Notes for a lecture delivered March 18, 2004:

Notes for a lecture delivered March 18, 2004:

THE BENEFITS AND COSTS OF

AFFIRMATIVE ACTION IN HIGHER EDUCATION:

LESSONS FROM THE UNITED STATES AND INDIA

OUTLINE

(click on any section title below to reach the corresponding section)

INTRODUCTION

*****************************************************************************************

INTRODUCTION

Affirmative-action policies in the United States and reservation policies in India are the two most important examples of positive discrimination in favor of members of under-represented racial/ethnic groups. It is therefore of considerable interest to compare how such policies have worked over the past 40 years in each country. (This is the period during which such policies have been in effect in the US; the Indian experience goes back much further, but it has been most significant over these 40 years.)

In Affirmative Action in the United States and India: A Comparative Perspective (Routledge, 2004) I undertook a comparative review of the existing literature and research on affirmative-action policies in the US and reservation policies in India, focusing especially on the sphere of higher education. Here I will report on key conclusions from the book; the sources and details of the reported empirical evidence can be found in the book.

The key question motivating my work is:

Should affirmative-action/reservation policies in admissions to higher educational institutions be maintained?

To answer this question, as in the case of any public policy, we should identify and compare benefits & costs – as best we can.

NB: I will use the following abbreviations throughout this document:

AAHE = affirmative-action/reservation policies in admissions to higher educational institutions

UREG = under-represented racial/ethnic (or caste or tribal) group

A. THE BENEFITS AND COSTS OF AAHE

In general, relevant evidence is a good deal more plentiful in the US than in India.

In the US, the UREGs favored by AAHE are African, Hispanic and Native Americans.

In India, the UREGs favored by AAHE are the Scheduled Castes [SCs], Scheduled Tribes [STs] and Other Backward Classes [OBCs]. I confine my attention to SCs and STs, because these are the groups most comparable to US UREGs (~25% of the population) and because of the very limited extent of data on OBCs.

1. What AAHE does

In US undergraduate schools: it increases the prospects of admission and enrollment of UREG applicants, and it decreases the prospects of admission and enrollment of non-UREG applicants, only to the best – the top 10-20% – of colleges and universities.

In US graduate and professional schools: it redistributes students (as above), and it adds significantly to the total number of enrolled UREG students.

In India, at all levels of higher education: it both redistributes UREG students upward in the quality hierarchy and results in the enrollment of significant numbers of UREG students who would not otherwise pursue higher education.

It is an inevitable consequence of AAHE that conventional measures of the academic qualifications of incoming students will on average be lower for AAHE beneficiaries than for other admitted students – and lower than those of the non-UREG applicants displaced by AAHE.

NB: There is evidence from both the US and India that the entry-test-score gap between UREG and non-UREG students has been narrowing over the past few decades.

2. The most often cited possible benefits of AAHE

(the main underlying goals served by the benefits are indicated in italics)

(a) compensation of victims of past discrimination & mistreatment (equity)

(b) redistribution from the well-off to the poorly-off (equity)

(c) greater accuracy in appraising applicant qualifications, so that talented UREG members have the same opportunities to acquire a good higher education as talented non-UREG members (efficiency)

(d) better quality of education for all students, due to greater student diversity (efficiency)

(e) more even spread of social capital, helping UREG students overcome historically limited access to prominent persons and institutions (equity, efficiency)

(f) a more integrated societal elite (by race and ethnicity), as more UREG members move into top positions previously dominated by non-UREG members – generating a variety of benefits:

-- it enables people from all salient communities to participate in the society's decision-making processes, directly or via representation in key decision-making positions (democracy);

-- it promotes better performance of those people-oriented jobs involving representation, service, etc., which are better performed by UREG job-holders who best understand and empathize with UREG clients or customers (efficiency);

UREG members in responsible positions serve as role models and mentors (efficiency);

UREG members in responsible positions contribute more voluntary community service than non-UREG members (equity, efficiency);

UREG decision-makers offer more job opportunities to UREG members, helping to overcome the opposite bias of non-UREG decision-makers (equity)

N.B.: Benefits of type (c) and (d) are cited by proponents of AAHE in the US but not in India.

