A Second Century of Dishonor : Federal Inequities and California Tribes, ch.III

III. Service Population: Undercounting and Inequity in California

When the Bureau of Indian Affairs and other federal agencies allocate scarce program dollars among different tribes and geographic areas, they often base these allocations on numbers of individuals eligible to be served by these programs. Particularly where programs are designed to serve individual needs -- as welfare, education, housing, health care, and many other Indian programs do -- the number of persons eligible for service is an appropriate criterion for distributing funds. 1 Not surprisingly, then, studies of funding equity for California tribes typically compare federal dollars allocated per person for different area offices of the Bureau of Indian Affairs or for comparable divisions of other federal agencies. 2 Significantly, these comparisons rarely question the population figures put forward for each Area Office or other division by the relevant agency. Thus, for example, the 1976 study by the California Department of Housing and Community Development, the 1977 Report to the Commissioner of the BIA Regarding Funding of Bureau Programs in the Sacramento Area, and the 1989 BIA Resource Allocation Effectiveness Study all take as a given the Bureau's own service population figures in determining whether California Indians were unfairly denied federal program support. 3 The analyses that we provide later in this report, based on federal data, follow the same methodology.

Using the official service population figures, all of these studies conclude that California Indians have been denied their fair share of federal program dollars. Our own analyses reach the same conclusion. Such reports will understate the true extent of the inequity, however, if the Bureau's population figures systematically undercount California Indians. Hence, it is critical to determine whether California Indians are counted in an appropriate way by the Bureau. If either the eligibility criteria themselves are biased against California Indians, or these criteria are applied in a way that leaves California Indians undercounted, the equity problem is even greater than the previously published studies and the analyses in this report suggest.

In fact, the eligibility criteria and population figures used by the Bureau of Indian Affairs for California are difficult to defend, especially insofar as the Bureau places geographic limits on eligibility. Even the limitation of benefits to members of recognized tribes is suspect in the California context, where individuals who can prove that the federal government views them as Indians nonetheless have been denied recognition. We recommend that both of these limits on the service population count be relaxed. To understand how these Indians came to be excluded and why the practice should be changed, it is necessary to review the history of Bureau eligibility policy and its application in California.

In the earliest years of federal Indian policy in California, there were no treaties and no reservations. Every member of an indigenous California group was treated as a federal responsibility, regardless of place of residence or formal federal recognition. During the first seventy years after California statehood, as small areas of land were set aside for tribes throughout the state, Congress periodically appropriated funds for the benefit of California Indians, again without specifying eligibility criteria related to residence or recognition. 4

In 1921, Congress regularized appropriations for Indian benefits nationwide by adopting the Snyder Act, which gave general authorization for such expenditures. The Snyder Act likewise had no geographic restrictions on eligibility, stating that it was for "the benefit, care, and assistance of the Indians throughout the United States." Since most Indian programs funded through the Bureau rest on the authority of the Snyder Act, debate has ensued over whether the Bureau may introduce geographic restrictions, such as requirements that beneficiaries live on reservations or "near" reservations. There have also been challenges, focused largely in California, on the limitation of benefits to members of recognized tribes.

Geographic Restrictions.

Although Bureau practices have not always confined benefits authorized by the Snyder Act to reservation Indians 5, Bureau pronouncements have sometimes articulated an "on reservation" requirement. For example, in 1970, Assistant Secretary of the Interior Harrison Loesch wrote to the Commissioner of the Bureau of Indian Affairs, "It is a long-standing general policy of the Bureau of Indian Affairs and the Congress that the Bureau's special Federal services are to be provided only to the reservation Indians." 6 Where such limitations found their way into BIA manuals, tribal members who were living near reservations and functioning as part of tribal communities brought lawsuits challenging denials of benefits. In one of these cases, Morton v. Ruiz 7, the Supreme Court declared that the federal trust responsibility to Indian tribes required clear articulation and justification for any policy that excluded groups of Indians from benefits under the Snyder Act.

