Nominal Nonsense
It’s always a good day when Paul Krugman throws a nice easy pitch over the fat part of the plate. In this post commenting on Beckworth he combines all of the worst features of his blog, in one nice package with a bow on top.
Here’s my problem. Underlying the focus on nominal demand or GDP is some notion that there’s a quantity equation:
MV = PY
where M is the money supply, V the velocity of money, P the price level, Y real GDP. And of course this always holds true, by definition. But the temptation is to take it as a causal relationship “” to say that real GDP fell because nominal GDP fell, and that this in turn was caused by either a fall in M or a fall in V; and furthermore that any such decline is a failure of monetary policy, because the central bank should have either prevented the fall in M or increased M enough to offset the fall in V.
The second sentence has to be one of the weirdest things I have ever read by a famous economist. I have no idea the point he is trying to make. It is essentially saying that underlying the statement that A*B is important is the implication that A*B = M*(A*B/M). Okay . . .
The problem in this post is that Krugman confuses the view that falling nominal GDP is a problem, with various monetarist dogmas. He is essentially saying that if you mention that falling NGDP is the problem, you might be led to think of the MV=PY equation, and that if you even think about the (completely true) MV=PY equation, then you might start thinking about some sort of monetarist dogma that he thinks is false, and that he thinks he proved is false in a 1998 paper, even though the 1998 paper actually shows that the monetarists are completely correct in asserting that a once-and-for-all increase in M will cause PY to increase. Got that?
In fact, almost everything Krugman has written on the crisis has implicitly assumed that falling NGDP is precisely the cause of our recession, and that stimulus aimed at boosting NGDP is the solution. How does he suddenly reach the conclusion that falling NGDP is not the problem? First he points out that in his 1998 model of liquidity traps the price level is fixed in the short run. Of course in the real world the price level is not even fixed in the short run, and responds to changes in AD. But let’s put aside that little objection, and see what the fixed price assumption implies. If the price level is fixed then movements in RGDP and NGDP are identical. Thus any policy that effectively prevents a fall in NGDP will also prevent a fall in RGDP. Of course Krugman has been arguing that fiscal stimulus can do this more effectively than monetary stimulus, but that’s a completely unrelated issue. Even if fiscal stimulus is the only solution, it would not in any way change the fact that falling NGDP is the cause of our crisis. A turnaround in NGDP is both a necessary and a sufficient condition for economic recovery in Krugman’s model, and falling NGDP is both a necessary and sufficient condition for any demand-side recession in his fixed-price Keynesian model.
Even worse, when opponents of the NGDP view have attempted to come up with real explanations of the recession, he has ridiculed those views. These alternative theories include Kling’s “recalculation hypothesis” which suggests that a recovery in RGDP requires time-consuming and costly restructuring, downsizing, retraining, etc. If Kling is right then the current recession is not just a shortfall in nominal spending (as Krugman implicitly assumes in his analysis) but rather represents a real problem. Kling’s argument and some related Austrian and RBC explanations suggest that higher NGDP might lead to higher inflation rather than more real output. So if Krugman is contemptuous of all those right-wing “real” explanations of why fiscal stimulus won’t work, how in the world can he claim NGDP is not the problem? Of course it’s the problem.
Sometimes it seems like waving MV=PY in front of a Keynesian is like waving a red flag in front of a bull. It so enrages them that they lose sight of economic logic. Keynes’s entire General Theory is basically a theory of PY, and hence necessarily a theory of MV. (In fact V doesn’t have any independent meaning; it’s just PY/M.) Sure Keynes often assumes fixed prices in the GT, but not always. And what happens when he relaxes the fixed price assumption, which variable does he see AD affecting, RGDP or NGDP? The answer is obvious. Keynes relaxes the assumption at full employment, or what he calls “bottlenecks” in the economy. If the economy has reached capacity then any further increases in AD continue to cause increases in NGDP, but RGDP no longer rises. So the General Theory is first and foremost a theory of nominal income determination. The impact of AD on RGDP is entirely contingent on the slope of the SRAS curve. For the millionth time, the equation MV=PY has zero monetarist implications, and that’s doubly true for the concept of nominal GDP.
Now let’s look at the more important part of Krugman’s post, his argument that money can’t save us now, and that this result was proved in his 1998 paper, which he increasingly describes as a sort of a road to Damascus moment in his intellectual career. There is just one little problem with this story, the article he cites makes the opposite argument from what he keeps claiming in these posts that link to it. It claims monetary policy can be effective in a liquidity trap, and that an appropriate monetary strategy seems a more promising option than fiscal stimulus. First let’s see what he says in the article, and then figure out what’s really going on.
In his 1998 paper there is only one type of monetary stimulus that is ineffective, a temporary currency injection. Some Krugman defenders will undoubtedly say “well he did concede in his blog that unconventional monetary strategies might work, but they could be risky, etc.” But the problem with that argument is that his 1998 paper says that the most conventional, the most garden-variety monetary strategy imaginable will in fact be highly effective in a liquidity trap. What is the most basic plain vanilla monetary policy? Think about the simple thought experiment that most monetary theory courses use to motivate the discussion. They begin with a once-and-for-all 10% increase in the money supply. You can’t be much more basic than that. Increase the money supply and then leave it at the new and higher level. And in Krugman’s 1998 model that sort of one-time policy shock would be highly effective, even in a liquidity trap. So how does he get the monetary policy ineffectiveness result? By assuming that a current increase in the money supply has no effect on the future expected money supply:
A good way to think about what happens when money becomes irrelevant here is to bear in mind that we are holding the long-run money supply fixed at M*, and therefore also the long-run price level at P*. So when the central bank increases the current money supply, it is lowering the expected rate of money growth M*/M
In other words he makes an assumption (analogous to Ricardian equivalence in fiscal stimulus) that monetary injections are expected to be temporary. But that doesn’t mean monetary policy is ineffective in a liquidity trap, it just means that a silly and misguided strategy might be ineffective in a liquidity trap. Elsewhere I give Krugman credit where credit is due, the Fed has adopted exactly that misguided strategy, promising to pull the recent base injections out of circulation long before NGDP returns to its previous trend line. But it didn’t have to be that way, they could have followed the monetary strategy that Krugman recommended in his 1998 paper. He should be using his column to bash Bernanke, not praise him. Take a look at this passage from 1998, and notice how Krugman first considers the pros and cons of fiscal stimulus, and then ends up suggesting that monetary policy is more promising:
Fiscal policy: The classic Keynesian view of the liquidity trap is, of course, that it demonstrates that under some circumstances monetary policy is impotent, and that in such cases fiscal pump-priming is the only answer. The framework here is rather different in its implications for monetary policy, but it does suggest that fiscal expansion could work. Obviously the model is subject to Ricardian equivalence, so that tax cuts would have no effect. But government purchases of goods and services in the first period, while they would be partly offset by a reduction in private consumption expenditures, could indeed increase demand and output.
While this policy could work, however, is it the right one for Japan? Japan has already engaged in extensive public works spending in an unsuccessful attempt to stimulate its economy. Much of this spending has been notoriously unproductive: bridges more or less to nowhere, airports few people use, etc.. True, since the economy is demand- rather than supply-constrained even wasteful spending is better than none. But there is a government fiscal constraint, even if Japan has probably been too ready to use it as an excuse. And anyway, is it really true that it is impossible to use the economy’s resources to produce things people actually want?
Monetary policy: It may seem strange to return to monetary policy as an option. After all, haven’t we just seen that it is ineffective? But it is important to realize that the monetary thought experiments we have performed have a special characteristic: they all involve only temporary changes in the money supply.
This point needs enlarging upon. Because the traditional IS-LM framework is a static one, it cannot make any distinction between temporary and permanent policy changes. And partly as a result, it seems to indicate that a liquidity trap is something that can last indefinitely. But the framework here, rudimentary as it is, suggests a quite different view. In the flexible-price version of the model, even when money and bonds turn out to be perfect substitutes in period 1, money is still neutral – that is, an equiproportional increase in the money supply in all periods will still raise prices in the same proportion. So what would a permanent increase in the money supply do in the case where prices are predetermined in period 1? Even if the economy is in a liquidity trap in the sense that the nominal interest rate is stuck at zero, the monetary expansion would raise the expected future price level P*, and hence reduce the real interest rate. A permanent as opposed to temporary monetary expansion would, in other words, be effective – because it would cause expectations of inflation.
