United States General Accounting Office Testimony For Release Analysis of Crop Insurance Proposals on delivery H.R. 4360 and H.R. 4592 Expected at 10:00 a.m. EDT Tuesday May 1, 1990 Statement of John W. Harman, Director Food and Agriculture Issues Resources, Community, and Economic Development Division Before the Subcommittee on Conservation, Credit, and Rural Development Committee on Agriculture House of Representatives GA0Foi-m 180(12;~~) GAO/T-RCED-90-76 Mr. Chairman and Members of the Subcommittee: We are pleased to testify before this Subcommittee on ways to reform the federal crop insurance program. As you requested, our testimony addresses how well two legislative proposals for replacing the current crop insurance program--H.R. 4360, introduced by Representative English, and H.R. 4592, introduced by Representative Grandy-- meet eight criteria for assessing the federal role in providing disaster assistance to farmers. These criteria were initially identified in a September 1989 GAO report-l Our analysis focuses only on the two proposals you are most interested in. Other alternatives could meet more criteria than these proposals. In summary, both proposals would generally meet four of the eight criteria. However, these proposals differ in the degree to which they meet these criteria. Like current programs, both proposals strongly rely on federal resources for premium subsidies and the payment of excess losses but rely on private companies to deliver services. We believe that any proposal that uses a mixed private/public system requires a strong federal regulatory, oversight, and audit function to ensure program integrity and to protect the government's financial interests. In particular, we believe H.R. 4592 requires stronger federal oversight. lDisaster Assistance: CrOR Insurance Can Provide Assistance More Effectivelv Than Other Proqrams (GAO/RCED-89-211, Sept. 20, 1989). CRITERIA FOR ASSESSING CURRENT DISASTER ASSISTANCE PROGRAMS Let me now briefly discuss our criteria for assessing federal disaster assistance programs to farmers. In developing these criteria, we determined that the policy principles of equity and efficiency are essential elements of any desirable disaster assistance program. These principles suggest that an equitable disaster assistance policy ensures that aid is provided consistently to victims suffering from similar losses over time. An efficient disaster assistance policy ensures that benefits are provided at the lowest possible cost to government and to society as a whole. In our opinion, an equitable and efficient disaster assistance policy should exhibit the following eight characteristics: -- (1) determine compensation by the amount of a farmer's loss, not by the severity of the disaster, -- (2) provide similar amounts of assistance to farmers suffering similar amounts of losses, -- (3) not provide farmers more assistance than the amount of their disaster losses, 2 -- (4) not create incentives to encourage farming practices that increase the likelihood and extent of losses, -- (5) make the programs consistently available over time to allow for long-range planning, -- (6) help farmers withstand and recover from the effects of natural disasters, -- (7) provide predictable annual costs, and -- (8) meet their objectives at the lowest possible cost. RESULTS OF OUR ANALYSIS Both proposals meet our first criterion, which is that the amount of disaster assistance provided should be determined by the amount of a farmer's loss, not by the severity of the disaster. Both meet this criterion because the conditions for compensation are agreed to before the growing season. No legislative or administrative action-- such as a presidential disaster declaration-- is needed after the fact to determine the level of compensation. Both proposals also meet our second criterion, which is that disaster assistance programs should provide similar amounts of 3 assistance to farmers suffering from similar amounts of 10ss.~ In a broad sense, they meet this criterion because all crop insurance purchasers would be treated equally according to the terms of their policies. Neither proposal meets our third criterion, which is that disaster assistance programs should not provide farmers more assistance than the amount of their disaster losses. In determining the amount of insurance policy coverage, both proposals permit producers to substitute ASCS county yield data for actual production histories. A producer would be expected to use the ASCS yield when it is higher. Therefore, producers under both proposals can be compensated for more than they would have been entitled to if they had used their actual production histories to determine their yields. Neither proposal meets criterion four, which is that disaster assistance programs should not create incentives to encourage farming practices that increase the likelihood and extent of losses. Subsidized disaster assistance programs discourage farmers from taking risk-reducing measures because, with subsidies, farmers may be able to obtain disaster assistance that provides protection at a cost lower than prevention. Generally, the more a program is subsidized, the less likely it is that 2For the purpose of this analysis, we are ignoring the tax system's effect on meeting this criterion. 