3. The most often cited possible costs of AAHE

(the main underlying goals dis-served by the costs are indicated in italics)

(a) mismatch of UREG students with universities, placing them in educational environments in which they are set up for failure (efficiency, equity)

(b) dependence and complacency on the part of UREG members (efficiency, equity)

(d) resentment about the unfairness of admissions procedures in higher educational institutions (social harmony)

(e) exacerbation of racial/ethnic divisions (social harmony, democracy)

(f) devaluation of the accomplishments of members of racial/ethnic minority groups (equity, efficiency)

Next I will bring theory and evidence the bear on the validity in the US and India of the most often cited benefits and costs of AAHE.

4. The validity of the six main benefit claims

(a) compensation of victims of past discrimination and mistreatment

The main beneficiaries of AAHE are most often children of relatively well-off UREG families, who are less likely to be exposed to bias and discrimination than most UREG families. Furthermore, there is no good reason why only those non-UREG applicants who are displaced by AAHE should be singled out to pay the compensation.

So I reject the compensation claim; AAHE is poorly tailored to achieve just compensation.

BUT: AAHE does represent an effort to ameliorate the position of UREG members in a society; its abolition would therefore be seen by many as a major setback – unless it were replaced by a significant alternative policy to achieve the same end.

(b) redistribution from the well-off to the poorly-off

-- AAHE does not reduce inequalities within UREG communities. There is much evidence from both the US and India that the vast majority of AAHE beneficiaries come from a UREG "creamy layer."

-- AAHE may reduce inequalities across UREG and non-UREG communities. There is some evidence from US that AAHE beneficiaries are less well-off than AAHE-displaced applicants. There is little relevant evidence from India. Even if favorable redistribution results from AAHE, there are obviously much more direct and effective ways to accomplish this objective.

So I reject the redistribution claim; AAHE is poorly tailored to achieve progressive redistribution.

Evidence from the US shows that Black students tend to achieve lower GPAs not only relative to other students but also relative to predictions on basis of their own average SAT scores. Evidence from India shows that SC and ST students perform considerably less well academically than non-UREG students – more so than in the US. This suggests that the true potential of AAHE beneficiaries for good academic performance is not undervalued by conventional qualification indicators, and that AAHE beneficiaries will not perform as well as displaced non-UREG applicants.

BUT: There is evidence from the US and India that the gap in performance between UREG and non-UREG students is considerably less in post-university career achievements than in within-university conventionally-measured academic performance. The same is surely true as between AAHE beneficiaries and displaced non-UREG applicants. This suggests that a more comprehensive measure of performance than GPA would show less evidence of "underperformance."

In any case, AAHE is not well-tailored to identify those applicants whose conventional admissions characteristics most under-estimate their potential (either for academic or career performance). If this were main objective, one should look for evidence of poorer past learning environments, adverse circumstances overcome, etc., which would be better indicators than simply the racial/ethnic identity of an applicant.

So I reject the claim that AAHE provides greater accuracy in appraising applicant qualifications for academic work and probably also for ultimate career success.

(d) better quality of education for all students, due to greater student diversity

In the US: there is plenty of evidence that AAHE has significantly increased student racial/ethnic diversity.

In India: there is strong evidence that AAHE has significantly increased enrollments of SC&ST students.

It is important to distinguish academic learning and ability to function well in a multicultural society. Theoretical reasoning suggests racial/ethnic diversity is more likely to generate benefits of the second type – except in fields in which multicultural awareness important element in understanding of the subject.

In the US (but not in India) there is some highly suggestive evidence that greater student diversity increases student ability to function well in a multicultural society, and there is some (contested) evidence that greater student diversity increases student academic learning. Such positive outcomes require informal inter-ethnic interactions of a kind more likely to occur in US than in Indian higher educational institutions.

