Scott Sumner  

Market monetarism is not (primarily) about NGDP targeting

Closed borders and the liberty... Meer vs. Galbraith on the $15 ...

I was very excited to read about a new San Francisco Fed study that supported one key tenet of market monetarism. A number of commenters seemed confused as to why it was a big deal. To see why, we need to consider what market monetarism is all about.

I think it's fair to say that my blogging is associated with the proposal for NGDP level targeting. It's true that I favor that policy, as do many other market monetarists. But I also think the Fed did a pretty good job during 1983-2007, without an explicit NGDP target. And I believe that it's quite possible that a closely related policy, such as targeting total nominal labor compensation per working age adult, might be slightly better.

In fact, market monetarism is mostly about another set of ideas:

1. Expected NGDP growth is the best measure of the stance of monetary policy.
2. Monetary policy causes changes in NGDP.
3. Market indicators are better than structural models at predicting NGDP growth.
4. A fiat money central bank never runs out of ammunition.
5. Interest on reserves is contractionary, negative IOR is expansionary.
6. Market interest rates are not a reliable indicator of the stance of monetary policy.
7. Interest rate targeting doesn't work well at the zero bound, and hence interest rates are not a reliable policy tool.
8. Fiscal stimulus is usually offset by monetary policy when the central bank is targeting NGDP (except for supply-side effects.)

Of course this is just my list, and not all market monetarist agree with each and every point. But notice that NGDP targeting is not on the list. I don't see it as a core idea, although I freely concede that it's a part of the appeal of MM.

Why was the San Francisco Fed paper such a bombshell? Not because they suggested the Wicksellian equilibrium interest rate might be negative 2.1%. As some commenters pointed out, others have made similar claims. Instead the real breakthrough was that the SF Fed paper conceded that this meant policy was actually contractionary (and by implication had been contractionary since 2008.)

When we claimed that money was very tight in 2008-09, people thought we were crazy. Almost everyone else (liberal, conservative, Keynesian, Austrian, old monetarist, etc.) thought policy was expansionary, and that the recession was caused by the financial crisis. If money was tight in 2008-09, then it's quite plausible that the Great Recession was caused by Fed policy, not financial turmoil.

Much of market monetarism is basic textbook economics, which has been forgotten by the rest of the profession (with a few exceptions such as Robert Hetzel.) Low interest rates don't mean easy money. Monetary policy is still highly effective at zero rates. An inflation targeting central bank will generally offset the impact of fiscal stimulus. We used to teach this stuff to our students back in 2007.

Another EC101 idea is that there is nothing wrong with trade "imbalances." The current account deficit represents the gap between domestic investment and domestic saving, and there is no reason to expect that gap to be zero for a household, a small town, a big city, a province or a country. It's simply not a problem. Except most economists seem to have convinced themselves that "imbalances" are a problem. Tyler Cowen has a new post today, which quotes someone pushing back on this bizarre conventional wisdom. Here is the Martin Sandbu quote he provides:

In an integrated regional economy like Europe's, it is improbable that every country is able to offer just the right investment opportunities to match the country's own savings. Countries that want to save more than they invest need to find a productive outlet for their savings. Countries that can productively invest more than they are willing or able to save must find funds from the outside. And so long as the fund that flow across borders are invested well, such flows can benefit lenders and borrowers alike. Indeed, large asymmetries are not only compatible with efficient economic development but they can be vital for making it happen: Norway's current account deficit reached 14 per cent of GDP in the late 1970s, but the capital is imported enabled it to build up one of the world's largest oil industries.
Paul Krugman made a career out of defending EC101, back in the 1990s (now he tends to do the opposite.) Here's Krugman back in 1996:
(ii) Adopt the stance of rebel: There is nothing that plays worse in our culture than seeming to be the stodgy defender of old ideas, no matter how true those ideas may be. Luckily, at this point the orthodoxy of the academic economists is very much a minority position among intellectuals in general; one can seem to be a courageous maverick, boldly challenging the powers that be, by reciting the contents of a standard textbook. It has worked for me!
And the same is true of market monetarism; we are mostly trying to defend basic textbook ideas that have been almost completely abandoned by the rest of the profession. We have to get back to a place where it's not "bizarre" to claim that a big fall in AD might have been caused by tight money.

