Imbeau, Louis-Marie (1989), Donor Aid – The Determinants of Development Allocations to Third World Countries: A Comparative Analysis, New York: Peter Lang

These findings emerged from the comparison of the two cases and from the comments of the legislators and other key actors interviewed as part of this research:

Furthermore, while the majority of respondents from the early 1960s were quick to give credit to their parties for electoral assistance, none discussed the parties in terms of serious legislative voting commitments. Willful agents, rationally pursuing self-determined and sometimes personal goals were much in evidence in both periods of the study, and the issue coalitions of actors are not described accurately by party or interest group simplifications, even in the early 1960s. An important aspect of this analysis is that the parties could not deliver on policy changes by themselves.

We have seen that aid allocations from a variety of donors and for a variety of time periods, samples and definitions of aid show surprisingly consistent biases against poorer and larger recipients. While much of the cause of these biases is political per se, there are a surprising number of non-political factors which appear to contribute substantially to the biases. From the perspective of those who feel that a poor person living in a larger country is no less worthy of assistance than a poor person in a smaller country and that poorer countries should receive more rather than less aid, it would be desirable if the ‘development community’ were more conscious of these biases and their causes, and of what might be done to mitigate, if not eliminate them.¹⁸ In this connection, the recent emphasis on countries ‘Most Seriously Affected’ by the oil situation and on the (terms as well as volume for the) ‘Least Developed’ countries is fairly encouraging for prospects for further initiatives of this type.

It should be noted that mitigation or elimination of these biases does not require that other, otherwise desirable criteria be dropped. For example, a medium-term shortage of foreign exchange, taking account of increased needs from accelerated growth, should be a necessary condition for concessional foreign resource transfers. Donors should still be quite concerned about the quality of projects, but should consciously bias their own project development activities to favour the lower-income countries which have serious human resource and institutional constraints on project development. Similarly, donors could take account of ‘self-help’ or ‘performance’ (although it should be recalled that the measure of self-help used in this study turned out not to be significant in explaining past allocations). Such self-help ratings would, of course, be quite different from conventional past ratings which focused almost entirely on factors relating to GNP growth per se, and would include commitment to, and steps to meet, the needs of the poor. There is a danger, though, of inadvertently reintroducing the middle-income bias. For example, given the five criteria used to measure performance in Cline and Sargen (1975) – savings, inflation, exports, taxes, and capital-output ratio – it is not surprising that the four countries in their 19-country sample which would have their aid increased as a result of good performance were Taiwan, Korea, Iran and Malaysia, in that order. This inadvertent apparent middle-income bias occurs in an article which, otherwise, comes out strongly against the middle-income and country-size biases. Similarly, some of the poverty-focused criteria which have recently been suggested to measure performance, such as infant mortality (uncorrected for level of development), would also strengthen the middle-income bias; many such ‘social indicators’ are highly correlated with per capita income, and would in effect allocate aid as a reward for past performance, to those who need it the least.

While the quantitative anlaysis here has been on the basis of DAC assistance, it follows from the behavioural arguments for the middle-income bias and from the trade-related behavioural argument for the country-size bias, that the distribution of non-aid economic benefits, as from trade preference or commodity agreements, are likely to suffer from these biases even more than aid allocations do; if sufficient care is not taken by rich and poor countries, measures undertaken as a part of the ‘New International Economic Order’ may leave poorer and larger countries relatively worse off.

The analysis here is also likely to be relevant to allocations made by OPEC donors. While oil-exporting nations obviously have their own political uses of aid, one can hope that with the advantage of being ‘late-comers’ they can avoid at least some of the more inadvertent biases of past allocations. For example, some among the new donors may have an understandable preference for financing ‘bankable’ projects but limited project-development staffs, which would thus introduce a bias toward the middle-income countries best able to generate a large volume of bankable projects.

In conclusion, there is a need now for development economists and others with an interest in the ways in which the rich nations can assist the poor to focus specifically on the desirable inter-country allocation of such assistance and the political and practical issues of the transition to the allocation desired.

The grant component of bilateral aid has increased considerably over the last decades, reaching 97% in 2005. Although bilateral donors in general provide aid in the form of grants, the grant ratio varies greatly across recipient countries. Recipient need explains this variation of aid composition to a certain extent, as poorer countries appear to receive a higher grant component of aid between 1975 and 2005. However, the study finds no evidence that recipient-country indebtedness explains the variation.

