Sumner Arguing That Low Interest Rates Are Contractionary
Scott has been posting some interesting stuff on this topic (one, two, three). Many of you–and Scott himself, at the tail end of this post–are bracing for me to go nuts, accusing him of all kinds of inconsistency. But nope, I am far more judicious than you realize.
However, I do think there is something not quite right in how Scott is discussing these matters, particularly when it comes to the Fed’s new policy (since 2008) of paying “interest on reserves” (IOR). Here’s the comment I just left:
Scott,
I’m actually mostly on your side on this one (believe it or not). But I do have one major doubt in the way you’re handling it.
How can you be sure that even a textbook rate hike (accompanied by reduction in reserves) is contractionary relative to the status quo? Wouldn’t it depend on the elasticity of velocity (or something like that)?
For example, if the Fed raises rates from (say) 4% to 5% and it does so by reducing the monetary base by (say) 2.3%, then to know the impact on NGDP, wouldn’t you need to pit the two forces against each other to see which dominates? Isn’t it conceivable that the point rate hike causes velocity to increase by more than 2.3%, and thus the smaller base is still generating more NGDP?
And then, it seems to me that it should be a no brainer that IOR is expansionary, since it doesn’t reduce the monetary base and increases interest rates. If you want to make an argument about fractional reserve banking and how a hike via IOR reduces M1 and the other broader aggregates, OK that at least makes sense, but I haven’t seen you make that argument.
Do you understand my confusion? I hope I’m at least making sense.
I thought Scott was meaning that higher rates stimulate the banks to lend if the IOR is not increased or if it is diminished.
Mr Murphy, I posted an article contrasting Austrians with Market Monetarists and Keynesians with Market Monetarists. Seems to me the analysis of the housing bubble is pretty good with the Austrians, but the market monetarists are right about tightening by the Fed making things worse. And Keynesians are right that low rates can help big business, but that is only a start and rates should be raised to help small business get loans they would not otherwise get, and house buyers as well. What do you think? I tried: http://www.talkmarkets.com/content/us-markets/sumner-and-his-market-monetarists-compared-to-mish-libertarians-and-keynesians?post=81981
My response would be that interest rates are prices, and all prices contain valuable information, the best thing the central bank can do is provide a lender of LAST RESORT which should rarely be used, and only then at higher than market rates. This offers stability, avoiding excessive interference with market price discovery, and allowing price signals to go ahead unimpeded.
https://youtu.be/tDkNPNSgiaY
The Fed cannot deliver a better price than what market discovery can deliver. Interference is effectively systematic dishonesty designed to fool people into doing what they would not have voluntarily chosen to do.
Hi Tel, I am watching the video. If banks don’t lend at low rates, then Prof Murphy’s argument is sort of irrelevant. It is funny though, he is a comedian.
Also, do you believe this?
Federal Reserve notes are not redeemable in gold, silver, or any other commodity. Federal Reserve notes have not been redeemable in gold since January 30, 1934, when the Congress amended Section 16 of the Federal Reserve Act to read: “The said [Federal Reserve] notes shall be obligations of the United States….They shall be redeemed in lawful money on demand at the Treasury Department of the United States, in the city of Washington, District of Columbia, or at any Federal Reserve bank.” Federal Reserve notes have not been redeemable in silver since the 1960s.
The Congress has specified that Federal Reserve Banks must hold collateral equal in value to the Federal Reserve notes that the Federal Reserve Bank puts in to circulation. This collateral is chiefly held in the form of U.S. Treasury, federal agency, and government-sponsored enterprise securities.
So, Tel, if the above is true, and collateral is important for all lending and clearinghouses, etc, then hoarding bonds lowers rates so low that banks don’t want to lend. That is a BIG problem. JMO.
