Just as in the case of rice, the same general degree of concentration can represent quite different distributions of benefits by allotment size. Farmers with allotments of 200 acres or less receive 97 percent of all benefits in the North Central region, 61 percent in the Great Plains, and only 34 percent in the Northwest. Farmers with allotments of 1,000 acres or more receive only 0.1 percent of all benefits in the North Central region, they receive 3 percent in the Great Plains, and 11 percent in the far Northwest (see table 9). Variation between States is considerable.

The average U.S. Gini ratio for wheat price supports is .566, for wheat certificate payments is .544, and for wheat diversion payments is .480. These statistics, of course, refer to distributions among program cooperators. But, in an industry in which a quarter of the producers choose not to participate in a price and income support program, it is relevant to ask how the nonparticipating allotment holders differ from those in the program. While the distributions are not presented in tabular form, Lorenz curves and Gini ratios were computed on allotment acreages of nonparticipating allotment holders with interesting results.15

Using the same type of value weighting system as we did for price supports, we find that over the whole of the United States, wheat allotments held by farmers outside the program exhibit a concentration ratio of .355-well below the .566 for pricesupport program cooperators. The concentration ratios of the regions for nonparticipating farmers generally follow the pattern previously observed for those in the program: the Northwest had a Gini of .693, the Great Plains one of .573, and the North Central region is .202. In other words, allotment concentration among nonparticipating farmers is greater in the Northwest and Great Plains than among farms in the program in those regions. In all other regions and for the United States, concentration is lower among nonparticipating farmers than those in the program.

Clearly it is the smaller allotment holder who is the typical noncooperator. Less than 1 percent of noncooperators hold allotments of more than 100 acres; less than 5 percent hold allotments of more than 30 acres. Among program cooperators, 99 percent of farmers hold allotments of 100 acres or better. Thus, not only is concentration on the whole lower, but the average size of wheat allotment is far smaller among those wheat farmers who chose in 1964 not to participate in the wheat program.16

15 The data came from the same source as the rest of the wheat program distributions. See source (a) of table 4. The data came from tables 4 and 5 of the source.

16 One must be careful in interpreting this. Because a nonparticipating farmer has a small allotment does not mean he planted wheat acreage equal to his allotment. He may have planted far more-or none at all. Also, the farmer who holds a small wheat allotment may not be a small farmer, wheat may just be a minor enterprise on his farm. All we have done here is compare distributions of allotment size.

This suggests that if concentration were measured over all allotment holders, program participators and nonparticipators, the measure of benefit concentration in wheat would be greater than that reported here for program participators.

The interesting general conclusion suggested by this analysis of the 1964 wheat program is that the combination of direct payments (as designed) with price supports results in an even greater concentration of the distribution of benefits from the wheat program than results from price supports alone.

The Feed Grain Program

The feed grain program is very similar to that of wheat, for the wheat program was designed in part with the feed grain program as its model. The primary difference is that in feed grains there is a price-support payment in place of the wheat certificate payment. Thus, in feed grains there are three types of program benefit streams: price supports, price-support payments, and diversion payments.

We have feed grain program distributional data for the 1964 crop year, just as in wheat. The same type of data is available for computing pricesupport benefit distributions as well as diversion. payments. Distributions of price-support payments by acreage allotments are not available. However, it is possible to construct such a distribution using data (from another source) on the distribution of total feed grain price-support payments by State (14) combined with the distribution of planted acreage by allotment size on participating farms. The necessary assumption is that price-support payments are distributed between allotment size groups within a State in the same proportion as planted acreage on participating farms. This is a perfectly valid assumption in this case since price-support payments are made on each bushel produced on participating farms and should exhibit distributions. and a degree of concentration identical to that of benefits from price supports.

There is a difference from wheat also in the procedure for estimating the distribution of price-support benefits. In the case of feed grains, actual planted acreage distributions are available by allotment size group by State. This avoids problems of failure to plant and crop failure, which one is involved with when only a distribution of allotment acreage is available (as in wheat).

The weights for feed grains also had to be computed in a different and less satisfactory fashion. Season average prices by State are available only for the components of feed grains: corn, oats, barley, and grain sorghum. Allotment data are available by State, only for total feed grains-not by corn, oats, barley, or grain sorghums. Consequently, one is forced to use a value of total feed grain production as State value weights for aggregation purposes.

The weight for a State is an aggregate of the products of State average price and production data on corn, oats, barley, and grain sorghums. It is possible that this introduces some distortion into the relative relationship of the value weights if the proportion of nonparticipating farmers and acreage outside the program varies greatly between allotment size classes. As noted earlier, experiments using various weighting systems indicate that the Lorenz curves are not very sensitive to changes in the weights. Nevertheless, it would have been better to be able to use a wheat type of weighting system.

As with wheat, we have the opportunity in this analysis to observe the impact on the concentration of benefit distribution that results when direct payments are combined with price supports.

The Gini concentration ratios for all of the Lorenz curves computed for feed grains can be seen in summary form in table 10. The pattern is utterly consistent, for every State, every region, and for the United States. The degree of concentration of diversion payment distributions is well below that for price support benefits. The concentration ratio for the combined or total program benefit distributions falls consistently between that for diversion payments and that of price-support benefits.17 In contrast to the results in wheat, the diversion payments in feed grains tend to reduce the concentration of overall program benefits, though perhaps not by as much as one might expect for the volume of diversion payments that are involved in the feed grain program.

Price-support benefit Lorenz curve computations are presented in table 11. Similar presentations for diversion payments are found in table 12 and for total program benefits in table 13. The lower half of farmers receive only 8 percent of all price-support benefits, but 24 percent of all diversion payments. The top 10 percent of farmers receive 37 percent of all price-support benefits and 30 percent of all diversion payments. There is considerable variation by State and region, but the variance is far greater in the case of price-support benefits than diversion. payments-which is reflected in the lower concentration ratios for diversion payments.

Again, looking at the distribution of benefits by allotment size groups (in table 14) one sees a varied picture. Nationally, half of all program benefits go to producers with allotments of 100 acres or less. This is also true of the North Central and South Central regions. But in the Western States only 14 percent of the benefits go to farmers with allotments of 100 acres or less, and in the North Atlantic region these farmers receive 83 percent of all benefits.

Nationally, 6 percent of all feed grain program benefits go to farmers with allotments of 500 acres

Arkansas. Kentucky. Louisiana. Mississippi. Oklahoma. Tennessee.

Colorado.. Idaho..

Oregon. Utah. Washington Wyoming

Western.

United States.

Sources: Tables 11, 12, and 13.

For an explanation of the Gini concentration ratio see section on procedure in this paper.

1

Lower

Gini concen

farmers

farmers

farmers

farmers

20% of 10% of farmers

1% of

tration

farmers