The latter finding is surprising as the recent grant-versus-loan debate focuses on debt sustainability in recipient countries. While the debt-sustainability problems in developing countries have called for the initiation of wide debt-relief programmes, this study fails to provide evidence that the allocation of the bilateral grant–loan mix has been tailored to avoid future debt problems. In addition, the recent debate focuses on the shift from loans to grants for multilateral organisations such as the International Development Association of the World Bank (see, for example, the Meltzer Commission report). This study shows, however, the importance of focusing not only on increasing grants but also on the allocation of the grant–loan mix.

Recipient need in terms of income level is nevertheless a significant determinant of the grant component of aid. Poorer countries receive a higher grant component in all three periods. This finding is in line with the suggestion of Radelet (2005) who argues that the income level of recipient countries should be the crucial factor in determining the allocation of the grant–loan mix, as it avoids perverse incentives and provides a simple allocation rule. Although a significant determinant, the analysis shows that the effect of recipient-country income level is quite limited. At most, a decrease in GDP per capita of 10% increases the grant component of aid by about one percentage point.

In all, the results suggest that the recent grant-versus-loan debate should focus not only on a total increase in the grant component of aid but also on the actual allocation of the grant–loan mix to recipient countries. By allocating a higher share of grants to already indebted countries and a higher share of loans to countries able to absorb loans, donors could scale up development assistance and at the same time probably reduce future debt problems in recipient countries. The policy recommendation is thus not to scale up grants on a general level, but for donors to be more selective in allocating grants and loans with respect to recipient needs and characteristics.

Jolly, Curtis M. and Gadbois, M. A. (1989), “Foreign Aid as a Promotional Strategy,” in The Review of Black Political Economy, Vol. 18, No. 1, pp. 59-74, New York: Springer.

The study has shown that the commercial and economic interests of aid can be modelled if trade is expressed as a function of aid. An advertisement model can then be used to examine the response function. If such an approach is adhered to, and donors are more open about their real motives for aid-giving, more information will be made available for long-term distribution planning. Long-term plans will result in a constant flow of external aid investment in recipient countries and thus will have a greater effect on both donor and recipient countries.

The results of the study showed that aid flows and per capita income of the recipient countries have positive effects on trade flows. Hence, it is in the self-interest of the donor country to stimulate economic growth in the developing economies, and therefore, contribute to a healthy international trade market. Another conclusion from the study was that the aid donors studied have been efficiently allocating their aid budgets to the recipient countries, and that donors will be willing to increase allocations only if they see that it is to their best interest. The use of an advertisement model in measuring the economic and commercial interest of aid shows promise.

Kang, Seonjou (2007), “Agree to Reform? The Political Economy of Conditionality Variation in International Monetary Fund Lending, 1983-1997,” in European Journal of Political Research, Vol. 46, pp. 685-720.

IMF conditionality has been criticized for its inflexibility and insensitivity. However, a close look at conditionality agreements reveals that conditions attached to loans from the IMF have varied qualitatively and quantitatively. This study attempted to explain the thus-far-overlooked variation in IMF conditionality. Understanding conditions attached to IMF loans is important because it paves the way to assessing not only the appropriateness of IMF programs for addressing economic problems in borrowing countries, but also compliance with the programs. With this study, the whole spectrum of IMF conditionality is covered, from borrowing countries’ decisions to participate in IMF programs to the outcomes of those programs.

This study used a political economic framework to examine the variation in IMF conditionality. Although the IMF has its official economic rules to determine conditionality, its institutional arrangements steer it away from those rules. The five countries that fund most IMF operations exert influence to determine IMF conditionality outside the official rules in order to advance their strategic interests vis-à-vis developing countries and, as a result, IMF conditionality comes to vary across programs. Thus, between 1983 and 1997, such influence from the five countries made IMF conditionality easier for borrowing countries that showed national attributes important to the five countries than those that did not. The five countries and borrowing countries have a common interest in promoting goals with IMF conditionality beyond economic reform, and this is why, despite the poor compliance and disappointing results, conditionality agreements do not disappear, but continue to be contracted even today.

