Noah Smith thinks David Romer is batty
At least I think he does. Here’s what Smith says:
I find Scott Sumner making this batty claim:
“Keynesian economists have never been able to accept my assertion that the fiscal multiplier is roughly zero because the Fed steers the (nominal) economy.”
Translated roughly from Monetarese into English, this means: “If Congress tries to boost output by spending money, the Fed will counteract this effort by enacting tighter monetary policy. Thus, stimulus can never work.”
Here’s David Romer:
As Robert Solow stresses in his remarks in this session, we should not be trying to find “the” multiplier: the effects of fiscal policy are highly regime dependent. One critical issue is the monetary regime. Consider estimating the effects of fiscal policy over the period from, say, 1985 to 2005. Central banks were actively trying to offset other forces affecting the economy, and they had the tools to do so. Thus if they were successful, one would expect the estimated effects of fiscal policy to be close to zero.
Now I suppose one could argue they weren’t always successful. But that doesn’t really help. It just means that sometimes the fiscal multiplier will be 0.5, and sometimes it will be negative 0.5. The expected fiscal multiplier will be zero. Smith doesn’t seem to realize that he’s not just arguing against my “batty” views, but also against mainstream new Keynesian economics. That’s why the new Keynesians moved from fiscal stabilization to inflation targeting in the late 1980s; they came to realize the futility of using fiscal policy when monetary policy is already steering the economy.
Noah Smith is a very smart guy, but he has a bad habit of jumping into disputes without first discovering what the other side is actually saying. Indeed he recently admitted this after DeLong pointed out that his criticism of Robert Lucas was inaccurate.
In his long post he spends a lot of time setting up foolish arguments that he imagines I might make, and then shoots them down. Unfortunately for him, I don’t actually make those foolish arguments.
His main criticism is that I can’t explain why the Fed would have allowed NGDP to fall so sharply after mid-2008, and remain low, if they were in fact targeting NGDP. In past posts I’ve made two arguments. First, that the decline in late 2008 was partly due to the Fed’s excessive fear of supply-side inflation, and partly due to the fact that policy was excessively backward-looking. But that’s not the main problem. Rather, the Fed seems more reluctant to use unconventional monetary stimulus than conventional stimulus. That’s why NGDP remains low. I don’t think that’s at all controversial, indeed Paul Krugman makes the same argument.
But Krugman goes on to argue that because the Fed is reluctant to use unconventional stimulus, fiscal authorities can fill in the gap. I’ve always conceded that this argument might be true, but on balance I think it’s more likely to be false, or at least mostly false. It seems to me that the Fed’s reluctance to use conventional stimulus has pushed them to use unconventional stimulus instead. But since they aren’t as comfortable with unconventional stimulus, they only pull it out when conditions seem especially bad. Again, this is not particularly controversial. We all saw the Fed pull out QE1 when the economy seemed very bleak in March 2009. We saw them pull out QE2 when core inflation fell to 0.6% in the summer of 2010. Each time the inflation rate rose modestly after the stimulus. They are still steering the economy, but along a lower track than otherwise. I argue that if the Congress had not done the 2009 stimulus, Bernanke would have been much more aggressive with unconventional stimulus, in order to avoid a repeat of the Great Depression under his watch. I can’t prove that; but it seems plausible, especially given his earlier critique of the BOJ, which allowed persistent deflation to develop. I may be wrong, but where are the Keynesian counterfactuals for Bernanke’s response if Congress doesn’t do any stimulus?
This isn’t to say that one can’t make plausible assumptions for why fiscal stimulus might work right now, but in my view the Keynesians have not done so. Indeed Andy Harless is the only blogger than I have seen effectively refute my argument. The others just wave it away. Krugman keeps insisting the multiplier is positive because monetary policy is ineffective, even as he keeps demanding the Fed do something to boost spending! Interestingly, Harless finds Noah Smith’s critique to be so weak that he actually starts defending me in the comment section.
As far as the basic problem with Smith’s post, I think anon/portly nailed it in my comment section:
The Andy Harless comments are terrific, but if you read Noah Smith’s post carefully, what comes across is a deeply ingrained incuriosity to Sumner’s ideas. Noah Smith is saying, in essence, “I don’t have to be very familiar with Sumner’s views to know that slagging them is a good idea.”
Who would know better about these “Ketman” type issues than a smart grad student? I assume that Smith’s incuriousness is representative of the wider econ profession, and that market monetarism” has really gotten very little traction with the Macro world in general.
