Executive Summary
The U.S. Department of Education Office for Civil Rights (OCR) released the findings of its seven-year investigation into undergraduate admissions at U.C. Berkeley on March 15, 1996.*1* The OCR declared, among other things, that UCB "does not maintain illegal quotas for Black, Hispanic and Filipino applicants"*2* and that its admission policy does not unduly discriminate based on race. The OCR report was welcomed by proponents of racial preferences as "a model of prudence"*3* and "an historical vindication of one of the most scrutinized affirmative action programs in the country."*4* Although OCR declared that UCB's use of race and ethnicity falls within the parameters of existing law, California taxpayers should not believe that this means all applicants are subjected to the same standards. From the information provided in the OCR's Letter of Finding (LOF), it is clear that targeted minorities, primarily Hispanics (Chicano/Latino), American Indians and African Americans, were granted huge preferences over other applicants, regardless of these targeted applicants' state residency or financial status.

The matrix of criteria used to grant undergraduate admissions to UCB in 1993, according to the OCR document, shows:

• Wealthy African American, American Indian, and Chicano applicants who were California residents were automatically admitted in three Academic Index Score (AIS) ranges in which other California residents were automatically rejected. (page 8)
• Although UCB is a state supported institution which costs California's general fund more than $8,000 per student, wealthy non-resident minorities were granted huge preferences over middle-class California residents who were not members of the right ethnic group. Wealthy non-resident American Indian, African American or Chicano applicants were automatically accepted in two AIS categories in which other California residents were automatically rejected. (pages 8-9)
• Although much discussion was given to how applicants' files were reviewed to discover qualities such as leadership and musical talent, it is clear from the UCB matrix that very few middle-class California residents who were non-minority were granted the opportunity to have their files read in 1993. Most other California residents only had their files reviewed in one out of eight AIS categories. In contrast, African Americans, American Indians, and Chicanos, regardless of state residency and financial status, were guaranteed either admission or a full file review in all but the lowest academic index categories. (page 11)
• Although UCB claims to grant preference based on economic status, this preference seemed to be minor in comparison to the preference it granted based on race. Poor, non-minority California residents merited only a file review in two AIS categories in which wealthy African Americans, American Indians, and Chicanos were automatically accepted. These poor residents were automatically rejected in two AIS categories in which targeted minorities receive full file reviews. (page 8)

Although the OCR's Letter of Finding is supposedly based on extensive data review, it does not include even the most basic data on which it must have formed its conclusions (applicant pool versus admission data, SAT and GPA data for the various admission categories and ethnic and racial groups). While it is clear from the information disclosed in the Letter of Finding that race and ethnicity play major roles in the admission process at UCB, the exact magnitude of these preferences cannot be evaluated without further data. From the OCR's own document, however, it is clear that it had to be creative to conclude that on the grounds of race, UCB doesn't treat "an individual differently from others in determining whether he satisfies any admission, enrollment, membership or other requirement ..."*5*

The Complaint that Started it All
"There is wide agreement that if whites have higher admission rates and lower test scores than Asians, it is evidence of discrimination against Asians. But if this applies to black students relative to white students, it is called affirmative action, even though it is exactly the same procedure."
Arthur Hu, Original Complaint May 22, 1989*6*

In 1989, Arthur Hu, a software programmer in Seattle and columnist for Asian Week, filed an official complaint with the U.S. Department of Education charging discrimination against Asians in admissions to Harvard University. He also claimed that UCB and UCLA maintained "quotas and differential admissions standards at the University of California at Berkeley and Los Angeles which discriminate not only against Asians, but against all racial groups."*7* This report only deals with his complaint against UCB.

In brief, Hu's complaint accused the UC system, primarily through UCLA and UCB, of maintaining quotas for minority group admissions based on representation among the state's high-school graduates. In the wake of the Assembly Concurrent Resolution (ACR) No. 83 in 1984, which the UC interpreted to mean that it must strive for racial proportionality, the number of minority admits increased dramatically while that of whites fell, especially at competitive schools. Hu based his complaint on this evidence.

