Cullen on Krugman on getting the Keynesian "liquidity trap" wrong. For whatever reason, Brad has decide to put Post Keynesian analysis in play. Another step in the last mile.
Grasping Reality
Must-Read: Cullen Roche: Does a “Liquidity Trap” Ever End?: "Brad Delong has a very smart post over at Equitable Growth
Brad DeLong | Professor of Economics, UCAL Berkeley
Also
Must-Read: The hawk-eyed Mark Thoma picks up and develops a very important point: Mark Thoma: Krugman vs. DeLong: "...
Is the tide turning?
24 comments:
It's obvious from DeLong's article that he has no idea what adaptive expectations is all about. If anything, Krugman is applying a forward-looking expectation model when he argues that the central bank can hit any inflation rate it wants just by announcing its target (and being...err..credible; how it's supposed to achieve this credibility is never explained)
Discussions on macro have become truly pointless. Putting Tobin and Feldstein together in the same group? What's that all about?
Krugman was being polite in his rejoinder; he knows DeLong is way out on left field here.
Actually, I posted it because I thought it interesting that he cited Cullen as a must-read. Otherwise, I would have posted Cullen's piece directly instead of by way of Brad.
Nothing against you or the choice of article you posted. I just quickly commented on the discussion. The bottom line is that neo-keynesians (yes, the neoclassical synthesis guys) knew not to put an expiry date on depressions. This talk of a 1-year short-run, 2-year medium run and 5-year long-run is total nonsense. Old keynesians understood quite well that it can be a very, very long time before an economy recovers from a deep downturn.
Hi circuit
I also think this distinction DeLong makes is kind of strange. But I'd like to ask you a question, not directly related.
What do you make of the distinction Cullen Roche makes between Keynes' liquidity trap and Krugman's liquidity trap:
http://www.pragcap.com/would-keynes-have-called-this-a-liquidity-trap
Maybe it's really subtle, but -- frankly -- I have difficulty seeing the difference.
Hi Magpie,
I saw this but must run to work so will check out and respond later.
BTW, I enjoy reading your blog.
Magpie,
Not an economist here, but let me say what I know.
Keynes defined liquidity trap as a situation in which the CB cannot lower interest rates b/c they are already so low that everybody only expects them to go up, so bond prices are not attractive.
Krugman ignores that and made up his own definition: "monetary policy loses traction" because you can't lower rates much anymore (they are so low).
So, Keynes:"you can't lower rates to zero"
Krugman:"you can lower rates to zero, but then what?"
http://www.nakedcapitalism.com/2014/07/philip-pilkington-paul-krugman-understand-liquidity-trap.html
The point that Cullen makes is that Krugman assumes a Wicksellian natural rate (and general equilibrium) while Keynes agreed against this in the GT. This distinguishes old Keynesianism and Post Keynesianism from most contemporary people in the mainstream calling themselves Keynesians. And the Hicksian ISLM model is based on loanable funds, which Krugman also assumes.
So those folks are all monetarists except some admit the suitability of fiscal intervention in a liquidity trap when the policy rate cannot be set lower than zero to approximate the supposed natural rate that would bring the economy back into equilibrium using all resources efficiently, including labor, based on market forces. This implies no need for government intrusion into markets, which are best left alone to adjust iaw "natural" processes that are therefore assumed to be the proper study of economics, although the exception for some would be the so-called liquidity trap where monetary policy is no longer operative owning to the no negative rate constraint. Some would even like to banish cash so that the policy rate could be set negative in an effective way that could not be avoided.
MMT economists have written on this but mainstream economists haven't picked up on it. I found it interesting that DeLong picked up on Cullen's post about it now, while having ignored the MMT position for years. Not that MMT economists are alone in this.
LK, Post Keynesians Reject the Liquidity Trapl
And as circuit points out above, old Keynesians did, too.
Randy Wray, Reconciling the Liquidity Trap With MMT: Can DeLong and Krugman Do the Full Monty With Deficit Owls?
Bill Mitchell, Whether there is a liquidity trap or not is irrelevant
Bill Mitchell, The on-going crisis has nothing to do with a supposed liquidity trap
Bill also discusses the S-LM frameworkI in several posts
I wonder if BDL is creating an opening here?