Let us now bring this discussion back to earth, and to Japan in particular. Of course the Bank of Japan does not announce whether its changes in the monetary base are permanent or temporary. But we may argue that private actors view its actions as temporary, because they believe that the central bank is committed to price stability as a long-run goal. And that is why monetary policy is ineffective! Japan has been unable to get its economy moving precisely because the market regards the central bank as being responsible, and expects it to rein in the money supply if the price level starts to rise.
The way to make monetary policy effective, then, is for the central bank to credibly promise to be irresponsible – to make a persuasive case that it will permit inflation to occur, thereby producing the negative real interest rates the economy needs.
This sounds funny as well as perverse. Bear in mind, however, that the basic premise – that even a zero nominal interest rate is not enough to produce sufficient aggregate demand – is not hypothetical: it is a simple fact about Japan right now. Unless one can make a convincing case that structural reform or fiscal expansion will provide the necessary demand, the only way to expand the economy is to reduce the real interest rate; and the only way to do that is to create expectations of inflation.
By the way, it is odd to describe this policy as “ineffective.” The BOJ said it wanted a stable CPI and for the most part got one. The term ‘misguided’ would be much better than ‘ineffective’ as the latter term suggests (wrongly) that Japanese policy failed achieve its objectives. It succeeded in achieving its objectives; the problem was that it had the wrong objectives.
If we now turn to Krugman’s blog we find him singing a completely different song from the 1998 paper. Now he has one post after another arguing that fiscal stimulus is the answer. And he creates the strong impression that monetary policy is ineffective in a liquidity trap:
In that model, prices are assumed sticky in the short run, so P is predetermined. What, then, determines Y? Well, it’s a real thing “” as opposed to a nominal thing. In the model it’s actually tied down by an Euler condition, by future consumption and the real interest rate (which is stuck thanks to the zero lower bound). Monetary policy has no traction at all against the right hand side of the equation.
Now, the equation still holds. But all that tells us is that any changes in the money supply are offset one for one by changes in velocity. Focusing on nominal spending makes you think that low nominal spending is the problem, a problem with a monetary solution; but actually it’s the symptom, and monetary policy doesn’t matter (unless it can affect expected future inflation, but that’s another story).
I suppose Krugman defenders will again point to the final parenthetical remarks. But how many of his readers would even notice that, or understand what it means? How many people realize that in a rational expectations model that has an upward sloping SRAS curve all monetary stimulus, liquidity trap or no liquidity trap is only effective if it can influence the future expected inflation rate. As a practical matter if monetary injections have no impact on the expected future inflation rate then they won’t impact current AD. So all he is really saying is that monetary policy is ineffective, unless it is effective. I can’t argue with that!
Now let’s review Krugman’s criticism of Levitt and Dubner’s chapter on global warming. Recall that their chapter opened with a charming and entirely accurate anecdote about how “some” scientists had predicted global cooling in the 1970s. But then Levitt and Dubner went on to emphasize that the current science on global warming was quite solid, and therefore we needed to consider some pretty extreme policy options for dealing with this very serious problem. Krugman argued that their opening 1 1/2 pages of this 45 page chapter were misleading, leaving the clear impression that the current consensus for global warming was suspect. I don’t see how anyone reading the entire chapter could conclude that, but let’s say Krugman was right. Even Levitt and Dubner’s worst enemies would have to concede that they also presented lots of evidence in favor of the global warming hypothesis.
Now let’s compare that to Krugman’s post. He makes the very misleading statement that his 1998 paper shows monetary policy doesn’t work in a liquidity trap, when the paper actually ends up arguing that monetary stimulus is one of the most promising policy options for Japan. And then he hides the relevant policy option in a parenthesis. That is far more misleading than anything Levitt and Dubner were accused of doing. After all, Levitt and Dubner’s chapter contained comments like the following:
There is essentially a consensus among climate scientists that the earth’s temperature has been rising and, increasingly, agreement that human activity has played an important role.
And then they spent 20 pages discussing promising options for dealing with the problem. Even worse, Krugman did this when discussing one of the most important issues facing the world today. The unemployment rate hit 10.2% today because of a lack of NGDP. Monetary stimulus plans discussed in Krugman’s 1998 paper offer a very promising way of addressing that nominal shortfall (although NGDP targeting would be even better than inflation targeting.) And Krugman gave the average reader little reason to be hopeful about monetary stimulus.
Here’s how Krugman scolded L&D:
Levitt now says that the chapter wasn’t meant to lend credibility to global warming denial “” but when you open your chapter by giving major play to the false claim that scientists used to predict global cooling, you have in effect taken the denier side. . . . And that’s not acceptable. This is a serious issue. We’re not talking about the ethics of sumo wrestling here; we’re talking, quite possibly, about the fate of civilization. It’s not a place to play snarky, contrarian games.
That’s how I feel about Krugman. He creates the impression that monetary stimulus can’t work, even though his scientific writings suggest otherwise. And a recession causing tens of millions of workers around the world to become unemployed, perhaps for many years, is also a very serious issue.
PS. Check out Beckworth’s excellent post and see how many times he mentions the hated MV=PY equation, the equation that one can’t help thinking about anytime one mentions nominal GDP.
PPS. I am assuming someone else (like Kling) will handle Krugman’s proposal for a new WPA and CCC. How would this work in the 21st century? Someone with a better imagination than me can run with this one. We need monetary stimulus, and we need it soon.
Update 11/9/09: Zubin Jelveh has done what I suggested here.
Tags:
6. November 2009 at 15:26
And see this for the Economists response to Krugman – very similar to what you are saying here.
http://www.economist.com/blogs/freeexchange/
November
6th
15:59 GMT +00:00
Crank up the helicopter
Posted by: Economist.com | WASHINGTON
Categories: Monetary policy
6. November 2009 at 16:05
“although NGDP targeting would be even better than inflation targeting”
As I think you’ve noted before, price level targeting would probably be better than inflation targeting. I think I like NGDP level targeting better.
6. November 2009 at 16:19
Here’s a citation I recalled when reading Krugman’s post:
“It is, indeed, always possible to for the Central Bank, by open market operations, to force out money into balances held by the public. But in times of deep depression, when industrialists see no hope anywhere, there may be no positive rate of money interest that will avail to get this money used. The proper money rate, i.e. the rate which would maintain the standard monetary system, may, in short, be a negative rate, and, therefore, one which is impossible to introduce. In these circumstances attempts to uphold the standard monetary defences, are bound to fail. If, however, at the same time that the banking keeps money cheap the Government adopts a policy of public works, the risk of failure is greatly reduced. For this policy, providing, as it does, new openings for real investment, pushes up the proper rate of bank interest above what it would otherwise have been”. (A. C. Pigou, The Theory of Unemployment, 1933 (p. 213))
It’s ironic that Paul Krugman, who clearly sees himself a modern day evangelizer of pristine, 1936-style Keynesianism, is in fact proselytizing policy advice written by that Keynesian whipping boy Pigou in 1933. The economy fluctuates; some economist’s move in circles.
6. November 2009 at 16:50
JimP, Thanks for the link. David also has a reply to Krugman. It’s getting pretty hard to argue that monetary policy is ineffective when the dollar is falling rapidily. Most people seem concerned about inflation. I think they are wrong, and I recall you also think we need more stimulus, but Krugman is in an increasingly uinteniable position suggesting that tightening would be a disaster but further easing is not possible. At least he’s half right.
dWj, Yes, that is an important point. Level targeting is far better than growth rate targeting regardless of whether P or PY is used.
Torgeir, Thanks for digging up that quotation. Keynes was incredibly unfair to those he disagreed with, often badly mischaracterizing their views. (Interesting that Krugman likes him so much.) In the General Theory Keynes acted like he was one of the few economists who saw flaws with the simple flexible price classical model. Indeed all the good interwar economists had sticky price models of the business cycle, and policy recommendations for dealing with deflation. Often their ideas were similar to those of Keynes, and in some cases even more “progressive” by modern standards. For instance Fisher favored moving to a pure fiat money regime (something Keynes opposed throughout his life.)
6. November 2009 at 17:00
Torgeir:
But that gets to the whole point of the importance of permanence. If the Fed creates money and convinces the public that they will never, ever withdraw that money, then its a loan that never needs to be repaid. If the loan never needs to be repaid then assets depreciate because deflation is irrelevant.
