Statement Before the U.S. Senate Agriculture, Nutrition, and Forestry Committee

Jerry R. Skees

University of Kentucky

March 10, 1999

ABSTRACT

Chairman Lugar and members of the committee, I appreciate the opportunity to appear before you today to discuss the important topic of agricultural risk management. Having spent the better part of my career working in this area, I feel it is an extremely important dimension of the agricultural sector and merits your special attention. I remain dedicated to work that will offer US farmers more market-based risk management alternatives. I have attached a paper that provides a comprehensive analysis of several important issues. There are three central themes that I wish to focus on in my remarks: 1) policy intervention can not remove the risk from agriculture; 2) there are unintended consequences of many of the subsidies that have been added to the USDA portfolio of risk management alternatives; and 3) many opportunities for innovation in private sector risk management are being hampered by government regulation and subsidies.

Research shows that risk reduction innovations will be partially offset by subsequent risky behavior that returns the individual to pre-risk innovation levels of risk. In short, there is a level of risk that people are comfortable with accepting. The logical consequence of this is that it constrains your ability to engineer the risk out of agriculture with subsidized risk management alternatives. When farmers are provided subsidized risk reduction alternatives, to the extent that they properly evaluate the value of the transfer, they will move to higher levels of risk. For example, they will borrow more and grow more crops on riskier sites. Expected rates of failure will change only slightly You must ask yourself if your policies are reaching the desired goals.

There are clear unintended consequences that have emerged from the subsidized risk management programs. In 1999, farmers will pay only 25 cents per dollar of the total cost for crop and revenue insurance programs. Some of my farmer friends who have never purchased crop insurance are now telling me - "I can't afford not to buy some type of subsidized insurance". These are farmers who know how to manage risk by other means. As with any subsidy, such as the tobacco quota system, the benefits have been ultimately bid into land prices. About half of US crop land is not farmed by the owner. If agricultural subsidies were properly characterized as transfers from taxpayers to land owners, many of whom are not farmers, increasing land prices and hampering entry into farming, political support might diminish. Moreover, there are fundamental differences in providing subsidies for farm-level yield risk versus national price risk. Insurance risk subsidies, as they are currently designed, favor the highest risk regions, the highest risk farmers, and the farmers who are most likely to abuse the insurance programs. Further, the income transfers in insurance programs have increased output, putting downward pressure on price. The price declines hurt the regions that are most productive and the risk transfers help the regions that are most risky.

Despite my concerns, I believe US agriculture needs well-functioning risk-sharing markets. The challenge is how to facilitate these markets. Further challenges involve the need for protecting multiple year risk and a better understanding of how and why farmers may not recognize certain types of risk that occur very infrequently. It is my view that government subsidies have constrained innovation in the private markets. US agriculture is simply too diverse to expect basically the same approaches to work everywhere. Further, I believe that many farmers would use more market instruments, if they did not have access to subsidies. In the attached paper, I offer some approaches to motivate more true market-oriented innovation. Some things that appear to be innovative in today's world are more targeted at getting access to taxpayer dollars. Regulatory reforms are needed. Changing the subsidy structure for insurance products should also be considered.

Agricultural Risk Management Policies

Jerry R. Skees⁽¹⁾

University of Kentucky

Crop insurance reform is once again on the agenda of agricultural policy makers. Some members of Congress want to add at least $1 billion in subsidies to a program that is currently budgeted at approximately $1.7 billion annually. Others want to introduce subsidized livestock insurance for the first time ever. The President is also on record calling for crop insurance reform in his 1999 State of the Union address. Everyone is talking about shoring up the safety net with more subsidized insurance.

