I teach in a public policy school, so hear constantly about leadership. But economists don't even have the language to talk about how or why leadership matters in economics--even though I guess that deep down most of them think it is very important.
Recently there has been some interesting work that may signal a change. In an empirical paper, Ben Jones and Ben Olken link shifts in growth performance in countries to changes in the national leader. They find that exogenous leadership transitions (induced by the natural death of existing leaders) generate effects on policy and growth, particularly in autocracies. Surprise, surprise, right? Well, at least it has drawn economists' attention to it.
And now, there is a theoretical paper by Sumon Majumdar and Sharun Mukand which looks at how individual characteristics and "objective" circumstances interact in creating leadership "opportunities." Here is the abstract:
Individual leaders have been central to the transformation of organizations, political institutions and many instances of social and economic reform. In this paper we take a first step towards analyzing the role of leadership to ask: when and how does a leader engineer change? We show that while underlying structural conditions and institutions are important, there is an independent first-order role for individual agency in bringing about change and thus transforming the institutions. We emphasize the key nature of the symbiotic relationship between followers decisions' to willingly entrust their faith in the leader and the leader's initiative at leading them. This two-way interaction can endogenously give rise to threshold effects; slight differences in the leader's ability or the underlying structural conditions can dramatically improve the prospects for successful change. Given the centrality of this leader-follower relationship, we further explore conditions under which an individual may deliberately prefer to follow an ambitious leader with divergent interests rather than a benevolent one with congruent preferences. Thus by virtue of having followers, both `good' and `bad' leaders may be effective at bringing about change.
A couple of nice things about the paper. One is that it shows how bad leaders can emerge in equilibrium as well as good leaders. Mugabe versus Mandela, in other words. Another is that it identifies an important threshold effect: small differences in leaders' ability and in circumstances can make a big difference to the possibility of reform.