Plugging the gaps in flood-control policy.
Just as the waters receded, interest in and efforts for reform have ebbed as well. The badly needed corrections in the system will, history has taught us, remain largely unmade. Then, with the next great flood, very much the same cycle of hand-wringing, outpouring of sympathy and money, and the subsequent anguish about costs, irrationalities, and inequities will be repeated. It's too bad. The solutions are neither expensive nor technologically difficult. They take, however, the resolution of some sticky issues, firm and even-handed dealing in the execution, and, above all, more political backbone than is usually seen when flood-control decisions are typically made - during the flooding.
Assessing the damage
In his 1994 State of the Union address, President Clinton called the 1993 flood a "500-year flood," which led many people to believe that such a storm could happen only once every 500 years. That is simply not true. Such a storm could conceivably happen next summer. Our streams are dotted with United States Geological Survey (USGS) streamflow gaging stations (154 in the region of the Midwest flood alone) that have recorded the heights of flood waters for much of the century. Looking at the record of a given gage, hydrologists can determine the probability that a particular height (peak stage) or volume (discharge) would be reached by flood waters passing that location. Those probabilities are more commonly referred to as "frequency" events. Thus, if there is a one percent probability that a particular flood stage will occur in any year, the actual event will be referred to as a 100-year flood. The so-called 500-year flood has a 0.2 percent probability of occurring in any year, the 50-year flood has a 2 percent probability, and so on. Such terminology is misleading because the person once flooded out by the "100-year flood" sometimes figures, wrongly, that he or she is home-free for the next 99 years.
The 1993 flood was a 500-year flood at some gaging stations. But it was no more than a 100-year flood at many more, and flows were even below 100-year level at one-third of all the USGS stations in the flooded area. Hood waters didn't even reach the heights of the 50-year flood at four of the six stations on the Mississippi River itself. Obviously, there is a lot of room left in the basin, given the unpredictability of rainfall and weather patterns, for plenty of "great floods" in the future.
The 1993 flood was caused by a combination of events. These included unusually heavy summer rains; soils that were saturated by spring snowmelt and could not absorb the additional moisture; and bad timing that joined the largest flows from one part of the basin with heavy flows from other parts. The pattern was unusual in 1993, but nothing about it was unique. There are so many variations that could occur that there is no doubt that something similar will happen again.
The 1993 flood caused a lot of damage, as floods always do. A federal task force report put estimated flood losses at $12 billion; the National Weather Service's best guess was $16 billion. It was generally agreed that the majority of the damage was agricultural and that a good portion of those crops were destroyed (or their planting was prevented) by soggy soil conditions rather than by "overbank" flooding. The Federal Emergency Management Agency (FEMA) estimated that about half of the 100,000 damaged homes were affected by something other than overbank flooding, such as backed-up sewers and seepage through walls and cracks due to high water tables. Whatever the source, thousands of people suffered, and millions of federal taxpayers paid.
The "flood relief" bill passed in August 1993 provided almost $6 billion in aid to the flood victims. At least three quarters of that money went for disaster aid, doled out by FEMA to private individuals and local governments and by the U.S. Department of Agriculture (USDA) to farmers in the forms of disaster aid and heavily subsidized federal crop insurance. Smaller amounts of money were also distributed by more than 20 other agencies, including the National Flood Insurance Program. Aid from other federal sources over the following 12 months, including appropriations tacked on to the Los Angeles earthquake relief bill, have raised the taxpayers' contribution to perhaps as high as $8 billion - a lot of money, particularly since it is exempt from congressional pay-as-you-go requirements and thus adds to the federal debt.
In the immediate aftermath of the flood, politicians fell over one another to express sympathy and provide compensation to flood victims. Yet once the nightly drama of human suffering had faded from our TV screens, those personal complaints and criticisms of flood-control policy that lie dormant during dry spells began to surface. These criticisms were expressed in congressional hearings and in a flurry of poorly thought-out legislative proposals. They also yielded a small number of important legislative changes, the formation of a task force by the Clinton administration, and the initiation of a two-year U.S. Army Corps of Engineers study of the upper Mississippi basin where the flooding had occurred.
The findings of the Interagency Floodplain Management Review Committee established by the task force were released in June 1994. "Sharing the Challenge: Floodplain Management into the 21st Century" (more commonly referred to as the "Galloway Report" after General Gerald E. Galloway who headed up the effort) contains five conclusions, 35 recommendations, and 60 proposed actions, most of which have some merit. The report has something for everyone and avoids stepping on anyone's toes. It has not been officially approved, endorsed, or adopted by anyone, so that for now it represents nothing more than the efforts of the 31 staff members from participating agencies. Many of those agencies, however, have responded to the report's suggestions: working committees have been formed, legislation introduced and programs developed. The professionals within government who know what should be done will struggle onward, but their ability to make big changes is limited.