For at least the past 20 years, every Secretary of Agriculture has had crop insurance reform as a major agenda item. However risk management took on new significance after the 1996 Farm Bill. The 1996 legislation was designed to end government price supports that were tied to product decisions in agriculture markets over a six-year period through a program of declining transition payments fixed in law. Farmer cropping decisions would no longer be influenced by government payments. As a result, farmers would need to learn to manage their own risk. The 'old game' of protecting subsidies that were tied to price movements appeared to be largely over. The 'new game', which sounded more market and less rent seeking oriented, allegedly was risk management accompanied by a limited 'safety net'.

The Federal multiple peril crop insurance program now insures much more than crop yields. After being introduced in 1996, gross revenue insurance grew quickly from regional experiments to national programs. Revenue insurance, insurance against declines of any combination of within year price declines times farm-level yields shortfalls, is now a major component of the federal crop insurance portfolio. Approximately 17% of the 1998 federal crop insurance premium was for gross revenue insurance programs. Some congressional members are calling for converting all yield insurance into revenue insurance. Others in the policy realm are calling for whole farm (multiple crop) revenue insurance. The US is now pilot testing revenue insurance for a basket of crops. A pilot program that uses income tax gross revenue is underway for 1999. New crops are being added every year. The nursery insurance program has been enhanced and expanded. When we insure woody ornamentals and fresh cut flowers using taxpayer subsidies, one may ask how this fits into a "food and fiber policy".

Before Congress moves forward with yet more subsidies in the name of risk management, it is important to take stock of the existing programs and their subsidies. What experience have we had? Are the risk management programs reaching their goals? Are there unintended consequences? To what extent might these programs be putting downward pressure on commodity prices? Do subsidized insurance programs affect production decisions? If risk management were the goal, how can we move toward improved risk management alternatives with fewer subsidies?

What are the Goals of the Risk Management Programs?

Agriculture is risky. Many variables contribute to the risk beyond yields and prices. Still, when yields are low or prices drop quickly farmers may go bankrupt if they do not have a willing lender to help them survive the shock. When farmers understand that they must maintain good relations with bankers so that they have ready access to credit during bad times, there is less likelihood that they will go out of business due to short-term shocks. When farmers go out of business, there is considerable media attention and a sympathetic American public. It is still politically popular to implement policies that will 'save the family farm'. The accountability of government programs to that end is rarely an issue.

Risks can and do create inefficiencies in markets. Having effective risk-sharing markets is important for improving the efficiency of the farm sector. Kenneth Arrow demonstrated long ago that society can gain from insurance (and other contingent claims) markets. When decision makers are risk averse, they are willing to give up some income to protect from future events that may cause them to lose large amounts of income. Well functioning risk-sharing markets allow firms to protect themselves from risk and pursue the advantages that come from becoming more specialized. Since Adam Smith, it has been a fundamental argument of economists that society gains from the specialization of firms.

Performance of US Agricultural Risk Management Programs

Subsidized Insurance Will Not Take the Risk Out of Agriculture

While there are good economic arguments that governments might be necessary to help the formation of risk-sharing markets, it is unlikely that either Adam Smith or Kenneth Arrow would advocate subsidies for risk-sharing instruments. When decision makers must pay for risk protection, at least what they expect to get back over the long term, then the risk will be internalized into the decision processes. When they pay less than they will get back in indemnities due to insurance subsidies, society is paying people to take on additional risk. The resources used in taking on those additional risks, by farmers for example, really cost the economy much more than just the income transfer imbedded in the subsidy. The additional cost becomes the inefficiency created by the subsidy.

Public sector subsidies to various markets often reflect the political and cultural desire to raise the incomes found in a particular sector in the economy to a level above that created by simple market forces. Such political rationales, however, reflect a profound lack of understanding about how subsidies affect the operations of markets. Insurance subsidies are no different and, may in fact, be worse.

A fundamental truth about markets is that "good deals" cannot persist, be they jobs with higher than expected wages or assets with greater than normal profits. Once markets become aware of such deals people compete for the good jobs and reduce the wage or compete for the desirable asset and bid up its price. When the government uses agricultural subsidies, the rewards to the land, labor, and capital involved in agriculture will be greater than they otherwise would be, but this "good deal" does not persist because people compete to get the "good deal."