Hu wrote:

UC claims that it uses goals, but not quotas to admit minorities according to state proportions. In 1988, UC has actually admitted all minorities in equal to, or even greater than graduate proportions. Since this is caused by affirmative action, and is not based on qualifications, but on numerical goals, this policy is a fourth illegal quota in fact, if not in name. *9* Hu pointed out that by concentrating minority students at UCB and UCLA, the UC system "discriminates against whites and Asians who apply to these campuses."*10* Hu claimed that this "also discriminates against minorities in other campuses such as UC Santa Cruz because it makes them [the campuses] primarily white and Asian."*11* This is a direct consequence of the high acceptance rates at UCB and UCLA. Hu also claimed that the UC was "mismatching abilities by putting the weakest students in the toughest schools," thereby reducing their groups' respective graduation rates by admitting African American and Filipino students with considerably lower grades and SAT scores than other students.

The OCR's Findings

The U.S. Department of Education's Office for Civil Rights is charged with ensuring that colleges and universities that take federal money do not discriminate. It carries out this mission primarily by enforcing Title VI of the 1964 Civil Rights Act. It determined, in this investigation, that the relevant implementing regulations reside at "C.F.R. ¤¤100.3(a), (b)(1)(v), and (b)(6)(ii). The first section, as paraphrased in the LOF, states "that no person in the United States shall on the grounds of race, color, or national origin be excluded from participation in, be denied the benefits of, or be otherwise subjected to discrimination under any program to which this part applies."*12* The second section, as paraphrased in the LOF, states, "a recipient [of federal money] may not directly or through contractual arrangements on the ground of race, color, or national origin treat an individual differently from others in determining whether he satisfies any admission, enrollment, quota eligibility, membership or other requirement or condition which individuals must meet in order to be provided any service, financial aid, or other benefits provided under the program."*13* Complicating, if not contradicting, these first two prohibitions against classification and differential treatment based on race, is the third section stating, "that even in the absence of a prior finding that a recipient has discriminated in the past, a recipient, in administering a program, may take affirmative action to overcome the effects of conditions which resulted in limiting participation by persons of a particular race, color, or national origin."*14*

These were the standards against which the OCR judged the legality of UCB's admission policy. The first two prohibit treating people differently depending on their race and ethnicity while the third states that a university "may take affirmative action." If these sections are to be internally consistent, then racial preferences, such as those used by UCB, cannot be considered "affirmative action." That is, affirmative action programs are proscribed from using racial preferences. They can, however, concentrate on recruitment. The OCR in its pursuit of fairness, didn't interpret these regulations in this manner. For the OCR, even though they didn't explicitly recognize it, there is a contradiction between the first two prohibitions against discriminating based on race and the third which states, in its interpretation, that preferences can be granted on the basis of race. In this light, the OCR sacrifices the first two standards on the altar of the third.

In a letter dated March 1, 1996, nearly seven years after the original complaint was filed, the OCR released it findings. Among other things, the OCR found, "The program employed not quotas or set-asides, and all students, regardless of race or national origin, competed with all other students."*15* Furthermore, any problem associated with not being admitted to UCB was mitigated by the fact that applicants were free to apply again, offered admission at a later date, or, if UC eligible, guaranteed a spot at a less competitive UC school. The LOF found:

Any impact of the affirmative action component on students who were not admitted to UCB was substantially diminished by the fact that all UC-eligible students must be admitted by the System to one of the UC campuses, and many students achieved admission to UCB subsequent to an initial denial of admission.*16*

To arrive at its conclusion of non-discrimination, the OCR's investigation focused on the 1993 admissions process. In 1992, UCB switched to a matrix admission process that was "quite different" from those under the old system. While this paper examines the matrix process, it should be noted that the OCR did not release a Letter of Finding directly related to Hu's charges, which were filed under an older admission process. Similarly misleading, if the OCR investigated and found a discriminatory process, but UCB was willing to alter its policies to mitigate this discrimination, the OCR wouldn't necessarily disclose that it had deemed the former policy discriminatory.*17* As a result, one should be suspicious of the OCR's complete neglect of UCB's admissions policies prior to 1993, which were, after all, the policies that prompted the complaint. As Hu recently wrote in Asian Week, it is as if the police investigated "OJ Simpson by checking on whether he killed anybody in the next year."*18* Even so, there is plenty in the 1993 matrix standards to make one wonder how the OCR concluded that UCB does not "on the ground of race, color, or national origin treat an individual differently from others in determining whether he satisfied any admission, enrollment, ... requirement."*19*

