APACAgricultural Policy Analysis Center

Back to Articles

September 27, 2002

Crop Insurance and Disasters

No doubt you have heard the assertion: "There will not be another disaster program for agriculture; disaster programs are things of the past." Such statements have been confidently declared at various points in time over the thirty years that I have paid attention to such things. The assertions usually follow a "reform" of the crop insurance program or a "significant improvement" in the farm bill. Each time, I would bet that disaster payments will continue to be paid.

As along as there are systemic, severe problems that beset agriculture (i.e. as long as weather is variable and uncertain), we will have disaster programs, especially when disasters involve widespread short-fall in production. In fact, to me, the astonishing thing is not that we continue to have disaster programs but that we think it is reasonably possible to concoct a crop insurance program that would eliminate the need for disaster programs.

Insurance is best suited to providing protection against random negative events; random in the sense that only a few of those purchasing the insurance will experience the event. Fire insurance works because relatively few houses, not all houses, catch fire in a given year. Generally, anything that would cause all houses to catch fire is excluded in the fine print of the insurance policy.

That is why crop insurance seems best suited to compensate for hail damage and other localized events that could affect a given farm but not all farms nationally. When insurance is discussed as protection against generalized droughts or low prices, its scope is expanded from random events affecting a few farms and farmers to systemic, widespread problems that affect all or a large share of farmers.

Using insurance to protect against low prices or low incomes sounds great. And it would be great except that when prices plummet, it is crop agriculture's version of all houses burning down in the same year. Furthermore, in the corresponding application to agriculture, next year all the newly rebuilt houses could burn down again if prices keep falling. Now that would be expensive insurance, whether paid for privately or by Uncle Sam. And if the level of coverage is based on previous-years' market incomes and prices, the level of coverage could spiral downward as insurance coverage chases declining crop prices and incomes. Many of the price/income insurance proposals of recent years seem to assume that, on average, farm prices and incomes are about evenly divided between "good" and "bad" price years. This is, of course, a very bold assumption, especially if there are few additional farm program instruments available for use.

While the chances of a widespread drought is unlikely to happen in consecutive years, this use of insurance also does not fit the random-event-across-all-farms scenario. Widespread drought is systemic, not random. Of course, in years in which weather-based yield reductions are not widespread, losses are by definition not agriculture-wide, even if losses are not strictly random geographically. But in years like this one, even though about two-thirds of program crop production have some kind of crop insurance coverage, the economic losses are so pervasive and severe and affect so many farmers and rural communities that crop insurance proceeds are judged to be inadequate to maintain the health of the industry and disaster payments are invariably paid.

It may be a good idea to step back and reconsider what can be realistically accomplished with private crop insurance. Then, we could consider how to deal with disasters that can't reasonably be handled well by private/public insurance-with or without the kinds of subsidies that are currently being paid. Years ago, we knew what private crop insurance was for: it was to provide a degree of protection should a hail storm strip the corn or soybeans down to twigs or harvest the wheat for us. Using insurance in this way should mean that rates will vary by geographical areas and depending on which events are insured against, which could include drought reduced yields on an insured farm but possibly at a relatively high rate. This is the kind of thing that private insurance companies do and they are good at it.

As discussed above, it is the agriculture-wide events that don't fit the true insurance model. So, let's admit that and quit trying to fit a square peg in a round hole. Since we know that agriculture-wide or near-agriculture-wide production disasters are going to happen occasionally AND disaster payments will be made, why don't we plan for such events, rare as they hopefully will be, as part of the farm bill? Farmers could be paid a percentage of a "target price" for production below a certain percentage of normal production caused by a widespread drought or other widespread event. The local FSA (Farm Service Agency) folks have the farm-level data and other information they would need to do the work. No sales commissions to pay either.

Daryll E. Ray holds the Blasingame Chair of Excellence in Agricultural Policy, Institute of Agriculture, University of Tennessee, and is the Director of the UT's Agricultural Policy Analysis Center. (865) 974-7407; Fax: (865) 974-7298; dray@utk.edu; http://www.agpolicy.org.

Reproduction Permission Granted with:
1) Full attribution to Daryll E. Ray and the Agricultural Policy Analysis Center, University of Tennessee, Knoxville, TN;
2) An email sent to hdschaffer@utk.edu indicating how often you intend on running Dr. Ray's column and your total circulation. Also, please send one copy of the first issue with Dr. Ray's column in it to Harwood Schaffer, Agricultural Policy Analysis Center, 310 Morgan Hall, Knoxville, TN 37996-4500.