As for the Corps study, it is under way and will go on through at least half of 1995. The agency's critics predict that because the Corps builds the levees, it will see structures as the solution to every problem, but the scope of the study is much broader. The study is called an "assessment" because it will describe how the water levels and damages of 1993 would have been affected if different programs and policies had been in place. Although the study is not designed to propose solutions to the problems, it may provide a new way of looking at the issues.
Congress did pass two important pieces of "buyout" legislation soon after the flood. The first increased the proportion of its money that FEMA could use to remove damaged structures from the floodplain; the second established an Emergency Wetland Reserve Program that allows USDA to buy out farmland behind destroyed levees that are not economical to repair. Congress allocated $152 million dollars to the first program and $25 million to the second - two drops, so to speak, in the $6-billion disaster-relief bucket.
Two other congressional actions were intended to increase the assumption of risk by the property owners themselves. Revisions to the flood-insurance program will make it more difficult for people to avoid buying flood insurance and still receive disaster assistance. In addition, surcharges can now be imposed on properties that flood repeatedly. Changes in the Federal Crop Insurance Program will also deny farmers disaster aid unless they first purchase a minimum amount of crop insurance. However, the government will pay the premium for the farmer, whose only cost will be a $50-per-crop handling fee.
The flood's lessons
Some of these first steps toward reform were important, and all would move flood-control policy in the right direction. But 1994 was relatively dry and the momentum generated by the flood has slowed. By the time the next big one hits, the lessons may have to be learned again. Four major lessons need to be understood.
Floods produce too much damage, and no matter what we do, the damages continue to grow. As far as we know, the 1993 flood was the most costly in history. The Weather Service's best-guess historical record, beginning in 1903, shows that although flood damages vary dramatically from year to year, the 30-year averages of damages, in equalized dollars, are steadily increasing. Annual damages in the first 30 years of record averaged $1.4 billion per year; in the second 30 years, $2.4 billion per year; and in the 30 years ending in 1993, $3.4 billion per year, about a 250 percent increase.
The weather hasn't changed much over that 90-year time span but everything else in the upper Mississippi River basin has. Certainly the character of the overall basin has changed. Agricultural (destruction of forests and cultivation of fields) and urban (replacing natural surfaces with asphalt and concrete) economic activity make floods worse by reducing absorbing surfaces and accelerating runoff. In pre-Colonial times, millions of acres of wetlands and thousands of beaver dams controlled the hydrology and held the water far longer in the upper watershed, releasing it to the great rivers at a far slower rate than that of today. And it is argued, with some evidence, that the levees that confine the waters of midwestern rivers in narrow channels actually raise the water heights and cause increased flooding downstream.
But although modern economic activity in the basin has certainly worsened the flooding, it has also brought ample economic benefits. By its very nature, the floodplain of the upper Mississippi, flattened by the ebb and flow of the river and rich with fertile deposits from countless floods, makes ideal seedbed for corn, soybeans, and other midwestern crops. Flood-control works have transformed the basin into an economically productive area. This land is protected by levees, built first by the thousands of drainage districts organized at the end of the past century and the beginning of this one, and later by the Corps, whose work was mandated by a series of expensive and wide-ranging congressional actions.
The assumption behind these efforts has always been that flood damages could and would be reduced, but that is not how it worked out. Yet, as the Corps points out, the damages prevented by the projects are increasing as well. The Corps estimated that in 1993 its projects had prevented $19 billion in damages. Does that mean that without those projects damages would have totaled an additional $19 billion? Not at all. Without the levee projects the damages would have been far less because no economic development would have taken place on the land.
The real problem is how much damageable activities have increased, and the main reason they increase is because we build protection structures in the first place. The mechanism is simple. Take, for example, a piece of land that floods whenever 10-year-frequency river crests occur. No one farms there and so those floods produce no damages. The landowners then band together and get the Corps to build a levee high enough to protect the land from a 50-year flood. Farmers rush in to plant, laying down seed, fertilizer, and pesticides. They build storage bins and install dryers and join together to build roads and cut drainage ditches. In the first year, flood crests reach only the 25-year flood level, and all that new farm income becomes "prevented" damages. Then along comes the 100-year flood. The levee blows, the pumping station floods, ditches erode, roads wash out, storage bins shatter, and equipment disintegrates. All those investments translate into millions of dollars of damages. The trade-off was farm production for flood damage; when the first is expanded, the second is increased as well.
