William Niskanen, a proponent of NGDP targeting

In the last couple of day’s there´s been quite a few critical comments on NGDPT followed by replies from Bill Woolsey, Nick Rowe, Ryan Avent, David Beckworth and Scott Sumner. Instead of adding to the list I´ll take the hint provided by Bill Woolsey in his own eulogy of William Niskanen who passed away on October 26. Bill put it like this:

Reading the many eulogies to William Niskanen on the web recently, I have not seen too many mentions of his support for nominal expenditure targeting. It is sad to lose a long time advocate, just when the approach is gaining attention.

Niskanen´s first article on NGDPT was in 1992. In the conclusion he writes (my bolds):

One of the most important lessons of political economy is that a government must have at least as many policy instruments as it has goals. The Fed has only one policy instrument, specifically the level of the monetary base. It is important to focus this instrument on the single, most important, achievable goal of monetary policy. That goal, I suggest, is a stable path of nominal domestic final sales (my note: call it NGDP level targeting). At the same time, it is important to recognize that we need to put our fiscal and regulatory house in order if we are to achieve our other economic goals.

Nine years later, in 2001, he wrote “A test of the demand rule”. It is a very easy to read and straightforward article, full of insights for the present. His conclusion, referring to the “instability” caused by Fed actions beginning in 1998 echoes what Market Monetarists have been saying for some time:

We have had an important test of the demand rule. It worked remarkably well for six years subject to various shocks. And it ended when the Federal Reserve chose to respond to several other shocks. Why did the demand rule work so well for so long? And why should the Federal Reserve depart from the target demand path by responding to some types of shocks but not others? We should learn from this experience.

It appears we haven´t. The present state of “instability” is magnitudes worse and more painful than the one Niskanen was analyzing (and that one was still four years away from being “resolved”).

Niskanen begins by making a very straightforward, no frills, argument (my bolds):

It is important to recognize that a demand rule is consistent with any desired price-level path, including a stable price level. My primary point is that a demand rule is potentially superior to  price rule, whatever the desired price-level path, because of the different response to changes in supply conditions. A central bank following a demand rule would not respond to either positive or negative supply shocks; such shocks would lead to a one-time change in the combination of price and output changes in that year, but would not lead to a long-term change in the inflation rate. A central bank following a price rule, in contrast, would increase the monetary base in response to a positive supply shock and would tighten the base in response to a negative shock, thereby increasing the variance of output. Similarly a demand rule is potentially superior to a money-supply rule because it accommodates unexpected changes in the demand for money. The general case for a demand rule, thus, is that it minimizes the variance in output in response to unexpected changes in either supply or demand.

In my illustrations of Niskanen´s “test of the demand rule” I extend the period backwards. The story starts with the strong NGDP growth following the 1981-82 recession. By early 1987 the economy had been placed on a level path along which growth proceeded at a rate close to 5.5%.

The picture illustrates.

In late 1987 there was the stock market crash and in 1990 the invasion of Kwait and the spike in oil prices. Those were significant shocks that jostled aggregate spending, but the “plane” remained under control, so when 1992 came along “flying” was “smooth and easy” i.e., the demand path was maintained, all the way to early 1998. As Niskanen states: “In retrospect this was an extraordinarily successful period of monetary policy, marked by a very steady growth of demand and a continued decline in both inflation and unemployment”.

The illustrations below confirm.

Then, suddenly, after early 1998, demand veers off its path. Looking at the pictures above one could say that the Fed reacted to the steep drop in inflation and lowered interest rates. But on the other side, real growth remained robust (unemployment was still falling), so that posed a dilemma. Maybe, as Niskanen surmises, the Fed appears to have responded to the near collapse of LTCM and the Russian default, with the monetary stimulus that ensued leading to significantly higher demand growth.

In mid 1999, with inflation on the rise and unemployment still falling, there was no “dilemma” so the Fed put rates up. In mid 2000 demand growth reversed. Niskanen´s analysis ends in early 2001. Much was still to come to compound on the 1998 MP “mistake. Nevertheless Niskanen leaves us with a thought and a suggestion that have a bearing on the present predicament. This is his thought about “Fed bias”:

The issue whether monetary stimulus was a necessary or appropriate response to the several financial crises of 1998 (but not, interestingly, to those of 1997) will have to be deferred to a later discussion.

For the moment, I will leave you with a personal episode that illustrates the bias of those who serve in the Federal Reserve. During the period that I served as acting chairman of the Council of Economic Advisers, I was “summoned” to the American Enterprise Institute office of the venerable Arthur Burns. He may have heard that I had opposed the bailout of Continental Illinois in the spring of 1984; in any case, his primary guidance to me was that I should never, under any circumstance, fail to support a bailout of a major financial institution. He claimed that the bailout of Franklin National Bank was the most important decision during his own disastrous experience as chairman of the Federal Reserve. Robert Rubin is reported to have had the same obsession about avoiding the collapse of a major financial institution. My point is that a fear that the collapse of any major financial institution may lead to a general financial panic overwhelms any concern about the deferred moral hazard effects of bailouts and may introduce a bias in favor of a more expansive monetary policy.

Avoiding the collapse of a major financial institution and the possibility of a more general financial panic may be worth a little inflation, but I do not know. I am skeptical about the alleged contagion effects of the failure of a major financial institution, but I simply do not know the answer to this question. In 1998, in addition, but not in 1997, the Federal Reserve seems to have been confused about whether it was the central banker for the world or only for the United States, and I hope that it has recovered from that confusion.

And his suggestion on “what the Fed should do”:

For the moment, given the acceleration of demand growth since early 1998, the important remaining issue is “What should the Fed do now?” My position is that the Fed should not try to offset the high demand growth of the period from early 1998 to mid-2000. This would require unusually low demand growth for a period until demand was back on its prior trend; the future output costs of such a policy seem higher than any value of reversing the prior price-level effects of high demand growth. Maintaining a demand rule, I suggest, requires offsetting small differences from the target path but not as large a difference as has developed since early 1998.

Instead, the Federal Reserve should try to establish and maintain a new 5.5 percent demand growth path based on the current level of demand. For several quarters through early 2001, this seemed to be what was happening. However, given the sharp decline in demand growth in the second quarter of 2001, some additional monetary stimulus may be necessary. Alan Greenspan may yet pull off another miraculous “soft landing.” I keep reading about a recent firming of demand but I do not yet see it in the data through the second quarter of 2001. In any case, it is important to restrain future demand growth to no more than a 5.5 percent rate if we are to avoid a new higher trend rate of inflation.

With regards to Niskanen´s “thought”, it appears Bernanke tried to avoid “a little inflation” from rising demand when avoiding the collapse of a financial institution by “shooting a neutron bomb” i.e., keeping financial “houses” intact, only “killing” demand.

With regards to his suggestion, the results would be terrible. In the present case we should be content with remaining far below any reasonable trend level. Unemployed people, I´m so sorry!