An economics, investment, trading and policy blog with a focus on Modern Monetary Theory (MMT). We seek the truth, avoid the mainstream and are virulently anti-neoliberalism.
Pages
▼
Pages
▼
Monday, November 4, 2013
Randy Wray — Did Scott Sumner Find MMT’s Achilles’ Heel?
L. Randall Wray | Professor of Economics and Research Director of the Center for Full Employment and Price Stability, University of Missouri–Kansas City
"Nay, Sumner’s battle is with economics—both mainstream and heterodox. There isn’t any theory or evidence in support of his claim that a 500bp reduction of the overnight rate would lead to a doubling of GDP and the price level. It is the claim of the unschooled or the ideologue."
Great article by Randall Wray. He simply finishes off Sumner in a very elegant - though demolishing - way.
Also, the paper he quotes at length, that he co-authored with Linwood Tauheed, is great research on a key topic: the impact of interest rate changes on aggregate demand. (Should be made mandatory reading for orthodox economists, who are usually somewhat allergic to empirical studies, evidence-based theories, etc...).
There's just one minor technical point I'd like to raise. According to Wray,
overnight interbank borrowing is collateralized by safe treasuries
However, Stigum (and Perry Mehring, who uses her book for teaching his money and banking course at Columbia) says otherwise.
Here's Stigum, page 540, Fourth edition of "Money Market":
a repo transaction is in essence a secured loan, whereas the sale of fed funds is an unsecured loans
So, was Wray really referring to the repo market, instead of the federal funds market? Do banks ask for collateral to lend reserves to one another or is it just really based on trust?
A federal funds transaction, or fed funds transaction, is an unsecured loan of U.S. dollars to a depository institution (DIs) by another DI or other eligible entity. The average rate at which these transactions occur is called the fed funds rate.
"In the United States, federal funds are overnight borrowings between banks and other entities to maintain their bank reserves at the Federal Reserve... These loans are usually made for one day only, that is, "overnight". The interest rate at which these deals are done is called the federal funds rate. Federal funds are not collateralized; like eurodollars, they are an unsecured interbank loan."
Merhling raised another point which I'd never heard before - that the market repo rate is normally a bit lower than the Fed Funds rate. Did you see that bit, Jose?
"Stigum says that in general the overnight repo rate* is a bit lower than the overnight Fed Funds rate, and a bit higher than the three month Treasury bill rate. Why should this be?..." etc
The way I understand it is that when the Fed isn't paying IOR and wants to set a target rate, the result is a corridor system, whereas when the Fed pays IOR it’s a floor system since no one is going to lend rb lower than they can get just holding them.
When the Fed is not paying IOR, then banks with excess reserves compete to lend in order to get a return on their excess rb and are willing to take a bit less than the Fed pays. Banks would rather borrow from other banks, too, since the Fed requires collateral. This keeps excess rb loaned up and near the Fed's target. If there were an abundance of excess rb, then the lower bound of the corridor would alls (tend toward zero). So the Fed conducts OMO to keep the level of excess rb sufficient for liquidity but tight enough to keep the the corridor close to the target rate.
I appreciate Mr. Wray's piece but why didn't he attribute my piece from yesterday that was featured here on Mike Norman by Tom Hickey and clearly had some level of inspriation for his own-certainly at least the title?
I think Merhling sees this difference between the repo and the FF rate as an instance of his cherished description of the monetary system as perpetually fluctuating between two poles - the ying of discipline and the yang of elasticity.
In 2012 we had repo rate > FF rate, so he interpreted this as the Fed trying (not very successfully) to promote more elasticity in the system by encouraging banks to borrow at FF and lend at repo to "support credit markets more generally".
Seems like an interesting and plausible interpretation.
Perhaps a deeper study of the issue - maybe including regressions of (FF-repo) on the economic and financial cycle etc. - would help clarify this subject further.
With regards to fed funds versus repo, I forget the details (I never got involved in the details of the US money markets), but there market segmentation issues. The GSE's do not have access to the fed funds market, I believe. Therefore there is a big pool of cash that can be deployed in one market, but not the other. I think the Fed opened up the reverse repo facilities in order to reduce these segmentation issues. In any event, the spreads are too small to make any difference to anyone other than someone on a money market desk.
