The Roles of Aid in Politics Putting China in Perspective

The findings of this study suggest a complex relationship among aid, trade, and political compliance in Soviet-Third World relations. It has long been noted that the Soviet Union uses aid as a direct instrument of influence, by seeking to induce Third World states to adopt policies favorable to the Soviet Union. On the one hand, in order to encourage Third World states to adopt anti-Western or pro-Soviet policies, the Soviets have offered assistance (Walters, 1970:33). On the other hand, ‘the Soviet Union has demonstrated a willingness to cancel credits abruptly or exert severe economic pressure on countries of importance to the USSR which adopt anti-Soviet policies’ (Walters, 1970:167; Dawisha, 1979:181).

Yet, this direct use of Soviet aid appears to be only part of a more complex causal nexus. According to the theory of Soviet-centered dependence and the results presented here, Soviet aid apparently can also induce compliant behavior indirectly through the creation of trade dependence. Indeed, specific features of the Soviet aid programs appear designed to accentuate their trade dependence effect. First, almost all Soviet aid has been tied, in the strictest sense, to purchases of Soviet machinery and equipment. This enhances the import dependence effect. Second, the Soviets have permitted many Third World recipients of arms and economic aid to repay these loans with local currencies or commodities (Cooper and Fogarty, 1979:650; Stevens, 1976:78). This enhances the export dependence effect.

The results of this study suggest that while the importance of the Soviet arms program as a direct instrument of influence has grown, its importance as a tool for creating trade dependence among recipients has declined. The diminished export dependence effect of Soviet arms assistance may result from the shift in Soviet arms sales policy toward commercial terms with repayment in hard currency. Originally, the Soviet Union was likely to offer liberal terms of repayment, including a repayment period of eight to ten years at 2 per cent and, importantly for its export dependence effect, repayment in commodities. In the 1970s, the Soviets increasingly required repayment in hard currencies (Cooper and Fogarty, 1979:650). Ericson and Miller (1979:214), both of the Central Intelligence Agency's Office of Economic Research, estimate that hard currency receipts rose from about 10 per cent of total arms deliveries in 1970-1972 to about 43 per cent in 1976-1978. This has meant that recipients, on average, are less likely to redirect their exports to the Soviet Union in order to pay for Soviet arms.¹⁷ In short, while Soviet arms transfers appear to have accentuated the growth of Third World trade dependence on the Soviet Union in the 1960s, they do not appear to have slowed its decline in the 1970s.

Conversely, the results suggest that the offer or threatened loss of Soviet economic assistance has not directly increased compliant behavior among recipients, but that the Soviet economic assistance program has increased in importance as a tool for building trade dependence among recipients. The second may be a result of important shifts in Soviet policies since the fall of Khrushchev away from his ‘political approach to development’ (Valkenier, 1970:424). Specifically, the growing export and import dependence effect of Soviet economic assistance may be the consequence of more effective administration as well as growing pragmatism and commercialism in the Soviet program. Improved administration – institutionalized in such practices as ‘extensive long-term planning’ through joint permanent commissions with individual recipients (Valkenier, 1970:417-418) – means that projects are more likely to be completed successfully and efficiently (Cooper and Fogarty, 1979:653; see also Walters, 1970:116). Growing pragmatism has meant fewer ‘showy’ projects that are unproductive or uneconomical. It has been reflected in a less dogmatic approach to questions of nationalization, agricultural development, and state planning within Third World economies (Valkenier, 1970:423-431). The greater commercialism of Soviet economic assistance policy has meant that the Soviet Union gives closer attention to projects that promise production that complements the Soviet economy. Indeed, some Soviet analysts have ‘viewed aid-giving primarily as an alternative to domestic investment and argued that it would be cheaper for the USSR to import certain goods and materials than to produce them at home’ (Valkenier, 1970:416-417; cf US Central Intelligence Agency, 1980:8; Pockney, 1981 :54). These shifts in policy seem to have increased the import and export dependence effect of Soviet economic aid.

This apparent difference in effect between the two aid programs suggests that the relative importance of the two forms of Soviet aid as instruments of influence varies with time. That is, in the short term Soviet arms appear to be the more significant instrument; in the longer term, economic assistance may be the more consequential. This difference implies that in a broader Soviet strategy its aid instruments can complement one another in the Third World. A rational Soviet strategy concerned with building political influence in the Third World can balance both forms of aid, using economic assistance to build longer-term ties that bind Third World states to the Soviet Union for political purposes, while using arms transfers as a more immediate inducement of favorable policies (cf Albright, 1982:301-302).

Finally, the findings of this study raise questions about the claim of the theory of Soviet-centered dependence that the longer-term Soviet objective behind its aid programs is to build an ‘international socialist division of labor’. This theory argues that the ultimate Soviet objective in building trade dependence is to integrate LDC economies with the socialist system and to deny those sources of raw materials and markets to the capitalist economies.