Brief Overview of Admissions Policy

In 1993, UC policy mandated that no less than 40 percent and no more than 60 percent of the first year entering class be selected on academic criteria alone, which include required course-work, GPA, and SAT scores.*20* The remaining students were to be selected on the basis of supplemental factors. According to the OCR, these factors included:

1. special talents, interests or experiences beyond the academic criteria that demonstrate unusual promise for leadership, achievement, and service in a particular field such as civic life or the arts;
   
2. special circumstances adversely affecting applicant's life experiences, including personal difficulties, low family income, refugee status, or veteran status; and
   
3. ethnic identity, gender, and location or residence. *21*

Notice that, as is nearly almost always the case when supplemental, non-academic factors are delineated, race and ethnicity are placed at the end of the list. As we shall soon see, this is a deliberate inversion of the order of importance.

In addition to those who meet the UC Eligibility criteria, the UC maintains the right to admit up to 6 percent of the entering class by "exception." According to the LOF, "This policy is designed for the purpose of admitting students who have not met the eligibility requirements who are either athletes, have special talents, are disadvantaged (i.e. from minority and/or low income or low education backgrounds), have demonstrated exceptional ability in certain areas, or have shown marked improvement in academic performance."*22* Notice that middle class and wealthy minorities can be considered disadvantaged simply on the basis of their ethnicity and race.

Through this admissions process, UCB seeks to ensure "diversity" in its incoming class. In 1988, the University of California's Academic Senate and the Admissions and Enrollment (A&E) Committee adopted the recommendations from a report entitled "Freshman Admissions at Berkeley: A Policy for the 1990s and Beyond." It is also referred to as "the Karabel Report," after UCB Sociologist Jerome Karabel, its principal drafter. The Karabel Report recommended flexible targets for admissions based on the following characteristics (notice the order): "low socio-economic status (SES), athletic ability, age, disability, re-entry students, students from rural and non-traditional high schools, and ethnicity and under-represented minority status."*23* These targets were soon adopted.

Here is how these goals manifested themselves in the admission process. Below, is the 1993 admissions matrix for UCB.

SDS A: 1) California residents who are American Indian, African American or Chicano AND low socio-economic status (SES)*24* or disabled; 2) California residents who are low SES and disabled; 3) California residents who are re-entry applicants (over 24 years of age).
SDS B: 1) California residents who are American Indian, African American or Chicano; 2) California residents who are Latino and SES or disabled.
SDS C: 1) California residents who are Latino; 2) Non-resident American Indian, African American or Chicano applicants.*25*
SDS D: Very low SES.*26 *
SDS E: Rural and Other High School
SDS F: Non-resident Latinos;*27* 2) Other Low SES.
SDS G: California Residents
SDS H: Domestic Non-residents.
SDS I: Foreign Applicants

Under this matrix, an applicant is placed in a box depending on two variables: 1) an Academic Index Score (AIS) and 2) his or her social diversity category. The AIS is determined by multiplying an applicant's GPA, capped at 4.0, by 1,000, and adding to this the combined total of the SAT scores and three required Achievement Test scores.*28* The socio-economic categories, which are on the vertical axis, are determined by an applicant's birth, parental income and place of residence. The LOF states that the AIS scores for the first box are 7160 to 8000 ( 8000 is a perfect score), but didn't disclose the other cut-off scores. The Social Diversity Scores (SDS), which are under the matrix, are quoted directly from the LOF:*29 *

In 1993, applicants were admitted in the following manner. All those in AIS 1, regardless of SDS, were admitted. This accounted for roughly 50 percent of the acceptances. This groups satisfies the requirement that 40 percent to 60 percent of the students be admitted on academic qualifications alone.