Competition to get the "good deal" reduces wages, raises asset values, and lowers the prices of outputs from the subsidized sector just enough to offset the benefits of the subsidies. Further competition restores wages, profits, and land rents to their pre-subsidy levels. The effect of the subsidy is to reallocate factors from one part of the economy to another (to induce more land, labor, and capital to be devoted to farming than would occur under simple market forces), but it is the owner of the assets is who advantaged by the subsidy. Profits and wages are the same as before the subsidy. In addition to benefiting landowners, the subsidies also induce too much land, labor, and capital to be used in the subsidized sector, and too little in the other taxed sectors of the economy.

Under these conditions one must ask who is really being helped. About half the land that is farmed is farmed by a non-owner. If agricultural subsidies were properly characterized as transfers from taxpayers to land owners, many of whom are not farmers, raising land prices and making entry into farming more difficult for the small farmer, political support might diminish.

As decision makers recognize the value of crop insurance subsidies, the effects on risk-taking behavior become counterintuitive. Economists have demonstrated that attempts to force people to reduce risk simply cause people to take on more and different risks until their risk level returns to the pre-policy intervention level. Consider tractor accidents. Twenty years ago farmers often died or suffered serious injury when pinned beneath wayward tractors. Upon examining the problem, the engineering solution was straightforward -- a steel-reinforced bar over the driver's seat would prevent the tractor from crushing the driver. A few years after mandated roll bars on tractors were in place, the death rate and rate of serious injury remained roughly the same. People with roll bar-equipped tractors were driving harder, faster, and on steeper slopes, resulting in the same rate of fatal accidents as before the engineered response.

Analogously, farmers with subsidized risk management will push harder and faster taking on more risk. They will borrow at higher rates, reducing The likely result is the same failure rate as existed before the existence of the subsidies. For example, when farmers have access to subsidized crop insurance, their bankers will loan them more money. Loaning more will reduce the access to credit when there is a problem. With lower debts, lenders will be more likely to lend when there are temporary shocks. Again, such credit reserves are one way farmers manage risk. Subsidized crop insurance replaces many management practices that farmers use to manage risk.

In short, the public provision of subsidies that mitigate risk creates incentives for farmers to restructure their operation to produce roughly the same level of risk that existed prior to the subsidy. This is particularly true when decision makers fully recognize the value of the subsidy. The most common response may be to borrow more and expand production into either the riskier part of the farm or into new and more risky farms. The risk subsidy makes society neither more efficient nor more equitable. Farmers are at just as much risk as they were previously and the subsidy has been bid into asset values creating barriers to entry for new farmers.

The efficiency consequences of crop insurance subsidies may be even more perverse than the tractor safety story suggests. Crop insurance subsidies are greater on an absolute basis for those farmers whose premiums are greater because they are calculated as a percentage of the premium. Free disaster payments can be thought of as no different than a 100% subsidy on crop insurance. Subsidies that are a direct function of premium will transfer relatively more money to the highest risk farmers and the highest risk regions. Is this the reward structure we want for US agriculture? Even more troublesome may be the farmers who abuse the risk management programs. As will be developed below, risk management programs that are tied to individual farm yields are more easily abused than the traditional price programs that were tied to national prices.

Can the Congress Resist Giving Free Disaster Aid?

Since 1980, Congress has provided numerous 'fixes' for the federal crop insurance program - arguing that if more farmers were buying crop insurance, the Congress would not need to provide ad hoc disaster aid during years with major crop failures. While the first Federal crop insurance dates to 1938, by 1980 Congress abandoned a standing program of federal funding of crop-disaster assistance in favor of a subsidized crop insurance program. With the new crop insurance, farmers and the government would share in the cost of protecting against natural disasters through privately provided insurance. During the passage of the 1980 Crop Insurance Reform Act, the Congress explicitly set a goal of 50 percent participation on an acreage basis.

