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.
No this is the TGA... TTL accounts have been zero since the zero rate policy put on as a result of the GFC... as the govt cant earn any interest on the govt balances with the rates at zero (to them)...
So when they jam these reserves down the throat of banks in short periods of time like they have recently done for $400B in short order, the question is can banks react with the same frequency response in their regulatory capital account to accommodate the new reserve assets?
I dont think they can.
Its not like they can refuse someone who wants to deposit a government check for $400B...
The frequency response of bank assets is much higher than the frequency response of regulatory capital so you have a classic frequency mismatch from Systems Theory standpoint... BAD!
Obverse, when they ran the TGA up in late 2016 by 100s of $Bs, that opens up some room on the balance sheets of banks to add more loan and securities assets... so we saw good quarters for the banks at the end of 2016... lots of autos, etc... so we had a decent 3rd and 4th qtr in 2016 with 2-3% + on the NIA GDP ...
Now this has reversed.
Mike is tracking a collapse in bank assets held for trading recently in his report... they are having to shed other assets and cut back on new lending as they may have moved towards their asset limits with this govt policy in late 2016.... growth in loans and leases is headed towards the flatline...
This is what is causing the current drop in GDP growth here in 1Q with GDP growth looking lucky to be positive at all...
Growth in leading flows ex-interest income looking shitty also... barely positive at all...
No new highs in stock indexes in over a month.... :(
Banks might need to add 25 to $50B of new capital or so to get things going again... might have to wait until after the 1Q earnings come out and see how much they have left over to put back in... perhaps raise some more from outside...
Still seems only hope we have is rates increasing for the rest of the year increasing interest income and maybe if they start to draw down the Fed factors effecting the reserve balances that would open up some room on asset side...
More "muddle thru" for now seems like banks have to build up the capital accounts for a while..
There has been a rise in delinquent mortgages, auto loans, leases and C&I loans. low but rising.
I get the point here along with the consequences for banking but also some perspective on how small 400bln in new assets are compared to the capital and assets of the banking system. JPM averages around ~60bln a year in operating cash flow. That is one big bank, but many more too.
I Showed it to Kim ( MMT) who used to work with these sort of things in the UK.
His response was...
I don't know what that account means (why would the US Treasury have so much on deposit with commercial banks?). Sharp fall, but what was behind the sharp rise in 2016?
It's such a huge amount that I suspect that there's some operational rather than fundamental reason. For example, was the 2016 expansion of deposits offset by repos, which would reverse the deposits and have the effect of injecting US Treasury securities into the market to use as collateral? Ie a bond lending operation where the lending fee is the spread between the repo and the deposit rates.
Though I'd have expected the Fed to operate such things, using its QE portfolio. Has such an operation just been shifted from the Treasury to the Fed? Some bureaucratic thing?
Ryan they have to pay dividends out of that etc...
If they add capital I'm sure it's not some low level decision it's somewhat of a big deal and those decisions can take time... but perhaps they can shed other assets pretty quickly if they have to..
I think rather this represents the amount Treasury has on deposit at the Fed rather than commercial banks... I believe the TGA is at the Fed...
"what was behind the sharp rise in 2016?"
They increased the rate of UST securities issuance well in excess of what they were withdrawing from the TGA so USD balances built up in the TGA.... "debt ceiling" deadline imminent March 15th? Who knows...
Certainly this was unprecedented in scale or otherwise if you look at the chart...
Go back and look at RESBAL in late 2008, it was 9B on Sep 10th and a month later it was like $300B...
so they stuffed 300B of reserves onto banks balance sheets in a month causing liquidations in the other bank assets and the whole system locked up and caused the whole gd GFC...
Matt, this is probably a silly question, so apologies in advance! What you guys are discussing sounds fundamentally important, but I don't understand a word of it. Where can I find out in detail, but in an easy Dummies-style way, about this stuff? Or is it yet another one of those things (the economy rather than fancy models) that mainstream economics and finance have no interest in, and so searching it out in the mainstream is pointless. Is it a case of finding out understanding the mechanics of these accounts (how and when they're debited, credited, etc), or is it just a case of simply understanding that these accounts work and then applying a bit of MMT and MMT-friendly finance?
I can probably answer questions but I currently don't have the communication skills to write it up appropriately.... if you (or anybody) have questions about my opinions here just ask I'll do my best....
Matt, I don't even know where to start! MMT-friendly finance is pretty much only available at one place - here! A friend of mine has Magic Mike's "Understanding the Daily Treasury Statement". I'll have a gander at that and see if that allows me to even articulate the problems and then try and get back to you! By the time I get back to you, you won't even remember what the hell i'm talking about. Many thanks for the offer.
John, Mike's DTS course is worth its weight in gold. 5-hr video.
Also read Frank Newman's "Freedom from National Debt." 87 pages. But it took me four months to read it. Altho' in easy English, I took a day/page to comprehend it.
MRW, thanks. I read both of Frank Newman's books last year in hospital. They may be like Warren Mosler's books, in that there is far more information to ponder than is immediately obvious, and, as you yourself have noticed, Newman's style of delivery makes you read the same sentence again and again, leaving you wondering is it the prose or the subject matter that's the problem. Plus I was as high as a goddamn kite! It was difficult judging how MMT-friendly Newman is. I did note that as often as he seems a paid up member of the MMT community, there did seem to be significant departures.
