The academic literature in monetary economics is replete with the Taylor Rule and its alternative forms. When academics profusely cite a fellow academic’s work it means it has substantial relevance, because academics are usually very reluctant to give their colleagues credit. Their bar is very high. This is the way the scientific method usually works. Ideas, when subject to systematic scrutiny, are either elevated at some point in their life to theories from the draft status of scientific hypotheses or are discarded. Only genius becomes law. It is not personal. Even the greatest were not exempt from it, as long as intellectual integrity in the process is maintained.
Often scientific hypotheses become de facto laws (or self-fulfilling models which distort reality because of misapplication) and this is a pitfall largely due to politics in academia rather than the intrinsic worth of the ideas themselves. It is useful, therefore, to separate the politics from the reality of economic hypotheses such as the Taylor Rule.
The Federal Reserve does not make monetary policy using the Taylor Rule. The Rule is only a useful benchmark, as John Taylor himself calls it. It is a macroeconomic benchmark because it uses macroindicators of the economy: output gap and deviation from an inflation target. And output gap is a treacherous thing. It is the difference between two variables, not between two, certain, constants. Both the variables, the current output and the potential output, are endogenous to policy, which, in the case of the Federal Reserve, is the federal funds rate. The Rule ends up as a circularity which cannot be resolved without the microeconomics of financial regulation.
The inflation gap between expected inflation (and at times current inflation) and the inflation target, whether that target be implicit or explicit and transparent, is also a policy choice. The deviation could be zero or of little concern if the central bank prefers to tolerate higher inflation to support growth (setting aside the usual debate about its consequences for the purposes of this argument, because in some circumstances higher inflation can be explicitly tolerable as long as the reasons for the level which would be tolerable are ex ante clear. This is one of those times.).
Taken together, the reality of output gap and the deviation of inflation make the Taylor Rule itself a policy choice, not a rule. The mechanistic view of the Taylor Rule breaks down in the face of judgment in real time, especially in the econometric white noise of a crisis.
Then what is a central bank to go by? The answer is judgment under incomplete information taking into account all indicators to make policy, which is inherently forward-looking and normative in the immediate term with which central banks are usually concerned by their mandates. The baseline comparison to the Taylor Rule is always backward looking, once the data becomes available, which is with a lag just as monetary policy also acts with a lag of at least 6 to 9 months per the Milton Friedman conjecture (a conjecture, not a law), under normal circumstances.
At the recent American Economic Association (AEA) meetings, the Chairman of the Federal Reserve, was indeed correct (even though he should not have used the Taylor Rule to make his case but a narrative of the policymaking process: “what was Ben thinking when he did what he did?” and that is a pertinent debate to have at the moment if the Fed is to change how it can do things differently going forward by building on its successes such as TALF and discarding its failures such as TARP) that oversight matters to ensure that risks in the markets are managed better. The enforcement of proper financial regulations can mitigate the effects of bubbles, even if regulation may not be able to prevent them, through the monetary transmission mechanism itself: by raising the quality of broad money because of better regulations and their enforcement, contrary to the recent opinion of a distinguished academic economist turned media pundit.
Just as there are detailed indexes for any number of economic variables, even if these are not compiled by the Fed, it is feasible to achieve an aggregated index of systemic risk based on the risk levels of the various types of financial market functions engaged in by the markets. If this were not true, we will not have a measure we can rely on for either inflation or output, the very essence of the Taylor Rule.
The individual risk levels of a few thousand financial institutions are far easier to compile and analyze than the spending or investment habits of millions of people within the borders of a country. Therefore, if inflation and output can be expressed quantitatively, so can both individual risk and systemic risk. Some of this is already done by the Fed at the level of individual banks which it supervises and the financial institutions also measure their risks using established methods. If a systemic measure can therefore be made available due to transparency in the markets instead of the currently pervasive text book causes of market failure, perhaps it could be useful to include it in the Taylor Rule as a third variable to put a weight on it.
Just as was the case after the Great Depression, the government should not have to fly half-blind to make policy to deal with the now more complex financial markets. Better data is one of the purposes of financial regulatory reform. In devising a system for obtaining it, we could also achieve better regulation to catch up with the many changes in the market place since the ‘80s.