Tuesday, November 30, 2010

The State of Macro

It's hard to tell who is more out-of-it, the author of this WSJ article, or Paul Krugman. To the extent that the WSJ guy talks to any serious economists, he gets reasonable answers. Lucas is quite polite, and Mark Gertler puts it nicely:
It strikes me as not productive to say that all we have done is a complete waste. The profession is extremely competitive. If you have a better idea, it's going to win out.
Krugman should take that to heart. As usual, here, he sees himself as the King of Prescience:
More specifically, we knew all about liquidity traps, and had at least thought about balance-sheet crises, a decade ago.
In fact, liquidity traps have been well-known to monetary economists for some time. Twenty years before Krugman thought about liquidity traps, Charles Wilson wrote this paper, which tells us most of what we need to know about the phenomenon:
“An Infinite Horizon Model with Money,” in General Equilibrium, Growth and Trade, edited by Jerry Green and Jose Scheinkman, Academic Press, New York, 1979, pp.81-104.
Krugman also says this:
It’s true that if you bought completely into rational-expectations macroeconomics, the crisis in the economy should be causing a crisis in your faith —
Now, the people who "bought completely into rational-expectations macroeconomics" would be essentially the whole profession. Krugman of course is stuck in 1978. He doesn't realize that Keynesians, new classicals, and whoever else, "bought in." This is what Woodford, Gertler, Gali, Kiyotaki, Lucas, Prescott, and all of the rest of us, do. There is no crisis of faith, only pressing and interesting problems to work on.

The WSJ guy finishes with this:
Many economists think the next big idea will more likely come from the ranks of younger Ph.D candidates, who are producing reams of work examining the financial crisis. Established academics—such as Mr. Gertler, Nobuhiro Kiyotaki of Princeton, Marcus Brunnermeier of Princeton, Michael Woodford of Columbia and Robert Hall of Stanford—are making progress on including banks, financial markets and even a bit of irrationality in traditional models.
I'm glad to hear that we are finally making some progress in getting banks and financial markets into "traditional" models. Of course, Gertler, Bernanke, yours truly, and Bruce Smith started doing that in the early 1980s.

Monday, November 29, 2010

More on Sectoral Reallocation and Labor Markets

DeLong, Krugman, and Thoma, were recently fuming about this op-ed in the LA Times. The relevant quote is this:
The day before his party's shellacking in this month's elections, President Obama sat down with his economic team to examine the single most important issue for voters across the country: jobs.

But the question on the agenda was not how to accelerate the recovery or target job creation to the depressed Rust Belt. It wasn't even the challenge of how to persuade corporations to spend their cash piles on investments and jobs — although both have been extensively debated for many months. The president had called the meeting to grapple with what he and his propeller-head economists have been debating for some time: the wonkish question of whether today's high unemployment rate is structural or cyclical.

Such questions are no doubt vital to shaping economic and trade policies as well as the general worldview of the West Wing. But on the eve of an electoral disaster, the esoteric discussion in the Oval Office points to a larger challenge facing the Obama White House as it tries to stage its own revival.
If you thought that DeLong/Krugman/Thoma are upset about the abuse of propeller-headed wonks, you would be wrong. Apparently, the problem has to do with the fact that Obama and his economists would deign to mull over "structural" unemployment.

According to Thoma:
The administration is still allowing the other side to take control of policy debates.


DeLong stays in character by going nuclear:
Let me point out that I think that the senior Obama advisor quoted is a liar.


Krugman comments here and here. Here's a quote:
More generally, I can’t think of any Democratic-leaning economists who think the problem is largely structural.

Yet someone who has Obama’s ear must think otherwise.

No wonder we’re in such trouble. Obama must gravitate instinctively to people who give him bad economic advice, and who almost surely don’t share the values he was elected to promote. That’s what I’d call a structural problem.


This is all very interesting. DeLong/Krugman/Thoma are confused, and on several dimensions. I've discussed sectoral reallocation before here, and in other posts, and I think it is important for our current predicament. Here are the key issues:

1. Unemployment is unemployment. Attempts at organizing our thinking using such concepts as "structural unemployment," "cyclical unemployment," "the natural rate of unemployment," etc., is unproductive. These terms have no precise meaning, and therefore cannot be measured. We do, however, know what unemployment is, i.e. the number of people who answer "no" to a particular question put to them by the BLS surveyor (roughly: are you employed?) and answer "yes" to another question (roughly: are you actively searching for work?). Unemployment is a particular activity - search - which involves choices. How much effort should I allocate to searching for work? When should I stop searching and exit the labor force? If I receive a job offer, should I accept it, or continue searching? Economists who put themselves in the shoes of the unemployed came up with models that helped us understand unemployment and its determinants. Those models are search models, and economists think so much of search models that we gave Nobel prizes to three smart guys who worked on them. The Mortensen-Pissarides search model has become a workhorse in dynamic macroeconomics for studying problems in labor/macro.

2. Labor market behavior in the US during the recession and up to the current time has been anomalous. Lee Ohanian does a nice job of describing some of the key anomalies here (though I am not much into his views on financial factors and on the effects of recent policies). In particular, relative to the average post-WWII recession in the US, and relative to what occurred in other developed countries during this recession, the fall in employment has been very large relative to GDP, and very persistent. Ohanian does not discuss this, but as I point out here, the drop in residential construction is also extremely large and persistent. The housing sector, which was at the center of the financial crisis and the recession, can also be the center of a sectoral reallocation story about the recession. The housing slump was certainly not uniform across regions in the country, and residential construction is notably low-skill-intensive. Both of these features make the frictions associated with sectoral reallocation particularly important currently.

3. Buying the idea that sectoral reallocation is important does not mean that I throw out the idea that government intervention can be productive, nor does it mean that I must be a Republican. Here's what Krugman says:
As far as I can tell, the only economists who believe that we’re suffering largely from a rise in structural unemployment are those who are ideologically committed to the view that the demand side of the economy doesn’t matter — and so by definition, in their universe, any large rise in unemployment must be structural.

But you have to ask, why do these people have a voice in the Obama administration?
Well, baloney. I take it as encouraging that people in the Obama administration are actually struggling to understand what is going on in the US labor market. I know people in the Federal Reserve System who are doing the same thing, and they should be commended for it.

Now, it's clear we need some better models. The Mortensen-Pissarides (MP) model I mentioned above is useful on certain dimensions, but it has little to say about sectoral reallocation and its role in this or any other business cycle episode. The idea behind search theory is that heterogeneity on both sides of the labor market creates a friction that makes finding the right job (or finding the right worker) time consuming. MP embeds that friction in the matching function. Taking the matching-function approach avoids what is most interesting, and key to the problem. What we need is a multi-sectoral model with heterogeneous labor and firms, but of course building such a model would be difficult (otherwise someone would have already done it).

However, suppose we had such a model. What might it tell us? There may be a particular role or roles for the government that we have never contemplated. Maybe, if applied to our current predicament, such a model would tell us that we need a much more generous unemployment insurance program, or that we should have government subsidies for relocation. Who knows? In any event, the proper reaction to the current state of the world is not to shut off the parts of our brains that might want to think about sectoral reallocation.