Can you break down what you said here so that it is easier to understand.
"Go back and look at RESBAL in late 2008, it was 9B on Sep 10th and a month later it was like $300B...
so they stuffed 300B of reserves onto banks balance sheets in a month causing liquidations in the other bank assets and the whole system locked up and caused the whole gd GFC...
1. Why does increasing bank reserves cause liquidations in other bank assets? How does that relate to 2008-9 subprime crisis? 2. Based on the graph above how come there wasn't a crisis in 2011 and 2013-2014 when there was significant increases in reserve balances.
I don't understand the capital issue. Reserves forced on the banks don't affect their capital positions; they are zero risk-weighted and (in the UK at least) excluded from the simple leverage ratio calculation (total assets divided by equity, regardless of asset quality).
Capital Adequacy Ratio is a minimum (which an aside for this discussion, but mentioning it for completeness's sake) and doesn't include reserve assets.
If government spending created excess reserves in banks and that forced them to not lend because their assets went up while their capital remained the same, it can't be solved by banks selling other assets for cash i.e. more reserves.
The reserve position of a bank is completely irrelevant to a bank selling assets for reserves. Since reserves don't count towards assets while computing the capital adequacy, a bank selling non-reserve assets for reserves allows it to meet capital requirements without infusing additional capital. This is exactly what they did during QE, they offloaded their non-reserve assets to the Central Bank, got reserves for them reducing the assets a percentage of which they have to hold as capital and thus meeting the capital requirements.
Some more points of view from guys who have worked in banking in the UK.
The Fed has less than 20% of the outstanding treasuries on its balance sheet. In any case when banks are swimming in excess reserves, one bank falling short of the capital adequacy ratio can sell non-reserve assets to another whose capital position is good. If all the banks are short of capital, that has nothing to do with the government spending leading to increased reserves causing a capital shortfall because banks don't have to hold any capital against reserves, only other kind of assets that can go down in accounting value (reserves can never go down in accounting value). If all banks are under capitalised, the Fed may have to run another QE operation buying other kinds of assets than Treasuries. They bought MBSs last time, not sure why can't expand the category of assets they take on their balance sheet.
You may have discovered a correlation, but that is not causation. The debt limit has absolutely nothing to do with banks selling non-reserve assets for reserves.
Maybe I misunderstand, but it's almost as if we've gone from "reserves increase bank capacity to lend" to the absolute reverse!
Both are madness.
Absolutely. Reserves have absolutely no role in the banks' capacity to lend, not since currencies stopped being backed by metals. They neither increase nor decrease that capacity. Their capacity to lend is solely a function of their capital, the risk-weighted assets they already have on their balance sheet and the capital requirements set. If the capital requirements are relaxed, they can lend more, if they are tightened they cannot lend more and may even have to get rid of the risky assets.
I am not sure, but if there is an urgency to sell assets on the banks' part it may have to do with the BASEL 3 rules, some of which came into effect this year.
What is the 'Tier 1 Leverage Ratio' The Tier 1 leverage ratio is the relationship between a banking organization's core capital and its total assets. The Tier 1 leverage ratio is calculated by dividing Tier 1 capital by a bank's average total consolidated assets and certain off-balance sheet exposures. Similarly to the Tier 1 capital ratio, the Tier 1 leverage ratio is used as a tool by central monetary authorities to ensure the capital adequacy of banks and to place constraints on the degree to which a financial company can leverage its capital base.
Then it further defines 'core capital' as:
"Core capital consists of equity capital and declared reserves."
Are 'declared reserves' the same thing as 'reserves'? Or does a bank have to segregate a certain amount of its reserves as 'declared' via bank internal policy for any regulatory period?
So apparently you have two different regulatory ratios at work here, you have the Capital Ratio and the Leverage Ratio...
I think what I am talking about is the Leverage Ratio...
Read more: Tier 1 Leverage Ratio http://www.investopedia.com/terms/t/tier-1-leverage-ratio.asp#ixzz4dTCsTKHY Follow us: Investopedia on Facebook
Decrease in wholesale funding and/or increase in assets (It buts debt i.e. US Treasuries). Usually a combination of both. For every $8 in deposits a bank requires $1 in US treasuries and its capital ration remains unchanged.
You make a classic error in believing that Central Bank rates go up, cost to consumer goes up. In the early part of the raising cycle the inverse is true.
"21. Includes vault cash, cash items in process of collection, balances due from depository institutions, and balances due from Federal Reserve Banks."
'balances due from Federal Reserve Banks' I view as aka 'Reserve Balances'
So if these "balances due from Federal Reserve Banks' are treated just as any part of bank consolidated assets and they are not automatically part of 'declared reserves' at a bank upon receipt, then to maintain a constant Leverage Ratio, imo it appears a bank would have to liquidate other assets if bank deposits from TGA were to spike ....
"and would be why the Fed is looking at balance sheet action."
Well yes they should very well be as they (to them) are probably freaking out and thinking they are soon going to be "out of money!" and bankrupt if this continues...
Well yes they should very well be as they (to them) are probably freaking out and thinking they are soon going to be "out of money!" and bankrupt if this continues...
Or they are just going to reduce QE. They are not paying banks to hold funds so all the interest they receive from the Government is paid back. The vast majority of those assets on the Fed's Balance Sheet are therefore not needed.
