To understand the implications of different aid modalities, we need to analyse the bargaining logic between donors and recipient governments

Budget support has given rise to vigorous debates among practitioners and academics. Scholars have sought to analyse why this aid modality was adopted (e.g. Clist et al., 2012; Dietrich, 2013); whether or not it is effective (e.g. Dijkstra, 2018; Koeberle et al., 2006; IDD and Associates, 2006; Ronsholt, 2014); and under what conditions it is likely to be abandoned and/or suspended (e.g. Koch et al., 2017; Molenaers et al., 2015). Beyond budget support itself, these debates raise important questions about why donors and recipients favour specific aid modalities, and the impact of their choices on development policy. Dijkstra critically engages with our article “The rise and fall of budget support: Ownership, bargaining and donor commitment problems in foreign aid” (Swedlund & Lierl, 2020), published as a part of a special issue on “Ownership in a post-aid effectiveness era: Comparative perspectives”, guest-edited by Niels Keijzer and David Black. We appreciate the discussion of our work and the evolving debate, which will help to advance our understanding of different aid modalities and their implications for development policy. In Swedlund and Lierl (2020), we argue that the key to understanding the implications of budget support for development policy priorities is the underlying bargaining logic between donor agencies and the governments of recipient countries. Our premise is that foreign aid, regardless of whether it is provided in the form of budget support or project aid, always involves political bargaining between donors and recipients. In other words, aid never comes without political strings, and those strings are not unilaterally determined by the donor. For example, in exchange for aid, recipient governments may agree to political reforms that make governance more inclusive—an outcome that western donors value highly, but recipient governments are reluctant to engage in, because it limits their power and their ability to cater to their domestic political clientele. Donors, on the other hand, may make concessions with respect to demands for political reform, if in exchange they get more say over how the government spends its money; for example, if the government is willing to shift more resources towards development goals, such as health, infrastructure, education or social protection. We argue that donors are particularly concerned about the fungibility of aid. Hence, they care not just about reforming the political process, but also concretely about how and on what the government spends its money. Moreover, we argue that donor commitment problems—i.e. donors’ incentives to renege on their aid promises in the long-term—constrain bargaining between donors and recipients. When this constraint is relaxed, different bargaining compromises become feasible. This is where we see the central transmission channel through which budget support affected donor–government bargaining: budget support was perceived to decrease donors’ commitment problems. In other aid modalities, donors have far greater incentives to dishonour aid commitments or shift their goalposts once the recipient government has satisfied their demands, because donors continue to control the planning of the interventions and the flow of resources. Through budget support, once disbursed, donors cede control over the use of aid provided to recipient governments. Our empirical research in Rwanda and Tanzania confirmed that, while other forms of aid commitments were heavily discounted by the recipient governments, budget support was initially seen as a much more reliable form of aid and therefore more attractive to them. Since budget support was more valuable to recipient governments, they were willing to give donors greater policy influence in exchange for it. But, this also had implications on political inclusivity, particularly the inclusion of civil society actors in debates about fiscal spending. Dijkstra takes issue with our proposition that budget support, by altering the bargaining compromise between donors and recipients, could lead to less emphasis on political inclusiveness. She argues that this proposition is “counter-intuitive,” asking “Why would donors give up the possibility of having political influence if they provide more predictable aid?” While we agree that our proposition is counter-intuitive, our model provides plausible conditions under which this would happen. If donors’ bargaining power increases, because budget support allows them to lock in their commitments, they will not automatically insist on greater political inclusiveness. This is because the recipient government is willing to offer them far greater concessions in another area they care about: a seat at the table in government budget discussions and thus more government spending in line with donors’ priorities (also see Swedlund, 2013, 2017). This is very attractive to donors because it helps to alleviate their concerns about aid fungibility. Hence, donors are at least to some extent willing to trade insistence on political inclusiveness for greater influence over the government’s expenditure decisions. Why would governments offer donors greater influence over spending decisions, but not over political reform issues? In our model, the first reason is that governments receive positive utility from development spending (because more spending is always better), but negative utility from political inclusiveness (because governments prefer to be subjected to less accountability). Hence, as donors’ bargaining power increases, it is more attractive for governments to commit public funds to conform with donor priorities than to allow for greater political inclusiveness. The second reason is that political inclusiveness would diminish the government’s marginal utility from the remaining discretionary spending, because it implies greater accountability. If the government faces an unabated need to satisfy its political clientele, a shift of public spending towards donors’ priorities would imply that the government now needs to satisfy its political clientele with even fewer discretionary funds. Hence, the remaining discretion over public spending becomes all the more valuable to the government, which becomes more determined to defend it against donor demands for political inclusiveness. This is consistent with what we observed in Rwanda and Tanzania. Dijkstra further argues that our article did not prove that the main reason why aid-recipient governments preferred budget support is the reduction in commitment problems. We do not dispute—and would in fact agree—that recipients favoured budget support for a variety of reasons. We never claimed that there was only one reason. Our model formalizes one mechanism through which budget support can affect development policy—by reducing donor commitment problems in donor–government bargaining. The choice of aid modalities, however, is exogenous to the model and no assumptions are made about the reasons why budget support is adopted. Rather, the model highlights that budget support, if it is believed to reduce donor commitment problems, facilitates a bargaining compromise between donors and recipients that leaves both sides better off. While not designed to explain the adoption or abandonment of aid modalities, the bargaining model does have implications for the conditions under which donors would find it worthwhile to provide budget support—namely, if it leads to a more attractive bargaining outcome with the recipient government than other aid modalities. This may depend on many factors, but donor commitment problems appear to be one of them. When exogenous factors, such as the global financial crisis and changing political priorities in the donor countries, made it evident that budget support was unable to fully resolve the problems of donors’ commitment, our empirical research suggests that previous bargaining compromises that were predicated on donors’ ability to make credible promises tended to break down and were followed by renewed bargaining. Finally, Dijkstra critiques our article by claiming that newer work on budget support proves its effectiveness. While aid effectiveness is an important point of discussion, it is not what we studied in Swedlund and Lierl (2020). Rather, we provided a theoretical framework for understanding the implications of different aid modalities, such as budget support or project aid, for bargaining on development policy between donors and aid-recipient governments. Our research thus complements, rather than contradicts, previous research that has analysed budget support from an aid effectiveness perspective (e.g. Dijkstra, 2018; IDD and Associates, 2006; Koeberle et al., 2006; Orth et al., 2017; Ronsholt, 2014). Nor is our bargaining framework limited to budget support, but can be adapted to study other aid modalities. It was never our intention to make a normative judgment about budget support, but rather to provide a novel theoretical perspective. In the future, we hope to continue a fruitful debate on the implications of different aid modalities for development policy

Business · Economics · Modalities · Political science · Public economics · Social science · Sociology · Computer Science · International Development and Aid · Poverty, Education, and Child Welfare · Religion, Society, and Development

• How the money flows

  Open Access•Rahwa Kidane, Robbert Biesbroek et al.•Global Environmental Change•2026

  Add to list

• The Development Dance

  Open Access•Haley J Swedlund•Development Dance•2017

  Add to list

• The rise and fall of budget support

  Open Access•Haley J Swedlund, Malte Lierl•Development Policy Review•2020

  Add to list

• What Determines the Suspension of Budget Support

  Open Access•Nadia Molenaers, Annie Gagiano et al.•World Development•2014

  Add to list

• Bypass or Engage? Explaining Donor Delivery Tactics in Foreign Aid Allocation

  Open Access•Simone Dietrich•International Studies Quarterly•2013

  Add to list