Saturday, November 20, 2010

More on Leverage, Debt Constraints, and Asset Prices

Given some of the issues raised in the Eggertsson-Krugman paper, a useful reference point is the paper Kocherlakota gave at the Jekyll Island conference. Kocherlakota's paper gives us some detail on the fundamental determinants of debt constraints and how they relate to the financial crisis. Check out his reference list, which is particularly useful.

Friday, November 19, 2010

Eggertsson and Krugman

I was asked by a couple of readers to comment on this paper by Gauti Eggertsson and Paul Krugman. Krugman gives us the impression here that he is quite proud of this. We should certainly encourage Krugman if he wants to get back into the game. I would love to think that there is nothing holding back the senior members of the profession, and that we can learn new things, but when I can't remember what happened yesterday I have my doubts.

Eggertsson is a former Mike Woodford student, and a committed New Keynesian who works at the New York Fed (and appears to be visiting at Princeton, which explains the Krugman connection). He's supplying the New Keynesian technology to help Krugman to flesh out his thinking. The basic structure in the underlying model (some of which you need to go to the appendix to understand) is a standard New Keynesian framework. We have some infinite-lived optimizing consumers, and monopolistic competition. Some fraction of firms can set prices at will, and some must set prices one period ahead, i.e. there is time-dependent pricing. The nominal interest rate is set by the central bank according to a Taylor rule, and there is no money in the model (Woodford "cashless economy"), but of course goods prices are set in units of money.

The novelties here are the following. First, there is some heterogeneity among consumers, i.e. consumers can be one of two types, and types differ according to discount factors, so we have patient and impatient consumers. Second, there is an exogenous debt limit faced by each consumer, set in real terms. Third, debt is denominated in nominal terms, by assumption.

What we get from this is the following. Suppose the economy is in a steady state where the impatient consumers are borrowing from the patient ones, and impatient consumers are debt-constrained. Then, suppose that there is an exogenous shock to the debt limit. There is then a "deleveraging" effect. Debtors have to reduce their debts, and they do this by reducing consumption. The real interest rate falls, and could fall sufficiently that, in this sticky price framework, the nominal interest rate hits the zero lower bound. There is a further effect from debt deflation in that the price level falls, increasing the value of the debt, and requiring further deleveraging. Fiscal policy can have a big multiplier in these circumstances, in part for reasons Mike Woodford has studied, and in part because the debt constraint implies that Ricardian equivalence does not hold.

My comments are as follows:

In the introduction, the authors state:
Given both the prominence of debt in popular discussion of our current economic difficulties and the long tradition of invoking debt as a key factor in major economic contractions, one might have expected debt to be at the heart of most mainstream macroeconomic models– especially the analysis of monetary and fiscal policy. Perhaps somewhat surprisingly, however, it is quite common to abstract altogether from this feature of the economy1.
In the footnote they cite a few papers, including Bernanke and Gertler, Kiyotaki and Moore, and Gertler and Kiyotaki. It is certainly true that much of mainstream macroeconomics ignores the frictions that make credit, banking, and monetary exchange important. Indeed, this has been one of the key drawbacks of New Keynesian economics. Since the mid-1990s, Keynesian economists have focused their attention on sticky wages and prices, and have neglected other frictions, and the financial crisis made it clear that they had missed the boat. It's good that they are trying to make up for lost time now, but they have a lot to learn. Some neoclassicals were not any better at recognizing the importance of financial and monetary factors, for example the "Great Depressions of the 20th Century" volume gives short shrift to monetary and financial factors.

There is a lot of relevant work that Eggertsson and Krugman are unaware of, or are ignoring. Bruce Smith spent his career working on models of credit market frictions in monetary frameworks. I have work closely related to Bernanke-Gertler, and predating it (including my 1987 JPE paper). Many people have studied related debt-constrained problems. These include Kehoe and Levine (1993) and Kocherlakota (1996). There is a large literature that uses standard incomplete markets models (e.g. Aiyagari QJE 1993) to study problems associated with bankruptcy (e.g. this paper). New Monetarist Economics is all about monetary and financial frictions (see this).

In constructing and analyzing the model, the authors cut a lot of corners. First, the debt constraint is imposed exogenously. The authors are clearly aware that something deeper is required, but they don't seem to think this is a big problem. Second, the debt constraint is set in real terms, but for some reason (also not in the model) the debt contracts are nominal. Third, in analyzing the model, there is some linearization around a deterministic steady state, but the dynamics are not worked out from the model - these are essentially pseudo-dynamics.

Cutting corners matters. In particular, in environments where we are explicit about debt constraints, for example with explicit limited commitment, Ricardian equivalence can be hard to escape. For example, borrowers can face binding debt constraints in equilibrium, but if the government has no advantage over the private sector in collecting its debts, then the government is faced with the same problem in tax collection. If someone defaults on their private debts, they also default on their taxes. The economy will only be non-Ricardian if the government has some advantage in collecting on its debts. Maybe it does in practice, but it is important to model this so that we can understand it properly.

This is a kind of chicken model. We assume an advantage for the government in the credit market, in that the government can cut taxes today and increase taxes in the future, which effectively relaxes the credit constraints of borrowers, but only because the government can always collect the future taxes at no cost. This helps to give us big fiscal policy multipliers. In addition, there is some sleight-of-hand associated with an effect similar to what is in this paper by Woodford, where it is essentially monetary policy that is doing the work.

I can add to this my usual complaints about New Keynesian models. First, they are not explicit about monetary quantities and transactions, which are critical to how we need to think about monetary economies and monetary policy. This paper is more about fiscal policy, but the liquidity trap comes into play. If we want to understand that properly, we need a full-blown story about monetary frictions and central banking. Second, of course the constraints on pricing are exogenous. This obviously violates the Lucas critique. If pricing is so important, we need to think about how it responds to changes in policy rules. Third, Keynesian output effects are essentially by assumption. Firms have to, by assumption, supply whatever output is demanded at the price the firm is stuck with.

I think what the paper needs is the following:

1. Work out the results for the case with flexible prices. Indeed, one could start with a non-monetary economy. A basic Aiyagari incomplete-markets model will have the property that tightening up the borrowing constraint will lower the real rate, for example, but not much is known about dynamics. One could even compute solutions.

2. The linearization needs to be done for a fully-specified underlying stochastic model.

3. Where are those shocks coming from? It is not very informative to introduce a shock to the debt limit. The key thing is to understand how an increase in credit market frictions occurs as a result of factors that were occurring during the financial crisis.

4. The hard part is the debt-deflation story. In early versions of Bernanke-Gertler (1989), they tried to tell a debt-deflation story, but without success. The problem is that the best we have in terms of optimal contracting models are setups that deliver real debt contracts.

There are certainly some interesting ideas in here, but there are too many moving parts, and too many corners cut to feel very confident about the results the authors want to extract.

Thursday, November 18, 2010

Gregory Mankiw on QE2

Mankiw writes about QE2 here. It came as a surprise to me to learn, at one time, that Mankiw was a Republican, as I had assumed that hard-core Keynesians had to be Democrats. Of course, this should not have surprised me, as I know many hard-core new-classical types who are Democrats.