COMMENTS (12 to date)
Dan W. writes:


It should not be bizarre to claim a big fall in AD might have been caused by tight money. But it also should not be presumed that tight money is the cause for a fall in AD nor should it be presumed that loose money is the cure.

After I pointed out to you that (a) there is real inflation in key parts of the economy and (b) government revenues are increasing quite rapidly you answered that money was still tight because monetary policy was failing to reach the objective of 2% inflation. This answer is a non-sequitur. Or, as you like to say, it is reasoning from a price change, or lack there-of.

Your recent posts demonstrate that you are well aware that increased investment in shale oil has greatly increased the supply of oil in the US which has lowered the price of oil and gasoline. This increased investment was a consequence of loose money, otherwise known as cheap credit. For if money had been tight then credit for shale oil would have been scarce and the extra supply of oil would not have happened. Tight money would have resulted in higher energy prices and higher reported inflation!

I would offer that if the Fed wanted to realize its objective of 2% inflation it ought to truly make money tight. The decrease in available credit would lead to a decrease in supply which would lead to an increase in prices.

Put another way, you are convinced that monetary levers can cause AD to increase faster than AS. But what if that is not so? What if loosening money leads to AS increasing faster than AD? If your measure of tight money is inflation then loosening money will fail to ever get you where you want to go!

BC writes:

Doesn't (1) imply NGDP targeting? If expected NGDP growth is the best indicator of monetary policy stance, then the central bank implements policy by altering expected NGDP growth, no?

End writes:

[Comment removed for supplying false email address. Email the to request restoring this comment and your comment privileges. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

Scott Sumner writes:

Dan, You are confusing tight money and tight credit. They are completely unrelated concepts. It is possible that tight credit could raise prices, as you suggest. But the Fed controls monetary policy, not credit. It is not possible for tight money to raise prices.

Because the Fed has not had a loose monetary policy in recent years, there is no possibility that loose money contributed to the fracking boom.

BC, Perhaps I should have said "NGDP, or some related concept" is the best indicator of easy and tight money. It need not be precisely NGDP, that is just a useful metric that many of us have latched on to. But you make a good point; one does seem to imply the other.

One can also speak of money being loose or tight relative to the central bank's actual policy goals, rather than relative to macroeconomic stability. For instance, I don't favor a 2% inflation target, but have recently pointed out that current policy is tight relative to what would be required to hit that 2% goal. So there probably is some ambiguity here, depending on whether I am talking about policy relative to a optimal regime (NGDP), or relative to the current regime (2% inflation).

James writes:

"A fiat money central bank never runs out of ammunition."

If a currency is devalued suffcently by a central bank, people will switch to some other currency. This is basically what is happening in Zimbabwe now. At this point, the central bank can still print all it wants, but since no one will accept the local currency, the central bank is basically firing blanks.

ThomasH writes:

I think you give far too much importance to whether Fed policy is labeled "tight" or not. If the Wicksellian interest rate were easily observed then their would be nothing wrong with judging "tightess" by how much the actual (comparable) interest rate differed from the Wicksellian rate. It's not, so the Fed needs to target something that is observable and reasonably related to real variables that we are interested in. Whether the target should be NGDP, the price level (general or just non-traded goods), wage levels depends on the way the target variable relates to the variables of interest under different regimes of shocks.

Whether the Great Recession was caused by Fed policy has not been proven. A very reasonable "cause" is the financial panic causing the Wicksellian rate or the expected ngdp or expected inflation or whatever to fall below its actual level. For some reason, the Fed did not start buying something soon enough long enough to get the variable back on target. If that failure is what you mean by "cause" then I understand what you mean by saying Fed policy caused the GR.

It would be more persuasive to see this in a model that shows the GR not happening if the Fed had perused a different set of buy and sell decisions.

