Reply to Thoma on NGDP targeting
Mark Thoma recently asked the following question:
So, for those of you who are advocates of nominal GDP targeting and have studied nominal GDP targeting in depth, (a) what important results concerning nominal GDP targeting have I left out or gotten wrong? (b) Why should I prefer one rule over the other? In particular, for proponents of nominal GDP targeting, what are the main arguments for this approach? Why is targeting nominal GDP better than a Taylor rule?
The entire post is rather long, and Thoma raises issues that I don’t feel qualified to discuss, such as learnability. My intuition says that’s not a big problem, but no one should take my intuition seriously. What people should take seriously is Bennett McCallum’s intuition (in my view the best in the business), and he also thinks it’s an overrated problem. I think the main advantage of NGDP targeting over the Taylor rule is simplicity, which makes it more politically appealing. I’m not sure Congress would go along with a complicated formula for monetary policy that looks like it was dreamed up by academics (i.e. the Taylor Rule.) In practice, the two targets would be close, as Thoma suggested elsewhere in the post.
Instead I’d like to focus on a passage that Thoma links to, which was written by Bernanke and Mishkin in 1997:
Nominal GDP targeting is a reasonable alternative to inflation targeting, and one that is generally consistent with the overall strategy for monetary policy discussed in this article. However, we have three reasons for mildly preferring inflation targets to nominal GDP targets. First, information on prices is more timely and frequently received than data on nominal GDP (and could be made even more so), a practical consideration which offsets some of the theoretical appeal of the nominal GDP target. Although 20 collection of data on nominal GDP could also be improved, measurement of nominal GDP involves data on current quantities as well as current prices and thus is probably intrinsically more difficult to accomplish in a timely fashion. Second, given the various escape clauses and provisions for short-run flexibility built into the inflation-targeting approach, we doubt that there is much practical difference in the degree to which inflation targeting and nominal GDP targeting would allow accommodation of short-run stabilization objectives. Finally, and perhaps most important, it seems likely that the concept of inflation is better understood by the public than the concept of nominal GDP, which could easily be confused with real GDP. If this is so, the objectives of communication and transparency would be better served by the use of an inflation target. As a matter of revealed preference, all central banks which have thus far adopted this general framework have chosen to target inflation rather than nominal GDP.
1. I believe the federal government could estimate monthly nominal GDP numbers that are accurate enough to be useful for policy purposes. However, even if they could not I’d still favor NGDP targeting, because like Lars Svensson I believe the Fed should be targeting the forecast, that is, setting policy in such a way that the Fed’s forecast of future NGDP is equal to their policy target. I also favor level targeting (recently recommended by Woodford), and I think this would reduce the overshooting problem associated with futures targeting.
2. It seem to me their second point (which isn’t really a criticism at all) was actually disproved during the recent crisis. Between mid-2008 and mid-2009 NGDP fell over 8% below trend (or about 3% in absolute terms.) On the other hand core CPI inflation fell only slightly below trend. Because the Fed is an (implicit) inflation targeter, the slight slowdown in CPI inflation did not present an unambiguous signal (in their view, not mine) for aggressive stimulus. Hence they waited until November 2010 to undertake QE2. If they had been targeting NGDP along a 5% growth trajectory, it would have been immediately obvious that NGDP was coming in well below target, and would remain below target for many years. In my view the QE2 program would then have been adopted much sooner and in larger amounts, and I think it retrospect it is clear that additional stimulus would have been welcome in late 2008 and 2009.
3. The third point is where I most strongly disagree with Bernanke and Mishkin. In the current crisis we’ve seen just how difficult it is to communicate the need for higher inflation. The public interprets that as the Fed trying to raise their cost of living. I’m not surprised the plan is unpopular. I’d guess that in 1997 Bernanke and Mishkin were thinking about the central bank communicating the need for lower inflation, not higher inflation. In contrast, NGDP is essentially nominal income. The Fed can tell the public they are trying to raise nominal growth to 5%, because a healthy economy requires the incomes of Americans to grow by about 5% per year. That’s much less negative sounding that trying to raise the cost of living. Of course the opposite could be argued on the upside, but the Fed has shown a much greater ability to hold inflation down that increase it, as the zero rate bound has left them spinning their wheels when inflation has fallen below target. I think it would be easy to explain to the public that an excessively rapid growth in nominal incomes could be inflationary, and raise rates when needed. Especially given that they were widely criticized for not raising rates enough during the housing bubble.
