Krugman on credibility and expectations
We market monetarists have frequently argued that the ECB rate increases that took place last spring were a turning point. Now Paul Krugman comments:
Note the peak in April 2011. It wasn’t very high; slightly above the ECB’s target, but arguably still too low to make the needed adjustment within the euro area feasible. Nonetheless, the ECB raised rates “” and that was when the euro really began falling apart. The direct effects of the rate increase can’t explain that unraveling, but the effect on expectations “” aha, so they really are that fanatical about price stability! “” can.
This is a very good post. Note that the direct effect of the change in interest rates is small, it’s mostly about signals of the future path of monetary policy. That explains why rates have fallen over the past year, even as money has gotten tighter. The expectations of weaker NGDP growth have reduced interest rates, and those expectations come from too much policy credibility–the markets trust the ECB’s resolve to destroy the eurozone economy keep inflation below 2%.
In this post Krugman’s policy views are actually very close to market monetarism. If interest rates matter, it is mostly as a signal. I suppose the main difference is that he thinks moves in the expansionary direction would be less credible than we do. But we can both agree that “refraining from insane contractionary moves” is a highly effective policy, at least relative to the alternative!
HT: Steve
Update: Tim Duy also sees that there’s more to policy than low rates. Here he responds to Eric Rosengren’s call for lower long term rates:
I guess Rosengren is not seeing the same Bloomberg numbers that the rest of us are getting, but long-term interest rates are already falling rapidly. The ten year rate is holding behold 1.5% as I write. The Fed needs to give up its fetish with low interest rates. Is Rosengren completely ignorant of the situation in Japan? At least I can argue that the Bank of Japan was navigating uncharted territory, which is how they wound up where there are. Sadly, the Federal Reserve has no such excuse. Their leader, Federal Reserve Chairman Ben Bernanke knows what to do, but just won’t do it.
Exactly. (And Rosengren’s one of the good guys.)
Tags:
2. June 2012 at 12:43
Your point that low interest rates are indicative that policy has been too tight I always find a subtle one. Not that I vehemently disagree-far from it I’m not nearly sure enough about this to have a confident pronoucement.
I mean when the economy is slipping-NGDP is tanking-conventional monetary policy has been to lower interest rates in the belief this will spur investment. I get the idea of expectations.
What is true is that when times are good interest rates are higher-but then of course the Fed wants them higher.
I also see how China during it’s boom years had very high interest rates-the Chinese people put a high premium on savings.
Then again, of course high interest rates in a way give a high incentive to save-this is why the Chinese liked their high rates.
There’s very little incentive to save right now other than fear of any other asset alternative. Right now you pay the bank or treasury to hold on to your money.
The bottom line is there are so many ways to look at it that I haven’t really properly coneptualizes what happens the casuality behind your point. This point I think is easy to not grasp at least for me as it seems kind of chikcen or egg-isn’t it rates low because money has been tight so the Fed lowers it to make it easier?
So are low interest rates a sign of tight money or a kind of reaction to tight money?
2. June 2012 at 12:53
“In this post Krugman’s policy views are actually very close to market monetarism.”
Well, or market monetarism is close to what New Keynesians have been saying for a long time and close to what Keynes said originally.
2. June 2012 at 13:07
Mike, Read the response to Kimball. You need to get away from interest rates as a causal mechanism. Interest rates don’t do anything. They are a price, a rate of exchange. Never reason from a price change. If someone tells you, “lower rates will stimulate investment”, think about what would actually be happening if that were true. The rate is just a ratio of exchange. Investment is a flow of expenditure which leads to the accumulation of a stock (capital). If it is said that spending is increasing, you need to think about what spending is (the sum of all monetary flows) and whether that sum is actually increasing. Maybe there is an inflow of new money in the economy, which is borrowed by firms and spent on capital goods, adding to total nominal expenditures. Or maybe expectations of the long run path of spending are still tight, and velocity falls to offset the injection of money. This means that this new flow of spending is cancelled out by a reduced flow somewhere else. Or perhaps the Fed didn’t “do” anything at all – perhaps the fundamentals simply changed, the economy’s natural balance between savings and investment, supply and demand for loanable funds aside from net monetary inflows. Then the Fed is simply “chasing down the natural rate”. More precisely, think of a downward shift of the IS curve due to reduced demand for investment (perhaps because the markets know the Fed is not interested in doing what it takes to sustain long term nominal spending at an adequate level).
