Foreign Relations of the United States, 1964–1968, Volume XXV, South Asia

450. Memorandum From the President’s Special Assistant (Rostow) to President Johnson¹

Washington, August 2, 1967, 1 p.m.

SUBJECT

• Food Aid for India—Act II

At Tab A is a complicated memorandum from Gaud and Freeman²—supported by Rusk and Fowler—recommending a new PL 480 agreement to supply India with a million tons of wheat, along with other minor odds and ends. The memorandum is complex because it is very hard as a technical matter (1) to define precisely what we mean by “matching”, and, whatever our definition, (2) to be sure how much matching is actually going on. Your advisers have concluded that by any measure we now have enough matching to justify another one million tons of U.S. wheat.

The Matching Problem

You will recall that your India food message—and the Congressional Resolution which followed—spoke in terms of 3 million tons of additional wheat this calendar year if we got “appropriate” matching from other donors. By May, we had collected pledges totalling $97 million in new food or food-related aid from the others. Therefore, you authorized a first agreement providing for 1.5 million tons of wheat, worth about $95 million.

We now have a further $122 million in aid pledges from other donors—most of it in the form of debt relief. (See table at Tab B for breakdown.)³ Both the World Bank and the Indians argue that this is more than enough in matching pledges to justify releasing another 1.5 million tons, completing the full 3 million tons contemplated in the Resolution.

However, our matching proposal was designed to draw additional aid resources from the other donors—resources that otherwise would not have been supplied to India. Neither we nor the Indians gain anything from ear-marking aid as food-related which is clearly not over and above what these countries have traditionally provided through the consortium anyway. We just don’t know how much of this “new” $122 million is additional. (Indeed, we know that some of it definitely is not additional—the British have told us outright that their contribution will be subtracted from their consortium pledge.) Until we see what our brethren do at the consortium meeting in October, we just won’t know whether they are really providing something extra or whether they are simply putting new labels on old benefits.

Despite a hot debate, your advisers have not been able to agree on precisely how much more U.S. wheat we can provide now without threatening the matching principle. But they do agree that we can clearly justify at least a million tons. The acid test of additionality is how much new grain from non-PL 480 sources the Indians have come up with since our last PL 480 agreement was authorized. Any new grain from other donors is clearly eligible for matching. Any Indian purchases beyond what they had programmed in May reflect the Indians’ belief that the new aid they have received has freed more of their foreign exchange for buying food. This is not a foolproof method of measuring additionality. The Indians may be diverting money from their development budget to buy food, or they may just be wrong about what the new aid will do for them. But our people doubt that much diversion is going on, and they point out that there are tremendous pressures on the Indians to be right about their foreign exchange situation.

Applying this test, we find that India has scared up 1 million tons in new grain from non-PL 480 sources since last May. They have a new Australian donation of 150,000 tons, 200,000 tons in the process of being donated by the Soviet Union and 650,000 tons in additional Indian commercial purchases. (There is some question about whether the Russians will come through, but the Indians have officially informed us that they will buy enough more themselves to cover any Russian shortfall.) Accepting the tough principle that the best proof of additionality is how much a new non-PL 480 grain the Indians actually import, we can clearly justify another million tons from the U.S. This is what your advisers recommend.

The Need

The Congressional Resolution and the entire matching exercise are aimed at an overall target of 10 million tons of grain imports into India this year—through food aid and commercial purchases—without forcing India to slow down her development program. According to our latest estimates, she may be able to squeeze by with 9.5 million tons without mass starvation. Even with her greatly expanded commercial purchases (44% more than last year), this will require the full 3 million tons from us mentioned in the Resolution. The central Government’s stocks are now down to about two weeks’ supply and dropping. Food shortages are very important factors in touch-and-go political situations in several major states. The outlook for the next four months is bleak at best. Without our grain it is hard to see how Mrs. Gandhi could handle it without completely derailing the fertilizer imports and other development expenditures which will stimulate larger Indian crops in the future.