4. Regarding revealed preference, NGDP targeting is more desirable the larger and more diversified the economy. If an economy is dependent on just a few industries, then a major price shock in one export industry might force a dramatic contraction in other industries under NGDP targeting. Price level targeting might provide for a better macroeconomic outcome in that case. Thus I wouldn’t expect small countries to be the first to adopt NGDP targeting. And would I be out of line in noting that the BOJ and ECB haven’t become famous for creativity and boldness?
Update: I just noticed that Bill Woolsey has a long and very informative reply to my discussion of NGDP futures, and also a response to some of the points made by Brad DeLong. Bill knows my plan better than I do, so I usually defer to his judgment. He’s correct that I cut some corners in selling the idea to the National Review’s readers, and that the actual plan is more complicated than I suggested. Indeed, I believe he was the first to use the phrase “index futures convertibility.”
Tags:
19. December 2010 at 14:47
Given so much online data, my guess is that we can track GDP more accurately than even 1997. I don’t know if an Amazon, Google, Mastercard etc would want to give up data, even if treated in a proprietary manner, but it seems to me we should be able track many, many proxies for economic output much better than in 1997. Real estate loan data is far better collated than back then.
On the other hand, I can think of incredibly complicated arguments that could be made whether we are having inflation or not. Shopping at dollar stores? Cell phone only vs, landlines? Computers? Improved health care? Chicken vs. beef?
Egads, wake me when you have settled the arguments.
The fact that Bernanke was willing to consider NGDP targeting in 1997 I take as a hopeful sign. He only said that he preferred inflation targeting.
The argument keeps shifting in Sumner’s court. Bernanke strikes me as cautious and circumspect to a fault–but not close-minded. We may yet see informal NGDP targeting soon.
Sumner, as usual, is right about selling inflation to the public–there ain’t a rookie salesman in the world that would try that gambit.
Try selling economic growth.
19. December 2010 at 17:38
Seems like a somewhat better target for the fed to fall short of.
19. December 2010 at 18:08
Is “e” channeling Triumph the insult comic dog? “That’s a great Fed target…FOR ME TO POOP ON!”
20. December 2010 at 05:02
“Literal Scott”
http://www.marginalrevolution.com/marginalrevolution/
“Scott’s views remind me of a concern of Robin Hanson’s. “Why can’t those writers just come out and say what they mean?” Robin asked me once about the classic great books. It was a plea for a more literal discourse. Yet more literality is not always possible and not always more effective”.
20. December 2010 at 08:41
Scott,
This is all a non-issue. A nominal GDP target is just a special-case Taylor rule.
Steve
20. December 2010 at 09:59
“3. The third point is where I most strongly disagree with Bernanke and Mishkin. In the current crisis we’ve seen just how difficult it is to communicate the need for higher inflation. The public interprets that as the Fed trying to raise their cost of living.”
That is exactly how I see it. More negative real earnings growth for the lower and middle class but not the rich, the bankers, and the economists. Where is the communication that everyone should have positive real earnings growth and everyone should be saving (not just the rich)?
“I’m not surprised the plan is unpopular. I’d guess that in 1997 Bernanke and Mishkin were thinking about the central bank communicating the need for lower inflation, not higher inflation. In contrast, NGDP is essentially nominal income. The Fed can tell the public they are trying to raise nominal growth to 5%, because a healthy economy requires the incomes of Americans to grow by about 5% per year.”
Now you are going to have to deal with wealth/income inequality (just because national income is growing 5% does NOT mean lower and middle class workers’ income is rising 5%) from an oversupplied labor market and whether debt is being used to make up for negative real earnings growth on the lower and middle class, especially workers. Do you really believe greenscam, bernanke, or mishkin would allow lower and middle class workers’ income to rise about 5% a year? Would they start screaming like idiots “Don’t you remember the 1970’s”? It does not seem to me you get what the fed and congress have been doing for about the last 30 years.
Just from what I have read, it seems to me that nominal GDP targeting suffers the same problems as price inflation targeting. That is assume aggregate demand is unlimited and all NEW medium of exchange should be debt.