Like Krugman, Mankiw is worried about deflation, and he has a Keynesian view of how QE2 works, i.e. through interest rate effects on "aggregate demand." The interesting part is in this paragraph, where he nails some of the potential problems with QE2:
Moreover, I do see some potential downsides. In particular, the Fed is making its portfolio riskier. By borrowing short and investing long, the Fed is in some ways becoming the hedge fund of last resort. If future events require higher interest rates, the Fed will end up making losses on its portfolio. And even if doesn't recognize these losses (by not marking to market), it could end up paying more interest on newly expanded reserves than it is earning on its newly acquired portfolio of long bonds. Such a cash-flow deficit could potentially undermine the Fed's political independence (which is already not very popular in some circles). Yet if the Fed tries to avoid these losses by failing to raise rates when needed, inflation could indeed become a problem down the road. I trust the team at the Fed enough to think they will avoid that mistake.
He's basically in tune with some of the issues that Kocherlakota raises here. I like this paragraph, but disagree with the last sentence. The Fed is trustworthy, in some sense, but I don't think they will be able to withstand the pressure to do the wrong thing when the time comes to do the right thing.

Wednesday, November 17, 2010

What do the German Finance Minister, the Chinese Government, Narayana, and Sarah have in Common?

According to your friend and mine, Paul Krugman, the Keynesian enlightenment is over, and we have entered the Dark Ages of economic thought. He first gives us a Schumpter quote:


Now, Krugman claims that Schumpeter was long ago discredited, but this passage is not totally crazy. Schumpeter is telling us that caution in policymaking can be appropriate. We can do more harm than good by intervening to slow down adjustments that will have to occur in any event, and sometimes allowing events to take their course can be the best strategy. Clearly, this need not always be the case, but as a general principle, it's hard to argue with it.

Schumpeter also states, presumably also as a general principle, that "the trouble is not fundamentally with money and credit," and that I would argue with. Obviously, we don't have to look any further than the recent recession to find an episode where money and credit seem to have been important elements in what was going on. However, one could make a case that what currently ails the US economy - principally dismal performance in the labor market - has little to do with money and credit, and monetary invention will do little, if anything, to cure the problem, whatever it is. There is a lot we do not know of course, and there may be something going on that we do not understand about how financial ailments interact with labor market ailments. Who knows?

Now, the funny part of Krugman's piece is the conclusion:
But here we are, in 2010 — and something very much like that position is being forcefully advocated by Wolfgang Schauble, the government of China, Narayanan Kocherlakota, and Sarah Palin.
Now, I'm assuming that what Krugman has in mind here is that all of these people (or organizations) have in common a distaste for QE2. However much Narayana may enjoy the company of these other individuals (maybe he and Sarah might enjoy a good game of table hockey), I don't think he belongs in that group. Kocherlakota's latest posted speech is this one. In it, he discusses some policy options, and his conclusion is this:
I’ve talked about three possible tools—lowering the IOER, strengthening the forward guidance in the FOMC statement, and quantitative easing. As I mentioned earlier, Chairman Bernanke observed in his August 27 speech that each of these tools has benefits and drawbacks that must be balanced against each other. With QE, I would say that the multiple novel effects make the calculus even more difficult than usual.
Before coming to that conclusion, he has some thoughtful things to say about quantitative easing and its possible effects, using available economic theory, and framing the debate in terms a lay audience should be able to understand. Basically, he's saying that we have some theory and empirical work to bring to bear in understanding what QE might do, but there is an awful lot that we do not know. There's not much in there to disagree with, and it's far from Dark-Ages economics.

Tuesday, November 16, 2010

The Fed's Dual Mandate

Some Republicans in Congress have come out in support of altering the Fed's dual mandate to focus solely on price stability. The dual mandate is written into the Full Employment and Balanced Growth Act (a.k.a. Humphrey-Hawkins Act) of 1978, which was an amendment to the Employment Act of 1946. Humphrey-Hawkins is behind this statement in the last FOMC statement:
Consistent with its statutory mandate, the Committee seeks to foster maximum employment and price stability.


Now, what does "fostering maximum employment" mean? Would we have maximum employment when all working-age individuals have a full-time job, when all working age individuals have a full-time or part-time job, when some specified fraction of the working-age population has a full-time or part-time job, when total hours worked exceeds some quantity? If we specified exactly what maximum employment is, would it even be feasible for the Fed to achieve this? What a can of worms!

It is certainly widely-recognized that central banks can control inflation, though of course the short-run connection between central bank actions and inflation can be loose at times. Some central banks in the world that have adopted explicit inflation targeting have achieved some success in conforming to their targets. Generally, it seems that it is feasible for a central bank to keep inflation within some target range, whatever it is, so long as the range is sufficiently broad.

Inflation targeting was first adopted in New Zealand in 1989, and since then the U.K., Canada, Australia, South Korea, and Brazil, among others, have adopted it. We would think of most of these countries as more left-leaning, and with more extensive social insurance than is the case in the United States, so inflation targeting is by no means part of a general right-wing agenda. In the past, many economists, inside and outside the Federal Reserve System, have been proponents of inflation targeting, including Ben Bernanke.

Given the support in Congress for focusing the Fed's mandate on price stability, the time might be ripe for Bernanke to get behind this. This is certainly consistent with things that Charles Plosser (Philadelphia Fed President) has been saying about a new Accord between the Fed and the Treasury. This is an important opportunity for the Fed, and getting on board with an inflation-targeting agreement could also keep the Fed out of trouble (with everyone).

QE2 Complaints

Post-financial crisis, most economists appear to have been supportive of the Fed. Of course, one of the potential costs for the Fed of doing unusual things is that they are going to find more people disagreeing with what they are up to. Yesterday, a number of economists signed on to a public statement criticizing the Fed, which reads:
We believe the Federal Reserve’s large-scale asset purchase plan (so-called “quantitative easing”) should be reconsidered and discontinued. We do not believe such a plan is necessary or advisable under current circumstances. The planned asset purchases risk currency debasement and inflation, and we do not think they will achieve the Fed’s objective of promoting employment.

We subscribe to your statement in the Washington Post on November 4 that “the Federal Reserve cannot solve all the economy’s problems on its own.” In this case, we think improvements in tax, spending and regulatory policies must take precedence in a national growth program, not further monetary stimulus.

We disagree with the view that inflation needs to be pushed higher, and worry that another round of asset purchases, with interest rates still near zero over a year into the recovery, will distort financial markets and greatly complicate future Fed efforts to normalize monetary policy.

The Fed’s purchase program has also met broad opposition from other central banks and we share their concerns that quantitative easing by the Fed is neither warranted nor helpful in addressing either U.S. or global economic problems.
The signatories of this document include Michael Boskin (Stanford), Charles Calomiris (Columbia), and John Taylor (Stanford), but these academic types (who in some cases have done time in Republican administrations) are in the company of people like William Kristol. What unites these people seems mainly to be the Republican party, so we can interpret what they are up to in that light.

I have some sympathy for what these people are saying, but I think that the Fed should be doing what it can to increase the rate of inflation. The Fed's implicit inflation target has been 2%, and Fed officials, including Bernanke, have actually made that target explicit in public statements (though not in the FOMC statement itself). By any measure, the current inflation rate is below 2%, and the Fed should then do what it can to increase it, to make good on its commitment. Under current circumstances, with a large stock of excess reserves in the financial system, it is widely recognized that conventional open market operations (purchases of T-bills) will have no effect, and the interest rate on reserves is close to zero. Thus, there is no other option available but purchases of long-term assets. Purchases of private assets by the Fed is a dangerous game, so the Fed needs to purchase long-term Treasuries, which is what it has proposed to do. How much Treasuries should it purchase? That is the million-dollar (or six hundred billion dollar) question, and I don't know of anyone who has a good answer.