There is also another way in which this works. Krugman often gets confused into thinking that the Fed purchases t-bills from banks to implement policy. This is wrong on two levels–one which I’ve mentioned extensively already on this blog: the Fed normally trades notes not bills for its permanent OMOs. The second is just as important: the Fed trades assets with the PUBLIC (via investment banks) not commercial banks specifically.
This is no need for a loan intermediary. No need for borrowing. The newly created money enters the economy immediately (the Fed ‘overpays’ for the assets it buys) and it can arbitrarily overpay for them–that’s the privilege of being the buyer! This action directly satisfies the ‘excess’ demand to hold cash balances by the public which is the nominal problem.
What can and does happen during a recession though (and not just at the zero-bound) is that business demand for credit slackens. When this happens the marginal money-multiplier declines–which means that it takes strictly more base money to achieve the same macro-scale effect.
6. November 2009 at 17:02
This was a more difficult read than your usual posts, perhaps because I find Krugman almost as interesting as Sumner, but the thing that struck me when I read Krugman’s post, this bit of it — “Underlying the focus on nominal demand or GDP is some notion that there’s a quantity equation… where… the temptation is to take it as a causal relationship “” to say that real GDP fell because nominal GDP fell…” — was his use of the words ‘real’ and ‘nominal,’ the worst words in economics, words my son wisely calls doublespeak because nominal output is what we pay for and what we get, whereas real output is some calculation that tells us what output would have been if prices never changed, though the words themselves suggest just the opposite — the opposite! — and I think, if we are ever to make a science of economics, it will not be because we use words that mean the opposite of our meaning, but words that mean the same as our meaning, though i am too bold to refer to “we” and “our” for i am the child who thinks that (if only he can reach a bit farther) he can touch the clouds, and your blog, sir, is the sky.
6. November 2009 at 17:05
Scott
Yes – I do think we need more stimulus – as I said in my last 5 replies to your previous post – in rising frustration. I will repeat myself – I think Obama has to act more like the early Roosevelt – making his desire for inflation clear. Financing health care reform with a 60% tax bracket in the middle of the worst drop since the depression is complete madness – and the House is going to vote on that tomorrow. The democrats have simply lost their minds.
Finance health care reform with a vat. Obama should give us an income tax cut and Bernanke should finance it with his chopper. Its either that or Obama is just going to be out of a job. Thank God we still have a democracy and can toss these morons out – and replace them with different morons.
6. November 2009 at 17:22
Fiscal Stimulus JimP?
Thats a terrible idea.
I can understand monetary stimulus and we can debate that, but fiscal stimulus is just plain a horrible idea.
From this paper.
“All news measures suggest that most components of consumption fall after a positive shock to government spending. The implied government spending multipliers range from 0.6 to 1.1. ”
1) This isn’t the only paper to come to a similar conclusion
2) This means that at best government spending is more or less a net wash in the short term, and at worst it causes actual damage to the economy.
3) I’ve said it before, elsewhere, government spending crowds out individual spending.
4) Government deficit spending crowds out individual investment AND individual spending.
6. November 2009 at 17:25
Sorry point 4 should read:
“4) Government deficit spending crowds out individual borrowing AND individual spending.”
6. November 2009 at 17:41
Jon, Those are good observations. I would add that under NGDP targeting if there is a recession then inflation rises as RGDP falls. So nominal interest rates might be fairly stable. The real rate would fall during recessions (as you say) but the inflation component would rise. So demand for base miney would probably be much more stable than under our current regime (even apart from the crazy interest on reserves policy.)
The Arthurian, Terminology will be a problem as long as there is confusion over the underlying concepts. And I don’t see any signs that that confusion is going away, indeed economic crises always set us back.
JimP, Obama somehow picked the one University of Chicago professor who doesn’t understand supply-side economics as his advisor. (Austen Goolsbee.) I think Obama is getting bad advice. The health care bill is shaping up to be a massive increase in marginal tax rates on the working class and the rich.
Doc, I believe JimP supports more monetary stimulus (a VAT is not stimulus.)
I agree that fiscal stimulus is not the way to go. In fact I think even the Free Exchange post was too favorable toward fiscal stimulus (although in the end I recall he leaned toward monetary stimulus as being preferable.)
6. November 2009 at 17:42
Doc
Not fiscal stimulus. Monetary. Positive inflationary expectations, best done exactly as Scott would have them done.
6. November 2009 at 17:43
Not long ago I heard a baseball manager say something to the effect: “You are only judged harshly if you do something that meets 2 conditions: 1) It is unconventional; 2) It doesn’t work.”
The specific context I believe was batting the pitcher in the 8th spot instead of 9th (in a situation where the specific composition of the lineup caused it to make statistical sense). The broader context was resistance to change, despite what makes more sense on paper a la Bill James.
Somehow I suspect it is meeting the above 2 conditions that both Krugman and Bernanke are afraid of. Despite what the numbers may tell them, they are at bottom terrified of batting the pitcher in the 8th slot AND losing the game. By continuing to bat the pitcher 9th the public won’t judge them as harshly if they lose. In other words, what they lack is simply balls.
6. November 2009 at 17:50
rob, I seem to recall Krugman actually saying he’d support a higher inflation target, but thinks modern central bankers are too conservative to implement such a policy. Perhaps someone else can confirm that. I do agree with you about Bernanke, although I think it may be more complicated than it seems from the outside. He faces enormous instuitutional inertia, which would limit his ability to act even if he was so inclined. I guess we’ll have to wait until we read his memoirs. But given that so many academics outside of the Fed seem happy with their policy, it wouldn’t surprise me if a relatively moderate fellow like Krugman has much the same views as Mishkin, and other non-Fed economists.
6. November 2009 at 18:46
“Here’s my problem. Underlying the focus on nominal demand or GDP is some notion that there’s a quantity equation:
MV = PY
where M is the money supply, V the velocity of money, P the price level, Y real GDP…..”
All this confusion, and all this wonderment at which measure of money to use. It comes from one thing alone. One item out of these four.
“Y is real GDP”
We should not be using GDP for Y. Okay so Y is usually doubling for national income. But another variation of that same formula is MV=PQ.
See we are using GDP for Q. We must not and we cannot do that. This is where all the confusion comes in. If we substitute in Gross Domestic REVENUE for Q…. then within about a year of doing statistical work on the figures, everything will fall into place.
And these years of thinking maybe we ought to us monetary base, or wider monetary aggregates for M…. Well this will all fall away like a bad dream.
You think I’m being ridiculous laying forth on ten different subjects at once. But comparing paradigms can become a bit of a specialty in and of itself. I cannot do the quant. work that needs to be done. I’m just trying to make it easier for everyone.
When people get familiar with the statistics again after making this basic mental-shift, they will see that production is the production of producer and intermediate goods as well as consumer goods. That broad monetary aggregates are not money but what money buys. And that monetary base is the banks money whereas M1 is the publics money.
You guys ought not waste your time by trying to see patterns between metrics that do not belong together.
6. November 2009 at 19:03
@ JimP and Scott:
Hrm, I must have misread, I think his comparisons to FDR threw me off.
6. November 2009 at 22:03
Note well, including Hayek.
You write:
” all the good interwar economists had sticky price models of the business cycle”
6. November 2009 at 22:08
“MV = PY…And of course this always holds true, by definition. ”
“The second sentence has to be one of the weirdest things I have ever read by a famous economist. I have no idea the point he is trying to make.”
Well, maybe he’s just saying:
“Keynes’s entire General Theory is basically a theory of PY, and hence necessarily a theory of MV. (In fact V doesn’t have any independent meaning; it’s just PY/M.)”
as you yourself say just a few paragraphs later. Seems pretty straightforward.
6. November 2009 at 23:03
The Arthurian: “[T]he words ‘real’ and ‘nominal,’ the worst words in economics, words my son wisely calls doublespeak because nominal output is what we pay for and what we get, whereas real output is some calculation that tells us what output would have been if prices never changed, though the words themselves suggest just the opposite”
Not to derail the conversation, but I disagree. I think the words are actually very accurate, and if they don’t seem so (which I will agree that they don’t) it is only because people think wrong.
Think of a company sending a truckload of goods across the country. Along the way, every time the truck stops to refuel, bandits sneak in the back and steal some of the cargo. When the truck reaches its destination, the receiver checks the manifest and, sure enough, 1 truckload was sent and one was received. Perfect transaction.
