"tackling inflation would be easier if the Fed could pull money out of people’s checking accounts"
The Fed can't do this, but the Feds certainly could, through the mechanism of a wealth tax or raising marginal rates. Your entire argument (which I certainly agree with) offers robust support for managing the economy more through direct fiscal transfers and taxation; much less through the Fed and rates; and not at all through QE. Replace modern monetary policy with old-school fiscal policy, in other words.
Taxation is the opposite of stimulus - if stimulus is a contributor to inflation than taxation would be a helpful way out. The trouble of course is that politics doesn't necessarily abide by those rules. Ideally, the Fed is the "adult in the room" that doesn't have to bow to political pressure and can to the unpopular things when necessary.
The Fed's job is easier if fiscal policy is rowing in the same direction.
It is complex questions that I am after to explain the macro world we live in, and everyone seems to have dull answers to questions I didn't ask. Your writing is data driven, intelligent and with a different perspective. Makes me ask the questions that make me better prepared for economic events.
Is the fed/government more afraid of inflation or deflation/depression? - (it depends). I have never been more convinced of the trajectory of the economy when I realized that the difference between previous downturns and this one, is severe inflation. And you hit the nail on the head on this question in this article.
Or maybe I am just a perma-bear. Either case, these articles are one of my favourite economic sources, and recommend to anyone interested in understanding things a little better.
Thanks for the kind words, Phil, and I hope you will keep reading.
In recent history, inflation has not been a challenge and so it has receded on the list of central banking worries. But over the long term the Fed is more afraid of inflation, because its harder to tackle. As Bernanke said, the cure to deflation is the money printer. This has been the paradigm of central banking for the last several years, and it has been a long time since there was a serious inflation problem to worry about.
Now, inflation is concern number 1, compounded with second guessing of recent policy mistakes.
The Furnace was my first exposure to your writing. I would like to tell you that after reading many explanations of why QT is dangerous, including The Fed Guy, I found yours to be the most clear, concise and easy to understand.
The connection between deposits and liabilities at the banks and the Repo market is one that has been alluded to a great deal – but the simplicity of your explanation made it very apparent to me the corner that the fed could find themselves in.
I am concerned greatly that any backstopping of the Repo market will be considered a pivot by market participants who are very focused on the currency implications of higher rates, at the expense of tamping down inflation.
Is it possible that if the fed backstops the repo market to ensure liquidity that the result is a much higher stock market, resulting inflation and corresponding longterm severe financial issues or am I over simplifying the problem?
appreciate any insight you could provide...now i am off to read the rest of your substack.
I would argue that the scenario you are asking about (Fed intervention leading to higher stock market, which in turn creates risk of popping), has been what's already occurred since 2008, and supercharged over the past two years.
From your linked article "From the middle of 1928 to its peak in September 1929, the Dow Jones Industrial Average soared 90%."
Well from March 2020 to December 2021 the Nasdaq 100 soared 140%.
That *doesn't* mean I think we are on the verge of a depression, but the dynamics between loose fed policy and a soaring asset bubble have already taken place (in stocks, housing and bonds!)
Thanks again for reading, and I hope you enjoy the column.
Great write up as always Bear! I just had one question:
Bank cash went up by 10x between 2008 and 2014 from the previous QE, but demand deposit did not have a similar spike. As I understand from Down the Drain, QE provide cash to owners of financial assets who then deposit their cash into commercial banks. Would this deposit not increase the demand deposit? If so, then why do we not see demand deposit following the path of bank cash?
The easy rule is that if the Fed is buying securities from an actual commercial bank, than there is no change in money supply or demand deposits - it is just an "asset swap". This was the case for the "original" QE, was primarily a bailout of the banking sector in 2008. However, in the later periods of QE in 2011 - 2014, you see that there was a significant uptick in deposits which corresponds with the Fed's asset purchases. In the COVID era it was almost 1:1.
Hey Bear, great piece. Such a pleasure to always read the pieces AND the interaction between you and your readers! Keep up the great work :)
You wrote:
"On the short end, government money market funds meanwhile have reduced their holdings of USTs in favor the RRP."
