My latest Macro Musing podcast is with John Cochrane, a Senior Fellow of the Hoover Institution and former professor of finance at the University of Chicago. In this week's episode I got to sit down and talk the fiscal theory of the price level (FTPL) with John Cochrane. I have been thinking a lot about FTPL lately--thanks in part to John's work on it--so it was a real treat to discuss it with him. We also got to chat about how he got into economics and the benefits of blogging for academics. It was a fascinating conversation throughout.
Here is a previous post I did on the FTPL that may be helpful in thinking about its implication. I also have a paper on FTPL coming out soon so stay posted.
You can listen to the podcast via iTunes or Sound Cloud, or through the embedded player above. And remember to subscribe since more guest are coming!
John Cochrane's Website
John Cochrane's Blog
John Cochrane's Twitter Account
Really enjoyable podcast. After listening to the podcast and rereading your linked post on the FTPL, it seems like there is a focus on finance - an equilibrium between financial products including gov debt - and how this affects psychology. Also, some ideas stood out as similar to other theories I have heard of. For example, the “taxes soak up inflation” and “money is gov debt” components sound similar to what advocates of MMT describe. On the other hand I thought his preference for a price level target was similar to what market monetarists advocate for. Although, to be honest, there seems to be overlap among macro theories and I get confused when trying to think about the distinctions between them.ReplyDelete
And indeed you seem to view FTPL and MTPL as complimentary views. I can imagine where you might use these ideas to further articulate a framework for understanding macro from both a fiscal and monetary perspective.
JonathanH, I think there is a way to reconcile these views and am working a paper that shows how. Will post it here once it is done.Delete
Another good one. But I feel like the different guests are kind of going in different directions. I have no complaint with that, it's interesting, but I hope there will be some attempt to reconcile or compare/contrast the views of different guests.ReplyDelete
At a first glace, FTPL and market monetarism would not seem to have a whole heck of a lot in common, but then I wonder if it isn't more of a "different parts of the elephant" kind of thing.
Regardless, the first few weeks of this have been great.
Thanks Michael. Yes, different views are being expressed here, but that is the nature of the beast when it comes to macro. However, I do think, as you suggest, there are are some areas of overlap.Delete
A download button would be nice. :-) If I pause this podcast too long, I lose the session and have to start over. The streaming could be fixed, but offering a download would be the best remedy.ReplyDelete
If FTPL adds to the picture, how does an interplay with a $500 billion a year US national balance of payments deficit?ReplyDelete
Also, the markets seem to accept the US government monetizing debt, without responding in any way that suggests inflationary impulses. This suggests to me that markets respect currency, as the US government can print more currency to honor existing debts!ReplyDelete
Sorry last comment: the idea of a CPI that measures 100 over centuries strikes me as impractical due to rapid evolutions in goods and services. In addition, as a practical matter in the United States there are housing shortages caused by local property zoning. This translates into inflation as measured. A central bank may indeed hit 100 on the CPI but what does that 100 mean?ReplyDelete
On other matters, the FTPL remains unconvincing due to empirical observations. Even as national government debts mount, and indeed debts of all kinds, inflation is retreating as measured. John Cochrane is obviously a highly intelligent and engaging individual, but he pretty much laid out the ground rules that he won't talk about FTPL in the current context. I thought Cochrane also got away with claiming that QE is just a swap of Treasuries for reserves, and he continues to ignore that Treasury bond holders sold their Treasuries to the primary dealers, and received freshly printed (digitized) cash in exchange.
What did the sellers of Treasuries do with the $3 trillion in new cash received?
I also wish David Beckwith had asked John Cochran about the use of cash in a no-inflation economy. The use of cash is already exploding, and is an excellent way to avoid taxes. this will leave the above ground economy exposed to regulations and taxes as larger fractions of the economy submerge.
I appreciate the effort made in this podcast but I must confess some dissatisfaction with John Cochrane positions and attitude.
A constant price level target would lead to liquidity traps in case the natural real rate went negative no? I was surprised Cochrane didn't mention thisReplyDelete
This, among other reasons, is why I favor a NGDP level target.Delete
This, among other reasons, is why I favor a NGDP level target.Delete
An increasing price level target would be fine (as is recommended in Egertsson & Woodford http://www.columbia.edu/~mw2230/Japan.pdf) - I'm highlighting not NGPD vs price level, but rather constant vs. increasingDelete
OK, so here is my question, which will probably reveal my ignorance, but here goes.ReplyDelete
Doesn't the fiscal influence on the price level still act through a monetary channel?
A rising (accelerating?) deficit growth will eventually force the government to print money in order to finance its deficit, since apart from congressional chaos it won't actually default. Money growth will be higher than what is needed to maintain a reasonable price level trend and therefore high inflation.
