A blog called Canucks Anonymous has a new post up that exemplifies everything wrong with modern macroeconomics.  The basic argument is that we market monetarists have fled in terror from Lawrence Ball’s demonstration that NGDP targeting would lead to greater economic instability:

Greg Ip refers us to this excellent paper from Laurence Ball which shows that targeting NGDP is a pretty lousy idea. One can see immediately that Ball must be spot on from that fact that none of the usual suspects, Scott Sumner or Nick Rowe for example, has even acknowledged the existence of the paper even though Nick actually did a post responding to Ip’s article!

I don’t respond to everything I read, as I don’t have time.  And I’m especially unlikely to respond if the macroeconomist I most respect (Bennett McCallum) thinks the argument has no validity.  But let’s take a stab at it anyway:

Ball writes down the simplest model you can imagine:

y = -b*r(-1) + l*y(-1) + e
pi = pi(-1) + a*y(-1) + n

Here, y is real income, pi is inflation (both as deviations from trend) and e and n are shocks. The notation y(-1) denotes one period lagged income, etc.

I’m already highly irritated.  What makes this canuck think that “the simplest model you can imagine” has any bearing on whether NGDP targeting is a wise policy for our highly complex economy?  Theoretical macroeconomists love to play with their little toy models, but unfortunately these models don’t actually describe the world we live in.  In their models the IS curve slopes downward.  Output depends on real interest rates.  Inflation depends on output gaps. I don’t know about you, but that’s not the world I inhabit.

Haven’t we seen ultra-low real interest rates for several years?  Where is the robust growth?  I know, the “error term” takes care of all these pesky little problems.  And why is inflation not falling right now?  That’s right, the error term explains why.  And why did prices rise sharply after March 1933, despite the biggest output gap in US history?  Yes, the error term explains why.  I prefer models where all the really big phenomena that I’d like to understand don’t have to be explained away by references to error terms.

Balls’ model has monetary policy impact Y with a one period lag and P with a two period lag.  In fact, monetary policy affects P with a lead.  Or if you consider expectations of future monetary policy to be “policy,” then the effect is contemporaneous–as when rumors of QE2 drove up commodity prices, and hence the headline CPI.

I don’t think the real interest rate measures monetary policy.  To the extent that interest rates matter at all, I believe it is the nominal rate minus expected NGDP growth.  But I’d rather just leave interest rates out of my “simplest imaginable model,” and stick with NGDP shocks and sticky wages as my explanation of changes in real output.    And I’d rather model inflation (or better yet NGDP growth rates) via expected future NGDP, and hence expected future monetary policy.  The hot potato effect.  I’m not saying the output gap plays no role, but it’s much more complicated than the second equation implies.

Come back to me later when Ball has a model in which output fluctuations are caused by NGDP shocks and sticky wages, not real interest rates.

I also found this amusing:

You can’t cheat this, inflation targeting is simply better.

So inflation is best.  Or is Canuck saying that inflation is best in Ball’s model?  I can’t quite tell.  Canuck doesn’t seem to realize that inflation is something that exists only in the minds of economists, not out there in “reality.”  Or perhaps I should say there are as many inflation rates as there are economists.  There is the US CPI inflation, which is built on the assumption that using housing prices have risen 7.5% in the past 5 years.  Is that the inflation that Canuck wants us to target?  Then there is the inflation rate that uses Case-Shiller data, which shows housing prices down 32% in the past 5 years.  Given that housing is 1/3 of the CPI, it might be nice to know which of those two inflation rates we should target model.  But in my search of Ball’s model I didn’t find any answer to that question.  Indeed there are a million inflation rates; are we to believe that all of them are “optimal” policy targets?  Is that what Ball’s equations prove?

I don’t work with toy models; I try to stay grounded in the real world.  I notice that periods of above 5% NGDP growth (like the 1970s) are viewed as periods where monetary policy was too expansionary.  And when NGDP plummets, like in the 1930s, money was too tight.  And when NGDP grew at a steady rate of 5%, we achieved the best macro performance in history, the so-called “Great Moderation.”  And when we let NGDP collapse in late 2008 and 2009, we had a very severe recession.