Banks are paid a deposit rate to hold funds at the Fed. The Fed pay the deposit rate from the interest they earn from the assets on their balance sheet.
The remainder of unpaid interest is returned to the Government so the Fed balance sheet doesn't continue to grow. If it is continually returning interest to Government the balance sheet is way to big.
The other thing is you have to assess is if the balance sheet reduction would necessarily result in increased net UST securities issuance which btw is currently impossible as they are at their "debt ceiling!"...
Matt: "imo it appears a bank would have to liquidate other assets if bank deposits from TGA were to spike ...." Hunh? Bank deposits from TGA spike every time there's a Congressional appropriation, and the Fed dispenses them to vendor accounts. Two weeks to a month later (in the main), the US Treasury issues treasury securities in the amount of the appropriation (in the main), and those are sold at auction to the public. Money supply restored to balance.
The most important economic fact you should ever know; "People do not make decisions based on statistics"
A cycle last s as long as it does, past cycles are not a guide to longevity.
We are still in the early stages of phase 2, way too early to call time for me. I am still awaiting animal spirits in the third phase and capitulation in the forth
""imo it appears a bank would have to liquidate other assets if bank deposits from TGA were to spike ...." Hunh? "
USD balances in the TGA are not counted as reserves.... so when Treasury ran the TGA up to over 400B in umprecedented fashion last year that would have the effect of reducing reserve balances on bank balance sheets.... you could see this on the H.8.... , then disgorged them in about a month, it had a short term effect of increasing reserve assets on bank balance sheets by 400B in a very short period of time as the Treasury didnt do a 'reserve drain' by issuing bonds before they withdrew from the TGA... they in effect did the 'reserve drain' well in advance by running up the TGA by 400B+ over the previous year...
Withdrawals from the TGA add system reserves if they are not first offset by Treasury issuance $4$ in advance...
In normal circumstances, the Treasury does a reserve drain (by issuing bonds) BEFORE they do the reserve add via TGA withdrawal...
Also we have reserve balances very high (see Tim's statement " If it is continually returning interest to Government the balance sheet is way to big.") with the QE which is also unprecedented... so 400B of that ended up in the TGA last year by UST issuing bonds well in advance of necessary withdrawals....
If the sharp increase in reserve assets caused by this TGA policy effected the regulatory "Leverage Ratio" then that would perhaps be a problem for banks if it was not well coordinated between Treasury and the member banks beforehand...
I have to look further into the difference between Capital Ratio and Leverage Ratio from the standpoint of 'reserves' and 'declared reserves'....
A leverage ratio is a specific type of capital requirement—it does not take risk weights into account. Or put another way, a leverage ratio is a capital requirement that treats all assets as if they had 100% risk weights. If the leverage ratio for a bank is 5%, then it would be required to have $1 of equity for every $20 in assets regardless of the riskiness of those assets."
"three separate regulatory bodies, the FDIC, the Federal Reserve and the Comptroller of the Currency, review and restrict the leverage ratios for American banks. "
Are reserves specifically considered assets of the commercial banks, such that they would be considered part of their (commercial banks) cash assets? Anyone know?
”In normal circumstances, the Treasury does a reserve drain (by issuing bonds) BEFORE they do the reserve add via TGA withdrawal...
Not according to Frank Newman, former Deputy SecTreas in his latest book. He said they are done AFTER for a bunch of reasons; namely, that sometimes they dont need to issue bonds in the full amount of the congressional appropriation (TGA withdrawal amt to vendors, etc).
And Warren is fond of repeating the Federal Reserve/US Treasury mantra: “Reserve add before reserve drain.”
When the Treasury distributes funds, the nation’s deposits are initially increased. Where can the bank money go? Let’s look at an example, excluding the portion covered by taxes. Typically, before the Treasury issues $ 20 billion of securities, the government has distributed $ 20 billion to the public from its account at the Fed: redeeming maturing Treasuries, paying companies that provide goods and services for the government, for payments to individuals, etc. Many investors simply “roll over” their Treasury securities, replacing maturing ones with newly issued ones, and taking just the interest. For example, perhaps $ 10 billion of the $ 20 billion issue might be in that category. The Treasury pays out the other $ 10 billion to the private sector. At that point, a set of participants in the U.S. financial system will have the extra $ 10 billion in their bank accounts and will look to place those funds. The money supply has been increased by $ 10 billion, and the new dollars move around within the overall US financial system. All the Treasuries previously available are already owned by investors, and prior auctions had demand that exceeded the amount offered. As the new Treasuries are auctioned, the demand is filled by exactly the $ 10 billion offered, and the money supply returns to its prior level. In the whole of the U.S. financial system, the only place to put the money is into the new Treasuries that are being auctioned— or otherwise just leave the funds in banks.
Newman, Frank N. (2013-04-22). Freedom from National Debt (pp. 18-19). Two Harbors Press. Kindle Edition.
MRW we need to keep looking into this #1... I'm going to still assert I'm correct here for now....
I'm talking about BANK reserve assets reported on the H.8 while Newman is talking about "money" which is a figure of speech.
My language is more precise and scientific than Newman's which is required in order to do an acceptable analysis....
What do you think that when the FRS/OCC/FDIC do their regulatory work they go into a bank and say "Hey! How much money do you guys have?!"
Please.
USD balances in the TGA do not even count as bank reserve assets so what is Newman even talking about?