If this is, indeed, the Soviet objective, the findings of this study argue that the Soviet Union is failing. While Soviet economic aid has been associated with increasing trade dependence on the Soviet Union, its effect has not been enough to overcome those forces that are eroding that dependence. As noted above, the proportion of Third World trade with the Soviet Union has been declining among recipients of Soviet aid as well as non-recipients. In our study of 62 Third World states, only 18 saw the proportion of their exports going to the Soviet Union grow between the beginning and the end of the 1970s and only three saw this grow by more than five percentage points (Argentina, 14.4 percentage points; Ethiopia, 8.7; and Ghana, 5.6). In this period, only 24 states saw the proportion of their imports coming from the Soviet Union grow and only one saw this grow by five percentage points or more (Ethiopia, 6.9 percentage points). Of these three states, only Ethiopia was a major recipient of aid in the 1970s. Pockney (1981 :3) notes:

The 1976 Report of the United Nations Conference on Trade and Development shows that whilst the trade of the USSR has expanded in physical and monetary terms, as has the trade of most developing countries, the proportion of the world Import-Export market each group has taken has been falling for most of the years since 1960. The importance of this cannot be over-emphasized. Even with all the problems of oil prices, inflation, stagflation, and the biggest depression in the trade cycle since 1945, the developed market economies have been taking, together with OPEC countries, larger shares of an expanding world market turnover (emphasis in original).

In short, the growth in Soviet-LDC trade has failed to keep pace with the far more dynamic growth of Western trade. It would appear that in the aggregate Third World trade dependence on the Soviet Union is declining.

In fact, this claim by the theory of Soviet-centered dependence may be based on a misunderstanding of both Soviet and Third World objectives. First, this theory may fail to understand Third World intentions in opening economic ties with the Soviet Union. Stevens (1976:38) argues that in Black Africa ‘commercial relations with the East have been thought of. . . not as an alternative to Western markets but as an adjunct to them’. Third World states, for the most part, have remained interested in trade with the West, which earns them hard currency and provides commodities of higher quality. Yet they cannot continue to expand their sales of primary products to the West without depressing the prices they receive. To avoid such gluts that might jeopardize their trading relations with the West, many Third World states have turned to the Soviet Union and the East as a second-best market to which they can sell their primary products without depressing the world market price. In short, Third World states oftentimes turn to the East not to redirect their trade from the West but to protect that trade. This limits Soviet opportunities to build trade dependencies and build a separate ‘international socialist division of labor’.

Second, the claims of this theory may be based on Soviet theories that are now dated and have been revised. Valkenier’s research (1970, 1979) into Soviet foreign economic policies points out that the Soviet view of the international division of labor has changed fundamentally. She argues persuasively (Valkenier, 1979:17) that:

The new trends in aid and trade policies and much of the attendant discussion suggest that the aim is not so much the displacement of the existing economic order as its modification to create appropriate conditions for greater and more advantageous Soviet participation. Instead of trying to enlist less developed countries (LDCs) of the Third World in an international socialist division of labor . . . the Soviets are increasingly discussing and instituting new forms of economic exchanges wherein mutual advantage, modified market forces, multilateralization, and commercial gain will benefit not only the LDCs and the socialist bloc but the West as well, with all three categories considered and treated as parts of an interdependent world (cf Albright, 1982: 302).

The decline in Third World trade dependence on the Soviet Union may be closely tied to this change in policies. In particular, the Soviet Union, in its commercial relations with the LDCs, may increasingly be seeking to complement its growing trade with the West by earning hard currency and attempting to support economic development at home by satisfying consumer demands and providing primary products more economically. With this shift in purpose, the Soviet Union is less concerned with building a separate international socialist division of labor.

Taken as a whole these findings suggest a second policy implication: the Soviet Union, in seeking political influence, may find reliance on military, rather than economic, assistance more profitable in the future. The apparent continued value of arms assistance as a direct inducement to political compliance (just as economic assistance seems not to have had that value) gives priority to the first in the short run. The inability of economic assistance (even with an apparent increase in its ability to build trade dependence) to compensate fully for all those forces eroding the proportionate position of Soviet trade in the Third World may lead the Soviets to question the value of economic assistance for even long-term political purposes. Indeed, the growing economic rationality in Soviet economic assistance programs may, in part, be a reflection of the growing recognition that these programs have only limited political value. And the rise of Soviet military involvement in the Third World during the last ten years may be a manifestation of growing Soviet recognition that they are unable to extract great political value from economic asistance [sic] or to build an international socialist division of labor (cf Albright, 1982:301).

