channels to be the price-sustaining influence it should be. The other view is that the building up of reserves in a succession of good crop years will create such a large reserve as to form a constant threat to the market. A few additional remarks on the subject, therefore, may be in place that this time.

It is a foregone conclusion, as several witnesses have pointed out, that if more wheat is produced than the domestic and foreign markets can consume, that the carry-over of wheat will be built up. That will be true whether there is crop insurance or whether there isn't. That is a production and marketing problem to be taken care of in any complete agricultural program. The question with respect to grain reserves in connection with crop insurance is whether they will add to or subtract from the difficulties.

It is first to be observed that the premiums paid in any year will be partly consumed in the payment of the losses of that year. Only the balance will be available for the purchase of reserve stocks of grain. Senator POPE. It would be expected in any year, even with good crops there would be a certain amount of indemnities to be paid.

Mr. GREEN. Yes, that would be true, and even though there are areas like Illinois and Ohio where they have steady production, they will not be so much interested in insurance and in losses, yet, if you refer to the seed loan record you will find that practically every State in the Nation, including Rhode Island 1 year, had some seed loans at least 5 years out of the last 11 or 12. So that even in the good States there was some losses.

I believe it was 1928 when Illinois had a winter kill to the extent of 5 or 6 percent in some of their counties.

The heaviest premiums will be paid in those areas where annual losses average the largest. It should be the policy to maintain grain reserves as near areas of probable loss as practical and in such quantities as loss experience indicates may be necessary. Increased premiums from any increased acreage will merely offset increased insurance liabilities. Partial or only a percentage coverage, and coordination with the soil conservation program further restrict possible building up of reserves.

It therefore appears that the chief danger is from that portion of production that is for sale, and from acreage outside the insurance plan rather than from the insurance reserves. The premiums used in building up reserves are taken in at the market and do not therefore obstruct the flow of wheat into consumption except to the degree that the removal of a part of supplies from trade channels in good crop years raises prices. Even here the competition for supplies is only that necessary to cover most probable future crop losses. Only more good crop years in succession than usual and fewer bad crop years would build reserves in excess of loss liabilities. This would mean premium rates were too high and should be lowered somewhat.

Now with regard to the second viewpoint, which is concerned mainly with the disposal of reserve stocks of grain. To the extent that premium rates are accurate, to that extent, there was a predetermined outlet for the wheat stocks when they were put there, namely, to meet future crop losses.

If, as suggested, the policy of maintaining reserve stocks of grain near the area of production is followed, the price effect of disposal of stocks in short crop years will be absorbed partly in market differential

adjustments between local and terminal points rather than completely as a change in the basic price.

Now, what is meant by that, I can point out by this example. As an example, disposal of wheat stocks held at Dodge City, Kans., and you are going to find higher rates in southwestern Kansas, if you follow the plan of storing those stocks as near points of production as possible. You would expect in Kansas a fairly sizeable share of your stocks to be Dodge City, Hutchinson, and Wichita, that territory to which southwestern Kansas is tributary. But to repeat, as an example, disposal of wheat stocks held at Dodge City, Kans., in a year of subnormal wheat crops in that area would tend to put the local price at nearer 12 cents under Kansas City, the normal shipping difference, rather than leave it at only 6 cents under or less than the shipping difference as is the case now in years of short crops locally.

Farmers with insurance would have something to sell at this price whereas, now even though the local price is high, most farmers have little or nothing to sell in the bad crop years. Such price stability would be in line with progress that improved transportation, financing, storage, and other instrumentalities have made possible toward lessening the disturbing effects of local surpluses and shortages. I think the point to be considered even in those short crop years when you are disposing of that surplus you will be disposing of it in a territory where the loss was high and a lot of price increase will be observed by charging the differential rather than completely depressing the central market or basic price and that has the effect of keeping that local point nearer the steady relationship to the central market. It does not go up in position with 1 year as normal shipping, and at the end of the crop year, a similar shipping distance and therefore the mills or industries in that section as well as the farmer has to put up with that fluctuation due largely to local surpluses and shortages.

