that kind of expression of intent probably reflects also an indication that all those wishing to buy insurance should be eligible to purchase

it.

Senator WILLIAMS. When you get private companies acting in concert, or you may call it a pool, there are some people that raise their eyes and say: Is this not some kind of combination that might have serious questions of antitrust?

Mr. WOOD. We have explored that, Mr. Chairman, with the Department of Justice. My understanding of this is that, given the voluntaristic character and the provision for easy entry that I think will be part of the negotiations, at the present time this does not seem to raise the problem, and in the course of negotiations we can certainly make sure it does not happen.

Senator WILLIAMS. There are precedents for this kind of pool with private companies?

Mr. Wood. Yes. For instance, with the Atomic Energy Commission. Senator WILLIAMS. Is that in the record anywhere?

Mr. WOOD. Yes, it is. It is in my statement.

Mr. KAPLAN. Not only that, Mr. Chairman, but Mr. Wood indicated quite clearly in his statement that the pool would be open to both kinds of participation-one, risk takers, and also those who wanted to act as fiscal agents who did not want to take risks; and the participating members of the company would assume the risk for such business as might be produced by fiscal agents.

Senator WILLIAMS. On the criteria that will be used to establish flood-plain regulation findings, HUD-Housing and Urban Development-will use as an agent the Army Corps of Engineers, is that right?

Mr. KAPLAN. Or the Geological Survey or the TVA.

Mr. WOOD. Whichever area is appropriate for the type of inundation.

Senator WILLIAMS. Will the agencies forward the findings on the flood-plain areas?

Mr. Wood. Your method for determining both the damage, the frequency and getting the data for the formulas, as Mr. Kaplan presented in his charts, I think have now been thoroughly studied.

Mr. KAPLAN. We will be guided largely by the experts with hydrologic experience as to the criteria of whether an area is a special flood hazard or not, sir.

Senator WILLIAMS. I think it would be helpful for the members, that cannot be present at these hearings if the record does include this.

Mr. KAPLAN. I might say in this connection that in this area there is an extraordinary amount, in my humble judgment, of competence within the Federal Government on hydrologic problems. There are a number of agencies of the Government which have for many, many years developed a special and highly respected expertise in measuring inland and coastal flood problems.

Senator WILLIAMS. You have tabulated the number of dwellings that are in the different risk zones that have been analyzed?

Mr. Wood. That is correct.

Senator WILLIAMS. I have a table here that was handed to me. I think we will include that table in the record, also, I have two tables

81-080-67

showing the amounts that various agencies of the Federal Government have spent during the last 11 years to assist victims of flood disasters. (The tables referred to, follows:)

Estimates of dwellings in the 6 zones of flood frequency, 1966

Source: Friedman and Roy, "Simulation of Total Flood Loss Experience of Dwellings on Inland and Coastal Flood Plains" (designated app. H-5, Secretary's report on "Insurance and Other Programs for Financial Assistance to Flood Victims'). Office of Emergency Planning disaster aid under Public Law 875,

1 For the years 1962 through 1966, flood relief loans were made as follows: on homes, 4,015 loans for $16,521,000; on businesses, 2,276 loans for $63,737.000.

2 481 rural housing loans for $4,659,830 for the period June 1, 1964, through April 30, 1967.

Senator WILLIAMS. I reiterate what I said earlier. You gentlemen certainly have given this profound deliberation and given us a great deal to think about, and we appreciate it. We have only addressed ourselves to the bill that we consider the administration bill that I

introduced for myself and for 28 other Members. We have not dealt particularly with two other bills that are pending before the subcommittee, one of which I had introduced to get things moving on this subject. I had an idea in that bill that does not come into your bill, that there should be no profit until the Government got paid back a certain amount of the funds it had expended. I do not think we will go into that, because I know that you have put a lot more time and effort into industrywide discussions, which was an obligation under the original bill.

Mr. Wood. The net effect is probably the same.

Senator WILLIAMS. The what?

Mr. Wood. The net effect is probably the same.