Well, according to Richard Werner (widely cited in the recent BoE paper on banks as "not intermediaries of loanable funds") it's not easy to pin down Keynes' position on money:
The literature review has identified a gradual progression of views from the credit creation theory to the fractional reserve theory to the present-day ubiquitous financial intermediation theory. The development has not been entirely smooth; several influential writers have either changed their views (on occasion several times) or have shifted between the theories. Keynes, as an influential economist, did little to enhance clarity in this debate, as it is possible to cite him in support of each of the three hypotheses,through which he seems to have moved sequentially
Anyway, in the GT Keynes - as quoted by Werner - was unambiguous:
… No one can be compelled to own the additional money corresponding to the new bank-credit, unless he deliberately prefers to hold more money rather than some other form of wealth.…Thus the old-fashioned view that saving always involves investment, though incomplete and misleading, is formally sounder than the newfangled viewthat there can be saving without investment or investment without ‘genuine’ saving.
Perhaps it's time to move on, forget about "what Keynes really meant" and concentrate instead on "how does the system really work".
Just like the recent BoE paper did.
I forgot to mention that the (very good) paper by Richard Werner on banks as creators of money can be found here:
http://www.sciencedirect.com/science/article/pii/S1057521914001070
Thanks, y'all! -- as Southern Americans say
@Jose G. Thanks for the link! Food for thought.
@Tom
"So those folks are all monetarists except some admit the suitability of fiscal intervention in a liquidity trap when the policy rate cannot be set lower than zero to approximate the supposed natural rate that would bring the economy back into equilibrium using all resources efficiently, including labor, based on market forces."
This may be a general conclusion, but it was not what happened after the housing crash in the US and in Australia. Both, in the US and in Australia, fiscal and monetary stimulus (with all the political opposition and with the limitations imposed by it) took place at the same time: both CBs lowered interest rates together with increased fiscal spending by the respective Federal Governments. In Australia, in contrast with the US, monetary stimulus was partially removed before fiscal stimulus (in fact, in my opinion, the mere increase in interest rates, limited and early as it was, signalled to our local Austerians that they would be justified into renewing their whingeing and carping against deficits and asking for their holy SORE+ -- Aussie pronunciation of the word).
-------------
What I find interesting is that Krugman has been speaking of liquidity trap for years now (as Cullen Roche rightly observes and DeLong agrees) but neither Hicks nor Keynes ever considered their models to be long term models.
So, what gives? A permanent liquidity trap?
So, what gives? A permanent liquidity trap?
This is where MMT comes in. The so-called liquidity trap is created by the assumption that monetary policy is dominant and fiscal policy is basically impotent other than a a short term palliative. That assumptions wrong, so the approach is wrong and the conditions persists long after to the mystery of all other than those who reject that assumption based on operational analysis and causal reasoning. Asl explained in the links above.
Krugman has his own view of liquidity tarp based on his Hicksian IS-LM heuristic model. That allows him to see that once caught in the liquidity trap that results from ZIRP, fiscal is the only way out until monetary policy begins working again. Then fiscal can be dropped. In Krugman's view then Christina Romer got it right but by the time it got through Larry Summers and then the WH political operatives it was too small and didn't last long enough to get to the point of recovery that monetary policy would start working again.
The difference between Krugman and MMT is his assumption of monetarism is except in a liquidity trap based on his assuming a Wicksellian natural rate in a general equilibrium theory and loanable funds.
So Krugman's analysis is partially right but for the wrong reasons. MMT points out why this is so.
Magpie,
I like the way PeterP described it. I would just add that, in Keynes, the liquidity trap involves the long term bond yield. The so-called 'trap' stems from the uncertainty facing investors as they attempt to balance money holdings against investments in bonds. Given that bond prices and yields move inversely, investors who purchased bonds at a high price (a low bond yield) face the risk of capital losses if the price of bonds subsequently falls. In other words, it's the fear of capital losses that makes investors shy away from purchasing bonds at higher prices. And this could occur whether or not the central banks is injecting money in private hands via open market operations.
That said, I sometimes find the distinction between Keynes's original view and the popular Krugman view (zero lower bound and ineffectiveness of conventional monetary policy) to be somewhat of a distinction without a difference. Both are meant to imply the weakness in resorting to monetary policy under conditions of low interest rates and high unemployment.
I sometimes find the distinction between Keynes's original view and the popular Krugman view (zero lower bound and ineffectiveness of conventional monetary policy) to be somewhat of a distinction without a difference.
The difference from the PKE-MMT perspective is the reasoning behind the views. They are different explanations for the same phenomenon. Then the question becomes which explanation fits the case more closely both operationally and causally. Krugman concedes that his position and the MMT position look the same at ZIRP but he contends that the proof will be what happens when the economy recovers and monetary policy becomes effective again. MMT contests that on the grounds that Krugman misunderstands the underlying operations and also gets the causality wrong since it is based on a Wicksellian rate and loanable funds as the foundation for equilibrium. My understanding anyway.