Furthermore, a large share of the wheat now carried over from one crop year to the next is mixed to a grade deliverable on future contracts, and is hedged in these futures. Bankers require that before they will permit the wheat to be carried over from one year to another, it is mixed down to the contract grade so it can be hedged. Otherwise, they are part of the losses that might be impaired by carrying wheat 2 percent premium over the future and in 3 months it may be down to the future or not.

Financing requirements that this wheat carried over be hedged leads to this situation. The insurance reserve stocks of train could be kept much more suitable for the mill trade. This would apply particularly in the spring-wheat and hard-winter wheat areas. Soft-wheat areas to the east and northwest would gain from having some of the wheat in the spring- and hard-winter wheat areas removed from the market in good crop years. When there is a good crop over the spring- and hardwinter wheat areas, it usually means a large total crop. In years of short crops the shortage is most extensive in the spring and hard winter wheat areas as a general rule. Reduction of storage stocks in those areas to pay losses there would tend to reduce differentials between the different wheat areas to a greater extent than it would act as a general price-depressing influence.

Considering then what has been said and the fact that there is no overbidding the market to induce supplies to be dumped onto the insurance corporation, I am inclined to the first-mentioned view, that the major problem, as far as the insurance program alone is concerned, will be getting volume of wheat stocks large enough in good crop years

to lend price support. It will be other carry-over stocks of wheat with no predetermined outlet that will constitute the greatest disposal problem, and not the insurance reserves.

Now, there is one other point brought up this afternoon that is not touched upon in the bill or the committee's report. It is left to regulations, but since we have been giving it some thought, I judge you would be interested in what is being attempted for recommendation, anyway. That is, with regard to grades in which these premiums or indemnities are to be paid. We have discussed with several grainmen, discussed with Mr. Thatcher because of his experience in grains, the problems of the past and the most practical way to go at the problem, and we are inclined to the view, at the present, that in each area certain basic grades should be designated in our winter-wheat area.

I will mention that as an example because I know more what I am talking about when I talk about it. It would be number two hard winter. We would say to the farmer, "Your premium is one bushel per acre, and that means one bushel of no. 2 hard winter wheat. Now, if you bring in a load of no. 1, you can sell that and pay your debt with no. 2 hard winter wheat. When indemnities are paid, they will be paid on the basis, the same basis", and so in each area. If a man had an unusual good production of wheat, that is, the increase is based on basic grade, both as to payment of premiums and payments of indemnities.

I thought that might help to clear that up.

Senator MCGILL. Mr. Green, this is on an entirely different line from what you have just been testifying. I think by reason of some questions that have been asked outside of the committee room by different ones, it might be well to have something in the record about it. That is, how does the Department or the Bureau go about it to determine from year to year how much wheat is raised in a given territory, a given county, or we will say in a State. How do you know how much was produced?

Mr. GREEN. I will tell you; if you are interested in having that in the record, I think we can get some from the division of crop estimates. I am sure they can give you a better background.

How

Senator MCGILL. The reason I brought that up was this. does the Department know how much is raised in any State? Mr. GREEN. Of course, those figures are based on estimates made from crop reports.

Senator MCGILL. You are figuring on this insurance to be based on a year's performance?

Mr. GREEN. That is right.

Senator MCGILL. And, in order to start off it would seem, taking the base period as to how much the farmer produces, you would have to have the statistics?

Mr. GREEN. Yes, that is why we took that period 1930 to 1935. We, through your triple A program did get for that short span of years, records from 100 of those individual farms as to what they produced in 1930, 1931, 1932, 1933, 1934, and 1935. Now, it is true-or very likely true at least that 1933, 1934, and 1935 figures are a little more accurate than the earlier ones because there was some check-up measurements in the later years.

Senator POPE. You heard the testimony of Mr. Talbott today; in North Dakota they have some very accurate data on individual farm production?

Mr. GREEN. Yes, we have that.