Low interest rates generally are a sign of tight money. But really you shouldn’t be using them as an indicator at all, as you simply can’t always tell whether they are moving due to the liquidity effect, the fisher effect, or the income-and-price level effect, or some exogenous shift in the social discount rate. Instead, in the absence of a Sumnerian futures market, the TIPS spread is the best way of seeing whether money is tight or easy – as long as longer TIPS show <2% the Fed is effectively tightening.
Again, it might help to think in terms of what Nick Rowe calls "machine language". Contrary to widespread opinion, the Fed doesn't actually lower interest rates at all. What it does (always) is buy and sell assets. When it buys assets, new money enters the economy. This puts downward pressure on interest rates, initially, as the supply of loanable funds shifts right. But whether the interest rate actually does fall, and when, and for how long, depends on a host of things. Don't believe the simple Keynesian models, or the even simpler mindless popular recital. Interest rates don't do anything – flows of money do. And the level of nominal spending is determined by the path of nominal spending. And the path is determined by supply and demand for money in the long run. The Fed buys assets, and the rest depends on expectations. Insofar as the Fed talks about targeting rates, that's just a communications tool for those expectations.
"I also see how China during it’s boom years had very high interest rates-the Chinese people put a high premium on savings.
Then again, of course high interest rates in a way give a high incentive to save-this is why the Chinese liked their high rates."
Never reason from a price change. Rates can be high because investment is high (a boom) or because saving is low (capital-constrained economy, as in autarky). Of course high rates give an incentive to save, just as high apple prices give an incentive to sell apples. But can you tell how well apple production is doing from the price? No.
Again, you could either learn an expectations-augmented ISLM model – or just listen to Scott when he says that the target is everything, and the rest is endogenous, so don't worry your head about it. If the market doesn't expect recovery, then policy hasn't been set to produce recovery, and never mind what rates are, or how much money has been printed, or where John Taylor thinks rates should be.
2. June 2012 at 13:11
Daniel, Keynesians don’t understand that money is special, they don’t even model it. They obsess about interest rates, which are highly misleading. Keynes lived in a commodity currency world, he didn’t even believe in the Fisher effect. His whole doctrine rested on monetary impotence. And he didn’t have anything to say about what determined NGDP in the long run, and didn’t understand rational expectations, which says that what determines NGDP in the long run determines it in the short run as well.
2. June 2012 at 13:15
See this: http://www.themoneyillusion.com/?p=220 and this: http://worthwhile.typepad.com/worthwhile_canadian_initi/2011/09/whats-wrong-with-new-keynesian-macroeconomics-from-an-mmqm-perspective.html
Amongst many other posts by the two of them.
2. June 2012 at 13:43
Saturos –
“Keynes lived in a commodity currency world”
To be fair to Keynes, he opposed the gold standard. But your point is an imporatant one. When monetary policy is made impotent, fiscal policy is the only thing left. If NDGP drops and you rule out increasing the money supply to raise NGDP, then the government can attempt to increase velocity by borrowing and spending. I recall Friedman saying the reason Keynes favored fiscal policy was because he didn’t trust central bankers to do the right thing.
Even with a gold standard, the Fed can respond to shocks by raisng or lowering the multiple of money to gold.
Even before the Fed, Congress would sometimes issue legal tender US Notes during a panic to increase the money supply. They’d just cover some or all of the deficit that naturally occured during a downturn. I’m not saying we should go back to that system, just that people have always known about using monetary policy to prevent depressions, and I’m sure Keynes knew that.
2. June 2012 at 14:45
This is a very good post.
???
In this post Krugman’s policy views are actually very close to market monetarism.
Riiiight.
If interest rates matter, it is mostly as a signal.
…of time preference, which inflation into the banking system distorts, thus leading investors to making errors.
2. June 2012 at 14:51
Ineptitude is seriously underrated as a reason for this depression. What are these people thinking?!
PS Who’s pretending poorly to be Major Freedom? That post was only eight lines long, not at all believable.