The same scenario is often acted out in urban areas. Building a dam or reservoir results in some land that was wet becoming dry, and someone builds housing or opens a trailer park. It's not prime real estate but it's usually cheap, and the people who move in can probably least afford the loss when that one big flood overtops the level of protection. And because the levee satisfied the National Flood Insurance Program's level-of-protection requirements, the properties are treated as if they were at no risk from flood.
"Level of protection" is a key concept because it is driven by economics. A project is built because its annualized benefits derived from the protected land are greater than annualized costs of the construction of protection structures. It makes no sense, the theory goes, to spend more on protection than the value of what is being protected. So far, so good. The economically justified level of protection may be the 50-year flood. The land is expected to flood when the 100-year flood hits, because it is not worth protecting at that level; in fact, it's supposed to flood then, the computer might tell you, to provide the needed extra flood storage area. But try telling that to the landowner who is out there sandbagging. And once the levy has burst, the property owner expects, demands, and receives disaster aid to compensate for losses: federally subsidized flood insurance; ad hoc disaster assistance; Corps repair of the levees; federal funds to repair pumping stations, roads, and ditches and to remove sand and silt; and local property-tax relief for several years. All that assistance, a regular occurrence in certain levee districts, produces costs that never would have been incurred without the levee. Yet they are never factored into calculations of cost/benefit ratios. If they had been, the levee might not have been built at all.
New floodplain development is very expensive to the taxpayer and should be curtailed. In 1968, the federal government launched the National Flood Insurance Program to attack the problem of floodplain development, at least in urban areas. Federally subsidized flood insurance was provided to floodplain property owners but only where local zoning laws restricted new development. The idea was to give local governments incentives to restrict new building in flood areas and to help people already living there. Unfortunately, the legislated mechanisms for requiring insurance are weak, administratively awkward, and difficult to implement. Its role in disaster-damage reduction is insignificant.
The two new buyout programs adopted after the 1993 flood are part of the new federal emphasis on "mitigation"-activities that reduce or remove flood damages themselves. By combining its emergency allotment of $150 million with some $500 million from special Housing and Urban Development (HUD) Community Development Block Grants, FEMA has organized a promising new program to buy 8,000 structures, which is 15 percent of those that were heavily damaged along the riverbanks. The property will be turned over to the local community, which promises to never develop it again. FEMA also held hearings across the Midwest in the fall of 1994 in order to gather suggestions for a new mitigation strategy.
The second program, the Emergency Wetland Reserve Program, expects to spend about $50 million in 1995 to buy permanent easements on damaged farm lands that are too expensive to repair. Although the money is adequate, farmers are becoming less willing to sell their land as flood memories fade and fields dry out. The program will be lucky if it takes 45,000 acres out of production.
The success of both these programs depends upon the willingness of property owners to give up their economic rights, a state of mind more prevalent amid the stink of muddy fields and basements than in later drier times, particularly if one's pockets have in the interim been filled with generous disaster checks. There's the rub. Disaster assistance, distributed generously to farm and residential property owners, has weakened the incentive to move from the floodplain. The economic value of properties has always been maintained, flood or no flood, by the generosity of the federal government.
The many (taxpayers) pay too much to benefit the few who freely choose to live, work, and farm in flood-prone areas. Flooding can be good business. The rest of the country might be more inclined to stand up for the freedom of choice and property rights of floodplain owners if the latter took some of the responsibility for those same choices. But floodplain owners gamble, and the rest of us pay when the dice roll against them. Many property owners don't bother to buy insurance; those that do are heavily subsidized by the government; and a small number of property owners make a killing from government programs.
At the time of the 1993 flood, only 20 percent of the eligible residential structures were covered by the National Flood Insurance Program, and 57 percent of the eligible acreage was covered by the Federal Crop Insurance Program. Although the program for residential structures has been striving for financial independence, 40 percent of the policies are still subsidized, with taxpayers paying three dollars for every one provided by the crop-insurance policyholder.
In each insurance program, the people who benefit most are those who take the greatest risks. It has been estimated that 2 percent of the flood-insurance policies, typically repeat victims, account for 25 percent of the claims. In the agricultural disaster programs, one-half of one percent of the recipients received a startling 9 percent of the money, with each recipient paid between $100,000 and $250,000.
The low rates of participation in both insurance programs is easily explained. Despite the subsidies, farmers and homeowners consider premiums too high and coverage too limited. However, they would probably see it differently if the alternative were no disaster payments. But as long as government is willing to provide disaster relief for everyone, there is little incentive to pay for insurance.