A second group comprising SDS A through SDS F and AIS cells 2 through 5 were also automatically admitted. Those admitted in this group, in which race and ethnicity play a major role, accounted for 17 percent of the total acceptances. Explaining these acceptances, for which race was an important factor, the "OCR determined that while race/ethnicity was one factor in defining this group, placement in this group was not based exclusively on race or ethnicity."*30 *

A third group in cell SDS G and AIS 2 was also automatically accepted, accounting for 9 percent of total acceptances. A further 21 percent of the acceptances came from cells SDS A through SDS F and AIS 3 through AIS 7. These acceptances were granted after each file was reviewed and the essays read by two review teams. The Karabel diversity goals were taken into consideration in accepting this group.*31* In addition, three percent of total acceptances were granted on the basis of "special admission." The LOF didn't disclose the racial composition of this or any other admission group.

There are many things about this process indicating that while race is a factor among several, it is a significant factor. While all applicants are put through the same process, they are not all judged by the same standards. First, note that a California taxpayer who is white or Asian is in the seventh possible lowest SDS ranking, above only out-of-state white, Asian, and perhaps only non-minority foreign applicants.*32* In fact, any African American, Chicano, or Latino from anywhere in the United States will be automatically admitted with lower academic qualifications. For example, non-resident American Indians, African Americans, or Chicano applicants, regardless of economic status were automatically admitted in two AIS categories (C2 & C4). In stark contrast, non-poor whites were automatically rejected without even having their files reviewed. Wealthy out-of-state minorities were also automatically admitted in one AIS category (C4) in which "Very Low SES in-state whites and Asians" merely get their application read. It is surely strange for a state university which draws more than $8,000 from the General Fund per student*33* to grant a privilege to wealthy out-of-state applicants over poor and middle class California taxpayers solely on the basis of race.

Take a close look at SDS A and SDS B. Why are they separate categories? Automatic admissions for both categories extends to AIS 5 and applicants in both categories have their applications closely examined in as low an AIS category as 7. There is no difference between these two categories except, perhaps, their public relations utility. If these two categories were properly combined (as in practice they are simply one category for the purpose of the matrix), it would be obvious that wealthy California residents who are American Indians, African Americans, and Chicanos are automatically granted admission, without even a review of their file, at radically lower academic standards than whites and Asians. The proper SDS A category would then read, "1 ) California residents who are American Indian, African American, or Chicano; 2) California residents who are Latino and SES or disabled; 3) California residents who are low SES and disabled; and California residents who are re-entry applicants (over 24 years of age.) SDS B would disappear and each other category would simply move up the matrix while retaining the same structure. This would dispel the carefully crafted notion that in order to be considered in highest disadvantaged status, American Indians, African Americans, and Chicanos must not only be California residents, but must also be poor or disabled. "

Two other points. First, based on UCB's own admissions matrix, a computer automatically admitted nearly three out of four applicants in 1993. For all the talk of personal reviews of files, only one in four applicants was admitted after having his file reviewed. Unless an applicant is considered disadvantaged, the chances of an applicant even having their file read appears remote. For a white or Asian student who excels in the supposed subjective criteria -- "special talents, interests or experiences beyond the academic criteria that demonstrate unusual promise for leadership, achievement, and service in a particular field such as civic life or the arts" -- it is not clear how the University avails itself of this information. Non-poor whites and Asians are automatically rejected if not in AIS 1 or 2. Thus, their chances of being accepted based on non-academic qualifications appear remote.

This failure to review the files of non-minority applicants helps explain the recently exposed "back-door admissions policy" for the wealthy and well connected. It appears that a non-minority applicant who was a class president, volunteered at a homeless shelter, spent her spring break in Mexico building houses for poor peasants, and was fluent in three languages but was just shy of the GPA and SAT score needed for the first two AIS categories, would be automatically rejected without a file review. It is reasonable for parents to wonder why some of their children's peers with both lower academic credentials and fewer extra-curricular activities were granted admission and to inquire through whatever channels available. In many cases, such a review would not be a further review, but rather the first review. Each minority applicant is entitled to a review simply on the basis of his or her race. While this explains the backdoor policy, it does not condone it. After all, such a policy does little to help the majority of taxpaying Californians who are bereft of connections but, support UCB with their taxes nevertheless.

