Ongoing Supreme Court cases have cast doubt on the ability of universities to continue to use race-conscious affirmative action to achieve institutional diversity goals. Colleges now face pressure to argue that workable, race-neutral alternatives to racial affirmative action do not exist. Some state systems are already using potential alternatives. In this paper we focus on one such alternative—“percent plans”—and determine the extent to which it is a viable alternative to race-conscious affirmative action according to the demographic composition of each state’s school system.

Percent plans guarantee admission to a public, four-year institution for students from that state who graduate in some top percentage of their high school class. Research has shown that existing percent plans do not recreate diversity levels achieved under traditional race-conscious affirmative action; they discourage minority students from enrolling at elite public institutions; and minorities that do enroll are less likely to persist and graduate. In a national simulation, Howell (2010) found percent plans similarly ineffective. This approach obscures that percent plans may vary in effectiveness according to racial segregation between schools in different states. It is an open question whether percent plans will fail in all states, or if their failure is a function of the states that use them. Alternatively, because current percent-plan states relatively racially diverse, the results of each may represent the best possible outcome if such plans were to be adopted more broadly.

Using data on high school and college enrollments by race, as well as class rank data from the National Education Longitudinal Study, we examine the effect of the adoption of a hypothetical percent plans in each state on public postsecondary enrollment. We do this under three different scenarios. First, we assume that the racial composition of the top students in a school reflects the racial composition of the school as a whole. Second, we use regression analysis to predict the composition of the top percent of a high school according to the racial composition of the entire school using the class ranks in NELS. Finally, we use proficiency data from public high schools to predict the racial composition of a class for more recent cohorts than NELS. In each scenario we “admit” a top percentage of students to a hypothetical college class and compare the composition of this class to the actual composition of public university students formed without percent plans. We then link the extent to which the hypothetical percent plans alter postsecondary racial diversity to measures of racial segregation between high schools in a state.

Under assumptions that are generous to the success of percent plans, we find that percent plans are not viable options for maintaining the levels of racial diversity achieved under race-conscious affirmative action policies in any state. We also find a weak connection between school segregation and the degree of percent-plan failure. We conclude by discussing the policy implications of these findings.