Let me start this post with a warning. As indicated by the title, it will involve semi-serious thoughts, which in this case is equivalent to semi-humorous thoughts. So the contents are intended as a sort of economists’ joke (which may not be funny to that many, if any, besides me). Also, in order to understand the fun, it will require some knowledge about graduate dynamic macroeconomics, more specifically the continuous-time Ramsey-Kass-Koopmans model. With this warning, I proceed.
In recent macroeconomic literature on monetary and fiscal stabilization policies, researchers often characterize the optimal stabilization policy in a conventional public-finance fashion. Then, many argue that such a policy is too complicated to understand for the general public, and as a consequence, they consider the performance of so-called simple rules. These are equations that relate the policy instrument(s) to some (preferable only a few, hence, “simple”) macroeconomic variables. A well-known example in monetary policymaking is the Taylor rule, which prescribes that the nominal interest rate should respond to inflation and an output measure only. The strength of the responses is determined by what mimics the nominal interest rate movements in the US in the late 1980s.
It is, of course, of interest to assess the performance of various simple rules for macroeconomic stabilization, and for example assess the robustness of their performance across different models. (Although I have always had the reservations that a) if a simple rule performs almost as well as the optimal policy, the model is probably unrealistically simple, and b) if a simple rule performs well across many models, then these models probably do not differ that much after all.) From one slightly deeper methodological perspective, however, I have problems with simple policy rules. In particular when they are applied to models with optimizing private-sector agents under rational expectations. Why is it that these agents presumably are unable to understand the properties of optimal policy, but “only” a simple rule? This is usually not spelled out clearly.
Indeed, in Lars Svensson’s critique of Taylor rules, “What is Wrong with Taylor Rules? Using Judgment in Monetary Policy through Targeting Rules”, Journal of Economic Literature (2003), he argues that it is long overdue that researchers modeled policymakers as being as sophisticated as the private sector (the “targeting rules” referred to in his title are essentially optimality conditions for policy). I would stretch that argument by saying that if one loosely (very loosely) thinks about who is capable of doing dynamic optimization, then it should be the policymaker rather than the private sector. Therefore, why not turn everything upside down, and free the agents in the economy from solving dynamic optimization problems and just let them follow simple rules? Then, the policymaker could perform a policy that will result in optimality. Ideally, this will relieve the private agents from use of brain resources (which can then be spent on something else), and burden the policymaker with some extra computations. Or maybe not, because in the standard analyses, the policymaker actually does all the complicated math. Afterwards, however, it ties its hands and follow a simple rule. So, it could actually be a pure win-win situation.
I thought about this on a sunny afternoon during a beautiful conference in Spain in 2000, and I wrote down a small note that applies the Ramsey-Cass-Coopmans model. In the note, I free the private agents from intertemporal optimization exercises, and let them follow a simple Keynesian consumption function. The policymaker then takes on the task of choosing a tax policy that secures that the optimal allocation is attained. It works nicely!
The note was called “How to lift the burden of intertemporal optimization from consumers to the policymaker” and can be downloaded here (pdf, 80 Kb). Please note that it is meant as fun, but it nevertheless reflects my dissatisfaction with the arbitrariness, which in some instances, plagues the application of simple rules as a representation of public policy.