Of course it isn’t. The actual value of the shipment is the value of the cargo that remains on board, not the arbitrary value of a “truckload.” In nominal terms, thing look alright; only by looking at the real value of the shipment can you see how much you lost along the way. In other words, people too often confuse dollars with something of value.
7. November 2009 at 03:02
Scott,
As soon as I saw that Krugman post I guessed it would drive you up the wall. I put in a mention for you in comments. But getting riled isn’t helping you to understand his economics, or Keynes’s.
In fact, almost everything Krugman has written on the crisis has implicitly assumed that falling NGDP is precisely the cause of our recession, and that stimulus aimed at boosting NGDP is the solution.
I think I’ve read nearly everything Krugman has written on the crisis and I never noticed that he was implicitly assuming that falling NGDP is the cause. How could it be? If variable X is seen as a cause of some phenomenon, doesn’t it have to be an exogenous variable in any sensible model of that phenomenon? Has Krugman ever presented a model in which NGDP is exogenous?
Sure Keynes often assumes fixed prices in the GT, but not always.
You surely can’t mean that Keynes “often” assumes fixed goods prices. Keynes assumes fixed wages for most of the book, but he stresses that he is doing so for purposes of exposition and that his conclusions do not depend on fixed wages. Goods prices are flexible, as indeed they must be since the goods market clears, unlike the labour market. Here and there Keynes considers the possibility that there are unintended changes in inventories, but that’s taken to be a very transitory thing. For the most part prices are flexible, short-term expectations are fulfilled and firms are producing profit-maximising quantities.
So the General Theory is first and foremost a theory of nominal income determination.
Words fail me. So I’ll steal those of Alan Coddington: that’s not just reading between the lines, that’s reading off the edge of the page. The GT is what it says on the cover: a theory of employment etc. Employment is determined by aggregate demand, which is a real variable measured in wage units. NGDP barely gets a look-in.
But the problem with [Krugman’s] argument is that his 1998 paper says that the most conventional, the most garden-variety monetary strategy imaginable will in fact be highly effective in a liquidity trap.
For central bankers “to credibly promise to be irresponsible” isn’t conventional at all. Didn’t Mishkin say somewhere that the Bank of Japan was so shocked at Krugman’s suggestion that it made them even more intransigent? If that’s true, or if Krugman merely believes it could be true, it would explain why he isn’t keen to press the same advice on the Fed. But I don’t know where Mishkin got his information.
7. November 2009 at 05:44
Scott:
Nominal expenditure targeting and expected inflation targeting-real interest rate targeting aren’t the same.
You think the Bank of Japan had the wrong goal–price level stability–and that is why their economy was depressed. If the Bank of Japan targetted nominal expenditure at a rate consistent with their growth of capacity, then price level stability could be achieved–more or less. If the problem in Japan was low nominal expenditure, then reversing that should be possible without inflation.
Permanent increases in the quantity of money? Why should the Fed respond to a temporary increase in the demand for money with a permanent increase in the quantity of money? You are complaining that the Fed is promising to reverse the increase in base money. Well, I think that the demand for base money will almost certainly fall in the future from its current high levels, and I think the Fed should decrease the quantity of base money when that happens and they should be openly committed to that future policy. Nominal expenditure targeting would imply just that policy for base money.
That the Fed is targeting expected CPI inflation (maybe,) _is_ a problem. And promising to reverse the increase base money when the CPI is expected to grow 2% from its current level rather than when nominal expenditure theatens to pass its previous growth path, may well be a mistake. But that has to do with the growth rate/level targeting issue not permanent vs. temporary increases in base money.
I do think there has been a capacity shock. In an ideal world, with nominal expenditure targeting, we would have had some inflation. And so, if we think about that price level as the implicit target, I suppose we should be targeting some inflation now. (And if and when that capacity shock dissipates, there will be deflation or extra low inflation.) But if there hadn’t been a capacity shock, then there should be no goal of any sort to cause inflation.
Now, if we had deflation (which by many measures of the price level, we haven’t,) then reversing that deflation would imply inflation too. But I don’t see why the temporary decrease in real interest rates implied by the inflaton needed to reverse deflation will necessarily be consistent with returning nominal expenditure to target.
What would happen is that the price level must get to a sufficiently low level, so that the expected rate of increase from that low level results in a temporarily low real interest rate (perhaps very negative) that will result in sufficient real expenditures so that there isn’t any further downward pressure on the price level. With sticky prices and wages, we are left with a depressed economy. All one can say is that it would be even worse if a permanent decrease in nominal expenditures is expected.
I think you did a good job criticizing Krugman. But I think his response to you would be that the Federal Resreve needs to promise say, 7% inflation for the next 3 years, causing the price level to permanently increase by about 25%. This would make real interest rates on T-bills, FDIC deposits, and the like, temporarily very negative. And that would motivate plenty of real purchases over the next few years, causing a recovery. If the need for highely negative real interest rates is permanent, then a higher inflation target is needed. This entire approach is about driving the target for nominal expenditures and inflation to get the real interest rate where we need it. Horrible!
Anyway, the Fed can not promise a 25% increase in the price level over the next 3 years for political reasons.
Since I support 9% nominal expenditure growth over the next year, I suppose that one could try to estimate what that would imply for CPI inflation over the next year, and the value of the CPI one year from now. Perhaps that would imply quite negative real interest rates on T-bills and FDIC insured deposits and the like. Perhaps I should try to figure that out. What is the slope of sras over that range?
The real problem with Krugman is more basic. We don’t have “the” nominal interest rate at zero. We have some nominal interest rates very near zero. It is possible that open market operations in zero nominal interest rats bonds will result in an increase in the demand to hold money matching the increase in the quantity of money. (V will fall to offset increased M.) It has to do with the reality that those selling the assets to the Fed don’t have them anymore. Maybe they will substitute newly created FDIC insured depostis for the T-bills they no longer own.
But once all of those zero nominal interest instruments are purchased by the Fed, then either we give up, and the problem is there are no assets for the Fed to buy, or else we buy other assets. And the the nominal interest rates on those other assets fall too. When all nominal interest rates are at zero, or assets with yields that are not zero cannot be purchased by the Fed for some reason, then we are in the situation Krugman has in mind. And the problem clearly is that the Fed can’t do more open market purchases because there is nothing left that is suitable to buy.
We are nowhere near there yet.
7. November 2009 at 06:45
@JimP:
I have been thinking of your proposal to raise inflation expectations and I see one large problem. In inflation expectations and high and interest rates low the carry trade will absorb a lot of that excess liquidity, until there is a large bump in the dollar then it will unwind at once, Roubin style, causing a crash.
With freely floating exchange rates, weird things happen when Central Bankers try to manipulate markets.
7. November 2009 at 06:46
Bah, poor, editing on my part. It should read “If inflation expectations are high and”
7. November 2009 at 07:16
Graeme, Thanks for the advice, but I think I’ll keep wasting my time.
Greg, That’s right.
PR. Replace nominal GDP with nominal interest rates. Here’s how it would read:
Underlying the focus on nominal interest rates is some notion that there’s a Fisher equation:
i = r + infl.
where i is the nominal interest rate, r is the real interest rate and infl is expected inflation And of course this always holds true, by definition. But the temptation is to take it as a causal relationship “” to say that real interest rates fell because nominal interest rates fell.
Of course I agree that MV=PY is an identity. The question is why Krugman even brings it into the discussion of nominal GDP. Why does that make NGDP a misleading concept? Are nominal interest rates also a misleading topic? Krugman talks about nominal itnerest rates all the time.
Mario, Nice analogy.
Kevin; You said;
“I think I’ve read nearly everything Krugman has written on the crisis and I never noticed that he was implicitly assuming that falling NGDP is the cause. How could it be? If variable X is seen as a cause of some phenomenon, doesn’t it have to be an exogenous variable in any sensible model of that phenomenon? Has Krugman ever presented a model in which NGDP is exogenous?”
In Krugman’s model of this particular recession sharply falling NGDP growth is both a necessary and sufficient condition for the recession. That makes it a proximate cause. Obviously there are always deeper root causes explaining why NGDP fell.
The point is that in Krugman’s model any policy that prevented NGDP growth from falling below 5% would have prevented a severe recession. He doens’t buy real models of this recession. His model is a nominal shock model, as economists generally understand the term ‘nominal shock’.