2. June 2012 at 15:20
Left Outside,
It was me, I’m just really depressed today.
2. June 2012 at 18:06
Why wasn’t Krugman calling for monetary stimulus? One answer is that rates were already at zero, and hence could not be cut any lower. But there are lots of other ways to stimulate the economy.
Like what? The fed deals almost exclusively with lenders. Monetary policy in our economy consists primarily of the Fed’s dependence on the banks to expand credit. That is where Fed policy is primarily channeled into aggregate spending. They don’t print money to purchase everything. They print primarily for the banks, and the banks are expected to expand credit as a result.
Krugman is only half right. Yes, the Fed does become essentially powerless in a recession when general profitability is zero or negative, if banks refuse to lend into losing projects, and not because of any barrier of the lower bound in the fed funds rate. And you are only half right. Yes, the Fed can in principle give the banks as much money as they want, despite the fed funds rate being close to zero, but that doesn’t mean they can bring about any aggregate demand they want in this way. Not if the banks refuse to lend more.
If after a financial collapse the economy has negative profitability, and banks refuse to lend more than they are already lending, for fear of incurring losses, then the Fed goosing bank reserves cannot possibly coax banks into lending more. The banks will hoard as much money as the Fed prints that cannot be profitability lent until costs fall below current aggregate demand.
Only if the Fed starts buying things from non-lenders, such as consumer goods, or stocks, can the Fed have any direct influence in increasing aggregate demand. Depending on lenders is a no win situation if lending will incur losses at current prices.
Unlike Keynesians, monetarists don’t believe the short term rate is the key mechanism for the transmission of monetary policy. Rather we look at things from a supply and demand for money perspective.
A sustained, general increase in prices requires inflation of the money supply, but inflation of the money supply does not necessarily imply prices will rise. Not if the demand for money rises, as people await a sufficient fall in costs that will restore profitability, before expanding credit becomes attractive again. If NGDP falls during this process, then short of the Fed printing money to buy output, the Fed will have to accept a fall in NGDP because that’s what depending on banks can result in.
Would you have the Fed buy cars and computers if that’s what it takes to increase NGDP, because banks are not lending enough to make NGDP rise via credit expansion? I’ve asked this before, but never got a satisfactory answer. Asked another way, how far will you go into “unconventionalism” when it comes to the Fed targeting NGDP? Suppose credit is insufficient. What should the Fed buy, and how does unearned money being printed and spent, constitute a gain for those who produce for the sake of those who spend without producing for that money? We’re talking trillions of dollars here.
3. June 2012 at 00:20
I think that everybody (who’s interested in the subject ;)) should see first chart from the article
http://www.economonitor.com/blog/2012/05/does-history-support-ngdp-targeting-now/
and see that with one’s own eyes how NGDP (US NGDP) is volatile.
And what NGDP targeting framework would do to fix that.
3. June 2012 at 02:41
Daniel Kuehn,
The funny thing is that (as Greg Mankiw set out very well in an article 22 years ago!) New Keynesianism could have been called “New Monetarism” and no-one would have batted an eyelid. Much of New Keynesianism is just monetarism with numerous epicycles to keep money causally irrelevant e.g. Woodford’s work.
So Market Monetarism, New Keynesianism, Old Monetarism, and the Economics of Keynes are a big bunch of brothers. While no-one fights quite like brothers, the underlying theories of recessions and how the economy works are all close. That’s why both Market Monetarists and the smarter New Keynesians have such similar views on what policy should be right now.
It’s also why Friedman had to invent a “Chicago Oral Tradition” – it would have been embarassing to admit that Old Monetarism was just Keynes* plus the neutrality of money and a more realistic theory of liquidity preference.
* I need to re-read it, but the way Keynes sets out the liquidity trap in the GT assumes that broad money can be an exogenous variable. That’s one difference from Keynes and modern Keynesians, who tend to frame the liquidity trap as being a situation where the CB can’t change interest rates in order to change monetary aggregates, as opposed to Keynes’s trap of being unable to change monetary aggregates to change (long-term) interest rates.