By 1988, only 25 percent of the acreage was insured. A major Midwest drought created significant hardships. Despite strong language in the 1980 act saying there would be no more disaster aid, Congress provided ad hoc disaster aid at a record level of nearly $4 billion. One reason many farmers do not purchase crop insurance is that they believe that when things get bad enough, the government will provide free insurance in the form of disaster payments. And the evidence, of course, suggests that these farmers are correct.

After the 1988 drought and ensuing ad hoc aid, calls were made to reform crop insurance. A Congressional Commission investigated the issue over a two-year period. Many administrative and legislative reforms were made in 1989 and 1990. Still, some form of ad hoc assistance was provided in every crop year between 1988 and 1994. In one year, the assistance was even made retroactive for a previous crop year.

In 1994 another major crop insurance reform was undertaken. This legislation added the average historical funding of ad hoc disaster assistance, approximately $1 billion, to the crop insurance program that was already costing nearly three quarters of a billion. Much of the $1 billion was used to provide catastrophic insurance (CAT) to farmers for no premium cost - farmers were only required to pay some of the administrative cost ($50 per crop). Some of the additional subsidy was also used to encourage participation in higher levels of coverage.

Even though the CAT insurance, in effect, was free disaster aid, it had potential to be subject to less abuse than disaster aid for several reasons. First, more 'underwriting' is involved and farmers are required to enroll early and provide some evidence that they actually are growing the crops and have some records to prove their yield potential. Second, Congress once again attempted to precommit not to fund disaster assistance. Budget constraints were imposed to add teeth to these commitments. Farmers who chose not to 'sign-up' for the CAT insurance policies were required to sign statements saying they understood they were waiving their rights to any future disaster assistance.

By the 1998 crop year all bets were off again. Serious drought in Texas was once again hurting the cotton crop. Cotton prices were flat. Persistent disease problems in the Red River Valley of North Dakota and Minnesota were hurting wheat. Wheat prices were down. The President and Secretary of Agriculture were basically saying 'no farmer should go out of business because of an act-of-God'.

By going 'off budget', Congress was able to pass an emergency aid packet last year that approached $6 billion. About $2.4 billion was disaster aid for crop losses. Some of those funds will be used to pay for losses in previous years. Another $400 million was allocated to reduce already subsidized premiums by another 30 percent for the 1999 crop year.

By continually breaking its commitment not to fund ad hoc disaster aid, Congress has undermined the incentive for farmers to pay for crop insurance. When the 1998 appropriations were added, Congress ignored the fact that many farmers had signed statements saying they would no longer be eligible for disaster aid. Education programs that were designed to help farmers understand that they would have to manage their own risk must look foolish to the savvy farmers who understood agricultural policy far better than their instructors. The central message that flows from Congressional action over the past two decades is that subsidies for taking risks increase, and when things get bad, Congress provides still more free aid.

The inability of our political system to resist ad hoc disaster subsidies is not confined to agriculture. Take flood insurance. People in Grand Forks, North Dakota had strong warnings that the Red River was going to flood in 1997. Still many did not purchase federal flood insurance. And sure enough, the government provided disaster payments. After major hurricanes and earthquakes, the government also provides assistance.

As a consequence of free governmental disaster programs, people are more likely to locate in disaster prone areas, which, in turn, creates more pain and more "need", which, in turn, requires more taxpayer payments.

Supporting anecdotal evidence comes from several examples. A recent issue of National Geographic mapped the locations of the US population relative to locations of major natural disaster risk. Over time people are moving to locations where disasters occur. One reason they do so is because they expect the government to bail them out when they suffer losses from a natural disaster.

A second example comes from our experience with the Mississippi River. At the turn of the Century we attempted to eliminate flood risk from the Mississippi River with levees and dams. People responded by residing and farming closer to the riverbanks. The 1993 flood was much worse than it would have been without our attempts to engineer solutions. Because the costs of river management and flood relief have been socialized and people don't have to internalize the cost of risky decisions they take on more risk. And when the disaster strikes, losses and the pressure for further assistance are greater than they were before the initial federal involvement.