I also notice that those who slavishly follow inflation targeting seem to often give the “wrong” advice.  And I don’t just mean wrong in the sense that I don’t agree, but also wrong in the sense that almost all sensible economists, even those who don’t believe in NGDP targeting, would beg to differ.  Like right now in the UK, when an inflation target would call for much tighter monetary policy.  Or indeed even in the US, where the implications of current inflation are ambiguous, but most economists (including I’d guess Ball) think that more demand is obviously desirable.

And that’s not even touching on the political problems associated with inflation targeting.  In theory inflation and NGDP are similar targets; whenever either aggregate is too low, the Fed calls for an increase.  But as I pointed out in this paper, calling for higher inflation led to a political firestorm in late 2010.  In contrast, most Americans would be accepting of a policy aimed at higher NGDP.  (An argument I have made many times, and which was recently picked up by Paul Krugman.)  But the messy complications of the real world don’t matter to those living in a world of pure mathematics.

I make no apologies for ignoring these little toy models, and having my policy analysis incorporate a complex mixture of politics, macroeconomic history, well-established basic economic principles, and logic.

HT:  Bill Woolsey

Update: Josh Hendrickson sent me some comments that might be a more effective rebuttal:

Four points:

1. In my paper on nominal income and the great moderation, I show that a strong responsiveness to nominal income reduces the variability and the output gap in both a New Keynesian model and McCallum’s P-bar model.

2. It’s hard to assess nominal income targeting when nominal income isn’t defined in the model. Monetary transmission has nothing to do with nominal income in these models so we shouldn’t expect nominal income targeting to be optimal.

3. A long standing reason to support nominal income targeting is because there is uncertainty about how fluctuations in nominal income are divided between output and prices. Ball’s model assumes a specific process for inflation.

4. I continue to go back to the point I made regarding Svensson’s paper. The transmission of monetary policy in the model is through output in the first period and only effects inflation with a one period lag. This is the source of instability and I don’t think its a realistic assumption.

Some days I want to just shoot myself, like when I read the one millionth comment that easy money will hurt consumers by raising prices.  Yes, there are some types of inflation that hurt consumers.  And yes, there are some types of inflation created by Fed policy.  But in a Venn diagram those two types of inflation have no overlap.

So here’s my plan.  Beginning tomorrow, November 1st, I will ban all discussion of inflation from the comment section. I won’t respond to questions on inflation.  (God knows how Bob Murphy will react to this—something tells me it won’t make me look good.)

Before everyone starts whining, I am about to provide a substitute language for you to use, and tell you when to use it:

1.  Let’s start with the easiest type of inflation to consider, and the only one people really care much about—supply side inflation.  Suppose the AS curve shifts to the left because of a cutoff in oil production, crop failures, bad government tax and regulation policies, etc, etc.  Real GDP will fall, if we do nothing to aggregate demand.  And prices will rise.  People think it is the price rise that is making them worse off, but that’s an illusion; it’s really the drop in RGDP.  How do we know?  Because consider the case where the Fed responds with tight money, which shifts AD far enough to the left to prevent any inflation from the adverse supply shock.  In that case there are two possibilities; the drop in real GDP and real living standards would be just as bad (if money is neutral) or it would be even worse, if money isn’t neutral.  It is the fall in RGDP that is the key problem, and any price change is incidental to what’s really going on.  So if you ever envision an inflation problem that makes consumers worse off, please don’t call it “inflation,” call it “falling real incomes.”

2.  Now let’s turn to demand-side inflation, which either makes people better off in the short run (via higher RGDP, higher real incomes), or has no effect (if money is neutral.)  Here there are many cases to consider:

2.a.  Some argue that low inflation makes a liquidity trap more likely.  But as we can see by comparing Japan, China, and Hong Kong, mild deflation creates liquidity traps only when real growth is persistently low.  The problem isn’t low inflation, it’s low expected NGDP growth.  That’s because real interest rates are strongly correlated with real GDP growth, and of course expected inflation in nominal rates is perfectly correlated with the expected inflation component of NGDP growth.  So from now on please talk about the danger of low NGDP growth leading to a liquidity trap, not low inflation.  China had deflation in the late 1990s, but fast RGDP growth—hence no LT.