The key problem is not that the Fed is engaging in purchases of long-term Treasuries, but how much of it they are doing, and the level of commitment implied by announcing specific quantities to be purchased over a specific period of time. I think a better approach would have been to say that purchases were going to be made to achieve a particular inflation target (say 2%), or a target for the price level path. Given the announced purchases, the Fed will have to do some talking if they want to deviate from what was announced in the last Fed statement, and that could be costly in terms of their credibility.

An odd feature of the document criticizing the Fed's policy is that John Taylor appears to have played a prominent role. Taylor's rule, which plays a central role in New Keynesian economics, is explicitly interventionist. It formalizes the Fed's "dual mandate" in a rule that sets the fed funds rate target in response to actual inflation and the "output gap." Implicit in the Taylor rule is the idea that the Fed can and should intervene to smooth the path of real GDP over time. Taylor has stated (see this) that: "To establish Fed policy going forward, the best place to start is to consider what has worked in the past." Indeed, some people, including Glenn Rudebusch, have used Taylor rule predictions, which tell us, for example, that the fed funds rate should currently by -6%, to justify quantitative easing. They use this justification in spite of the fact that typical New Keynesian models with Taylor rules do not capture the effects of purchases of long-term Treasuries by the Fed. Now, how John Taylor thinks QE2 works is a mystery. One might expect that he thinks like a typical New Keynesian, but it seems like most of those people want to give QE2 a shot, so apparently Taylor's views are different. Maybe his ideas have changed, but it's unclear how or why.

Some Fed officials have made public statements supporting QE2, and these statements appear to be respones to criticism coming from other countries. Here, Discussing QE2, Janet Yellen says:
The purpose of it is not to push down the dollar. This should not be regarded as some sort of chapter in a currency war.
As well, as reported here, New York Fed President William Dudley "said that the Fed’s move was not intended to affect the value of the dollar..." Right. This sounds like what happens when your 8-year-old is clowning around and spills her milk ("I didn't mean to do it.") By their nature, monetary expansions tend to make the currency depreciate. I know it, Ben Bernanke knows it, Janet Yellen knows it, and so does William Dudley. If the FOMC was not thinking about the international repercussions of QE2, it should have been.

Thursday, November 11, 2010

Reply to Richard Serlin

Richard Serlin, who deserves an award for time spent writing blog comments, asks:
You know, Stephen, you would be doing a great service to economists as well as the general public if you would finally answer the question asked of you many times, in many ways, how can we get inflation -- in the US, right now, in our current situation, with our current employment level, inflation level, etc. -- without also getting an increase in employment and real GDP.
Here's my stab at an answer.

The key question here is: "how can we get inflation ... without also getting an increase in employment and real GDP." The question is a bit odd, as we usually ask it the other way around. Monetary policy seems to be about manipulating nominal quantities, so how on earth can that have anything to do with real quantities like employment and real GDP? As for many economic questions, it is useful to start with the simplest case and then make it more complicated. An uncontroversial notion in macroeconomics is that money is neutral in the long run. If the central bank increases the level of the quantity of outside money - currency and reserves - in a one-time fashion, in the long run all nominal quantities, including prices and wages, will increase in proportion to the money supply increase, and there are no real effects.

Now, think about something more complicated. Imagine that the central bank permanently increases the rate of growth rate in the stock of outside money. Would this be neutral in the long run? Well, clearly not, as otherwise why do we care about inflation? If increasing the level of the money stock increases the price level, then increasing the rate of growth in the money stock increases the inflation rate in the long run. Leave aside the question of what these long-run nonneutralites from changes in money growth are (more on that later when I have time), but suffice to say that a higher money growth rate implies a higher inflation rate in the long run, and this relationship is roughly one-for-one.

Another complication is thinking about what the short-run non-neutralities of money are. This is much more controversial. There are many different ideas about short-run nonneutralities, and the implications for monetary policy. Among these are:

1. Lucas (1972 "Expectations and the Neutrality of Money"): Monetary policy has short run effects because monetary intervention confuses price signals for private sector producers. I am producing a product for which the market price goes up. I produce more thinking that this is because my price went up relative to other market prices, but I was fooled into thinking so. Actually, what happened was that the Fed increased the money supply and all prices went up. All producers are in fact doing the same thing. Policy conclusion: The Fed should behave in a predictable fashion. This monetary intervention may move output around, but those fluctuations are inefficient. Is this theory any good? I don't think many economists pay much attention to it any more, principally because it seems the information friction in this type of model is not plausible.

2. Sticky prices: The New Keynesian story is that only some firms can or wish to change their prices in the short run. If I am a sticky-price firm, I produce more when demand for my product rises. Demand for the products of all firms can rise in the present due to monetary intervention which reduces the nominal interest rate, which also reduces the real interest rate (remember the prices are sticky), and therefore induces intertemporal substitution of goods, increasing the demand for goods in the present. Policy conclusion: It need not be efficient for the central bank to increase output, even though it can. However, there is a role for monetary policy in minimizing the welfare losses from the relative price distortions induced by sticky prices. A key flaw in the theory is the link between monetary policy and output. We have to think that is somehow less costly for the firm to hire more factor inputs and produce more output in the face of higher demand than to increase its price.

3. Sticky wages: In modern macro, this works much like (2), with the same caveats.

4. Market segmentation: Early models of this type were Grossman and Weiss and Rotemberg. Lucas worked on this, as did Fernando Alvarez, Andy Atkeson, Pat Kehoe, and Chris Edmond, among others. The idea here is that different people, firms, and financial institutions have different degrees of participation in financial markets. Some hold stocks, others don't; some trade frequently in sophisticated asset markets, others don't; some have transactions accounts with banks, but others don't. What this leads to is a distributional wealth effect from monetary policy actions. For example, an open market operation will have its first round effects on the economic agents who trade frequently in financial markets, and thus have initially large effects on asset prices, e.g. nominal interest rates. Over time, the effects become more diffuse, affecting everyone in the economy. These models never gained much traction, though Christiano and Eichenbaum did some empirical work in the early to mid-1990s on this stuff. I think they are interesting, but I don't think this mechanism will give you much in terms of effects on aggregate output and employment.

5. New Monetarist Models (shameless advertising): You can read about that stuff here, here, and here. One idea in here is that asset liquidity and liquidity premia are determined by how assets are used in exchange, and monetary policy actions can have subtle effects on asset prices and output that we can understand by being explicit about the exchange process in retail markets and financial markets.

6. Models with Illiquidities: Here, I am thinking of work by Mark Gertler, Nobu Kiyotaki, and John Moore, among others. This is somewhat related to (5), but is less explicit about the financial frictions. Interesting ideas though.

Where does that leave us right now? I more or less dismissed (1) as a useful theory of monetary policy. What does (2) tell us about what the effects of QE2 will be? This theory tells us that, once we are at the zero lower bound on the nominal interest rate, there is nothing the central bank can do. We can lower the interest rate on reserves from 0.25% to zero, but that won't do much, according to the theory. Of course there are some nominal interest rates - the yields on long maturity bonds - that are positive, but New Keynesian models are silent about how the quantities on the central bank's balance sheet can move those long-term interest rates. Ditto for (3).

