The sun was coming down over the ancient forum of Rome as I finished reading Haley Swedlund’s book. I am not sure I could have imagined a more appropriate setting, given the focus of The Development Dance on the arena in which donors and recipients come together to reach agreement.

Swedlund’s core insight is that both donors and recipients have trouble honoring the promises they make to one another. This uncertainty over whether counterparties will actually honor their commitments leads both aid givers and aid receivers to seek mechanisms that bind the other to agreements. Swedlund further argues that the inability of many of these mechanisms to resolve the commitment problem is central to understanding the faddish rapid rise and fall in popularity of particular aid delivery mechanisms.

The author’s writing is crisp and engaging, and she weaves her central metaphor of the negotiation process as a dance between donors and recipients throughout. Drawing on more than 150 semistructured interviews in four countries (Ghana, Uganda, Tanzania, and Rwanda) and a novel survey of more than a hundred Heads of Cooperation from 23 donor agencies, Swedlund convincingly argues that while donors and recipients may not have equal power, neither donors—in her framing, the lead in the dance—nor recipient government partners can unilaterally impose their will on the other. As she puts it, “It takes two to tango” (p. 18).

After introducing the argument (Chapter 1), the theory underlying it (Chapter 2), and her research methods (Chapter 3), Swedlund turns to empirics. In Chapter 4, she illustrates that both sides have power in the aid bargaining relationship. Particularly compelling is her description of recipient countries as active and important, if unequal, participants.

In Chapter 5, Swedlund convincingly argues that credible commitment problems exist and are salient to both sides as they engage in aid negotiations. Recipient governments are justifiably concerned, for example, that donors may “move the goalposts,” requiring more from recipient countries than initially agreed; if so, promised aid may also fail to materialize. For their part, donors similarly worry that recipient governments may not carry through reforms to which they have committed; this may be because agreements were made in bad faith, or because well-intentioned reformers overestimate their ability to alter long-standing equilibria, or the speed with which they will occur. For both aid donor and recipient nations, governing coalitions or policy may change; a new head of state, or a new minister, may abandon the plans of his or her predecessor.

In Chapter 6, Swedlund argues that commitment problems are a key factor in the rise and fall of budget support—direct transfers from donors to recipient country treasuries for recipients to spend, rather than hiring nongovernmental organizations or consultants to deliver aid projects, for example—as a popular aid delivery mechanism. In this chapter, Swedlund’s data seem at times underutilized. While the author illustrates both the rapid rise and rapid decline of the budget support “craze,” as she puts it, the causal role that commitment problems played in this craze is less clear.

At issue, in my view, is not whether budget support programs suffered from commitment problems; they clearly did. But did these commitment problems cause the rapid fall in the use of budget support as an aid delivery mechanism? We could imagine a number of potential causes for budget support’s rapid decline, including lobbying by rich world NGOs that had been cut out of the aid delivery process, or increasing worries by governments in the Organization for Economic Cooperation and Development regarding the possible risk of funds flowing to bad actors in the wake of 9/11. While Swedlund describes her methodology as process tracing, she does not walk chronologically through any narrative of budget support’s rise and fall. Instead, she samples her data in what I think of as repeated cross sections, making it hard to see the causal logic of any particular case or the narrative of any given country.

Throughout the book, I wondered whether Swedlund might have done more to exploit the richness and novelty of her data. One question central to her thesis that the data might have explored in greater depth is why some donors (e.g., the European Commission) have remained champions of budget support. Have these continued users of budget support avoided the credible commitment problems that have befallen other donors, or do these donors persist in their use of budget support despite these commitment problems? In either case, the answer might have shed further light on the credible commitment problem central to the book and the role of this commitment problem in the choice and sustainability of aid delivery mechanisms.

More generally, as Swedlund acknowledges, donor agencies are a heterogeneous group, as are recipient country governments. This heterogeneity offers empirical leverage on the central concerns of the book: We might imagine that some donors, like multilateral aid agencies or bilateral donors with stable ruling coalitions, have longer time horizons and face fewer commitment problems. Recipient countries, including the four countries where Swedlund’s interviews take place, also have varying levels of stability. To extend the author’s dance floor metaphor, the shape, agility, and desires of the dancers clearly impact how the dance proceeds. To me, at least, not using her rich data to explore these dynamics feels like a missed opportunity.

In the concluding Chapter 7, Swedlund notes the ability of new and growing donors (e.g., China) to alter existing bargaining dynamics. She also discusses the rise of a new trend in aid delivery, Results-Based Aid, through the lens of credible commitment problems. These discussions suggest but a few of the directions for future work opened up by the work. Swedlund’s bargaining model, and focus on credible commitments as a key feature of aid bargaining, can serve as tools for practitioners and scholars alike. As she puts it in her typically elegant prose, “if we want to improve the delivery of foreign aid, we have to break open the black box of donor-government relations” (p. 131).

While I believe that this book may not be fully convincing in what Chapter 7 describes as its second purpose—to tightly link credible commitment problems to the political sustainability of aid delivery mechanisms—I believe it is quite convincing on its first aim. It does indeed advance “our understanding of what donor agencies and recipient governments bargain and negotiate over and how these negotiations affect choices regarding how foreign aid is actually delivered to beneficiaries in recipient countries” (p. 123). It is the rare book that prompts one to realize assumptions implicit in a literature one knows well but had never considered. Swedlund is right, in my view, that almost all work on aid agencies and their interactions with recipient governments frame the problem through delegation: of aid agencies to their agents, or from agencies to recipient countries. The author’s bargaining frame, by simultaneously animating both donors and recipients, opens the door to a more nuanced understanding of donor–recipient dynamics: to understanding what steps each takes in the dance, with what results, and how they affect the couple’s position on the dance floor.

Scholars interested in the interactions of developing and developed countries or those interested in the outcomes of aid efforts from a wide variety of disciplines—for example, those with interests in international relations and political economy; public administration and service delivery in developing counties; development studies and development economics—will all benefit from this engaging book. Practitioners involved in the aid bargain will also almost certainly benefit from this work, which brings what I suspect many will find to be a useful, broader frame to their lived experience. Swedlund has taken a big step forward in explaining the dance of development assistance, and her book deserves to be widely read.