Being a long time reader of yours, I remember you writing in your piece "Down the Drain" that the Fed essentially chose to pay interest to MMFs that park funds in the RRP. Specifically, what drew my attention back then was the phrase (in quotes): " This week’s rate hike increased the RRP rate from 80bps to 155bps. With the 1-month UST yielding just over 105bps, there is now 50bps of incentive to sell UST and allocate to the RRP."
The Fed rate is currently at 3-3.25% and the 1-month T-bill yields 2.67%, so essentially the arbitrage opportunity is around the same. as it was then. This implies that MMFs would still be incentivized to forgo buying USTs in favor of truly riskless RRPs.
What I do not get is, why would the Fed chose to do so? Wouldn't it be convenient for the Fed to stop raising the interest rate in the RRP with the same amount that they do during each rate hike? This way, the "truly riskless" RRP would yield less than T-bills at some point and this provide an incentive to some MMFs to gradually move funds from the RRP right into USTs. As I understand it, there is around 2-2.5 trillion USD worth of liquidity parked in RRP at the moment which could help absorb liquidity shocks.
And then, I would like to hear your thoughts on whether USTs could be used as a weapon from foreign holders of USTs against the US, as part of an economic war if things were to escalate. For example, China holds around 13% of USTs (according to this site ----> https://ticdata.treasury.gov/Publish/mfh.txt). In a scenario where China decides to respond back to US waging an economic war to them (tarrifs, ban of EUV machines, ban of export of high end chips, delisting of certain entities from US stock markets) and escalate by dumping their USTs in the market, wouldn't that further exacerbate the UST funding problem you mentioned in your piece? And what would be possible scenarios from there onward? I suppose funds parked in RRPs could act as a backstop against that, if I have understood the mechanics correctly.
First - yes you are correct that the RRP award rate after last week's rate hike is much higher than the current 1Mo UST rate which will cause money market funds (MMFs) to reallocate to the RRP. They will likely do this by reallocating their exposure once the 1mo bills come due, rather than outright selling them, which causes it to be a more gradual transition (rather than immediate selloff to parity). This should continue to occur over the coming months.
Why does would the Fed do this? Because need the RRP to make sure that short term rates meet their target, which today is 3.05%. They injected a ton of liquidity over the past two years to reduce rates, and now the only way raise rates to their target is to let all of that liquidity be released into the RRP.
This may not sound like a huge problem until you realize that there are things that were built on all those dollars being around that aren't going to work when they are removed. You've already seen this in the most speculative area's of the market (think crypto or unprofitable tech), but will become more problematic when it becomes bank balance sheets.
As to the last point on economic weapons. We must acknowledge that economics and markets are an important lever in geopolitical power. For example, the West collectively have used their sanctions as a financial weapon against Russia - Russia has responded by cutting energy supply to europe, and there are big consequences for each as a result. This same concept could apply to countries that hold USTs, and they could certainly wield them in an unfriendly way. Russia, for example, sold out of all of its treasuries over the course of the last decade likely to avoid the brunt of future sanctions like the ones that were imposed for recently.
Hey Bear, late reply here as I have been away on a long trip. Thanks for the explanation, helps a lot.
Regarding liquidity and RRP - do you think then that in the event of a true UST liquidity event, the chance of which seems to be growing as days pass, the flows of funds parked in the RRP could easily reverse back into USTs if the Fed decided to do so? And do you think reversing their decision to have an RRP yield that follows the rate hike cycle would damage their credibility?
An additional problem I suppose is that the Fed might not actually know the real size of the Eurodollar market anyway. As I understand it, they do know the amount of debt in relation to USDs issued within the US jurisdiction, but they could be off in their estimates of the Eurodollar debt market.
And then, regarding the use of economic warfare. I suppose a strengthening dollar and the liquidity problems that creates in emerging markets is in itself a weapon the US can use to influence political leaders within those markets. The US can probably use that as a very efficient weapon but need to do that elegantly, to avoid a "loss of trust" event.