USD in GSE accounts do not exist as bank reserve assets so again what is he talking about?
Right now, Feds H.4.1 has total factors effecting reserve balances at Iirc like 4.3T, yet RESBAL is at about 2.3T so where are the other 2T of "reserves"?
They are either in the TGA or with the GSEs... maybe govt employee retirement account?.... NOT with the banks as banks only have 2.3T....
Ok , Newman: "When the Treasury distributes funds, the nation’s deposits are initially increased."
Ok, what happens here, AT THE BANK?
Some social security person receives a Direct Deposit or wtf...
At the bank:
On the right they increase "customer checking account payable", AND on the left they increase "reserve assets"...
IF, and I repeat IF, the bank has fixed capital for the period in question, then the banks ratio of total assets to capital has just increased over said period...
Else equal, to maintain a constant Leverage Ratio at 20:1, banks would need to have added an additional 100B to their capital accounts to comply... or reduced other non-reserve assets by same amount...
So then let's look at the effect of ending the QE, if the effect of that would be to reduce bank reserve assets from current 2.5T to perhaps 500B, then that would allow banks to put an additional 2T of assets on their books to maintain a constant Leverage Ratio with no additional capital needed...
Plus 100B+ of interest income going back to savers instead of back to Treasury....
USD balances in the TGA do not even count as bank reserve assets so what is Newman even talking about?
USD in GSE accounts do not exist as bank reserve assets so again what is he talking about?
Right now, Feds H.4.1 has total factors effecting reserve balances at Iirc like 4.3T, yet RESBAL is at about 2.3T so where are the other 2T of "reserves"?
They are either in the TGA or with the GSEs... maybe govt employee retirement account?.... NOT with the banks as banks only have 2.3T....
The mistake is counting entires on the side of government as money when they don't appear as entries in any of the measures of money in the US e.g, M0 (base money, HPM), M1 and M2.
The only government liabilities that appear in the measures of money in the US are Fed liabilities as either settlement balances in nongovernment accounts (this includes Treasury deposits in the TT&L accounts) and cash (vault cash and currency in circulation).
(Actually, the Treasury does "have some money" in the form of the TT&L deposits that count toward bank reserves.)
the Fed doesn't "have money" since it issue the currency.
So ending QE gives more interest income to savers and allows additional bank lending. Is that what Tim Browning was getting at when he says early interest rate rises DO NOT lead to credit tightening but instead to looser credit?
Blogger Matt Franko said...
So then let's look at the effect of ending the QE, if the effect of that would be to reduce bank reserve assets from current 2.5T to perhaps 500B, then that would allow banks to put an additional 2T of assets on their books to maintain a constant Leverage Ratio with no additional capital needed...
Plus 100B+ of interest income going back to savers instead of back to Treasury....
#MOONSHOT
Blogger Tim Browning said...
Let me set you on another path.
Do you think early rate rises lead to credit tightening.
Look at past cycles, I will assure they do not. In fact what you tend to see is increased bank margins and looser credit.
Former Dept. Secretary of the U.S. Treasury Says Critics of MMT are "Reaching" - New Economic Perspectives http://neweconomicperspectives.org/2013/10/former-dept-secretary-u-s-treasury-says-critics-mmt-reaching.html
Not sure if this will be read, but I'd really like to get in touch with Matt to ask a clarifying question about the reserve effect on bank leverage ratio.
My question is when Fed increases reserve balances it seems that bank assets rise (reserves go up), but what happens to liabilities and capital? Do they stay the same or does capital/equity also go up since assets (reserves increased) and Assets = Liabilities + equity. Apologies if this is a novice question but just trying to make sure I'm on the same page with some of the excellent info you have posted. Thanks!
51 comments:
Why the sharp drop after Nov 2016? Looks very suspicious to me.
Looks like drawing down the TT&L accounts?
No this is the TGA... TTL accounts have been zero since the zero rate policy put on as a result of the GFC... as the govt cant earn any interest on the govt balances with the rates at zero (to them)...
See here Tom ytd:
https://www.fms.treas.gov/fmsweb/viewDTSFiles?dir=w&fname=17040300.pdf
"Transfers to Depositories": zero ytd...
So when they jam these reserves down the throat of banks in short periods of time like they have recently done for $400B in short order, the question is can banks react with the same frequency response in their regulatory capital account to accommodate the new reserve assets?
I dont think they can.
Its not like they can refuse someone who wants to deposit a government check for $400B...
The frequency response of bank assets is much higher than the frequency response of regulatory capital so you have a classic frequency mismatch from Systems Theory standpoint... BAD!
Obverse, when they ran the TGA up in late 2016 by 100s of $Bs, that opens up some room on the balance sheets of banks to add more loan and securities assets... so we saw good quarters for the banks at the end of 2016... lots of autos, etc... so we had a decent 3rd and 4th qtr in 2016 with 2-3% + on the NIA GDP ...
Now this has reversed.
Mike is tracking a collapse in bank assets held for trading recently in his report... they are having to shed other assets and cut back on new lending as they may have moved towards their asset limits with this govt policy in late 2016.... growth in loans and leases is headed towards the flatline...
This is what is causing the current drop in GDP growth here in 1Q with GDP growth looking lucky to be positive at all...
Growth in leading flows ex-interest income looking shitty also... barely positive at all...