Senator WILLIAMS. We have decided to zero in on S. 1985, our bill. Thank you very much.

Mr. WOOD. Thank you, sir,

(The following memorandums were subsequently submitted by the Department of Housing and Urban Development:)

TENTATIVE ESTIMATED COSTS OF THE FLOOD INSURANCE PROGRAM PROPOSED UNDER S. 1985

The cost to the Federal Government of the proposed flood insurance program in any one year can be broken down into two basic elements: (1) liability of the Government in the insurance program itself, essentially measured by the Government's share in payments of claims for losses, industry expenses, and reinsurance payments in years of heavy flood losses; and (2) the Government's own expenses of administering the program, which, in turn, can be broken down into (a) staffing and related expenses to assist the Secretary of Housing and Urban Development in carrying out program responsibilities and (b) the cost of the rate-making and other studies called for under the legislation.

1. BASIC INSURANCE COSTS

The inurance liability of the Government is directly related to the size of the program in any given year, and to the amount of flood damage which insured properties suffer in that year. These are, of course, a number of uncertainties which would make precise cost predictions in any given year difficult, including how many homes will be insured, in what zones will they be located, how many new homes will there be which are insured at full cost rates, how many floods there will be, and what damage will be caused.

It is, however, possible to make tentative estimates of insurance liability costs within reasonable limits. In the first year of operation of the program, for example, there could be some 40 areas in which insurance may be sold. Based on information contained in the Secretary's report on "Insurance and Other Programs For Financial Assistance to Flood Victims," perhaps 32,000 dwelling units might be insured in these 40 areas. If it is assumed that about 40 new areas would be added each year for the first four years of the program, this rate of growth permits the following estimate to be made regarding dwellings eligible for coverage in the first four years:

Based on an average coverage of $14,000 per dwelling unit for structure and contents, the amount of insurance coverage for these dwelling units might be estimated to be as follows:

Because of the actuarial soundness of rate data which will govern the proposed program, it can be assumed that over the long run, average annual net losses should be equal to the risk portion of the premium. On this basis, and on the basis of the coverage estimates above, losses could amount to perhaps $7,200,000 in the first year, and up to $33,200,000 in the fourth year.

If the Government's share of these losses, exclusive of reinsurance payments is, for example, based on a 75-25 percent relationship (reflecting the difference between true aggregate risk premiums and actual aggregate premiums paid) then the annual cost to the Government based on these estimates could range from $5.4 million in the first year to about $24.9 million in the fourth year. If the distribution is nearer to a 60-40 percent relationship costs would range from $4.3 million in the first year to about $19.9 million in the fourth year.

Estimates of costs for the Government's reinsurance payments to industry and for the Government's share of industry administrative expense for flood insurance cannot be made with any accuracy until negotiations between Government and industry, as provided for in the legislation, reach detail stage. However, reinsurance payments over a long period of years should be fully covered by reinsurance premiums paid to the Government by the industry.

If there were no significant floods in the first year, the flood insurance fund, instead of incurring net costs, would benefit from net credits arising from reinsurance premiums received. If, on the other hand, there were heavy floods in the first year to the extent where claims reached twice the level of total risk premium for the coverage-possible but unlikely-the Government's cost could reach levels as high as $11 million.

Similarly, for the fourth year of the program, if severe flooding was experienced to the extent of claims for damage equal to twice risk premium, the flood insurance fund might have to pay out about $50 million.

These examples are based on the estimates of growth of the program, average annual damage, and risk premium rates shown in the Secretary's report.

2. ESTIMATES OF ADMINISTRATIVE EXPENSES OF THE DEPARTMENT IN CARRYING OUT THE FLOOD INSURANCE PROGRAM

Expenses under this category include internal administrative expenses and studies related to rate-making and other responsibilities under the legislation. While internal administrative expenses would include such general classifications as general administration, legal, program, actuarial and fiscal, program organization, service and travel, these expenses cannot be with any degree of accuracy simply projected because the first year's operation of the program will involve a number of activities such as negotiating with the industry-which would not be typical of the program operation in later years. A judgment estimate as to internal administrative expenses for the first year of program operations would be $1.5 million, of which at least two-thirds would be for studies required to further the program.