But Krugman doesn't deny the effectiveness of fiscal policy when rates have left the zero bound area.
He'd likely say that - at that point - it would be much easier for the CB to lower rates to stimulate the economy than it would be for Congress to approve a fiscal stimulus with the same objective.
Tom,
Two points.
First, I think Krugman argues that recoveries are different (than in periods of ZIRP) because the CB's reaction function will act to push it to increase its short term rate, a move that will in most instances cancel out in part or entirely the effect of fiscal policy (and other components of aggregate demand). This, I believe, is a perfectly sensible characterization of the effect of current strategy. For instance, if the CB's reaction function could be tweaked to allow its short term rate to stay put until the economy reaches the point of full employment, Krugman would probably sound a lot like MMT.
Second, it's true that Krugman would disagree with the (strong form?) MMT view that interest rate policy (conventional monetary policy) is always ineffective and functional finance is the only appropriate approach. But, it seems to me, is due to a misunderstanding of the real effect of monetary policy. It's true business investment isn't sensitive to changes in monetary policy, but this is not the case for consumer spending and real estate 'investment'. Using the language of ISLM, it's well accepted that the IS curve slopes downwards once you include consumer and real estate spending.
Also, Krugman doesn't think monetary policy is entirely ineffective; rather he believes only the conventional part of it is ineffective because the nominal short term rate can't go any lower via conventional open market operations. We can debate whether additional QE (of other securities) or measures affecting the expectations channel work (which I'm skeptical about), but I don't see anything highly unusual with his view.
I meant to end the first point by saying Krugman's view on this is not because of loanable funds (in other words, it's not a crowding out issue)
He'd likely say that - at that point - it would be much easier for the CB to lower rates to stimulate the economy than it would be for Congress to approve a fiscal stimulus with the same objective.
Right. And that is something that MMT economists object to on political grounds as well as economic. This is political economy after all. Monetary policy is a technocratic solution that avoids the messy democratic process by making the central bank politically independent. It's anti-democratic. It's also a bad idea economically in that at ZIRP the political authority has to come in with fiscal policy to alleviate, but they have been taken off the hook. BB knew this, but he was reluctant to push Congress on it for whatever reason.
That may be true, circuit. Krugman is a smart fellow and he is the best of a bad bunch (Wicksellian monetarists). But as I see it the issue is about the preferred economic explanation. Are you saying that the explanations stand on the same level and preference is coin flip?
I think that the MMT economists would say that MMT is a superior explanation as a general theory that accounts for a body of data and changing conditions, whereas Krugman is working with a heuristic gadget that is not the same level of generality because the analysis operations and causality it is based on by his own admission is not correct. It is closer to neoclassical rather than Keynes, for one thing, and secondly, it ignores the contributions of PKE and MMT that clarify and extend the work of Keynes.
Ironically, Krugman put up a post today, Why I Am Keynesian.
I think Jan Kregel's critique of PK's Japan paper many years back is still one of the best at laying out the different views of the liquidity trap. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=219494
Also, there are no MMTers that think monetary policy is necessarily ineffective. We argue there are many effects, and the balance of those effects can be unclear. I would agree that the negative relation to rates on housing/consumption can often out pace all the other effects, though.
At the same time, if the Fed is only dealing directly with the overnight rate (as it normally tries to do), it's no guarantee that rates paid by actual people will adjust in a significant way--this has frequently occurred in the past. A quick look at the data shows the spread of mortgage and corporate bonds over the 10y Tsy follows the unemployment rate fairly closely, which imparts an effect that is the opposite of what the Fed is intending, for instance.
Also, there are no MMTers that think monetary policy is necessarily ineffective.
Thanks, Scott.
The way I understand it is that the MMT position is that monetary policy cannot be tightly targeted, as fiscal policy can. Moreover, monetary policy doesn't address demand leakage and the uncertainty about future demand that inhibits firm investment.
Monetary policy alone is like trying to build a house with only one tool. No single tool is versatile enough to handle the job. Why bother trying when other more appropriate tools are available.
The policy rate operates chiefly through the housing channel with some lag. As you say, there is no guarantee that lowering the policy rate will translate into a corresponding reduction in mortgage rates, or an increase in investment in housing. As a matter of fact, the post-crisis experience has been that even though the Fed reduced the policy rate to near zero and mortgages are priced at historical lows, housing still lags because 1) banks have significantly tightened credit, 2) incomes are lagging, and 3) government is not accommodating increased saving desire owing to deleveraging.