Ok the Sumner has responded in any case-though I still think Wray should give attribution. For me, it's not a small mattter.
Wray hasn't printed my response yet which also quotes Sumner's response to his post. Here it is:
"My rebuttal is that whoever wrote that nonsense is a liar, a moron, and a jerk, all wrapped up in one. Life’s too short to respond to someone who claims I asserted that lowering interest rates to zero would cause NGDP to double. (It’s got several names attached, so I don’t know who wrote it.)"
"Rates just fell to zero in 2008 and I don’t see NGDP doubling."
"Mike, That quote says doubling the money supply causes NGDP to double, which is not a very controversial claim. The claim that cutting rates to zero causes NGDP to double is absurd."
Wray claims that if the CB doubled the monetary base the only effect would be to reduce the overnight funds rate from 5% to zero. He proceeds to argue that this wouldn't have much of an effect.
However Wray doesn't give a time frame. Is he talking short term or long term? In Sumner's example the monetary base is doubled permanently, so you could argue that even if it didn't have a huge effect in the short term, in the long term the effect could be that claimed by Sumner.
But the main problem is that Sumner claims the effect of doubling the monetary base permanently would not be due to the overnight funds rate being zero, but due to the size of the monetary base - i.e. the so-called "hot potato effect' of base money.
Wray completely ignored this aspect, because he has no knowledge of Sumner's theories.
Y I think it is a case of people talking past each other and that's mostly what Market Monetarists and MMTers always do anyway.
Sumner wants to point out that he didn't say cut the FF to zero and the NGDP and the price level will double but rather double the moetary base and the interest rate will do what it will.
I guess Wray is sort of assuming that doubling the MB would lead to the FF dropping to zero.
Even so, this doesn't make Sumner's claim that doubling the MB would double NGDP right.
Wray claims to have had no knowledge of Sumner until now-which if true should doubly require him to link to my blog-I'm just syaing.
It's not Wray who moderates the comments over at NEP. I have no idea why they moderate every comment but there you go. If you posted it it will probably appear sooner or later, but sometimes they take their time. They really should stop moderating all comments. What a waste of time.
If the cb doesn't pay IOR and allows excess reserves to exceed demand for overnight loans, the rate will fall to zero.
We can assume that that if the Fed did not pay IOR and doubled the monetary base, then this drive the overnight rate toward zero and if this resulted in the supply of excess reserves exceeding demand then the rate would approach zero very quickly.
Sumner is correct that doubling the M while holding V and Q constant would also double P. But he mistakenly thinks that either M is the money base or that changes in the monetary base drive M.
He doesn't seem to grasp the function of the monetary base. He seems to think that since rb are currency that changing the level of the monetary base is the same as dumping cash into people's pockets who realize that the dump will be inflationary so the new cash becomes a hot potato, with everyone trying to exchange it for real assets or goods before it loses more purchasing power.
Also Sumner was asking us to take his thought experiment where during a healthy economy the Fed doubled the MB. The Fed wouldn't do this during a healthy economy anyway.
In 2008 the Fed almost tripled the MB and this led to nothing like a tripling of prices or NGDP
Sumner is correct that doubling the M while holding V and Q constant would also double P. But he mistakenly thinks that either M is the money base or that changes in the monetary base drive M.
Right we have just been living through a major real-world experiment that has seen a huge increase in the money base - and are all still waiting for that "hot potato" effect to take place. Those guys have not learned a single thing since the beginning of the crisis.
Ok, I love Randall Wray again as he gave me attribution now! LOL.
Dan Sumner's 'thought experiment' was of an economy not in recession and no IRR. So his claim was that in a good, healthy economy doubling the MB doubles prices and NGDP.
What use this is when the economy isn't helpful and we do have IRR I don't know but this is what he wanted to prove.
What do economists do in countries that simply don't require CB reserves?
Reserve more time for afternoon tea?
Issue unemployment insurance for invented-conflict mediators?
:(
ps: were there any North American economists yet, back in the early 1600, to screw up the colony's issuance of diverse & inventive paper scrips?