While the program is designed to place essentially all operating functions with the insurance industry, the Department will be assuming responsibility for determining rates and rate data, auditing, and essentially assuming a supervisory role in the program.

3. STUDIES

Assuming that the insurance program will begin slowly, studies called for in sections 104, 301, and 402 of S. 1985 (necessary for the establishment of rates, for identifying flood plain areas with specific flood hazards, for establishing flood risk zones, for establishing criteria for land use management and studying the feasibility of insurance protection against natural disasters other than floods) might require approximately $1,000,000 for the first year of operation. A very large portion of the studies will be performed under agreement by other Government agencies. Approximately half of this estimated amount for studies might be required for mapping in the first year, and similar amounts for the four succeeding years.

Total estimated Department of Housing and Urban Development expenses for the first year of the flood insurance program can therefore be summarized as follows:

Administration, travel and equipment purchases_

Studies

Total

$500,000 1,000,000

1,500,000

FUNDING FOR DISASTER STUDIES UNDER SECTION 402

The Secretary of Housing and Urban Development was directed by Section 5 of the Hurricane Disaster Relief Act of 1965 (P.L. 89-339, November 8, 1965) to study alternate programs to provide disaster assistance to those suffering property losses in earthquates and to prepare a report to the President for submission to the Congress on such studies by October 1968.

Funds appropriated under P.L. 89-309 for studies of flood insurance and other disaster programs were used to finance a preliminary study on the feasibility of earthquake insurance by the Coast and Geodetic Survey, ESSA of the Department of Commerce.

The Congress in considering the Department of HUD's FY 1967 appropriation failed to provide the $600,000 requested to continue the earthquake insurance investigation.

The Department has received a report on the preliminary study of earthquake insurance feasibility and recommendations for the continuation of these studies required to prepare the report due in October 1968.

If adequate funds to continue the earthquake studies are not appropriated under existing authority, the authority under section 402 would be the basis for requesting earthquake study funds in the future.

Senator WILLIAMS. By the way, we do have two more witnesses, and it is 20 after 12. We have Mr. Frank E. Clarke from the Geological Survey, and Robert F. Shea from the American National Red Cross. I am willing to stay if the witnesses are. This does not put any mandate on Mr. Wood, Mr. Kaplan, Mr. Schechter, Congressman McGrath.

Mr. Clarke, I think, because we have no amplification, if you would come up here. Are you a geologist, Mr. Clarke?

Mr. CLARKE. No, sir. I am a hydrologist. I brought along my friend Walter Langbein, and he is a civil engineer and I am a chemical engineer. We are neither geologists.

Senator WILLIAMS. I was just going to be envious of you if you were hydrologists. It was the only course I got an A in college.

I am advised that our room tomorrow for the hearing will be 2221 in the New Senate Office Building. That is the Finance Committee's room, rather than room 342. We had better let all the committee know that.

Mr. Clarke?

STATEMENT OF FRANK E. CLARKE, GEOLOGICAL SURVEY, DEPARTMENT OF THE INTERIOR; ACCOMPANIED BY WALTER LANGBEIN

Mr. CLARKE. Mr. Chairman, members of the committee, I am pleased to present this brief statement on behalf of the Geological Survey of the Department of the Interior.

The results of the many existing studies of the feasibility of a program of insurance against flood hazards establish the fact that frequency of flood occurrence can be determined. This finding is based on the extensive basic data collected over the years by the Geological Survey of the Department of the Interior. The Geological Survey has available more than 80 years' records of the flood heights and flood discharges of the rivers of the country and has pioneered in the research needed to estimate the magnitude and frequency of floods anywhere in the United States.

The first definitive study by any Federal agency on flood frequency was published by the Geological Survey in 1936 as Water Supply