4 farmers will try to reduce risks. H.R. 4360 does not meet this criterion because it makes substantial use of subsidies, particularly in its provision of a free disaster assistance plan to producers. H.R. 4592 also includes direct and indirect federal subsidies. Because the proposal is different from and less specific than the current system, it is difficult to determine the extent of program subsidy that would be involved. H.R. 4592 also does not meet this fourth criterion because it has features that increase the risk of excessive loss. This proposal emphasizes a more privatized insurance program, characterized by (1) a commissioner who will have apparently fewer resources than the current FCIC manager, (2) a separate office of crop insurance operations, (3) an advisory committee with few members representing the federal government's interests, and (4) a mixed federal/state regulatory structure that may not provide the federal government with sufficient authority to regulate the program effectively. Our experience in reviewing loss adjustment practices--which have shown significant amounts of overpayments-- indicates that strong federal oversight and control is necessary to protect the government's financial interests.3 Generally, both proposals meet criterion five, which is that disaster assistance programs should be consistently available over 3Croo Insurance: Private Company Loss Adiustment Imnrovina, but Overuavments Still Hiqh (GAO/RCED-90-32, Nov. 7, 1989). 5 time to allow for long-range planning. Actual implementation experience, however, might reveal significant differences between them. H.R. 4360 lessens the likelihood of an ad hoc disaster assistance program by providing most producers with a limited amount of disaster assistance protection at no cost. Furthermore, the effect of H.R. 4592 on participation and the comprehensiveness of coverage (both in terms of numbers of crops and locations) is not clear to us. The failure of either proposal, however, to attract sufficient participation or provide sufficient protection could lead to periodic pressures for ad hoc disaster legislation, creating uncertainty for producers. Both proposals meet criterion six, which is that disaster assistance programs should help farmers withstand and recover from the effects of natural disasters. These proposals meet this criterion because they provide cash assistance to producers to enable them to continue farming. For several reasons, neither proposal meets criterion seven, which is that disaster assistance programs should have predictable annual costs. First, both proposals subsidize the purchase of crop insurance and, thus, program costs will vary with participation rates. And second, neither proposal significantly addresses how to make crop insurance actuarially sound, a situation in which premium costs reflect the risk underlying insurance coverage. The federal government would absorb excess losses under each proposal. 6 Therefore, federal expenditures could increase to the extent that premiums do not fully reflect insurance risks. In particular, H.R. 4592 requires the Crop Insurance Commissioner to establish an actuarially sound premium for a catastrophic reinsurance fund, yet some insurance officials have told us that this cannot be done. Neither proposal meets our last criterion, which is that disaster assistance programs should meet their objectives at the lowest possible cost. Both proposals are heavily subsidized, which increases the likelihood and extent of losses. The federal government would ultimately have to compensate insurance companies for some of these excess losses. In addition, the H.R. 4592 regulatory structure raises questions about how effectively program abuse can be controlled, which can also lead to extra government expenditures. This proposal also includes a federally funded hardship allowance for premium increases above 35 percent. Although rates are subject to federal approval, there appears to be minimal incentive to keep premium costs down once this level is reached. As we understand the proposal, the policyholder pays nothing additional out of pocket, and funds do not have to be appropriated before the fact to cover these expenditures. Mr. Chairman, this concludes my prepared statement. My colleagues and I will be happy to answer any questions you may have.
Analysis of Crop Insurance Proposals H.R. 4360 and H.R. 4592
Published by the Government Accountability Office on 1990-05-01.
Below is a raw (and likely hideous) rendition of the original report. (PDF)