If the banks were really constrained because of legal collateral requirements there wouldn’t be 2.8 Trillion dollars of excess reserves in the system.
https://research.stlouisfed.org/fred2/graph/?g=33uk
That said, it is getting a tiny bit difficult to figure out which laws still apply and whether anyone at that level is making a serious attempt to follow the law. One might go a step further and ask what is the meaning of swapping a government IOU note for a quasi-government Federal Reserve Note then using either as collateral for the other. There might be a law requiring the count of one paper to match the count of some other paper, and perhaps you can even find people conscientious about that… but it’s still all piles of paper. There’s never a practical limit to what you can do with piles of paper.
Finally, if you are worried that the US treasury is not writing enough IOU notes, then before long this “problem” will be fixed… even if nothing else is. Social Security is going to call in the internal IOU notes, and since there’s no tax available, the deficit is set to balloon. Not sure how this will fix the rest of the economy but anyhow there will be no shortage of fake collateral out there.
“… what is the meaning of swapping a government IOU note for a quasi-government Federal Reserve Note then using either as collateral for the other. … but it’s still all piles of paper.”
Excellent point.
I see your point, Tel, but I am thinking the excess reserves can only go out into the economy with collateral of some sort exchanged. It is really Fed funny money until that collateral is offered. Right?
“… the best thing the central bank can do is provide a lender of LAST RESORT which should rarely be used, and only then at higher than market rates.”
The high-ness of rates, alone, can’t tell you whether or not they are higher than the market would set.
If interest rates are oppressively high because of consumer demand, then that’s the correct interest rate. At a high enough interest rate, someone is going to be able to profitably offer a lower one, assuming the government isn’t preventing new lenders – to include unregulated ones.
There doesn’t need to be a lender of last resort for the same reason there ddesn’t need to be laws against price gouging.
“And then, it seems to me that it should be a no brainer that IOR is expansionary, since it doesn’t reduce the monetary base and increases interest rates.”
start with Y, M, i, V(i) and no IOR
now add IOR
i increases, M doesn’t change
but Y increases only if i increases by more than IOR at least according to Sumner’s assertion that V=V(i-IOR)
So maybe it’s all about elasticities
“Isn’t it conceivable that the point rate hike causes velocity to increase by more than 2.3%, and thus the smaller base is still generating more NGDP?”
Conceivable perhaps, but, it seems to me, that this is the exact opposite of what all the poor little Econ 101 students are taught about the effect of increasing M. But, I guess you could say that here too it is about elasticities.
So, my question is: Is this Dr. Murphy’s point? Sumner should be writing “Low Interest Rates are Contra..er..Expan um…Well It’s all about Elasticities!”
Capt. Parker, right, I probably should’ve elaborated more: Scott merely asserts that the relevant value is (i-IOR) without really explaining why. I understand why he might think that’s right, but I’d like him to spell out the argument so I can make sure I agree.
With mortgage rates dropping, people can borrow more and house prices go up. Existing home owners benefit from the windfall and a lot of it goes into consumption. Pretty simple equation.
https://research.stlouisfed.org/fred2/graph/?g=33kl
Stops working when :
[A] insufficient new borrowers can come along to bid up house prices (e.g. no jobs for young people, can’t afford the ridiculous entry prices, or just won’t commit to that level of debt)
[B] crisis in the mortgage industry caused by poor risk assessment leads to banks that clam up and refuse to lend
[C] artificially high prices encourage oversupply which eventually operates as negative feedback
[D] mortgage rates hit zero (i.e. Japan) and the bubble just cannot continue anymore.
I mean, it already happened didn’t it? In more than one country…
This was my thought, as well. And I’ll add that [B] is frequently accompanied by a lot of deleveraging in the financial industries.
“How can you be sure that even a textbook rate hike (accompanied by reduction in reserves) is contractionary relative to the status quo? ”
Assuming that the velocity of money NOT held in cash balances does not increase as rates increase I think you can be sure it will be contractionary.
.When the base shrinks, people spend (and lend) less to build their balances up again – which is contractionary.
The smaller base will mean that rates increase and people will hold smaller cash balances – which will mean a higher proportion of money is circulating – which is expansionary
But when the base shrink and rates rise – the second effect can never be bigger than the second.
