Invasion of the New Keynesian Mind Snatchers
Wars make people think and do stupid things. So does deflation caused by tight money. The intellectual low point of 20th century macro occurred in 1938, when a promising recovery in 1936 turned into renewed depression and deflation. Interest rates fell to zero, feeding the view that money was irrelevant. A 1938 EJ article by Joan Robinson provides a good example of the zeitgeist. Here she criticizes Bresciani-Turroni’s argument that the German hyperinflation was caused by their government printing too much money:
“An increase in the quantity of money no doubt has a tendency to raise prices, for it leads to a reduction in the rate of interest, which stimulates investment and discourages saving, and so leads to an increase in activity. But there is no evidence whatever that events in Germany followed this sequence.”
So easy money couldn’t possibly have caused the German hyperinflation because German interest rates were not very low. And everyone knows that easy money is associated with low interest rates. I won’t insult the intelligence of my readers by explaining what is wrong with her reasoning.
But perhaps we shouldn’t be too hard on poor Joan Robinson. Unlike some of the more wimpy Keynesians, she at least had the courage of her convictions. If interest rates are the right indicator of monetary policy; then doggone it money must have been really tight in Germany during the early 1920s. Let’s not be distracted by a few wheelbarrows full of cash.
Fortunately, after WWII economics gradually crawled out of this intellectual morass. Friedman and Schwartz explained how money was actually very tight in the 1930s, despite the low interest rates. By the 1980s the new Keynesians had acknowledged that interest rates were not a good indicator of the stance of monetary policy. The best selling money textbook (Mishkin, p. 606) informed students that:
“It is dangerous always to associate the easing or the tightening of monetary policy with a fall or a rise in short-term nominal interest rates.”
For a while it seemed all was well with macroeconomics. It was a golden age of inflation targeting; the Great Moderation of 1982-2007. And then Fed policy lost credibility and we slipped into deflation. Deflation always brings out the worst in economists. We started to hear that opportunity costs didn’t matter—there were free lunches out there waiting to be consumed. And suddenly interest rates again started to be good indicators of monetary policy. A few days ago I even saw an example in Krugman’s blog. He posted a graph showing that the fed funds rate is negatively correlated with unemployment (i.e. is procyclical.) Then he argued that we shouldn’t view the recent rise in unemployment as evidence fiscal stimulus doesn’t work, after all we don’t view the recent rise in unemployment has evidence that monetary stimulus doesn’t work. What was his evidence that monetary stimulus had been tried? Low interest rates!
Of course, we all understand that the correlation runs the way it does because the Fed cuts rates in an effort to fight recessions.
But here’s the thing: a lot of people are asserting that because unemployment has risen along with the budget deficit, fiscal expansion has failed or even made things worse. Why don’t we apply the same standards to monetary policy?
BTW, is it really true that interest rates fall in recessions because the Fed cuts them? Why did rates fall in recessions before there was a Fed?
It seems to me that Krugman is moving further and further away from the sort of new Keynesian research he did back in the 1990s. Consider the following progression from new Keynesianism to the crudest version of 1930s style Keynesianism:
1. In the late 1990s Krugman develops an explanation of the Japanese liquidity trap based on rational expectations theory. He suggests that fiscal stimulus leads to “bridges to nowhere,” and recommends that the BOJ adopt an explicit inflation target.
2. Then Krugman seems to have some doubts about whether monetary stimulus can be effective in Japan. He discusses how an expectations trap can form if a central bank has a conservative reputation and cannot convince the markets that a monetary stimulus will be permanent. Monetary policy is still highly effective, but only if expected to be permanent.
3. Then he starts leaving expectations out of his analysis. Now we are back in the 1930s where swapping cash for zero interest T-bills has no effect. “Period. End of story.”
4. And then a few days ago he started arguing that low interest rates indicate that the stance of monetary policy is expansionary. Now he is back in Joan Robinson’s world. Tight money couldn’t have caused the crash of 2008; after all, interest rates were low! Perhaps in a few weeks we can expect a graph showing the correlation between interest rates and inflation in the German hyperinflation, disproving the theory that easy money was to blame. And let’s not forget the high inflation of the 1970s; the cause couldn’t possibly have been easy money, weren’t interest rates really high in the 1970s?
This progression reminds me of the scene in 2001 where the astronauts are deprogramming the HAL 9000. Krugman’s new Keynesian intellectual components seem to be disappearing one at a time.
[Let’s just hope his next post doesn’t start out Daisy, Daisy . . . ]
No sooner did I finish writing this vicious smear of our newest Nobel Prize winner than I realized it was all a horrible mistake. Krugman’s not the one who’s lost his marbles, I am. Or at least I think that is the only plausible explanation. I’ll lay out the evidence, and you decide which view seems more likely.
