Scott Sumner  

Roger Farmer on NGDP futures targeting

The Pathos of Doing the Bes... What's Killing Us? A Huemer Gu...

I had the good fortune of meeting Roger Farmer last year, when he was still teaching at UCLA. We had a great discussion of Keynes's ideas. (I seem to recall we both thought he was misunderstood, and that the General Theory focused more on wages and nominal spending, rather than prices and real GDP.)
Prosperity for All.jpg
Unfortunately I've been so busy with two book projects that I haven't had much time to revisit his work, which includes a fascinating recent book called "Prosperity for All". Cloud Yip recently interviewed Farmer as part of his "Where is the General Theory for the 21st Century?" series, and the interview included this interesting passage:

When they started intervening in the MBS market, the stock market began to rise. When they stopped intervening in the MBS market, the stock market slowed down again. A policy of this kind can and should be pursued in the future. There is more than one way to intervene and I am not sure which is the best way. I have advocated intervention in the equity markets, but others, Scott Sumner and Bob Shiller, for example, have advocated instead that we create a market for GDP futures. That market doesn't exist yet. If it did exist, it would be a good substitute for operating in the stock market.

Q: Do you think that the central banks buying and selling NGDP futures would be a better policy, compared to interventions in the stock market?

F: If you could create a thick enough market for NGDP futures, then yes. There is a lot of skepticism over whether that would be feasible.

The reason for both kinds of interventions is related to the connection between the asset markets and consumption. Traditional Keynesians think, or thought, that consumption depends on income. In the 1950s and the 1960s, with the work of Milton Friedman on the Permanent Income Hypothesis and Franco Modigliani on the Life Cycle Hypothesis, we learned instead that consumption does not depend on income, it depends primarily on wealth.

The asset markets are highly developed in western economies, and those people who would be buying and selling NGDP futures will also be buying and selling stocks. Arbitrage opportunities would cause interventions in one of those markets to spill over to all of them.

When wealth fluctuates and stays up or down persistently, those wealth changes feed into consumption, and consumption feeds into employment. Asset price fluctuations, caused by animal spirits, become self-fulfilling. In my view, intervention in the asset markets operates through wealth effects. I remain eclectic as to the best way to intervene in these markets to stabilize asset price movements.

I don't agree with everything in Farmer's book (I am more sympathetic to natural rate models, for instance), but it's exactly the sort of thought provoking, outside the box work that we need more of. The last thing the profession needs is a new DSGE model with a slight twist, which tells us almost nothing about how the profession was so far off base in 2008.

One of the things I like best about Farmer's work is the focus on asset prices. I believe that an increased focus on asset prices offers a way forward for macro in the 21st century.

PS. Bloomberg recently quoted me in a piece on liberal/conservative opposition to appointing Kevin Warsh as Fed chair:

"It is not obvious why he would be a good choice," said Scott Sumner, the director of the monetary policy program at the Mercatus Center, a free-market oriented research center at George Mason University in Fairfax, Virginia. "He was given a chance to do a good job at a lower level, and did poorly."
Reporters must boil down long interviews to single quotes, so let me mention that I also pointed to the fact that he did not have any qualifications for a job in monetary policy when he was appointed to the Fed in 2006. I'm not someone obsessed with credentials, and I'd be willing to support someone lacking normal credentials if there was some other evidence of their ability. What's so troublesome about Warsh is that he did very poorly in his time on the Federal Reserve Board. No credentials and a poor track record---do we give that man the most important economic policymaking position in the world, or keep the woman who is highly talented and is doing a decent job in getting the economy close to the Fed's inflation/employment targets?

Comments and Sharing

COMMENTS (10 to date)
Kevin Erdmann writes:

I would contend that in the short term, equity markets are, roughly speaking, an NGDP futures market, and that the wealth effect from rising equity prices is a side effect of stable or recovering NGDP expectations. One big benefit of a robust NGDP futures market would be the added clarity it would bring to that question.

Mark Bahner writes:
...or keep the woman who is highly talented and is doing a decent job in getting the economy close to the Fed's inflation/employment targets?

Yes, has a Fed chair ever left the position so close to the inflation/employment targets?

Scott Sumner writes:

Kevin, I'm not convinced by that. I've seen lots of examples of huge swings in equity prices where NGDP expectations didn't seem to move much at all---like the past 12 months, or 1987.

Mark, Perhaps Greenspan 2006, but then he retired after 19 years.

Kevin Erdmann writes:

"Roughly speaking".

But, that is why the NGDP futures market would be so useful. It would help to identify times when shifts were due to NGDP expectations and times when shifts were due to other factors. Certainly many shifts are due to NGDP expectations.

It seems to me that being able to see that relationship would be a huge help in getting past populist sentiments about the Fed "bailing out" Wall St. when accommodation is needed. And, if the stock market rose without a corresponding stabilization of NGDP expectations, that would be informative too.

Cloud Yip writes:

Thanks a lot, Prof. Sumner!!

Actually, there is another thing that I have talked with Roger Farmer that I would also love to hear your opinion.

Farmer prefers the notion of "Involuntary Unemployment", which is quite rare in recent days. What do you think about the term "involuntary unemployment"?

mariorossi writes:

It seems optimistic to assume that NGDP markets would always be perfect forecasts to be honest. Any market is only as rational as the people trading it and I doubt that it would be immune from excess volatility in some cases.

Of course the main difference is that the equity market is a long term NGDP forecast, while you are proposing a more short term forecast. Short term forecast are inheretly lower volatility, but again I am not entirely certain the central bank can ignore the long term altogether...

I am not totally sold on the NGDP future market viability: many investment banks tried clever market structures in the past and it never worked (Goldman had a totalizer based one in the 2000s which was very clever). I think embedding them in debt instruments (like inflation linked bonds) seems more likely to be succesful, but I have not paid enough attention to your proposals I freely admit.

Mark Bahner writes:
Mark, Perhaps Greenspan 2006, but then he retired after 19 years.

And he was 80 years old. :-) So if Janet Yellen doesn't get another term, it would appear to be unprecedented.

Thaomas writes:

I think the instrument used -- purchase of government ST liabilities, government LT liabilities, foreign exchange, equities, NGDP futures, Price level futures -- is secondary to the target being followed. I suspect that any instrument could be made to work to achieve steady PL growth, steady NGDP growth, or some other combination of of price level and real GDP growth. (NGDP= 1*PL*1*RGDP).

Farmer's ideas are interesting but he utterly fails to consider the political economy of how the Fed operates. Would a change in instruments automatically lead the Fed to fulfilling its current mandate or any other mandate it were given?

Steve F writes:

The case you make against Warsh as quoted in the Bloomberg piece is probably the kind of case that would resonate with the Trump administration. They don't necessarily care about credentials, but they do care about track record and results.

Focusing on stuff like that might get the content into Trump's ear.

Scott Sumner writes:

Kevin, I agree.

Cloud, I don't have strong feelings either way. There is a sense in which lots of unemployment is involuntary and a sense in which it is not. I think I understand the point of view of both sides.

Mariorossi, You said:

"It seems optimistic to assume that NGDP markets would always be perfect forecasts to be honest."

No one claims they are perfect. I'd add that it's clear you are not familiar with my NGDP futures targeting proposal, as previous private attempts to set up such markets have no bearing on my proposal. Indeed the failure of these markets makes it MORE likely that my proposal would work.

Comments for this entry have been closed
Return to top