penditures lead to rapid growth, nor does it imply that rapid growth leads to high-arms expenditures.

But notwithstanding that caution, and some other cautionary remarks that I make in the prepared statement, about these data and about the unwarranted inferences that one should avoid drawing from them, I think the data are at least of interest in testing the intuitions or preconceptions about the relationships that are involved.

I do not want to be misunderstood in what I have said so far. Specifically I do not want to be interpreted as saying that military programs, either ours or those of other countries, should be justified by virtue of their favorable economic impact. Even if there are such favorable impacts, it seems to me that military programs are a high cost way of buying development. There are more efficient ways of doing this.

The primary justification for military assistance or security assistance programs, I think, must be found in other directions than in terms of their economic impact.

It seems to me this justification lies in helping other countries to meet reasonable defense goals based on two considerations: one based on their own demonstrated credentials for helping themselves; and the other based on the premise of some diminuation, some reduction, in our own involvement abroad. It seems to me, against that background, the justification for security assistance stems basically from the principle of comparative advantage. Where the United States and another country or countries have some security goals in com'mon, it makes sense (a) for us to contribute to some extent at least, to those goals, and (b) to provide that contribution in forms that are efficient for us to provide, to provide it in a relatively efficient way, such as military equipment and technical advice. And it makes equal sense for the other partner country or countries to make their contributions in ways, such as military manpower, that are more efficient for them to provide. In the example of Korea that I cite in my prepared statement, it turns out that the operating costs of ground forces in general, are something like one-twentieth to one-thirtieth as much for Korean forces as for our own forces, whereas in very gross terms, the effectiveness of those forces is perhaps one-third the effectiveness of our own. So there is a leverage factor of somewhere between seven and 10 against which to trade off some reductions in our expenditures and forces and manpower against some increases in theirs.

I think this general principle of comparative advantage has provided the underlying rationale for most of our military assistance in the past several decades. And I believe this is also the rationale at the core of the Nixon doctrine that Mr. Irwin has referred to in his statement.

Let me say finally that it seems to me if handled properly security assistance programs can be quite generally complementary rather than conflicting with international development programs. There are a number of ways in which each of these programs can help to advance the purposes of the other program. For example, economic development programs can raise the effectiveness of military security programs by increasing the resources on which local defense budgets can draw, and thereby diminishing the need for U.S. assistance, as

well as raising the educational and literacy levels on which the military can draw.

On the other hand, if development is to move forward in these countries, security has an important role to play. It is important that the environment be more predictable for savers, investors, workers, and other economic decisionmakers. And security programs that contribute to improved law and order and reduced insecurity, have this effect.

Furthermore, military programs can complement and advance economic development programs by adding to the stock of joint civil and military infrastructure, and also by raising the quality of human resources through military service and training that then is turned over to the civilian economy. I think something of this sort has had a very major effect in the case of Korea's economic development as I mentioned earlier.

So, in conclusion, I think one point that has not come up, at least in the testimony that I have had the privilege of reading before coming, is that potential complementarities between these two programs are considerable. They ought to be more carefully identified and more fully exploited than they may have been in the past. Thank you very much.

Chairman PROXMIRE. Thank you, Mr. Wolf.

We will place your prepared statement in the record at this point. (The prepared statement of Mr. Wolf follows:)

PREPARED STATEMENT OF CHARLES WOLF, Jr.1

Mr. Chairman and members of the committee, it is a pleasure to appear before this Committee in response to the Chairman's letter of invitation of January 11, 1971.

Of the three subjects that I understand these Hearings to be principally concerned with (namely, the costs, management, and impact of U.S. military assistance programs), the Committee staff has asked me to concentrate on the third, and more particularly to concentrate on the impacts of these programs on the economic development of recipient countries.

The remarks I will make are based on more than a dozen years of familiarity with these programs. In 1959, I served on the Asian Subcommittee of President Eisenhower's committee to study the Military Assistance Program, under the Chairmanship of William Draper. (As you may recall, the Draper Committee was one of the nine presidential committees that have been organized within the past two decades to investigate and report on various aspects of U.S. foreign assistance programs, economic as well as military, and on the relationships between them.) Since my association with the Draper Committee, I have from time to time done analytical studies on various aspects of these programs. However, I would ask the Committee to bear in mind that I have not been in close touch with these programs for the past two or three years.

Let me turn to the specific question you have raised. There are two sharply contrasting views of the effects of military assistance on development in the recipient countries.

One view, which I shall refer to as the "resource diversion" (RD) view, is that military assistance diverts resources from economic development. Both local resources and foreign exchange may be diverted by the recipient, for example, in order to operate and maintain the equipment received under the military assistance program. The diverted resources may also include the time and attention of top decisionmakers, as well as non-negligible fractions of the limited supply of administrators and skilled technologists in these countries.