3. June 2012 at 04:31
W. Peden, Old Monetarism was a lot more than that – see Clower and Yeager. As Scott points out Keynes had nothing to say about what determines NGDP. Keynesians are Wicksellians, they reason from price changes, whereas Monetarism always recognizes money as special (see how Milton differed from the Keynesians on Japan). Really, the tradition we belong to has more in common with Hawtrey and Cassel, and all the pre-Keynesians that were ignored once the General Theory came out- including the old Hayek. In fact our lineage goes straight back to Hume – the real “inventor of macroeconomics”.
3. June 2012 at 04:32
Who’s that pretending to be Morgan, you’re doing a terrible job.
3. June 2012 at 04:49
Saturos,
I distinguish between Old Keynesianism and the Economics of Keynes. Money is very special in the GT: so special that Keynes models the economy at one point with nothing other than money and bonds, which was the source of his confusion regarding the liquidity trap.
I’m not sure about the determination of NGDP- I’ll defer to those better read in Keynes’s economics.
3. June 2012 at 05:18
W. Peden, read this: http://worthwhile.typepad.com/worthwhile_canadian_initi/2011/04/walras-law-vs-monetary-disequilibrium-theory.html
As you can see, Daniel Kuehn is already familiar with this. So he should know better.
Then again, I see a gap between Scott and Nick. Market Monetarism is a broad church, Nick is the more “monetarist” monetarist. He understands that general gluts are all about the medium of exchange. So perhaps Keynes can be a Market Monetarist without being a monetarist.
3. June 2012 at 05:28
While I’m at it, Scott, I posted this before, I don’t know if you saw it then. Could you comment briefly on whether you agree with what Nick said to me here http://worthwhile.typepad.com/worthwhile_canadian_initi/2012/06/counterfactuals.html#more ?
I see some similarity. On the one hand, you’ve always said that the liquidity trap doesn’t boil down to one thing, that a number of things need to be tried. On the other hand, you’ve referred to fiscal policy as a “fifth wheel”. So I wasn’t certain you’d agree with Nick.
And again, I thought the equity premium post was really interesting.
3. June 2012 at 06:09
Mike Sax, You said;
“What is true is that when times are good interest rates are higher-but then of course the Fed wants them higher.”
How do you explain the fact that interest rates tended to be higher in good times even before the Fed existed?
Daniel, The view that monetary policy works through changes in interest rates is one of the defining characteristics of Keynesianism (new and old.) You are right, however, that Krugman focused on the importance of expectations back in the late 1990s. But I rarely see that work show up in his blog posts.
Negation, I mostly agree, although it would be more accurate to say Keynes favored a flexible gold standard, something like Bretton Woods. He strongly opposed fiat money.
Left Outside, Yes, it can’t be MF, the comment is too short.
Morgan, No, banks don’t play an important role in our monetary system (prior to 2008). The Fed traditionally injects base money by swapping cash for T-bonds. That need not involve banks at all.
Wadolowski, That post does have a good graph. But there is a serious error. He said Bernanke called NGDP targeting “reckless”. That’s false, nor does Bernanke believe it’s reckless.
W. Peden, That’s right, DeLong made a similar observation.
Saturos, Yes, that’s weird too–doesn’t seem anything like Morgan.
Nick is both more monetarist and more Keynesian than me (only slightly), I’m more in the Fisher/New Monetary Economics church.
A lot could be said regarding Nick’s suggestion that fiscal stimulus be a back-up plan–I’ll do a post on that.
3. June 2012 at 09:24
Actually Morgan sounds like the Major in that one comment. Some of that I seem eto remember Major saying word for word
3. June 2012 at 09:28
And Mrogan never talks about Krugman only “DeKrugman”
3. June 2012 at 09:32
maybe i am starting to think Morgan and Major are the same person, some odd bipolar self debate.
3. June 2012 at 09:46
Mike and dwb:
I can assure you that Morgan and I are different people. The reason why you think we’re the same is because I have successfully incepted you, and so now every little off chance detail of similar words, tone, argument, pretty much what happens all the time with everyone else to some degree but you just don’t notice it, sends your minds off on playing detective, as if something funny is going on and you just can’t figure it out.
sumner:
Morgan, No, banks don’t play an important role in our monetary system (prior to 2008). The Fed traditionally injects base money by swapping cash for T-bonds. That need not involve banks at all.