The upshot of Morton v. Ruiz was that the Bureau formalized and expanded its eligibility and priority criteria to include Indians living "near" reservations as well as those living within reservation boundaries. 8 In addition, the Bureau established procedures for officially designating particular areas as "near" reservations, including application of specified criteria, 9 and requirements of consultation with the relevant tribe and publication in the Federal Register. 10 Consequently, when the Bureau does its biannual determinations of service population and labor force, it directs tribes to identify the "Total Resident Indian Population" by adding together the Indians "within the reservation" and those "adjacent to the reservation." 11

Oklahoma Indians and (more recently) Alaska Natives have received special treatment under these Bureau policies regarding geographic eligibility. In each case, the special treatment responds to a distinctive history of legal actions affecting the existence of reservations in that state. For example, the Bureau has taken account of the fact that Oklahoma reservations (but not tribes) were abolished near the turn of this century, by defining "reservations" to include "former reservations in Oklahoma." 12 Furthermore, the Bureau has accommodated the fact that Congress in 1974 established a regime of regions and Native villages in Alaska in lieu of reservations, by defining the term "reservations" to include "Alaska Native regions established pursuant to the Alaska Native Claims Settlement Act." 13 As a consequence of such special treatment, there is a close approximation between the census statistics for Indian population in those states and the Bureau's figures for service population. 14

In California, by contrast, the BIA service population statistics and the census figures for all Indians in the state diverge widely. For 1990, the census shows 236,078 Indians living in California. For 1989 (service population is compiled in odd years only), the Bureau's service population is only 12 percent of that figure, or 28,815. This discrepancy has existed for many decades. Thus, while the census counted 91,018 Indians in California in 1970, approximately 45,000 of them native California Indians, the Bureau's 1971 service population was only 6,100, or 7 percent of the census figure. Why have so many Indians in California been excluded from the Bureau's service population?

One hypothesis is that the Bureau's counting methods are seriously biased toward members of local tribes, leading to understatement of the total number of Indians. The standard protocol is for the Bureau to circulate forms to individual tribes, asking tribal leaders to identify the resident Indian population on and adjacent to the reservation. Tribal leaders may indeed find it easier to identify tribal members than non-member Indians. While these non-member Indians may be eligible for federal services, they may simply not be counted for service population purposes because the tribal counters don't know who they are. The difficulty with this hypothesis is that a count of California Indians from the 1990 census still produces a figure that is more than double the 1989 Bureau service population count, or 59,011. 15 Thus, even counting only California tribal members, the service population numbers should be considerably higher than they are. One significant reason why the Bureau's service population figures do not include even all tribal members is that tribes lack staff and resources to undertake their population surveys. If resources were provided and tribes were given incentives to count all Indians, the numbers would doubtless rise even higher.

Another hypothesis that might explain the disparity between service population and census figures in California is that large numbers of California's Indians do not live "on or near" a reservation. 16 In fact, the census allows us to determine the number of Indians who live in rural parts of counties in which reservations are located, an approximation of the "on or near" reservation category. For 1990, that figure is 70,860, or approximately two and one half times the 28,815 service population for 1989. Thus, even accepting the Bureau's definition of service population areas, there is a serious undercount in California. This undercount means, in turn, that California Indians are denied their appropriate share of federal benefits, at least when such benefits are allocated on the basis of population numbers. 17

Even a figure that is two and one-half times the service population count in California would not necessarily provide the most appropriate tally of Indians for purposes of assessing funding equity, however. There is good reason to question whether the "on or near" limitation ought to be applied at all to members of indigenous California groups. And if that limitation were relaxed, the number of Indians moves closer to the 236,078 figure for statewide Indians in the census or, if one excluded the 54,440 Indians in Los Angeles, San Francisco, and Alameda (Oakland) counties, 181,638. Excluding these large cities makes some sense if the focus is on indigenous California Indian groups, because masses of Indians from tribes outside of California were relocated to the Los Angeles and San Francisco Bay areas during the 1950s. It is also true, however, that members of indigenous California groups live in these cities as well. 18

Should the "on or near" limitation be applied to California Indians? As discussed above, such a criterion was not in force in the early years of federal responsibility for California Indians. Indeed, in 1970, a Bureau official filed a sworn affidavit in the course of litigation, in which he stated, "There has been no instance where a California Indian, otherwise eligible, was denied available federal boarding school or scholarship assistance by the Sacramento Area Office for failure to meet the 'on or near' criteria in the regulations." 19 In the post-Snyder Act period, however, the Bureau has in fact invoked the "on or near" limitation to restrict service population counts and eligibility for California Indians. As a consequence, many federal officials and Indian advocates have argued that special treatment is warranted for California Indians, just as it has been instituted for Oklahoma Indians and Alaska Natives. This special treatment could be provided either by an expansive interpretation of the terms "on or near" (for example, to include all counties in the state where trust lands are located, or even to include the entire state), or by dropping the requirement of "on or near" altogether.

