Governments around the world support a wide range of activities through credit and insurance markets. Though credit support comes in a range of guises, all methods have the same goal: to provide financial support to enterprises and/or activities at more favorable terms than they would be able to obtain by borrowing on their own. In some cases, access to credit allows industries or technologies to exist when they otherwise wouldn't. Insurance subsidies attempt a similar improvement in competitive conditions for selected firms/industries by shifting some or all of the risks of these activities to the public sector. While all of these subsidies are usually justified on the grounds that they support socially-beneficial activities, there are many less admirable reasons that they are used so widely across the world:
- Less Visibility. Unlike direct grants to industries, subsidized loans, insurance, and especially loan guarantees, can be done with very little budget appropriation and much less oversight than occurs for direct spending. While problems will eventually arise through defaults on the loans or claims on the insurance policies, the outlays associated with such programs often occur many years into the future, beyond the political horizon of current decision makers.
- "No Cost." Where governments lend money to industries at the federal cost of borrowing, politicians may argue that there is no real cost to the government, and that such loans should be promoted as examples of "public-private" partnerships. Unfortunately, this argument is false. More loans means more expected defaults. Administrative costs also rise, and are not always recovered through lending rates. Finally, because funds are always limited at some point, the choice of which industries get access to the government's borrowing rate and which are left to fend for their own in private markets can create substantial distortions in the marketplace.
- Export Promotion. Many countries have export-import banks that lend to purchasing countries at subsidized rates to enable them to purchase domestically-produced capital goods. Often, these decisions are politically-influenced and can impede market access for competitors with improved or more environmentally sound products.
Additional information on how these programs work is below.
Loan programs provide subsidies through interest rates, repayment terms, and defaults. Loans may be lent directly to corporations at favorable interest rates, sometimes significantly below the government's cost of borrowing. Loan repayment schedules may also be highly favorable to the lender. For example, debt to federally-owned Power Marketing Administrations in the US extended 50 years or more.
Favorable repayment conditions generate subsidies by allowing borrowers to defer repayment until the end of the loan life. Even though the nominal value of the loan will eventually be repaid, since the duration of these loans is longer than what was normally available in the capital markets of the time, the federal government will have to refinance the loan at new market rates, but will not be able to adjust the interest rates on its loans to the long-term borrower. Thus, deferring repayment until the end of the lending period shifts all interest-rate risk to the federal government and allows borrowers to pay back loans in inflated (perhaps greatly so) dollars.
Government loan guarantees eliminate the default risk to the lender by shifting it entirely to the government, enabling the borrower to obtain much more favorable loan rates. Often, without the guarantee, the loan would not have been approved at all. In other cases, the interest rate would have been higher. Although one may try to incorporate the value of guarantees by estimating only the expected defaults that the government must pay back, this approach understates the true value of loan guarantees to the energy sector. Even if there is no default, the guarantee actually gives the borrower access to lower cost financing than would otherwise be available. Equally important, the guarantee often makes it possible for the borrower to modify its capital structure to one using more debt and less equity. Since equity is more expensive, this shift as well provides immediate and recurring subsidies to the borrower.
Comprehensive evaluation of the value of loan guarantees is not easy. As a rule-of-thumb, the Organisation for Economic Cooperation and Development has valued loan guarantee subsidies at 1 percent of the loan value. This value may be reasonable for guarantees to larger firms in more developed countries, but is far too low for loans supporting smaller firms or projects in high risk business sectors or high risk regions of the world.
For both loans and loan guarantees, the interest rates and other fees charged to borrowers have rarely included enough of a premium to cover the large defaults on which the government must make good. These high default rates are probably due both to the fact that the government lending targets more higher risk ventures than do private lenders, and to less stringent risk assessment prior to approving loans. For example, defaults on direct and guaranteed loans from both the U.S. Export Import Bank and the U.S. Rural Electrification Administration (the predecessor of the Rural Utilities Service) have historically been a large percentage of outstanding obligations.
Lenders normally charge higher interest rates for riskier loans, and the differences in rates between borrowers of different grades can be quite large. This is graphically illustrated using actual data from the oil industry. The private cost of capital (PDF file) to oil and gas extraction companies is twice the cost of government debt. Rates to smaller exploration firms would be even higher. For many industries, the larger the portion of capital that can be met through access to federal loans, the better.
Government-provided insurance programs have many of the same characteristics as loan and loan guarantee programs. Premiums often don't cover policy losses, and government risk-bearing is often cheaper than the equivalent service on the private market due to economies of scale and no required rate-of-return. As with loan programs, not every competing energy service has similar access to federal insurance, introducing inter-fuel market distortions.
Some government insurance programs are run by a government agency, collect at least some premiums, and make at least some attempt to set those premiums based on actuarial assessments of the risk being insured. While net subsidies often exist, there are some revenues being returned to the government as well. In contrast, indemnification programs hold a private entity harmless for the costs of particular activities (e.g., a nuclear reactor accident) by agreeing to pay damages, or by shifting risks to the public. These programs usually require more extensive analysis to value. This is because they often shift low probability, but extremely expensive, risks off of private industry. Unlike insurance programs, these market interventions do not have premiums, and do not have any federal agency responsible for regularly assessing risk exposure. Even the government itself may not have a comprehensive view of its exposure. Insurance programs exist on a continuum, with full indemnification at one extreme, and private insurance (with no public subsidies) at the other.
All of these loan, guarantee, and insurance programs provide an intermediation benefit to borrowers, since the federal government can borrow funds and absorb risks more cheaply than most private entities can. Since not all participants in natural resource markets have equal access to advantageous rates on federal loans, guarantees, and insurance, this access becomes an additional barrier to entry and is properly recognized in subsidy assessments. The value of government intermediation can be seen graphically as the difference between private market rates and the government's full recovery interest rate in "Subsidies Through Government Lending Programs" (PDF file).
Measuring the Benefits
Interest rate subsidies are calculated by comparing what the borrower actually paid for funds and proxies for what the real cost of the funds would have been without the government program. To bound the true subsidy value, it is useful to estimate both a high and a low value. The low estimate are compares the government's cost of funds to the interest rate charged. Within the US, we use the difference between the Treasury (or, in some cases, the Federal Financing Bank) borrowing rate and the interest rate charged to the borrowing entity. The high estimate uses the cost of funds to power borrowers in the private capital markets. The weighted average cost of new gas, power, and light bonds from the Moody's bond rating service is a good proxy for this value. For other countries or industrial sectors the most appropriate reference values will differ.
The high estimate better reflects the net advantage received by the subsidized entity relative to substitutes through its access to government borrowing. Defaults on loans and loan guarantees are based on estimates by the relevant federal agencies in the form of their "provision for losses" entry in their financial statements.
As with direct grant programs, loan and loan guarantee programs may sometimes be justified on the grounds that they improve societal equity. For example, the Rural Utilities Service within the US subsidizes rural sub-sectors of the country to improve their quality of life. Again, while these expenditures may be justifiable (or have been justifiable at one point in time), they do distort market choices. For example, subsidizing rural electrification will obscure the point at which extending transmission lines becomes more expensive than building small scale decentralized power or on-site renewable sources. Furthermore, such a program contains an embedded assumption that access to power lines is a critical element in rural development, as assumption far less valid in an era of cellular and satellite communication.