This morning I noticed that I was linked to by a blogger way over in England. It seems he wanted to find an economist who was crazy enough to think that Fed policy has recently been tight, despite the low interest rates, and I was the only one he could think of. You’ll have to admit that it would be odd to cite an authority as obscure as me, if there were more famous economists making the same point. Here is what Leigh Caldwell said:
Scott Sumner has argued that money did become tighter later in the decade, more so than signified by interest rate movements; if true this could be one reason that asset prices are not recovering despite very low interest rates
Despite very low interest rates? Wait a minute, I though asset prices were usually low when rates were low, and high when rates were high. High rates and high asset prices in 1929, low rates and low asset prices 1932. Ditto for 2000 and 2002, and ditto for 2007 and 2009. Why the term ‘despite,’ I thought my view was conventional wisdom?
Now I’m starting to feel like Kevin McCarthy in the film Invasion of the Body Snatchers. Have I “misremembered” my history of progress in 20th century macro? Has some quantum fluctuation plunged me into a parallel universe where post Keynesian theory accurately describes the laws of macroeconomics?
I try to remain dispassionate and look at things logically. What are the odds that some pod-people from space have not only rewired Paul Krugman’s brain, but that of most other macroeconomists? Isn’t it more likely that I am going crazy? Like a character in a Borges story, I am now afraid to open Mishkin’s textbook, for fear that the passage I remember may not be there. If you see a tall thin guy at the Harvard or MIT economics parking lots, jumping on the windshields of cars, please call the appropriate authorities.
Tags:
13. July 2009 at 11:58
Hi Scott.
I suppose some may forget that it is the relationship between money demand and money supply (and not simply what is happening in absolute terms to the supply of money) that determines whether monetary policy is easy or tight. They see the expansion of M2, for example, by (say) 15% and a huge expansion of the monetary base and think that therefore money is easy without thinking too much about how the demand for money may also have changed or whether it has itself been affected by policy.
Krugman’s point (in the post you referenced above) about dogma applies to pro-fiscal-policy types as much as it does to pro-monetary-policy types in that the former may be a little too eager to say “oh well, told you monetary policy probably wouldn’t work (or has reached the zero bound or whatever), better ramp up fiscal stimulus and the size of government”.
I think he makes a fair point about monetary policy being viewed by almost everyone as ideologically neutral even though it may be the most significant and fundamental intervention of all, even without considering its more exotic forms like credit easing.
13. July 2009 at 13:33
David, I should clarify that there is nothing wrong with the basic point Krugman is making. I don’t think fiscal stimulus is very effective, but he is well within his rights to argue that it has never been given a fair chance. That’s my argument for monetary stimulus. My only problem is that he assumes monetary policy has been expansionary merely because interest rates are low. If that is right, then why is Joan Robinson nuts? Or is she not nuts?
I do think money is much more ideologically neutral than fiscal stimulus. Monetary policy has not been a political issue in America for decades, perhaps since 1933. Fiscal stimulus is a huge issue because it involves government spending and taxes. Monetary policy is primarily a technical problem, not an ethical problem.
13. July 2009 at 13:35
Scott,
You are not alone. Someone once wrote:
“…The answer is that Keynes never envisioned such a monetary regime as being possible. And even if he had, he did not believe in the Fisher effect (except under hyperinflation) and thus would not have seen Meltzer’s proposal as accomplishing anything…”
Oh… wait… that was you… so maybe you are alone and going crazy. Have you considered the possibility of being schizophrenic? That would certainly explain why some of your last posts include the words mind snatchers, depression and paranoid in their titles. Are you wearing a aluminum foil helmet?
Alex.
13. July 2009 at 14:43
Scott,
Your blog is awesome. The writing is superb and witty, and it could even be that you’re not crazy.
But let me see if I can get the narrative straight now:
The market crashed last October not because of the bad loans and the collapse of the housing bubble, but because the market realized the Fed was taking too tight a monetary policy in response to the monetary crisis created by housing, etc.?
If so, I have trouble understanding who all these market participants moving the market were. I was lucky enough to short the market last summer because the reasoning of the doom & gloomers sounded more plausible to me than the reasoning of the optimists. E.g., The Economist pointed to the 2004 IMF study showing how housing bubble crashes usually result in a nasty equity market crash. In other words: there was a conventional wisdom to the extreme pessimism at the time and that conventional wisdom was: the problem is housing, consumers with too much debt and bad loans. But you are saying that was not the real problem. But you are also saying that the market foresaw the real problem, which was monetary. So perhaps I have my answer: all these speculative suckers are in aggregate wise. Those like myself who shorted the market were right but for the wrong reason?