Senator MCGILL. I think you will find that is true in some States and in other States they haven't any.

Mr. GREEN. Two States, North Dakota and Kansas, have the best data over a period of years. We have the data from North Dakota and Kansas.

Senator MCGILL. That is largely due to State law.

Mr. GREEN. That is right.

Senator MCGILL. And taking States that have no statutory requirements generally, this is done in estimating return, whether there is a tax based on it or not. We require a farmer to tell how many horses he has and how many colts, to keep a fairly accurate record.

Your figures based on production under the A. A. A. program, that would only be applicable to those farmers who entered into that program?

Mr. GREEN. Yes; that is about 78 percent of the acreage at that time. Senator MCGILL. I think we ought to have statistics in this record that will enable us to answer how the Department knows how much wheat is produced in a State.

Mr. GREEN. Well, that is the Division of Crop Estimates.

Mr. TALBOTT. Senator McGill-pardon me, Mr. Chairmanisn't it a fact that there are States without laws that have a cropreport system gathered more or less accurately for a good many years?

Senator MCGILL. I presume that is true, but the State in question made no record, so whatever information there is or any statistics as may be available, would have to be only such as are gathered by the Department of Agriculture.

Mr. GREEN. That has been one of the problems in this matter.

Senator POPE. Now, there are two questions, but before suggesting those I will say that we will get in touch with the Bureau of Crop Estimates, and I think we can have someone here tomorrow who can answer any questions with reference to that.

Mr. GREEN. They can give the details.

Senator POPE. There is one other question now, Mr. Green. With the data, quite complete data that you have from North Dakota and from Kansas with reference to individual farm production, does it occur to you that there would be very much difficulty in using a longer base period than we have perhaps contemplated in our discussions?

Mr. GREEN. We are doing that. That is why we got the North Dakota and Kansas data and the Division of Crop Estimates are now working it up for a number of other States, data that has never been worked up. We are at least going to get a 10-year period for all the States. They are doing some work now on data that has never been worked up, so we will have a 10-year basis. In fact, we have done some work on North Dakota and Kansas and taking the average figures for the 1930-35 period. We have triple A data on it, too, and comparing that for the 10-year period; and find in a number of cases, you have to step up your yields. There are some cases where the 1930 to 1935 data was the average yield, or three or four, and when you correct it, it is seven or eight. The same is true in Kansas. There is a section in Kansas, and some cases in Oklahoma from Enid, Okla., where they had a good crop in 1934 as well as a bumper crop in 1931, as the 1930-35 figures had about the same as 10 or even 25-year period in Kansas.

But if you go down into southwestern Kansas, those 14 counties down there, then, you have a situation similar to the one in North Dakota where you have to make correction. Now, we are making that correction. We are making that correction on the 1930-35 period. In addition to that, you have a few other adjustments in some cases that you have to make. For instance, there are two counties, namely, in central Iowa, where the rate in the report of the President's committee is as high as western Kansas, and we had the Bureau of Entomology to spot all insect outbreaks for the 2 years 1930 and 1935, and it was upon those two counties that the worst outbreaks centered. In this case, after you have done everything else, they will have to be brought into line similar to the counties around them. So that there are a few of this kind of adjustments that have to be made after you have done all the figuring, but that is the purpose, to get that at least on a 10-year basis for all States. If we had as good data as we do from some State like North Dakota, you could spread it out over a longer period.

Senator POPE. Are there any other questions, Mr. Green? Did you say all you wanted to say about the matter of grading?

Mr. GREEN. Yes, sir; I think that is about where we are in this matter of grades and that has appeared to be the most practical solution of that problem.

Senator POPE. All right, Mr. Green, if agreeable to the committee, we will meet tomorrow morning at 10 o'clock. We have one or more insurance men, and perhaps one or two other witnesses for a brief time. We hope to close the hearings tomorrow.

(Whereupon, at 4:15 p. m., an adjournment was taken until 10 a. m., Tuesday, Mar. 9, 1937.)