The flood-control program leads to substantial environmental damage. As flood flows have been constrained and confined by structural devices, the valuable aquatic ecosystems that once lined our rivers have been destroyed. The millions of acres of cropland protected by agricultural levees today in the Midwest represent lost wetland areas. Any flood-prone portion returned to natural conditions by the elimination of damageable activities would generate environmental benefits. Wetlands provide the habitat and breeding grounds for a rich variety of plant and animal species. They cleanse the water as well, filtering out pollutants and sifting sediments. In times of heavy floods, they provide broad storage areas from which the waters gradually move back into the river channels. These benefits, long recognized, have been lost with the destruction of almost 60 percent (26 million acres) of midwestern wetlands in the past 200 years.
With the recognition of the value of wetlands, our national priorities have changed. Laws have halted new destruction. Wetland protection policies are now so stringent that they can prevent someone from filling an acre here or there to make a home more functional or facilitate farming operations. A 1992 National Research Council report recommended a strategy that would eventually restore the entire 117 million acres lost nationwide. Yet no federal program is focused on wetland restoration. USDA's Wetlands Reserve Program provides benefits to the farmer but results in little genuine wetland restoration. We have not even considered making a major effort to restore the thousands of acres of those rich ecosystems where they are needed most: along our rivers. We prefer instead to prop up development so marginal that it requires repeated bailout by the taxpayers. It makes no sense.
Simply understanding why our flood-control policies have failed will not be enough to forge real change. We must also tackle the political issues and policy conflicts that underpin our failing policies. Three issues in particular must be confronted.
Farm policy. The crop insurance and disaster relief programs for farmers are but a part of the overall farm program, which has battled the price-depressing effects of farm surpluses for much of this century in an attempt to protect and sustain our agricultural economy as well as to preserve noneconomic values such as the rural way of life. Farm programs thus subsidize activities that would not survive in the marketplace: They reward farmers for setting aside millions of acres of cropland when surpluses build up. The same farmer who received crop insurance and disaster payments from the 1993 flood would also have received assistance if a drought had reduced yield or if conditions had been so good that high yields reduced prices dramatically. In any case, the excessive generosity of the crop insurance and disaster relief programs are only part of the complexities of the total farm policy problem.
The "takings" issue. If farmers didn't hate environmentalists before the flood, they probably do now - and with some cause. Strident talk about allowing their income-producing property to revert to habitat for plants and animals adds insult to the physical, financial, and emotional injury farmers suffered in the summer of 1993. A solution that reduces damages by removing cropland from the floodplain means disrupting peoples' lives and taking land. It will not be done without a costly and politically contentious fight. Although a farmer may not choose to forfeit economic value for wildlife habitat, in cases where that value derives from diminishing subsidies, he may be willing to sell at a reasonable price. No workable solution will be reached unless all sides are allowed to participate in floodplain-usage negotiations.
Takings policy also affects flooded homeowners. People prevented from building (or rebuilding) or from living on the floodplain are likely to perceive that they have been injured, particularly if they believe themselves to be the only disaster victims ever asked to change their ways. They are much more likely to be willing to give up their land if they are simply required to shoulder the risks of flood damage on their own.
Disaster relief policy. Until disaster relief of all kinds is administered equitably, flood-policy reform will not be possible. In areas where the probabilities of natural disaster are known to be above average, bad decisions should not be rewarded by disaster relief unless that aid is tied to actions that would mitigate future damages. The country cannot continue to cover the losses of people who heedlessly continue to live on the banks of the Mississippi or to put flimsy roofs on their Florida homes or to build or purchase structures in Los Angeles that do not meet earthquake-protection standards.
Above all, we must seek to restore aquatic environments to those portions of the floodplains where economic activities proceed today at greatest risk. The fact that excess production is such a persistent problem for U.S. farmers should reinforce the rationale for removing cropland from the floodplain. In the midwestern states, for example, the amount of farmland set aside in 1992 was the same as FEMA's estimate of the flood-prone land in those states: between 25 and 30 million acres. There was no set-aside requirement in 1994, but 1994 produced the largest corn and soybean crops in history, and set-asides for 1995 will be 7.5 percent. It's hard to argue that we need to farm the floodplain to keep up production.
Removing substantial portions of midwestern floodplain from production and restoring them to riverine wetland would appear to be a win-win national strategy: agricultural production reduced, wetlands restored. Restoring 2 million acres (less than I percent of total cropland and 10 percent of flood hazard areas) would be a nice first step - one that could have provided storage area for as much as 15 percent of the overbank floodwaters in 1993, reducing damages elsewhere. Restoring 2 million acres would cost about $2 billion, based on the typical $ 1,000-per-acre value of land in the floodplain.
