According to the U.S. Bureau of the Census, American Indians comprise slightly under 1 percent of the U.S. population (an estimated 2.3 million persons in 1997) and, correspondingly, slightly less than 1 percent of U.S. households nationwide. Many Indian households are found in and around reservations/tribal lands, but Indian households are also distributed throughout the country. In considering initiatives to support the Indian population, one must consider Indian access to and costs for energy, especially electricity. This chapter summarizes available information on these issues from the 1990 Census of Housing (the so-called "long form" of the Decennial Census) and the Energy Information Administration's (EIA) 1997 Residential Energy Consumption Survey (RECS), a national sample survey of household energy use that enables energy data to be evaluated according to household characteristics. Both the Census of Housing data and the RECS data for U.S. households as a whole are quite precise, but RECS data for Indian households are subject to larger uncertainty because Indian households are a small proportion of households in the country, and thus, in the RECS sample.

Figure 1. Percent Distribution of U.S. and Indian Households by Electricity Access/Payment Status, 1990

Of the approximately 600,000 Indian households in the United States in 1990, almost 20 percent were on Indian reservations with over 500 households. Another 10 percent were located in so-called Tribal Jurisdictional Statistical Areas (TJSAs) in Oklahoma, and the remaining 70+ percent were spread across the country. The Indians in households not on Indian lands, as well as those in the TJSAs, had access status for electricity similar to U.S. households as a whole (Figure 1). Only a small percentage of households were recorded by the 1990 Housing Census as having no cost for electricity/no access. However, Indians on reservations were another story. Fully one in 7 households, about 16,000 total, were in the no cost/no access category, and authoritative sources state that the reason is the lack of access to electricity, not central purchasing organizations or other arrangements that might provide Indian households with access at no cost.⁽¹⁾ Over three-quarters of these 16,000 households were located on the Navajo reservation of Arizona/New Mexico, where over one-third of the 34,000 households did not have access, even though generation and transmission facilities are located within the boundaries of the reservation. Detailed information from the 1990 Census of Housing, which include data for individual TJSAs and reservations of over 500 Indian households, is given in Appendix B.

Indian households that did pay for electricity in 1990 had costs that were similar to U.S. households as a whole, whether or not they were on reservations. However, because of their generally lower household incomes, electricity costs were a greater burden for Indian households, especially those on reservations (Figure 2). While the distribution of electricity costs relative to household income was only slightly more burdensome for Indian households outside of reservations than for the general population, fully 10 percent of Indian households on reservations spent 20 percent or more of their income on electricity.

The access issue is much less clear for natural gas and other fuels (e.g., coal, wood, and propane). These are not considered as crucial for the modern lifestyle as electricity and are not used by virtually all U.S. households, as electricity is. However, for Indian households on reservations that do pay for these fuels, the burden remains. The most-burdened 10 percent of these households paid a much higher proportion of their income for these fuels in 1990 than did U.S. households as a whole or non-reservation Indian households (Figure 3). It should also be noted that 37 percent of Indian households on reservations used and paid for one or more fuels besides natural gas and electricity in 1990, a percentage that is twice as high as for U.S. households in general (18.4 percent) or for Indian households not on reservations (17.5 percent).

Figure 2. Percentiles of Electricity Costs Relative to Total Household Income, for Households That Pay for Electricity, 1990

Figure 3. 90th Percentile of Energy Costs Relative to Total Household Income for Households Paying for the Energy, 1990

The next portion of this chapter focuses on more recent data from the EIA's 1997 Residential Energy Consumption Survey (RECS), which includes energy consumption as well as expenditures data, but is much less precise due to its relatively small sample size, especially for Indian households.⁽²⁾ The 1997 RECS estimates that Indian households consumed about 101 trillion Btu (TBtu) of major energy sources (electricity, natural gas, LPG, fuel oil, and kerosene) in 1997, roughly the same amount as households in the Kansas City metropolitan area. Over half this energy, about 54 trillion Btu, is electricity (including power station use and transmission losses), and another 41 trillion Btu is natural gas. Together, these two sources account for 94 percent of major energy use (Figure 4), about the same fraction that they repre-sent for all U.S. households. The electricity quantity is equivalent to 5.2 billion kWh, about the amount that could be generated by a 600-MW power plant operating at full capacity 24 hours per day throughout the year. The total 1997 energy bill for Indian households was about $750 million. Electricity accounted for three-fifths of Indian household energy expenditures, and natural gas about one-third.

Figure 4. Energy Consumption and Expenditures for Major Energy Sources in All Indian Households, 1997

Figure 5. Consumption in 1997 Per Household Using Energy Source: All U.S. Households and All Indian Households

The average Indian household consumed about 143 million Btu of primary energy (including electricity losses), 28 million less than the average across all U.S. households, geographically adjusted.⁽³⁾ This difference across major energy sources is of marginal statistical significance. However, the average Indian household supplied with electricity clearly uses less than the average U.S. household. On average, Indian households use only about 75 percent of the electricity used by U.S. households in general, a statistically significant difference (Figure 5).

