Zambian chiefs play a key role in delivering development, but they are not a neutral technology for achieving this end. They are also political actors, as evidenced by the attention they receive around elections. Candidates prioritize visits with chiefs in every area they visit during their campaigns. Politicians of all stripes bring gifts – in kind or in cash – to the chief in the run-up to elections. Political leaders are quick to publicize statements from chiefs suggesting that they have their support.

These types of behaviors understandably may lead observers to conclude that chiefs are powerful electoral vote brokers. By many accounts, politicians try to mobilize votes by winning over the support of chiefs, who then use their material and psychological influence over voters to ensure that they vote as directed. It would be deeply troubling for the democratic process if chiefs are able to deliver the votes of community members to whichever politician they chose.

However, electorally powerful chiefs are also consistent with the alternative model of chiefs as development brokers. In this view, traditional chiefs have political influence not because they can direct voters to support their preferred candidates via coercion or pressure. They have electoral power because of the crucial role they play in facilitating the work of elected politicians. Traditional chiefs play a key role in helping politicians to coproduce local public goods. As a result, political candidates must be able to work effectively with chiefs if they are to be successful representatives. Other things being equal, politicians will be able to work better with chiefs with whom they have a better preexisting relationship. For this reason, chiefs' opinions of politicians are politically influential, but their effects on elections are an indirect function of the influence of these relationships on the delivery of development.

This chapter examines the influence of Zambian chiefs on electoral results, considering whether it is more consistent with the vote-broker or the development-broker model. It begins by showing – as both theories would predict – that there is a relationship between chiefs' preferences and electoral support for candidates. But historical examples and interviews with politicians suggest that the influence of chiefs is more qualified than the vote-broker model would predict.