Beautiful analogy. I will now recommend this to anyone who wants to understand this (arguably complex) topic. Humor me, though. Adding too much fuel creates a lot of smoke because of incomplete combustion (which is the goal, so that you can remove the fuel later if its too much). In your analogy, what would the smoke be?
Ha! I'm not sure - obviously the analogy is only a crude representation so I'm trying to think how the comparison could apply, and I'm not coming up with much. I'm not sure if there is a "waste product" type that would be continue to the analogy, but happy to hear if you could think of something.
Loved the real life analogy of your grandbears’ fire place Bear.
The reason you developed those skills is because either the wood was not infinite,and comes at a cost,and of course Bears like the temperature ‘just right’
Assume the wood heap (coal)is infinite in your application.
So is it that all the doors and windows are open,and despite burning at a high rate,no warmth is accumulating to allow the fire to be turned down?
I almost made this point about how the analogy doesn't hold because we had limits to the wood, where the fed can print to infinity. Yet, while there is no upper limit of the Fed's balance sheet, the "cost" comes in inflation or currency devaluation.
2. Agree that the declining bank reserves set the stage for sept 2019, though in combination with intraday bank liquidity requirements established post GFC. Fabulous paper on the subject. https://www.nber.org/system/files/working_papers/w29090/w29090.pdf . One notable difference sept 2019 to now though is the Fed became a repo lender on an emergency basis in 9/2019, there was no standing facility/systematic liquidity relief valve to police the top end of the range like there is now with the SRF.
3. I generally think all the liquidity being "sterilized" in the RRP will eventually transfer to cover the liquidity hole caused by QT. I think the govt. MMFs are parking the money in RRP because its often a better play for yield (not to mention its effect on MMF liquidity requirements) than short term UST given it keeps increasing by 75 bp every 6-8 weeks. When we get to the terminal rate I think that dynamic shifts and WAMs of the MMF will get much closer to their 60 day max. Even before then though, If repo rates rose above RRP due to funding stress I dont know of any reason why those MMFs wouldnt shift deployment of their cash to grab the extra yield.
4. On treasury liquidity. As a technical matter I dont think there can be insufficient liquidity in the primary market. As I understand it the primary dealers have no choice but to provide it even if nobody else bid, though that doesnt mean the yield cant be super high. On the secondary market. I think the MMF/RRP liquidity would eventually deploy since if price of UST got low enough an attractive carry trade funding through repo from MMF (at rate higher than RRP) would develop for entities with the B/S space to do it. SRF probably an eventual backstop here too
That said, I am very new to the world of finance systems (i literally first learned how QT worked in relative detail from reading dispatch 8 of your blog in June) so its possible im saying ill informed things.
1. Interesting point on the maturity schedule vs. caps that I will have to check out.
2. Yes the SRF will help (which was also referenced in the post here). But it doesn't cover all market participants. Further, the problem in general with Fed liquidity facilities is that no one will use one for convenience - they will be forced into them after things go south, so while it is still helpful, it is not fully preventative.
3. I've written quite extensively on the RRP - I think its by far the best asset (assuming equal yields) because of its duration, counterparty. If 4 week tbills start trading well above the RRP expected rate, than MMFs may start to buy t bills. But again, this assumes that everything remains orderly. If there is some blowup, will MMFs be the one to step in to save the day with liquidity? I doubt it.
4. As a practical matter, there can absolutely be insufficient liquidity in the treasury market. It has happened several times before and forced the Fed to intervene to support prices. Its hard to see how the treasury market is in better shape than it was at those points in time.
In theory you can always say that there is a "carry trade" to be had in a liquidity blowup, and yet they still happen. I think it has the causation backward though. Prices go down because people are trying to access liquidity and no one is willing to provide it. If there was a ton of liquidity waiting on the sideline to pounce on treasuries, then the treasury market wouldn't have blown up in the first place. The dynamics are a bit different today because of "idle" liquidity at the RRP, but MMFs could only step into short dated treasuries, and I doubt they would choose to in a moment of crisis anyways.