These findings have significant implications for how to improve IMF conditionality. Critics have ascribed the disappointing results of IMF conditionality only to the IMF’s economic model from which specific conditions are shaped, and they have recommended that the IMF modify its economic model for successful structural adjustment. However, we now know that the political economic influences were as responsible as the economic model for the failed structural adjustment in developing countries. Loaded with a necessity for compromise from the complementary strategic interests between the five countries and borrowing countries, conditions were not properly formed to address economic problems in developing countries. The seeds of failure were planted by the political economic process of IMF conditionality as well. Therefore, the solution is ‘let the IMF do what it does best with autonomy’. Improving IMF conditionality should incorporate chances for the IMF to design macroeconomic reforms that truly address economic problems in developing countries. To do so, the IMF has to enhance its institutional capacity to counter internal and external factors that will pervert its program design. However, there should be no illusion that this can be expeditiously achieved.

In this part of the paper, we shall sum up the topics treated in the foregoing pages with special reference to the attitudes of the political parties. Further on, we shall try to identify the forces underlying these attitudes and finally we shall give our conclusions regarding the alleged trend described as a change from the ‘traditional Swedish system of development assistance’ towards the ‘general OECD model’.

It is fairly easy to describe the policies of the political parties. It is far more difficult to discuss and analyze the determinants of these policies. No research has been undertaken on these questions. What we shall say here is based to a large extent on our own experience and judgments.

First, the volume of development co-operation. Among the political parties, the Moderate (Conservative) Party has proposed at least a temporary abandonment of the one per cent target. This position is logical, bearing in mind that the party is the only Parliamentary group advocating a substantial reduction of taxes and consequently in the role of the public sector in the economy. Taking into account the fact that the Moderates also argue in favor of increased defense spending, almost no other public-sector activity can, in their view, be spared reduced spending if budgetary equilibrium is to be reached and maintained. In Moderate Party thinking, foreign borrowing plays an important role in the back ground to such proposals. Since the oil crisis, Sweden has resorted to foreign borrowing and, so the argument goes, it is irrational and almost immoral to finance development aid with borrowed money. Bearing in mind the existence of unused carry-overs in the aid appropriations, the Moderate Party argument goes on, a decrease of the share of development assistance in the GNP would have limited, if any material consequences.

The Moderates have had only a limited and temporary success in obtaining a reduction of the share of development assistance in the GNP. On one occasion, a Parliamentary majority consisting of Moderates and Social Democrats succeeded in decreasing the share of development aid, but only temporarily. Violent opposition from the Liberals, the Centre Party and the Communists and – probably more importantly – from groups within the Social Democratic Party forced the government to return to the one per cent target.

The one per cent target has been considered to be a first stage. If and when economic conditions allow, the appropriations should be increased over and above that target. However, only the Liberals and the Communists are now talking about a two per cent target, though they do not mention a timetable for its attainment.

It seems reasonable to assume that there will also be in future a strong majority in Parliament in favor of maintaining the one per cent target. Whether there will be a majority for a higher target in the foreseeable future is very doubtful.

Now the quality of the development co-operation programme. To those who favor a lessening of the aid burden on the Swedish economy, an increased share of Swedish-oriented allocation is a measure to be considered, bearing in mind the near impossibility of abandoning the one per cent target. Logically, the Moderates have advocated increased tying and also a kind of tying of local-cost-financing. Such financing means in practice that convertible Swedish kronor are placed at the disposal of the recipient country, which uses the corresponding sum in its own, non-convertible currency. Thus, the recipient country will receive convertible currency, which can be used for purchases in countries other than the donor country. This problem, however, has not been of major importance in the debate, as the other parties accept local-cost-financing as an efficient means in the aid programme. The Social Democrats have been very responsive when in government to demands for aid which will produce increasing return flows to Swedish industry. This attitude can largely be explained by the Party’s giving priority to creating employment and by the strong connections between the Party and the blue-collar trade unions.

It would be tempting to explain the Moderate Party’s attitude as a result of the Party’s links with industry. No doubt such links exist, but to attribute the Party’s policies to them only would be over-simplistic. As an example, the Moderates have criticized the alleged concentration of the aid programme on so-called Socialist countries, in spite of the fact that the return flows from such countries are comparatively high. A special case in point is the development co-operation with Vietnam, in which the Party advocates an immediate cessation of the aid programme because of the Vietnamese military presence in Kampuchea and, as they see it, the Vietnamese Government’s lack of respect for human rights. One of the biggest projects supported by Sweden is the Vinh Phu pulp and paper mill in Vietnam, which is being built in close co-operation with Swedish industry. The immediate abandonment of that project would result in the loss of a good showpiece in the region for Swedish industry and, worse, in the probably failure of the project, with possibly negative consequences for Sweden’s industrial reputation in the region.

