He contends the following. That the Great Moderation was due to adherence to the Taylor Rule [and to ‘rules-based’ fiscal policy]. That during the early 2000s, monetary policy was set looser than that prescribed by the Taylor Rule. This caused the build up of debt and risk-taking, which ultimately led to the bust, and the end of the Great Moderation. Weak activity following the crisis has been due to departures from rules based monetary policy, in the form of unconventional monetary policy. And departure from rules based fiscal policy, in the form of the fiscal stimulus enacted by Obama in 2009. These departures have created uncertainty that has weighed against activity. Tighter policy on both counts would have led to more buoyant activity during the recovery on account of being more certain.
I think he’s wrong on every point. And I doubt many at all in the mainstream macro profession, even the conservative strands of it, will agree with him.
1. Research on the causes of the great moderation are split. Those that take explicit macro models to the data tend to stress good policy. (But sill leave a sizeable role for good luck). Those that are more agnostic, and in the empirical macro tradition, stress good luck. Even to the extent that good policy is stressed it does not follow necessarily that adherence to a Taylor rule was what delivered the goods. John’s claim that macro performance during that era was due to adherence to his rule was highly dubious even before the Great Moderation’s end.
2. John’s rule was shown to deliver pretty good results in variations on a narrow class of DSGE models. The crisis has cast much doubt on whether this class is wide enough to embrace the truth. In particular, it typically left out the financial sector. Modifications of the rule such that central bank rates respond to spreads can be shown to deliver good results in prototype financial-inclusive DSGE models. But these models are just a beginning, and certainly not the last word, on how to describe the financial sector. In models in which the Taylor Rule was shown to be good, smallish deviations from it don’t cause financial crises, therefore, because almost none of these models articulate anything that causes a financial crisis. How can you put a financial crisis in real life down to departures from a rule whose benefits were derived in a model that had no finance? There is a story to be told. But it requires much alteration of the original model. Perhaps nominal illusion; misapprehension of risk, learning, and runs. And who knows what the best monetary policy would be in that model.
3. In the models in which the TR is shown to be good, the effects of monetary policy are small and relatively short-lived. To most in the macro profession, the financial crisis looks like a real phenomenon, building up over 2-2.5 decades, accompanying relative nominal stability. Such phenomena don’t have monetary causes, at least not seen through the spectacles of models in which the TR does well. Conversely, if monetary policy is deduced to have two decade long impulses, then we must revise our view about the efficacy of the Taylor Rule.
4. John puts great store on the effect of uncertainty on weak-post-crisis activity. But in the models in which the TR is shown to be good, the effects of changes in uncertainty are small. Highly unlikely to be a major cause of a drop in activity in the region of 5-10%. Of course, it’s possible that the effects of changes in policy uncertainty are large and the models are wrong. But in which case, we need to revise the class of models we used to evaluate the TR in the first place. A fair argument can also be made that QE and fiscal stimulus did not aggravate policy uncertainty. It provided the stimulus that monetary policy was not able to, but would normally, absent the zero bound, have been expected to. The coincidence of the Fed and Treasury’s actions with the elimination of elevated spreads on risky assets is at least good circumstantial evidence of this alternative view. It is plausible that the fiscal stimulus helped aggravate the wars between the left and right in Congress over the debt ceiling, and that, other things equal, this depressed demand. However, relative to a situation in which there had been no discretionary stimulus at all, the US economy was surely still better off [goal variables closer to target]. Of course, the ideal policy is one in which there is certainty over the fact that discretionary fiscal stabilisation steps in when conventional monetary policy has run out of room. Freshwater macro people don’t agree with that; but this is the conclusion in the family of models in which the Taylor rule is shown to do well. Taylor can’t draw comfort from the freshwater rejection of the efficacy of fiscal policy, since that line of thought asserts that prices are flexible and active central bank policy of the kind JT wants is at best an irrelevance.
5. John seeks tighter monetary policy in the form of no quantitative easing, and no fiscal stimulus. In the models in which the TR was shown to be good [rational expectations models with sticky prices], doing this would have moderate, but damaging short run contractionary implications, taking goal variables further away from target than they were or are.