Now, market segmentation might be able to tell us something here. Indeed, old-fashioned "preferred-habitat" models (though I don't want to call these things models - they are really just stories) are essentially about market segmentation. Given enough market segmentation, the Fed could in principle reduce long bond yields by purchasing long bonds. Also, (5) and (6) may have something to say as well. Exchanges of outside money for long-maturity Treasury bonds by the Fed could indeed increase the average liquidity of the outstanding debt of the Treasury and Fed combined. To the extent that the marginal value of liquidity in financial markets is high, this will matter, and could reduce long bond yields.

Now, how much will any of these things matter for real activity in the short run given the size of the QE2 operation? I think not much. With short-term nominal interest rates close to zero, and the market already flooded with liquidity from the previous large-scale asset purchases conducted by the Fed, I can't see that it will matter much, for real activity. It can certainly matter for inflation though, and this depends on the extent to which the outside money injections are not just held as reserves. One way you can think of the price increases happening without any increase in output is that price increases feed through the chain of supply. The price increases start in the markets for commodities - a storable commodity is just another asset, and we can think of this fitting into the array of available assets. Even if there are few good investment projects to fund, monetary policy actions can have the effect of bidding up the prices of commodities. These commodities are used in production processes, and increases in their prices increase the costs of production, and therefore lead to increases in final goods prices and in wages. All prices and wages go up together with little or no real effects.

QE2 Details

As was promised by the Fed, details of the first month of QE2 purchases are spelled out here. This is a bit like what the Fed does in normal times on a daily basis. In pre-crisis mode, the Fed would make a prediction about what would affect reserves on a daily basis, and then intervene in the overnight market in an attempt to hit the fed funds rate target. Now, they plan to buy $75 billion in Treasury securities each month, and a prediction is made concerning how mortgage prepayments will affect the stock of agency securities and mortgage-backed securities (MBS) in the Fed's portfolio. The open market desk then attempts to offset that. For the next month, the forecast is that $30 billion in agency securities and MBS will run off.

Tuesday, November 9, 2010

Sarah Palin and the Fed

Well, everyone else is doing it, so I guess we shouldn't have a problem with Sarah sounding off about monetary policy. I wouldn't go asking her questions about it though, as I'm sure she didn't write this stuff. It's more or less coherent, though misleading about the rate of increase in the price of groceries.

I'm going to focus on this, which is a very standard sound bite that I have heard before:
I’m deeply concerned about the Federal Reserve’s plans to buy up anywhere from $600 billion to as much as $1 trillion of government securities. The technical term for it is “quantitative easing.” It means our government is pumping money into the banking system by buying up treasury bonds. And where, you may ask, are we getting the money to pay for all this? We’re printing it out of thin air.
You can forgive people for this view as, to some extent, it is propagated by the approach people often take in undergraduate money and banking courses, where "money creation" is treated as some kind of mysterious process. The Fed is no more "printing it out of thin air" than is any economic entity when it issues a liability. For example, General Motors could issue a corporate bond to purchase new capital equipment. A liability is issued, out of thin air, and an asset is acquired. How is that different from what the Fed is up to? Well, the corporate bond issued by General Motors is a promise to pay something specific in the future. The Fed's liabilities - currency and reserves, principally - are not promises to pay anything. However, the Fed can, if it chooses, sell the assets in its portfolio at any time, and retire the money it has issued. In this sense, the money issued by the Fed has something in common with stock issued by corporations, which is also not a promise to pay anything specific in the future. A corporation can issue dividends on its stock, at its discretion, but this is not required. Further, just as the Fed can retire money, a corporation can buy back its stock.

The Fed is just a financial intermediary and, as such, it behaves like other financial intermediaries. It issues liabilities (out of thin air) and buys assets, and it has to worry about the liquidity of its assets and liabilities, their maturity, their rates of return, etc. What makes the Fed different is that it has been assigned a monopoly on the issue of particular kinds of liabilities - circulating small-denomination pieces of paper, and the stuff that is used daily in interbank clearing and settlement (reserves). Printing things out of thin air is certainly not the Fed's distinguishing characteristic.

Monday, November 8, 2010

SOMA

You probably know this, but in my quest for information on QE2 on the New York Fed web site, I found all the details of the Fed's security holdings. The statement on the details of how QE2 will be carried out is here. Then, if you go here, you get the a complete listing of the SOMA (System Open Market Account) Treasury and agency security holdings. This gives you the details of the quantities of each issue held, and how much of the total issue is in the SOMA. This will be interesting to watch in the coming months as QE2 unfolds. Note that the planned monthly purchases will account for a large fraction of the quantity of debt the Treasury will have to issue, on average, each month to finance the federal deficit. I'm guessing that the open market desk always purchases newly-issued Treasuries, i.e. "on the run," but I'm not sure. Anyone know?

Sunday, November 7, 2010

Thoma - Last Comment

Thoma is back again with this. Replies as follows:

Ah, I see, the surge in 2008 is why we are having such a problem with inflation today!
He's talking about the surge in currency in circulation in 2008. Yes, exactly. This is why I'm not an old-fashioned quantity theorist. What has to be going on here is a large increase in the world demand for US currency during the financial crisis. All the more reason to be worried about inflation, as the crisis-driven demand goes away.

o what has been happening to (the log of) M2 (the monetarists' favorite measure of money -- he does call himself a New Monetarist after all)?
No, New Monetarists don't care about broader monetary aggregates. Neil Wallace taught us that.

I also like the contention that calling Bernanke a "wuss" is not "disparaging Bernanke's character."
This seems a bit strange. I'm not sure how Mark comes by all this respect for authority. In this context, I think it's healthy to be skeptical about what these central bankers are telling us. They have a penchant for secrecy, and I don't think we should take everything they say at face value, or necessarily trust them. We've given them an important job, and I think they are taking some big risks. If they screw up, we'll all suffer for it.

Despite Williamson's wishes to the contrary, there's no evidence here that inflation is just around the corner, or even down the street.
Now the guy has me wishing for inflation. In a sense this is actually true. Early last year I bought a load of TIPS, and I continue to put my money where my mouth is. My TIPS have been doing quite well (not quite as well as the stock market, but a pretty good annual return, and particularly nice over the last few weeks). The current TIPS break-even inflation rate has gone up to 2.11% on 10-year bonds, and 2.62% on 30-year bonds. I'm expecting better things in the future. If I were really nefarious, I would be behaving like Mark, and complaining about how tiny the QE2 program is.

More Thoma

Here is Thoma's reply to my last post. Let's see what he says.

Saying that inflation is always and everywhere a monetary phenomena, and that prices depend upon the amount of outside money in the system, doesn't answer the question about how we get inflation before aggregate demand kicks up.
I don't know what "aggregate demand" is. That's in a language I can't make any sense of.

Exactly how the excess reserves leave banks in a depressed economy is not explained other than through reference to some vague friction that says banks won't want to hold reserves.
Presumably Mark would characterize the current state of the economy as "depressed." And, the fact is that reserves are leaving banks in the form of currency as we can see in this chart. Note in particular that reserves have recently been leaving banks at a more rapid rate. It's possible that we would get more inflation even without QE2.