Good lord!
I guess the banking lobby is at fault? If you stick ENOUGH pins in an economy, I guess even a zombie will float on all the dancing heads? Divided, distracted & conquered.
Sumner's law in plain English: "When banks double their deposits at the central bank in a moment when the economy is not at the zero lower bound, then nominal GDP will eventually double".
he thinks that would still be the result 'at the zero lower bound', so long as the increase in base money was permanent and the CB doesn't pay interest on reserves.
"Sumner is correct that doubling the M while holding V and Q constant would also double P. But he mistakenly thinks that either M is the money base or that changes in the monetary base drive M."
Trouble is, that equation has very little relevance to any real-world scenario. In a addition, the individual variables within the MV=PQ relationship are themselves functions of each other.
Too many moving parts to make sense of, and setting some variables as constant while looking at the change in others is not giving one a view into real-world processes.
Chasing your tail would be more productive.
Then there's the conventional definition of money supply, since in practice the money supply is unlimited, and the only money supply that matters is the money consumers have available to spend.
"Nay, Sumner’s battle is with economics—both mainstream and heterodox. There isn’t any theory or evidence in support of his claim that a 500bp reduction of the overnight rate would lead to a doubling of GDP and the price level. It is the claim of the unschooled or the ideologue."
ReplyDeleteOuch. That was brutal and entirely called for.
Randy likes the word 'Nay'.
ReplyDeletehttp://www.youtube.com/watch?v=0vSmiq7CrCk
Great article by Randall Wray. He simply finishes off Sumner in a very elegant - though demolishing - way.
ReplyDeleteAlso, the paper he quotes at length, that he co-authored with Linwood Tauheed, is great research on a key topic: the impact of interest rate changes on aggregate demand. (Should be made mandatory reading for orthodox economists, who are usually somewhat allergic to empirical studies, evidence-based theories, etc...).
There's just one minor technical point I'd like to raise. According to Wray,
overnight interbank borrowing is collateralized by safe treasuries
However, Stigum (and Perry Mehring, who uses her book for teaching his money and banking course at Columbia) says otherwise.
Here's Stigum, page 540, Fourth edition of "Money Market":
a repo transaction is in essence a secured loan, whereas the sale of fed funds is an unsecured loans
So, was Wray really referring to the repo market, instead of the federal funds market? Do banks ask for collateral to lend reserves to one another or is it just really based on trust?
FRBNY:
ReplyDeleteA federal funds transaction, or fed funds transaction, is an unsecured loan of U.S. dollars to a depository institution (DIs) by another DI or other eligible entity. The average rate at which these transactions occur is called the fed funds rate.
http://www.newyorkfed.org/aboutthefed/fedpoint/fed15.html
http://en.wikipedia.org/wiki/Federal_funds
ReplyDelete"In the United States, federal funds are overnight borrowings between banks and other entities to maintain their bank reserves at the Federal Reserve... These loans are usually made for one day only, that is, "overnight". The interest rate at which these deals are done is called the federal funds rate. Federal funds are not collateralized; like eurodollars, they are an unsecured interbank loan."
y,
ReplyDeleteThanks.
So it was probably a typo by Wray.
Anyway, I also posted the question on NEP. Let's see if he'll answer (apparently, he does not much follow blogs, perhaps not even his own...).
Merhling raised another point which I'd never heard before - that the market repo rate is normally a bit lower than the Fed Funds rate. Did you see that bit, Jose?
ReplyDeleteFrom Merhling's course notes:
ReplyDelete"Stigum says that in general the overnight repo rate* is a bit lower than the overnight Fed Funds rate, and a bit higher than the three month Treasury bill rate. Why should this be?..." etc
*(Not the Fed repo rate)
https://d396qusza40orc.cloudfront.net/money%2Flecture_notes%2FLec%2007--Repos%2C%20Postponing%20Settlement.pdf
The way I understand it is that when the Fed isn't paying IOR and wants to set a target rate, the result is a corridor system, whereas when the Fed pays IOR it’s a floor system since no one is going to lend rb lower than they can get just holding them.