And the timing is urgent, as usual. The 1.5 million tons authorized in May will be fully delivered by the middle of September. Because of the closing of Suez and the strain on tanker supply, it now takes six weeks’ leadtime to book ships, rather than 3–4. Thus, if the pipeline is not to break, we need a decision this week.

The Politics

The political case for food aid to India is neither better nor much worse than when you last reviewed it. The Indians played an irritating and often stupid role in the Middle East crisis. Internally, Mrs. Gandhi remains as weak and indecisive as ever; her party is having more and more difficulty maintaining its power in the states. Nobody is certain how long she can hang on. However, there is a powerful case that this makes food aid more rather than less imperative. If change is to come to India, it is very much in our interest that it be as peaceful as possible. But peaceful change is most unlikely in a context of starvation.

The major immediate political point is that if we do let the pipeline break, our relations with the Gandhi government—and any successor—will turn very sour indeed. And our public explanation of why we didn’t come through (presumably because we don’t think we have been matched) would not be supported by the World Bank or other “neutral” observers. We could also expect some trouble at home.

The Alternatives

You should know that this recommendation is the result of careful examination of four alternatives:

1. Do nothing until after the October consortium meeting.

2. Do a million tons now and worry later about the additional half-million tons required to meet the 3 million tons mentioned in the Resolution.

3. Do 1.25 million tons now, on condition the Indians buy the other 250,000 tons.

4. Do 1.5 million tons now, arguing that we have matching for at least 1 million tons and that we are not committed to 1–1 matching. (On this alternative, we would have at least 6–5 matching for the full year.)

No. 1 was dismissed as impossible in view of the urgent need and the dangers of delay. No. 3 was rejected as too time-consuming and too likely to force India to dip into development funds to pay for wheat. No. 4, strongly favored by State and AID, was rejected because Fowler and Freeman believe it would amount to abandoning the matching principle.

That left us with No. 2 as the lowest common denominator. Its principal drawback is that it leaves us with a further 500,000 ton requirement late this year for which we now have no matching prospects in sight. The choice between Nos. 2 and 4 (1 million tons versus 1.5 million tons) reduces to whether we want a 1.5 million ton package now with an unmatched component or risk a half-million ton package later with no matching at all. You may wish to consider going ahead with the full 1.5 million now.

Indian Purchases from the U.S.

The Indians have taken the procurement lesson to heart. Their commercial grain purchases in the U.S. this year will be at least four times⁴ what they bought last year. And they are following through on their promise to buy at least half of all their commercial grain from us, beginning last May.

All of the self-help conditions we have worked out for the earlier agreement would also apply to this one.

Recommendation

I recommend you approve the Gaud-Freeman memorandum.⁵

Walt

Approve Gaud-Freeman memo (1 million tons)

I want to hear more; set up a meeting for me with Rusk, Freeman, Gaud, etc.⁶

No

See me⁷

1. Source: Johnson Library, National Security File, Country File, India, India’s Food Problem, Vol. IV. Secret. A handwritten note on the memorandum reads, “Rec’d 3:32 pm.”
2. Memorandum from Freeman and Gaud to the President, August 1. (Ibid., Memos to the President, Vol. 37, August 1–10, 1967)
3. The table, which lists the offers of food assistance on a country-by-country basis, is ibid., NSC Histories, Indian Famine, August 1966–February 1967, Vol. IV.
4. Johnson circled “four times” and wrote in the margin “How much?”
5. Rostow added a handwritten postscript that reads: “We started out to match $190 million. We have matched somewhere between (say) $160 million and $219 million. We are taking the lower figure here. In fact, the exercise has been more successful than we could have hoped.”
6. Johnson checked this option.
7. Johnson checked this option and added the following handwritten note: “Get me someone to argue the other side please.” On August 4, in response to the President’s instruction, Rostow sent him three memoranda drafted by the NSC Staff that made the case for three options: no more food now, one million tons now, and 1 1/2 million tons now. (Johnson Library, National Security File, Country File, India, India’s Food Problem, Vol. IV)