In any case, these are all great points, and thanks for reading!
Hey John, as a person trying to understand how financial plumbing works I can say that your posts looks great and have just been added in my to-read list.
Just a question though. You wrote: " Even before then though, If repo rates rose above RRP due to funding stress I don't know of any reason why those MMFs wouldn't shift deployment of their cash to grab the extra yield." Does this mean that the repos can have different yields than reverse repos? But how is that possible if a reverse repo is the unwinding (maybe settlement is a better word?) of a repo purchase?
My understanding was that each single repo transaction is associated with a reverse-repo transaction. Specifically:
1) A repo purchase is the act of selling an asset to a counterparty with an agreement to buy back that asset at a later point in time at a specific price.
2) The duration is typically quite short (1 day) and that is rolled over on a day-to-day basis.
3) Repo transactions do not officially pay a yield, as they are not loans. However, the system is kinda "rigged" as the fact that the price of repurchase is agreed a-priori, and there is a duration in the transaction, essentially is a form of yield.
Could you please point out to me where/if I am wrong on any of the above?
Just nomenclature. In both instances the MMF plays the role of the Repo buyer/lender (considered the reverse side of the transaction) with either:
1. The FED on the other side of the trade as the Repo seller/borrower via the RRP facility or
2. Some Repo mkt participant who has collateral and needs cash who is the Repo seller/borrower in that instance (usually the trade is intermediated by a dealer)
Your understanding is generally correct, though I would change "repo purchase" in 1 to "repurchase agreement" or just "repo". Repo Purchase implies to me the purchase of the collateral/lender of money which is actually the reverse side of the trade. Best to avoid that confusion.
Kinda late on this, as I was away on a long trip :) Thanks for the great explanation. Will definitely read Zoltan's piece and your substack posts, which are high on my to-read list.
On a side note: do you know any place where one can find a systematically organized archive of Zoltan's pieces?
Interesting angle! I have not thought about it (and frankly haven't spent much time at all thinking about a central bank digital currency). Perhaps a good idea to explore in a future week!
Loved it. I've listened to heaps of Jeff Snider so I'm extremely familiar with the bank reserves are not money argument. Thus it was interesting to read your piece and get a bit of perspective on how those reserves can seep into the economy. It your article is correct, this situation we are facing could be even worse than imagined.
As with any complex topic like this, the truth is probably not black and white, and there is disagreement among people who spend their whole life studying it. Is QE money printing? In my opinion - yes, but with the relevant caveat that it does not immediately translate into the most liquid forms of consumer spending immediately on its own. Combining QE with fiscal is a different beast because it puts those deposits immediately in people's pockets, and could be harder to reverse as a result.
The reality is these tools are so new (and extreme) that we can't say with certainty what impact they will have - we have to look to the data as it evolves.
Sure, as an example, an individual could buy a UST directly from TreasuryDirect.gov, by transferring cash from a checking account. More likely, this would be happening between the the bank account of an institutional investor and a market maker like a primary dealer.
"tackling inflation would be easier if the Fed could pull money out of people’s checking accounts"
The Fed can't do this, but the Feds certainly could, through the mechanism of a wealth tax or raising marginal rates. Your entire argument (which I certainly agree with) offers robust support for managing the economy more through direct fiscal transfers and taxation; much less through the Fed and rates; and not at all through QE. Replace modern monetary policy with old-school fiscal policy, in other words.
Taxation is the opposite of stimulus - if stimulus is a contributor to inflation than taxation would be a helpful way out. The trouble of course is that politics doesn't necessarily abide by those rules. Ideally, the Fed is the "adult in the room" that doesn't have to bow to political pressure and can to the unpopular things when necessary.
The Fed's job is easier if fiscal policy is rowing in the same direction.
TLBS, you are a true gem in a sea of dull rocks.
It is complex questions that I am after to explain the macro world we live in, and everyone seems to have dull answers to questions I didn't ask. Your writing is data driven, intelligent and with a different perspective. Makes me ask the questions that make me better prepared for economic events.