Second, without hard data on the academic parameters for the various categories, (automatic admission on AIS alone, automatic admission on AIS and SDS, admission based on extensive review of file, and special admissions), the exact magnitude of the differing standards cannot be evaluated. Similarly, although the two-step intensive review process as described in the LOF seems race and ethnic neutral, there are points in this selection process in which race bias could play a significant role. Until hard data on the composition of the pool and the acceptances from this pool are provided from the OCR, the public should not conclude that race and ethnicity do not play a major role in an applicant's chances for admission in this category. As always, the devil is in the details. In this case, the details are the data, which have yet to be released by the OCR.

OCR's Conclusion

After review of the process and still undisclosed data, the OCR concluded that UCB does not discriminate based on race.*34* It stopped short, however, of saying that all applicants were judged by the same standards regardless of race, which would, of course, be untrue. This is the central issue in which most Californians are interested.

The OCR concluded, "the process ensured that no applicant was considered or excluded from consideration for selection solely on the basis of race or national origin."*35* This simply means that UCB put all of the applicants on the same grid before it gave enormous preferences depending on racial and ethnic classification. The OCR dismissed Hu's complaint that by concentrating minorities at two schools it was denying other minorities of diversity at other schools by pointing out that it is the individual schools and applicants who make the decisions. Hu's complaint of "academic mismatching" causing low graduation rates for targeted minorities admitted with lower standards was dismissed by the OCR, which stated, "UCB provided additional data which showed the current six-year graduation rates for African American students at UCB is 62 percent compared to a national average [for African American students] of 37 percent."*36* This, of course, neglected the fact that the proper basis for comparison is not against an amorphous national average, over which the University has no control, but against the ethnic groups it holds to higher admission standards. The six year graduation rates for the entering class of 1988 showed very different rates depending on race, with Whites and Asians graduating at rates over 80 percent and Chicano/Latinos, American Indians, and African Americans, who we have seen are admitted by lower standards, all in the neighborhood of 60 percent.*37* UCB's six-year graduation rate, independent of race, for all regularly admitted students in Fall 1987 was 81 percent. For those admitted by exception, the graduation rate was 48 percent.*38* In fact, I suspect that differing graduation rates for different groups do not reflect racism or differences in academic abilities inherent in race, but simply the different standards at which students are admitted. This hypothesis, which asserts that if white and Asian students were admitted to UCB with the same academic qualifications as targeted minorities, they would fail to graduate at similar rates, can only be tested with further data. At any rate, the OCR didn't investigate this issue in its LOF.

Finally the OCR determined that UCB's racially determined admission matrix "did not impose an undue burden on non-beneficiaries." It came to this conclusion because "by state law, all UC-eligible students must receive admission to a UC campus." It continued, "Thus, no student would experience exclusion from the opportunity to receive a University of California education."*39* This "UC Riverside is just as good as UCB" logic should apply to targeted minorities as well. As it stands, though, UC eligible targeted minorities seem to have a presumptive right to attend the system's most prestigious schools to ensure that these schools exhibit the right color scheme. At the same time, it is acceptable for the system's other schools to be filled with Asians and whites. It is never explained why UC Santa Cruz equals UCB for whites and Asians, but not for preferred minorities. Or why, for example, deferred admission to UCB is considered the equivalent of a direct admission for California's taxpayers who are white and Asian, when this same standard apparently isn't applied to targeted minorities, some of whom are no doubt wealthy and from out-of-state.

Conclusion

The OCR's finding of non-discrimination was welcomed by proponents of racial preferences in university admissions as conclusive proof that UCB's admission process does not discriminate based on race. But in reality, the OCR's investigation showed neither that the process was race neutral, nor that race did not play a major role in an applicant's chances for admission. It merely meant that in the interpretation of the OCR, race was only one factor in this process. Just how large a factor race is, as determined by the objective results the process produces, cannot be determined until the OCR discloses the data it used to form its conclusion. Until that time, however, California residents can be assured that race-based admissions are alive and well at UCB.

