Surely Krugman can do worse than this?
Paul Krugman continues his habit of assuming the worst about those with whom he disagrees. In a new post, entitled “The Worst Economist in the World” he discussed the following quotation from the WSJ:
When one country devalues its currency, others tend to follow suit. As a result, nobody achieves trade gains. Instead, the devaluations put upward pressure on the prices of commodities such as oil. Higher commodity prices, in turn, can cut into global economic output. In one ominous sign, the price of oil is up 8.7% since August 27.
OK, it’s not the best thing I’ve ever read. There are two possible interpretations, neither of which are entirely satisfactory. First, he might have been referring to the fact that all currencies could simultaneously depreciate against goods and services. Unfortunately, it seems extremely unlikely that a modern journalist would refer to inflation as “devaluation.” More likely, he had the following scenario in mind: The US depreciates first by using QE, then other countries do the same. This would bring dollar exchange rates back to their original equilibrium, but leave all prices higher.
Does Krugman have any idea, any clue, as to how difficult it would be to win the title of most economically illiterate journalist? With all due respect to Mr. Krugman, I don’t think the WSJ quotation even comes close. By analogy, I was always the last person picked for basketball in high school, despite being 6′ 4”. But was I the worst player in the world? After all, there are people who are blind, or 105 years old, or in comas. I expect better, er I mean worse, from Mr. Krugman next time.
I prefer Matt Yglesias’s take on the same quotation. He senses what the WSJ was trying to get at, but points out that right now higher oil prices might be a positive sign:
If you try to reason in this direction, you end up tying yourself into knots. Is a higher price of oil “ominous.” Well it depends why it’s going up. If a bunch of equipment in the North Sea breaks, then the price of oil is increasing because the quantity of oil available to the world economy has declined. That’s bad because oil is useful. But conversely, if the US economy were to start growing rapidly that would increase the demand for oil and lead to a price increase. That, however, would be a good thing. If the world’s central banks engage in coordinated monetary stimulus, that will result in some inflation (and hence higher nominal oil prices) but some inflation would be helpful at the moment. But if Israel and Iran go to war, that will also increase the price of oil and it’ll be terrible.
In general, higher supply of useful commodities is good (and leads to lower prices) and higher demand for useful commodities is a side-effect of good things (and leads to higher prices) so you can’t just look at commodity prices and draw any conclusions about what’s happening.
That’s right. Never reason from a price change. That’s the real problem with the WSJ quotation.
Tags:
24. October 2010 at 18:08
If we all devalued together then that would raise AD. Increased AD would raise the price of commodities. That would be a good sign (a global recovery).
The point Krugman seems to want to make (and fails at) is that a commodity price increase doesn’t necessarily undo an increase in AD. Unless you’re heavily invested in markets that would incur net losses from a commodity price increase why would you be so vehemently opposed to increases in commodity prices? There must be some other (more sinister) reason.
24. October 2010 at 18:57
Professor Sumner: Can you elaborate on your point that “never reason from a price change.” I think you have made this point a couple of times and I have never really caught on as to why this important. The reason I raise this question is because many finance and accounting related studies take stock prices changes to infer the reason or causes of the price change.
24. October 2010 at 20:08
This quote is certainly no worse or more blind to the process through which economic value changes actually occur than the Keynesian idea that if “workers” tried to cut their nominal wages, it wouldn’t work because they wouldn’t be cutting their real wage. Talk about goofy. Labor isn’t “L”. I can’t imagine there’s ever been an even where every single working person took an identical nominal wage cut. Similarly, the notion that a process of competitive devaluation would be uniform and instantaneous seems pretty unlikely. Too much aggregation. That Krugman would rail against too much simplification or aggregation in an analysis is, ehem, ironic.
24. October 2010 at 20:34
“When one country devalues its currency, others tend to follow suit. As a result, nobody achieves trade gains. Instead, the devaluations put upward pressure on the prices of commodities such as oil. Higher commodity prices, in turn, can cut into global economic output. In one ominous sign, the price of oil is up 8.7% since August 27.”