PS. The worst day of 2011 was October 27th, the day Andy Harless and Matt Rognlie stopped blogging. They tend to lean a bit more in the Keynesian direction, but actually do try to find out what the other side is saying.
PPS. I don’t mean to suggest that Romer agrees with this post. He goes on to argue that fiscal stimulus can be effective at the zero bound. But Smith is making a much more general argument.
Tags:
1. January 2012 at 11:09
Excellent blog.
Once again, the criticisms of Market Monetarism appear to be feeble, peevish or (pointlessly) partisan.
I have yet to read a thoughtful, compelling critique of Market Monetarism, or even a palatable alternative.
The worst scenario critics can come up is that the Market Monetarism-directed NGDP increase would be only inflation, with no real growth (despite vast unused domestic resources hungry for business, and a globalized supply side hungry for dollars).
Even in this unlikely worst-case “inflation only” scenario, inflation would help underwater property owners, banks who lent on property, and US taxpayers through deleveraging.
I think we are now facing criticism from those so deeply encrusted into reflexive positions (tight money! inflation is bad! we need a strong dollar! the federal government should not be activist! gold is precious!) that contemplative reflection is not possible for them.
Of course, we cannot print money to prosperity, we all know that. But when you have a huge engine of economic output (the USA) starved for money and growth,,,,you gotta step on the gas.
Sure, gasoline only won’t do it. But we have the huge motor, and not enough gas….
We must present Market Monetarism as the optimistic and prudent, conservative option going forward.
Which, in fact, it is.
1. January 2012 at 11:19
I’ve always thought it’s misleading to describe the multiplier in terms of the counter-factual situation where there is no stimulus and the fed may or may not loosen policy instead. I think a better term is usefulness; i.e. you should be saying that if the Fed can stabilize NGDP just as well as fiscal stimulus can, and does so when fiscal authorities refuse, then fiscal stimulus is effectively useless. That’s not quite the same thing, in my mind, as saying the multiplier is zero, because this is really a strictly Keynesian concept regarding how many times an initial increase in government spending is re-spent by private agents, which is independent of any counter-factuals.
1. January 2012 at 11:58
Indeed Andy Harless is the only blogger than I have seen effectively refute my argument.
Strange word choice. Is there a link? And it must not have been that effective a refutation if you’re still sticking to the argument.
1. January 2012 at 12:37
Scott and William,
“Indeed Andy Harless is the only blogger than I have seen effectively refute my argument”
I assume you mean that he is the only blogger to effectively REBUT your argument. To rebut is to offer an opposing response. To refute is to prove that a proposition is wrong.
1. January 2012 at 14:00
Response: http://noahpinionblog.blogspot.com/2012/01/fiscal-helplessness-cont.html
I apologize if I ascribed to you a straw-man point of view!
1. January 2012 at 14:17
I didn’t so much stop blogging in 2011 as never get off to much of a start. (One problem with trying to be fair is that it makes blogging more difficult.) Anyhow, thank you for the compliment. (There ought to be a pun on the word “complement” here, but I can’t think of one right now.)
William and W. Peden, the responses to Scott’s zero multiplier argument are in the comment sections of various of his posts, such as this recent one. I don’t think I’ve actually written about that in my blog. (I agree he probably meant “rebut” rather than “refute,” although it’s a matter of degree: Scott wouldn’t claim that the multiplier is exactly zero when we are at the zero bound, and I may have convinced him that it’s a bit further from zero than he thought, as he has convinced me — with some help from Noah Smith bringing out the devil’s advocate in me — that it’s closer to zero than I thought.)
1. January 2012 at 15:10
Ben, I agree.
Brito, It’s impossible to be free of any counterfactuals. You must make SOME assumption about monetary policy.
William and W. Peden, I meant to offer an argument that I found plausible. I’ve always argued that there is a lot of uncertainty surrounding the multiplier concept. I think his assumptions are quite plausible, but I still think my assumptions are a reasonable first approximation of reality. But he’s convinced me that a reasonable set of assumptions can produce a small but positive multiplier–say a few tenths of a percent. I hope that’s not too vague.
Noah, Don’t worry about it. When I show outrage it’s usually feigned.
Andy, Yes, that’s about what I meant–see also my response above.
1. January 2012 at 15:25
And when I launch into bombastic antagonistic rants I’m usually tweaking on caffeine and stressed out about the job market… 😉
1. January 2012 at 15:36
“Brito, It’s impossible to be free of any counterfactuals. You must make SOME assumption about monetary policy.”