Stripped of its sometimes apocalyptic fears, such as the denial of raw materials to the West and the ‘inevitable crisis of capitalism’, the theory of Soviet-centered dependence describes consequences of military and economic aid that are not unique to the Soviet programs. While this theory and other studies of aid, dependence, and compliance have largely developed independently of one another, each posits a set of relationships that might be argued to exist for all great powers. The effort here has been limited to formulating the theory of Soviet-centered dependence as a series of hypotheses, testing those hypotheses, and drawing parallels to the extant literature on dependence in a manner that is faithful to the original, albeit inchoate, theory. Yet, the parallels found to exist in the empirical evidence suggest that future research on either Soviet or Western aid, dependence, and compliance must consider the effect of competing efforts by other great powers dispensing aid, expanding trade, and building political influence.

Each of the papers examined here embodies a set of choices about model specification and data. Aid is measured a certain way. A certain epoch is studied. Periods have a certain length. And so on. Some of these choices imply certain assumptions about the world, such as, say, that aid is not endogenous to growth. All limit the scope of a strict interpretation of the results.

To wit, the Burnside and Dollar results can be stated more precisely as follows:

Aid was associated with higher GDP growth in a good policy environment during 1970–93, on average, in countries and periods where the necessary data was collected, except for outliers (unless aid²×policy is included to allow for diminishing returns to aid), when aid is defined as “Effective Development Assistance” as a share of real GDP and polices are defined by inflation, budget deficit, and the complex Sachs-Warner “openness” variable, controlling for log of initial real GDP/capita, assassinations per capita, ethno-linguistic fractionalization, the product of those two, money supply/GDP, and period effects, assuming that no unobserved country-specific effects simultaneously and substantially influence aid, policies, and growth, and no variables other than aid aid×policy are endogenous to growth.

This is not quite “aid has a positive effect on growth in a good policy environment.” In fairness, such qualifiers could be tacked on to the conclusion of any study in the cross-country literature on aid and development, or in econometrics more generally (though perhaps not always quite so many). Moreover, Burnside and Dollar did test some of their assumptions, such as the exogeneity of the policy variables. Nevertheless, a question of great scientific and practical importance remains, and it is how many of such implied qualifiers in studies of aid and growth can be dropped without harming the conclusions. This study attempts to contribute to answering that question.

The test results reported here suggest that the fragility found in Easterly et al. (2004) for the case of Burnside and Dollar is common in the cross-country aid effectiveness literature.

In surveying the results, it is tempting to ask which results are robust and which are not. But the test data are best seen in shades of grey. The results tested here break roughly into five groups, ranging from weakest to strongest. The weakest group consists of the results on aid×policy in the Burnside and Dollar, Collier and Dollar, and Collier and Dehn regressions, which lose significance at 0.05 in all but the weakest tests. In the second division I put the Collier and Dehn result on Δaid×negative shock, which passes more tests but is quite sensitive to changes in the control set, as well as to removal of a minority of the negative shocks in the sample. The Collier and Hoeffler result on post-conflict 1×aid×policy (or the collinear post-conflict 1×aid), the Hansen and Tarp results on aid and aid², and Guillaumont and Chauvet result on aid×environment seem stronger. They generally pass the “whimsy”-based tests at or near 0.05. Most survive the sample-expansion test, but with autocorrelation—and then fail the AR-robust test meant to address this problem. The Guillaumont and Chauvet result could not be put to the sample-expansion tests, so the degree of its robustness is less certain and I add it to this middle category.

In the fourth and fifth categories are the GMM results of Hansen and Tarp and Dalgaard et al., respectively. Both fare well under the sample expansion. The Hansen and Tarp GMM results, especially those on aid and lagged aid, generally persist through the test suite, though not always significantly at 0.05. The only test that completely eliminates the Hansen and Tarp GMM results is that of defining aid as EDA/real GDP, but this is a misleading measure of aid. As for the Dalgaard et al. results, they come through powerfully in all the tests but the 12-year aggregation that reduces the sample size to 116.

Does this mean that the statistically weaker stories of aid effectiveness should be dismissed? Are recipient policies, exogenous economic factors, and post-conflict status irrelevant to aid effectiveness? No. There can be no doubt that aid sometimes finances investment (Hansen and Tarp, 2001), and that domestic policies, governance, external conditions, and historical circumstances influence the productivity of investment. Why then do such stories of aid effectiveness not shine more clearly through the numbers? The reasons are several. Aid is probably not a fundamentally decisive factor for development, not as important as, say, domestic savings, inequality, and governance. Moreover, foreign assistance is not homogenous. It consists of everything from in-kind food aid to famine-struck countries and technical advice on building judiciaries to loans for paving roads. And some aid is poorly used. Thus the statistical noise nearly drowns out the signal.

