mortgages, and I would say guaranteeing the building of new houses is not as serious a problem as taking care of people after they have lost their homes.

This is to be said of Senator Kennedy's bill-I am speaking quite frankly that it is a sort of an experimental bill. What he is really saying is, "We will have to start slowly in order to get anything of this sort through."

I will show you in a minute the figures that are involved. It may very well be that even a $500-million amount, then a billion, and then going on to a billion and a half may not be adequate even at the beginning.

But I certainly would agree that in terms of our lack of experience that we should not plunge into this thing completely at the beginning but see how the thing comes out and adjust year after year.

I am impressed by what has happened with the crop-insurance program. If you look at the last few lines of page 8, you will see in the first 5 years indemnities under the crop-insurance program as a percentage of premiums were 164, 151, 168, 149, and 182, but in the years 1950-53 the figures were 91, 112, 97, and 115.

This program went through the mill a number of times, and it improved each time, and I am hoping the same sort of thing could happen with our flood insurance.

Senator LEHMAN. May I point out for the record that the bill which I drafted and I think the bill drafted under Senator Bush's direction had a limit of $2 billion, not $500 million. The amount originally suggested I believe by the administration limited it to $500 million, and I believe yours does, Senator Kennedy.

Senator KENNEDY. Yes; that can be moved up at the discretion of the President.

Mr. HARRIS. There are two problems here. I would like to put this question to the distinguished Senators. This is a case of the man biting the dog. When you say a billion or a billion and a half or 2 billion, you mean this is how much the Government is interested, not how much they are obligated to pay. When you consider something like $900 billion or $1,000 billion of wealth in this country, if you look at the table at page 9 at the bottom, you see real estate is $349 billion, producers' durable goods $29 billion, consumers' durable goods $99 billion, and inventories $32 billion, you have a total right there of $509 billion. You can cut that some in terms of the property that is not in the flood-plain areas. But the amounts involved are hundreds of billions of dollars, and I wonder whether even a billion and a half insurance is going to make much of a dent on this.

You might try to limit the Government obligation some other way, but the coverage of insurance, I should think, would even at the beginning be more than a billion and a half or even more than $2 billion.

This is purely a personal position, and I know the two Senators, with much more political experience, have their eyes on what we can get by. I am looking at this from a somewhat more academic viewpoint.

Senator LEHMAN. Of course, the measure I have drafted permits the Government to write up to $2 billion flood insurance, flood-damage insurance. They can, of course, if authorized by the Congress, write a larger amount.

Mr. HARRIS. Yes.

Senator LEHMAN. But that is the limit without further legislative approval. But as far as the other items of manmade disaster insurance, the amount, of course, is vastly greater; I believe $10 billion.

Mr. HARRIS. I am going to finish pretty soon. I am coming in a minute to the problem of general-disaster insurance. I just wanted to point out that you can look at the problem in this way: You can say, for example, "We will insure only property in the flood-plain areas. This would be roughly $400 billion, according to the Hoover report, but I hope the committee would get more careful estimates. But if we estimate the total average is going to be $400 billion, this would mean a rate of 75 cents per $1,000.

If we assume we are going to cover all wealth, this would reduce the charge to 372 cents per $1,000.

If we assume that the Federal Government will pay part of the bill, and by that I mean they would pay probably no more than they would save through tax income because of the reduction of losses, and if you allow for how much disaster relief would be necessary, it is probably true that if the Government took about 25 percent of this cost plus administration this would get the charge down to 30 cents per $1,000, and this would roughly be $3. The limits would be $3 to $7.50 for the average $10,000 house.

That is one way of looking at the problem. As I say in paragraph 13, there is a great deal to be said for the Saltonstall-Kennedy

Senator BUSH. Mr. Harris, before you get on to paragraph 13, you say at the bottom of paragraph 12 that the ideal situation would be to add the payment to the general property tax. I do not quite understand that statement.

Mr. HARRIS. It is a tough one, and I don't know whether this would be practical or not. I think the first reaction would be, "No, this would not be practical."

Senator BUSH. I haven't any reaction, but I don't know what the general property tax is that you refer to.

Mr. HARRIS. The real-estate tax.

Senator BUSH. Those are all local taxes?

Mr. HARRIS. Those are local taxes, but frequently there is some kind of control by the State Government, and very often the State collects part of the tax. What I was hoping is that the program might be sold in such a way with rates adjusted in such a way with a sufficient Government indemnity or subsidy that each State would take a responsibility in adding a certain amount to the general property tax.