Okay, would you agree, as a starting point, that the only assumption that monetary policy can affect when counting the amount of times an increase in government spending is re-spent by individuals in aggregate is the return to saving (via interest rates, asset prices, deleveraging etc…) right? I mean, the size of the multiplier is dependent on the marginal propensity to consume, according to Keynesian macro, which in turn is presumably affected by the return to saving (it has been a while since I did old Keynesian macro).
1. January 2012 at 20:07
Scott Sumner,
That’s what I thought you meant. I’ve always thought that the proper monetarist position on fiscal stimulus is not that it is necessarily ineffective, but that it is always superfluous if monetary policy is correct and that under almost any situations fiscal policy introduces dangers e.g. patronage spending and timing problems.
1. January 2012 at 20:17
There are 2 cases where fiscal stimulus may be helpful. One of which is “finger in the dike” spending, which prevents some disastrous shock. There are certain economic situations where the consequences extremely and monetary policy is too slow or too inefficient.
The money to the states to reduce the number of lay offs was such a case. It gave monetary policy the time it needed to take effect. Most states are obliged by their constitutions to have a balanced budget (don’t laugh), which would have obliged them to lay off workers rather quickly. This would have disrupted government services and worsened public morale (which is one of the hurdles monetary policy can and usually does jump).
The second is capital improvements. Unfortunately the US doesn’t have a capital budget, which is ridiculous and has bad political consequences. There are hundreds of bridges that are long overdue for repair or replacement. This is a slow moving but very real crisis, since we could reach such a backlog that it would be impossible to repair some bridges before they failed. For the cost of moving investment forward the government gains the value of the use of the improvements for the extra time plus the negative of the savings from preventing failures. This is no difference from the probability of death, which we know how to insure, so the failure number is not conjectural, but merely a job for some actuaries.
2. January 2012 at 03:49
“Incuriousness” and confirmation bias go hand in hand. It boils down to conceit versus humility. If you already “know” the answer then you have no real incentive to genuinely consider other possible answers. This explains Noah’s inability to consider the possibility of, ceteris paribus, allowing taxpayers to directly allocate their taxes…The Ostrich Response to Pragmatarianism.
To steal Obama’s favorite analogy…it doesn’t matter who’s driving…the car will always end up in the ditch if the driver is driving under the influence of conceit. The trick is for libertarians to understand that Ron Paul also suffers from the same fatal conceit malaise that Obama suffers from…given that Paul thinks he “knows” for certain what the proper scope of government should be.
If we allow taxpayers to determine the proper division of labor between the private and public sectors then trying to figure out how to get the car out of the ditch will probably be a moot point. The problem is that we’re all just blind men arguing over the scope of government…so, unfortunately, Noah’s not a unique snowflake in this regard.
2. January 2012 at 16:03
“sometimes the fiscal multiplier will be 0.5, and sometimes it will be negative 0.5. The expected fiscal multiplier will be zero.”
Even assuming that were true Scott, consider Romer’s claim: “Consider estimating the effects of fiscal policy over the period from, say, 1985 to 2005. Central banks were actively trying to offset other forces affecting the economy, and they had the tools to do so. Thus if they were successful, one would expect the estimated effects of fiscal policy to be close to zero.”
This might also suggest that the active policy of central banks during this period of the Great Moderation was wrongheaded as I believe it was.
While I know you are one of those who look back on the 20s as an economic Golden Age, even going so far as to argue that if Benjamin Strong had not died there would not have been a depression, I remember the Great Moderation having lived it and know first hand, and it was not so great other than if you were a bond holder.
Fact is that for most Americans the previous period of 1948-77 was the Golden Age. That period just happens to coincide with the age of the “Keynesian talk” you hate so much.
2. January 2012 at 18:47
Noah, I do it when I’m stressed out by blogging, or when my alma mater loses the Rose Bowl.
Brito, I don’t understand your comment. Can you translate it into monetarism? If monetary policy is targeting inflation, then the multiplier is zero, regardless of the MPC.
W. Peden, That’s close, although I would have phrased it differently.
Peter N, I think monetary policy is far quicker than fiscal policy, which must work it’s way through Congress.
Xerographica, I agree about Ron Paul.
Mike, I lived through the 1970s, and it was far worse than the 1990s, not even close. I presume you are too young to remember. I’m not sure in what way those earlier decades were better. If it’s productivity growth, then yes, but that was true almost everywhere in the world except China and India, and had nothing to do with monetary and fiscal policy. Inflation was much more unstable and we had plenty of recessions. And it’s absurd to argue Keynesian talk stopped in 1977. The policy after 1977 was anathema to the monetarists. Carter was no monetarist. The Keynesian head of the Fed was fired in 1979 because Keynesian economics had driven the economy into a ditch with double digit inflation.
