COVID vaccines: an example of a most needed independent regulator

My modest experience in doing empirical work has taught me humility and skepticism. That is because often data tells you that what you thought was probably completely true, actually gets qualified a lot by empirical evidence.
That happened to me when I was studying the empirical effects of Independence in practice for telecom regulators in Latin America, with my friend and colleague Miguel A. Montoya (mostly in his PhD thesis, which I supervised). In my previous theoretical work on the topic, with Jon Stern and Paul Levine, we concluded that independent utility regulators were as important, if not more, than independent central bankers. However, once we accounted for a continuous measure of independence in practice (difficult), and once we accounted for the endogeneity of regulatory independence (difficult), the effect on investment in telecoms was significant, but very modest. There were probably things, at least in Latin American telecoms, that were more important for investment than the independence of the regulators.
There are many other examples of empirical results that disappoint the theoretical or policy expectations. That is the case for example of microcredits, as explained by Esther Duflo in her Nobel Prize speech (published in the last issue of the American Economic Review). Or it is the case of small class sizes to have a positive impact on learning results, as explained in the econometrics textbook by Stock and Watson. Does this mean that independent regulators, microecredits or class sizes never matter? No, they still may matter a lot in some cases, and actually empirical work can help understand the heterogeneous effects of these treatments.
Sometimes, we may  have to make a judgement about the effect of these treatments in particular cases, in the absence of data. It would be very interesting to study what is the impact of expert commissions on policy in general; my bet is that this impact is very small, as these commissions are many times an exchange of legitimacy for the politician for ego for the experts, but of course in some cases experts are a matter of life and death, like in a pandemic. One particular issue in which regulatory independence seems to me crucial now is in the approval and distribution of a vaccine for COVID-19 in the USA. The combination of a context where corporate lobbying is a massive phenomenon, vaccines are produced by private sector corporations, and there is a presidential election where the incumbent president is desperate for votes (and not known for his respect of democratic institutions) creates a context that will require, at least for once, really insulated Independence of the Food and Drug Administration, which, if I am not wrong, is the agency in charge of approving the vaccine. I can easily imagine a nightmare scenario where Trump accelerates the approval of a vaccine in collusion with a corporation to boost his election prospects some weeks before the election.