Actually, the general property tax yields about $10 billion a year. If $200 million, say, which would be about what would be required here, were added, this would be an increase in the general property tax of only 2 percent. If the present rate were $30, then the new rate would be $30.60.

Senator BUSH. That in effect would be the States' contribution? Mr. HARRIS. That is right.

Senator BUSH. That is based on the assumption all States would contribute?

Mr. HARRIS. The idea would be they would collect this from the local government.

Senator BUSH. Yes; but I mean it takes all the States into account?

Mr. HARRIS. That is right. The theory would be where the vulner ability was very small the rates would be very low. There would be some adjustment in this direction.

I just want to make one general remark about flood insurance and disaster insurance generally.

Senator, you recall in the 1951 hearings a bill was introduced which suggested the Federal Government put up a billion dollars for a general disaster program. That, of course, would include atomic bombing, all kinds of disasters. I gather you are also interested in this general problem.

There was a good deal of evidence on this, a good deal of dissatisfaction with various bills proposed, and finally the Government came across with a suggestion of its own, and this suggestion was that $1 billion would not anywhere near meet any likely damage done by even one atomic bomb in a big city and that, therefore, the wisest solution might be to say the Government ought to put up $20 billion and say, "We are ready to cover $20 billion worth of property." On some kind of basis for example, Senator Kennedy suggested some part of your loss would be coinsurance-some part of your loss would not be covered, and you might put a ceiling on the amount, and so forth.

But if you only allowed $1 billion, people might think they were getting much more than they were going to get. It would be better to announce you would make a certain amount of money available and when that amount was gone there was nothing else you could do so far as covering these losses.

The advantage of this kind of program is, Senator, that if you had this kind of universal disaster that the net results would be, for example, that the Government would have to feel free to take care of all kinds of emergency situations including loss of income.

I might say parenthetically-I meant to say this before-that when you are considering a flood insurance bill and talking about the losses that are involved, there is something to be said for taking account of income maintenance in this sense.

I think, for example, there ought to be provisions in unemployment insurance programs which make it possible to pay the workers who lose their jobs, as they have in New England through the flooding of the textile mills, so that instead of collecting, say, for 26 weeks, they could collect, say, for 15 or 20 weeks longer. This, of course, would be a State matter, but the Federal Government might make recommendations.

And the same way if we had, say, an atomic attack, you would want to be prepared to provide the minimum amount of subsistence for the people in the cities and so forth who have been damaged or who are in trouble. You would also want to be ready to deal with all kinds of emergency situations.

Therefore, it would be a great mistake in this kind of program for the Federal Government to try to cover anything more than a small percentage of all the property that is involved.

Twenty billion dollars is perhaps 4 percent or 5 percent of all the wealth of the country. We actually build each year about $40 billion worth of construction, so that the $20 billion program or something of that sort might not be bad.

I would also, finally, like to say that it would be a great mistake to collect large sums of insurance premiums to cover this kind of

disaster. We already have $100 billion tied up in insurance, pension, and retirement reserves. As a matter of fact, this was in 1949. It is larger now. It would be a great mistake to keep on piling up this kind of money, because it has certain serious effects on the economy.

Let me say in conclusion I think we ought to have a comprehensive disaster insurance program, that it should not be on a purely actuarial basis, that on grounds of equity there is a great deal to be said for having a subsidy program in part although the subsidy would probably be covered largely by savings on tax returns.

There should be wide coverage including unemployment insurance. The program should be related to other programs like disaster relief and the flood-control works and so forth.

Senator, I appreciate very much the opportunity to come here, and I hope I haven't been too long-winded.

(Mr. Harris' prepared statement follows:)

STATEMENT OF SEYMOUR E. HARRIS, CHAIRMAN, NEW ENGLAND GOVERNORS TEXTILE COMMITTEE AND CHAIRMAN, ECONOMICS DEPARTMENT, HARVARD UNIVERSITY

SUMMARY

[Numbers refer to paragraphs in statement]

1. Flood insurance results in an improved distribution of the costs of floods over time and place.

2. Costs of flood insurance would be reduced insofar as outlays on flood control, inclusive of rezoning and disaster relief are made available.

3. Private insurance is not practical.

4-5. The case for Government insurance rests in part on the fact that the Government can pay part of the costs out of savings on taxes otherwise lost and savings on disaster relief; partly on the fact that floods are a national disaster and the Government should require a sharing of the costs; and partly on the large reserves required by private companies when annual costs vary as much as 400 times, as they have over a period of 20 to 25 years. Rates based merely on vulnerability to floods would aggravate the migration of industry from the Northeast already stimulated by some Government policies.