So I accept in part the claim of better quality of education benefits, on several grounds:

-- it is plausible that there are such benefits in that students are enabled to function better in a multicultural society;

-- it is implausible that there are such benefits in terms of improved academic learning, except in fields where multicultural awareness is important.

(e) more even spread of social capital

The historical evolution of the US and India has left UREGs significantly lagging not only in financial and physical capital, but also in human and social capital. Educational institutions enable individuals to accumulate human capital; and higher educational institutions are especially important in enabling individuals to acquire social capital. Greater numbers of UREG members succeeding in such institutions – especially elite ones – result in a more equal distribution of social capital across groups.

The evidence shows that AAHE beneficiaries graduate from selective schools in both the US and India at somewhat lower rates than non-UREG students. The more selective the school, the smaller the differential.

In the US: Many studies have shown that students get a significant career boost from attending a relatively selective higher educational institution. (This is certainly implied by the intense competition for entry into prestigious institutions.)

In India: There is a paucity of systematic research on this question; but the US finding no doubt applies there as well. (Consider the violent reaction on the part of "forward caste" members against the expansion of reservations for "backward caste" members in 1990, when such recommendations of the Mandal Commission were adopted by the Government of India).

Also in the US: Several studies have concluded that the economic payoff to attending a more selective college is significantly greater for Blacks (and low-SES students) than for Whites (and high-SES students). There is no systematic research on the extent to which social capital vs. human capital accounts for the payoff to higher selectivity. However, the fact that the payoff to college selectivity is greater for UREG than for non-UREG students suggests that social capital is indeed an important part of the story.

So I accept the claim that AAHE spreads social capital more evenly, thereby enabling UREG students to compete for post-university jobs on a more level terrain with non-UREG students.

(f) a more integrated societal elite, generating benefits in terms of a more effective democratic system and more effective societal institutions

In both the US and India: the evidence shows that AAHE has substantially increased admissions of UREG applicants to elite institutions, and most UREG beneficiaries have done well enough to graduate.

In the US: The evidence shows that UREG students, and AAHE beneficiaries, have gone on to pursue successful careers in esteemed positions. The evidence is abundant and strong for elite colleges; it is less abundant, but fairly strong, for elite graduate and professional schools.

In India: The few studies done suggest that most UREG students who graduate from relatively elite institutions end up in responsible and well-paying positions – even though they do not do as well as their non-UREG peers.

BUT: have successful UREG graduates strengthened democracy and the effectiveness of societal institutions?

In both the US and India, UREGs are now somewhat less under-represented in society's elite positions than in past; so US and Indian democracy have certainly gained legitimacy, which is arguably important for effectiveness.

In the US: Many studies suggest that UREG graduates do better than non-UREG graduates in elite jobs working with multicultural co-workers or serving a multicultural clientele (e.g., law, medicine). Some studies suggest that elite UREG graduates make more community service contributions than non-UREG graduates, both to their own community and to society in general.

In India: there is scattered evidence that SC and ST graduates are more likely than other graduate to make service contributions, less likely to pursue purely materialistic goals, and more likely to pursue careers in a way in which they can be helpful to members of communities in need.

So I accept the claim that AAHE contributes to a more integrated societal elite, which is probably the most important AAHE benefit.

5. The validity of the six main cost claims

(a) mismatch of UREG students with universities

Ø Consider first: AAHE beneficiaries vs. other students.

In the US: There is much evidence that UREG students in general – and even more so AAHE beneficiaries – do not do as well (on average) as other students in conventionally-measured academic performance at higher educational institutions, both at the undergraduate and the graduate level.

BUT: UREG students and AAHE beneficiaries do nonetheless most often successfully complete their degree programs.

In India: The average academic performance of SC&ST students (almost all of whom are AAHE beneficiaries) is distinctly worse than that of other students, and graduation rates are considerably lower.

BUT: The differential between SC&ST and other graduation rates not as great in elite higher educational institutions.