Is the supply of credit the main constraining factor or is it the demand?
A contributing factor could be that some banks are capital-constrained. It would help to know where the reserves are in the system. Someone at the St. Louis Fed is working on this. I'll find out more this week.

But I still don't see how excess reserves are converted into real investments in plants and equipment...
The reserves aren't somehow converted into loans. You're thinking about the money multiplier, I think, which is misleading. If the opportunities are there, the banks will lend. In the absence of intervention that will happen, though probably not as fast as we would like.

Finally, if claiming someone is a "wuss" is a key component of your argument, I suppose that's fine, but we shouldn't pretend that an opinion about someone's character is based upon any sort economic reasoning.
No, I wasn't disparaging Bernanke's character. I know him well from the old days, though I haven't talked to him for a very long time, and he is a fine person. However, based on his behavior as a central banker, he may be a bit of a wuss. Buying more than $1 trillion in mortgage-backed securities was rather wussy, in my opinion. I'm thinking it's likely that, when push comes to shove, he's not going to be good on his word. There's not much science in that of course.

Grumpy Thoma

Apparently Mark Thoma didn't like my last piece on QE2. I've had a fairly peaceful time here for a while. Thankfully my fellow bloggers have not been paying much attention to me, and my readers are typically thoughtful and helpful in the comment box.

Now, as my mother (rest her soul) would have said, "Mark, did you get out of the wrong side of the bed this morning?" Hopefully my mother is not reading Thoma's blog, wherever she is, or she would think I had turned into a nasty piece of work.

Thoma was right about a couple of things, though. First, I did not lay out all the details of my arguments. Most of those are in previous posts, and obviously I can't assume everyone is reading all these things. Second, there is an inconsistency in there.

First, the details. What causes inflation? I'm with Milton Friedman on this one. Inflation is everywhere and always a monetary phenomenon. I'm not with Milton Friedman in the sense that I don't think the demand for an asset is anything like the demand for potatoes. Trying to find stable demand functions for monetary quantities is a waste of time. Think of the price level as being the terms on which the private sector is willing to hold the stock of outside money - currency and reserves. The Fed determines the total stock of outside money, and the private sector determines how that total gets split up between currency and reserves. What makes the price level go up? That would be anything that increases the supply of outside money relative to the demand.

Now, what is QE2 about? Under the current circumstances, with a large stock of excess reserves held in the financial system, it seems clear that a conventional exchange of reserves for T-bills cannot matter at all in the present. The Fed swaps one interest-bearing short-term asset for another, and nothing much should happen, short of some minor effects due to the somewhat different roles played by T-bills and reserves in the financial system. On the other hand, swapping reserves for long-maturity Treasuries, as in the QE2 plan, is a different story. We're now swapping a short-term interest-bearing asset for a long-term one. But what will the effects be? Unfortunately there is no good theory to tell us. To the extent that this matters in the present, for example by moving asset prices in the way that Bernanke seems to expect, this depends on some kind of financial market segmentation. Private financial intermediaries cannot be capable of undoing what the Fed is about to do.

Now, what I discussed in the previous paragraph is just about the current effects of the QE2 open market operations. What about the medium-term effects? There are two important points to note here about QE2. The first is that, while interest-bearing reserves, when they are held by banks, look essentially like T-bills, they have one feature that is very different from T-bills. This is that they can be converted one-for-one into currency. For a bank, a reserve account is a transactions account, and currency can be withdrawn from that account in the same way that you withdraw cash from the ATM. Thus, in contrast to T-bills, interest-bearing reserves can be converted into an asset that can be used in retail transactions.

Therefore, the more reserves that the Fed floods the financial system with, the more potential there is for inflation. As the economy recovers, other assets will become more attractive to banks relative to reserves, the demand for outside money will fall, and the price level must rise. Further, there could simply be a net increase in the supply of outside money relative to the demand at the outset of the QE2 operation. What I have in mind here is that, in spite of the fact that a QE2 open market operation simply swaps one consolidated-government liability for another, there may be some friction that implies that, on net, banks will not want to hold the extra reserves at market prices. Surely this is part of what the Fed has in mind. They think that long bond yields will fall. However, part of the adjustment should be an increase in the price level as well.

Now, if the inflation rate starts to rise, what happens then? There are three forces here that are going to make inflation control difficult. First, QE2 will have lengthened the maturity of the Fed's asset portfolio, so that the Fed has a lot more to lose from an increase in short-term interest rates. To tighten, the Fed will have to increase the interest rate on reserves (thus increasing all short rates), which results in a capital loss on its portfolio that will be larger the longer the average maturity of the Fed's assets. If the Fed continues to hold those assets, its income will fall, and if it sells the assets it will be selling them at a loss. If the Fed does not tighten, then inflation rises. None of these outcomes is very appealing. The second force at play is that Bernanke in particular thinks that monetary policy matters for real activity in a big way, and he will be very reluctant to tighten as he will think that he risks another recession. Third, and I may be wrong about this, but I think Bernanke is probably a wuss. He does not want to bear the short term pain associated with people screaming at him if tightening occurs.

Finally, on the inconsistency, I said here, by implication, that I did not think that QE2 would have much in the way of real effects. But I also said that it is costly to bring inflation down. Seems a little goofy, right? Some people think they understand nonneutralities of money well, but I don't feel like I do. Keynesians (new and old) have not convinced me that sticky wages and prices imply that a monetary expansion gives aggregate output a big kick in a positive direction. Some people, including me, made a case that market segmentation could imply a substantial redistributive effect of monetary policy, but this seemed to matter more for asset prices and allocation than for aggregate activity. New Monetarist ideas may give us short-run nonneutralities of money associated with asset trading and liquidity, and with credit market activity, but we haven't worked all of that out. Given what we know, my forecast is that the net real effects of QE2 will be insignificant. Now, what if inflation takes off, Bernanke is not a wuss, and substantial monetary tightening occurs? Do we have to suffer a lot to bring inflation down, or not? The "Volcker recession" was severe, but in the early 1980s inflation came down over a relatively short period from about 15% to 5%. There were plenty of people at the time who thought that the consequences of tightening would be much more severe. Possibly with the benefit of our 1970s and 1980s experience we can manage this inflation better. Who knows?

Saturday, November 6, 2010

More on QE2

After the Fed statement on Wednesday, which I discussed here, we have plenty of opinions on what the Fed is up to. Martin Feldstein (actually from the day before the statement, but presumably his opinion did not change with the actual announcement) thinks it's risky, and some of the business press is negative, particularly this guy, who claims Bernanke doesn't know any economics, but apparently his grasp is not the best either. Predictably, Krugman and this two buddies DeLong and Thoma think the asset purchase program should have been larger. DeLong has a particular complaint about the average duration of the the Treasury purchases. Now, in the interest of encouraging people when they say something useful, the guy actually has a point here. The New York Fed page where the relevant information resides is down for maintenance today, but you can find the details of what the Fed plans to purchase here. Very little of these purchases will be of Treasuries with maturities greater than 10 years, and the average duration of purchases will be 5-6 years. On the up side, there is then less maturity mismatch on the Fed's balance sheet (which of course is risky for them), but if QE works in the fashion the Fed hopes it will, then longer-maturity asset purchases would give the Fed much more leverage.