Maybe the market only seems predictive because in fact it is causative.
13. July 2009 at 16:14
Scott,
You raise a good point about the inability to use interest rates as an indicator of the stance of monetary policy. However, I think that one possible reason your point does not convince people is that people have become used to Taylor rules and CBs using interest rates to target inflation, and hence see interest rates as THE indicator of monetary policy.
Using this Taylor-rule paradigm, your point could be extended by noting that interest rates at zero – a level which is widely regarded as being the most ‘easy’ m/p – is significantly higher than the Taylor-rule implied level. This would also highlight your point that zero interest rates are not easy money.
This might be a better argument for people to follow than saying that the Fed is not conducting m/p in a manner consistent with the inflation target (although the 2 arguments are largely the same).
Alan
13. July 2009 at 16:35
I find a better method, I ignore money and consider it just another good. In my scheme, the yield curve represents the average growth of the inventory of goods (money included) over the various terms periods.
Then, in my world, tight money is differentiated from tight shoelaces. Shoelaces are tight because there is a constraint in the production network for shoe laces. Money is tight because there is a constraint in the money distribution network. Then I let price fall out by measuring the volatility of a particular goods inventory at a particular stage in the distribution network to the volatility of money inventory at the equivalent stage of the money distribution network.
Price is derived, but the essential components of the economy remain and can be analyzed as a queuing model without regard to price.
Then inflation falls out as too many stages of production with interference between the production terms (as in term structure). Deflation is the opposite, the economy is under-determined with too few stages of production.
Then the important concept becomes relative inflation, the growth of one inventory over another when both inventories are governed by the same term in the economic yield curve.
If there are insights into micro foundations, like money illusion, the tendency to gamble, fraud, etc; these end up being modeled in the inventory functions of the particular good.
13. July 2009 at 16:54
Scott:
“My only problem is that he assumes monetary policy has been expansionary merely because interest rates are low.”
I think you are right – the point I was trying to make was that a failure to consider the role of inflationary expectations is really just a subset of failing to explicitly consider the demand for money (or at least that’s how I see expectations entering the exercise).
It is beyond question that fiscal policy is more politically charged than monetary policy. From my perspective, that’s not the same thing as saying that monetary policy is in substance more ideologically neutral than fiscal stimulus. Monetary policy is certainly generally viewed and taught as a technical issue but I would suggest that that is largely because most economists with mainstream training (at least of my generation – graduating in the early ’80s) were not aware that there might have been an alternative to central banking. So my view is that, at least in recent decades, monetary policy came to be viewed as a technical issue essentially by default.
Regardless of what side of that debate (central vs free banking) one comes down on, it is the case that central banking and monetary policy represent a very significant government intervention. It doesn’t get any more fundamental in a capitalist market economy than to be mucking about with the medium of exchange and interest rates and acting as a lending institution. One needs to be awfully confident that one knows what one is doing if one is going to be engaging in that sort of stuff. Setting aside the question of the Fed’s track record, I would have thought that the relative lack of professional policy consensus alone is cause for considerable concern (you guys can’t all be right).
“I don’t think fiscal stimulus is very effective, but he is well within his rights to argue that it has never been given a fair chance.”
Governments have been at fiscal policy for quite awhile. I shudder to think how many tries at it they get before we get to conclude that it has been given a fair chance.
13. July 2009 at 18:02
Everyone on this blog writes as if the threat were coming from Keynesianism… It isn’t.
It’s coming from mercantilism (combined with hard-money fanatics run amok):
http://en.wikipedia.org/wiki/File:Cumulative_Current_Account_Balance.png
13. July 2009 at 18:21
The problem with Fiscal stimulus is that it usually carries a negative productivity.
13. July 2009 at 23:41
Jon,
You might like this:
Several months ago, the WSJ, in a tribute to Milton Friedman, recounted a story about when Friedman went to visit Asia in the 1960s, and came upon a construction site, where workers were busy digging a canal – but using shovels to do it. Somewhat shocked, since there were machines available that would make the arduous task much easier, Friedman asked what was going on. The response came, that this was a “jobs project”, so the workers were using shovels to make sure there was enough work available. “Oh I thought you were trying to build a canal,” came the reply. “If it’s a jobs project you should be using spoons.”
14. July 2009 at 00:32
I’m not sure where all this stuff about monetary policy lacking political charge. It’s certainly politically charged in the UK. It seems quite charged in the US too, and rightly so.