You said;
“You surely can’t mean that Keynes “often” assumes fixed goods prices. Keynes assumes fixed wages for most of the book, but he stresses that he is doing so for purposes of exposition and that his conclusions do not depend on fixed wages. Goods prices are flexible, as indeed they must be since the goods market clears, unlike the labour market.”
If this is true, do you believe Krugman is making a mistake in his assumption that prices are fixed? Regarding Keynes, I said that Keynes assumed prices were fixed at less than full employment, but that as one approached full employment prices would start rising. I don’t recall his specific assumption about relative prices. He definitely assumed that higher AD would raise output, not the price level, during recessions. If he allowed for some individual variation in relative prices as long as the overall price level was fixed, then that’s fine. But it doesn’t change anything material in my argument.
You said;
“Words fail me. So I’ll steal those of Alan Coddington: that’s not just reading between the lines, that’s reading off the edge of the page. The GT is what it says on the cover: a theory of employment etc. Employment is determined by aggregate demand, which is a real variable measured in wage units. NGDP barely gets a look-in.”
Then you need to reread the GT, as this is completely false. Keynes clearly says that higher AD determines employment only during recessions; at full employment he argues that more AD raises prices. So his “general” theory says that AD sometimes effects employment and always affects NGDP.
You said;
“For central bankers “to credibly promise to be irresponsible” isn’t conventional at all. Didn’t Mishkin say somewhere that the Bank of Japan was so shocked at Krugman’s suggestion that it made them even more intransigent? If that’s true, or if Krugman merely believes it could be true, it would explain why he isn’t keen to press the same advice on the Fed. But I don’t know where Mishkin got his information.”
This is false. The BOJ is the only central bank that I am aware of that targets zero inflation. Most have something closer to a 2% inflation target. Positive inflation is exactly what Krugman meant be “being irresponsible” The Japanese policy didn’t “fail,” it achieved what the BOJ wanted. The problem was that they wanted the wrong thing. Yet Krugman continues to argue falsely that monetary stimulus in Japan was “ineffective”, as he doesn’t seem aware of the fact that the BOJ succeeding in hitting their CPI inflation target once they started QE.
If the Fed was opposed to higher inflation then why in the world would they have advocated fiscal stimulus, which is even more inflationary than an equivalent amount of monetary stimulus? (I.e. a monetary stimulus with the same effect on NGDP as a given fiscal stimulus would be less inflationary.)
BTW, your argument about monetary policy is irrelavent to my argument the Krugman created the false impression that monetary policy is ineffective in a liquidity trap. If Krugman really means that monetary policy is highly effective but that central bankers are too conservative to do the right policy (and you may be right about his views) then that needs to be said EVERY SINGLE TIME HE MENTIONS THE TOPIC, as it implies a catastrophic worldwide recession is being needlessly caused by unelected morons at the central banks. That should be the number one political issue in America if that’s what Krugman really thinks.
He says L&D are global warming deniers, I say there is far more evidence that krugman is a monetary policy denier. Yes, it is hard to get central bankers to move. You may have also noticed that its hard to get the US and China to cut back on carbon. But we still need to try, don’t we?
More to come . . .
7. November 2009 at 07:42
Bill, Japan had rough CPI stability after QE started, but the NGDP deflator kept falling gradually. So I don’t see the inflation/NGDP dsitinction as being their key problem, rather that the target was too low, either way. Both the CPI and NGDP were roughly stable.
BTW, I don’t think monetary policy was their main problem, but rather supply-side factors. However, I think tight money aggrevated those problems in the early 1990s, the late 1990s, 2001, and 2008-09.
Of course not all of the increases in the money supply should be permanent, the Fed should promise to keep enough of it in circulation forever to make sure prices (or preferably NGDP) rise at the target rate.
I agree with you that NGDP targeting is better than CPI. I just mention inflation, because everyone else looks at that variable, and a higher CPI target would also be very helpful.
I agree that Krugman thinks high inflation is necessary, but I don’t get that argument, as it contradicts his own (correct) argument that the SRAS curve is currently fairly flat. What does Krugman think would happen with 3% inflation, rather than 7%? Stagflation? That’s Klings’ argument, but I thought he disagreed with that argument.
You said;
“Since I support 9% nominal expenditure growth over the next year, I suppose that one could try to estimate what that would imply for CPI inflation over the next year, and the value of the CPI one year from now. Perhaps that would imply quite negative real interest rates on T-bills and FDIC insured deposits and the like. Perhaps I should try to figure that out. What is the slope of the sras over that range?”
I don’t follow your argument here. I agree it would be great if we got a 1983-style recovery. But it surely would raise rates, not lower them. If NGDP was expected to grow 9% the demand for credit would soar. Remember that both the money supply and nominal interest rates become endogenous once you set the NGDP target. With 9% NGDP growth I would expect inflation of around 2%, maybe 3%. But after one year it would be disastrous to keep up that NGDP growth rate, as inflation would soar.
I am actually more sympathetic to Krugman than you are. I don’t think his basic flaw is overlooking higher yielding assets that can be bought. I agree with Krugman that the low inflation expectations are the key problem (and by implication the low expected NGDP growth as well.) And I agree that even buying higher yielding assets wouldn’t do a whole lot unless the policy was expected to permanently impact the price level (and hence money supply.)
Doc, If my policy of raising NGDP growth expectations worked, then nominal interest rates would also rise.
7. November 2009 at 09:55
It seems to me that when a fall in money velocity is offset by an expansion of the money supply, the *composition* of nominal expenditure should shift from consumption to capital goods. In other words, new money should be injected into sectors of the economy involved with the production of future goods in anticipation of future spending. In my opinion, one of the problems with a fiscal stimulus is that it is likely to distort the composition of nominal expenditure in an unsustainable manner.
A fall in money demand is not uniform — the blockage in the stream of money occurs in some places more than others. It is important for the efficient allocation of resources that money injections match the non-uniform rise in the demand for money. A lot of talk about the fiscal stimulus, however, seems to involve injecting money where demand has rose the least.
7. November 2009 at 12:05
Lee:
A fall in velocity is an increase in the real demand for money relative to real income. If there were a shift from the demand to hold corporate bonds to holding money, then there would be no impact on the allocation of resources between consumer and capital goods if the quantity of money increased to meet that demand.
You are assuming that increase in the demand for money is additional saving–reduced expenditure on consumer goods. That is possible, but really, it depends.
If the goal was to somehow make sure that the increase in the quantity of money has no impact on the allocation of resources, then worry about where the demand for money (reduced flow of expenditure) occured would be important.
But the relevant comparison is to an incease in the demand for some other asset. For expamle, if people eat out less, and instead purchase corporate bonds, there will be a shift in the allocation of resources away from restaurants to capital goods. If people choose to eat out less and accumulate money, there is no problem if this impacts the allocation of resources–less restaurants and more capital goods. That is what should happen.
The problem with money is that when people “buy” more of it, the result isn’t that they go to the money shop and buy more. The price doesn’t rise. And there is no obvious reason for the yeild to fall. People just spend less and don’t buy anything.
The notion that what should happen is that the allocation of resources is entirely uneffected is a mistake.
If the market reponds by deflation of prices, that is going to impact the allocation of resources too. But probably in a different way.
7. November 2009 at 12:46
Scott:
Krugman is explictly saying that real interest rates must be very negative to motivate an increase in real and nominal expenditure.
If the Fed promised 3% inflation, people would still not not spend much, and any increase in the quantity of money (aimed at generating that inflaiton) would be hoarded.
The Fed’s promise of 3% inflation would have little effect, and inflation wouldn’t be 3%.
People would be willing to bear that cost, and continue to hold T-bills, FDIC insured deposits and the like, rather than spend.
In other words, the fact that the Fed said things will be 3% more expensive next year will not motivate people to buy now, rather than hold money because they are afraid that they will lose their job.
This notion about temporary vs. changes in the quantity of money is useless if the change in the demand for money is temporary and the goal is to keep nominal expenditure or the price level on target.
The change in the quantity of money is supposed to accomodate a termporary increase in the demand to hold money at a given growth path of nominal expenditures and a given growth path of the price level. When the demand for money falls again, the quantity of needs to fall again. Making it permanenty, so that nominal expeniture mores to a higher growth path would be counterproductive.
So, “permanent’ purchases of T-bills or a permanent increase in base money is inappropriate with the increase in money demand is temporary. (I think you are implicitly assuming that a temporary increase in money demand cannot possibly cause a decrease in nominal expenditure.)