Bernanke felt the need to market QE2 to the public in this Washington Post piece. The basic argument for QE2 we have heard before, which is that the unemployment rate is stubbornly high, and the inflation rate is lower than the Fed would like. Further, Bernanke has some fear of deflation too:
In the most extreme case, very low inflation can morph into deflation (falling prices and wages), which can contribute to long periods of economic stagnation.
So, apparently he is with Jim Bullard in viewing a deflationary state as potentially absorbing and harmful to real activity.

How does QE work? By moving asset prices:
Stock prices rose and long-term interest rates fell when investors began to anticipate the most recent action. Easier financial conditions will promote economic growth.
Is Bernanke worried about the risks?
Although asset purchases are relatively unfamiliar as a tool of monetary policy, some concerns about this approach are overstated. Critics have, for example, worried that it will lead to excessive increases in the money supply and ultimately to significant increases in inflation.

Our earlier use of this policy approach had little effect on the amount of currency in circulation or on other broad measures of the money supply, such as bank deposits. Nor did it result in higher inflation. We have made all necessary preparations, and we are confident that we have the tools to unwind these policies at the appropriate time. The Fed is committed to both parts of its dual mandate and will take all measures necessary to keep inflation low and stable.
Note here that Bernanke is actually paying attention to some monetary quantities, in particular the stock of currency and bank deposits, which is unusual.

Now, I think it is quite possible that we will look back on QE2 as a severe error. In spite of the talk from some quarters about the intervention being too small, this is a very large-scale asset purchase for the Fed, on top of a previous very large purchase of mortgage-backed securities and agency securities. One possibility is that economic growth picks up, of its own accord, reserves become less attractive for the banks, and inflation builds up a head of steam. The Fed may find this difficult to control, or may be unwilling to do so. Even worse is the case where growth remains sluggish, but inflation well in excess of 2% starts to rear its ugly head anyway. Bernanke is telling us that he "has the tools to unwind these policies," but if the inflation rate is at 6% and the unemployment rate is still close to 10%, he will not have the stomach to fight the inflation.

My concern here is that, given the specifics of the QE2 policy that was announced, the FOMC will be reluctant to cut back or stop the asset purchases, even if things start looking bad on the inflation front. Once inflation gets going, we know it is painful to stop it, and we don't need another problem to deal with.

Wednesday, November 3, 2010

FOMC Statement, 11/03/10

Today's FOMC statement was much as expected. Here, the committee says why it wants to do what it is going to do:
Consistent with its statutory mandate, the Committee seeks to foster maximum employment and price stability. Currently, the unemployment rate is elevated, and measures of underlying inflation are somewhat low, relative to levels that the Committee judges to be consistent, over the longer run, with its dual mandate. Although the Committee anticipates a gradual return to higher levels of resource utilization in a context of price stability, progress toward its objectives has been disappointingly slow.
Note that, in spite of the fact that Bernanke and others on the FOMC have been quite explicit about the 2% inflation target, they do not put that in the statement. They also do not say what the unemployment rate is that they view as being consistent with "maximum employment."

Here is the key change in what they intend to do:
To promote a stronger pace of economic recovery and to help ensure that inflation, over time, is at levels consistent with its mandate, the Committee decided today to expand its holdings of securities. The Committee will maintain its existing policy of reinvesting principal payments from its securities holdings. In addition, the Committee intends to purchase a further $600 billion of longer-term Treasury securities by the end of the second quarter of 2011, a pace of about $75 billion per month. The Committee will regularly review the pace of its securities purchases and the overall size of the asset-purchase program in light of incoming information and will adjust the program as needed to best foster maximum employment and price stability.
Thus, the FOMC is retaining the change in policy from the August 10 statement, which would have simply replaced mortgage-backed securities and agency securities that run off with long Treasuries. In addition to that, the Fed is committing (as long as nothing unexpected happens) to the purchase of long-term Treasuries, at the rate of about $75 billion per month, until the second quarter of 2011. This represents an increase over an 8-month period in the stock of outside money of about 45% at an annual rate. Given the experimental nature of this action, it is of course sensible that this is a contingent plan that can be revised in light of new information. I'm as curious as anyone to see what happens.

Tuesday, November 2, 2010

Bizarre-Statement -of- the-Month Club

This month's winner is our friend Laurence Meyer, former Fed Governor and forecaster, previously discussed here, who wins for this statement, from today's Wall Street Journal (sorry if you don't have a subscription):
Former Fed governor Laurence Meyer, now a private consultant, estimated the Fed would need to buy $5.25 trillion of new assets if it wanted to accomplish the equivalent of cutting short-term interest rates by 4.25 percentage points, about what Mr. Meyer's model indicates is needed now.
This is funny on several dimensions. First, the guy has a model that is giving him a policy conclusion: short term interest rates should drop by 4.25 percentage points, which obviously is not feasible. Second, after making that statement, he's asking us to put some confidence in his prediction that what is "needed now" is $5.25 trillion in asset purchases by the Fed. Third, what exactly does he think that a more-than-tripling of the size of the Fed's balance sheet will accomplish and why is a $5.25 trillion asset purchase somehow equivalent to a non-feasible decrease of 4.25 percentage points in the fed funds rate?

Monday, November 1, 2010

Why Some People Should Not Run Central Banks

Sorry for beating a dead horse here, but these two posts were hard to resist. Krugman posts this piece on his blog. In particular, he says:
In particular, an individual businessman, no matter how brilliant, never has to worry about the fact that total income equals total spending, so that if some people spend less, either someone else must spend more, or aggregate income must fall.

This is why we have a field called macroeconomics.
Actually, this is why we have national income accounting.

Next he posts this piece. Krugman is going to tell us what he would do if he were in Bernanke's shoes. First he would commit to a price level target that implies average inflation of 5% per year. Not unexpectedly, he wants more inflation than Bernanke does, but of course we have no good reason to prefer 2% to 5%, as far as I can tell. Now, the key thing is that he doesn't inform us about how he hopes to achieve this. He says:
The sad truth, of course, is that the chance of actually getting anything like this are no better than those of getting an adequate fiscal stimulus — at least for now. QE as currently contemplated is mild mitigation at best. What one has to hope is that as the reality that we’re in a liquidity trap sinks in...
It's not clear what is "currently contemplated," but of course we will find out in a couple of days. In any case, Krugman doesn't tell us what he would actually do to get the 5% inflation. People at the Fed of course know about liquidity traps, which is why they are talking about purchases of long Treasuries, rather than purchases of T-bills. Does Krugman think that more than what is "currently contemplated" would achieve the result he wants? If so, how much? Please enlighten us.

Sticky Prices and Keynesian Economics

Since Keynes wrote his General Theory, other economists have tried, in various ways, to formalize what Keynes appeared to have had in mind. Hicks constructed the IS-LM model, which is a static framework in which prices are fixed in nominal terms. In the 1980s, Mankiw and Blanchard and Kiyotaki, among others, thought about Keynesian menu-cost models in which it is costly to change prices. John Bryant,Peter Diamond, Cooper and John, and the authors represented in this JET volume studied equilibrium coordination failure models with Keynesian features. Mike Woodford, among others, modified the neoclassical growth model, using some of the insights from the menu cost literature, to develop sticky-price New Keynesian models.