20. December 2010 at 11:31
scott, NGDP targeting has one fatal flaw…Congress can change the Fed’s mandate. All it would take would be one oil spike accompanied by high employment for public outrage to turn towards the Fed. Congress could easily take action to limit the Fed to a stable-money policy. Public opinion is already quite negative on the Fed so this doesn’t seem like a stretch. The markets would surely know the Fed could not commit to an NGDP target.
20. December 2010 at 11:32
meant to say “accompanied by high unemployment.
20. December 2010 at 11:52
[…] of the serious ailments””a decline in NGDP expectations””with the tools available. I also like his take on the question of the public's ability to understand and support an NGDP-based (rather than an […]
20. December 2010 at 15:07
Benjamin, Yeah, I’m increasingly fond of the argument that NGDP is an easier sell than inflation. It was only when I stopped thinking about the issue like a macroeconomist and started thinking about it like a normal person that I realized how off-key their message sounded.
e, Getting half way to the right NGDP is much better than getting half way to the right inflation rate.
TGGP, Another cultural icon I know nothing about.
Marcus, Thanks, I’ll try to respond.
Steve, NGDP is much easier to sell, but my main disagreement is that Taylor himself uses a backward-looking version, I prefer a forward looking version. Also, he argues that interest rates should be the operating target, I favor NGDP futures prices as a policy instrument.
Fed up, If falling NGDP was bad for the working class, and it clearly was, why wouldn’t rising NGDP be good for the working class?
I don’t favor the medium of exchange being debt, I like cash.
mlb, I hear this argument all the time, but it makes no sense to me. In 2009 we had the biggest fall in NGDP since 1938, and almost no one in Congress is pushing the Fed, except those who want tighter money. The 1970s are over, dead, buried.
20. December 2010 at 15:19
[…] of the serious ailments””a decline in NGDP expectations””with the tools available. I also like his take on the question of the public’s ability to understand and support an NGDP-based (rather than […]
21. December 2010 at 10:31
@Scott:
“I don’t favor the medium of exchange being debt, I like cash.”
This is a nonsense statement. Cash is itself an outstanding federal reserve liability.
21. December 2010 at 10:35
“Fed up, If falling NGDP was bad for the working class, and it clearly was, why wouldn’t rising NGDP be good for the working class?”
Because distribution and budgets matter with rising NGDP IN THE PRESENT BUT CAN FALL IN THE FUTURE. If the labor market is oversupplied and wages are not rising fast enough for the lower and middle class, a lot of people are depending on debt to make up the difference in their budget. This debt has to be serviced. If the goods market then goes from supplied constrained to demand constrained, people start losing hours worked as productivity grows. They can’t get a raise and can’t work more hitting their budgets, including debt service.
“I don’t favor the medium of exchange being debt, I like cash.”
I assume you mean currency. I like currency not debt as the medium of exchange too because currency can’t be directly defaulted on (or paid off) like demand deposits from debt can. However, the way the system is set up now, isn’t all NEW medium of exchange debt? If you want to talk about changing that, please explain. I’m all ears. I see about 1 trillion in currency and about 52 trillion to 58 trillion (depending on how gov’t debt is counted) in debt.
If you want to explain to me the difference between creating more medium of exchange from debt and creating more medium of exchange from currency, please do. That is the discussion to have. I’ll start. If an economy is price inflated with debt (assuming an aggregate demand shock), an economy gets rising NGDP IN THE PRESENT. In a few years the amount of debt may fall leading to falling or the same NGDP IN THE FUTURE. If an economy is price inflated with currency (assuming an aggregate supply shock), an economy gets rising NGDP IN THE PRESENT. In a few years the amount of currency does not fall leading to increasing NGDP IN THE FUTURE.
Can you see the difference from a time perspective and budget perspective (as in debt is future demand brought to the present and currency is present demand in the present)?
21. December 2010 at 10:43
One other thing is I do NOT assume aggregate demand is unlimited. I can’t think of any ecomonic model that even considers this. If you know of one, please let me know. I believe that is the fatal flaw in most to all economic models, and it has “worked” because most to all economies have been supply constrained since The Great Depression with the possible exception of Japan (until about 2000 or 2007).