No new highs in stock indexes in over a month.... :(
Banks might need to add 25 to $50B of new capital or so to get things going again... might have to wait until after the 1Q earnings come out and see how much they have left over to put back in... perhaps raise some more from outside...
Still seems only hope we have is rates increasing for the rest of the year increasing interest income and maybe if they start to draw down the Fed factors effecting the reserve balances that would open up some room on asset side...
More "muddle thru" for now seems like banks have to build up the capital accounts for a while..
There has been a rise in delinquent mortgages, auto loans, leases and C&I loans. low but rising.
I get the point here along with the consequences for banking but also some perspective on how small 400bln in new assets are compared to the capital and assets of the banking system. JPM averages around ~60bln a year in operating cash flow. That is one big bank, but many more too.
Sratching heads in the UK
I Showed it to Kim ( MMT) who used to work with these sort of things in the UK.
His response was...
I don't know what that account means (why would the US Treasury have so much on deposit with commercial banks?). Sharp fall, but what was behind the sharp rise in 2016?
It's such a huge amount that I suspect that there's some operational rather than fundamental reason. For example, was the 2016 expansion of deposits offset by repos, which would reverse the deposits and have the effect of injecting US Treasury securities into the market to use as collateral? Ie a bond lending operation where the lending fee is the spread between the repo and the deposit rates.
Though I'd have expected the Fed to operate such things, using its QE portfolio. Has such an operation just been shifted from the Treasury to the Fed? Some bureaucratic thing?
It's worth finding out, that's for sure.
Ryan they have to pay dividends out of that etc...
If they add capital I'm sure it's not some low level decision it's somewhat of a big deal and those decisions can take time... but perhaps they can shed other assets pretty quickly if they have to..
"so much on deposit with commercial banks?"
I think rather this represents the amount Treasury has on deposit at the Fed rather than commercial banks... I believe the TGA is at the Fed...
"what was behind the sharp rise in 2016?"
They increased the rate of UST securities issuance well in excess of what they were withdrawing from the TGA so USD balances built up in the TGA.... "debt ceiling" deadline imminent March 15th? Who knows...
Certainly this was unprecedented in scale or otherwise if you look at the chart...
Go back and look at RESBAL in late 2008, it was 9B on Sep 10th and a month later it was like $300B...
so they stuffed 300B of reserves onto banks balance sheets in a month causing liquidations in the other bank assets and the whole system locked up and caused the whole gd GFC...
https://fred.stlouisfed.org/graph/fredgraph.png?g=dgje
Yeah its all a big "neoliberal conspiracy!" LOL!
A brand new Boeing 777 would still crash with a chimpanzee at the controls....
It's crazy.
It makes you wonder why the stock market is where it is.
These commercial banks must know what's happened and see the complete picture. Surely they would be selling into this rally big time.
It's the marks that are buying at the tops.
"There's a sucker born every minute," is still true.
Matt, this is probably a silly question, so apologies in advance! What you guys are discussing sounds fundamentally important, but I don't understand a word of it. Where can I find out in detail, but in an easy Dummies-style way, about this stuff? Or is it yet another one of those things (the economy rather than fancy models) that mainstream economics and finance have no interest in, and so searching it out in the mainstream is pointless. Is it a case of finding out understanding the mechanics of these accounts (how and when they're debited, credited, etc), or is it just a case of simply understanding that these accounts work and then applying a bit of MMT and MMT-friendly finance?
John yes the latter...
I can probably answer questions but I currently don't have the communication skills to write it up appropriately.... if you (or anybody) have questions about my opinions here just ask I'll do my best....
The wonky guys at the Fed are getting more details.
Matt, I don't even know where to start! MMT-friendly finance is pretty much only available at one place - here! A friend of mine has Magic Mike's "Understanding the Daily Treasury Statement". I'll have a gander at that and see if that allows me to even articulate the problems and then try and get back to you! By the time I get back to you, you won't even remember what the hell i'm talking about. Many thanks for the offer.
John, Mike's DTS course is worth its weight in gold. 5-hr video.
Also read Frank Newman's "Freedom from National Debt." 87 pages. But it took me four months to read it. Altho' in easy English, I took a day/page to comprehend it.
MRW, thanks. I read both of Frank Newman's books last year in hospital. They may be like Warren Mosler's books, in that there is far more information to ponder than is immediately obvious, and, as you yourself have noticed, Newman's style of delivery makes you read the same sentence again and again, leaving you wondering is it the prose or the subject matter that's the problem. Plus I was as high as a goddamn kite! It was difficult judging how MMT-friendly Newman is. I did note that as often as he seems a paid up member of the MMT community, there did seem to be significant departures.
Hi Matt,
Can you break down what you said here so that it is easier to understand.
"Go back and look at RESBAL in late 2008, it was 9B on Sep 10th and a month later it was like $300B...
so they stuffed 300B of reserves onto banks balance sheets in a month causing liquidations in the other bank assets and the whole system locked up and caused the whole gd GFC...
https://fred.stlouisfed.org/graph/fredgraph.png?g=dgje
"
1. Why does increasing bank reserves cause liquidations in other bank assets? How does that relate to 2008-9 subprime crisis?
2. Based on the graph above how come there wasn't a crisis in
2011 and 2013-2014 when there was significant increases in reserve balances.
Would really appreciate your insight. Thank you.