The the two main ideas don’t really follow from each other, but the second part– “the devaluations put upward pressure on the prices of commodities such as oil.” Is what I have been arguing on this blog for months… I just don’t think its directly a result of competitive devaluations, although they can definitely cause it.
Commodities have store of value like money (although not quite as convenient), if the expected supply of money is growing faster than the expected future supply of the commodity, then it would be natural for the commodity to increase in price.
24. October 2010 at 21:04
Jesus. When oil prices were soaring the main Saudi Oil minister would waddle out and say, “for every cent the US dollar falls, the price of oil goes up $4, stop blaming us.”
Everyone is sitting on dollars trying to buy oil, whoever shows up with a extra newly printed one, bids it up.
He wasn’t lying. YOU CAN’T STEAL COMMODITIES from the producers, you can only make their last payment worth less.
You do understand this right Scott?
24. October 2010 at 22:25
I’m a Yglesias fan and a Sumner fan and a Krugman fan and a Delong fan. From the perspective of an interested layman from another academic discipline, y’all seem to be making versions of the same point: that increased stimulus is a must, but that there is obviously some dispute as to the way to achieve it.
To an outsider, it seems as if all sides are debating not over what would actually work and what would not, but rather over, to make a pedantic grammar point, what may be done and what can be done: Prof. Sumner seems to believe that analogues of the helicopter drop like money financed fiscal policy–favored by Delong, possibly Krugman (although I cannot recall any specific posts), and by Yglesias, Prof. Sumner’s favorite progressive commentator, CAN work, but that they are irresponsible, and thus we MAY not do them, because they may get the economy working, but they might also create unintended inflationary consequences. But by the same token, he also suggests that increased monetary stimulus CAN work, but that it MAY not be done because of political and other constraints. Similarly, there seems to be an implication that fiscal stimulus could work, but that because of the nature of the political process–i.e., that the fed would counteract the stimulus, or that the congress would not approve it–we may not do it.
On the other hand, New Keynsians like DeLong and Krugman feel that quantitative easing could work, but that we may not do them because the FED will not follow through properly. By the same token, they also feel that fiscal stimulus, including presumably money-financed stimulus, can work, but that we may not do it because of political constraints.
So at the end of the day, the question seems to be not one of economics but rather one of politics. I guess my question is: is the main problem behind everything the failure to repoliticize monetary policy? Prof. Sumner has quoted William Jennings Bryan approvingly, but yet there is no pro-monetary stimulus party, largely, it seems to me, because of the sense that economics is a scientific profession, and that we MAY not make this a political issue, because that would be irresponsible.
This is not a left-right issue, as everyone seems to be convinced that politicizing monetary policy would be a bad decision–and yet the consequence of this failure seems to be the rise of extremist anti-Fed views that advocate for a return of the gold standard. In such a climate, monetary policy will be politicized, regardless of what the “adults” want to happen.
To an outsider, the blockages to stimulus seem to result not from what we can do-everyone seems to know what would work–but rather from a perceived sense of what, because of political reasons, we may do.This is problematic for people of both political persuasions, because the Democrats, despite the Bryan heritage, show no sign of interest in political action about monetary policy, while the Republicans are only voicing incredibly counterproductive strategies. I love the monetary policy aspect of these blogs. But they need both to preserve their intellectual integrity and to present a more overt political understanding of the blockages to these issues.
Stimulus, even in the monetary realm, is a political rather than a policy problem, as everyone knows what we need to do. Just nominating Lawrence Summers would not do the trick. What we need is the Billy Carter/Roger Clinton of monetary policy: someone who has the negative credentials to prove that he will do what can work, despite what responsible people think may work.
Or in other words: shouldn’t the democrats, considering the delicate relationship between MAY and CAN that holds in our politics, simply have run on a policy of a money-financed tax moratorium? It may seem like something we MAY not do, but it is certainly something that COULD work.
