On-line contribution about independent regulatory agencies

I just recorded a piece for an on-line course on regulation directed by my friend and colleague Antonio Estache from Université Libre de Bruxelles. I had to talk through Skype about independent regulatory agencies and the result will be in this course. I took as a starting point my 2010 working paper where I tried to summarize mine and others' (Montoya, Stern, Gual, Levine...) research on the topic. In that paper I argued that independent regulatory agencies were one of the possible means to alleviate the commitment problem in regulation: firms being reluctant to invest in sunk assets anticipating opportunistic regulatory behaviour. That is not the only rationale for independence, as the need to appoint experts in complex industries is another one. Through history, there have been other ways to try to alleviate the commitment problem, such as public ownership, popular capitalism or detailed legislation, but the preferred one in the recent decades has been regulatory independence. In between conversations trying to record a good cut, I learned from Antonio that the World Bank (where he worked for many years) tried to promote the UK version of the institution rather than the US version, because in the latter there were too many political appointees. The institution has a moderate but positive impact on investment according to the empirical literature, but it is not free from problems. One of them is isolation from the rest of government, which may prevent necessary coordination. Another one is the enhanced risk of capture, since the revolving doors phenomenon may show up precisely because independent agencies need to be staffed by experts who may have also a career in the private sector. Beyond my 2010 paper, I argued that currently there is promising research in trying to incorporate insights from behavioural economics to the field, in particular three departures from full rationality by regulators: failure to optimize (because of satisfying or adaptive behaviour); intrinsic preferences (for example, regulators may have not only career concerns but also a public sector ethos); and expert bias (for example, overconfidence in their own skills or tunnel vision). And another hot topic is the controversy between regulators accepting having a variety of objectives (as argued by Massimo Florio), or regulators being given only one single objective, as argued by Jorge Padilla and others in a recent paper published by the Brussels think-tank Bruegel.