1 The author is head of the Rand Corporation's Economics Department. Any views expressed in this paper are those of the author. They should not be interpreted as reflecting the views of The Rand Corporation or the official opinion or policy of any of its governmental or private research sponsors. Papers are reproduced by The Rand Corporation as a courtesy to members of its staff.

According to the resource-diversion view, military assistance therefore tends to retard development, as well as to distort it by increasing the relative size of the military sector, and by reducing the margin available for civilian capital formation and skill formation.

The second position, which I shall refer to as the "resource addition" (RA) view, argues that resources provided through military assistance represent additions to the recipient's economic base that are likely to advance its development. These resource additions include not only such increments as roads, harbors, warehousing, trucks, jeeps, and power generators, which may directly augment what is available to the rest of the economy. They also include such increments as may become available indirectly through increased supplies of better trained and more productive manpower coming into the civilian economy after receiving military training. The resource-addition view suggests that military assistance is likely to accelerate economic development and technological progress in recipient countries.

Incidentally, it is worth noting that logically, rather than ideologically, neither of these two views implies any conclusion with respect to the desirability or "value" of military assistance. One might adhere to the resourcediversion view, but nevertheless argue that the programs were fully justifiable on other grounds, despite these adverse effects. Or, one might adhere to the resource-addition view, but still contend that the programs are too costly for the benefits they provide, or that they are accompanied by other undesirable side-effects.

I would guess that the resource-diversion (RD) view is probably more frequently encountered, although my own opinion runs along the resource-addition (RA) line, for reasons I will explain in a moment. In any event, choosing between the two views is better approached on the basis of specific country experience rather than general theorizing. Instead of thus "choosing" a single "view," we might search for the conditions under which the one and the other, respectively, hold; recognizing that the world is unlikely to conform to only one pattern of this kind.

Turning to specific countries, one can cite examples that appear to sustain each of the two views. For example, my own judgment would be that the resource-diversion view would apply during the 1960's in India and Pakistan, as well as in Castro's Cuba. If the U.S. had provided less military assistance to India and Pakistan (that is, less to both, with the reductions applied in a balanced way, or had provided none to either country), competition for constrained resources within those two countries would have been altered in such a way that defense programs probably would have been smaller, development efforts larger, and the countries' rates of economic growth probably higher. And similarly, if Soviet military assistance to Cuba, following the missile crisis of 1962, had been appreciably less, I would conjecture that that country's rather dismal development record would have been improved.

On the other hand, there are a number of cases in which the resource-addition model seems to be a more realistic interpretation of events. As examples, I would cite the cases of Korea, Taiwan, and Thailand. Over the past ten years, I have had an opportunity to do a fair amount of work on the Korean economy, beginning in 1961, when I assisted that country's Economic Development Council in structuring its first development plan. It is my considered judgment, which has been elaborated more fully elsewhere, that security assistance has had a substantial and beneficial impact on the remarkable performance of the Korean economy. As you know, Korea's rate of growth in real national product, as well as in real product per capita, has in the past three or four years been the highest of any country in the entire developing world.

2

2 See "On Aspects of Korea's Five-Year Development Plan," P-2288, The Rand Corporation, Santa Monica, April 1961, and "Economic Planning in Korea," Asian Survey, December 1962.

In my judgment, Taiwan provides a similar example of the resource-addition view, although the direct experience I've had in that country is less than in the case of Korea.3

Which of the two views, RD or RA, applies in any particular case, and which view one chooses to adopt in general, depends on the answers one gives to two critical questions:

1. Would the resources that are made available to security assistance programs by the United States be available for development purposes if the security programs were reduced?

2. Would the military expenditures made by the recipient country not be made if U.S. military aid were not forthcoming? (In other words, would all, or part, of the country's own defense expenditures be discontinued if foreign military assistance were not forthcoming?)

The answer to the first question is something which the members of this Committee are in a better position to provide than I am. To put it more bluntly, would in fact appropriations for foreign economic development aid (grants, loans, technical assistance, investment guaranties, etc.) be larger if military aid were reduced? I strongly suspect that the answer is negative; I would be interested in hearing the Committee's own judgment on this question.

The answer to the second question is, in the major military aid recipients that I'm familiar with, also negative. Partly because of genuine security concerns in these countries, and partly because of strong internal political and organizational realities, military programs in fact often command a higher priority than development programs. It is often the development programs, or parts thereof, that are the marginal claimants on resources, rather than the defense programs.

In countries like Israel or Korea or Thailand, the basis for the higher relative value placed on defense seems clearer and more convincing to us than in some other countries. Yet it is an important reality in the national priorities of many other small and insecure countries, too. In such cases, the effect of military aid may well be to release resources for development purposes. And reductions in military aid may well result in diminished resources for development.