Actuarial Performance and Taxpayer Costs

The loss experience for the federal crop insurance program has been so bad that Congress has nearly abandoned the goal of actuarial soundness. According to the new congressional definition, actuarial soundness exists as long as losses are not more than 107.5 percent of premiums (a loss ratio - indemnities divided by premiums - of 1.075). Even with a built-in subsidy of 7 percent, a number of regions have not yet reached the 1.075 loss-ratio target, meaning that even more unintended subsidies are present because of poor actuarial experience. When intended subsidies are taken from the total premium, the expected loss ratio was 1.57 in 1998 - the average farmer could expect to get $1.57 back over time for every $1.00 paid in premium. Of course, there are no average farmers - some expect to get much more and many less.

To be fair, recent national experience of the program has been very good. Since the 1993 floods, the national loss ratio has exceeded one in only one of the five years. However, no major disasters occurred in those years either. Some unpublished work at USDA suggests that the expected national loss ratio is now around 1.08. Again, some areas have expected loss ratios that are much higher.

Since costs are tied to how good or bad the crops are in any given year, actual costs of the risk management programs can vary greatly from year to year. When all sources of subsidies are summed, the expected taxpayer cost of the risk management programs is close to $1.7 billion. This $1.7 billion is distributed roughly as follows: 1) $130 million for the excess losses; 2) $950 million for intended farmer premium subsidies; 3) $400 million for delivery cost reimbursements to the crop insurance industry; 4) $160 million for crop insurance industry returns from the government for taking on risk; and 5) $63 million to USDA administrative costs.

Farmers paid $930 million in premiums in 1998 or about 35 cents on a dollar for the expected total cost of the program. When free disaster aid is added to taxpayer contributions, farmers paid about 20 cents on a dollar for the actual totals in 1998. For the 1999 crop year only, premiums will be reduced another 30 percent, an additional $400 million in subsidies. With the same participation as 1998, the total cost of the crop insurance program will exceed $2.5 billion in 1999. Farmers will pay about $650 million, 25 cents on a dollar.

Again, despite the current heavy burden on taxpayers, members of Congress and others are calling for still more subsidies - arguing that with higher subsidies more farmers will 'buy' crop insurance and that will reduce the likelihood that free disaster aid will be needed. This is a familiar argument. The more than doubling of subsidies that came with the 1994 reform did not prevent the 1998 ad hoc disaster aid. Are we destined to simply move our insurance programs to something that looks more like a multi-billion dollar standing disaster assistance program to which farmers pay even fewer pennies on the dollar?

Complexity of Program Constrains the Opportunities to Fix Problems

Federal multiple peril crop insurance is complex. Farmers can get paid for individual field losses that are as little as 15 percent below their proven average yield. In addition, they can get paid for quality problems, prevented planting, and replanting. With the advent of revenue products, they can receive payments when price times yield drops below a specified level. One of the revenue products will pay for lost bushels at the harvest price rather than the forecasted price when the insurance is sold. Many levels can be selected. Between 60 and 70 crops are now eligible for some type of subsidized crop insurance.

The complexity of the crop insurance programs increases the difficulty of considering sensible reforms. One objective in using the private sector to deliver federal crop insurance was to allow for more innovation. However, when the private companies do not bear the full responsibility for the risk there is a problem. Companies can be very creative in designing products that will increase sales when they do not have to take the full responsibility for the increased risk. Government oversight is needed.

Government oversight makes change difficult and increases the cost for private companies attempting to introduce new products. The government must approve insurance rates. The government must also approve new products, considering if they put the government at even more risk. Since it is a government program, it is rare that new products will be abandoned. When things go bad in a particular region or for a particular crop because of mistakes, abuse, or simply bad weather, just about the only correction processes available are rate increases or lowering coverage. Rates are based on 20 years of loss experience. Thus, any unanticipated losses in newly designed products will ultimately result in increased rates. In many cases, higher rates will not address the underlying reason for unanticipated losses.

