The see-saw effect and the difficulty of reforming bad governance

The chapter by Baland, Moene and Robinson in the Handbook of Development Economics, devoted to governance issues, has a very good explanation of the difficulties of reforming bad governance. One well-known issue is the need to take into account not only de iure institutions but also de facto practices (as their account of central banking in Sierra Leone or Zimbabwe illustrates). But less well-known mechanisms are the path dependence of practices even when the elites change, or the fact that elites find new ways to satisfy old objectives when their old instruments are removed. This is called the see-saw effect, and here’s is how they explain it: “Many poor growth experiences are accompanied by a system of dysfunctional laws and regulations and other aspects of governance. An obvious idea might be to directly intervene in these components of governance and promote change in laws and regulations. This was the sort of reasoning that led to the famous Washington consensus some of whose components, for example, privatization of state enterprises, deregulation, and legal security for property rights all seem related to governance. The first pitfall of reform is that directly reforming specific institutions, policies or aspects of governance may not be sufficient, and may even backfire. The reason why such reforms may be ineffective is that it is usually not a coincidence that some aspect of governance is bad. Bad governance is probably fulfilling some political objective. But there are many different ways and a multitude of instruments to achieve a specific goal. Taking away one instrument without altering the balance of power in society or the basic political equilibrium can simply lead to the replacement of one instrument by another with little net effect of the ultimate goal - economic performance.”