24. October 2010 at 23:26
Did you even read the whole post? I am no Krugman defender, but I don’t think you paid enough attention to what he was saying. Start with the first paragraph:
“A thought: it has occurred to me that we could use an economics equivalent of Keith Olbermann’s “Worst Person in the World” award. KO does not, of course, mean that the person he goes after on any given night really is the worst person in the world; he just uses the title to highlight some especially awful action or statement.”
I expect better from you next time.
So obviously he didn’t really mean the guy is the worst economist in the world, he just wanted to highlight the awful statement.
24. October 2010 at 23:59
Morgan, I believe that is what Scott said. More dollars, prices go up.
25. October 2010 at 03:24
It is supply _and_ demand.
Higher oil prices, higher inflation generally, and higher interest rates are all bad for real output and employment–ceteris paribus. Compare to what would happen if the relevant supplies were all perfectly elastic. However, greater money expenditures will raise real output as long as the relevant supplies aren’t perfectly inelastic.
Money expenditures are well below their growth path from the great moderation. They need to be much higher. It is likely that a recovery in money expenditures would result in higher oil prices, higher inflation, and higher interest rates. Still, none of these things are expecially desirable. With inflation, at least, I think it is clearly an unfortunately side effect of a desirable change.
Only from the point of view of central bankers, with their irrational attatchment to targeting short term and safe nominal interest rates, does higher (expected) inflation provide “benefits.” Their approach to policy, combined with their policy of issuing zero nominal interest currency on demand, has left them stuck. Higher expected inflation would pull them out of this trap of their own construction.
Well, I guess higher inflation makes the real yeild on zero nominal interest hand to hand currency more negative, and so provides more income to the central bankers who borrow by issuing it immediately, and their government owners indiredtly.
The economy doesn’t need more inflation now. It needs more aggregate demand–more money expenditures. More expected inflation is one way to generate it. Quantitative easing and lower longer term real rates is another. And, most importantly, as Sumner always insists, a clear committment to do it may allow it to occur with paradoxically higher real interst rates, short and long, a smaller quantity of base money, and incredibly, a smaller quantity of money more broadly defined. How is it possible? Because it is about supply _and_ demand.
25. October 2010 at 03:33
Scott. Maybe Krugman could do worse but Blinder takes 1st prize!!!
http://online.wsj.com/article/SB10001424052702303738504575568122743023374.html?mod=WSJ_Opinion_LEADTop
25. October 2010 at 03:46
Mishkin advocates an explicit inflation target in today’s FT:
The Fed Must Adopt an Inflation Target.
25. October 2010 at 04:40
JTapp JimP linked to that piece yesterday, and I commented:
“JimP has linked to it but I think there´s some “desperation” to Mishkin´s arguments. In 3 sequential paragraphs he uses the expression: “the FOMC should announce that this rate would only be modified for sound economic reasons” and in one repeats it as: “by stating its intention not to modify the rate without a clear technical rationale”. He also tries to dispel worries that the Fomc would put too much emphasis on inflation and not enough on growth/employment:”Some commentators have worried that establishing an inflation objective will soon lead to an overemphasis on controlling inflation, and not enough concern about stabilising real economic activity”.
All in all, a very confusing piece. Much better to propose the “all encompasing” NGDP stability! It would cover, simultaneously, both p and y! And forget about “sound economic reasons…”
25. October 2010 at 04:48
@ marcus nunes
Agreed, NGDP stability would be better than an inflation target. Although… politically, if the fed adopted an NGDP stability regime, it would become the politician’s whipping boy any time there was supply side caused recessions. I guess thats to be expected when you live in a centrally planned system, the central planner is used as a scape goat by the politicians when their system fails.