To make matters worse, political intervention is pervasive. One should expect this since the program is funded and supported by Congress and all politics are local. Heavy pressures come from members of Congress to add uninsurable crops in their districts. When constituents complain about rule changes that are designed to fix some problems that are creating unanticipated losses, Congressional pressures prevail and the 'fix' is either never tried or abandoned. A recent example involved an attempt to classify the most grievous abusers who persistently collected on crop insurance policies, in good and bad years. After a few years this special program was dropped due to political pressures. All farmers who buy insurance in the same county as the bad performer must now pay more since their experience will be added to rates. This cycle repeats itself time and again.

As another example, consider the recent administrative decision to allow farmers who planted fall wheat to buy insurance this spring. Since the additional subsidy was only announced in January, to be fair the sales closing was extended into the spring. Many farmers who planted wheat in the fall may have already experienced weather that will undoubtedly create a loss. This is akin to allowing someone to buy insurance when their house is already on fire. There are many cases where sales closing dates were extended to increase participation. And while you may ask how the private companies who are at put at increased risk when these changes are made respond, rest assured that they are generally made whole. No one can fault the private companies for seeking political means to assure compensation under these circumstances.

Adverse Selection and Moral Hazard

Farmers always know more about their yield potential and risk than anyone from the outside (either the government or a private insurer). Such asymmetry in information creates the dual problems of adverse selection and moral hazard. Adverse selection occurs when there are problems in classifying risk of potential purchasers. Since farmers know the most about their potential yields, they will look at the insurance offer and decide if it is fair or, maybe even more than fair. Those who conclude that it is more than fair will buy. Those who conclude it is over priced for their risks will stay out.

In the federal crop insurance program, farmers must prove their yields. They are required to have records to back up their reported historic yields. It is not uncommon for them or the agents who sell to them to conveniently 'lose' some historical yields so as to make the offer yield higher. Further, as few as four years of records can be used to set the average yield for the contract. Any student of statistics knows that significant measurement error can exist with such a limited number of observations. These errors can work in favor of the farmer or against the farmer - compounding adverse selection.

While adverse selection potential is in motion before the purchase, moral hazard occurs after the purchase. Moral hazard occurs when farmers who are insured change their behavior in ways that increase the chances of loss beyond the insurance contract terms. Farmers having insurance may not spray for insects as quickly as the non-insured farmer. Or farmers may prove their yields from acreage with good soils and then plant to the poorer soils once they purchase insurance. Beyond the rational and totally understandable changes in behavior, some fraud also exists. It is not difficult to misrepresent realized yields.

Controlling adverse selection and moral hazard requires more information. Obtaining information is costly. Those costs are higher on a per unit basis when insured acreage is smaller. Information cost will also be greater when a new crop is insured or when few farmers grow an insured crop in a particular area. None of these considerations are built into the current crop insurance program. The reimbursement of insurance-company expenses by the government is on a flat-percentage-of-total-premium basis- no matter what crop or what part of the country. This too contributes to a willingness on the part of private companies to simply accept the fact that some problems will never be fixed. Fixing some of the problems would require more monitoring and more monitoring would require more reimbursement expenses than those currently provided.

Sharing Risk with the Government

By now you should be asking why the private insurers are not more diligent in pushing the government to improve the underwriting of the federal insurance program. This is a particularly curious question since the private sector now shares the risk on $1.6 billion of the $1.88 billion of premium that was sold in 1998. Obviously the private sector has been diligent in some regions and there are many parts of the country where the program is actuarially sound. But Congress wants universal availability of crop insurance. No farmer can be denied. No region can be eliminated.