I don't understand the capital issue. Reserves forced on the banks don't affect their capital positions; they are zero risk-weighted and (in the UK at least) excluded from the simple leverage ratio calculation (total assets divided by equity, regardless of asset quality).
Capital Adequacy Ratio is a minimum (which an aside for this discussion, but mentioning it for completeness's sake) and doesn't include reserve assets.
If government spending created excess reserves in banks and that forced them to not lend because their assets went up while their capital remained the same, it can't be solved by banks selling other assets for cash i.e. more reserves.
The reserve position of a bank is completely irrelevant to a bank selling assets for reserves. Since reserves don't count towards assets while computing the capital adequacy, a bank selling non-reserve assets for reserves allows it to meet capital requirements without infusing additional capital. This is exactly what they did during QE, they offloaded their non-reserve assets to the Central Bank, got reserves for them reducing the assets a percentage of which they have to hold as capital and thus meeting the capital requirements.
Some more points of view from guys who have worked in banking in the UK.
The Fed has less than 20% of the outstanding treasuries on its balance sheet. In any case when banks are swimming in excess reserves, one bank falling short of the capital adequacy ratio can sell non-reserve assets to another whose capital position is good. If all the banks are short of capital, that has nothing to do with the government spending leading to increased reserves causing a capital shortfall because banks don't have to hold any capital against reserves, only other kind of assets that can go down in accounting value (reserves can never go down in accounting value). If all banks are under capitalised, the Fed may have to run another QE operation buying other kinds of assets than Treasuries. They bought MBSs last time, not sure why can't expand the category of assets they take on their balance sheet.
You may have discovered a correlation, but that is not causation. The debt limit has absolutely nothing to do with banks selling non-reserve assets for reserves.
And some more.....
Maybe I misunderstand, but it's almost as if we've gone from "reserves increase bank capacity to lend" to the absolute reverse!
Both are madness.
Absolutely. Reserves have absolutely no role in the banks' capacity to lend, not since currencies stopped being backed by metals. They neither increase nor decrease that capacity. Their capacity to lend is solely a function of their capital, the risk-weighted assets they already have on their balance sheet and the capital requirements set. If the capital requirements are relaxed, they can lend more, if they are tightened they cannot lend more and may even have to get rid of the risky assets.
I am not sure, but if there is an urgency to sell assets on the banks' part it may have to do with the BASEL 3 rules, some of which came into effect this year.
From Investopedia:
What is the 'Tier 1 Leverage Ratio'
The Tier 1 leverage ratio is the relationship between a banking organization's core capital and its total assets. The Tier 1 leverage ratio is calculated by dividing Tier 1 capital by a bank's average total consolidated assets and certain off-balance sheet exposures. Similarly to the Tier 1 capital ratio, the Tier 1 leverage ratio is used as a tool by central monetary authorities to ensure the capital adequacy of banks and to place constraints on the degree to which a financial company can leverage its capital base.
Then it further defines 'core capital' as:
"Core capital consists of equity capital and declared reserves."
Are 'declared reserves' the same thing as 'reserves'? Or does a bank have to segregate a certain amount of its reserves as 'declared' via bank internal policy for any regulatory period?
So apparently you have two different regulatory ratios at work here, you have the Capital Ratio and the Leverage Ratio...
I think what I am talking about is the Leverage Ratio...
Read more: Tier 1 Leverage Ratio http://www.investopedia.com/terms/t/tier-1-leverage-ratio.asp#ixzz4dTCsTKHY
Follow us: Investopedia on Facebook
And some more from the UK ex bankers.
The reaction to increased bank deposits is:
Decrease in wholesale funding and/or increase in assets (It buts debt i.e. US Treasuries). Usually a combination of both. For every $8 in deposits a bank requires $1 in US treasuries and its capital ration remains unchanged.
You make a classic error in believing that Central Bank rates go up, cost to consumer goes up. In the early part of the raising cycle the inverse is true.
Isn't the inter bank rate higher than the deposit rate? Hasn't the differential changed recently.
Why borrow from another bank and put it on reserve with the Fed when you lose money on the deal?
If I reduce my inter bank liabilities I reduce the need for capital. Not to mention that I increase my net interest margin and make more money.
Surely this is a natural function in a raising cycle and would be why the Fed is looking at balance sheet action.
"calculated by dividing Tier 1 capital by a bank's average total consolidated assets"
I believe the H.8 report accounts "reserves" as part of bank total assets under Line Item 34 'Cash Assets' here:
https://www.federalreserve.gov/releases/h8/current/
Line 34 has Note 21 which says:
"21. Includes vault cash, cash items in process of collection, balances due from depository institutions, and balances due from Federal Reserve Banks."
'balances due from Federal Reserve Banks' I view as aka 'Reserve Balances'
So if these "balances due from Federal Reserve Banks' are treated just as any part of bank consolidated assets and they are not automatically part of 'declared reserves' at a bank upon receipt, then to maintain a constant Leverage Ratio, imo it appears a bank would have to liquidate other assets if bank deposits from TGA were to spike ....
"and would be why the Fed is looking at balance sheet action."
Well yes they should very well be as they (to them) are probably freaking out and thinking they are soon going to be "out of money!" and bankrupt if this continues...
Well yes they should very well be as they (to them) are probably freaking out and thinking they are soon going to be "out of money!" and bankrupt if this continues...