2.b.  This also applies to the phony tears people shed for the savers hurt by inflation.  Let’s assume savers buy long term nominal bonds.  Those bonds promise a fixed amount of money, at specified futures dates.  The standard argument is that inflation hurts savers by reducing their real return on bonds.  But savers don’t care about the nominal interest rate minus the inflation rate, they care about the nominal interest rate minus the per capita NGDP growth rate.  I’ll give you an example.  Years ago the British government indexed the initial starting point for retirement pensions to the cost of living, not average wages (as we do.)  The Thatcher reforms led to real increases in living standards iGn reat Britain, and so over time the living standards of retirees fell further and further behind living standards of the employed (who received nominal wages increases that exceeded inflation.)  Eventually the old-timers looked at the flashy lifestyles of their younger neighbors, and revolted.  The UK government was forced to change the indexing scheme.  People don’t care about real incomes they care about how they’re doing relative to their neighbors.  If NGDP rises faster than expected, then a bondholder is paid back a smaller share of national income than he anticipated when he bought the bond.  And that hurts.

2.c.  The Fisher effect applies to nominal interest rates minus NGDP growth, not minus inflation.  Some Keynesians fear that the Fed can’t stimulate the economy, because it finds it politically difficult to increase the expected rate of inflation.  But they don’t need to increase the expected rate of inflation, just the expected rate of NGDP growth (and the SRAS is pretty flat right now.)  When expected NGDP rises you get more investment whether firms expect more inflation, or more RGDP growth.  So just shoot for more nominal growth.

2.d.  Some people say inflation and deflation are bad because wages are sticky.  A sudden bout of deflation will raise real wages, and lead firms to lay off workers.  But that’s only true if the deflation comes from falling NGDP growth expectations.  Suppose it is the “good deflation,” produced by rising productivity.  Then if wages are sticky the deflation raises living standards.  This occurred during the 1927-29 boom, when the price level fell in America.  Of course that was followed by a “bad deflation,” sharply falling NGDP during 1929-33.  So rather than talk about good and bad deflation, let’s just talk about what we really mean, rising and falling NGDP.

2.e.  Some people think the Fed should target inflation.  When you mention oil shocks they say “well that’s an exception, I favor a flexible inflation target that allows prices to rise during supply shocks.”  OK, but then why not just target NGDP, so you don’t have to make exceptions?  Why totally confuse the public?

2.f.  One of the supposed costs of inflation is the excess tax on capital caused by the fact that capital gains taxes and taxes on interest are not indexed.  But that’s really a problem of high nominal returns on capital, not high inflation.  You say the two are correlated?  I say they’re even more correlated with NGDP growth.  So let’s talk about how rapid NGDP growth imposes inefficient tax burdens on capital.

2.g.  Menu costs?  Maybe, but it’s ambiguous, which is not enough to overcome my presumption for NGDP growth.  After all, many economists think the biggest menu costs apply to wages, which seem very hard to adjust.  Costly strikes result from attempts to adjust wages.  Or workers occupy the capital building in my hometown of Madison.  And wages are arguably more closely linked to per capita NGDP than inflation.  In early 2008 inflation rose rapidly while NGDP did not.  Wages remained well contained.

2.h.  The inflation tax from printing money?  It comes from the opportunity cost of holding cash, which is the nominal interest rate.  And the nominal interest rate depends on NGDP growth.  (I should add that it also depends on lots of other things, like economic slack and budget deficits.  But those other things are also “not inflation.”)

2.i.  Inflation can reduce the burden of the national debt?  No, NGDP growth reduces the burden of the national debt.  Does unexpected disinflation trigger debt crises?  No, it’s unexpected falls in NGDP that trigger debt crises.

To conclude:  If you are about to type the word “inflation,” please stop.  If you have in mind something that implies lower living standards, please type “falling real incomes.”  If it is a demand-side inflation (all the items in category 2 above) then type “NGDP growth” or “nominal income growth.”

So you have 12 hours to convince me to rescind this ban, before it goes into effect.  If you cannot come up with a scenario where I need the word “inflation,” then it will be banned.

Happy Halloween!