6. Rates reflecting factors other than vulnerability to disaster (that is, rates below actuarial ones) are supported by the great contribution of the Northeast net to the Treasury. The Northeast has been the stepchild of the Treasury. Over a period of 18 years, 3 major Northern States received back from the United States Treasury one-third as much as 5 major Southern States.

7. New England has in recent years received from flood-control outlays about one-fifteenth of what might be expected on the basis of the size of her economy or needs.

8-9. The New England Governors' program for help on flood control is a minimum one. Despite inadequate outlays on flood control, the Hoover Commission has recommended great caution and larger relative contributions by State and local governments. But the financial position of these governments has deteriorated much more than that of the Federal Government since 1946.

10. Federal aid associated with recent floods is acknowledged; but aid is inadequate, especially for such peacetime industries as textiles. Hence an even greater need for insurance.

11. What kind of insurance? With experience, the bugs will be eliminated-as they have largely in crop insurance, a similar kind of insurance.

12-14. Coverage and rates.-The Saltonstall-Kennedy bill for flood insurance marks a large step forward. A more comprehensive proposal is the following. Depending upon whether all property values are covered ($800 billion), only real property, durable goods and inventories ($500 billion), property only in flood plains ($400 billion), and upon the proportion reasonably assessed upon the Government, the cost per $1,000 would average 30 to 75 cents.

15. Flood insurance and disasters generally.-Coverage should be limited, premiums should be restricted, and Government allowed a free hand.

FLOOD INSURANCE AND RELATED MATTERS

1. The advantage of flood insurance

The need of flood insurance arises because the damage done is large and insurance makes possible a wide distribution of costs both at one time and over time and thus reduces the sacrifices for those especially vulnerable.

2. Costs of insurance and related programs

Insofar as flood control is carried through, the damage would clearly be reduced and the costs of insurance cut. Insofar as the Government is prepared to meet part of the costs of floods through disaster relief or tax concessions, to that extent also the need of insurance is reduced. Insofar as zoning regulations or movements of population or property result in reduced hazards, to that extent the need of insurance or the costs are reduced.

3. Impracticability of private insurance

Private enterprise is not prepared to carry this kind of insurance, in part because of the difficulty of measuring vulnerability, in part because of the large proportion of those insured likely to collect damages (and hence the high actuarial rates, or adverse selectivity), and in part because of the need for balancing accounts over a short time under private insurance. Flood insurance, because of the large variation in costs, requires large reserves at the outset and balancing over long periods.

4. The case for Government insurance

The case for Government insurance rests on the following grounds:

The Government alone has adequate resources and can take a long view. Average costs of floods over 25 years have been $150 million annually-with a figure of $3 million in one year in the early thirties and in excess of a billion in one year. The estimate current now is $300 million per year in the future.

The Government can set rates which do not reflect purely actuarial considerations. A flood is a national disaster just like an atomic bomb, and the costs should not be borne disproportionately by those especially vulnerable. Hence Government insurance should be part insurance and part subsidy. But the subsidy would not necessarily be larger than the sum of savings on taxes and outlays for disaster otherwise payable by the Federal Government. Not only would losses uninsured result in reduced incomes, but imposition of actuarial rates would further cut investment and income and induce disinvestment with unfortunate effects on Treasury income. Finally, by offering subsidy rates the Government might well induce a wider coverage than would otherwise be possible. 5. Actuarial rates

Under the theory of classical economics, property in hazardous spots should be charged high rates so that the inducement to seek safer locations would be increased. But some reservations must be made to this position. Property owners in older industrial areas already operate under large handicaps, and in part these spring from Federal policies. An actuarial rate to cover potential flood damage might often be the final blow. Some variaitons in rates according to risks should be allowed, but the range of rates should be restricted. It might even be appropriate to charge the full actuarial costs for property newly established in vulnerable places.

6. Actuarial rates and issues of equity

One of the strongest reasons for not charging full actuarial rates is that the northeastern area, especially vulnerable to floods in recent years, has been the stepchild of the Federal Government. To give but two examples. Over a recent period of almost 20 years, the 3 major industrial States in New England (Massachusetts, Connecticut, and Rhode Island) received back but one-third as much relative to what was paid into the Federal taxes as 5 major southern industrial States (North Carolina, South Carolina, Georgia, Alabama, and Tennessee). Surely, the Northeastern States can fairly claim some help from the Federal Government to meet part of the costs of disaster. Again, from 1934 to 1953, Massachusetts, Connecticut, and Rhode Island received back 60 cents for every dollar of Federal taxes borne (Connecticut only 40 cents). Similar trends are to be ohserved in tax concessions under accelerated tax amortization.