I can't praise Poor Economics by Abhijit Banerjee and Esther Duflo highly enough. Although its subject is economic development, all economists interested in any aspect of public policy should read it.

The book describes the policy prescriptions development economists can derive from the randomised control trials advocated by the Abdul Latif Jameel Poverty Action Lab (J-PAL), which they co-founded. Its perspective is similar to that of another book reviewed here recently, More Than Good Intentions, although the latter was more focused on the application of behavioural economics to development policy, and in particular how to market different types of behaviour to “customers” for aid or policy intervention. However, what I found most interesting about Poor Economics was the evident applicability of its approach to economic policy anywhere.

The book is organised into thematic chapters – healthcare, education, financial services (both borrowing and saving) and entrepreneurship.

The chapter on education is outstanding. It describes a number of experiments to assess the effectiveness of different interventions – proving textbooks, school uniforms, deworming pupils, taking photos of teachers to reduce their absenteeism. It also addresses some puzzles about why it appears to be so hard to persuade people – either parents or schools – to adopt some behaviours which clearly deliver a high economic return. The answer lies in two insights. One is that low-income parents make a correct assessment of the average return to education but greatly over-estimate its variability: “They see education as a lottery ticket, not as a safe investment.” (p87) Often, families will educate only the one child they consider to be the best bet academically. The other is that school systems have low expectations of most poor pupils, and teach to the elite. “References to a certain old-fashioned sociological determinism, whether based on caste, class or ethnicity, are rife in conversations involving the poor,” the authors comment (p91). This is just as true of the English teaching system: the worst poverty is the poverty of expectations of many pupils. The good news is that there is evidence that it is perfectly easy to teach every child the basics in school as long as that is the focus of education, and as long as there is detailed and continual assessment of every child.

I also found the chapters on borrowing and saving very interesting. The book is mildly positive about microfinance, although this attitude has earned its authors criticism from some MFIs, which in some cases will not brook anything other than unthinking advocacy. The book notes that the requirement for zero default makes MFIs too inflexible for many would-be borrowers, but equally that the zero default structures are what enable them to keep their interest rates lower than a commercial or moneylender rate. The Indian system of requiring banks to lend a minimum proportion of their loan book to priority sectors gets some support here, despite the complaints of the banks, because it helps small but potentially growing businesses access to working capital. Often these are exactly the kind of firms (and this also is true in the UK) too small to be of interest to the banks. The chapter concludes: “Ultimately, this problem stems from the structure of banks. Because they are, by nature, large organisations, it is hard for them to provide incentives to their employees to screen firms, monitor projects and make worthwhile investments.”  (p180) With microfinance and mobile payments schemes having improved access to financial services for individuals and very small businesses, midi-finance is the next challenge.

Most of all, though, I liked the modest conclusions. Banerjee and Duflo end by saying there is no Big Idea solution to poverty, and much that we don't know. One critique of the experimental approach to development economics is precisely that it only delivers small answers to context-specific problems, rather than generalisable answers. (This theme did emerge in the recent Boston Review forum on the insights of behavioural economics for development, around an essay by Rachel Glennester and Michael Kremer.) However, the obvious answer is that there has been a dangerous tendency in economics to draw big, sweeping answers from context-specific evidence. Greater humility becomes us. Many of the lessons of the J-PAL approach should be applied to economic policy in the rich economies as well.

My one slight disappointment in the book is that its only perspective on the role of the private sector in economic development concerns policies for small enterprises, including microfinance. All poor countries need to industrialize in order to develop their economies, and will need to do so through a mixture of inward investment (for the capital but also the technology and knowledge transfer) and growing domestic businesses which start off as part of the supply chains to those multinationals. This might be classed as macroeconomics and therefore out of scope for this book, but big businesses also play a part in alleviating individual poverty, whether through Safaricom's M-PESA or Unilever's single sachet products. What's more, although the obsession with silver bullet solutions is one thing that has marred development economics over the decades, so is its obsession with solutions lying in the hands of governments and NGOs alone, with no role seen for the private sector.

It's a good sign, though, to want more of a book, rather than less! In addition to the book, there is a Poor Economics website with excellent teaching resources, which I'll certainly be using, and data.