I don’t recall saying there would have been no Great Depression if Strong had lived. Can you find the quotation, or did you just make it up?
2. January 2012 at 19:23
Volcker was the start of the phase where it was decideed that that we should prefer low inlfation to low unemployment.
You contrast the 90s to the 70s choosing the outlier years during the periods I mentioned as of course viewd in isolation it is most favorable to you. What is interesting abotu the 90s though is that Greenspan was criticized for not raising rates more aggressive when unemployment was back down at 6%-then it was believed that this was the “Nairu rate.”
The question that begs is why even if that was-it turned out that it wasn’t the Nairu rate assuming such a thing exists-the case we should find high unemploiyment preferrable to high uninflation. The 70s were not a Golden Age however the were preferrable to the 30s and-for most people-there were preferrable to today.
I didn’t say that “Keynesian talk”-admittedly a rather vague phrase but you had used it- stopped in 77 but it was no longer the dominant paradigm of policy makers either the fiscal authorities or the Fed.
Carter for his part was the start of budget austerity where what matters more than anything is a balanced budget-during Democratic Administartions of course. During Republican Administartions as Cheney put it, “Reagan proved deficits don’t matter.”
The way in which those earlier decades were better is income distribution as well as lower unemployment-the average of the rough period I used-1948-77-the average unemployment rate weas certainly lower than what we have had since.
The fact is that median income has been stagnant since the late 70s. We did not have unstable inflation until the 70s. Prior to then inlfation- after the end of WWII inflation- compares quite favorably to the Great Moderation.
As to the Strong quote I’m not making it up. Perhaps you meant it tongue in cheek. I’d have to look through the archives to find it-you said that in respoonse to a prior comment I made to you in an earlier post. I had made a negative comment about the 20s.
3. January 2012 at 06:32
“Brito, I don’t understand your comment. Can you translate it into monetarism? If monetary policy is targeting inflation, then the multiplier is zero, regardless of the MPC.”
Look at it like this, the fiscal multiplier remains constant, let’s say it increases nominal spending by 2, but an offset tightening by monetary policy decreases nominal spending by a half, so the overall effect on nominal income is zero. However, I wouldn’t say this means the fiscal multiplier is zero, rather, the combined effect of monetary tightening and fiscal expansion is zero, because the two forces offset each other. However, if the multiplier itself was zero, then the combined effect of fiscal expansion and monetary contraction would be less than zero, it would in fact be negative, because the only ‘force’ affecting spending is monetary, which is contracting, where as fiscal expansion has no force.
3. January 2012 at 13:14
Mike Sax, You said;
“Volcker was the start of the phase where it was decideed that that we should prefer low inlfation to low unemployment”
Since when is there a long run trade-off between the two?
I have many posts pointing out that median income data is highly unreliable, living standards have obviously risen sharply since the 1970s–I was alive then and can certainly recall that period.
I’m claiming you simply made up the Strong quotation. I’ll retract that if I made it, but I certainly don’t hold that belief.
If Carter’s austerity is anti-Keynesian, than I suppose Reagan’s big deficits are super-Keynesianism. Is that your point, or am I putting words in your mouth now?
Brito, I don’t agree with your characterization of the problem, as there is no “baseline” or neutral monetary policy on which to judge fiscal stimulus against. Suppose fiscal stimulus would have boosted GDP if investment stayed still, but investment actually fell due to crowding out. Would you say the fiscal multiplier is actually positive, but offset by business tightening?
3. January 2012 at 14:07
“Brito, I don’t agree with your characterization of the problem, as there is no “baseline” or neutral monetary policy on which to judge fiscal stimulus against.”
I still think we’re talking past each other here, there is something one of us is simply not getting (I’m not sure who). I don’t see how I’m suggesting there is a neutral monetary policy, I’m in fact saying that the monetary policy stance is irrelevant.