Notwithstanding these observations, or perhaps because of them, I would like to endorse a point made by several others who have already appeared during the course of these Hearings: in reorganizing and refurbishing our security assistance programs, a strenuous effort should be made to ensure that these programs, as well as the related local resources within the recipient countries, are allocated and evaluated within the standard budgetary procedures of the recipient countries. To my knowledge, this is rarely if ever done in the conduct of our military assistance programs, although as I mentioned earlier my direct knowledge on this point is at least a couple of years out of date.

To summarize the preceding discussion, my reasons for favoring the RA view over the RD view are based both on direct experience in the countries I've referred to, as well as the negative answers that I would judge should be given to the two questions concerning resource availabilities.

Let me now turn from the specific question of the impact of military assistance programs on development, to the broader, but related, question of whether a high rate of defense expenditures tends to be associated with a low economic growth rate in the developing countries? In some empirical work done on Latin America by a colleague of mine at Rand two years ago, an effort was made to answer this question. The results are summarized in the following table.

3 For a compatible, though not identical, view, see Neil H. Jacoby, U.S. Aid to Taiwan: A Study of Foreign Aid, Self-Help, and Development, New York, 1966, pp. 116-126. Jacoby's approach is rather different from that suggested in my statement in the text. since he considers the question of the "military burden" from the standpoint of whether, and by how much, Taiwan's growth would have been accelerated if all the country's resources that were devoted to defense purposes were instead to have been available for developmental uses. This assumption begs the fundamental question of whether that country's own priorities would have permitted such a reallocation. In any event, note the following interesting conclusion that Jacoby's study leads him to: The view has often been expressed that a ["semideveloped"] country cannot simultaneously maintain a powerful military force and also finance rapid economic growth. The case of the Republic of China suggests that both national purposes can be achieved simultaneously by a semideveloped country like Taiwan's in 1965 provided that major military hardware is supplied by an external source. (underscoring added) Ibid., p. 126.

Source: Excerpted from Thomas A. Brown, "Statistical Indications of the Effect of Military Programs on Latin America, 1950-65," P-4144, The Rand Corp., Santa Monica, Calif., July 1969, p. 10. "Highs" and "lows" are divided at the respective medians.

The first figure opposite each country in the table shows that country's defense outlays as a percentage of GNP, while the second figure after the semicolon shows the country's annual growth rate in real terms, for approximately the period 1950-1966. What the table suggests is that the answer to the question is emphatically negative: high arms burdens do not seem to be associated with low growth rates. In fact, the data suggest just the reverse: most of the countries with relatively high defense budgets have relatively high growth rates as well, and most of the slowly-growing countries in Latin America have low defense budgets! Of course, such an apparent relationship certainly does not imply a casual connection in either direction. Nevertheless, the results are perhaps surprising, although not as surprising as might be immediately inferred. In fact, for a number of sociological, organizational, and psychological reasons, one might expect such a result. However, to conjecture about these reasons would be too time-consuming and would take me too far afield to warrant further discussion here.

In any event, I would urge extreme caution in drawing any inferences from these data. One reason for caution concerns the data themselves: intercountry and intertemporal comparisons of the sort summarized in the previous table are often unreliable. Another reason is that there are obviously several countries that don't fit the apparent result: for example, Mexico and Bolivia both have low defense budgets and high growth rates, while Argentina and Paraguay have relatively high defense budgets and low growth rates. Anyhow, the data are at least of interest in testing one's intuitions, or prejudices, about the relationships that are involved.

I do not want to be misunderstood in what I've said up to this point; specifically, I want to make quite clear that I am not suggesting that military programs either ours or others should be justified by virtue of their favorable economic impact. Even if there are such impacts, military programs seem to me a high-cost means of obtaining them. There are more efficient ways.

The primary justification for military assistance programs, and more broadly for security lies in other directions. This justification lies in helping other countries to meet reasonable defense goals, based on their demonstrated credentials for helping themselves, and based on the premise of some diminution in U.S. involvement and expenditures for national security purposes abroad. Basically, the justification for our security assistance programs lies in the principle of comparative advantage. Where the United States and another country or countries share common security objectives, it makes sense for the U.S. to concentrate its contributions to those joint objectives in forms which we can provide efficiently, such as military equipment. And it makes equal sense for the other partner country or countries to provide their contributions in forms, such as military manpower, which they can furnish more efficiently. For example, in the case of Korea, it has been estimated that Korean forces are on the average perhaps one-third as effective as corresponding U.S. ground forces, yet their operating costs are one-twentieth, or a still lower fraction of U.S. operating costs. With the resulting leverage factor of 7 or more, it is efficient from both the U.S. and the Korean standpoints to substitute or upgrade Korean forces in place of higher-cost U.S. forces.