21. December 2010 at 15:13
@Scott
“Fed up, If falling NGDP was bad for the working class, and it clearly was, why wouldn’t rising NGDP be good for the working class?”
Thats fallacious, Scott. If food/energy/housing prices were rising more than NGDP, then rising NGDP could also be bad for the working class. Not that this happened this recession, but again, your logic doesn’t follow. Remember, for banks what matters most is NGDP for regular people RGDP matters more.
21. December 2010 at 18:51
Doc Merlin, I don’t agree. The Fed has made no commitment to pay it back. They can leave it in circulation forever.
Fed up, What the working class needs is JOBS. You get that with more NGDP.
I think the Fed’s big mistake was focusing on debt. I want them to issue more currency–which would have been far more effective.
Fed up, Zimbabwe proved that AD is unlimited, as it went up zillions of percent in just a year.
Doc Merlin, Food energy and housing prices fell sharply in 2009–how’d that work out for the working class?
21. December 2010 at 19:02
@Scott Sumner
“Doc Merlin, I don’t agree. The Fed has made no commitment to pay it back. They can leave it in circulation forever.”
Um, yes they have. Bernanke has said that he will pull it back after the crisis, he has said such on multiple occasions.
“Doc Merlin, Food energy and housing prices fell sharply in 2009-how’d that work out for the working class?”
Sigh, Scott. Reread what I said. You are making a freshman mistake. The converse of a supposition doesn’t necessarily have the opposite result as that supposition.
I said:
“If food/energy/housing prices were rising more than NGDP, then rising NGDP could also be bad for the working class. ”
I NEVER said:
“If food/energy/housing prices are falling more than NGDP is, it is good for the working class.”
21. December 2010 at 19:28
“Fed up, What the working class needs is JOBS. You get that with more NGDP.”
Not necessarily. Let’s say price inflation is 4% and real GDP is 1%. Now put productivity growth at 3%. Won’t jobs be lost?
If “THE JOBS” could speak, would they say there is about the right amount of us jobs but too many of you workers? Could “THE JOBS” say why don’t some more of you workers retire?
I’ll put it this way too. Let’s say there are about 6 billion people in the world, 4 billion need/want jobs, but only 3 billion are needed to produce enough for the 6 billion. What should happen?
“I think the Fed’s big mistake was focusing on debt. I want them to issue more currency-which would have been far more effective.”
IMO, the fed is always focused on debt. That is their preferred way to attempt to grow and stabilize an economy until more private debt owed to the rich doesn’t work any more than they get the gov’t to create more gov’t debt owed to the rich for them while minimizing defaults. Issuing more currency could help. It depends on how. I don’t believe swapping a treasury for currency will eliminate the person’s desire to save. The other thing is the way the system is set up now I don’t think they can issue currency at will and the fed probably wouldn’t do it even if it/they could.
“Fed up, Zimbabwe proved that AD is unlimited, as it went up zillions of percent in just a year.”
Hmmm… I think I forgot how you define it, and I’m probably not saying it correctly. I’ll try it this way. Why does Bill Gates not spend (not invest) all he has? Why doesn’t he invest more in Microsoft to produce more copies of Windows?
Maybe I need a phrase like real goods/services spending?
21. December 2010 at 19:32
I think Doc Merlin’s point is that real wages (and probably interest rates too) should be positive for all categories of a person’s budget.
Sorry Doc Merlin if I put “words in your mouth”.
22. December 2010 at 10:48
@Fed Up.
Thats not quite what I meant. Rather that real income growth should be at any given time positive if you want people to do well. This is a separate issue from nominal income, which should meet very long range nominal expectations (because things like 30 year loans are sources of very long term stickiness) rather than necessarily being up or down.
22. December 2010 at 10:50
@Fed Up:
“I’ll put it this way too. Let’s say there are about 6 billion people in the world, 4 billion need/want jobs, but only 3 billion are needed to produce enough for the 6 billion. What should happen?”
I don’t agree that this is possible. I believe that individual demand is unlimited, and thus that there is no “produce enough for the 6 billion.”
22. December 2010 at 19:33
Doc Merlin, No, he said he buy back the ERs, not the currency. Currency can stay in circulation forever.
