There is no one good way to give foreign aid. Simply giving individuals money is all the rage at present. It often works well but doesn’t build hospitals or roads. Donors can construct infrastructure themselves of course, but if the host country is poorly governed, today’s glistening infrastructure projects often turn into tomorrow’s white elephants. In theory, donors can solve this problem by using aid to improve governance. It’s excellent – in theory – but in practice it’s very difficult. Another alternative is to ignore the state and fund NGOs. Much like giving people money, this can work. But, once again, there are some things NGOs simply cannot do.

Aid can help. But if donors want it to, they need to choose carefully – matching different approaches to the contexts they find themselves working in.

Of the approaches donors can use, budget support has had a particularly turbulent history: the source of both idealistic aspirations and furious controversies. As a form of aid, it seems straightforward, involving aid agencies simply giving money to the governments of poorer countries to spend as they more or less please. (I’ll return to the “more or less” bit in a second.)

It’s simplicity is its charm: if all their aid were given as budget support, donors would no longer have to worry about managing thousands of projects. Recipients wouldn’t have thousands of other countries’ projects dotted around their social and economic landscapes either. But they would have money to spend on promoting development in ways that were responsive to the needs of their people. Perhaps even, free from the meddling of outsiders, and with enough money to actually do things, developing countries might start governing themselves better.

It’s charming, but nothing in the world of aid is ever that easy. Most aid-recipient countries have governance problems. Hopefully, they won’t one day but for now that’s the reality. Under these circumstances, it would be unwise for donors to simply write blank cheques. (“Oh, sure, you want to buy tanks, no worries, fire away.”) And so budget support has historically come with conditions (or “conditionalities” as they are called in the world of aid). At times these have been ineffectual, ignored by recipients. At other times, driven by contentious orthodoxies, they have been extremely controversial.

Budget support isn’t that simple after all. But there is still a case for using it, carefully, as part of a suite of approaches to giving aid. In their book Retooling Development Aid in the 21st Century: The Importance of Budget Support, Shahrokh Fardoust, Stefan G. Koeberle, Moritz Piatti-Fünfkirchen, Lodewijk Smets and Mark Sundberg make that case.

The book is not an impassioned piece of advocacy. It is thorough and scholarly. (The authors all have academic and/or World Bank backgrounds.) The book covers the history of budget support, situates it amongst the myriad of other similar, and not so similar, means of financing development, discusses different types of budget support in detail, distinguishes budget support given in times of crisis from longer-term approaches, and outlines the options for giving it. As it does all this, the book provides many useful examples.

Beyond the specific examples, the authors also bring systematic econometric evidence to bear on the question of what determines the performance of World Bank budget support programs. (Short answer, careful preparation within the World Bank, and perhaps policy quality at the recipient end; however a lot is unknown about the recipient end.) The authors also look into the ability of budget support to bring successful policy reforms in its wake: here the findings are very mixed (my view) or suggestive of success (the authors’ take).

Although my interpretation of the econometric evidence presented in the book is less positive than the authors’, I was still persuaded that budget support can work when it is well matched to context. There’s clearly a case for it. And the book is at its best when it builds on this case and discusses the practicalities of delivering budget support. In places, the book was too in-depth and a tough edit would have improved its readability. But the detail will be very valuable for readers with specialist interest in the topic.

Hopefully, some of those readers will be in the Australian and New Zealand aid communities. According to OECD DAC data, both countries make more use of budget support than most donors. What’s more, many Pacific countries receive comparatively large shares of their aid in the form of budget support. Like it or not, budget support is big in this part of the world.

Budget support is not the only way aid can be given, but it is an important tool in the kit, and it can be made to work well. Anyone who wants to help make budget support work, or even just understand it, should take the time to read this book.

This article was first published on the Australia National University’s blog site, Devpolicy.org. Terence Wood is a research fellow at the Development Policy Centre.