If there is one strong conclusion from this literature it is that on average aid works well outside the tropics but not in them. But just as it would be a mistake to conclude that the other stories of aid effectiveness contain no truth, it would also be mistaken to conclude that this result is the wholly, simply true. Indeed, the Dalgaard et al. result is more of a question than an answer. Presumably distance from the poles is not a direct determinant of aid effectiveness. Rather, the causal pathways are complex, and so it cannot be assumed that no kind of aid will work well in the tropics. Much the same can be said about the more optimistic, but somewhat less robust, Hansen and Tarp result on the overall positive effect of aid on growth. Even accepting it as true, it gives little guidance about where aid ought to be sent, and in what forms.

Perhaps further econometric work will disaggregate aid by types of program and recipient and unearth more robust answers to the fundamental questions of aid policy. Or perhaps re-searchers have hit the limits of what cross-country empirics can reveal about aid. The search for truth may need to rely more on the particularistic case study approach. Van de Walle and Johns-ton (1996), for example, synthesize conclusions from case studies on the use and effects of aid in seven African countries, each jointly conducted by researchers from donor and recipient countries. Of course, the lessons that emerge from case studies are particular to the country studied. But they can be generalized. Killick (1998) provides an excellent example by conducting a systematic survey of case studies that feeds into a trenchant analysis of the effects of IMF conditionality. Nevertheless, robust generalizations will not come easily.

The results reported here suggest that the fragility found in Easterly, Levine, and Roodman (2004) for Burnside and Dollar (2000) is the norm in the cross-country aid-growth literature. Indeed, in a counterpoint to the focus of Leamer (1983), Levine and Renelt (1992), and Sala-i-Martin (1997) on the choice of controls as a source of fragility, it turns out that modifying the sample generally affects results the most. For example, in the Collier and Dollar (2002) regression, half of the specification-modifying tests leave the t statistic at 1.49 or higher and two more lower it to near 1.00 (see table 5). But adding more years sends it to –0.19—and, after dropping outliers, to –0.81 (see table 6).

Does this mean that the various stories of aid effectiveness should be summarily dismissed? Are recipient policies, exogenous economic factors, and postconflict status irrelevant to aid effectiveness? Are there no diminishing returns to aid? Is helping the neediest countries a hopeless task? No. There can be no doubt that some aid finances investment and that domestic policies, governance, external conditions, and other factors these authors study influence the productivity of investment.

Why then do such stories of aid effectiveness not shine through more clearly? Aid is probably not a fundamentally decisive factor for development, not as important, say, as domestic savings, inequality, or governance. Moreover, foreign assistance is not homogeneous. It consists of everything from food aid for famine-struck countries to technical advice on building judiciaries to loans for paving roads. And much aid is poorly used—or, like venture capital, is like good bets gone bad. Thus the statistical noise tends to drown out the signal.

Perhaps researchers will yet unearth more robust answers to the fundamental questions of aid policy. Or perhaps they have hit the limits of cross-country empirics. Either way, robust, valid generalizations have not and will not come easily. Despite decades of trying, cross-country growth empirics have yet to teach us much about whether and when aid works.

Rothchild, Donald S. (2001), “The US Foreign Policy Trajectory on Africa,” in SAIS Review, Vol. 21, No. 1, pp. 179-211, John Hopkins University Press.

Despite an activist foreign policy orientation, the U.S. trajectory on Africa has, on the whole, tended toward low-profile involvement.⁷⁷ To be sure, there have been notable exceptions, such as the 1986 Comprehensive Anti-Apartheid Act, the 1988 mediation of the Angola-Namibia accords, and the humanitarian intervention in Somalia. In general, however, U.S. administrations, Republican and Democratic alike, have been cautious in their engagement with Africa. The Clinton administration, despite its rhetoric on partnership, held firm to the post-World War II trend of limited involvement. Its initiatives on enlarging democracy, promoting trade, backing a debt write-off, and sponsoring ACRI [African Crisis Response Initiative] represent low-cost and low-risk efforts that reflect U.S. interests and values. Its record on foreign economic assistance and peacekeeping are not what one would expect of a great power. By acting pragmatically, the Clinton administration had to limit its foreign policy objectives to what was acceptable to Congress and the public.

Under these circumstances, what can be done to promote a change in the United States trajectory toward a more constructive and life-affirming interaction with Africa? We do not want an activist foreign policy that does not respect Africa’s desire for autonomy and self-reliance, but rather one that supplements responsible African decision-making to achieve Africa’s own goals and purposes. Such a spirit of self-effacing cooperation and enlightened self-interest may be essential to achieve long-term African and U.S. interests, but it will not be easy to cultivate. To do this, it will be necessary for political leaders to commit greater energies to encourage an increased public understanding of Africa’s needs and aspirations, and for a stronger U.S. constituency on Africa to emerge. Only a concerned U.S. leadership and domestic constituency for Africa will be sufficient to overcome the trend toward drift and detachment that has marked U.S. policy toward Africa.