Is the fed/government more afraid of inflation or deflation/depression? - (it depends). I have never been more convinced of the trajectory of the economy when I realized that the difference between previous downturns and this one, is severe inflation. And you hit the nail on the head on this question in this article.
Or maybe I am just a perma-bear. Either case, these articles are one of my favourite economic sources, and recommend to anyone interested in understanding things a little better.
Cheers! (Phil)
Thanks for the kind words, Phil, and I hope you will keep reading.
In recent history, inflation has not been a challenge and so it has receded on the list of central banking worries. But over the long term the Fed is more afraid of inflation, because its harder to tackle. As Bernanke said, the cure to deflation is the money printer. This has been the paradigm of central banking for the last several years, and it has been a long time since there was a serious inflation problem to worry about.
Now, inflation is concern number 1, compounded with second guessing of recent policy mistakes.
Dear Mr.Bear,
The Furnace was my first exposure to your writing. I would like to tell you that after reading many explanations of why QT is dangerous, including The Fed Guy, I found yours to be the most clear, concise and easy to understand.
The connection between deposits and liabilities at the banks and the Repo market is one that has been alluded to a great deal – but the simplicity of your explanation made it very apparent to me the corner that the fed could find themselves in.
I am concerned greatly that any backstopping of the Repo market will be considered a pivot by market participants who are very focused on the currency implications of higher rates, at the expense of tamping down inflation.
I imagine you are well versed in what lead to the great depression, I was not familiar with the nuance until I read this recent history of the fed on Bloomberg. Please note: II.The Great Depression https://www.bloomberg.com/news/features/2022-08-15/how-the-federal-reserve-responded-to-8-economic-crises?sref=PHnbDlDG
Is it possible that if the fed backstops the repo market to ensure liquidity that the result is a much higher stock market, resulting inflation and corresponding longterm severe financial issues or am I over simplifying the problem?
appreciate any insight you could provide...now i am off to read the rest of your substack.
thanks again for the thinking and the writing
simon
Thanks Simon!
I would argue that the scenario you are asking about (Fed intervention leading to higher stock market, which in turn creates risk of popping), has been what's already occurred since 2008, and supercharged over the past two years.
From your linked article "From the middle of 1928 to its peak in September 1929, the Dow Jones Industrial Average soared 90%."
Well from March 2020 to December 2021 the Nasdaq 100 soared 140%.
That *doesn't* mean I think we are on the verge of a depression, but the dynamics between loose fed policy and a soaring asset bubble have already taken place (in stocks, housing and bonds!)
Thanks again for reading, and I hope you enjoy the column.
Great write up as always Bear! I just had one question:
Bank cash went up by 10x between 2008 and 2014 from the previous QE, but demand deposit did not have a similar spike. As I understand from Down the Drain, QE provide cash to owners of financial assets who then deposit their cash into commercial banks. Would this deposit not increase the demand deposit? If so, then why do we not see demand deposit following the path of bank cash?
The easy rule is that if the Fed is buying securities from an actual commercial bank, than there is no change in money supply or demand deposits - it is just an "asset swap". This was the case for the "original" QE, was primarily a bailout of the banking sector in 2008. However, in the later periods of QE in 2011 - 2014, you see that there was a significant uptick in deposits which corresponds with the Fed's asset purchases. In the COVID era it was almost 1:1.
Hey Bear, great piece. Such a pleasure to always read the pieces AND the interaction between you and your readers! Keep up the great work :)
You wrote:
"On the short end, government money market funds meanwhile have reduced their holdings of USTs in favor the RRP."
Being a long time reader of yours, I remember you writing in your piece "Down the Drain" that the Fed essentially chose to pay interest to MMFs that park funds in the RRP. Specifically, what drew my attention back then was the phrase (in quotes): " This week’s rate hike increased the RRP rate from 80bps to 155bps. With the 1-month UST yielding just over 105bps, there is now 50bps of incentive to sell UST and allocate to the RRP."