AND: In both the US and India, over time, the graduation rate differential between UREG/AAHE students and non-UREG/AAHE students has been declining.

Ø Consider now: AAHE beneficiaries in more selective vs. less selective schools.

In US colleges: AAHE beneficiaries at more selective colleges attain lower GPAs and lower class ranks than they would have at less selective colleges (the most likely alternative); but they tend to graduate at higher rates, they more often attain advanced degrees, and they gain greater earnings and career satisfaction. The evidence suggests that it is in the long-run interest of any student to attend the most selective college to which he/she can gain access; this is even more true for UREG than for non-UREG students.

In US professional schools: UREG students (many of them AAHE beneficiaries) perform less well in their studies than do non-UREG students; but the great majority do graduate, most go on to successful careers, and their prospects are greater the more prestigious is the school from which they receive their degree.

In India: The limited available evidence from elite higher educational institutions suggests that SC&ST students graduate at reasonable rates (though their academic performance is inferior to peers), and most go on to successful careers. Since a degree from an elite institution carries much greater promise of a good career than a degree from a run-of-the-mill school, Indian AAHE beneficiaries have surely been better off in the former. It is not clear whether this applies to AAHE beneficiaries in the great majority of mediocre higher educational institutions.

So I reject the claim that AAHE causes a mismatch of UREG students with universities.

(b) dependence and complacency on the part of UREG members

There is no systematic evidence from either the US or India bearing directly on how AAHE policies affect the motivation of UREG youths to develop their human capital.

But theoretical analysis of the payoff to academic effort by UREG students shows that is far more likely that AAHE policies cause potential UREG applicants for admission – as well as enrolled UREG students – to be more highly motivated for academic effort than that they cause students to become complacent. Even as AAHE raises the prospects of UREG applicant admission into any given higher educational institution, it does not reduce the payoff to harder work – because the better UREG applicant performance, the better the schools into which he/she will be admitted.

Complacency is likely to result among UREG applicants only if they are essentially guaranteed admission to the school they wish to attend; and complacency is likely to result among UREG students only if they are virtually guaranteed passing grades, whatever their academic accomplishments. Such guarantees are virtually nonexistent in US and rare in India; for even where there are quotas of reserved seats, there are usually minimum criteria for admission and for passing courses.

So I reject the claim that AAHE fosters dependence and complacency on the part of UREG members.

There is no systematic research in the US or India bearing directly on the effect of AAHE on negative stereotyping.

There is much evidence that AAHE beneficiaries as a group perform less well academically than peers; so the claim that AAHE exacerbates negative stereotyping of UREG members is prima facie plausible.

BUT: To validate the claim one must show that the extent of negative stereotyping is greater than it would have been in the absence of AAHE policies. In that case, UREG students would have constituted smaller minorities in selective institutions, and much smaller minorities in the most elite ones. This no-AAHE situation might well have precipitated as much negative stereotyping as AAHE alternative.

It seems most likely that negative stereotypes of UREG members are fundamentally attributable to persistent inequalities across racial/ethnic groups, such that there is UREG under-representation at the top and over-representation at the bottom of the SES ladder. The best way to end such stereotyping is to reduce the underlying group inequalities, which AAHE arguably helps to do by fostering more integration of societal elites.

Evidence that AAHE policies have led to expanded UREG representation in the societal elite, both in the US and in India, provides indirect support for the claim AAHE has reduced negative UREG stereotyping.

In the absence of clear-cut evidence one way or the other,

I do not accept the claim that AAHE carries costs in the form of aggravation of negative stereotypes about members of racial/ethnic minority groups.

(d) resentment about unfairness of admissions procedures in higher educational institutions

The conviction that AAHE is basically unfair arises from the belief that admission to a higher educational institution should be based on "merit," as measured by conventional quantitative indicators of academic achievement. This belief could be dispelled if (a) AAHE seen as necessary for more accurate assessment of applicant true potential or (b) AAHE is understood to generate significant educational or societal gains.