25. October 2010 at 05:01
Krugman did it. He topped himself:
http://krugman.blogs.nytimes.com/2010/10/25/monetary-base-and-short-term-debt-ultra-wonkish/
25. October 2010 at 05:05
@ Doc Merlin
That´s the problem with having an explicit target. That´s why Greenspan never was a fan of one. In practice (and unconsciouly) he managed to stabilize NGDP and in return got the Great Moderation! Remember the second half of the 90s when everybody (PK included) was clamoring for higher rates because unemployment was dropping fast and Greenspan spoke about higher productivity?
25. October 2010 at 05:54
Mark, I agree, which is why I liked Yglesias’ post better.
rob, It is all right to attribute stock price changes to shocks, if you have good reason to. What I meant was that one should not simply observe a price change, without other information on why it changed, and then assume that had some implication for the economy.
BTW, I sometimes violate this rule. The key is to always be aware (in the back of your mind) as to what you are assuming, and what evidence you have. Otherwise changes on commodity prices, exchange rates, and interest rates, by themselves, have no implications for the economy.
John Papola, Very good observation.
Doc Merlin, Yes, and the prices could rise for one of two reasons–expected inflation, or expected real GDP growth.
Morgan, Read my post, I said currency devaluation pushed up commodity prices–it’s Krugman you should be going after.
Shane, You make a number of good points, but:
1. Not everyone agrees monetary stimulus is needed. I think most economists oppose it.
2. My differences are not just political, not just a question of what policies are politically feasible. I believe monetary stimulus is far more powerful than fiscal stimulus–I don’t think Krugman and DeLong agree with that. Of course back in February 2009 my position was quite different from their’s. They were then vigorously arguing that the Fed was out of ammunition.
Vivi, You said;
“I expect better from you next time.
So obviously he didn’t really mean the guy is the worst economist in the world, he just wanted to highlight the awful statement.”
Yes, I know he didn’t really mean it literally. My post was intended to be a sort of joke–or at least the paragraph where I mentioned basketball. But even so, that paragraph he quoted wasn’t all that bad even considered on its own terms.
Next time I’ll try to add a PS to clarify which parts I was joking.
Lorenzo, That’s right.
Bill, You said;
“Higher oil prices, higher inflation generally, and higher interest rates are all bad for real output and employment-ceteris paribus. Compare to what would happen if the relevant supplies were all perfectly elastic. However, greater money expenditures will raise real output as long as the relevant supplies aren’t perfectly inelastic.”
I don’t agree. Prices never change for no reason at all. They change because of shifts in supply or shifts in demand. If the high oil prices or high interest rates come from more demand (oil demand and credit demand), that is bullish for the economy. If they come from less supply (OPEC cutback, tight money) that is bearish for the economy. It makes no sense to talk about a “ceteris paribus” price change. SOMETHING must change for prices to change. It all depends on what that something is.
You said;
“Money expenditures are well below their growth path from the great moderation. They need to be much higher. It is likely that a recovery in money expenditures would result in higher oil prices, higher inflation, and higher interest rates. Still, none of these things are expecially desirable. With inflation, at least, I think it is clearly an unfortunately side effect of a desirable change.”
That’s true, but suppose you are trying to boost AD through monetary policy, and you want an indicator as to how well the stimulus is working. If inflation does not rise, then it looks like you are pushing on a string. If you see inflation expectations rise, then the monetary stimulus is working. In that context more inflation is good news for growth.
I completely agree that it makes more sense to say we need more AD, than we need more inflation. But I also think that if commodity prices rise in response to QE, that’s a sign QE is starting to work.
If you want to say that it would be better for the US if the supply of oil was more elastic, I have no problem with that.
Marcus, I liked Blinder better when he was talking about negative IORs.
JTapp, That’s better than the last thing by Mishkin I read, but he needs to talk about price level targeting, not inflation targeting.
Marcus, Krugman always leaves himself an “out.” In this case he’s implicitly assuming any QE would be temporary. That’s why it’s not inflationary–the zero rate bound is actually a side issue. If the QE is permanent, it will be expansionary even at the zero bound.