It does involve banks and banks do play an important role, prior to 2008 and today in 2012. You’re talking about what ifs, but you first have to accept what is. Banks DO play an important part in establishing aggregate spending. Their credit expansion creates a very large component of the total money supply, and hence total spending.
Yes, in principle the Fed COULD purchase t-bonds from non-banks, but today in 2012 they deal almost exclusively, officially, with the primary dealer banks. So you’re wrong to claim that a potential what if somehow serves to refute an argument over what is. When the Fed buys T-bonds, they buy T-bonds almsot exclusively from the primary dealer banks.
They don’t buy T-bonds from you, from me, or from hedge funds. They buy them from banks. And the money the banks receive from the Fed are used directly and indirectly by the banks when they expand credit, and thus expand the money supply, and thus expend aggregate spending.
Of course, this doesn’t include the NY Fed secretly (at the time) sending over $40 billion to Iraq 2003-2008, to fund the war, and that’s just what we found out.
3. June 2012 at 09:58
I just read what the obviously fake “Morgan Warstler” wrote (notice how there is no picture of Morgan in that quote), and yes, that looks almost identical to what I wrote a while back. It’s probably a copy paste job.
I think it is the same person who used to copy and paste what I say, and I think back then he used the nick “Major Freedum” or something like that. My guess is that it is the same prankster. This prankster probably WANTS us to talk like this, so my recommendation is to ignore “Morgan Warstler” that doesn’t have the picture icon (hint: even if the picture icon is there, you can verify it by right clicking and opening the pic in a new tab, and then look at the avatar code. They’re all unique).
Maybe I will put in a picture icon of my own so that you know it’s me who is annoying you, and not some faker.
3. June 2012 at 09:59
Aaaaaaand he’s back.
3. June 2012 at 10:04
Aaaaaaand he’s back.
well, that was a lot less than the usual 14000 words, and also, it was not in the typical style where every sentence is put in italics and followed by a rebuttal. call me skeptical.
3. June 2012 at 10:30
Here’s empirical data consistent with my theory that central banks have to accelerate the pace of money printing in order to prevent an unsustainable economy – that was itself caused by money printing – from correcting (i.e. from going into deflationary depression):
http://i.imgur.com/zvzDO.jpg
3. June 2012 at 10:32
Saturos and dwb:
What does that even mean? Aaaaand I’m back? You two write posts, then leave, then come back. It’s like you’re saying there is something special about me doing the same thing you’re doing.
Or are you believing yourselves to be permanent residents of this place, tallying visitor times? That’s rather weird.
4. June 2012 at 07:55
MF, Since it obviously makes no difference whether the Fed buys from primary dealers who are banks, vs. primary dealers that are not banks, the role of banking per se is negligible.
7. June 2012 at 11:45
ssumner:
MF, Since it obviously makes no difference whether the Fed buys from primary dealers who are banks, vs. primary dealers that are not banks, the role of banking per se is negligible.
In principle, yes, but in practice?
I will agree that the line between commercial banking and investment banking has been blurred since the abandonment of Glass-Steagall, and I think that’s part of the reason why a given quantity of Fed inflation doesn’t pack the same punch as it used to when it comes to bringing about the business cycle. I think it’s also why the shadow banking system has grown so much larger with no corresponding increase in the real economy. Prior to around the 1990s, the way money from Federal Reserve affected the general spending stream was by way of bank lending. But not so much any more. Now, the Fed buys t-bills from investment banks that also carry commercial banking services. So money can now effect stock, bond and option prices directly.
So sure, the Fed doesn’t have to depend on the banks to affect aggregate demand as much anymore, but to a large extent they still do, and so when you say the fed should ease, you’re still calling for more bank lending to boost NGDP, which just brings about the business cycle. If NGDPLT targeting were adopted, then I argue there will be a revolution in the structure of finance, and the Fed will find itself having to inflate exponentially more and more to keep NGDP on a constant growth path, the same way they had to do so to target a constant price growth.
Does the Lucas critique (Goodhart’s Law) not apply to NGDP targeting?
26. December 2012 at 10:01
[…] monetary policy works primarily through shaping expectations, it makes sense to consider the QE2 hints at Jackson Hole as the functional start date for QE2. . . […]