The introduction of balance-of-payments support among the instruments of the aid programmes has also been greeted with much skepticism by all the non-Socialist parties, in spite of the fact that the support has hitherto been largely tied to procurements in Sweden.

Another important quality aspect has been the country-programming system. The system undoubtedly decreases the flexibility of the donor country, preventing it from easily switching aid funds from one country to another at short notice. All the non-Socialist parties have argued that donor flexibility should be increased and consequently that the country-programming system should be considerably modified or in practice abandoned. Lately, however, at least the Liberal Party seems to have accepted the system, which has already been slightly modified without changing its main characteristics.

Generally speaking, among the parties represented in Parliament, the Liberal and the Communist Parties have the best records with regard to the ‘purity’ and the ‘recipient flexibility’ of aid. The Centre Party has normally adapted itself to the Liberals’ views. At the other extreme is the Moderate Party. The Social Democratic Party has a rather mixed record. It was only with some hesitation that the Social Democrats accepted a fixed timetable for the attainment of the one per cent target. They have been very ready to listen to demands to increase the share of Swedish-oriented appropriations. With regard to the country-programming system, the Party has successfully defended ‘recipient flexibility’ ad coherent planning against their critics.

Against this background, it may seem rather astonishing that the non-Socialist governments between 1976 and 1983 did not substantially change the aid policy. The explanation is that during the whole period the Liberal Party was in command of development-aid questions. Another essential explanation is that development co-operation is not a very important item in Moderate Party policy, other aspects having a much higher priority, for example, the level and design of the tax system and the national defense. The Moderates thus seem to be able to accept the maintenance of both the quantity and the quality of aid, provided that they can have an influence on other matters which are of greater importance to them.

In this context, it is worth mentioning that the LDCs play a comparatively modest role as regards the Swedish economy. Sweden is greatly dependent on foreign trade – about 30 per cent of Sweden’s GNP is exported. The recipients of aid through SIDA have almost exclusively been chosen from among the poorest LDCs, which – with the exception of India – are not very important in the world’s foreign trade. Consequently, the actual ‘return flow’ of about 40 per cent of total ODA is probably very satisfactory from the point of view of export industry. A substantially higher rate of return flows would only be possible if the aid programme was considerably restructured with less regard for the recipients’ own needs. Such a restructuring is not likely to occur at any foreseeable political juncture.

What must be considered more important from the point of view of industry is the absolute and relative growth of Swedish-oriented appropriations. Industrial and trade-unionist pressure groups will probably try to influence the Moderate and Social Democratic Parties to maintain the present level of return flows and also to increase the share of Swedish-oriented appropriations. Another aspect mentioned in international discussions is the so-call in-effect tying, meaning that the donor tries to influence the use of non-tied aid in favor of industry in the donor country. As regards the Swedish attitude in this respect, it seems likely that a recipient who wishes to use Swedish funds for purchases outside Sweden, in cases in which Swedish industry is normally competitive and interested, must have rather strong arguments for doing so.

The importance of Swedish-oriented appropriations, as regards the attitudes of different actors, is also illustrated by the fact that discussions on projects and recipients are practically limited to aid through SIDA. BITS and SWEDFUND seem to be protected bodies in this respect. In principle, this is rather astonishing, as it has been underlined that the guiding principles of Swedish development aid are equally valid for all types of aid, that is not only aid through SIDA and SAREC but also aid channeled through BITS and SWEDFUND.

The Swedish administrative system and tradition normally afford vast spheres of action to the central government agencies. Bearing in mind the composition of their boards, it is obvious that a director-general – normally the president of the board – has a potentially powerful political base at his disposal, which can be used to exert considerable influence on political decisions. Being responsible for the day-to-day operations of the agency, the director-general can choose the ways in which proposals will be presented to the board. The system thus provides ample opportunities too for the bureaucracy to influence political decisions. It should also be mentioned that a director-general may be a person with a distinguished career in the civil service but also wholly or partly a politician.

