The Economics of Crop Insurance and Disaster Aid.

This book is a nifty example of the insights that careful economic analysis can bring to public policy debate and reform. Barry Goodwin and Vincent Smith, agricultural economists from North Carolina State University and Montana State University respectively, do an excellent job presenting in clear and straightforward fashion the economic history and implications of the federal crop insurance and disaster aid program. This program cost the federal government more than $25 billion in subsidies and payments in the period 1985-1993 alone.

Goodwin and Smith begin their analysis with a simple question: What is the purpose of federal multiple-peril crop insurance (including disaster aid, which they correctly describe as a form of premium-free crop insurance)? They report several possible answers: to rectify market failure stemming from the inability of the private market to deal efficiently with problems of moral hazard and adverse selection; to enhance equity by transferring income from well-off taxpayers to distressed and less well-off farmers; or, finally to transfer income to farmers as a way of garnering their political support. Goodwin and Smith systematically review the evidence for each of these arguments.

In Chapters Two and Three they describe the current crop insurance and disaster payment program and chart its evolution from its inception in 1938. The legislative history of the 1938 Crop Insurance Act reveals a concern on the part of the legislators of that time, not to address any market failures of the private crop insurance market, but to relieve widespread hardships caused by recent, severe droughts. From the beginning the program has been plagued by problems of moral hazard and adverse selection. Moral hazard problems result when farmers adopt poor farming practices because they are insured. Adverse selection problems result when low-risk farmers are charged too high premiums and therefore choose not to participate, leaving a higher proportion of high-risk farmers in the insurance pool and causing a upward spiral in premiums if actuarial soundness is to be maintained. These twin problems have doomed federal crop insurance to high losses and low participation rates. This has not gone unnoticed by Congress which regularly tries to reduce losses and increase participation rates with legislative tinkering (the most recent example is the 1994 Crop Insurance Reform Act).

Goodwin and Smith describe four distinct phases of legislative experience and reform. From 1938 until 1943, Congress got a first taste of the losses they could incur from actuarially-inefficient crop insurance. Local committees set the premiums and loss assessments for neighboring farms based on county-wide averages instead of individual farm data and participating farmers could delay paying their premiums until well into the growing season. As a result, premiums were set too low, expected losses were underestimated, and farmers with a good start to their crops simply dropped out. In 1943 Congress scrapped the program.

Congress restarted the program in 1944 with several initiatives for reducing losses and increasing participation rates. Some were successful, some were not, and Goodwin and Smith carefully show that low subsidies and high participation rates are mutually exclusive goals for a voluntary insurance program. In 1974, federal crop insurance was complicated with the institution of mandatory disaster payments. The disaster aid program quickly proved to be extremely costly because it encouraged farmers to plant in marginal areas, effectively insured farmers against inefficient farming practices, and weakened farmers' incentive to get as much yield as possible out of poor growing seasons. In 1980 Congress passed the General Crop Insurance Act designed to replace the disaster aid program with an expanded, universal crop insurance program. The starting premise for the reform was that some premium would reduce program losses better than no premium. Again, problems in developing an actuarially sound program combined with a series of ad hoc disaster aid bills led to low participation rates and high subsidies. The latest attempt to improve federal crop insurance is the 1994 Crop Insurance Reform Act, which Goodwin and Smith describe in depth.

Goodwin and Smith show how, with each attempt at reform, the problems of moral hazard and adverse selection have served as Achilles' heels to efficient federal crop insurance. Yet they find that the justifications for the various crop insurance bills rarely include arguments for government provision as a response to private market failure. In Chapter Four Goodwin and Smith explain why this is the case. They build a model of the supply and demand of crop insurance which implies that a private, multiple-peril crop insurance market is not viable. Farmers do not value crop insurance enough to pay its cost. Moral hazard and adverse selection are important factors in this result. For example, moral hazard problems raise monitoring costs and lead to high deductibles. High information costs for discerning individual farm risk and appropriate premium levels lead to adverse selection problems. However, Goodwin and Smith argue convincingly that these problems should be viewed as inherent market conditions, not sources of market failure. Any provider, be it government or private, will have to deal with them and there is no reason to believe that government can monitor farm behavior or assess individual farm risk better than the private market. In Chapter Five, Goodwin and Smith support their conclusion with an overview of the empirical evidence worldwide on the viability of the private crop insurance market.

In the end Goodwin and Smith conclude that the purpose of federal crop insurance is political rent-seeking. Market failures do not explain government insurance and the era when farmers were a low-income group deserving of income transfers is over. With the purpose of federal crop insurance discovered, they turn in Chapter Six to a review of alternative government-sponsored crop insurance programs in other countries. Their carefully supported conclusion is that none of these alternative forms has been successful at improving farmers' participation rates unless farmers' premiums have been heavily subsidized by taxpayers.

Claire Holton Hammond Wake Forest University