The New Institutionalism – Bates and Galiani (2014)

Published

Bates and Galiani reflect on the nature and influence of institutions, and ultimately their value in promoting economic development. Their starting point is a review of the ‘New Institutionalism’ literature (See North 2003). While this literature contains important insights, Bates and Galiani argue that it is now in “conceptual disarray.” Specifically, they point to the numerous and often divergent notions of what constitutes an institution, how it is created and what its influence is likely to be. They also argue that the literature does not sufficiently explore how “coercion can be socially productive,” how institutions role in shaping agents’ choices may be beneficial (or not).

From this two-pronged critique, Bates and Galiani draw positive insights. First, they argue that much of the conceptual confusion surrounding institutions—notably the question of whether they are chosen or inherited—can be resolved by identifying what temporal scale scholars are using. They suggest that a narrow temporal focus on the period when institutions form or are reformed by agents is consistent with a view of institutions as chosen; by contrast, a “longue durée” perspective illustrates how institutions, once formed, persist and also structure the strategic decisions available to agents.

Bates and Galiani’s second insight is to stress the need for a better sense of the role of politics in institutional creation and performance. Institutions—even the same institution—may have a positive or negative impact on economic performance depending on how political actors manipulate them. For instance, coffee marketing boards in Kenya under President Kenyatta were value enhancing, while in Uganda under Amin they were extractive and a drain on growth. The key determinant of performance was not the institutional form itself, but rather the incentives of the political leaders, who could either support growth enhancing activity or discourage it. For institutions to perform, the politics must be right.