Or they are just going to reduce QE. They are not paying banks to hold funds so all the interest they receive from the Government is paid back. The vast majority of those assets on the Fed's Balance Sheet are therefore not needed.
"They are not paying banks to hold funds"
What about the IOR?
2.4t of reserves now at 1% don't they now have a 24B annual liability?
"They are not paying banks to hold funds"
Banks are paid a deposit rate to hold funds at the Fed. The Fed pay the deposit rate from the interest they earn from the assets on their balance sheet.
The remainder of unpaid interest is returned to the Government so the Fed balance sheet doesn't continue to grow. If it is continually returning interest to Government the balance sheet is way to big.
The other thing is you have to assess is if the balance sheet reduction would necessarily result in increased net UST securities issuance which btw is currently impossible as they are at their "debt ceiling!"...
Why would issuance change?
Just the asset holder.
Matt: "imo it appears a bank would have to liquidate other assets if bank deposits from TGA were to spike ...." Hunh? Bank deposits from TGA spike every time there's a Congressional appropriation, and the Fed dispenses them to vendor accounts. Two weeks to a month later (in the main), the US Treasury issues treasury securities in the amount of the appropriation (in the main), and those are sold at auction to the public. Money supply restored to balance.
Matt, why dont you call your district Federal Reserve bank and ask them what 'balances due from Federal Reserve Banks means? I dont have time today.
Let me set you on another path.
Do you think early rate rises lead to credit tightening.
Look at past cycles, I will assure they do not. In fact what you tend to see is increased bank margins and looser credit.
That's an advantage of being old
The most important economic fact you should ever know; "People do not make decisions based on statistics"
A cycle last s as long as it does, past cycles are not a guide to longevity.
We are still in the early stages of phase 2, way too early to call time for me. I am still awaiting animal spirits in the third phase and capitulation in the forth
""imo it appears a bank would have to liquidate other assets if bank deposits from TGA were to spike ...." Hunh? "
USD balances in the TGA are not counted as reserves.... so when Treasury ran the TGA up to over 400B in umprecedented fashion last year that would have the effect of reducing reserve balances on bank balance sheets.... you could see this on the H.8.... , then disgorged them in about a month, it had a short term effect of increasing reserve assets on bank balance sheets by 400B in a very short period of time as the Treasury didnt do a 'reserve drain' by issuing bonds before they withdrew from the TGA... they in effect did the 'reserve drain' well in advance by running up the TGA by 400B+ over the previous year...
Withdrawals from the TGA add system reserves if they are not first offset by Treasury issuance $4$ in advance...
In normal circumstances, the Treasury does a reserve drain (by issuing bonds) BEFORE they do the reserve add via TGA withdrawal...
Also we have reserve balances very high (see Tim's statement " If it is continually returning interest to Government the balance sheet is way to big.") with the QE which is also unprecedented... so 400B of that ended up in the TGA last year by UST issuing bonds well in advance of necessary withdrawals....
If the sharp increase in reserve assets caused by this TGA policy effected the regulatory "Leverage Ratio" then that would perhaps be a problem for banks if it was not well coordinated between Treasury and the member banks beforehand...
I have to look further into the difference between Capital Ratio and Leverage Ratio from the standpoint of 'reserves' and 'declared reserves'....
Here:
http://www.thirdway.org/memo/capital-requirements-and-bank-balance-sheets-reviewing-the-basics
"Leverage Ratio
A leverage ratio is a specific type of capital requirement—it does not take risk weights into account. Or put another way, a leverage ratio is a capital requirement that treats all assets as if they had 100% risk weights. If the leverage ratio for a bank is 5%, then it would be required to have $1 of equity for every $20 in assets regardless of the riskiness of those assets."
Here:
http://www.investopedia.com/ask/answers/111414/how-do-leverage-ratios-help-regulate-how-much-banks-lend-or-invest.asp
"three separate regulatory bodies, the FDIC, the Federal Reserve and the Comptroller of the Currency, review and restrict the leverage ratios for American banks. "
Capital Ratio looks like a qualitative measure while Leverage Ratio is a purely quantitative measure...
Are reserves specifically considered assets of the commercial banks, such that they would be considered part of their (commercial banks) cash assets? Anyone know?
Matt,
”In normal circumstances, the Treasury does a reserve drain (by issuing bonds) BEFORE they do the reserve add via TGA withdrawal...
Not according to Frank Newman, former Deputy SecTreas in his latest book. He said they are done AFTER for a bunch of reasons; namely, that sometimes they dont need to issue bonds in the full amount of the congressional appropriation (TGA withdrawal amt to vendors, etc).
And Warren is fond of repeating the Federal Reserve/US Treasury mantra: “Reserve add before reserve drain.”
Newman say they are done "typically" two to four weeks after the TGA withdrawal.
Matt,
When the Treasury distributes funds, the nation’s deposits are initially increased. Where can the bank money go? Let’s look at an example, excluding the portion covered by taxes. Typically, before the Treasury issues $ 20 billion of securities, the government has distributed $ 20 billion to the public from its account at the Fed: redeeming maturing Treasuries, paying companies that provide goods and services for the government, for payments to individuals, etc. Many investors simply “roll over” their Treasury securities, replacing maturing ones with newly issued ones, and taking just the interest. For example, perhaps $ 10 billion of the $ 20 billion issue might be in that category. The Treasury pays out the other $ 10 billion to the private sector. At that point, a set of participants in the U.S. financial system will have the extra $ 10 billion in their bank accounts and will look to place those funds. The money supply has been increased by $ 10 billion, and the new dollars move around within the overall US financial system. All the Treasuries previously available are already owned by investors, and prior auctions had demand that exceeded the amount offered. As the new Treasuries are auctioned, the demand is filled by exactly the $ 10 billion offered, and the money supply returns to its prior level. In the whole of the U.S. financial system, the only place to put the money is into the new Treasuries that are being auctioned— or otherwise just leave the funds in banks.