If the only reason why T-bills and the like can be near the zero nominal bound is because of a monetary policy failure, then I suppose fixing the monetary policy failure avoids the problem. But I don’t think that is necessarily true. Nothing could cause the demand for safe and short term assets to rise enough to push their nominal yields to zero other than expectations of falling nominal expenditure?
7. November 2009 at 13:10
This Krugman post is the worst ever. He completely misses the reality of increased inflation expectations since depths of the crisis as measured by TIPS spreads.
7. November 2009 at 13:19
Speaking of leaving the wrong impression, Free Exchange at The Economist again links to here and to the economist burdened by a conscience:
http://www.economist.com/blogs/freeexchange/2009/11/the_heedless_fed.cfm
And summarizes with the statement: “America needs expected inflation if monetary policy is to have any effect!”
I’m hoping the cart-before-the-horse irony is intentional. The post says that PK and SS make the same point, but the points quoted seem to miss the ultimate point of both. It fails to arrive at the conclusion that it is not the monetary injections that matter per se, but monetary policy expectations.
7. November 2009 at 14:40
Bill,
—–quote—–
You are assuming that increase in the demand for money is additional saving-reduced expenditure on consumer goods. That is possible, but really, it depends.
—–quote—–
You’re right, I was making that assumption. Thanks for pointing it out.
7. November 2009 at 16:39
First some general remarks about that famous MV = PY. Let’s do a slight transformation on it:
M_new*V_new + M_old*V_old = PY
It makes sense to have such a distinction if V_new is very different from V_old, as it seems to be the case in the current circumstances. And since Fed has direct control only over M_new, and V_new is close to zero than, if you want to increase P (even if Y is falling)… than I guess you have a problem.
Mr. Summner, correct me if I’m wrong, but in order to have steadily increasing NGDP, you (we) need to have consumers steadily increasing their monetary expenses, right? Even if those, for whatever reason, cannot afford it?
Yes, most probably, the feds can create such conditions that consumers will increase their money expenses now, seeing it as a lesser (economic) evil, but one needs to create a greater (economic) evil for this to work.
Mr. Woolsey, you were very close to becoming my economic hero, but deftly avoided it. You are quite right when you place the blame for decreasing velocity/spending on increased demand for money, you even seem to see that demand for money is demand for delayed consumption, except that later you join the chorus of Evil Economists which frantically try to figure the best way of derailing such attempts.
You’re partially absolved by asking the right questions, like: What will happen, “if market reponds by deflation of prices”. Well, this is easy. People (and businesses) will see their real money balances increased and will start spending them again. This time because they want to, not because they are forced by an Evil Central Bank.
7. November 2009 at 18:20
Any comments on this?
http://www.clevelandfed.org/research/commentary/2009/0809.cfm
“New Approach to Gauging Inflation Expectations”
7. November 2009 at 19:17
Im trying to give Krugman the benefit of the doubt for a moment in figuring out the apparent inconsistency between the 98 paper on japan and his views on the US today. He described Japan as a country with “poor long term growth potential” says his conclusion is “radical” and uses the phrase “otherwise irresponsible policy” more than once. so maybe he viewed japan as a very different situation, one which required what would otherwise be radically irresponsible monetary policy for a country with good long term growth potential. if so, he is not inconsistent.
7. November 2009 at 20:52
In the past I’ve suggested that Krugman might be viewing Fiscal policy as a mechanism to inject money into the economy… In other words, a form of “unorthodox” monetary policy where fiscal spending (rather than bank lending) is the mechanism (on the presumption we have structural problems with banks). He _seems_ to have advocated QE. But now I can’t really tell what he’s saying. Nor can I tell what his explanation for the crisis currently is (if it’s not real, and not nominal (e.g. debt deflation), then what is it? Spontaneous self-reinforcing demand contraction created by a Panic? And then, how does it get fixed?
Blech.
7. November 2009 at 21:30
I would raise a challenge to one of the primary mechanisms suggested by others about how inflation increases expenditures. For example, David Beckworth (in his most recent post) writes:
“if folks think that inflation will be permanently higher going forward they are more likely to spend their money today. That is, money demand will fall and velocity will pick up.”
http://macromarketmusings.blogspot.com/2009/11/my-reply-to-krugman.html
The same sentiment is reflected in the Fed article that Larry linked, and elsewhere above.
This sounds intuitive, but the problem is that financial markets move much faster than household budgets.
Here’s a realistic example: Let’s say the Fed credibly announces that it’s going to do everything it can to achieve a 7% increase in prices next year, then return to a 2% increase thereafter. Pretty much instantaneously, the price of Oil and other internationally traded commodities jumps to reflect this new information.
Do consumers then spend the next year rushing out to buy more gasoline? Aside from the fact they can’t easily store it for a couple years, they are simply too slow. The price of oil will have jumped before most of them even realize what happened, and there is no reason to presume that demand will thus pick up. Indeed, after a week goes by and retail prices are updated, it’s quite plausible that consumers will cut back on oil-related purchases (like heating oil contracts) in the hope that oil prices might fall back down.
Likewise, while it might take time for house prices to adjust (those are sticky), mortgage rates adjust in less than 24 hours. Does that mean that people accelerate purchases (thinking that rates might go up higher) or hold off (thinking rates might drop lower). The process of expectation-formation among the general public is not entirely clear in this instance (and this is why behavioral research is important).
But this does not mean that monetary policy is not effective… it simply that this mechanism is limited (to situations where prices are very sticky), and there are other mechanisms at play. Notably, one major mechanism is ASSET PRICES, which includes the value of debt.
I’m confounded why this seems to be ignored frequently by macroeconomists as a major mechanism, when finance economists seem to get it. The issue of concern is Wealth, as well as Perceived Wealth. Maybe economists just don’t like talking about wealth, but any cursory review of conventional media reports identifies this a key (if not the primary) factor – we are spending less because we are poorer. And it wasn’t just that we woke up and “discovered” we were always poorer (e.g. the “Great Recalculation). We are REALLY poorer, because we are NOMINALLY poorer, because our DEBT is nominal.
Any economist who rejects this mechanism would – by necessity – need to reject a huge body of evidence that shows that households use wealth to smooth out intertemporal consumption when confronted with income shocks. They would also have to ignore the conventional wisdom that many people (in 2003-2007) increased expenditures dramatically due to an increase in home equity due to rising asset prices. They would also have to pretend that bankruptcy doesn’t exist and all households have no liquidity constraints (which the old classical models in fact assume).
Moreover, the asset price explanation (and the wealth effect) are entirely consistent with microeconomic theory, which suggest that households are risk averse due to decreasing marginal utility curves. (That is to say, poorer households save more – which is why poorer households save more even while incomes are lower.)
Yet the unfortunate truth is, unexpected changes in interest rates (or money supply) are a transfer of Wealth, and Bill Woolsey is precisely correct that this creates political impediments to doing what needs to be done.
But many economists share the blame for our current failure – they seem to be (with few exceptions, like ssumner) very squeamish when talking about asset prices and debt load. Perhaps these topics are politically inconvenient.
7. November 2009 at 21:44
“poorer households save more” should be taken to mean that when a given household becomes poorer due to a wealth shock it tends to save more (holding other things, such as income, constant).
7. November 2009 at 22:19
Suppose truth is a woman – what then?
(profundity is the last refuge of the drunk)
i believe Nietsche doesnt get the credit he deserves as an economist. he understrood the role of incentives among his fellow academics and didnt believe in the concept of truth for thuth’s sake.
Is there not not sufficient reason for believing that all philosophers so far, in as far as they are dogmatists, are very inexpert regarding women?
8. November 2009 at 00:48
@Bill:
“The problem with money is that when people “buy” more of it, the result isn’t that they go to the money shop and buy more. The price doesn’t rise. And there is no obvious reason for the yeild to fall. People just spend less and don’t buy anything.”
The ‘price’ of credit doesn’t accurately reflect the demand/supply of money/credit because money is somewhat price fixed through the federal reserve.
@StatsGuy:
You are correct:
Inflating lowers assets/liabilities that are inherently nominal. It also affects taxes that are nominal in nature (a lot of excise taxes). However, the tactic of using inflation to get rid of nominal problems won’t work for much longer, because a lot them are directly or indirectly indexed to inflation. Also, more importantly, I believe lags and stickyness is decreasing due to increased computerization and decreased unionization.