Much evidence from both the US and India rejects the claim that the addition of AAHE to conventional admissions qualifications results in greater accuracy in appraising applicant potential for academic work. More limited evidence tends to reject the claim that AAHE results in greater accuracy in appraising potential for ultimate career success. If one wants admissions decisions to be based solely on applicants' true potential, it would be better to seek and use direct measures of unrevealed potential.

It follows that AAHE leads to resentment about unfairness…unless the public agrees that AAHE generates significant educational or societal gains, even as it admits applicants with less potential for success.

So I accept the claim that AAHE generates costs due to resentment about unfairness of admissions procedures in higher educational institutions.

BUT: Such costs can be reduced by better understanding of the social and/or educational rationale for AAHE.

(e) exacerbation of racial/ethnic divisions

There is no doubt that AAHE policies focus attention on race/ethnicity/caste and thereby exacerbate consciousness of peoples' group identity, which contributes in some measure to potentially divisive identity group politics.

The comparative history of affirmative action in the US and reservation policies in India shows that competitive demands for group preferences have had a strong tendency to snowball in independent India (as evidenced by the growing demands for and implementation of reservation policies for OBCs), but far less so in the US.

So I accept the claim that there are some costs of AAHE in the likely exacerbation of racial/ethnic divisions, as AAHE spotlights inter-racial/ethnic competition for highly desired access to elite educational institutions.

BUT: Persistent inequalities across racial/ethnic groups, and the related lack of elite integration, also contribute to racial/ethnic divisions. To the extent that AAHE can help to reduce such group inequalities, it can also reduce racial/ethnic divisions (at least in the long run).

(f) devaluation of accomplishments of members of racial/ethnic minority groups

In the context of AAHE – and without knowing precisely who are the AAHE beneficiaries – it is rational to assume that any given UREG student may well have received a boost from AAHE that is to some extent independent of his/her individual achievements and/or potential.

(NB: The grounds for such an assumption would be weakened if AAHE provided a better way of measuring true potential; but empirical evidence from the US and from India does not support this proposition.)

Indeed, if AAHE serves a social/educational goal not directly linked to individual UREG academic performance – e.g. elite integration or improvement of educational quality – then the cost here is unavoidable.

Both in the US and in India, many UREG students are stigmatized by the presumption that their enrollment in a higher educational institution is due to preferential admission policies rather than solely to their own demonstrated capabilities. This is in fact true of AAHE beneficiaries among UREG students, but it is not true of other UREG students, whose true capabilities are therefore especially liable to be under-appreciated. Such under-appreciation is virtually inevitable, so long as non-AAHE beneficiaries among UREG students cannot readily be distinguished from AAHE beneficiaries. In the US it is indeed usually very hard to identify which UREG students are AAHE beneficiaries; in India it is often not so hard to do so.

There are several adverse consequences of stigmatization of UREG students via devaluation of their accomplishments. For one thing, at selective educational institutions in both the US and India, some people (usually non-UREG students) complain that UREG students – presumed to be AAHE beneficiaries – do not belong on campus. At least in the US, it is also charged that faculty disserve UREG students by holding low expectations of them and patronizing them.

BUT there is no evidence to determine the extent to which such views and actions might be attributable to AAHE policies, rather than to a general predilection toward negative stereotyping of UREG students.

So I accept the claim that AAHE generates some costs via the devaluation of accomplishments of members of racial/ethnic minority groups.

6. Summary of conclusions on benefits and costs

AAHE policies in both the US and India have resulted in a complex mixture of benefits and costs. The evidence is not broad and deep enough to permit measurement of the relative magnitudes of benefits and costs. But we can conclude, with some confidence, the following:

The main benefits from AAHE policies in the US and India have resulted from:

the racial/ethnic integration of society's elite, which fosters a more effective democratic system, more effective societal institutions as well as greater equity;

(probably also) the spread of social capital to UREG students, which promotes greater equity and greater efficiency;

(less certainly) greater motivation of UREG students, which promotes greater efficiency and greater equity;

(in the US only) a better quality of education as a consequence of greater student racial/ethnic diversity, which promoted greater efficiency.