The Fed rate is currently at 3-3.25% and the 1-month T-bill yields 2.67%, so essentially the arbitrage opportunity is around the same. as it was then. This implies that MMFs would still be incentivized to forgo buying USTs in favor of truly riskless RRPs.
What I do not get is, why would the Fed chose to do so? Wouldn't it be convenient for the Fed to stop raising the interest rate in the RRP with the same amount that they do during each rate hike? This way, the "truly riskless" RRP would yield less than T-bills at some point and this provide an incentive to some MMFs to gradually move funds from the RRP right into USTs. As I understand it, there is around 2-2.5 trillion USD worth of liquidity parked in RRP at the moment which could help absorb liquidity shocks.
And then, I would like to hear your thoughts on whether USTs could be used as a weapon from foreign holders of USTs against the US, as part of an economic war if things were to escalate. For example, China holds around 13% of USTs (according to this site ----> https://ticdata.treasury.gov/Publish/mfh.txt). In a scenario where China decides to respond back to US waging an economic war to them (tarrifs, ban of EUV machines, ban of export of high end chips, delisting of certain entities from US stock markets) and escalate by dumping their USTs in the market, wouldn't that further exacerbate the UST funding problem you mentioned in your piece? And what would be possible scenarios from there onward? I suppose funds parked in RRPs could act as a backstop against that, if I have understood the mechanics correctly.
Happy to hear your thoughts on the above.
First - yes you are correct that the RRP award rate after last week's rate hike is much higher than the current 1Mo UST rate which will cause money market funds (MMFs) to reallocate to the RRP. They will likely do this by reallocating their exposure once the 1mo bills come due, rather than outright selling them, which causes it to be a more gradual transition (rather than immediate selloff to parity). This should continue to occur over the coming months.
Why does would the Fed do this? Because need the RRP to make sure that short term rates meet their target, which today is 3.05%. They injected a ton of liquidity over the past two years to reduce rates, and now the only way raise rates to their target is to let all of that liquidity be released into the RRP.
This may not sound like a huge problem until you realize that there are things that were built on all those dollars being around that aren't going to work when they are removed. You've already seen this in the most speculative area's of the market (think crypto or unprofitable tech), but will become more problematic when it becomes bank balance sheets.
As to the last point on economic weapons. We must acknowledge that economics and markets are an important lever in geopolitical power. For example, the West collectively have used their sanctions as a financial weapon against Russia - Russia has responded by cutting energy supply to europe, and there are big consequences for each as a result. This same concept could apply to countries that hold USTs, and they could certainly wield them in an unfriendly way. Russia, for example, sold out of all of its treasuries over the course of the last decade likely to avoid the brunt of future sanctions like the ones that were imposed for recently.
Hey Bear, late reply here as I have been away on a long trip. Thanks for the explanation, helps a lot.
Regarding liquidity and RRP - do you think then that in the event of a true UST liquidity event, the chance of which seems to be growing as days pass, the flows of funds parked in the RRP could easily reverse back into USTs if the Fed decided to do so? And do you think reversing their decision to have an RRP yield that follows the rate hike cycle would damage their credibility?
An additional problem I suppose is that the Fed might not actually know the real size of the Eurodollar market anyway. As I understand it, they do know the amount of debt in relation to USDs issued within the US jurisdiction, but they could be off in their estimates of the Eurodollar debt market.
And then, regarding the use of economic warfare. I suppose a strengthening dollar and the liquidity problems that creates in emerging markets is in itself a weapon the US can use to influence political leaders within those markets. The US can probably use that as a very efficient weapon but need to do that elegantly, to avoid a "loss of trust" event.
Beautiful analogy. I will now recommend this to anyone who wants to understand this (arguably complex) topic. Humor me, though. Adding too much fuel creates a lot of smoke because of incomplete combustion (which is the goal, so that you can remove the fuel later if its too much). In your analogy, what would the smoke be?