6. John does not address worries about the fact, that, in models in which the TR was originally shown to be good, which ignored the zero bound to interest rates, it was also shown that adherence to the Taylor Rule could lead to the economy becoming perpetually trapped at the zero bound to nominal interest rates. This is unfortunate, since we know now that this trap is a real possibility.
7. It’s highly contestable that the Fed set too-loose monetary policy in the early 2000s. Bernanke made a stern and convincing case in favour of what they did while still Fed chair. He pointed out that if you substituted inflation for forecast inflation in the Taylor Rule, for which a convincing case can be made that one should, you find that Fed policy was not too loose. Specifically, rates were so low because the Fed were worried about deflation, and the zero bound. They had watched what they saw as slow and weak Bank of Japan monetary policy, and had seen its consequences, and were doing what they could to avoid that experience being repeated.
8. Not only does John’s ‘monetary policy caused the financial crisis’ thesis go against the theoretical models [in which his rule was shown to be good]. It also goes against the consensus of evidence about the vector autoregression evidence of the effects of an identified monetary policy shock. So far as I know, monetary policy shocks are not measured to have large, 1.5 decade-long [2000-2015], destabilising effects on the financial sector.
9. When pressed about the meaning of the #audittheFed requirement for the Fed to declare a monetary policy rule, he responds that this rule is not meant for religious adherence, but to provide guidance. Yet the research which he and others did studying central bank conduct has nothing to say about central banks that do and should depart in discretionary and judgemental ways from the rule. If those departures happen and are thought necessary, then this has to be because we think the world works in ways different from the models in which the TR was observed to do well. What are those differences? How important are the discretionary departures from the rule in a model that articulates them? This could be a detail, or it could be very important. If it’s a detail, I’d like to have it explained to me why.
10. JT’s thesis, that the TR is great in theory, and was responsible for the Great Moderation, expresses a confidence in this research that is not warranted. Although I personally don’t subscribe to the freshwater view of macro that prices are flexible, I do have a lot of sympathy for their view that the modelling of sticky prices is often superficial and question-begging. And certainly not sound enough on which to build a case for legislating to tie the hands of the Fed. Moreover, JT ignores doubts cast on the framework by John Cochrane’s hard but profound papers on equilibrium selection in monetary policy rule modelling. His view of the story that lies behind the equilbrium JT and others study so much is that it is a load of nonsense. And he can’t be dismissed lightly.
11. For reasons that escape me, John Taylor seems to relegate the misunderstanding of finance, and financial regulation, to the status of a detail in the list of likely causes of the financial crisis. Yet there are so many reasons to believe that this was its primary causes. Not least that in the models we know and love, real, low-frequency problems generally have real causes. John also has to contend with the freshwater economists on this matter, since they – mostly believing in flexible prices – would view monetary policy as the detail. To give a few examples, we are asked to believe that keeping interest rates a few tens of basis points below what JT argues – controversially – should have been the case, was responsible and not: politically motivated subsidies for sub-prime lending; failures in credit ratings of sub-prime securities; failures in the regulation of retail banks with new, aggressive wholesale funding models, and lax lending standards; failures in the regulation of investment banking functions which had unappreciated and systemic importance; instability caused by the failure of institutions to internalise the effects of their fire sales on the system as a whole once the crisis had begun; and by Knightian uncertainty over things like the value of asset backed securities and who was exposed to what. The impact of the export of vast flows of savings ‘uphill’, from emerging market countries, on asset prices, asset pricing models, regulatory capacity and so on. This is a short extract from a very long list of causal factors unrelated to monetary policy. But the general idea should be plain.
For me, it’s ironic that John calls his talk ‘a monetary policy for the future’. John does not seem to have absorbed the lessons of the recent past. The crisis has reopened questions about the appropriate macro model and the macro policy conduct within it that were previously thought to be settled. That there should be closer adherence to this old answer to monetary policy questions is a conclusion that I think very few indeed will draw.