In this context, it is interesting to note that the government has reserved for itself a comparatively strong position on the boards of the institutions responsible for the use of ‘Swedish-oriented appropriations’, that is BITS and SWEDFUND, and also seems prepared to envisage measures which will enable it to control better the management of the central agencies. There is no reliable information available with regard to the role of the bureaucracy in this context. Experience seems to show, however, that the influence of the bureaucracy is limited or virtually nil when it comes to very important political questions, such as the choice of recipient countries. There are indications that the bureaucracy has played some part in such issues as the quality and quantity of the total aid appropriations.

It is obvious that no political party can successfully operate without regard to public opinion within or outside the party’s constituency. Swedish public opinion over the years has had and still has a positive attitude to development co-operation and the 1986 opinion poll shows an increase in this positive attitude. It may seem surprising that the desire to cut the appropriation to development assistance is less than the willingness to decrease defense spending. It is also interesting to observe that environmental deterioration is the first concern expressed by participants in the opinion poll, followed by poverty and starvation in the world. Concern about unemployment risks has increased considerably in comparison with the 1985 poll. This has evidently not negatively influenced the willingness to give development aid. Furthermore, positive attitude towards the effectiveness of aid showed a marked increase in 1986.

It is difficult to determine whether these positive changes in public opinion represent a lasting trend or whether something has happened to influence it temporarily. It might well be argued that media reporting on, for example, the drought catastrophes in Africa and conspicuous relief activities, such as Band Aid, have had a spill-over effect on development aid as such, including government aid.

It is quite clear, however, that young people, women, older age-groups, members of religious organizations and other NGOs for an aid constituency sufficiently large to prevent serious qualitative and quantitative decreases in the aid programmes. Groups who in principle would be prepared to argue in favor of such changes probably find themselves in a reasonably satisfactory position, bearing in mind the political situation.

The aid constituency has a considerable influence within the Liberal Party, the Social Democratic Party, the Centre Party and probably also the Communist Party. The small Christian Democratic Party is also heavily influenced by such groups, while the Moderate Party seems to be more influenced by groups which are not particularly interested in development aid or have accepted the present structure and volume of development assistance as a fact of political life.

The Government’s information policy has greatly increased the opportunities for pro-aid NGOs to argue their case. Two-thirds of the SIDA appropriation for information on developing countries and development aid is given to NGOs to be used as they see fit.

Questions may also be raised regarding the relationship between development of aid policies and that of public opinion on aid. The trend towards an increasing use of aid appropriations for Sweden-oriented programmes has no doubt shocked many of the most fervent adherents of development co-operation in its pure form. On the other hand, Sweden-oriented activities, including not only those mentioned above but also other forms of exchange, no doubt contributed to opening up Third World countries to new groups of Swedes.

In it January 1987 bill on development co-operation concerning the programme for the fiscal year 1987/88, the government, under the heading of ‘mutually beneficial aid’, refers to the changes which have taken place since the mid-seventies in the following terms:

Within the aid programme new instruments have been added implying an increased emphasis on extended economic, technical and scientific co-operation and personal contacts. … This also contributes to a growing understanding of the needs of developing countries and the often difficult conditions of development as well as to an improved awareness of the know-how and the resources that Sweden is able to provide. This helps, together with the engagement of NGOs in aid activities, to maintain and strengthen the will to give aid in Sweden.

Has there in reality been a drift of Swedish development co-operation from the ‘traditional Swedish system’ towards the ‘general OECD model’? This question has more to do with the quality of the aid programmes. An answer to the question must take into account the fact that the ‘general OECD model’ has also changed over the years. There is also a greater willingness among donors other than Scandinavians to make long-term commitments, sometimes on longer terms than, for example, Sweden is willing to do. A good example is the European Community. A greater willingness to provide non-project assistance and sometimes also assistance with local costs may be discernible. It would be interesting to analyze to what extent the discussions within the DAC have influenced donors in these respects.

This being said, it is our conclusion that Swedish development assistance has drifted towards the ‘general OECD model’ during the period dealt with in this paper. Nevertheless, the aid constituency in Sweden has been comparatively successful in defending genuine development co-operation. There is still a ‘Swedish model’, and this model is being shared and sometimes bettered by some other donors.