The case for such special treatment rests on the distinctive and tragic history of loss of California Indian lands. California Indians left their traditional homelands between 1851 and 1853 and moved to eighteen large areas, comprising over 8 million acres, promised as reservations in treaties that tribal representatives and federal agents had signed. The Senate refused to ratify the treaties, however; and non-Indians soon laid claim both to the traditional Indian homelands and to the lands promised the Indians in the unratified treaties. Courts then failed to protect the Indians who sought to remain on or return to their historic lands. Thus the tribes were left homeless, dispersed, and starving. Between 1870 and 1930 the federal government created approximately one hundred small reservations and established numerous public domain allotments as homelands for the native California population. But these areas were too small, too remote, too arid, and too unsuitable for occupation or habitation to accommodate the entire native population. As Senator Henry Jackson stated in 1971,

Federal officials have advanced the view that the situation in California thus closely parallels the situation in Oklahoma, where "former reservations" are counted as territory "on or near" reservations. Maps prepared in 1971 by an Assistant to the Commissioner of Indian Affairs, depicting Indian reservations in California, territories promised in the unratified treaties, and areas of Indian community concentration, demonstrated three things: (1) that "the extent of Indian trust land and public domain allotments is very limited, emphasizing the notorious landless status of California Indians"; (2) the areas of Indian community concentration "are generally near to and in many instances overlap" the territories promised in the unratified treaties, "a situation comparable to that found in Oklahoma [with the "former reservations"]"; and (3) in addition to those Indians in areas of community concentration, "there are significant numbers of California Indians now residing throughout the State, also comparable to the Oklahoma situation." 21 The Assistant concluded, "Just as in Oklahoma...communities of descendants of the original Indians continue to live within or near the reservation boundaries originally delimited but, in California, never established by treaty."

If use of the "on or near" limitation is unfair in California, the question remains how best to count California Indians. Several distinct but conceptually related remedies have been proposed by federal officials and Indian advocates.

All of these proposals would move the Bureau in the direction of counting, for budget allocation purposes, a broader set of Indians than those satisfying the technical requirements for "on or near" a reservation. Viewed as a group, these proposals typically sweep into the service population net either all Indians in rural parts of the state, or all native California Indians regardless of location within the state. Remarkably, several federal agencies besides the BIA as well as the Congress have proceeded to institute such broad geographic criteria for California Indians, while the BIA has done nothing. Some illustrations follow:

It is apparent from this history and from the practice in other federal agencies that the BIA has been too slow to recognize and rectify the problems of program eligibility and service population undercount in California. The most modest adjustment -- including all Indians identified by the 1990 census as residing in rural parts of counties in which reservations are located -- would result in more than doubling the 1989 service population. Some adjustments have been made in the past six years -- the 1993 BIA service population figure for California is 45,568 -- but the current figure would still have to be increased by 55 percent to reach the 1990 figure for rural California Indians in counties with reservations. Were all such Indians to be counted, California Indians would constitute at least 6 percent of the BIA's national service population, rather than 3.8 percent using the Bureau's 1993 service population figures. 33 Adopting the most expansive definition used in the Indian Education Act -- all Indians in the state -- would result in multiplying the service population by many more times, and making California Indians 20 percent of the Bureau's total service population. Several intermediate possibilities exist as well. For example, matching the 1991 IHS service population would almost triple the BIA's numbers for that same year. Comparing the IHS and Bureau figures for 1994 and 1993, respectively, would also lead to multiplying the BIA figures by almost three times. Alternatively, if the BIA were to include all Indians in California from the 1990 census outside the Los Angeles and San Francisco Bay areas, it would still need to multiply the 1989 service population count by almost six times, or the 1993 figure by nearly four times. Whichever measure is used, the Bureau will have to revise its formulas for allocating funds to the state and to clarify or restate its program eligibility criteria. The consequence of such readjustments will be more equitable treatment for California Indians.
                                                              #                  %*
1989 BIA Service Population for CA 28,815 3.0
1993 BIA Service Population for CA 45,568 3.8
1990 Census Indians in CA from CA tribes 59,011 6.0
1990 Census Indians in rural parts of CA counties 70,860 7.2
1991 IHS Service Population for CA 88,675 8.3
1994 IHS Service Population for CA 113,465 9.0
1990 Census Indians in CA outside LA & SF Bay Area 181,638 16.5
1990 Census Indians in CA 236,078 20.4