ReplyDeleteWhen the Fed is not paying IOR, then banks with excess reserves compete to lend in order to get a return on their excess rb and are willing to take a bit less than the Fed pays. Banks would rather borrow from other banks, too, since the Fed requires collateral. This keeps excess rb loaned up and near the Fed's target. If there were an abundance of excess rb, then the lower bound of the corridor would alls (tend toward zero). So the Fed conducts OMO to keep the level of excess rb sufficient for liquidity but tight enough to keep the the corridor close to the target rate.
I appreciate Mr. Wray's piece but why didn't he attribute my piece from yesterday that was featured here on Mike Norman by Tom Hickey and clearly had some level of inspriation for his own-certainly at least the title?
ReplyDeletehttp://diaryofarepublicanhater.blogspot.com/2013/11/sumner-on-achilles-heel-of-mmt.html
The NY Fed has been testing reverse repos with non-primary delears as well to better set a floor in the repo markts.
ReplyDeletey,
ReplyDeleteI think Merhling sees this difference between the repo and the FF rate as an instance of his cherished description of the monetary system as perpetually fluctuating between two poles - the ying of discipline and the yang of elasticity.
In 2012 we had repo rate > FF rate, so he interpreted this as the Fed trying (not very successfully) to promote more elasticity in the system by encouraging banks to borrow at FF and lend at repo to "support credit markets more generally".
Seems like an interesting and plausible interpretation.
Perhaps a deeper study of the issue - maybe including regressions of (FF-repo) on the economic and financial cycle etc. - would help clarify this subject further.
For more background see here
ReplyDeletehttp://diaryofarepublicanhater.blogspot.com/2013/11/randall-wray-speaks-of-sumner-on-mmts.html
Probably would admit he doesnt have the "blog etiquette" down yet Mike...
ReplyDeletersp,
Well if he wants to blog maybe he ought to try learning it. I mean why not just admit this is where he got the quotes from?
ReplyDeleteYou got to understand I was jazzed he wrote the piece till I saw he failed to even link to where he got it from.
ReplyDeleteWith regards to fed funds versus repo, I forget the details (I never got involved in the details of the US money markets), but there market segmentation issues. The GSE's do not have access to the fed funds market, I believe. Therefore there is a big pool of cash that can be deployed in one market, but not the other. I think the Fed opened up the reverse repo facilities in order to reduce these segmentation issues. In any event, the spreads are too small to make any difference to anyone other than someone on a money market desk.
ReplyDeleteOk the Sumner has responded in any case-though I still think Wray should give attribution. For me, it's not a small mattter.
ReplyDeleteWray hasn't printed my response yet which also quotes Sumner's response to his post. Here it is:
"My rebuttal is that whoever wrote that nonsense is a liar, a moron, and a jerk, all wrapped up in one. Life’s too short to respond to someone who claims I asserted that lowering interest rates to zero would cause NGDP to double. (It’s got several names attached, so I don’t know who wrote it.)"
"Rates just fell to zero in 2008 and I don’t see NGDP doubling."
Here is the link to Sumner's response.
ReplyDeletehttp://www.themoneyillusion.com/?p=24551#comment-290539
"Mike, That quote says doubling the money supply causes NGDP to double, which is not a very controversial claim. The claim that cutting rates to zero causes NGDP to double is absurd."
ReplyDeleteWray claims that if the CB doubled the monetary base the only effect would be to reduce the overnight funds rate from 5% to zero. He proceeds to argue that this wouldn't have much of an effect.
However Wray doesn't give a time frame. Is he talking short term or long term? In Sumner's example the monetary base is doubled permanently, so you could argue that even if it didn't have a huge effect in the short term, in the long term the effect could be that claimed by Sumner.
But the main problem is that Sumner claims the effect of doubling the monetary base permanently would not be due to the overnight funds rate being zero, but due to the size of the monetary base - i.e. the so-called "hot potato effect' of base money.
Wray completely ignored this aspect, because he has no knowledge of Sumner's theories.
So it's a case of people talking past each other.
Y I think it is a case of people talking past each other and that's mostly what Market Monetarists and MMTers always do anyway.