I was reading a new book by Tim Congdon and came across this interesting quotation, discussing the flaw at the heart of New Keynesian economics:

In the New Keynesian schema, it [the interest rate] became in effect the only policy instrument, the factotum of macroeconomics.

Why did we have to end up with the worst possible policy instrument?  The only instrument that has a zero lower bound.  We could have chosen the monetary base, or the trade-weighted exchange rate, or the price on NGDP futures, or the TIPS spread, or the price of zinc.  But no, we had to pick nominal interest rates.

We ended up with a steering mechanism that locks up just when you most need it to work.  Even worse, central banks have so fallen in love with the mechanism that they can’t seem to shift to a different target.  Instead we end up with never-ending attempts to manipulate interest rates, even when short term rates have hit zero.  Promise to hold rates at zero for X number of years.  Or attempts to lower longer term rates.  Or to reduce interest rate risk spreads.

These proposals have a slightly pathetic quality, because (as Nick Rowe reminds us in this recent post) a policy that is expected to be successful will actually raise nominal rates.  So you have the Fed announcing that the goal of QE2 was to lower long term rates, and then when they start rising the Fed announces that the policy must be working.  It’s a wonder the Fed still has any credibility.  How’s this for communication?

WASHINGTON “” The Federal Reserve made a rare promise on Tuesday to hold short-term interest rates near zero through at least the middle of 2013, in a sign that it has all but written off the chances of an expansion strong enough to drive up wages and prices. . . .

By its action, the Fed is declaring that it, too, sees little prospect of rapid growth and little risk of inflation. Its hope is that the showman’s gesture will spur investment and risk-taking by convincing markets that the cost of borrowing will not rise for at least two years.

The Fed’s statement, with its mix of grim tidings and welcome aid, contributed to wild market oscillations as investors struggled to make sense of the economy and the path ahead.  (emphasis added)

Well that will certainly whip up those animal spirits!

The zero rate bound doesn’t occur for variables like the monetary base.  Some Keynesians argue that this doesn’t matter; open market purchases become ineffective when rates hit zero, as one is merely swapping one asset for another.  But that’s not true, as a permanent increase in the base is inflationary.  And if the Fed had already been using the base, it would have been able to continue signaling future policy intentions as if nothing had happened.  In contrast, once rates hit zero the Fed can’t signal anything with changes in interest rates, because it can’t change interest rates.

Of course the New Keynesians also insist that interest rates are the transmission mechanism.  Not so.  When there’s a big apple crop, NGDP in apple terms soars.  No need to invoke interest rates.  Ditto for a big crop of Federal Reserve Notes.  And the mechanism that causes nominal shocks to have real effects is sticky wages and prices, not interest rates.  When I point out that rates hardly budged during the most expansionary monetary policy in US history, Keynesians start talking about rates falling relative to their Wicksellian equilibrium value.  Yes, but that’s pretty much true by definition, and true for any price.  If the Wicksellian equilibrium zinc price is the one consistent with 2% inflation, then the Fed can boost inflation above 2% if and only if it can raise zinc prices above their Wicksellian equilibrium.

The next meeting will be a big test for the Fed.  I don’t expect miracles, but I’d hope for at least some sign that they understand there’s nothing more they can do to generate recovery by fiddling with interest rates.  They need to indicate that they are at least attempting to communicate in some other language.

Most people seem to assume nothing major will be done until the three hawks leave in January.  In fact, Fed stimulus would be more credible if they could get at least one of the three to vote for it.  It seems to me that Kocherlakota offers the best hope.  At times he seems to indicate that he’s aware of the unemployment problem, but doesn’t like the lack of a nominal anchor in open-ended promises, such as two years of near-zero interest rates.  He might be willing to support stimulus, as long as there is an explicit promise to maintain prices or NGDP along a particular trajectory.  If I’m right, it’s quite possible that the fate of 100,000s of unemployed people might depend on what he decides.

It’s no way to run monetary policy.  We should have an explicit 5% NGDP target, and let the market set the money supply and interest rates.  But you go into recession-fighting with the Fed you have, not the Fed you wish you had.