The main costs of AAHE policies in US and India have resulted from:

exacerbation of racial/ethnic divisions, which undermines social harmony and democracy;

devaluation of the accomplishments of members of racial/ethnic minority groups, which reduces equity and efficiency;

failure to select the most academically qualified applicants, which reduces efficiency;

resentment about unfairness of admissions procedures, which reduces social harmony.

The main differences between AAHE-induced benefits and costs in the US vs. India are:

the costs arising from the exacerbation of racial/ethnic divisions are more significant in India than in the US;

the costs arising from the degree to which AAHE results in failure to select the most academically qualified applicants are more significant in India than in the US;

the benefits of better quality of education from student ethnic diversity are realized in the US, but not in India.

Thus the overall net benefits of AAHE policies have most probably been greater in the US than in India.

BUT: The uncertainty and complexity of these conclusions point to the need for more empirical research on the consequences of AAHE policies – especially in India, where the evidence particularly scant.

OVERALL CONCLUSION:

The main benefits of AAHE stem from the integration of society's elite.

The main costs of – and need for – AAHE would be mitigated, if not eliminated, by reduction of persistent racial/ethnic group inequalities.

AAHE seeks to achieve such reduction precisely by fostering more integration of society's elite.

B. OPTIMAL CHOICE OF AAHE POLICY

The relevant policy question not just: AAHE – yes or no?

One must also ask: If AAHE – then what kind of AAHE policy?

Given the uncertainty about benefits exceeding costs, careful attention should be devoted to how AAHE structured; the better is the AAHE policy, the more likely it will have a (net) beneficial impact.

1. What is the best sphere and level for positive discrimination policies?

For several reasons the net benefits from such policies are likely to be greater in the sphere of education than in the spheres of employment or contracting:

at an earlier stage in the life cycle, people's potential capabilities are more important than their realized capabilities, so beneficiaries of positive discrimination are better able to catch up with less disadvantaged peers;

there is a greater element of development – as opposed to pure preference – involved;

the costs of poor performance are borne to a greater extent by the beneficiaries, and to a lesser extent by others;

less passion and resentment is likely to be generated from positive discrimination in the educational sphere.

Within the sphere of education, the net benefits from positive discrimination policies are likely to be greater in higher education than in elementary or secondary education:

-- The main benefits of positive discrimination arise from change in the racial/ethnic composition of the societal elite, for which (elite) higher education is preparatory.

-- The development vs. (pure) preference argument would favor positive discrimination at an earlier stage of education; but, as a practical matter, patterns of residential racial/ethnic segregation make it difficult to practice positive discrimination in K-12 schools – in all but elite college-preparatory schools.

2. How to determine which beneficiary groups should be eligible for AAHE?

Since elite integration is the most important objective of AAHE, one should first of all consider the strength of the need for a particular UREG to be integrated into the societal elite:

-- This is related to the degree of under-representation of the UREG in desirable positions, which is linked to the extent of its past mistreatment and current stigmatization by dominant groups, which is in turn closely related to the degree of UREG residential segregation.

To the extent that improving the quality of education via a diverse student body is salient, one should consider also the socioeconomic and cultural distance of the UREG from the dominant groups in society.

Ø If there are distinct subgroups of UREGs, which ones should be eligible for AAHE?

There is a strong case for simplicity in limiting the number of eligible UREGs, confining eligibility to the cases that are clearly most compelling.

Ø Should all AAHE-eligible groups receive the same magnitude of preference?

To avoid divisiveness, it is best to work with single magnitude of preference.

Ø What about a means test for eligible UREG applicants?

This is not good idea, because it would remove from the ranks of AAHE beneficiaries precisely those UREG members likely to be best prepared to perform well – and thus achieve the main objectives of AAHE.

It would be best to address resource-redistributional objectives by policies oriented directly to the disadvantaged, rather than by changing the scope of eligibility for AAHE.