*1* United States Department of Education, Office for Civil Rights, Letter of Finding (LOF) March 1, 1996, to Dr. Chang-Lin Tien, Chancellor, University of California, Berkeley, from John E. Palomino, Regional Civil Rights Director. Those interested in examining a copy of the LOF, which is available under the Freedom of Information Act, need only write John E. Palomino, Regional Civil Rights Director, United States Department of Education, Office for Civil Rights, 50 United Nations Plaza, Room 239, San Francisco, CA 94102. Refer to Docket Number 09-89-2099. Requests can also be faxed to (415)-437-7783, with a hard copy following in the mail.
*2* LOF p.1
*3* Jerome Karabel, "Perspective on Affirmative Action," Los Angeles Times, April 3, 1996, B-9
*4* Jerome Karabel quoted in Elaine Woo, "Probe Finds No Bias in Berkeley Admissions," Los Angeles Times, March 21, 1996, A-1
*5* Qtd. from LOF p. 2.
*6* Original Complaint by Arthur Hu dated May 22, 1989, p. 4. It was addressed to Gary Curran at the United States Department of Education. It can be found in cyberspace at http://www.halcyon.com/arthurhu/
*7* Hu complaint, p. 1.
*8* Michael Lynch "UC Dilemma: Manufacturing Equal Outcomes in a World of Unequal Inputs" May 1996, Pacific Research Institute Ideas in Action Fact Sheet.
*9* Hu complaint, pp.1-2.
*10* Hu complaint, p. 2.
*11* Hu complaint, p. 2.
*12* Qtd. from LOF p. 2.
*13* Qtd. from LOF p. 2.
*14* Qtd. from LOF p. 2.
*15* LOF p. 2.
*16* LOF p. 2.
*17* Personal conversation, January 23, 1996, with OCR staff.
*18* Arthur Hu, "Civil Rights: I See Nothing," Asian Week, April 4, 1996.
*19* LOF p. 2.
*20* These proportions have since been increased to no less than 50 percent and no more than 75 percent by the regents' 1995 resolutions.
*21* LOF p. 4.
*22* LOF p. 4.
*23* LOF p. 5.
*24* SES is defined as "1) parents, if employed, employed in clerical, operative, service, homemaker, or laborer positions or neither parent having education beyond high school graduation, and 2) an annual family income below $50,000." (letter from Bob Laird, Director, Office for Undergraduate Admissions and Relations with Schools, April 12, 1996.)
*25* It's not clear whether non-resident included foreign or not. I have requested a clarification on this point from the OCR. Notice that this may be the case as SDS H makes the distinction, which is lacking here, of Domestic Non-Resident.
*26* Very Low SES is defined as "1) neither parent holding a four-year college degree, 2) neither parent employed in a professional or managerial occupation, and 3) an annual family income below $39,635." (letter from Bob Laird, Director, Office for Undergraduate Admissions and Relations with Schools, April 12, 1996.)
*27* See footnote 25.
*28* LOF p. 7.
*29* LOF p. 8.
*30* LOF pp. 8-9.
*31* For a full discussion see LOF pp. 9-10.
*32* See footnote 25.
*33* Governor Wilson Administration "University of California (UC) Admissions Policy Facts."
*34* LOF p. 1.
*35* LOF p. 13.
*36* LOF p. 13.
*37* Ben Wildavsky, "UC Campus Debates Affirmative Action," San Francisco Chronicle, May 16, 1995, A1.
*38* "Information Digest: A Reference Guide for Student Affirmative Action Efforts at the University of California," Fall 1995, p. 38.
*39* LOF p. 14.

ABOUT PACIFIC RESEARCH INSTITUTE

The Pacific Research Institute is supported entirely by the private sector through voluntary contributions from foundations, corporations and individuals, and from the sale of its books. The Institute is a tax-exempt, 501(c)(3) organization; all contributions are tax deductible. The Institute neither solicits nor accepts government funding.

Nothing contained in this briefing is to be construed as necessarily reflecting the views of the Pacific Research Institute for Public Policy or as an attempt to thwart or aid the passage of any legislation.