Ha! I'm not sure - obviously the analogy is only a crude representation so I'm trying to think how the comparison could apply, and I'm not coming up with much. I'm not sure if there is a "waste product" type that would be continue to the analogy, but happy to hear if you could think of something.
Loved the real life analogy of your grandbears’ fire place Bear.
The reason you developed those skills is because either the wood was not infinite,and comes at a cost,and of course Bears like the temperature ‘just right’
Assume the wood heap (coal)is infinite in your application.
So is it that all the doors and windows are open,and despite burning at a high rate,no warmth is accumulating to allow the fire to be turned down?
Thank you
I almost made this point about how the analogy doesn't hold because we had limits to the wood, where the fed can print to infinity. Yet, while there is no upper limit of the Fed's balance sheet, the "cost" comes in inflation or currency devaluation.
Another fabulous post. I super enjoy reading your blog. A few comments:
1. At the risk of being that guy...(though in truth i kinda am that guy :) https://johncomiskey.substack.com/p/projecting-qt-balance-sheet-impact) QT pace will almost certainly never reach 95b. A more realistic go forward pace given MBS prepayment speeds is 80b.
2. Agree that the declining bank reserves set the stage for sept 2019, though in combination with intraday bank liquidity requirements established post GFC. Fabulous paper on the subject. https://www.nber.org/system/files/working_papers/w29090/w29090.pdf . One notable difference sept 2019 to now though is the Fed became a repo lender on an emergency basis in 9/2019, there was no standing facility/systematic liquidity relief valve to police the top end of the range like there is now with the SRF.
3. I generally think all the liquidity being "sterilized" in the RRP will eventually transfer to cover the liquidity hole caused by QT. I think the govt. MMFs are parking the money in RRP because its often a better play for yield (not to mention its effect on MMF liquidity requirements) than short term UST given it keeps increasing by 75 bp every 6-8 weeks. When we get to the terminal rate I think that dynamic shifts and WAMs of the MMF will get much closer to their 60 day max. Even before then though, If repo rates rose above RRP due to funding stress I dont know of any reason why those MMFs wouldnt shift deployment of their cash to grab the extra yield.
4. On treasury liquidity. As a technical matter I dont think there can be insufficient liquidity in the primary market. As I understand it the primary dealers have no choice but to provide it even if nobody else bid, though that doesnt mean the yield cant be super high. On the secondary market. I think the MMF/RRP liquidity would eventually deploy since if price of UST got low enough an attractive carry trade funding through repo from MMF (at rate higher than RRP) would develop for entities with the B/S space to do it. SRF probably an eventual backstop here too
That said, I am very new to the world of finance systems (i literally first learned how QT worked in relative detail from reading dispatch 8 of your blog in June) so its possible im saying ill informed things.
Thanks again for your posts.
1. Interesting point on the maturity schedule vs. caps that I will have to check out.
2. Yes the SRF will help (which was also referenced in the post here). But it doesn't cover all market participants. Further, the problem in general with Fed liquidity facilities is that no one will use one for convenience - they will be forced into them after things go south, so while it is still helpful, it is not fully preventative.
3. I've written quite extensively on the RRP - I think its by far the best asset (assuming equal yields) because of its duration, counterparty. If 4 week tbills start trading well above the RRP expected rate, than MMFs may start to buy t bills. But again, this assumes that everything remains orderly. If there is some blowup, will MMFs be the one to step in to save the day with liquidity? I doubt it.
4. As a practical matter, there can absolutely be insufficient liquidity in the treasury market. It has happened several times before and forced the Fed to intervene to support prices. Its hard to see how the treasury market is in better shape than it was at those points in time.
In theory you can always say that there is a "carry trade" to be had in a liquidity blowup, and yet they still happen. I think it has the causation backward though. Prices go down because people are trying to access liquidity and no one is willing to provide it. If there was a ton of liquidity waiting on the sideline to pounce on treasuries, then the treasury market wouldn't have blown up in the first place. The dynamics are a bit different today because of "idle" liquidity at the RRP, but MMFs could only step into short dated treasuries, and I doubt they would choose to in a moment of crisis anyways.