*% refers to percentage of total BIA service population for the closest approximate year, adjusted to reflect the additional numbers from California

Tribal Recognition as a Limit. Wholly apart from geographic limitations, BIA service population is determined by who is counted as "Indian." As a general rule, the BIA has used membership in a federally recognized tribe as the criterion for determining who is "Indian." 34 Occasionally, the Bureau has used "blood quantum" as an alternative or additional criterion, with the usual requirement being one-quarter degree "Indian blood." 35 Over the past two decades, the courts and Congress have begun to reject the "blood quantum" standards in favor of sole reliance on membership in recognized tribes. 36 This development disproportionately disadvantages California, which has the largest percentage of Indians from unacknowledged tribes in the nation. 37

What makes the situation of these unacknowledged California Indians particularly compelling is that most of them are on, or descended from individuals who were on, the judgment rolls for land claims cases involving the 1852 unratified treaties with California tribes. Indeed, some of these Indians also have public domain allotments or other forms of trust allotments, which strongly suggests a federal responsibility. Thus, although these individuals have one or more forms of federal certification of their status as Indians, they are not eligible for BIA benefits under contemporary agency standards because their tribal groups are not acknowledged. Accordingly, they are not counted for purposes of BIA service population statistics. The history of federal treatment of California Indians, described above, suggests that this is not a just result.

Judicial and Congressional authority supports the view that the BIA should move toward including certain California Indians from unacknowledged tribes. In Malone v. Bureau of Indian Affairs, 38 decided in 1994, the United States Court of Appeals for the Ninth Circuit held that the BIA had not properly promulgated regulations for higher education grants and loans that limited eligibility to members of federally recognized tribes. A member of an unacknowledged California tribe who had been denied a grant asked the court to enjoin the BIA from conditioning eligibility for grants on membership in a federally recognized tribe. Although the court refused to issue such an injunction, claiming that it lacked authority to promulgate eligibility criteria, it did offer some guidance to the Bureau as it attempted to establish valid regulations. In particular, the court admonished the agency to "adopt criteria consistent with the broad language of the Snyder Act, and we encourage the BIA to look to eligibility criteria used in other Snyder Act programs, such as those set forth in the 1988 Amendments to the IHCIA [Indian Health Care Improvement Act]...." As noted above, those amendments include among eligible recipients California Indians from unacknowledged tribes. In particular, the provision includes,
As the Ninth Circuit suggested, a similar definition should be adopted by the BIA for eligibility for its services. Such an expanded definition of eligibility would justify a larger service population count, such as one that followed census figures, described above, or the IHS service population figure. Census figures are useful for this purpose because the individuals counted by the census as Indians are individuals who self-identify. That group would include individuals who believe themselves to be Indian but who may not be members of federally acknowledged tribes. The IHS figure is also relevant because it follows from the broad eligibility criteria approved by the Malone court.


In establishing what is equitable funding for California tribes, service population figures are a crucial variable. Undercounting eligible Indians leads to an understatement of potential equity problems. The way the BIA determines eligibility in California has been rejected by other federal agencies and in some instances by the federal courts. Census data as well as data from other parts of the Executive Branch suggest there is an undercount of at least 55 percent. Larger estimates that are still conservatively based on Indian Health Service numbers suggest that the Bureau's service population count for 1993 should be nearly tripled.