3. If there is to be AAHE, what form should it take?

Ø Quotas vs. preferences; mechanical/formulaic vs. nuanced/holistic procedures

It is best to make selection processes as sensitive and nuanced as possible, to maximize the potential for determining which UREG applicants – underqualified in terms of conventional indicators – have the greatest potential to be successful.

NB: This calls for some investment of resources.

Ø Identifiability of beneficiaries

It is best not to render AAHE beneficiaries easily identifiable as such by others, given the adverse effects of being perceived and stigmatized as a beneficiary.

Ø Support for under-prepared AAHE beneficiaries

Preferences should be accompanied by:

§ outreach and aid to secondary schools with high UREG enrollment;

§ financial aid to beneficiaries;

§ academic support for beneficiaries, to promote high rates of enrollment, persistence to graduation, and subsequent success.

NB: All of the above require some investment of resources, which are not all captured by the investor – hence a need for government subsidies.

NB: A means test would be appropriate and desirable in the case of financial aid for beneficiaries.

4. How to determine the optimal extent of the preference?

There is a trade-off between two types of errors:

-- To achieve greater numbers of beneficiaries requires a higher preference magnitude;

-- To achieve a higher success rate of beneficiaries requires a lower preference magnitude.

There is therefore a need to strike a balance that will maximize net benefits – the benefits from successful beneficiaries vs. the costs from unsuccessful ones.

More research is needed on the relationship between preference size and success.

OVERALL CONCLUSION:

How one does affirmative action is critical to its net benefits.

C. ALTERNATIVES TO (RACE/ETHNICITY-BASED) AAHE

Is there not a better alternative – even to well-structured AAHE – that would achieve its objectives, without taking account of (morally irrelevant) racial/ethnic characteristics?

If so, we should certainly prefer it!

1. Class-based AAHE

This would lead to the admission of more (poorly-off) non-UREG applicants, and fewer (better-off) UREG applicants, resulting in more socioeconomic diversity, less racial/ethnic diversity.

There are several reasons why this would be problematic.

§ The point of AAHE is to reduce inter-racial/ethnic group inequalities – because:

Racial/ethnic group disparities are especially likely to be attributed (with much factual basis) to past or present mistreatment of under-represented groups;

Racial/ethnic groups are most vulnerable to damaging negative stereotyping;

Racial/ethnic groups are most vulnerable to stigmatization;

Racial/ethnic groups are most vulnerable to residential segregation.

§ Class-based preferences are likely to lead to poorer performance by beneficiaries

The admission of more low-SES (non-UREG) students, and fewer higher-SES (UREG) students, would mean admitting more under-prepared students – due to the high correlation of SES and preparedness for higher education.

This would result in lower rates of enrollment, graduation and subsequent success, which would reduce the net benefits of the AAHE policy.

§ Class-based preferences are more costly

Given the effect of low SES on financial capacity to enroll and persist in higher educational institutions, class-based AAHE beneficiaries would need significantly more support than race/ethnicity-based AAHE beneficiaries in order to achieve reasonable rates of success.

The race/ethnicity-disadvantaged lack both respect and resources, while the class-disadvantaged are less lacking in respect and more lacking in resources; thus a given amount of resources will stretch across greater numbers of race/ethnicity-disadvantaged than class-disadvantaged.

§ Classed status can be altered (or falsified) more readily than racial/ethnic identity

Hence there is greater scope for perverse incentives – to alter or falsify status.

Ø STILL – if there is to be class-based AAHE, how should class (SES) be determined?

It is best to use indicators of long-term family SES (e.g., family wealth, grandparents' as well as parents' education) rather than short-term family SES (e.g., parents' occupation and income).

This is because opportunities available to children are highly dependent on their family's accumulated financial, physical, human and social capital, all of which are closely linked to the cumulative effect of past history.

It is also desirable to use indicators of neighborhood SES – as opposed to family status – such as average income of the residential neighborhood or census tract.