Is what is perceived as FTPL, then, just that fiscal policy can, in effect, force the hand of the monetary authority under certain circumstances (rapid increas in the rate of deficit growth)?
Cochrane did not seem to explain it this way. His explanation, I think, was that the deficit growth is seem and causes people to realize that the value of their money will erode via inflation.
Michael, questions like yours is why I am writing my paper (actually coauthoring) on the FTPL. Cochrane would probably disagree with me, but I see the FTPL biggest contribution in helping us better understand the velocity of money. And so from that regards, it is in my view still a monetary theory in that it better explains the velocity of money.Delete
The idea underlying FTPL is simple: the real value of government debt liabilities (think bonds + money) is the present discounted value of future real surpluses (i.e. net tax receipts). I think the two things that are "tricky" about this, from an intuitive perspective are:Delete
(1) Mechanically how does this actually determine the price level?
(2) Why is it true that the real value of government debt liabilities is equal to present discounted value of real surpluses?
Typical models of FTPL (and competitive equilibrium models more generally) do not answer (1) explicitly, and I think that is causing some confusion here. They simply answer (2) by showing that the only price consistent with optimal behavior + market clearing must change. They do not describe explicitly how each individual asset changes hands etc.
In your scenario, literally, when the government starts printing more debt but people don't expect future surpluses, higher inflation becomes necessary to "clear" markets. To be explicit, if inflation did not increase, the real value of government liabilities would grow exponentially, which is inconsistent with optimality on the part of the consumer because they could shift some of that wealth to the present time and improve utiltiy (i.e. spend some of the real wealth they are holding).
This is kind of hard to wrap your head around (and I probably haven't done a great job making it intuitive bc of lack of space) but this is the mechanism / way of thinking you need in order to understand the fiscal theory.
David Beckworth: If you are attempting some sort of unified field theory with NGDPLT and FTPL you must read the below:ReplyDelete
There was an interesting "pro" fiscal theory of the price level piece published June 2014 by the Dallas Fed, "Inflation Is Not Always and Everywhere a Monetary Phenomenon" by Antonella Tutino and Carlos E.J.M. Zarazaga
(you have to scroll to right year and title)
Befitting any tale about inflation, the Weimar Republic is dutifully mentioned, and the Rentesbank solution. In conclusion the authors state:
"The fiscal theory of the price level argues that what’s true about hyperinflation is valid more generally: Fiscal policy can prevent inflation from rising or falling too much by backing all outstanding nomi- nal government liabilities—interest bear- ing or not—with a stable level of expected future primary government surpluses. By formally incorporating fiscal policy in the analysis of price-level dynamics, the fiscal theory of the price level is better equipped than the conventional monetarist ap- proach to explain why the recent large expansion of the money supply in the U.S. has not caused higher inflation.
The theory implies that the quantitative easing programs, which created money to purchase mortgage-backed securities from the public, preserved price stability because that money is backed by the returns from real estate investments. Similarly, Germany restored price stability after its interwar hyperinflation with its real-estate-backed currency."
That is fascinating So when the Fed bought the MBS, they were fighting inflation!
It also seems to me that monetizing debt, counterintuitively, is anti-inflationary under the FTPL. The government is less indebted, more able to meet obligations.
Indeed the authors say, "Likewise, any money created to pur- chase government debt from the public at market prices is backed by the same primary surpluses that the public already expected would service that debt. As long as the expected primary surpluses back- ing existing government liabilities haven’t changed, there is no reason for the price level to change either."
Maybe the authors are onto something, I have long wondered when the Fed pays down the national debt, and "gets away with it"--ie not much inflation---if that does not increase the credit-worthiness of remaining debt, and also lower tax burdens into the future.
Hoo-haw. Maybe the Fed should every year buy $100 billion in MBS.
You are very confused.Delete
Possibly, but I am only citing the Dallas Fed paper.Delete
Interesting episode. So here's my questions you, David (and John),ReplyDelete
1. How can John claim that "money is debt" when... it's not actually debt in any way that I can see. My holding of cash does not leave me indebted to anyone, including the government. Moreover, I don't hold money because I have to pay taxes with US dollars. I seek and hold money to buy goods and services... because it's a generally accepted medium of exchange. Debt is debt. Money is a medium of exchange, be it US dollars or cigarettes in prison.
2. How is this an actual macro theory rather than a theory of political economy and a subset of reasons for changes in the demand for money? As I'm understanding it, Cochrane is saying that the "fiscal" part of the "theory" is the fact that holders of a currency will be inclined to dump it if they see the government running up debts it can't hope to repay and will thus be expected to inflate-away the debt by printing money. That's not a theory of the price level. It's an example of a good reason for demand for a particular currency to fall, leading to a fall in the value of that currency and a rise in the demand for everything else, pushing up prices. That very well might be the primary reason why modern central banks end up inflating, due to political pressure, but it doesn't seem to contradict anything about normal, Hume/Menger/Hayek/Market Monetarist monetary theory.