  1. Funds for some federal Indian programs, such as programs for forestry management and land conservation, should not be distributed on the basis of population. Rather, distribution should depend on the presence or absence of resources as to which the federal government has a trust obligation. See infra.
  2. Since California comprises an entire area office of the BIA, it is possible to juxtapose the federal dollars that reach California Indians with the federal dollars that reach Indians in other parts of the country.
  3. For references to and discussion of these reports, see Section II of this report, supra.
  4. Report of the Commissioner of Indian Affairs for 1866 at 94.
  5. 1973 Interior Department Task Force, Ernest Stevens and John Jollie co-chairs at 12-14.
  6. Loesch ascribed this policy to treaty provisions and to the inability of states to raise funds for benefit programs by taxing reservation lands. In fact, few Indian treaties prescribe the range of benefits provided under the Snyder Act. Furthermore, many Indians who live on reservations, such as members of California tribes, have no ratified treaties with the federal government.
  7. 415 U.S. 199 (1974).
  8. See, e.g., 25 C.F.R. secs. 20.20(a)(3) (social services programs); 26.5 (employment assistance services); 31.1 (enrollment in BIA boarding schools). The Bureau's choice of "on or near" the reservation echoes legislation that Congress passed in 1956 regarding vocational training programs for Indians. The purpose of this statute, 25 U.S.C. sec. 309, is "to help adult Indians who reside on or near reservations to obtain reasonable and satisfactory employment...." See 25 C.F.R. sec. 27.5 for the implementing eligibility criterion.
  9. The criteria set forth are are follows:
  10. See 20 C.F.R. sec. 20.1(r).
  11. Form 5-2119, Department of Interior, Bureau of Indian Affairs, December 1993.
  12. See, e.g., 25 C.F.R. sec. 20.1(v). Many of the tribal members in Oklahoma had been given allotments of land, carved out of the reservation land base and placed in trust for a period of years, late in the nineteenth century. Although large numbers of these allotments found their way into non-Indian ownership through fraud and tax sales, some continued in trust status even after the reservations were abolished.
  13. Id.
  14. The 1989 BIA service population for Oklahoma was 231,952, and the 1990 Census population for Oklahoma was 252,089.
  15. This number was arrived at by compiling the number of Indians who identified themselves as belonging to an indigenous California tribe (49,254), and then adding to that number an appropriate fraction of all Indians from California who failed to report or specify their tribal affiliation (a total of 39,035). The fraction used was 25 percent, because 25 percent of all Indians in California who did identify their tribal affiliation indicated an indigenous California tribe. Adding 25 percent of 39,035 (9,758) to 49,254 yielded the figure of 59,011.
  16. Since the Bureau began formally designating areas as "near" a reservation, no areas in California have secured that status. In counting service population, however, the Bureau does not seem to differentiate between "adjacent" areas that have been formally designated and those that have not.
  17. To compile the service population data in California, the Bureau relies on forms filled out by the tribes themselves. Many of these tribes lack the staffing and expertise to conduct a thorough survey.
  18. The Gabrielino/Tongva tribe, for example, is an unrecognized group that is indigenous to the Los Angeles basin.
  19. This affidavit was introduced into the Congressional Record by Senator Alan Cranston in 1971. Congressional Record S21326 (December 11, 1971).
  20. Congressional Record S21327 (December 11, 1971).
  21. Memorandum from Roderick H. Riley, Assistant to the Commissioner, to the Commissioner of Indian Affairs, May 13, 1971.
  22. Memorandum from BIA Sacramento Area Director William Finale to the Commissioner of Indian Affairs, August 14, 1970.
  23. Congressional Record, S8591, May 31, 1972.
  24. 1973 Interior Department Task Force, Ernest Stevens and John Jollie co-chairs at 33.
  25. Memorandum from the Assistant Secretary, Indian Affairs to the Sacramento Area Director, January 31, 1984.
  26. 25 U.S.C. sec. 1679.
  27. 42 C.F.R. sec. 36.15(c).
  28. Letter from Michael H. Trujillo, Director IHS to Carole Goldberg-Ambrose, November 16, 1995. Trujillo goes on to explain that service populations between census years are estimated by a smoothing or averaging technique in order to show a gradual transition between the two periods.
  29. See IHS Circular 92-5.
  30. 56 Federal Register 26194 (June 6, 1991), codified at 45 C.F.R. sec. 98.80(e).
  31. 20 U.S.C. sec. 366, implemented in 34 C.F.R. part 771.1.
  32. See 100 P.L. 297; 102 Stat. 130, sec. 5312(2)(B) (1988).
  33. The words "at least" are used because the 70,860 figure is from the 1990 census, and there may well have been some population growth or decline between 1990 and 1993.
  34. See, e.g., Indian Financing Act of 1974, 25 U.S.C. sec. 1452(b).
  35. See, e.g., 25 C.F.R. sec. 40.1 (regarding eligibility for higher education grants and loans).
  36. See, e.g., Zarr v. Barlow, 800 F.2d 1484 (9th Cir. 1986) (striking down the one-quarter degree blood quantum requirement for higher education grants and loans as applied to members of federally recognized tribes).
  37. See Section XIV of this report, infra.
  38. 38 F.3d 433 (9th Cir. 1994).
  39. 25 U.S.C. sec. 1679(b).

A Second Century of Dishonor :
UCLA WWW Home Page

Social Sciences Division

American Indian Studies Center

For more information contact aisc@ucla.edu or (310) 825-7315
Last Updated :  Monday, October 14, 1996