Primarily because of this substantial difference in electricity use, the average 1997 energy bill for Indian households was almost $1,100, nearly $200 less than the average bill for all U.S. households--a significant difference. Natural gas, however, does not show a significant difference (Figure 6). Indian households paid prices for electricity (8.7 cents per kWh) and natural gas ($6.36 per thousand cubic feet) that were not significantly different from the prices paid by U.S. households as a whole (Figure 7). In other words, while Indian households use less energy, and specifically, less electricity than U.S. households as a whole, RECS shows no evidence of differential price experience for Indian households.

Figure 6. Expenditures in 1997 Per Household Using Energy Source: All U.S. Households and Indian Households

Figure 7. Electricity and Natural Gas Prices: All U.S. Households and All Indian Households

Retail Electricity Rates Paid by Indians
Living on Indian Reservations and TJSAs

For comparison with the RECs survey statistics gathered from the Indian households, additional information on electricity prices on Indian Reservations and TJSAs in 1998 was estimated from the electric utilities serving those areas. EIA's "Annual Electric Utility Report" (Form EIA-861) collects data from all electric utilities on their residential sales and revenues in each State. The survey does not specifically identify Indian lands, however, so average prices have been estimated using several sources to identify the utilities serving each Indian land. These prices should thus be viewed as approximate.

Figure 8. Distribution of Residential Electricity Rates on Indian Lands

Roughly 40 percent of Indian households on reservations and TJSAs pay between 7.0 and 7.5 cents per kWh for electricity in their homes (Figure 8). Ninety-two percent of the 175,000 Indian households on Indian lands are located in just four of the North American Electric Reliability Council subregions (Table 1). Electric utilities servicing counties containing Indian lands in three of those four subregions have higher rates than all utilities with residential customers in the subregion. From these data, it is impossible to determine whether the higher costs are due to the cost of service for sparsely populated rural areas, including Indian lands, or other factors.

Not surprisingly, the reservations with the lowest prices are those in the Pacific Northwest where extensive hydropower is available, while the highest prices are in Maine, New York, and California (Table 2). Reservations in New Mexico are also estimated to have relatively high rates.

Methodology for Estimating Electricity Rates Paid by Indian Land Households

The EIA-861 data were used as the source of all utility average residential electricity prices. To assign one or more utilities to each Indian land as the most likely provider, EIA used the following sources:

• A list from the Rural Utilities Service (RUS) of Cooperative Utilities that serve Native Americans
• Information from various web sites about utilities that sold electricity to Indian lands
• Specific information on the utilities serving the Navajo Tribal Utility Authority and the Tohono O'Odham Tribal Authority
• General information and assumptions about electricity providers to Indian lands.

Figure 9. Number of Non-Municipal Utilities Serving Counties with Indian Lands

For the majority of Indian lands, counties in the Indian land were matched with utilities having residential sales in those counties. However, in most counties, there are multiple utility providers, including investor-owned utilities (IOUs), cooperatives, and other providers, such as public utility districts, State agencies, or Federal agencies. Municipals were excluded because they serve single towns or cities and would overstate the number of potential providers to the reservations. Of the 319 counties which contain populated portions of reservations or TJSAs, almost 40 were served by only one utility, but 61 had 5 or more utilities (Figure 9).

If there was no information to determine which utility was the provider, it was assumed that if a cooperative were in a county in which there was an Indian land, then it was the likely provider. In cases where more than one of these cooperatives was listed for a county, their rates were averaged. The methodology further assumed that if the Bureau of Indian Affairs (BIA) sold power in the county, it was a likely provider. For the remaining counties, where no RUS borrower or BIA was designated, the residential rates of all the IOUs and cooperatives serving the county were averaged. If there were no IOUs or cooperatives in a county, which occurred in a few cases in the Northwest, then the Public Utility District average rate was used.

Average residential electric rates by county were combined into an average for each Indian land by weighting each county according to the number of households in the county.

Endnotes

1. Based on conversations with Richard Wilson of the Bureau of Indian Affairs on February 17, 2000 and David Lester of the Council of Energy Resource Tribes on February 25, 2000.

2. RECS data cover all Native American households, whether on tribal lands or not.

3. The remainder of this section compares average household energy use and expenditures and average energy prices for Indian households and U.S. households as a whole. One major problem with such comparisons is that Indian households are distributed much different geographically across the U.S. than households of other groups. Energy use and energy costs vary widely by geographic area, due to availability of energy sources, utility cost structures, etc. Thus, unadjusted comparisons between Indian households and U.S. households as a whole might reflect geographic differences rather than any difference attributable to Indian experience. To address this concern, all data for U.S. households are adjusted so that they are based on a distribution in the geographic areas measured by the RECS that is the same as the distribution of Indian population in those areas as of July 1, 1997, according to the U.S. Census Bureau's population data set PE-65, "Estimates of the Population of States by Age, Sex, Race, and Hispanic Origin: 1990 to 1997."