How do we get there? It could be done in one easy step by eliminating disaster aid. Without federal help, property owners, forced to make rational economic decisions, would abandon the floodplain in droves. Although straightforward, such drastic action would be unfair to those who have geared their behavior and choices to existing policy. The strategy must be more like a gradual drawdown. Taking the following five steps would go a long way toward resolution of the problem.
First, steadily reduce ad hoc disaster aid, establish a means test for distributing it, and use the savings to mitigate damages and restore wetlands. Although the Galloway report says little about the overall levels of disaster assistance, it makes a number of proposals for strengthening the federal ability to acquire and restore floodplain. It suggests that buyout and mitigation programs be established and funded independently of disaster declarations, that land purchases and restoration be coordinated by a lead agency, that the mechanisms for land acquisition be streamlined, that an ecosystem management demonstration project be selected within the upper Mississippi River basin, and that further evaluation be made of the impact of wetlands on river flooding.
Second, incorporate disaster funding into the annual budget process. Unless we do this, even a gradual reduction in disaster relief will be difficult, because when the next big flood occurs, politicians of all stripes will again throw reason to the wind, violate principles of fairness, overturn program decisions, and reward the undeserving to prove that they care the most. During negotiations over the 1993 relief bill, the administration increased its initial $2.2-billion request to $4.8 billion after the Senate proposed $2.8 billion. Congress finally settled on $5.7 billion. Yet even this was small potatoes compared to the $12-billion Los Angeles earthquake bill which funded a laundry list of programs including more money for the Midwest disaster, Kuwait cleanup, widows' pensions, and the U.S. trade representative's travel expenses.
Third, tighten and toughen the flood-insurance and crop-insurance programs, gradually make them pay for themselves, and turn them into the exclusive source of aid to eligible victims. The Galloway report recommends that FEMA reduce disaster aid to those who failed to buy flood insurance and suggests a number of small measures by which FEMA might increase participation and reduce the costs of the flood-insurance program. It urges Congress to put restrictions on public assistance grants to communities not in the flood insurance program, to reduce disaster aid for those who didn't buy flood insurance, to strengthen the mandatory provisions of the flood insurance program, and to reform the Federal crop insurance program. The new flood insurance reform legislation makes many of these improvements to the flood insurance program. As for the Crop Insurance Program, until a premium structure requires farmers to share the risk, it may as well be disaster assistance. In fact, in a report on deficit reduction, the Congressional Budget Office recommends that crop insurance be dropped in favor of direct aid. As things stand today, they're practically the same thing.
Four, limit new structural protection to critical facilities and reexamine the economic justification for repairing existing ones. The Galloway report recommends that the Office of Management and Budget prohibit the use of future damages in calculating cost/benefit ratios and that the Corps investigate procedures to minimize impacts associated with levee overtoppings and come up with appropriate criteria for evaluating the economics of levee repairs. The Corps' assessment of the upper Mississippi basin can be expected to deal with all of those issues and more.
Finally, even though a large amount of flood-related research has been done, more and even better data are still needed. It is obviously much easier to resolve political conflicts when good information is available. The Galloway report makes two key suggestions. It calls for a full accounting, funded by the National Science Foundation, of all public and private benefits and costs of floodplain occupancy and associated floodplain management measures, including monetary and nonmonetary methods of accounting. In addition, the report recommends that USDA evaluate the impact of federal farm programs on agricultural land-use decisions in and out of the floodplain.
Beyond this set of proposals, the Galloway report contains numerous suggestions for planning and operational improvements; all it lacks is focus. There is much that can be done. But it will take a lot of political courage - and perhaps another flood.
National Research Council, Restoration of Aquatic Ecosystems, Washington, D.C.: National Academy Press, 1992. "Sharing the Challenge: Floodplain Management into the 21 st Century." Washington, D.C.: Report of the Interagency Floodplain Management Review Committee, 1994. W.L. Hoffman, C. Campbell, and K. Cook, Sowing Disaster. Washington, D.C.: Environmental Working Group, 1994. D.L. Hey and N.S. Philippi, Reinventing a Flood Control Strategy. Chicago, Ill.: The Wetlands Initiative, 1994.
|Printer friendly Cite/link Email Feedback|
|Publication:||Issues in Science and Technology|
|Date:||Dec 22, 1994|
|Previous Article:||Star wars redux.|
|Next Article:||The Bell Curve: Intelligence and Class Structure in American Life.|