The result is both a lagging recovery and historically low housing investment, with a corresponding rise in rental demand and therefore a rise in the cost of renting.
Sober Look, Looming rental crisis in the United States
This is putting more and more people behind the eight ball.. Neither Increasing net exports or (inclusive disjunction) private sector borrowing seem to be in the cards to offset demand leakage, and without firm exceptions of rising demand, investment won't be forthcoming to offset either. That leaves government to step up fiscally. That's seemingly not in the cards either, given the political situation.
Therefore, the "new normal." Certainly, monetary policy should be favorable for construction borrowing and affordable mortgage rates. But the so-called recovery shows that that this is not enough. Moreover, the Fed is jawboning about tightening to get ahead of future inflation.
Even with the employment figures increasing, the structural quality is declining, with higher wage jobs being replaced by lower, underemployment high, participation rate historically low and incomes stagnant with labor share declining.
MMT explains this in terms of a general framework. Does Krugman's IS-LM gadget?
What about the 1980s recession, generally presented as solid evidence of the contractionary consequences of a rise in the federal funds rate?
What is MMT's understanding of this kind of evidence?
The Economist sums it up thus:
in July of 1980, Mr Volcker orchestrated a series of interest rate increases that took the federal funds target from around 10% to near 20%. What followed was an extraordinarily painful recession. Unemployment rose to near 11%. Manufacturing states were battered by the downturn; (...) Mortgage lenders were devastated by high interest rates. The banking system was pushed to the point of insolvency.
http://www.economist.com/blogs/freeexchange/2010/03/volcker_recession
STF: "...there are no MMTers that think monetary policy is necessarily ineffective. We argue there are many effects, and the balance of those effects can be unclear. I would agree that the negative relation to rates on housing/consumption can often out pace all the other effects, though"
Yes, thanks for clarifying this point, Scott: I would certainly agree with your statement. I did read one of Randy's paper (with a co-author) on this matter and found it well nuanced in its conclusion, as you also suggest.
Thanks also for reminding me of the Kregel paper, which also reminds me that, in addition to Krugman's belief in the merits of the New Keynesian Philips Curve, he also quick to give in to the quantity theory of money. Anyway, my critique of Krugman, I think, lays out the main challenge with the view he expresses in his Japan paper: http://fictionalbarking.blogspot.ca/2014/11/its-baaack-pauls-japan-paper-monetary.html
Tom: My views on Krugman are described in the post above. I don't feel his use of ISLM, loanable funds, natural rate of interest are the problem. If you ask, the main problem is his application of the New Keynesian Philips curve, which is completely devoid of realism.
I don't deny that, circuit. The point is that Krugman is trumpeting his use of IS-LM as a simple gadget that has allowed him alone to get it right when heavy-duty econometric models like DSGE failed. It seems pretty to me that Krugman is not just using IS-LM to the table but a lot of other unacknowledged assumptions that provide the context for his analysis. That's what Cullen was poking at, and DeLong thought in interesting enough to post as must-read. Of course, it is will known that Krugman doesn't recognize PKE and MMT success in this area because "no model." The reason that Krugman has had the degree of success he has had is due to being a smart guy rather than the extraordinary power he claims for IS-LM, which he attributed to Hicks even though Hicks himself backed away from it later on.
BTW, with no formal training in econ, I called an impending crisis about the same time as the FBI was warning of the obvious massive fraud. I was not an investigator either. I just happened to on the ground in an area where I was observing what was going on and was attuned to it because I knew some people that were availing themselves of the opportunity of cheap money with few questions asked for a sure thing in a hot market. Being some more circumspect I stayed on the sidelines since it was clear that the market could blow up at any moment. But it ran on for almost two years and some of the people I knew made a lot of money. A few got caught holding stuff they couldn't sell after the blow off, but that was offset by much greater gains. I also called the top within a few weeks when the market was still hot. No training, no analysis, other than some trading experience in my youth.
Admittedly I did not realize it would eventually bring down the big banks and nearly blow up the global economy, but it was pretty clearly a replication of the S&L crisis. Now that Bill Black and other have filled in the blanks and connected the dots, it's clear what happened. Economists are still not taking cognizance of this and trying to figure it out with their models, which don't include this aspect at all. they seem to be to be as foolish as Alan Greenspan, who blew off the FBI warning and attributed it to some "frothy" markets that would self-correct. He at least admitted later that he made a mistake. Economists, not so much. They are still scratching their heads and looking at their models since of looking at the now established facts.
Post a Comment