Skip Navigation

Should the Federal Government Sell Electricity? November 1997

This appendix describes the derivation of the net cash flows that serve as the basis for the study's market valuations. It also presents a sensitivity analysis for those valuations and for the budgetary effects described in the study.
 

Net Operating Income, Net Cash Flows, and Discount Rates

The Congressional Budget Office (CBO) used program data to estimate net operating income for each federal power agency. Those data are shown in Tables B-1 and B-2. Net operating income, along with assumptions about taxes and future income growth, yield estimates of future net cash flows that are the basis for high and low market valuations.
 

Current operating data for the power marketing administrations (PMAs) come from the statements of revenues and expenses appearing in the PMAs' 1995 annual reports. For the Tennessee Valley Authority (TVA), the data represent revenues and expenditures for 1996. The reason for using 1996 information for the TVA was to obtain a picture of that agency's cash flow after the Browns Ferry 3 and Watts Bar 1 nuclear units had entered service. The study assumes that the remaining levels of capital expenditures that the TVA and the other agencies report represent a routine amount that a new owner would have to spend every year, along with outlays on operations, to keep producing power. (That approach does not distinguish new capital expenditures from the new operating costs that such expenditures would produce.)

For the study, CBO assumed that a new owner would not take on the debt obligations or program costs of an agency that were unrelated to power generation and transmission. The largest costs of federal agencies unrelated to supplying power but paid for by power revenues come from three sources: the residential exchange program of the Bonneville Power Administration (BPA), in which the BPA subsidizes the cost of power to certain utilities that are not preferred customers; certain environmental activities of the BPA; and the payments that the TVA makes to local governments in lieu of taxes. (Other nonpower programs of the TVA for environmental activities, economic development, and stewardship of the river system are funded by Congressional appropriations, not power revenues.)

The net operating income that a potential buyer would look at would not include the costs for nonpower activities and interest on agency debts. CBO subsequently inflated those estimates of current net income to reflect assumptions about productivity increases, initial rate changes (increases from current federal rates to investor-owned utility rates), and future growth in net income (zero growth or growth at inflation). Productivity increases for the PMAs--equal to 5 percent of current sales at no extra cost--are related to assumed efficiency gains at federal hydropower projects, not to growth in demand. There is little untapped hydrocapacity on the nation's waterways. The assumed sales increase for the TVA pertains to increased use of existing capacity in response to growing regional demand. Assumptions about net cash flow for the TVA reflect additional costs for generating nuclear power.

In addition to net operating income, the net cash flow to a new owner reflects tax liabilities. In this study, CBO assumed a combined liability for federal and local taxes of 40 percent of income, less deductions for capital depreciation. Capital depreciation is based on the sales prices of the federal assets and an assumed remaining project life of 30 years.

The representation of tax liabilities treats all routine operating and construction costs as expensible. If the new owners must depreciate certain of those construction costs--that is, deduct them from taxable income over time--the present value of net cash flow will be lower than that presented here. Making that adjustment, however, requires distinguishing future capital costs from the additional operating costs for the new units, identifying which of those capital costs is depreciable, and determining the appropriate depreciation schedules.

For the study, CBO assumed a discount rate of 10 percent to convert future cash flows to their present-value equivalents. The actual discount rate that a potential buyer of federal assets would apply would reflect the weighted average cost of capital for that business.⁽¹⁾ Weighted average cost of capital is a measure of the rate of return that the business could obtain by investing its resources elsewhere. That discount rate reflects the cost of borrowing for the business, the amount of the purchase price the business will finance by borrowing, and the return on equity (or nonborrowed funds) that the business requires.

A major uncertainty underlying the choice of discount rates is the fact that in relation to other businesses, investor-owned utilities (the most likely group to express interest in acquiring federal assets) have low borrowing costs (around 7.5 percent) and therefore tend to finance much of their acquisitions with new debt (around 50 percent). Utilities' returns on equity are about 10 percent, which suggests a weighted average cost of capital of about 8.75 percent. But some non-utility groups that have higher costs of capital may bid. Furthermore, growing competition in power markets may be changing the borrowing costs and debt structures of utilities themselves. Aside from those concerns, the range of discount rates that appear to explain major acquisition decisions by U.S. businesses is very broad.⁽²⁾ For those reasons, CBO decided that the discount rate should be higher than 8.75 percent, but the choice of 10 percent is not based on explicit assumptions about future borrowing costs, debt structures, or returns on equity.
 

Sensitivity of Valuations of Net Cash Flow to Major Assumptions

Market valuations based on net cash flow are subject to many uncertainties. Chapter 5 reviewed the consequences of changing some of the basic assumptions of the cash flow analysis: higher and lower growth in power rates (and net cash flow), limitations on a sale to account for liabilities for nonpower programs, temporary freezes on power rates, and delays in the start-up of hydropower operations (see Table 12). Market valuations are also sensitive to small changes in discount rates, tax rates, and other elements of cash flow. Table B-3 summarizes the sensitivity of the high and low market valuations to those changes. (The results are approximately symmetrical for increases or decreases in the assumed values.)
 

In general, the results of the sensitivity analysis are not additive. All combinations of assumptions are not equally likely; many are extremely unlikely. The sensitivity results are presented here because CBO cannot know which businesses might bid for federal power assets or what their specific market assumptions might be. For anyone who has that information, the sensitivity analysis provides a quick way to see how the conclusions of this study could change.
 

Sensitivity of Budgetary Effects to Major Assumptions

The budgetary value to the government of keeping power assets--or the present value of future budgetary income, as discussed in Chapter 6--is also subject to many uncertainties. The study discusses how the budget in future years would be affected, on net, by receipts that the power agencies must collect to repay their debt obligations, assuming that power receipts would fully cover operating costs as they occur. Budgetary effects were estimated as the present value of future debt repayments (including interest). That approach describes the maximum contribution to the budget that power agencies can make under current law.

Future market conditions--as they affect power rates, sales, or costs--may complicate or ease the agencies' task of collecting sufficient revenues. Table B-4 summarizes influences on the maximum budgetary effects that could result from a rise in federal power rates to current market levels in each region, growth in power rates from current levels (without growth in costs), growth in sales by the TVA to fully use existing capacity (recall that the hydrocapacity of the PMAs will not support sustained sales growth), a loss of federal customers caused by competition, or growth in costs for operation and routine construction (without growth in rates). For the sensitivities on power rates, the analysis assumes the agencies have sufficient discretion to make those changes under current law.
 

As with the sensitivity analysis for market values, the particular budget sensitivities are illustrative, and the results may not be additive. Most important, current law would constrain federal rate setting in all cases: the maximum budgetary impact is based solely on current requirements for debt repayment. Thus, the numbers in Table B-4 simply represent potential sources of constraint on federal power income--not sources of additional income. In other words, factors associated with a positive influence on budgetary effects, such as growth in power rates or sales, can only offset the negative influence of other factors, such as growth in costs or loss of customers.

1. See Thomas E. Copeland and J. Fred Weston, Financial Theory and Corporate Policy (Reading, Mass.: Addison-Wesley Publishing, 1979) or other standard textbooks on corporate finance.

2. See Eugene F. Fama and Kenneth R. French, "Industry Costs of Equity," Journal of Financial Economics, vol. 43, no. 2 (1997); Steven N. Kaplan and Richard S. Ruback, "The Valuation of Cash Flow Forecasts: An Empirical Analysis," Journal of Finance, vol. 50, no. 4 (1995); and J.C. Bosch, "Alternative Measures of Rates of Return: Some Empirical Evidence," Managerial and Decision Economics, vol. 10, no. 3 (1989).

---