Scott Sumner  

Arnold Kling on the Fiscal Theory of the Price Level

PRINT
Herbert Spencer, the misunders... Market monetarism continues to...

Arnold Kling recently responded to a point I made about the Fed not determining the path of interest rates:

He wrote,
Unfortunately, the Fed doesn't get to decide the path of interest rates. It looks like they do, but that's a cognitive illusion. The bond market determines the path of interest rates, reflecting factors such as global credit markets, as well as NGDP growth and the level of NGDP in the US.

I am fond of Winston Churchill's remark about someone who "stumbles across the truth, but then picks himself up as if nothing happened."

That is what I feel took place here. I think that the implication of the quoted sentences is that it is the bond market, not the Fed, that is in control. As I write in the Book of Arnold, the Fed is just another bank. It has no more ability to "target" macroeconomic variables than does Citibank.


He's wrong, but it's partly my fault as I was using a bit of poetic license. Of course the Fed does have some control over the path of interest rates. My point was that if the Fed wants a reasonably stable economy, then it pretty much has to follow the bond market. And that insight has no implications at all for the Fed's ability to steer the macroeconomy, because it doesn't steer the macroeconomy by controlling interest rates, it steers it by controlling the supply and demand for base money. More importantly, I was discussing the case where the Fed wanted 2% inflation, which limits its discretion over the path of interest rates, and Arnold is concluding that my claim has implications for whether the Fed could hit a different inflation target.

I'm much more concerned about this claim by Arnold in the comment section:

2. Central banks by themselves do not cause high inflation. High inflation is a fiscal phenomenon. When the government runs out of cheap credit but still runs a deficit, it starts printing money in great quantities, and that leads to high inflation.
That's true for Zimbabwe, but the US? The budget deficits were not particularly big during the Great Inflation of 1965-81. But monetary policy was very expansionary. LBJ's tax increase of 1968 failed to slow inflation, but Volcker's tight money of 1981 succeeded---despite Reagan running massive fiscal deficits. Around 1990, the Fed decided to target inflation at about 2%. Since 1990 it's averaged 2%. And how did this miracle happen? The Fiscal Theory of the Price Level proponents would presumably say it had nothing to do with the Fed, nothing to do with the Taylor Rule. Instead, Newt Gingrich and the other geniuses in the House of Representatives got together and masterfully produced an inflation rate of 2% through the budget process. And not just in the US. Other countries with very different fiscal policies also got inflation under control in the 1990s.

Or maybe Milton Friedman and Ben Bernanke and Paul Krugman and John Taylor and Janet Yellen and Friedrich Hayek and Michael Woodford and lots of other economists (including me) are right, and the central bank determines the trend rate of inflation. Maybe the financial markets have good reason to react strongly to hints of Fed policy shifts. Maybe because prices are measured in money terms, not apple terms or copper terms, the supply and demand for base money might actually have an important impact on the value of base money.

As Arnold said in a more recent post:

Supply and demand is an example of an interpretive framework that is very strong. That is, it seems to explain a lot, one rarely encounters anomalies, and it is consistent with other beliefs that we tend to hold.
I couldn't put it better myself.

PS. Take a look at the fiscal situation in Europe and Japan, and then the inflation rates in Europe and Japan, if you are still skeptical that monetary policy drives inflation.

HT: TravisV


Comments and Sharing






COMMENTS (16 to date)
Mike Sproul writes:

"When the government runs out of cheap credit but still runs a deficit, it starts printing money in great quantities, and that leads to high inflation."

That's a widespread misconception about the fiscal theory of the price level. In truth, budget deficits mean there is less backing for the government's money. This causes the money to lose value. If the money is worth 10% less, then people demand 10% more money to conduct their business. A central bank that wants to avoid unemployment will accommodate this 10% rise in money demand by printing 10% more money. Quantity theorists see the resulting correlation between money and prices and think the causation runs from money to prices, when in fact it runs from prices to money.

dlr writes:

I agree with you, and there is a way to tell the FTPL story without making overstatements like Kling's "Central banks themselves do not cause high inflation." Of course they can, though empirically it's more often fiscal for reasons that seem obvious.

In a FTPL world where we presume money won't be treated super-senior in the event of debt default, the price level can be thought of as the PV of future surpluses per unit of government liability plus a premium for the convenience yield of the medium of exchange. Either the fiscal authority or the Fed has the potential to cause high inflation. Even if the current convenience yield is nil as it is at the zero bound, the Fed can still cause high inflation if people believe there will be a future convenience yield and also that the Fed will then create more money than is demanded at the expected price level path. But it's also possible for the fiscal authority to create inflation that the Fed cannot prevent with even the tightest policy, as long as people believe the Fed will eventually be forced to accommodate insolvency.

John Cochrane's money as stock paper is a great parable, so much better than the recent hyper-modeling of equilibria stuff, but it needs to add the convenience yield. Imagine government liabilities were Amazon.com B shares and base money were A shares. A shares were trade at a premium value because people could also wear them at parties as necklaces. The price level is the price of the A shares. Amazon shares could decline (inflation) either because Bezos decided to make less profit in the name of some alternative ambition (fiscal inflation) or because he issued enough A shares (monetary) that the marginal owner didn't want to wear one. As long as there is an expected future A share wearer, Bezos can create inflation with either tool, and in fact can do so by just issuing a press release giving guidance about either tool, as long as he doesn't simultaneously offset it with a press release about the other tool.

