This article surveys the role of learning as mechanism of policy diffusion in the context of the creation of a new political order. The author discusses policy learning against the background of recent research on the diffusion of deregulatory and regulatory policies and attempts to distinguish learning from other mechanisms of diffusion. She then surveys the challenges entailed in testing this mechanism and sets out her particular approach: a rational version of learning. She also reports the results of preliminary efforts to test learning as applied to the diffusion of regulatory policies. The author concludes that learning cannot be rejected as a plausible mechanism of the diffusion of policies, although it shares its explanatory role with less rational mechanisms of diffusion, in particular policy emulation. Further research and analysis is needed to test learning in either its rational or its bounded version and, in doing so, to delve into the politics of learning.
10 September 2009 in Review of International Political Economy
In this paper, I enquire whether 37 governments in industrial and in Latin American countries privatised as a result of learning from experience. Using a rational updating model, I examine whether the decision in the 1980s and 1990s to streamline the public sector was the outcome of a revision of beliefs about the effectiveness of privatisation or whether, alternatively, it was triggered by international pressures or mimicry. The results suggest that rational learning and especially emulation were two important factors in the decision to privatise. International pressures, here proxied by the presence or absence of an agreement with the International Monetary Fund and by European Union membership, are irrelevant to explanations of the decision to privatise. Finally, domestic political conditions appear relevant to the decision to launch privatisation but only when the analysis is carried out for each of the regional sub-samples. In the OECD countries, centre-left governments were more likely to privatise whereas in Latin American more repressive regimes were more willing to divest.
In political science, rational learning and bounded learning are commonly studied as two opposing theories of policy choice. In this article, I use a rational learning approach to reach conclusions about bounded learning, showing that the two theories are not necessarily incompatible. By examining a rational learning model and the decisions of a set of developing countries to open up their trade regimes, I show that countries are particularly influenced by the choices of neighbouring countries and by particularly successful policy experiences. These are two typical contentions of the bounded learning literature. I argue that bounded learning and rational learning yield the same results as soon as one drops the rational learning assumption that there are zero costs to gathering new information. I use the discussion on rational learning versus bounded learning as a basis for exploring more general issues concerning the diffusion of policy innovations.
In the 1980s and 1990s, market reforms swept the world. It is widely believed that the reformist wave can be partly explained in terms of the lessons learned from policy failures of the past. Whereas this interpretation of events is well established, it has never been empirically proved. Learning, Policy Making, and Market Reforms is the first study that tests the impact of policy learning on economic policy choices across time and space. The study supports the popular explanation that, on average, governments around the world adopted privatization and trade liberalization, and sustained open capital accounts, as a result of learning from the experience of others.