Newman, Frank N. (2013-04-22). Freedom from National Debt (pp. 18-19). Two Harbors Press. Kindle Edition.
MRW we need to keep looking into this #1... I'm going to still assert I'm correct here for now....
I'm talking about BANK reserve assets reported on the H.8 while Newman is talking about "money" which is a figure of speech.
My language is more precise and scientific than Newman's which is required in order to do an acceptable analysis....
What do you think that when the FRS/OCC/FDIC do their regulatory work they go into a bank and say "Hey! How much money do you guys have?!"
Please.
USD balances in the TGA do not even count as bank reserve assets so what is Newman even talking about?
USD in GSE accounts do not exist as bank reserve assets so again what is he talking about?
Right now, Feds H.4.1 has total factors effecting reserve balances at Iirc like 4.3T, yet RESBAL is at about 2.3T so where are the other 2T of "reserves"?
They are either in the TGA or with the GSEs... maybe govt employee retirement account?.... NOT with the banks as banks only have 2.3T....
Ok , Newman: "When the Treasury distributes funds, the nation’s deposits are initially increased."
Ok, what happens here, AT THE BANK?
Some social security person receives a Direct Deposit or wtf...
At the bank:
On the right they increase "customer checking account payable", AND on the left they increase "reserve assets"...
IF, and I repeat IF, the bank has fixed capital for the period in question, then the banks ratio of total assets to capital has just increased over said period...
I.e. Assets have increased while capital has not.
Go look at line 34 here today under QE it's almost 2.5T:
https://www.federalreserve.gov/releases/h8/current/
Then go back and look at line 14 here in early 2008 pre GFC and pre QE1,2,3 it's like 300B:
https://www.federalreserve.gov/releases/h8/20080111/
Else equal, to maintain a constant Leverage Ratio at 20:1, banks would need to have added an additional 100B to their capital accounts to comply... or reduced other non-reserve assets by same amount...
So then let's look at the effect of ending the QE, if the effect of that would be to reduce bank reserve assets from current 2.5T to perhaps 500B, then that would allow banks to put an additional 2T of assets on their books to maintain a constant Leverage Ratio with no additional capital needed...
Plus 100B+ of interest income going back to savers instead of back to Treasury....
#MOONSHOT
Meanwhile the mob will think "they are taking away the punch bowl!" and will all get short....
USD balances in the TGA do not even count as bank reserve assets so what is Newman even talking about?
USD in GSE accounts do not exist as bank reserve assets so again what is he talking about?
Right now, Feds H.4.1 has total factors effecting reserve balances at Iirc like 4.3T, yet RESBAL is at about 2.3T so where are the other 2T of "reserves"?
They are either in the TGA or with the GSEs... maybe govt employee retirement account?.... NOT with the banks as banks only have 2.3T....
The mistake is counting entires on the side of government as money when they don't appear as entries in any of the measures of money in the US e.g, M0 (base money, HPM), M1 and M2.
The only government liabilities that appear in the measures of money in the US are Fed liabilities as either settlement balances in nongovernment accounts (this includes Treasury deposits in the TT&L accounts) and cash (vault cash and currency in circulation).
(Actually, the Treasury does "have some money" in the form of the TT&L deposits that count toward bank reserves.)
the Fed doesn't "have money" since it issue the currency.
So ending QE gives more interest income to savers and allows additional bank lending. Is that what Tim Browning was getting at when he says early interest rate rises DO NOT lead to credit tightening but instead to looser credit?
Blogger Matt Franko said...
So then let's look at the effect of ending the QE, if the effect of that would be to reduce bank reserve assets from current 2.5T to perhaps 500B, then that would allow banks to put an additional 2T of assets on their books to maintain a constant Leverage Ratio with no additional capital needed...
Plus 100B+ of interest income going back to savers instead of back to Treasury....
#MOONSHOT
Blogger Tim Browning said...
Let me set you on another path.
Do you think early rate rises lead to credit tightening.
Look at past cycles, I will assure they do not. In fact what you tend to see is increased bank margins and looser credit.
That's an advantage of being old
Matt, get a grip.
Former Dept. Secretary of the U.S. Treasury Says Critics of MMT are "Reaching" - New Economic Perspectives
http://neweconomicperspectives.org/2013/10/former-dept-secretary-u-s-treasury-says-critics-mmt-reaching.html
Not sure if this will be read, but I'd really like to get in touch with Matt to ask a clarifying question about the reserve effect on bank leverage ratio.
My question is when Fed increases reserve balances it seems that bank assets rise (reserves go up), but what happens to liabilities and capital? Do they stay the same or does capital/equity also go up since assets (reserves increased) and Assets = Liabilities + equity. Apologies if this is a novice question but just trying to make sure I'm on the same page with some of the excellent info you have posted. Thanks!
Post a Comment