Say the base shrinks 20%, and rates rise in response. As rates rise people choose to hold smaller cash balances. Rates will rise until , in aggregate, people are happy to hold the new smaller base.
At the extreme the entire quantity of money (20% of the original base) will ,at the higher rates , stop being held in people’s balances and start circulating. If this happens then aggregate spending will, at best , remain constant.
But more realistically the smaller base will lead to a combination of higher rates , smaller cash balances, and a smaller pool of circulating money, and this will lead to a new equilibrium below the initial level of spending.
“the second effect can never be bigger than the second” = ” the second effect can never be bigger than the FIRST”
Example:
The money supply start out at 1M, 30% is held in cash balances , the rest circulates with V =10, giving total spending of $7M. Interest rates are 5%
The CB reduces the base by 20% to $800,000. Initially spending falls sharply as people still want to hold $300k cash balances.
Rates start to rise.
As rates rise people choose to hold less than $300k in cash balances, which frees up some money to circulate.
Eventually rates will rise to a point where people desired cash balances + money they wish to use for spending = $800k.
This will almost certainly be at a a level where money not held in balances (and therefore circulating) is lower than before. Even in the most extreme case, cash balances are reduced by the full amount of the reduction in the base, and spending will be the same as before the reduction but never more.
Transformer all money is held in cash balances. At any given moment, every dollar is in someone’s possession.
Yes, that is implicit in the example in my comment.
To clarify my point; Assume we live in a world where all money consists of gold coins (and all banking is 100% reserve). People keep some gold coins as a “cash balance” and the rest circulates around. Of course as any moment each gold coins is in someone possession, but that is not important.
When interest rate rise people want to hold less gold coins in the “cash balance”. They take some out of the hoard and lend them out, after which they get spent (or hoarded by someone else).
Conceptually , I think things are more-or-less the same in a world of CB money, and FRB as far as changes in the the money supply , aggregate spending and interest rates go.
Perhaps “cash hoard” would be a better term as the total cash balance will also include gold coins they hold at any moment from income, that they then plan to spend in the near future.
Strictly speaking, very few gold coins are still legal tender, so the majority of those gold coins are just pretty metal items that can be bought or sold for whatever value someone wants to exchange for them.
“And then, it seems to me that it should be a no brainer that IOR is expansionary, since it doesn’t reduce the monetary base and increases interest rates. If you want to make an argument about fractional reserve banking and how a hike via IOR reduces M1 and the other broader aggregates, OK that at least makes sense, but I haven’t seen you make that argument.”
Isn’t the fed, when it pays IOR , like a borrower who doesn’t spends the money they borrows – just holds it and pays interest when it is due. So while IOR , by raising rates , causes people to hold lower precautionary balances and lend more, as the money they lend doesn’t get spent by anyone – it is not expansionary.
Yep, the FED is borrowing from the banks and not spending the money is borrowing. But when doing QE, the FED uses banks excess reserves to buy Mortgage Backed Secutities.
Larry White has a presentation on this. Just fiscal policy on desguise.
Scott appears to be talking about the steady state. He’s comparing one steady state with interest rate X to another with interest rate Y. Steady states should be dealt with by general equilibrium. It makes no sense to say that a steady state is “contractionary” or “expansionary”, it’s neither it can’t change. Suppose we have to hypothetical steady states one with low interest and one with high interest. There must be endogenous reasons for this, exactly because it’s a steady-state argument. Interest rates are endogenous in the long term and in a steady-state. So, in our two different general equilibrias the savings-investment market has ended up at a different price. As Scott says, the demand for money (which is 1/V in MV=PT) must be lower with lower interest, because of holding cost. But, we can’t compare these two general equilibria without asking why the interest rate is different. That would be reasoning from a price change! So, there has to be a part of the argument that deals with a change, that explains why X and Y’s interest rate differs. That could also drown out the effect of the difference in V.
Bob,
i–IOR is th relevant number because IOR is the floor at which banks will lend. No one in their right mind will lend below IOR.
Before IOR, the floor was zero, so i–IOR = i. Now is not. So to compare historic data, your have to subtract IOR.