Examples of inflation linked nominal things:
Now some debt (ARMs) are indirectly indexed to inflation.
Government employee pay increases are indexed to CPI.
Some unionized employee pay increases are indexed to CPI.
Medical benefits’ costs will rise as inflation does so
8. November 2009 at 07:27
Lee kelly, You said;
“In other words, new money should be injected into sectors of the economy involved with the production of future goods in anticipation of future spending.”
I think this confuses money and credit. Almost all new money sits in people’s wallets. If the US had no reserves requirements then during normal times probably 98% of money would be in people’s wallets. Credit is different, but the Fed doesn’t create credit, people do.
Bill, I agree with you about money, changes in M and V have pretty much the same impact.
Bill#2; This confuses me:
“If the Fed promised 3% inflation, people would still not not spend much, and any increase in the quantity of money (aimed at generating that inflaiton) would be hoarded.
The Fed’s promise of 3% inflation would have little effect, and inflation wouldn’t be 3%.”
Suppose the Fed injected enough money so that the CPI futures (or TIPS) markets showed 3% expected inflation. What would be the expected real growth rate? One part of Krugman’s model says low, as you note, but the other part (flat SRAS curve) says growth would be high. His model is internally contradictory. In fact I may do a post on that problem, as I think it is really important.
You said;
“So, “permanent’ purchases of T-bills or a permanent increase in base money is inappropriate with the increase in money demand is temporary. (I think you are implicitly assuming that a temporary increase in money demand cannot possibly cause a decrease in nominal expenditure.)”
You have misunderstood my point. I completely agree with you that temporary increase in Md should only be accommodated temporarily. My point is that you must promise to keep enough money in circulation permanently to leave the expected price level higher than today. Right now the Fed is essentially promising to pull enough money out of circulation so that markets expect the price level in 2011 to be only about 1% higher than 2009. But the Fed should want the 2011 price level to be at least 4% higher than today. To achieve that at least some of the money injected must be kept in circulation. I agree that most will be withdrawn (if they stop paying interest on reserves) but perhaps 4% of the roughly 100% increase must be expected to be permanent.
You said;
“If the only reason why T-bills and the like can be near the zero nominal bound is because of a monetary policy failure, then I suppose fixing the monetary policy failure avoids the problem. But I don’t think that is necessarily true. Nothing could cause the demand for safe and short term assets to rise enough to push their nominal yields to zero other than expectations of falling nominal expenditure?”
I agree that this is a judgment call, and I might well be wrong. But I suspect that if NGDP was expected to grow fast enough to return us to even the 3% trend line you discussed (wasn’t that about 9% growth?) then nominal rates would actually rise.
More to come . . .
8. November 2009 at 08:16
123, Yes, one of many problems. He also pays no attention to how asset markets respond to Fed policy in a liquidty trap—perhaps he pays no attention because his model implies easing should have no effect.
rob, I just don’t know quite what to make of that. It looks like I need to do another post. My magnus opus on monetary theory (BTW, don’t expect anything new) will have to wait another day, although it is basically done.
The problem we face is not long term, it is cyclical. We are facing a couple years of suboptimal NGDP. The solution is much higher inflation expectations over the next 2 years, not 30 years. Again, (in case Bill is reading this) I much prefer focusing on NGDP, it’s just that I have to discuss inflation expectations because that’s the language everyone else uses. Most people talk about the importance of keeping inflation expectations low out of one side other their mouth, as they advocate fiscal stimulus out of the other side. My point is that fiscal stimulus will only work if it boosts AD. And higher AD raises inflation expectations.
Vasile, One thing I agree with Krugman about, the MV = PY equation isn’t a very helpful way of thinking through macro problems. I want a monetary policy stance that leads to on target expected growth in PY. I don’t much care what happens to M and V individually.
Larry, Thanks. Well it is nice that they show the huge spike in real rates in late 2008. Others had contested my observation that real rates soared. And I agree that long term inflation expectations are near 2%. But my enthusiasm is tempered by the one year forecast. Both TIPS spreads and CPI futures show that is very low. Can both be so far off? I doubt it.
rob, A few months ago he suggested the Fed should have a higher inflation target, so I doubt his views have changed much. He just tends to avoid emphasizing them in his recent posts.
Statsguy, I agree with your first point about Krugman. He’s saying it’s neither real nor nominal. In your second post you said;
“This sounds intuitive, but the problem is that financial markets move much faster than household budgets. Here’s a realistic example: Let’s say the Fed credibly announces that it’s going to do everything it can to achieve a 7% increase in prices next year, then return to a 2% increase thereafter. Pretty much instantaneously, the price of Oil and other internationally traded commodities jumps to reflect this new information.”
I think this confuses shifts in demand with movements along an demand curve. Easy to spot with individual markets, but a point often overlooked with AD. Imagine AD curve shifts right. Both P and Y increase. The rise in P doesn’t prevent a rise in Y, rather both are effects of more AD moving you along the SRAS curve. If commodity prices rise due to more demand, then output can still rise as the commodity producers respond by shifting out along their supply curves toward more production. In fact commodity prices wouldn’t rise in the first place unless markets expected the AD increase to lead to more demand for those commodities, and that almost surely means more GDP. Now of course for any given increase in NGDP, we’d like more Y and less of an increase in P. That’s why I think expected inflation is so misleading. The concept we are actually interested in is expected AD growth, i.e. expected NGDP growth.
When you start talking about asset prices, then I am on board. In an earlier post I spelled out my view of the transmission mechanism. A rise in M expected to be permanent will cause expected future NGDP to rise. This will immediately raise asset prices (stocks, commodities, real estate.) The higher asset prices are a key factor in motivating higher current expenditures.
rob#2, You’re lucky you put that comment in my blog, as I think 90% of bloggers would think you were drunk. I have what McCloskey would probably consider a “feminine” method of analysis. I hate the masculine “it all boils down to X” view of monetary policy, which I criticized in an earlier post. I like to look at monetary economics from 100 different perspectives. There are just two many important angles to boil it down into a single model.
Except of course NGDP. I really does all boil down to NGDP.
Doc merlin, Two points:
1. The Fed doesn’t fix prices, it influences prices by fixing quantities. The fed funds rate can and often does diverge from their target.
2. Yes, there are less unions, but many white collar jobs have fixed one year contracts.
8. November 2009 at 13:01
@Scott
Yes I am aware of how the fed adjusts prices. That is why I said, “somewhat price fixed” not “price fixed.” It still suffers from similar problems. The fed’s price will give you nominal supply and demand information. However, it doesn’t include information about market expected growth and inflation into its supply curve because it is disconnected from savings/investment. This means the price doesn’t reflect expectations and is disconnected from the market. While I would prefer free banking, one of the reasons I support you and Bill’s convertibility and expected NGDP targeting ideas because they add expectation information back into the price of credit/money.
I’ll expand further on this some other time on my blog.
8. November 2009 at 14:25
Scott:
I want a monetary policy stance that leads to on target expected growth in PY. I don’t much care what happens to M and V individually.
OK, as you say Scott. You’ve done your part, your pointed to the problem.
The rest is Fed’s concern. Except that… if they ask me if they should sacrifice virgins or economists in order to bust NGDP, I’d say, definitely economists.
StatsGuy, you’ve raised some very good questions. Exactly the ones which people which swear by MV (NGDP) seem to forget to consider.
Scott, and Bill Woolsey is exactly right when he says that an increase in expected inflation by 3% will not get the people to spend now. They will most probably just sigh and accept that small dent in their future purchasing power.
8. November 2009 at 16:14
ssumner:
“commodity prices rise due to more demand… …In fact commodity prices wouldn’t rise in the first place unless markets expected the AD increase to lead to more demand for those commodities, and that almost surely means more GDP.”
This is the part I would challenge in the case of goods that act like assets (e.g. oil). It’s quite possible for the dollar-denominated price of oil to rise almost instantly in response to expected inflation _without_ a shift in the domestic demand curve or even an expected shift in the demand curve. That’s because oil is traded internationally. In that sense, it functions like an asset (as well as a production/consumption good), and its asset-like characteristics cause its price to move much faster in response to a credible inflation promise than “sticky” prices of things like consumer goods.
Having said that, a nominal rise in asset prices (and devaluation of debt) increases net wealth, which is likely to increase AD and expected AD – so these dynamics reinforce each other.