In any case, these are all great points, and thanks for reading!
Hey John, as a person trying to understand how financial plumbing works I can say that your posts looks great and have just been added in my to-read list.
Just a question though. You wrote: " Even before then though, If repo rates rose above RRP due to funding stress I don't know of any reason why those MMFs wouldn't shift deployment of their cash to grab the extra yield." Does this mean that the repos can have different yields than reverse repos? But how is that possible if a reverse repo is the unwinding (maybe settlement is a better word?) of a repo purchase?
My understanding was that each single repo transaction is associated with a reverse-repo transaction. Specifically:
1) A repo purchase is the act of selling an asset to a counterparty with an agreement to buy back that asset at a later point in time at a specific price.
2) The duration is typically quite short (1 day) and that is rolled over on a day-to-day basis.
3) Repo transactions do not officially pay a yield, as they are not loans. However, the system is kinda "rigged" as the fact that the price of repurchase is agreed a-priori, and there is a duration in the transaction, essentially is a form of yield.
Could you please point out to me where/if I am wrong on any of the above?
Thanks in advance!
Hi Cat,
Thanks for the kind words.
Just nomenclature. In both instances the MMF plays the role of the Repo buyer/lender (considered the reverse side of the transaction) with either:
1. The FED on the other side of the trade as the Repo seller/borrower via the RRP facility or
2. Some Repo mkt participant who has collateral and needs cash who is the Repo seller/borrower in that instance (usually the trade is intermediated by a dealer)
If the MMF lends to the FED, its at the RRP rate. If the MMF lends to some other Repo mkt participant its at some mkt rate broadly measured by SOFR though the rate complex under the hood is more complicated (highly recommend reading Zoltan Poszsars Global Money Notes #25 https://research-doc.credit-suisse.com/docView?language=ENG&format=PDF&sourceid=em&document_id=1081711591&serialid=MLFU%2FU7Do9ID7nDETfx4syeW2JsAgCxrRqbX5Uk9Gjo%3D&cspId=null )
Your understanding is generally correct, though I would change "repo purchase" in 1 to "repurchase agreement" or just "repo". Repo Purchase implies to me the purchase of the collateral/lender of money which is actually the reverse side of the trade. Best to avoid that confusion.
Hope that helps!
Hey John,
Kinda late on this, as I was away on a long trip :) Thanks for the great explanation. Will definitely read Zoltan's piece and your substack posts, which are high on my to-read list.
On a side note: do you know any place where one can find a systematically organized archive of Zoltan's pieces?
Does the Fed's current limited capacity to rein in deposits demand further legitimize (in their eyes) the need to institute a CBDC conversion?
Interesting angle! I have not thought about it (and frankly haven't spent much time at all thinking about a central bank digital currency). Perhaps a good idea to explore in a future week!
Loved it. I've listened to heaps of Jeff Snider so I'm extremely familiar with the bank reserves are not money argument. Thus it was interesting to read your piece and get a bit of perspective on how those reserves can seep into the economy. It your article is correct, this situation we are facing could be even worse than imagined.
You may also be interested in some prior posts that cover this topic in more depth.
https://thelastbearstanding.substack.com/p/qe-for-dummies
As with any complex topic like this, the truth is probably not black and white, and there is disagreement among people who spend their whole life studying it. Is QE money printing? In my opinion - yes, but with the relevant caveat that it does not immediately translate into the most liquid forms of consumer spending immediately on its own. Combining QE with fiscal is a different beast because it puts those deposits immediately in people's pockets, and could be harder to reverse as a result.
The reality is these tools are so new (and extreme) that we can't say with certainty what impact they will have - we have to look to the data as it evolves.
Good point about these tools being new. I agree, not black and white. Thanks for the article recommendation, I'll check it out.
Sure, as an example, an individual could buy a UST directly from TreasuryDirect.gov, by transferring cash from a checking account. More likely, this would be happening between the the bank account of an institutional investor and a market maker like a primary dealer.