How can the dual goals of high participation and actuarial soundness co-exist? Because the program is not actuarially sound in many parts of the country, how can a private sector with $1.6 billion at risk survive? The government offers a special reinsurance arrangement.

The Standard Reinsurance Agreement (SRA) is the heart of the Federal-private partnership in the Federal crop insurance program. The SRA allows the private companies to pick and chose the risks that they wish to retain. The essence of the SRA is that it allows the companies to pass the polices that are expected losers to the public sector and keep those policies that should make money in their portfolio. Through a complex set of quota shares and stop losses, the SRA shares risk across three different reinsurance funds with settlement on a state-by-state basis.

While the companies are certainly at significant risk, they and the international reinsurance community have learned to optimize the SRA in ways that provide expected profits even in areas where the expected indemnities exceed premiums.

The SRA illustrates classically the arguments of public choice economists about the design of public programs. There is limited transparency; there are considerable start-up costs to understand the SRA. Even the government, who designed the instrument, has limited understanding and no real analytical model to understand the SRA until only recently. Information is power and information on how to use the SRA enhances the power of the astute crop insurance companies. The balance of power is with the private companies.

Still, the private sector simply would not make crop insurance universally available without an arrangement like the SRA. The SRA, however, provides limited incentives for the private sector to help clean up the actuarial and underwriting problems that exist in some of the worst areas. Thus, in the worst performing states, it is little wonder that the private sector is generally willing to help farmers change the rules to make the program more attractive and increase participation, even though this may increase the aggregate loss ratio in those states. The private companies must make the program attractive to a large number of farmers to maintain political support. Is this the type of public-private relationship we want?

More on Unintended Consequences

Federal insurance programs are largely income transfer programs whose social objectives are unlikely to be fulfilled because of the logical response of market decision makers. As potential owners of farmland recognize the subsidies they will bid more for farmland. This creates barriers to entry for new farmers. While these unintended consequences are common with subsidies in agriculture, there are more serious unintended consequences when the subsidies are tied to farm-level yield risk. It is quite easy to convince a Congressional colleague to vote for a program that will help save farmers in his/her district. This is likely true even when the newly converted member of Congress is unaware that the very policies she/he is voting for will disadvantage farmers in her/his district. Policy is complex. Short versus long-term consequences are difficult to understand. To highlight some of these issues, let's consider how the subsidies have increased production, where this increase has occurred, and what this means for the aggregate price of some basic commodities.

Let's focus on the top six crops in acreage in the US - corn, wheat, soybeans, grain sorghum, cotton, and barley and compare net transfers with gross revenues for the years 1988 through 1992. These crops account for nearly 90 percent of the crop acreage in the US. For purposes of this discussion, net transfers in crop insurance and disaster payments are define as all payments made to farmers minus all payments made by farmers. Some 300 Crop Reporting Districts are examined across the US - less than 58 CRDs comprise over half of the acres for these six crops. There are net transfers of 10 cents per $1 of gross revenue in 20 crop reporting districts (CRD). In 4 districts, these transfers exceed 20 cents on a dollar. A similar examination of 1998 subsidies on crop insurance transfers gives the same pattern with similar transfer levels. Increasing insurance subsidies will compound these numbers.

The risk subsidies alter farmers' production decisions. Figure 1 compares the change in land use from the late 1970s to the late 1980s. Again, the top six crops are used to compare the shift in production. The national share of the acreage used for these crops and for major federal acreage retirement programs is calculated for two five--year time periods - 1978-82 and 1988-92. The base is the national total of the six crops plus the national total of federal acreage retirement programs. Thus, the ratio reflects the percent of the national total in each CRD. To reflect how these acreages change between the two time periods, the 1978-82 CRD share numbers are subtracted from the 1988-92. Positive numbers indicate a growth in national share for the CRD over the decade.

Figure 1: Gains and Losses in Crop Share for the Top Six US Crops Between 1988-92 and 1978-82.

1.

¹ A version of this paper will be published by the CATO Institute in their magazine Regulation.