The followers of Keynes are sometimes motivated strongly by belief. You can see this, for example, in Ball and Mankiw's Sticky Price Manifesto. Some Keynesians look at a contractionary episode - the Great Depression or the recent recession - and cannot see anything other than Keynesian economics to explain it. The "fundamentals" in 1933 looked about the same is in 1929, so how could real GDP have been 40% lower, except as the result of a "deficient demand" phenomenon? There is also empirical evidence, for example on the stickiness in prices. Prices of goods are certainly much smoother than the prices of assets, and this forthcoming chapter from the new Handbook of Monetary Economics summarizes what is known about the empirical regularities in the changes in prices over time and across goods. Another interesting regularity is that much of the variability in the real exchange rate between two countries, even in close proximity (e.g. the U.S. and Canada), is explained by variability in the nominal exchange rate. Thus, it seems to matter how prices are denominated - e.g. in U.S. or Canadian dollars.

What do we want from an economic model? First, we want it to explain something to us. Why are people unemployed? Why do economies grow? Why is there inflation? Second, we would like to ask the model questions. In macroeconomics, some of the key questions are policy questions. How, if at all, can fiscal and monetary policy be used to make us better off? The answers to the first set of questions can have a lot to do with the answers to the second set. Once I understand what causes inflation and why it is bad for us, this can tell me how monetary policy should be conducted to control it.

Now, the class of coordination failure models I mentioned above explains something. These models tell us that aggregate fluctuations can arise due to people coordinating on extraneous information. Further, these fluctuations are generally inefficient. What should we do about this problem? Typically, fiscal and monetary policy rules in these models can be set in ways that eliminate undesirable equilibria. Coordination failure models are certainly Keynesian. Indeed, one can find passages in the General Theory that are essentially coordination failure stories. However, in spite of the large coordination failure literature, and some success in fitting these models to the data, most Keynesians are currently uninterested in coordination failures. This may be because of the technical demands required to work with these models. However, full-blown New Keynesian models appear no less demanding. To me, it's a puzzle.

The New Keynesian models in wide use now typically rely on Calvo pricing (a form of time-dependent pricing), whereby monopolistically-competitive firms receive random opportunities to change prices. Upon receiving an opportunity to change its price, a firm does this optimally (taking into account that it may not be able to change it again for a while), but is otherwise stuck with the price it charged last period. Some New Keynesian models go deeper and use state-dependent pricing, under which there is a menu cost, and a firm changes its price when it is optimal to do so. Time-dependent pricing and state-dependent pricing are just special cases of the same model. With time-dependent pricing the menu cost is random - it is either zero or infinite. With state-dependent pricing the menu cost is constant.

Neither state-dependent or time-dependent pricing actually explains why prices are sticky. Why should it be costly to change a price? This can't literally be the cost of posting a price schedule - surely that is trivial relative to other costs of producing and selling a good or service. Keynesians must have something else in mind, if they have thought about it. But what is it? And surely what is causing the price stickiness has to matter for whether and how policy should fix any inefficiencies arising from sticky prices.

The typical argument seems to be that actually explaining the sources of price stickiness is too hard - otherwise someone would have solved the problem. As the argument goes, a useful shortcut is to just fix the prices, perhaps in a Calvo fashion, and go from there. With the world falling apart and large numbers of people unemployed, why should we have to wait while the theorists work out the details? Of course, the problem could be that, once we know all the details, we might change our minds about how the policies work.

What would a theory of price stickiness, or more appropriately a general theory of pricing, look like? First, we have to have a model in which sellers want to quote prices in units of "money." While we have models of media of exchange, models of the unit of account do not exist, to my knowledge. In any monetary model I know about, the numeraire could be anything. It is irrelevant whether we quote prices in units of money, goats, or soccer teams. Clearly though, how we write contracts and quote prices has a lot to do with what we use as a medium of exchange. We want an environment that will imply that contracts with few contingencies are optimal, and possibly the nominal contracts can then be derived as optimal risk-sharing arrangements. Clearly, though, you have to be a long way from complete Arrow-Debreu contracts. The frictions will be critical.

Now, once we are past the nominal-contract hurdle (a very serious one at that), we have to worry about the stickiness. Suppose that I am selling toothbrushes at a particular location. There are many ways to determine how many toothbrushes I sell at what price. I could haggle with each customer who comes in the door concerning how many toothbrushes I will exchange for how many dollars. I could specify a time each day when I will auction off so many toothbrushes, possibly selling them individually using an English or Dutch auction. I could also post a price and sell toothbrushes at that price to whoever comes in the door. Of course we all know that it is not efficient to haggle over a toothbrush, or to sell toothbrushes in an auction (unless it's Elvis's toothbrush). However, haggling may be efficient when I am buying a house, and an auction may be efficient when I want to sell an idiosyncratic object. Surely, part of what the theory needs to tell me are the characteristics of a good or service that determines how it is exchanged - using price posting, an auction, one-on-one haggling, or some other mechanism.

Now, in the case where prices are posted, why would firms want keep a posted price unchanged for months at a time. It could be that it just does not matter much. Maybe the firm has other more important things to think about (inventory control, shoplifting, shirking employees) than the price of toothbrushes. Maybe the firm is trying to minimize effort for repeat customers. If I know that firms (such as the grocery store, where I buy the same goods with some regularity) are changing prices frequently, as a buyer I will have to reoptimize more frequently to make sure that I am purchasing the optimal basket of goods. Maybe I would rather buy at a store where the prices change infrequently.

Now, what could go wrong here? How are firms going to get the prices wrong, so that the government needs to step in to fix things? Possibly with infrequent price setting, we get phenomena like those in Calvo-pricing setups, where firms make staggered price adjustments, and relative prices are distorted. This is far from clear, but suppose that relative price distortions are the difficulty. Now, if the government has all the information, this problem is easy to solve. The government knows what the correct prices are, and it forces firms to price correctly. Of course that seems silly, as it is unrealistic to think that the government has access to enough information to accomplish this. We all know that market economies solve very complex allocation and pricing problems that it would be impossible for a central planner to solve.

If the relative prices are distorted, the government must necessarily rely on some indirect mechanism for dealing with this. If the prices are distorted due to inflexibility, why not subsidize flexibility? But how would this be done? Do some sectors of the economy need larger subsidies than others? Could firms game the system by making a price change today and reversing it tomorrow? How does the government monitor all these price changes? The more I think about this, the more unworkable it seems.

Now, the solutions that Keynesians, new and old, have come up with for solving the relative price distortion problem seem odd. First, Old Keynesians tended to (and tend to) focus on fiscal policy remedies. Why should the provision of public goods and services be an important element in correcting a relative price distortion? I know how typical Keynesian models work, but when I think more deeply about the pricing problem, this does not make sense. Second, New Keynesians argue for the use of monetary policy (within the constraint determined by the zero lower bound on the nominal interest rate) to correct the problem, when it seems that indexation would be feasible, and in fact optimal for the firms involved.

We need a serious theory that explains the behavior of the prices of goods and services. Such a theory is necessary, as it is far from clear that the observed behavior of prices implies some market failure and, even if it did, it is also not clear that standard Keynesian prescriptions would solve the problem. In the absence of the theory, I think there are good reasons to be skeptical about what Old and New Keynesians have delivered thus far.