ReplyDeleteSumner wants to point out that he didn't say cut the FF to zero and the NGDP and the price level will double but rather double the moetary base and the interest rate will do what it will.
I guess Wray is sort of assuming that doubling the MB would lead to the FF dropping to zero.
Even so, this doesn't make Sumner's claim that doubling the MB would double NGDP right.
Wray claims to have had no knowledge of Sumner until now-which if true should doubly require him to link to my blog-I'm just syaing.
maybe you should sue him, Mike.
ReplyDeleteIt's not Wray who moderates the comments over at NEP. I have no idea why they moderate every comment but there you go. If you posted it it will probably appear sooner or later, but sometimes they take their time. They really should stop moderating all comments. What a waste of time.
ReplyDeleteMathew Forstater and Warren Mosler, The Natural Rate of Interest Is Zero
ReplyDeleteIf the cb doesn't pay IOR and allows excess reserves to exceed demand for overnight loans, the rate will fall to zero.
We can assume that that if the Fed did not pay IOR and doubled the monetary base, then this drive the overnight rate toward zero and if this resulted in the supply of excess reserves exceeding demand then the rate would approach zero very quickly.
Sumner is correct that doubling the M while holding V and Q constant would also double P. But he mistakenly thinks that either M is the money base or that changes in the monetary base drive M.
ReplyDeleteHe doesn't seem to grasp the function of the monetary base. He seems to think that since rb are currency that changing the level of the monetary base is the same as dumping cash into people's pockets who realize that the dump will be inflationary so the new cash becomes a hot potato, with everyone trying to exchange it for real assets or goods before it loses more purchasing power.
Also Sumner was asking us to take his thought experiment where during a healthy economy the Fed doubled the MB. The Fed wouldn't do this during a healthy economy anyway.
ReplyDeleteIn 2008 the Fed almost tripled the MB and this led to nothing like a tripling of prices or NGDP
Sumner is correct that doubling the M while holding V and Q constant would also double P. But he mistakenly thinks that either M is the money base or that changes in the monetary base drive M.
ReplyDeleteRight we have just been living through a major real-world experiment that has seen a huge increase in the money base - and are all still waiting for that "hot potato" effect to take place. Those guys have not learned a single thing since the beginning of the crisis.
Ok, I love Randall Wray again as he gave me attribution now! LOL.
ReplyDeleteDan Sumner's 'thought experiment' was of an economy not in recession and no IRR. So his claim was that in a good, healthy economy doubling the MB doubles prices and NGDP.
What use this is when the economy isn't helpful and we do have IRR I don't know but this is what he wanted to prove.
So many macro tempests in faux teapots.
ReplyDeleteWhat do economists do in countries that simply don't require CB reserves?
Reserve more time for afternoon tea?
Issue unemployment insurance for invented-conflict mediators?
:(
ps: were there any North American economists yet, back in the early 1600, to screw up the colony's issuance of diverse & inventive paper scrips?
Good lord!
I guess the banking lobby is at fault? If you stick ENOUGH pins in an economy, I guess even a zombie will float on all the dancing heads? Divided, distracted & conquered.
Sumner's law in plain English: "When banks double their deposits at the central bank in a moment when the economy is not at the zero lower bound, then nominal GDP will eventually double".
ReplyDeleteIs that it?
he thinks that would still be the result 'at the zero lower bound', so long as the increase in base money was permanent and the CB doesn't pay interest on reserves.
ReplyDelete"Sumner is correct that doubling the M while holding V and Q constant would also double P. But he mistakenly thinks that either M is the money base or that changes in the monetary base drive M."
ReplyDeleteTrouble is, that equation has very little relevance to any real-world scenario. In a addition, the individual variables within the MV=PQ relationship are themselves functions of each other.
Too many moving parts to make sense of, and setting some variables as constant while looking at the change in others is not giving one a view into real-world processes.
Chasing your tail would be more productive.
Then there's the conventional definition of money supply, since in practice the money supply is unlimited, and the only money supply that matters is the money consumers have available to spend.
Matt calls this MMT HPM.
Jose I think you get Sumner's 'law' and it doesn't really amount to much as the Fed wouldn't double the MB during a healthy economy anyway.
ReplyDelete