The question is simple: Does affirmative action work? The attempt to answer it, however, has created a tangled mass of executive orders, legislative actions, and judicial decisions from Massachusetts to California.
Into this setting comes one of the first comprehensive surveys of affirmative-action data, yielded by dozens of research studies done over several decades. Its conclusion: Affirmative-action programs have been moderately successful, and the trend toward rolling them back is a "costly and dangerous experiment."
The study, by Harvard sociologist Barbara Reskin, comes at a crucial time. In recent years, affirmative action has suffered a backlash propelled by charges that it has generated its own abuses, such as reverse discrimination. Indeed, California has already banned affirmative-action programs in state employment and public universities (in 1996), and Washington State will vote this November on whether it will do so. All told, some 25 states are considering legislative or ballot measures aimed at rolling back affirmative action.
In her survey, Ms. Reskin found that:
* Racial discrimination in employment remains widespread.
* Though enforcement of affirmative-action programs is weak, such policies have lessened racial bias in the workplace.
* Neither whites as the dominant racial group nor meritocracy as a workplace principle is suffering to any significant degree because of affirmative action, despite perceptions that they are.
* The debate over affirmative action has been inflamed by opposition to numerical race or gender-based quotas and preferences in hiring - even though such tools are rarely used. When they are used, they comply with tightly defined, temporary conditions set by a court to remedy proven violations of the law.
Action and reaction
The real problem of affirmative action is "too little, not too much," says Reskin, who unveiled her study last week at the American Sociological Association's annual meeting here.
Critics, however, are not about to reverse field. "It's against much of the real world to assert that no discrimination in favor of minorities is taking place," said Michael Rosman of the Center for Individual Rights in Washington.
Ward Connerly, the architect of California's dismantling of affirmative action, said too often programs that start with defensible goals and timetables "create a mind-set" that leads to rigid quotas in practice. "I prefer a policy that says we will not tolerate discrimination and that everyone who applies receives equal treatment and equal opportunity. But a policy that does not try to dictate outcome" such as specific quantities of specific race groups for a particular job, category, or company.
Tracing the history of affirmative action, Reskin says longstanding laws against racial discrimination have proved inadequate for workplace integration because they provide remedies only after the fact. Affirmative action required proactive steps to provide equal employment opportunity.
One of the most pervasive forms of affirmative action is the requirement that contractors with the federal government establish goals and timetables for hiring minorities in proportion to their share of the qualified job pool in that market. Contractors with the federal government tend to be large firms and represent about 20 percent of the nation's work force, according to a number of experts.
Another 20 percent of the US work force, according to Reskin, is employed in private firms that voluntarily use affirmative-action programs. These efforts, in some cases, use race or gender as "plus factors" for qualified applicants in job categories with histories of racial segregation.
The most stringent affirmative-action programs are those imposed by the courts on employers found guilty of discrimination. While these can include quotas, court-imposed affirmative action remains relatively rare, Reskin says.
All in all, fewer than half of American firms engage in affirmative action, according to the survey. This, despite signs that racial job segregation remains a problem - and for some groups is actually worsening. Reskin says the wage gap between African-American college graduates and their white counterparts has been widening since 1979. And though that wage discrepancy has many causes, one of them could be hiring discrimination that shunts minorities to poorer jobs, despite their qualifications.
In job contracting with the federal government, though, affirmative action is producing gains for minorities, says Reskin. Studies have shown African-Americans making faster employment gains at firms with government contracts than at firms without contracts. And in federal government as a whole, Hispanics' share of jobs increased by 3 percent from 1982 to 1995.