Steve Waldman, aka Interfluidity, has a sort of critique of market monetarism:

Self-fulfilling expectations lie at the heart of the market monetarist theory. A depression occurs when people come to believe that income will be scarce relative to prior expectations and debts. They nervously scale back expenditures and hoard cash, fulfilling their expectations of income scarcity. However, if everybody could suddenly be made to believe that income would be plentiful, everyone would spend freely and fulfill the expectations of plenty. The world is a much more pleasant place under the second set of expectations than the first. And to switch between the two scenarios, all that is required is persuasion. The market-monetarist central bank is nothing more than a great persuader: when “shocks happen”, it persuades us all to maintain our optimism about the path of nominal income. As long as we all keep the faith, our faith will be rewarded. This is not a religion, but a Nash equilibrium.

I’m not very well versed in game theory, but this doesn’t seem right to me.  Yes, expectations are important, but ultimately the path of NGDP depends on decisions by the central bank about the future path of monetary policy.  In late 2008 the public saw that the Fed was going to let the future (post-liquidity trap) monetary base (assuming no IOR) fall well below previous estimates.  It now seems the markets were correct, the Fed has no intention to go back to the previous NGDP trend line, even though everyone agrees they could do so once nominal rates rise above zero.

The Fed isn’t some sort of mesmerist, they are quite clumsy.  The markets are far more sophisticated, often signaling Fed moves before the Fed realizes that it needs to move.  The Fed matters for three reasons:

1.  They have a monopoly over the supply of base money.

2.  Base money can be produced at near zero cost and in nearly unlimited quantities.

3.  Base money is the medium of account.

These three facts give the Fed control over the long run path of nominal aggregates like prices and NGDP.  Expectations are important because current aggregate demand depends partly on future expected AD, which depends on the future expected path of the monetary base.  But it’s not self-fulfilling expectations, it’s forecasts about future Fed policy that may or may not turn out to be correct.

Steve continues:

I have a Minsky/Mankiw theory of depressions. The economy is divided into two kinds of people, spenders and savers. Perhaps some people lack impulse control and have bad character, while others are patient and provident. Perhaps structural inequality renders some people hungry but cash-constrained, while others have income in excess of satiable consumption. Let’s put those questions aside and just posit two different and reasonably stable groups of people. Variation in aggregate expenditure is due mostly to changes in the behavior of the spenders. Savers spend at a relatively constant rate and save the rest. Spenders spend whatever they can earn or borrow, which varies with the level of wages, the cost of servicing debt they’ve accrued in the past, and the availability of new credit.

In this world, a central bank that targets something “” NGDP, inflation, whatever “” doesn’t regulate behavior via expectations. Instead, the central bank regulates access to credit and wages. When the economy is “overheating”, the central bank raises interest rates to increase debt servicing costs, tightens credit standards to diminish new borrowing, and if absolutely necessary squeezes so hard that a recession reduces spenders’ wages via unemployment. When the economy is below potential, the central bank reduces interest rates and relaxes credit standards, encouraging spenders to borrow and leaving them with higher wages net of interest payments.

This is a pretty good gig, it works pretty well, especially when the marginal dollar of expenditure is borrowed and easily regulated by the central bank. But if there are lower bounds on interest rates and credit standards, the scheme is not indefinitely sustainable. Even when spenders hold consistent, reasonably optimistic expectations about the economy, it becomes continually more difficult to persuade them to maintain their level of spending. The cost of debt service grows as their indebtedness grows, reducing their ability to spend. New borrowing becomes more difficult as wages are dwarfed by liabilities. Individuals become more nervous that some blip in their complicated lives will leave them unable to meet their obligations. In order to hold expenditure constant, interest rates must fall, credit standards must loosen, the value of spenders’ one consumption good that survives as pledgeable collateral “” their homes “” must be made to rise. Stabilizing expenditure requires continual easing. Any sort of lower bound provokes a “Minsky moment”, as expenditures that can no longer be sustained unexpectedly contract, rendering maxed-out spenders unable to service their debts.