In the end it is still whoever acts last. Or maybe, whoever has the reaction function in the first place (the CB) presuming the fiscal authority doesn't making acting last impossible.

ThomasH writes:

"if the Fed wants a reasonably stable economy, then it pretty much has to follow the bond market."

Whatever the Fed wants, it is constrained by the power of those who want tighter money. What other explanation can there be for the delay in dropping interest rates to zero in September 2007? For not starting QE sooner and making it larger and longer? For allowing the price level to drift below its pre-crisis trend?

ThomasH writes:

Unless taken horribly out of context (maybe he WAS talking about Zimbabwe) or has some special meaning of "by themselves,"

Central banks by themselves do not cause high inflation.High inflation is a fiscal phenomenon.

is a pretty shocking thing for an economist to say.

Lorenzo from Oz writes:

ThomasH: without disagreeing, there is something to what Arnold Kling has to say in that hyperinflation usually does occur in disastrous fiscal situations -- e.g. state with war debts way beyond its ability to pay which had also inflation-financed its war effort (Germany, Austria and Hungary after WWI for example).

A table of 56 hyperinflation (pdf) episodes does indicate that regime collapse, imperial collapse, military defeat are all common associations with hyperinfllation.

But that creates at best the motive, it does not create the hyperinflation; the actual issuing of money at extraordinary rates does that.

Scott Sumner writes:

Mike and dlr, I don't see any evidence that fiscal policy affects the inflation rate in the U.S.

Thomas, "What other explanation"? I just ordered Bernanke's new book. I'm sure he'll have plenty of other explanations.

CMA writes:

"the central bank determines the trend rate of inflation"

Nominal inflation only. It doesn't determine real inflation after it is adjusted for inflation.

Michael Byrnes writes:

Assuming there was a healthy dose of sarcasm there, I nominate CMA for comment of the year. (Or at least for nominal comment of the year, before any adjustment for inflation.)

Mike Sproul writes:

Scott:
The US evidence from the 1970's shows a moderate correlation between money and inflation, but the evidence from the last 10 years shows a complete disconnect. On the other hand, Lorenzo's data on 56 hyperinflations strongly supports the primacy of fiscal policy.

ThomasH writes:

I'll be looking forward to Scott's take on the Bernanke book.

From what I've heard about it he does not say much about the internal politics of the Fed board and the external [Paul "debase the currency" Ryan] pressure for hard money. [Can you imagine the uproar if a prominent Democrat had called for aggressive monetary policy or even for the Fed to keep the price level growing on target!]

Absence politics, it did not take much courage for Bernanke to reduce interest rates in December 2007 and to start QE in 2008; it's what any reasonably competent monetarist or Keynesian would have done. Unless Bernanke really explains what he was up against, he will not look very courageous.

Scott Sumner writes:

CMA, Yes, the Arabs discovered the real rate of inflation, or was it the Indians?

Michael, I agree.

Mike, I agree that hyperinflations are caused by governments monetizing fiscal deficits. But again, I believe the evidence strongly supports the notion that fiscal policy plays little or no role in US inflation.

Just to be clear, I'm not claiming that the rate of inflation is equal to the money supply growth rate, just that open market purchases make inflation higher than it would be otherwise, regardless of fiscal policy.

America's Great Inflation was not caused by monetizing fiscal deficits (which were not very big.)

I would add that there is a big difference between rapidly accelerating the rate of base growth when interest rates are well above zero (the 60s) and when interest rates are at zero (recently).

Paul writes:

I would recommend Fischer Black's "Business Cycles and Equilibrium" for background on these issues. Modeling inflation is hard. Assessing the efficacy of Fed policy is hard. Having strong opinions on what the fed should do is easy.

BTW @CMA if you deflate the CPI by the real price of gold you get a better estimate of real inflation ; )

Lorenzo from Oz writes:

Not my data, I just linked to it.

Even on the backing theory, it is the issuing of money at extraordinary rates relative to output which causes the hyperinflation. A fiscal motive is not quite the same as a fiscal theory.

Lorenzo from Oz writes:

Also, Australian inflation history and fiscal policy provides an excellent rebuttal to any fiscal theory of the price level, since Australian ran substantial and continual budget surpluses under the Howard Government while inflation bubbled along bang within the Reserve Bank of Australia's target range (2-3% on average over the business cycle), as it continued to do when the budget swung back to deficit under the Rudd, Gillard & Abbott Governments.

Mike sproul writes:

Lorenzo

Not your data, I know, but excellent choice of reference.

Actually, on backing theory principles, hyperinflation happens when money issuance outruns the issuer's assets, not when it outruns output of goods.

The Australian experience is consistent with money being a very senior claim on the issuer's assets, like bonds. The issuer's assets can go all over the place, and it's bonds ( and its money) could stay steady.

J.V. Dubois writes:

Inflation is obviously product of both supply and demand for money. Many times huge negative supply shocks may in fact cause inflation even in absence of any meaningful change in supply of money. One clear example is inflation in Warsaw ghetto in gold terms. It is also true that large part of Zimbabwe inflation was utter collapse of economy (based on agriculture) after massive confiscation of farms from western owners.

But to be honest I do not think this is what Arnold Kling had in mind - because sometimes supply shocks don't have to have anything with fiscal policy at all.

Comments for this entry have been closed
Return to top