counting anyone checking multiple boxes as a member of the relevant minority group. When enforcing civil rights laws, the federal government will ignore people's self-identified race, and collapse them into the same race boxes that started the brouhaha.

To better understand the problems created and solved by the OMB's 1997 and 2000 decisions, Perlmann and Waters invited scholars from academia and government to their conference. Many attendees were actors in the continuing drama over how to count people who identify with more than one race. Despite the impressive credentials each participant brought to the discussion, the list of presenters betrayed an unmistakable bias: no representatives of the multiracial community had been invited. Moreover, the unspoken sentiment of virtually every attendee was that race and discrimination continue to distinguish America at the dawn of the 21st century.

Discussion during virtually every session turned to presentations by David Harris and Roderick Harrison. David Harris' research highlighted two phenomena. First, the socio-economic characteristics of multiracial people depend on how one defines multiracial. Second, and more importantly, Harris noted that individuals alter their racial identity depending on the social context.

Harris' first point is rather mundane: how one defines any population determines the characteristics of that population. His second observation, however, is fraught with greater import. If social context determines a person's race, race really is an artificial category.

This contextual vision of race is particularly important for civil rights enforcement. The Department of Justice (DOJ) and other federal agencies rely on the race data generated by the Census to enforce federal civil rights laws. Harris' insight, however, notes that these data would vary significantly, depending on the context. The racial make up of a county would differ significantly, depending on whether DOJ was enforcing the Equal Credit Opportunity Act, or whether the Department of Education was suing for a school district to integrate itself. Defenders of America's racial data collection regime would be hard-pressed to defend the system in the face of such volatility.

Roderick Harrison noted a second source of volatility in Census 2000's race data. He noted that the OMB's March 2000 guidance amounts to a modern resurrection of America's infamous "One Drop" rule. According to the "One Drop" rule, states classified any person as black who had "just one drop of black blood." While OMB's modern incantation might serve less venal purposes, it nonetheless inflates the number of minorities. Whether this inflation can withstand legal challenge is an open, and ominous, question.

The 2000 guidance inflates the number of minorities by dropping the "white" from anyone who checked "white" plus another race category. While the number of people who checked white plus another minority category will likely be very small, in some counties the OMB guidance could significantly alter the racial make up. Using race data from earlier surveys, Harrison compared the number of minorities tabulated according to OMB's March 2000 guidance, and the number tabulated if the white category had not been ignored.

His data show that in some counties, the variation can be quite dramatic. For example, in Sevier county, Tennessee, the ratio of blacks tabulated by the 2000 guidance and the number of blacks without the 2000 guidance is more than 1.5. In other words, the 2000 guidance tabulates 1.5 black people for every 1 black person tabulated without the guidance. In Pima County, Arizona, the Native Hawaiians and other Pacific Islanders ratio is 2.1, and in Yakima County, Washington, the ratio for Asians is 1.4.

These local ratios are important, because the enforcement agencies must use local populations. While the OMB is aware that their decision inflates the size of minority populations, it is not at all clear whether this inflation can withstand legal challenge. Indeed, David McMillen, a senior staff member with the House Committee on Government Reform, noted his belief that Congress and the federal agencies would likely have to revisit both the 1997 and 2000 decisions within 5 years. Many conference attendees expressed similar concerns.

How the problems noted by Harris and Harrison play out over the next decade remains to be seen. People who believe government should divorce itself from racial classification should take courage from these findings. The priests of governmental race classifications are beginning to recognize the fragility of their temple. It is only a matter of time before the whole system implodes of its own weight.

©2001 all rights reserved.
Reproduction in whole or in part prohibited without
the express written consent of Interracial Voice.