Keynes thought that he’d developed a “General Theory,” but Hicks and Friedman argued that the only really distinctive innovation in the GT was the zero rate trap.  I fear Steve has done the same.  He’s describing what seems to be a sort of general theory of fiat central banks, but it all hinges on the idea of a zero rate trap (which I think he and others misunderstand.)  Recall that money is roughly superneutral, i.e. changes in the trend growth rate of inflation and NGDP should leave real interest rates unchanged.  That means a country can easily avoid a zero rate trap with a suitably high rate of inflation, even using the clumsy tools of modern NK central banks.  Australia has a trend rate of NGDP growth of about 7%, and thus much higher trend interest rates than the US.  They never fell to the zero bound, and thus avoided the 2008 recession, indeed the 2001 recession as well (and that success can’t be explained away with high commodity prices.)

The implication of Steve’s model is not that we should abandon market monetarism but rather that we should have a higher trend rate of NGDP growth, so that we never go up against the zero bound.

Now I happen to disagree with even that argument.  A 5% NGDP target, level targeting, is plenty high enough to avoid the zero bound.  The problem is that the Fed didn’t doing level targeting, hence NGDP fell 9% below trend in mid-2009, and even further below since then.  In Waldman’s world the economy plunged because the housing bubble popped, and thus the equilibrium rate (needed to preserve stable NGDP growth) fell below zero.  That’s possible, although I think it much more likely that rates fell because the markets correctly realized the Fed was going to allow NGDP to plunge, and then stay at much lower levels.  But even if I am wrong, I would recommend the Fed accommodate the demand for currency and reserves for a 5% NGDP growth path, no matter how large.  It seems unlikely that it could ever exceed the national debt, albeit not theoretically impossible.  If it did, then I’d favor negative IOR, or buying foreign debt.

Steve continues:

The market monetarists might retort that a sufficiently determined central bank, if given license to lend and purchase assets as it sees fit, can always meet a nominal spending target, and therefore can always set expectations of nominal demand. That may be true. But in the context of an economy structurally resistant to increasing expenditure, expectations of stable nominal income become equivalent to expectations of continual central bank expansion. NGDP expectations can be maintained, if and only if the central bank demonstrates its willingness to continually intervene.

If intervention will be frequent and chronic, precisely what instruments the central bank intends to use becomes a matter of great public concern, rather than a technocratic detail best left to professionals. Central banks may significantly shape patterns of consumption and investment by choosing to whom they are willing to lend and on what terms.

I see all sorts of problems here.  Once again, the “structurally resistant” point seems to confuse nominal and real variables.  The Fed can choose whatever nominal trajectory it likes.  After we left the gold standard the nominal trajectory became steeper, albeit quite unstable (except 1982-2007).  The Fed continually intervened during the Great Moderation by purchasing government debt.  This was quite uncontroversial, and didn’t have important allocative effects.  The Fed wasn’t picking winners and losers.  A continually rising monetary base, century after century, works just fine.

I’d like to end on a positive note, however.  Steve does point out correctly that interest rates seem to show a secular decline in recent decades.  I would add that this even may be true of real interest rates (and I expect it to continue for Cowenesque Great Stagnation reasons.).  One can envision a scenario where zero nominal rates become somewhat more frequent at relative low NGDP growth trajectories (5%, or even more so for 3%, Woolsey’s proposal.)  If that occurred the government would face two choices.  Set a higher nominal growth target, or be prepared to do much larger asset purchases than during the Great Moderation.  I lean toward the much larger asset purchases, but if America eliminated taxation of capital (as we should) then the argument for higher trend inflation would mostly evaporate.  In that case we might want to follow the Aussies.  Of course under our current regime (non-level targeting) we don’t know how to operate at the zero bound.  So if we aren’t going to do market monetarist reforms, I’d suggest going to the Australian NGDP trajectory right now, even without tax reform.  “It takes a lot of Harberger triangles to fill an Okun gap;” and even worse, output gaps lead to really bad public policies that create more Harberger triangles.

PS.  Steve also argues for “helicopter drops.”  But as the Japanese have learned, even that doesn’t work of you’ve got a perverse central bank, sending out the wrong signals about future monetary policy.  In the end you need the right expectations.  And that can only happen with the right monetary policy.

Here’s Paul Krugman:

I would submit, by the way, that the quasi-monetarists “” QMs? “” have actually backed up quite a bit on their claims. They used to say that the Fed can easily and simply achieve whatever nominal GDP it wants. Now they’re more or less conceding that the Fed has relatively little direct traction on the economy, but can nonetheless achieve great things by changing expectations. That’s pretty close to my original view on Japan. But changing expectations in the way needed is hard, especially when the Fed (a) faces massive sniping from the right and (b) has a number of hard-money obsessives among its own officials.