One may believe that actually following a certain policy the Fed would fully offset macroeconomic effects of fiscal policy but this does not take into account that the Fed may be politically constrained from following the policy. In 2007-present the Fed was supposedly following an inflation target, but it was constrained both by an inflation rate ceiling and, more vaguely, in the amounts of non-short term government paper it could buy with the result that the PL has drifted away from the level it would have if the Fed had followed its policy. In that circumstance a higher federal deficit may allow the Fed to buy more stuff and so not necessarily offset the fiscal deficit.

Now I may be misunderstanding you and my analysis may be wrong, but I don't think its a unique misunderstanding or analysis.

CMA writes:

"1. Expected NGDP growth is the best measure of the stance of monetary policy."

NGDP growth is an effect or a consequence of monetary policy. Stance seems to imply an action which is cause of the effect. Cause and effect seem to be getting mixed up. I think changes in the MB are the best measure of stance.

"2. Monetary policy causes changes in NGDP."

Yes it's a cause like in previous point. It seems misleading to word it the way you did. Changes in ngdp can also be caused by changes in money demand. Maybe what you mean to say is monetary policy is the ultimate determinant of NGDP because it can override other factors affecting money demand.

4. A fiat money central bank never runs out of ammunition.

It can misdirect bullets and achieve little/nothing though.

5. Interest on reserves is contractionary, negative IOR is expansionary.

If we analyse IOR on its own. IOR increases demand for M so is contractionary. It also expands M supply through interest payments so is expansionary also in a sense.

fralupo writes:

Maybe a silly question here, but wouldn't interest on reserves be contractionary only if the interest paid was greater than the risk-free rate? Seems to me it even a positive rate could be expansionary in the right environment.

Scott Sumner writes:

James, You said:

"At this point, the central bank can still print all it wants, but since no one will accept the local currency, the central bank is basically firing blanks."

Not at all. The term "running out of ammunition" refers specifically to a situation where a central bank cannot debase its currency. It does not refer to an inability to boost RGDP. In the Zimbabwe situation you describe they were able to debase their currency to an incredible degree.

Thomas, You said:

"Now I may be misunderstanding you and my analysis may be wrong, but I don't think its a unique misunderstanding or analysis."

I'm afraid you did misunderstand me. I don't expect everyone to fall over and accept the market monetarist interpretation. But nonetheless this really does make a big difference. In 2008 people weren't even willing to ENTERTAIN the MM hypothesis that tight money caused the Great Recession, and that was because they said money was "obviously" accommodative." Now we find out it was tight. That doesn't prove it caused the recession, but it certainly helps. If the Fed had raised interest rates from 2% to 8% during 2008, I guarantee that my claims would not have been scoffed at like they were back in 2008 and 2009. That's the point I was trying to make; not that it was a smoking gun for the Fed causing the recession, just that it made the hypothesis plausible

CMA, You said;

"I think changes in the MB are the best measure of stance."

You are certainly entitled to your opinion, but I'll make two points:

1. Very few people agree with you. (On the other hand very few people agree with me and Bernanke, that NGDP is a good indicator.)

2. That's not a useful definition, as the change in the supply of base money might simply reflect a change in demand, being accommodated by the central bank.

fralupo, No, any positive IOR is contractionary relative to no IOR. Any IOR increases the demand for the medium of account (base money).

B Cole writes:

Good blogging. Print more money. Lots of it.

ThomasH writes:

It still seems, and maybe even more after reading the comments, that avoiding the language of "tight" and "loose" monetary policy (in contradistinction to "tight or "loose" credit) would help avoid confusion. Rather than saying that Fed policy was "tight" in 2007 why not say that it should have been doing x rather than y. In a call-in program I heard Bernanke say that the Fed should have lowered interest rates to zero in September rather than December 2007. (If we only knew why!) That's the sort or thing I mean, not "admit" that policy was "too tight" in September 2007.

Scott Sumner writes:

Thomas, That is really important (I presume you mean 2008) Do you have a link? Or recall the station?)

Comments for this entry have been closed
Return to top