About the German hyperinflation this reminds me of a bit of “The Theory of Money and Credit”. Mises discusses variation in price level in the 1912 edition of that book. He adds the following note to the discussion in the 1924 edition:
“In a review of the first edition (Die Neue Zeit, 30th year, vol. 2, p. 1024-1027), Hilferding criticized the above arguments as “merely funny.” Perhaps it is demanding too much to expect this detached sense of humor to be shared by those classes of the German nation who have suffered in consequence of the depreciation of the mark.”
14. July 2009 at 05:17
Alex, I have considered that possibility. I think I see black helicopters outside my window right now. And by the way, I have noticed that you are one of my more “eccentric” commenters. 🙂
Rob, Don’t take this personally, but I definitely think the big players like the Harvard endowment that lost a fortune last fall are smarter than you are (and me.) So I think you were a bit lucky. But I am just assuming you are an average commenter. If you can consistently beat the market, then maybe you are exceptional.
The housing crash depressed stock prices modestly, but the real stock crash wasn’t associated with housing, but rather the fact that in late 2008 the trouble spread to other sectors, especially manufacturing all over the world. This was no longer a housing specific problem, it was a lack of AD. And its the Fed’s job to insure there is enough AD to keep NGDP rising modestly (while also avoiding an overshoot toward high inflation. My next post has more on this topic.
I think the 1987 crash (down 40% in 6 weeks) refutes the “causative” theory. The huge stock crash didn’t even cause a tiny blip in growth.
Alan, Good point. In previous posts I have argued that the Taylor rule idea is OK, the problem is that specific Taylor rules are backward-looking. The economy and expectations simply changed too fast in late 2008, and swamped the Taylor rule. If you have a forward-looking Taylor Rule, it would have called for a much more aggressive Fed response in late 2008. But there is also a circularity problem with (discretionary) forward-looking rules, so my first best option is still NGDP futures targeting. Thus with a CREDIBLE forward-looking policy of targeting 5% NGDP growth, a zero rate might have been enough to keep on track. That is one thing that makes monetary policy so confusing.
Mattyoung, I don’t think you can leave money out. I agree that some relative goods price changes are important, but I see money as the root cause of some of those changes. Monetary shocks produce all sorts of real effects in the short run, to both quantities and prices. I think you are picking up those effects, but you still need money in the model to see where nominal aggregates are going once prices have adjusted.
David, I understand your points much better now and tend to agree. Let me add a footnote. You were right to emphasize the Fisher effect, but I don’t want people to think that is all I am talking about. Krugman’s argument couldn’t be “fixed” by inserting the real interest rate, as that rate is also a flawed indicator. When markets are forward-looking a severely contractionary monetary shock can actually reduce real rates. How? By reducing expected RGDP growth sharply. This leads to less demand for credit (or a shift in the IS curve if you prefer.) The real rate may rise in the short run during tight money, and certainly did in late 2008, but I have also seen cases where even the real rate moved in the “wrong” direction.
Statsguy, Thanks for the insight. My blog is unusual in that it has something for everyone. I think much more stimulus is needed (like Krugman and DeLong) but I am a libertarian who thinks fiscal policy is often wasteful, isn’t necessary with good monetary policy, and isn’t even all that effective (although I have an open mind on the multiplier.)
Jon, I agree.
Adam P. I like that story. There are a lot of good Friedman anecdotes.
Current, Thanks for the Mises story. Monetary policy was the number one political issue in the 1890s, and again in 1933. It has almost disappeared as an issue today because politicians don’t really understand it. Even if they favor more stimulus, they don’t blame the Fed because they assume it has run out of ammo. Of course when policy debates were defined in terms of the price of gold (like the 1890s and 1933) there was no running out of ammo, as gold prices can be raised to infinity.
14. July 2009 at 11:53
Scott,
“Alex, I have considered that possibility. I think I see black helicopters outside my window right now. And by the way, I have noticed that you are one of my more “eccentric” commenters.”
The title and topic of your last post (panic attacks) confirms my suspicions. About my eccentric comments it is a simple strategy, if I avoid talking about economics people will not realize what a crappy economist I am. To quote Mark Twain … it is better to keep your mouth shut and be thought a fool than to open it and remove all doubt.
Alex.
20. July 2009 at 10:31
I dunno, Perfessor. You seem pretty lucid to me. I haven’t gotten to the point of understanding how your medicine is supposed to cure the disease, but I am more and more agreeing with your diagnosis of the problem and have been talking it up with some of the Cornell undergrads.
20. July 2009 at 17:24
Thanks D. Watson.