The part I take issue with is the notion that the primary mechanism by which expected inflation increases AD is because people want to spend their money faster because they fear it will depreciate in value. By the time most people realize inflation is actually going to occur, many product prices will have already adjusted.
Our economy is increasingly an “ownership” economy – one dominated more by wealth than labor-income (as reflected in the changes in income distribution over the last 25 years). As such, the wealth transmission mechanism has probably increased in importance during this time (as compared to the sticky prices/rush to buy transmission mechanism).
And yes – it’s precisely your post on asset prices as a transmission mechanism that caused me to observe this deviation between the standard story and reality.
9. November 2009 at 15:03
Doc Merlin, Thanks for clarifying that. I see what you mean.
Vasile, You said;
“The rest is Fed’s concern. Except that… if they ask me if they should sacrifice virgins or economists in order to bust NGDP, I’d say, definitely economists.”
Best of all would be virgin economists, we’re a pretty nerdy group.
Regarding inflation of 3%, people may not spend, but you can’t have 3% inflation over the next 12 months without much faster RGDP growth in my view. Somebody has to be spending or you won’t get 3% inflation.
Statsguy, I think you need to rethink your commodity argument. Suppose Americans don’t drive one more mile, and suppose oil prices only rise becasue the dollar is weak. Domestic oil producers are still almost price takers. So there is much more production from marginal fields, and exploration of the Gulf at $80 oil than $40 oil, regardless of whether we consume more. It’s a world market, so for price-taking US oil producers a higher price is effectively a shift upward in the perfectly elastic demand curve they face.
At the world level something has to be driving oil prices up. Even if it is just speculators putting oil into large containers, that’s still demand.
You said;
“The part I take issue with is the notion that the primary mechanism by which expected inflation increases AD is because people want to spend their money faster because they fear it will depreciate in value. By the time most people realize inflation is actually going to occur, many product prices will have already adjusted.”
We had this debate months ago vis-a-vis Keynes’ refutation of the Fisher effect. Steady state inflation does have a Fisher effect, and does reduce money demand because money is expected to lose value. A one time money supply increase is more complicated, as you suggest. Flexible prices rise right away, so I agree with you there. Sticky prices adjust slowly (that’s why a one time money supply increase causes expected inflation, not just a jump in the price level.) But I agree that beating the price increase on sticky prices isn’t a big motivation to spend money. If I suggested otherwise I erred. Sometimes I use the Fisher effect as a counterfactual to the liquidity trap. Either one of the two happens:
1. Prices jump immediately, in which case there is no expected inflation effect, but the higher prices boost NGDP.
2. Prices are all sticky, in which case expected inflation rises, real interest rates fall and AD rises. Either way AD rises, it doesn’t matter which assumption you make.
I agree with your last point about the importance of asset prices.
10. November 2009 at 14:16
Scott: Best of all would be virgin economists, we’re a pretty nerdy group.
No way. Those virgin economists can hardly offer a significant entertaining/emotional value. Most probably, already the second weekly sacrifice of virgin/geek economists will not even make it to the evening news. No, sir, if you want to make a splash you need to pick up bigger fish, say Nobel laureates. Wait, wait, will you ask but what if we run out of those, but the depression still goes this course? Well, as it is already a tradition in economics an idea is never wrong, it is just never tried hard enough.. So famous econobloggers will have to be used instead. Will you be a famous blogger by then, Scott?
Somebody has to be spending or you won’t get 3% inflation.
In this year, or in the next year? Well, the next year those bastards will be spending who have better money balances as compared to rest of us. Why and how? Short story: bastards.
And if the worst came to the worst, Fed will. After all he needs that bit of credibility, what a laughing stock is Central Bank who cannot increase prices!! Of course, it will have to dump all those consumers goods later into the ocean, but why should such a consideration stop a noble institution who’s trying to bring prosperity and inflation to an ailing economy.
On a more serious note, you scared me Scott. Maybe I’ve misread you, but my impression is behind all those talks about tinkering with NGDP is the old Keynesian desire to make customers spend now, whatever it costs. It is the “whatever it costs” bit that scared me.
Rightly or wrongly, but I call such views meta-socialism. The idea that regulators can and should regulate (tinker) not with individual prices but with aggregates. Because they know better? How? Where from?
10. November 2009 at 14:25
Yet some words about your easy/hard money distinction.
Suppose a shop has a sale and nobody came. Did the shop have a sale?
Obviously, Fed is now in an extended period of “loan sales”. Why do you insist in calling it a period of tough money is beyond me. Fed is all-powerful regarding the supply/price of money/loans, but for best or worst the demand for it is still in private hands and minds.
You don’t need a hundred cowboys to bring a horse to the water..
10. November 2009 at 14:54
@Vasile:
“Suppose a shop has a sale and nobody came. Did the shop have a sale?”
Wow thats great.
11. November 2009 at 08:55
Vasile, I want stable money, and a stable macroeconomy. Socialism is what happens when right-wingers get in charge, create tight money, drive NGDP down, and create a lot of unemployment. In desperation the public then turns to the left. When money is stable (1990s) left-wing governments are not able to implement any of their big government plans, because the public doesn’t want to rock the boat.
11. November 2009 at 12:02
I love that final response. Very well put.
11. November 2009 at 13:49
Scott, I’m not questioning your goals. Just your means. And between them, mostly, the one of getting the consumers spend again, whatever it costs. Yes, you don’t call it that, and you don’t thinks it is socialism of sorts. Now, that being said, I guess, I’m going to stop pestering you.
@Doc Merlin: Glad you liked it.
12. November 2009 at 07:11
Thanks D. Watson.
Vasile, I prefer ot think of it as getting producers to produce again. I actually think the term AD is very misleading, as it doesn’t necessarily mean consumers spend more (although in practice they probably will as we become more prosperous.)
5. February 2011 at 08:18
[…] year; Krugman argued that talk about NGDP would lead to dangerous monetarist ideas. At the time I argued that he was being illogical. Now I see that the field of macro can’t be understood in […]
9. February 2011 at 22:31
[…] quasi-マãƒã‚¿ãƒªã‚¹ãƒˆãŒå‹åˆ©ã™ã‚‹ãªã‚‰ã°ã€å‹åˆ©ã—ãŸã“ã¨ãŒã©ã†ã‚„ã£ã¦åˆ†ã‹ã‚‹ã ã‚ã†ï¼Ÿãã‚Œã¯çš†ãŒã“ã®è¨€èªžã‚’使ã„始ã‚ãŸã¨ãã 。マット・イグレシアスやブラッド・デãƒãƒ³ã‚°ã¨ã„ã£ãŸãƒ–ãƒã‚¬ãƒ¼ãŸã¡ãŒNGDPã®ã“ã¨ã‚’書ãã»ã©ã«ç§ã¯å¬‰ã—ããªã‚‹ã€‚彼らãŒã‚±ã‚¤ãƒ³ã‚¸ã‚¢ãƒ³ã®ã¾ã¾ã§ã‚‚構ã‚ãªã„””NGDPã®è©±ã‚’ã™ã‚‹ã¨ã€ã‚†ã£ãã‚Šã¨çŸ¥ã‚‰ã¬é–“ã«quasi-マãƒã‚¿ãƒªã‚¹ãƒˆã®è€ƒãˆæ–¹ã«ãªã£ã¦ã„ã。去年クルーグマンãŒãƒ‡ã‚¤ãƒ´ã‚£ãƒƒãƒˆãƒ»ãƒ™ãƒƒã‚¯ãƒ¯ãƒ¼ã‚¹ã‚’批判ã—ãŸã®ã¯ãれゆãˆã®ã“ã¨ã 。NGDPã«ã¤ã„ã¦èªžã‚‹ã¨å±é™ºãªãƒžãƒã‚¿ãƒªã‚¹ãƒˆæ€æƒ³ã«è¡Œãã¤ãã¨ã‚¯ãƒ«ãƒ¼ã‚°ãƒžãƒ³ã¯è«–ã˜ãŸã€‚ç§ã¯ã‚ã®æ™‚彼をéžè«–ç†çš„ã ã¨æŒ‡æ‘˜ã—ãŸã€‚マクãƒã®åˆ†é‡Žã§ã¯åŽ³æ ¼ã«è«–ç†çš„ãªè¨€è‘‰ã§ã¯ç†è§£ã•ã‚Œãªã„ã®ã ã¨ä»Šã¯åˆ†ã‹ã£ã¦ã„る。 […]