Professor Moglen, I've decided to write a new paper. My first paper was not well organized, confused, and not salvageable.

Cost Effective Affirmative Action

Much of the criticism of affirmative action is based on the assumption that affirmative action gives preferences to less qualified candidates. Opponents argue that as a result, the optimal output is not reached because the most productive are not rewarded. Instead of challenging the assumption that less deserving candidates are rewarded or the naïve conclusion that productivity is lowered, proponents of affirmative action argue that affirmative action is justified by the societal benefits of increased diversity. In Grutter, the court held that student body diversity in higher education is a compelling government interest. But, in deferring to the university, the Court did not address whether affirmative action actually generates suboptimal performance. However, there is a good deal of economic literature that investigates the effects of affirmative action on optimal outputs. The literature finds that even assuming that only less productive competitors are subsidized, affirmative action actually increases the overall level of production by forcing unsubsidized competitors to produce more than they would in the absence of affirmative action. For example, in Deficit Reduction Through Diversity: How Affirmative Action at the FCC Increased Auction Competition, 48 Stan. L. Rev. 761 (1996) (http://digitalcommons.law.yale.edu/cgi/viewcontent.cgi?article=2520&context=fss_papers), the authors show how at the FCC’s 1993 auction for radio spectrum licenses, giving “bidding preferences [to minority owned firms] increased the government’s revenue by more than 12 percent” (763). The moral is that affirmative action, far from being costly, may actually be cost effective: “Bidding subsidies for weak bidders – far from being ‘giveaways’ – can prevent giveaways by forcing relatively strong bidders to bid close to their reservations prices” (775).

The cost effectiveness of affirmative action has never, to my knowledge, been brought before the Court as a reason for its continuation. The question is whether affirmative action’s ability to increase efficient production will have any currency with the Court? The imposition of burden just for efficiency gains seems to violate “the basic concept of our system that legal burdens should bear some relationship to individual responsibility.” Frontiero v. Richardson, 411 U.S. 677 (1973) at 686 (internal quotations omitted). Of course courts have upheld the constitutionality of various compulsory programs that burden individuals for efficiency gains. Examples from torts include workers’ compensation and no fault auto-insurance. At first, workers’ compensation was found to be an unconstitutional deprivation of property without due process because of the lack of fault in placing liability on the employer (Ives v. South Buffalo Ry. Co. (NY, 1911)). But both workers’ compensation and no fault auto-insurance are now well established programs. Courts have variously held that access to a tort remedy is not a fundamental right. Importantly, programs that protect against injury are well established under a state’s police power. A 1974 article analyzing the constitutionality of the then newly enacted New York no-fault auto-insurance law reports that other state courts found that “compulsory "no-fault" insurance is no different from any other type of self-protecting legislation, such as that which requires motorcyclists to wear safety helmets or that which requires compulsory self-insurance under the Social Security system” (http://ir.lawnet.fordham.edu/cgi/viewcontent.cgi?article=2146&context=flr). Most importantly, compulsory no-fault systems are mainly challenged on due process grounds, not equal protection grounds as with affirmative action. Most equal protection challenges to no-fault auto-insurance for example, are based on discrimination against the poor or discrimination against other collateral sources of contribution for recovery. Unlike affirmative action, equal protection challenges to no-fault auto-insurance only require rational review. There is no explicit racial classification that activates strict scrutiny, as in affirmative action. Moreover, since there is no fundamental right to tort remedies and protective measures are well within the police power, due process challenges also do not activate strict scrutiny.

Perhaps closer comparison is the military draft, a program that burdens one group of individuals (only men can be drafted) based only on gender, regardless of potential individual contribution. Although arguably closer to affirmative action than no-fault auto-insurance because it involves a clear equal protection issue, the military draft is also not a good comparison. Firstly, unequal treatment on the basis of gender only activates intermediate scrutiny. In Rostker v. Goldberg, the Court examined the constitutionality of the Selective Service System (not the actual draft were it to be reinstated) and the majority applied a milder form of intermediate scrutiny (see Marshall’s dissent), giving lots of deference to Congress in carrying out its Article I powers to raise armies. Crucially, the Court stated that men and women were not similarly situated for equal protection purposes because women are not allowed in combat positions. (The exclusion of women from combat positions was not challenged.) In the context of affirmative action, it would much more difficult to argue that minorities and non-minorities are not similarly situated for the purposes of obtaining education or jobs.

Perhaps greater efficiency benefits could pass heightened scrutiny at all. The Court held in Reed and reaffirmed in Frontiero and Craig that administrative convenience is not a sufficient justification to withstand heightened scrutiny. High costs in shifting to a non-discriminatory system are not enough to justify discrimination based on a suspect or quasi-suspect classification. The question is whether there is any difference between reducing costs and increasing productivity. Fundamentally, both affect the bottom line. It could be argued that Reed and Frontiero stand for the proposition that the government cannot fail to remedy discrimination just because the remedy is costly. If this is true, surely the government cannot discriminate to generate profits.

It seems unlikely that the cost-effectiveness of affirmative action is enough to independently justify the program. However, it may help to further support diversity as a compelling government interest. Specifically, cost-effectiveness may show that affirmative action programs are more narrowly tailored than previously thought. The dissents in Grutter advocate percentage plans like those used in California and Texas because they produce similar student body diversity. These plans, however, would not increase the overall level of productivity they are based solely on the existing level of competition. Affirmative action, then, would be less replaceable and thus more narrowly tailored.