Doc Merlin, On the food prices I’m not saying it’s logically impossible for higher NGDP to hurt the working class, I’m saying there is no evidence. I provided a counterexample, that is all.
Your comment wasn’t out of the blue, it was a response to something I said to Fed up, it had a specific context. I’m sticking with that context. If you have another context, feel free to bring it up.
Fed up: You said;
“but only 3 billion are needed to produce enough for the 6 billion.”
Enough? I’m an economist, we don’t even know what sentences like that mean? What is enough?
You said;
“I don’t believe swapping a treasury for currency will eliminate the person’s desire to save.”
Not once in 700 posts have I ever said I’d like to see Americans save less.
You said;
“Maybe I need a phrase like real goods/services spending?”
I have to respond to the questions people ask, I’m not a mind reader. I still have no idea what you mean by “enough”.
23. December 2010 at 11:12
Doc Merlin said: “Thats not quite what I meant. Rather that real income growth should be at any given time positive if you want people to do well.”
I believe productivity gains need to be divided up in the right amount between workers, upper management, and shareholders.
And, “I don’t agree that this is possible. I believe that individual demand is unlimited, and thus that there is no “produce enough for the 6 billion.””
Then why do people voluntarily retire? Isn’t that the main reason people save (not spend everything) to retire someday? I’ll bring up Bill Gates (or someone else making that much) again. Does he spend everything he earns? It seems to me that his real demand isn’t unlimited.
23. December 2010 at 11:34
“Enough? I’m an economist, we don’t even know what sentences like that mean? What is enough?”
I can buy food, clothing, shelter (paid off, no mortgage), health care, tv, internet, vehicle or two, furniture, gas, electricity, and other. In the USA, I’ll estimate that at $30,000 to about $70,000 a year depending on location. If I can save enough to have the assets pay me enough, I can retire. At a certain point, the more I make the more I save the sooner I can retire. Not the more I make the more I spend.
It seems to me most economists assume real aggregate demand (??? right term ???) is unlimited. So, the faster real aggregate supply can grow without price inflation (whether from the future with debt or not) the better. What if that assumption is wrong? How do the models change then?
You can check out bernanke’s deflation speech to see what I mean.
23. December 2010 at 11:46
“”I don’t believe swapping a treasury for currency will eliminate the person’s desire to save.”
Not once in 700 posts have I ever said I’d like to see Americans save less.”
Sorry, maybe I wasn’t clear. I was referring to QE2. I don’t think it matters whether the swap is with bank reserves or currency. I’m also of the opinion that trying to affect other asset prices will not help the real economy.
23. December 2010 at 21:11
@Fed Up
“I believe productivity gains need to be divided up in the right amount between workers, upper management, and shareholders.”
Everyone believes that. The question is how you divide it up. Do you have someone arbitrarily set that, or do you let people decide for themselves based on their own choices?
The first way leads to corruption, stagnation and decay, the second leads to growth and prosperity.
23. December 2010 at 21:12
@Scott:
“Doc Merlin, On the food prices I’m not saying it’s logically impossible for higher NGDP to hurt the working class, I’m saying there is no evidence.”
The seventies serve as a evidence. I don’t think this is the case right now though.
24. December 2010 at 17:20
Fed Up, The issue of how much people work is one thing, and the desired consumption of all other goods is another. For any given level of leisure, people would prefer more goods. Most people would rather live in a palace in Beverly Hils than a trailer park–don’t you think? And most unemployed people would rather be working.
Doc merlin, Monetary stimulus was clearly excessive in the 1970s, but right now it’s clearly insufficient.
27. December 2010 at 22:28
“Fed Up, The issue of how much people work is one thing, and the desired consumption of all other goods is another.”
I believe wages ($ per hour) needs to be worked in there along with wealth/income inequality.
“For any given level of leisure, people would prefer more goods.”
It seems to me there are retirees who don’t need more goods.
“Most people would rather live in a palace in Beverly Hils than a trailer park-don’t you think?”
I’d rather live in neither, but somewhere in between.
“And most unemployed people would rather be working.”
Should there be some to a lot more people who are 55 and over who have worked long enough they should be retired (a “different” kind of unemployment)?