Am I missing something because I'm not a macroeconomist (likely).
1. Cochrane never said you become indebted when you hold money. That is absurd. Listen to the podcast again. Money is the government's debt. Why? Imagine a world with no government bonds. Suppose the government starts printing little pieces of paper. Of course it's possible for these pieces of paper to acquire value and be used as "money", even though in this case they are not debt. I think this is why you're confused.Delete
It's *also* possible for these pieces of paper to simply have no value as a medium of exchange. After all they're just pieces of paper.
What the FTPL does is it gives you conditions under which there is only ONE possible value for these pieces of paper. This requires that these pieces of paper be "backed" in some rigorous way by future income of the government
2. So it becomes a convenient shorthand to think of them as liabilities. That's the best explanation I can offer with no equations!
2. It's a theory of the price level in that it is a theory that gives you a single, unique equilibrium price level, something which is general not possible in models of competitive general equilibrium. It is a theory that literally tells you: there are X dollars in circulation, Y bonds in circulation (which pay some number N of coupons in units of dollars per year, etc.) and the present value of government surpluses is D, and monetary policy is this and that..... ---> the price level is P, and inflation is dP/dt. It tells you what the price level is.
I don't understand your argument, really. You agree that the fiscal theory tells you a mechanism by which prices change when certain things happen. How is that not a theory of the price level?
Money (at least current US currency) is not government debt. My US dollars are not an obligation of the US treasury or the Fed. I can't take my $20 to either agency and be expected to get anything other than even change in return. Yes, when I deposit that $20 in a bank, the cash becomes an asset for the bank and my balance and ability to withdraw becomes a liability. But that's still not a government liability.Delete
As far as the price level theory, the equations are inaccessible to me as a non-economist so forgive my ignorance on that language. But in plain english, what I understood is not a general theory but a particular pathway for both money demand and central bank behavior. I'm skeptical that such a mathematical model can hold without a X-factor variable that basically makes the whole thing a stylized story (like so-called "velocity"). MV = PT is a story, not a mathematical fact, as I understand it.
Is it impossible to imagine a world where the US treasury's reckless budgeting leads to a collapse in the demand for treasury bonds and then a default as rates skyrocket, yet the Fed through some amazing willpower doesn't inflate like crazy to buy maintains the price level? Again, pardon my limitations regarding the math.
If I take a bond to the US treasury or Fed, I also get nothing in return. That just isn't how bonds work! And that's the key point: there is no difference between (nominal) bonds and money. They are both just pieces of paper that promise other pieces of paper: bonds promise pieces of paper later, pieces of paper are just like bonds that pay no interest.Delete
The Fed can print a piece of money and use it to buy an asset. Similarly, they can print a bond, sell it for money, and use that to buy an asset. The point is the similarity between these two things.
If the Fed printed money and didn't buy assets but instead just spended it and wasn't going to make more money later, then (insert math here to make this rigorous, that's what the FTPL contributes) and there is inflation so that the real liabilities match the real assets. Simple as that.
This comment has been removed by the author.Delete
John Papola, even though the monetary base is not legally backed by some commodity anymore it is still a meaningful liability of the government to the extent the government cares about inflation. That is, the government has to have enough assets on hand so it can buy up the monetary base when inflation starts to run too high. Those assets include the Fed's assets but they also include the net present value of future government surpluses.ReplyDelete
To be concrete, consider a far-fetched scenario where some of the Fed's assets--say the MBS--lose their value completely. Now the Fed doesn't have enough assets to buy up the monetary base when its velocity picks up. If the government cared about maintaining price stability, what would happen next is that the Treasury Department would have to bail out the Fed. It would do so by giving the Fed more treasury securities to buy up the excess monetary base. This would work as long as the public expected the government to run large enough surpluses in the future to pay off the newly issued treasury securities. If the public didn't believe the Treasury would be able to run such surpluses, then the public would expect a future monetization of debt and the treasury bailout of the Fed in the present would not work--velocity would explode.
So yes, the monetary base is a 'debt' or real obligation of the government as long price stability matters. And if anything, the past seven years have taught us that low inflation is paramount for policymakers.
Repeating my reply on Facebook: "it is still a meaningful liability--or 'debt'--of the government to the extent the government cares about inflation." This is only true if "debt" is used figuratively rather than in its literal sense. A debt is an IOU or promise to pay up some definite thing or things. In what sense is a Federal Reserve Note an IOU? None that I can see. A mere "obligation" to avoid inflation (and a non-binding one at that) is hardly the same thing as a genuine debt. Let us not play with the ordinary meaning of words, and especially not for the sake of defending unsound theories.Delete