As an analogy, imagine we’re observing the effect of forces on a stationary hot air balloon, we have things like the upwards thrust of the hot air, the force of gravity, and other forces such as air resistance and weather effects. You seem to be concluding that since the hot air balloon is stationary (neither rising nor falling) the effect of gravity on the balloon is roughly zero. But we know for a fact that gravity has a significance downwards pressure on the balloon. I’m saying that if we removed gravity, the balloon would shoot upwards from the rising hot air being the only force now, thus this shows that gravity has an effect. You’re saying that this relies on the faulty premise that the operator of the hot air balloon will continue heating the air under the balloon in this scenario, whereas the reality is that the pilot will in fact stop heating the air such that there is no longer any upwards thrust, causing the balloon to remain stationary. This is fine, and I’d agree. But you’re then going on to say that because of this, gravity doesn’t have an effect on the balloon, do you see the problem here? Of course you’re not quite saying this, but this is the connotation of your statements since most people don’t share your definition of the multiplier.
If we use different terminology, do you agree that fiscal stimulus can be said to have ‘upwards pressure’ on nominal GDP? This doesn’t imply that NGDP will rise, because the central bank may still tighten in response, so it is consistent with what you’re saying.
“Suppose fiscal stimulus would have boosted GDP if investment stayed still, but investment actually fell due to crowding out. Would you say the fiscal multiplier is actually positive, but offset by business tightening?”
Well I thought under your framework the crowding out effect is caused by monetary policy? As in, if short term interest rates rise because of increased spending, that is because the central bank allowed it to, and thus committed passive monetary tightening. So we’re back to square 1.
3. January 2012 at 15:34
@Brito: Nitpicking, but, in your physical scenario, the net effect of gravity on that balloon is zero. The “upwards thrust of the hot air” (aka, buoyancy) is actually gravity pulling the air above the balloon down. Wikipedia explains well: http://en.wikipedia.org/wiki/Buoyancy
Ok, technically, the Earth is spinning, so it not *quite* zero, but…
3. January 2012 at 16:07
@Anthony DeRobertis, err good point. I guess you can replace the analogy with a man walking against the wind or something.
4. January 2012 at 08:52
Brito, You didn’t really answer my question about business investment and crowding out.
You said;
“If we use different terminology, do you agree that fiscal stimulus can be said to have ‘upwards pressure’ on nominal GDP? This doesn’t imply that NGDP will rise, because the central bank may still tighten in response, so it is consistent with what you’re saying.”
I don’t know what you mean by “tighten in response.” What does it mean for a central bank to “tighten” monetary policy? Some people say raise interest rates. Others say reduce the money supply. Milton Friedman says low interest rates mean tight money. Without knowing what you mean by tight money, I can’t respond.
Maybe I’ll do a post on this.
4. January 2012 at 11:41
Scott, I would say that if the crowding out is caused entirely endogenously, then the multiplier is zero. If the crowding out is caused exogenously by monetary policy, then the multiplier might be positive, but monetary policy is offsetting the upwards force on NGDP by raising interest rates, so the overall effect of two different policies is zero. But I don’t find the idea of crowding out particularly persuasive when the economy is in a recession.
“I don’t know what you mean by “tighten in response.” What does it mean for a central bank to “tighten” monetary policy? Some people say raise interest rates. Others say reduce the money supply. Milton Friedman says low interest rates mean tight money. Without knowing what you mean by tight money, I can’t respond.”
I mean exactly what you mean when you say tight money, I mean you’ve already developed a more sophisticated framework as to how the central bank can control the path of nominal spending that is more complicated than simply changing the monetary base or interest rates. I don’t see how it matters how. Maybe I should rephrase:
The central bank may still adopt a policy in response such that, were there no fiscal expansion, this policy would cause nominal GDP to decrease.
“Maybe I’ll do a post on this.”
Please do.
4. January 2012 at 19:03
Fact is that for most Americans the previous period of 1948-77 was the Golden Age.
As a person who lived through that period I can say you are either old like me but beating me into senility, or young and not yet outgrown belief in fantasy. Either way, dream on.
(This brings to mind Krugman’s lament over the passing of the political era of good feeling…
“Is this the same country that we had in 1970? I think we have a much more polarized political system, a much more polarized social climate …we’re probably not the country of Richard Nixon …”
… ah, no more race riots burning down inner cities, Weather Underground and SDS as domestic home-grown terrorists blowing up buildings, National Guardsmen shooting students dead at Kent State, Spiro Agnew soothing troubled political waters as Richard Nixon’s invaluable assistant … those were the days!)
9. January 2012 at 10:53
Brito, But if monetary policy is exogenous, then what it is maximizing?
9. January 2012 at 14:26
“Brito, But if monetary policy is exogenous, then what it is maximizing?”
I don’t understand what you mean with this question.
2. February 2017 at 08:45
[…] http://www.themoneyillusion.com/?p=12461 […]