27. December 2010 at 22:37
Doc Merlin, I don’t believe everyone believes that. It seems to me that upper management and shareholders want it all for themselves. caterpillar is the first thing I can think of.
I’ll let people decide for themselves if I can see their assumptions when they want to go into debt and not tricked into going into debt. If their assumptions about wages and prices involve the post WWII experience (especially the 1970’s), something may need to change.
27. December 2010 at 22:39
“Doc merlin, Monetary stimulus was clearly excessive in the 1970s, but right now it’s clearly insufficient.”
I’d like to see an explanation of that in terms of wages/budgets now and then.
28. December 2010 at 09:24
Fed up, You responded to my quote:
“”For any given level of leisure, people would prefer more goods.”
It seems to me there are retirees who don’t need more goods.”
Sure, but the words ‘need’ and ‘want’ have completely different meanings. I have all I need, but I want much more.
Being 55, I’m all for letting people retire at 55. But most people don’t want to retire that young.
28. December 2010 at 10:35
Make it “It seems to me there are retirees who don’t need or want more goods.”
“I have all I need, but I want much more.”
Exactly what more would that be? How about for bill gates, warren buffett, or a highly profitable corporation?
“Being 55, I’m all for letting people retire at 55. But most people don’t want to retire that young.”
I seriously doubt that, but it could depend on the job.
28. December 2010 at 22:07
About QE2 and financial asset speculation, this is from Mish’s blog and ZeroHedge.
http://globaleconomicanalysis.blogspot.com/2010/12/margin-debt-soars-to-highest-levels.html
http://www.zerohedge.com/article/nyse-october-margin-debt-jumps-highest-lehman-failure-investor-net-worth-lowest-april-highs
“Margin Debt Data is from NYSE Factbook Securities Credit
ZeroHedge discussed margin debt in “NYSE October Margin Debt Jumps To Highest Since Lehman Failure As Investor Net Worth Is At Lowest Since April Highs”
It is not just the stock market that is at the highest levels since Lehman. Probably just as importantly, NYSE margin debt has surged to $269 billion, an increase of $13 billion from the prior month, and the highest since September 2008 when it was at $299 billion.
We are confident that NYSE cash in November will be at the lowest level of the year, not to mention December, as hedge funds leveraged everything they could, in some cases hitting as much as 3-4x gross leverage, in pursuit of beta, now that unleveraged alpha strategies have ceased to work. Which means that with retail stubbornly missing from the picture, the only beneficiaries of the HFT and Fed facilitated melt up are the 1000 or so hedge funds, where average net worth is in the 6 digits, that will be profitable this year.
Moreover, mutual fund cash levels have been near record lows since September, and topping it off, a respected friend tells me NYSE cash levels are negative $35 billion.
Collectively, this sounds like “all in” to me, and then some. However, just as in 2007, no one knows for sure when excessive optimism gets punished, historically it always is.”
Quotes done.
It seems to me they are at or extremely close to “all in” (especially with the tax cuts) that the U.S. economy has had an aggregate demand shock and not an aggregate supply shock.
30. December 2010 at 08:54
Fed up, The owners of profitable companies want them to be even more profitable. There are very few people who are so rich that more wealth wouldn’t add to their consumption.
I don’t understand your last post. Stocks are always 100% owned by someone. There is no such thing as putting cash “into” the stock market. Money goes through markets, as one person sells and another buys.
12. February 2016 at 08:05
[…] Scott Sumner, an economist at Bentley University, came to prominence during the recession and recovery for observing that the Fed was doing far too little to support the economy, and that it had helped precipitate the deep recession by doing too little in 2008, during which time nominal GDP was falling sharply. While markets were flashing warnings in the summer of 2008 that NGDP expectations were sinking fast, the Fed waited until November 2010 to undertake QE2 because inflation only slowly reversed course and began falling. Between mid-2008 and mid-2009, NGDP fell at its fastest rate since 1938 at over 8% below trend (or about 3% in absolute terms) signaling an unambiguous need for further monetary expansion. If the Fed, for example, had been targeting NGDP along a 5% growth trajectory, it would have been immediately obvious that NGDP was coming in well below target, and would remain below target for many years. With such a view the QE2 program would then have been adopted much sooner and in larger amounts. […]