Of course we’ve always focused on expectations; that’s why we’re called market monetarists.  When I wrote an open letter to Paul Krugman in 2009, I discussed the need for QE, lower IOR, and a NGDP target, level targeting.  That’s still my view.  If Krugman thinks we’ve moved in his direction, it’s because he never really understood our views until now.  Come to think of it, there’s a lot of evidence that he didn’t understand what we were saying.

In fairness, market monetarists have been supportive of QE2, as that seemed much more politically feasible than NGDP targeting, level targeting.  Indeed even more feasible than price level targeting.  So it may have appeared we thought that was the optimal policy, at least at first glance.  But Krugman also supported QE2.

My areas of disagreement with Krugman have always been over fairly subtle questions.  He sees Japan as an example of an expectations “trap,” I see the problem as simply a central bank that has the wrong policy objective–an excessively low inflation target.  Not a central bank “trapped” by forces beyond its control.  He sees expected inflation as being needed; I see higher expected NGDP growth as being needed (but not necessarily anything close to 4% inflation.)  He sees monetary stimulus working through the real interest rate transmission mechanism, I see the interest rate as a sort of epiphenomenon, and instead see the mechanism as excess cash balances driving NGDP higher in the long run (hot potato effect), and expectations of future increases in NGDP driving current asset prices and current AD in the short run.  Other market monetarists obviously have diverse views on this question.

Our areas of agreement have always been much more important:

1.  We both see a need for much more demand stimulus.

2.  We both see temporary currency injections as useless and monetary aggregates as unreliable policy indicators.

3.  We both believe that the key is to raise expectations of future monetary stimulus.

4.  We both agree (I think) that if QE2 worked at all, it probably worked partly by sending a signal regarding the Fed’s future policy intentions.

5.  We both like Gauti Eggertsson’s work on the Depression (AER 2008.)  If Krugman thinks I’m a Johnny-come-lately to his expectations hypothesis, he might take a look at the 3 papers I wrote that Gauti cited in 2008, which argued (among other things) that the Fed’s 1932 open market purchases failed (contrary to the claims of Friedman and Schwartz), and that the reason they failed is that the purchases were viewed as temporary because of the constraints of the gold standard.

One final point.  There is a difference between Krugman and the market monetarists in terms of how we approach the thought experiment of an increase in the money supply at the zero bound.  He assumes the increase is expected to be temporary, and we often assume it’s permanent.  Thus the debate is often people talking right past each other, as even Krugman agrees that a permanent monetary injection would be expansionary.  In favor of our assumption, standard quantity theory models generally assume a permanent, one time money supply increase, when thinking about the effect on the price level.  To see why, consider how Irving Fisher or Milton Friedman would have reacted if told the Fed planned to double the money supply, and then remove the new money 12 months later.  Would Fisher and Friedman have predicted that the price of homes would double, but then fall back to the original level 12 months later?  Obviously not.  On the other hand neither Fisher nor Friedman seems to have fully understood the importance of the distinction between temporary and permanent monetary injections.  So Krugman’s 1998 paper did a great service by showing just how important this distinction really was.

But here’s something I never see Krugman discuss.  Expectations aren’t just important at the zero bound, they are always important.  If interest rates are 5%, and the Fed suddenly announces that they will immediately double the money supply, but then remove the new money a year later, there is little impact on prices.  So the expectations approach doesn’t merely challenge the naive QTM models at the zero bound, it does so at positive interest rates as well.  Expectations always matter a lot.  So why does the problem seem more severe at the zero bound?  Nick Rowe has pointed out that the Fed becomes mute at the zero bound.  As long as rates are positive, they can send signals about their future monetary policy intentions by adjusting short term rates.  Once rates hit zero, they don’t know how to communicate their policy intentions to the public, and the price level becomes unmoored, drifting around aimlessly.  At least until they find other tools (like QE) to communicate their future policy intentions.

PS.  I’m having lots of computer problems, so blogging may slow down for a while.