25. October 2010 at 06:46
I thought Krugman’s response was confusing because of this paragraph:
“There is, I think, a tendency to think of devaluations as reductions in the value of currencies relative to something external and eternal “” and hence as making us all poorer. But the reality is that my depreciation is your appreciation, and vice versa; we can’t all devalue at the same time.”
I guess he is trying to say that all economies can’t experience a real devaluation at the same time, but it is obviously true that they can all nominally devalue if they all print currency while the amount of real goods remains constant. (Many people say Krugman is the clearest popular writer on economics, if so our profession is a failure)
25. October 2010 at 07:22
@marcus nunes I agree the Mishkin piece is “confusing.” He’s basically saying adopt a target and then break it whenever you feel there is good economic reason. How exactly will that anchor expectations? I told my wife I’d take out the trash and then didn’t do it because I had better things to do. Now she doesn’t believe me whenever I tell her I’ll do it.
I agree with Scott, why doesn’t Mishkin just go for price level targeting here? Would make a more coherent argument.
25. October 2010 at 08:13
Morgan Stanley wants price level targeting:
http://www.morganstanley.com/views/gef/index.html#anchord4b07851-ddb8-11df-a74c-2bbdd0f0bcbb
25. October 2010 at 10:58
Scott, my annoyance is with Matty.
“If the world’s central banks engage in coordinated monetary stimulus, that will result in some inflation (and hence higher nominal oil prices) but some inflation would be helpful at the moment.”
In 2008, we had oil prices go past $145 in very short order as the value of the dollar plunged – while the economy was “booming.”
NONE of it was positive for us. Food prices spiked, people got freaked out, and then FINALLY everything fell back to NORMAL.
The GOOD part about Scott Sumner is the real part of pissing on booms, this idea that we generally are cozy with inflation and we’re going to cheer it pretends we’re not close to the target right now.
2% CPI level targeting VERY soon becomes – “hey it’s time to pull money out out of the system.” We should start to refer to this as HAWK CRACK.
The issue is this… your vision is being latched onto by people who are not committing to the pissing on booms part, and when they sense that unemployment STILL might sit up close to 9% – they freak out like Stieglitz, Ezra, DeKrugman, and Matty.
And you don’t cover this subject, you don’t draw out the fight.
25. October 2010 at 11:15
Scott, just got back from one of Dudley’s presentations. I asked him why not just target NGDP and tell people you are trying to raise incomes. I’ll be putting up some blog posts today and tomorrow about the rest of his comments (very interesting and informative) but wanted to let you know the answer to this question soonest: (paraphrased)
You have the right point. We would like to see about 1.75% inflation. We are worried about NGDP growth more right now because people can’t get the leverage they need. We could try to target nominal incomes. But would it be credible? Could we hit our targets? Then there’s the problem of communicating the metric. Most people would say “nominal income?? hunh? What’s nominal? What’s income?” This is a key issue. If we could guarantee 5% NGDP growth for the next few years, that would be great.
25. October 2010 at 11:16
Scott, just got back from one of Dudley’s presentations [ Dudley = NY Fed Governor]. I asked him why not just target NGDP and tell people you are trying to raise incomes. I’ll be putting up some blog posts today and tomorrow about the rest of his comments (very interesting and informative) but wanted to let you know the answer to this question soonest: (paraphrased)
You have the right point. We would like to see about 1.75% inflation. We are worried about NGDP growth more right now because people can’t get the leverage they need. We could try to target nominal incomes. But would it be credible? Could we hit our targets? Then there’s the problem of communicating the metric. Most people would say “nominal income?? hunh? What’s nominal? What’s income?” This is a key issue. If we could guarantee 5% NGDP growth for the next few years, that would be great.
25. October 2010 at 11:48
The important notes are here
25. October 2010 at 11:48
1.75%! my heart swoons.
