Campaigns in Africa are often referred to as ‘harvesting season’, with candidates distributing everything from cash and rice to T-shirts and calendars. These handouts are often characterised as vote buying, suggesting Africans value democracy so little that they are willing, as one headline from the DR Congo put it, to trade their ‘Vote for a Piece of Soap’. Eric Kramon's work presents a different story, with rational voters seeking to improve accountability and politicians’ performance.

There are, according to Kramon, at least two reasons to be sceptical of pandemics of vote selling in Africa. First, individuals frequently report taking handouts and not voting for the distributor, since secret, Australian ballots shield the content of their choice. Second, parties make little effort to monitor who gets what, with goodies often distributed en masse. An especially memorable story involves a politician dropping cash from a helicopter at a rally. These distributions don't match the quid pro quos we have in mind when we think of vote buying.

Kramon's innovation is his theoretical explanation for this behaviour. Electoral clientelism, according to the book's logic, represents attempts to bolster candidates’ credibility as future distributors of potentially more-valuable largess. In addition, candidates who are able to incur handouts’ expenses will seem more electorally viable, and thus more deserving of strategic voters’ consideration.

The author finds support for this theory through experiments in central and western Kenya, with hypothetical distributors winning broader support from subjects and enjoying higher viability ratings. Further, Kramon finds that electoral clientelism is more effective among poorer voters (who presumably care more about distributions) and that it benefits challengers and backbenchers more than powerful incumbents (who presumably have already established their ‘type’ and viability). Finally, the data also suggest that distributions are more impactful when voters face an ethnically homogenous candidate pool, and thus cannot differentiate between candidates on the basis of ascriptive identity.

Kramon's work presents myriad opportunities to explore other topics related to electoral clientelism, particularly from the perspective of politicians’ decision-making. First, like most literature on distributional politics, the book attempts to explain politicians’ behaviours solely with data on voters’ responses; the rich quotes here contain few, if any, perspectives from candidates themselves. Next, while much electoral clientelism in Africa is distributed en masse, a not insignificant subset is distributed in personalized ways, which would suggest attempts at vote-, turnout-, or abstention-buying. We therefore need a better understanding of how and why candidates mix different distributional strategies. Finally, the book closes with an optimistic – but somewhat tentative – finding, connecting electoral clientelism with higher post-election investment in local public goods. This suggests that electoral clientelism effectively conveys important information to voters. However, it is unclear why candidates who might not prefer to spend heavily on local public goods after victory would still not engage in the distribution of relatively cheap handouts during a campaign. In other words, it remains a bit of a puzzle as to why electoral clientelism might separate the redistributionists from the predators.