NB: this approach would increase the racial/ethnic diversity of class-based AAHE beneficiaries, since UREG members suffer particularly acutely from the legacy of a disadvantaged history.

2. Redistributive social policy programs other than AAHE

Why use preferential admission rather than more straightforward alternative policies to improve conditions for UREG members – such as redistributive monetary transfers or development assistance programs?

One must grant that preferential policies are problematical in that:

they concentrate losses on those applicants displaced by the admission of beneficiaries;

many unadmitted applicants are likely to attribute their rejection to preferential policy – even if that is not in fact true.

But preferential policies have compensating advantages:

they tend to be significantly less expensive, since they do not require substantial resource transfers or investments from policy-making organizations to beneficiaries; instead, a significant share of the costs are deflected away from such organizations.

they do not necessarily compete with resource-redistributional policies as a way of reducing UREG disadvantages. (These two types of policy may compete for political energies, but preferential policies – by enabling more UREGs to rise to occupy decision-making positions – can thereby result in the channeling of more resources to the benefit of UREG members.

3. X% plans

X% plans involve a guarantee of admission to students in the top X% of their high school graduating class, as measured by a quantitative indicator such as GPA. To the extent that high schools are segregated by race and class, an X% plan would increase the number of UREG and low-SES members admitted – as compared with admission of the top X% of all students. Such plans have been widely discussed, and in some states they have actually been implemented as an alternative to AAHE (e.g., 10%, in Texas, 4% in California, 20% in Florida).

But such plans have well-recognized disadvantages vis-à-vis preferential policies:

they don't work for out-of-state students;

they overemphasize narrow quantitative performance measures relative to broader qualification indicators;

they result in the admission of a significantly larger % of under-prepared UREG admits, who will therefore be less likely to succeed in their academic studies;

they generate perverse incentives for high school students to transfer to weaker schools, or take easier course loads.

4. Lotteries

Lottery admission systems select randomly a fraction of all applicants who meet a specified set of minimum qualifications (exclusively or primarily academic in nature).

Such lottery systems would result in higher enrollments of both UREG and low-SES applicants, as compared with a system admitting the applicants most qualified according to the chosen set of qualification indicators, since the proportion of UREG and low-SES applicants at any level of a quantitative academic achievement indicator varies inversely with the level of the indicator.

But such plans have some serious disadvantages vis-à-vis preferential policies:

they place strong emphasis on relatively narrow quantitative performance measures;

one such measure – the SAT score -- can be improved simply by taking test several times, and/or by paying for test-prep workshops, which results in significant advantages for high-SES applicants;

the inherently rather arbitrarily determined minimum qualification levels would become crucially significant;

there would be a great deal of resentment on the part of high-scorers who lose out in the lottery process.

5. Identifying high potential in low-credentialed applicants

This is in principle the best possible alternative to AAHE – but it is very difficult to implement successfully. It can work only to the extent that admissions officers can actually identify characteristics that predict good performance by applicants whose qualifications by conventional indicators are not very impressive. Some characteristics that may possibly point to high potential in low-credentialed applicants are:

the extent to which an applicant has overcome previous disadvantages;

evidence of persistence in achieving a difficult goal;

"relative" test scores – i.e., how high a person scores, controlling for background variables reflecting family and neighborhood SES.

Carefully analysis of past admissions decisions and outcomes may help in identifying applicant characteristics that correlate with subsequent success on the part of low-credentialed applicants. A key question in this regard is: what kinds of successful outcomes should be credited to students? In particular, to what extent should non-academic achievements be counted alongside academic achievements?

6. Conclusion

If AAHE is ruled out – e.g., by judicial decision, by legislation, or by ballot initiative – it will be important to find a good alternative policy for achieving the objectives of AAHE.

The most promising alternative policies are:

preferences for class-disadvantaged applicants, where class disadvantage is measured by such long-term class indicators as low long-term family SES and neighborhood SES;

extensive use in admissions decisions of indicators of high potential in low-credentialed applicants.