25. October 2010 at 14:58
Thanks for the reply Prof. Sumner. I have an honest, genuine, follow up question (if you have the time–sure you’re super busy), not a comment masquerading as one.
I read you here (http://www.themoneyillusion.com/?p=7155) saying that fiscal stimulus can and does work in some circumstances, but that it is unnecessary because the Fed should just handle it. Doesn’t this mean then that there is some level of fiscal stimulus that could work even now? Surely monetary policy can counteract decreasing AD produced by fiscal austerity (e.g., UK) but are you saying that the fed actually counteracts increasing AD produced by additional spending? Even if the fiscal multiplier is lower than the hype, isn’t is still just smart policy to front-load capital improvement projects at at time when interest rates and inflation are very low?
I agree that Krugman is more disingenuous about his on-again-off-again commitment to monetary stimulus. But isn’t his position really just a matter of semantics, namely, that monetary policy could work by raising inflation expectations, but in that case it is not monetary policy itself that is doing the trick, but rather the way the fed frames its actions in public statements? In other words, at this juncture, it would be better for Bernanke to have a degree in theater than in economics, and for him to transform his public persona from boring Ben to bling-bling Ben, showing up at press conferences in expensive cars with strippers on both arms, flinging wads of newly minted 100s at the audience and announcing endless QE. You both seem to agree that expectations are of primary importance at times like this, but disagree on whether to call targeting expectations monetary policy.
26. October 2010 at 05:41
James, You are right that we can all devalue against goods and services. Krugman would say he is talking about devaluing against other currencies.
Thanks JimP, That’s a good MS report.
Morgan, The economy is often in better shape when commodity prices are high, and in deep recessions when they are low.
D. Watson, That answer at least shows that he understands the attraction of NGDP targeting–which is a good sign. I understand that any Fed official is also going to raise some possible objections, as the Fed doesn’t plan on doing that anytime soon.
Shane, You asked;
“I agree that Krugman is more disingenuous about his on-again-off-again commitment to monetary stimulus. But isn’t his position really just a matter of semantics, namely, that monetary policy could work by raising inflation expectations, but in that case it is not monetary policy itself that is doing the trick, but rather the way the fed frames its actions in public statements?”
Good question. Actually, the nominal target (P or NGDP) is a far more important part of “monetary policy” than the setting of short term interest rates. So if that’s Krugman’s view, it is dead wrong. One can’t do monetary policy w/o a nominal anchor, or else the price level will shoot off to zero or infinity.
One of the criteria for fiscal stimulus to work is that the Fed not be doing inflation targeting. But they are now considering doing a large QE precisely to hit their inflation target. In that case, fiscal stimulus won’t work.
In theory we might want to do big capital projects when real rates are low. But when the government has to spend a lot of money quickly on capital projects, it tends to be wasted. In the US, truly valuable infrastructure takes many years to get off the drawing boards. When I was out west this summer I saw perfectly good roads being repaved–with ARRA money. There were also news articles about a very wasteful high speed internet project for Montana. These are the sorts of things that occur in the real world. So in theory you are right (if we aren’t inflation targeting)–in practice I am dubious.
26. October 2010 at 10:06
Morgan, The economy is often in better shape when commodity prices are high, and in deep recessions when they are low.
Scott, this is just noise.
OIL WAS $145 a barrel IMMEDIATELY before your oh so terrible NGDP July-Dec 2008.
Things were NOT ok… the US dollar the parachuting and prices were going up everywhere.
Scott, why don’t we just compromise… since you want to print money and I want to not hurt the savers:
Let’s print money and give it to the savers.
You want to sell the Tea Party of printing money… so argue for QE that ONLY BENEFITS THEM… come up with QE that the banks will HATE.
I’m confident this this exercise will be very helpful for you… instead of trying to change your “message” to make it politically palatable, craft your policy to the advantage of the Republicans.
27. October 2010 at 17:05
Morgan, Yes, just before the crash. And if they had stayed at $145 the crash probably would not have occurred.