wabuffo

The Fed can only buy "safe assets" because it is mandated by Congress to never take losses on its lending programs. The Federal Reserve has total assets of $6.9t against "equity capital" of $48b - so it is "levered" 144-to-1. Usually the Fed only buys US Treasury debt and the MBS of Federal agencies (with implicit Fed govt credit guarantees) and thus credit risk on these assets is effectively zero. While the US Treasury has stepped in to provide an additional "cushion" for some of the Fed's new asset programs - it is still only $114b of additional buffer and the Fed is still "levered" 43-to-1. So even 2% losses would make the Fed "insolvent". This would create an unacceptable technical risk that the Fed could not cover 100% of the liabilities it has created (mostly bank reserves and currency in circulation). It could result in a payment failing to clear in the inter-bank settlement markets that the Fed could not cover with an overdraft or discount window loan. Rule #1 for the Fed is "never allow a payment failure to occur in the inter-bank clearing market (Fedwire)". Of course, the Fed could get new equity from the US Treasury via Congress (which can print unlimited dollars) but it would cause a financial crisis as Congress deliberated over it. The "dodgy" stuff must be supported by acts of Congress via the US Treasury (eg. airline loans), because it exposes the US taxpayer to potential losses. wabuffo

Technically speaking though the money created by Fed is not borrowed money. When the Fed buys Treasury debt in the open market, it swaps new bank reserves in exchange for them. These bank reserves are assets of the commercial banks BUT liabilities of the Federal Reserve. The Fed even pays interest on those reserve liabilities. When looking at the Federal government from the point of view of the private sector, one has to consolidate the Federal Reserve and US Treasury. So whether its Treasury bonds, bank reserves on deposit at the Fed, or even currency in circulation - these are all liabilities of the Federal government. wabuffo

My own view is that Mr. Market is starting to clip BRK because of a likely increase in the US Federal tax rate from 21 to 28%. This affects BRK two ways: 1) It clips book value due to the increase in deferred cap gains taxes (reducing book value) 2) It reduces the after-tax earnings of both the operating subs (like BNSF) as well as the look-through earnings of some of its stock holdings (eg banks). This was a big factor on the way up in 2017-2018 as the tax rate fell from 35% to 21%. While at this point, its not a certainty that the Federal tax rate will increase, its becoming a higher probability (and thus a headwind to BRK). wabuffo

Also did this with PG - got Duracell business in tax-free return for BRK’s PG common shares. Buffett has done this a few times. In the WaPo tax-free exchange, BRK also received a Miami TV station. wabuffo

Hi benchmark - I just saw your questions. Perhaps we should set up a new topic on this one. I'll do that now. https://www.cornerofberkshireandfairfax.ca/forum/investment-ideas/pdli-(pdl-biopharma-inc)/#new wabuffo

What's the PDLI liquidation play? simple google didn't yield anything https://www.sec.gov/Archives/edgar/data/882104/000088210420000057/pdli-2020def14adoc.htm#se100117011f048748132db3631831519 You got some reading to do. Cop a squint at the sections entitled: "general information about the dissolution and plan of dissolution." "Proposal #3" wabuffo

PDLI’s Management mentioned something about benefiting from Cares tax changes, which can be substantially, since it is replied retroactively. Do you have any idea what sums we are talking about. Spek - the only comments I've seen PDLI mgmt make are from the recent 10-Q in May: and from the conference call for Q1: I can't tell from the 10-Q, if the tax benefit they quote is equal to the expected cash refund they expect to receive. It's also unclear if they have factored any tax refunds in their original liquidation estimate of $3-$6 per share (including the common shares of EVFM that have since been distributed to PDLI shareholders). PDLI has 116.3m common shares o/s, IIRC. wabuffo

bought more PDLI (a liquidation special situation) and some BRK.B at the lows at the end of the day (couldn't help myself - even though I already own some). wabuffo

One thing to keep in mind is that if the Democrats sweep the executive and legislative branches of US government in the fall (as seems likely at this time, though that may change), Biden has already indicated that he is raising the federal corporate tax rate to 28% - which he will be able to do if the Democrats have majorities in the House and the Senate. BRK was a huge beneficiary of the corporate tax rate cut in 2017 and will consequently get hurt if/when it is raised. It will get hurt two ways - large US subsidiaries like BNSF would see lower operating earnings and cash flows. But the after-tax values of the equity portfolio would fall as well. Not as big an effect as it was going from 35% down to 21% - but our partner, Uncle Sam is raising his stake from 21% to 28% without paying us for the larger ownership stake in all future pre-tax earnings. wabuffo

PDLI @ $2.87 wabuffo

Starter in MSGS Interesting - rationale? wabuffo

Dudley writes in a clear manner about the Fed's monetary operations. Here's another article he wrote in January after people were claiming the Fed's resuming repo operations and expanding its balance sheet was leading to a stock bubble at the end of 2019. https://www.bloomberg.com/opinion/articles/2020-01-29/fed-s-repo-response-isn-t-fueling-the-stock-market If one always remembers the quote above from Dudley every time some TV talking head blames the Fed's growing balance for one problem or another, one will experience less confusion about monetary policy. wabuffo

As I wrote in an earlier post, the effect on stocks after a strong force of deflation is stopped is always like this. In 2000-2002 there was a strong multi-year deflation/economic contraction, but starting at the end of Oct 2002 til end of 2003 reflation starts, and the median, non-S&P 500 stock was up +114%. In late 2008 we saw another great deflation due to the mortgage crisis, but starting in March 2009 til the end of 2009, the actions of the Fed and US Treasury stop the debt deflation/bank crisis, and the median, non-S&P 500 stock was up +117%. This year, the Fed and US Treasury acted much more quickly and basically stopped the economic contraction/deflation that started in early March by March 23rd. So its not a surprise that stocks are rising quickly. And as has been shown before, the lower-quality, more indebted companies recover faster and higher because they fell the most. Twas ever thus. We are seeing the same effect now and probably for the rest of the year. The cast of characters may change - but we are seeing the same underlying effect except the ducks that think they are great swimmers because of the huge rise in water levels in the lake are on social media these days and crowing about their swimming skills. Just my 2-cents. FWIW. wabuffo

I find it fascinating that every time we have a big crisis, people start worrying about inflation and deflation at the same time.... I am a little surprised that people are a lot more intellectually interested in papers and theory.... It is weird that when people are handed very valid solutions to their problems, Laacz, they kind of ignore it and instead try to go on some sort of philosophical search.... I have noticed this kind of indifference both on this message board and elsewhere. It's puzzling to me.... What I am saying is that if you want inflation protection, if that's precisely what you want, then it is one of the best instruments to protect you. From 9/1/2008 - start of GFC thru next 4 years (using the deficit-to-gdp as the key indicator) GLD vs LAACZ. From 9/1/2008 - to today (though in fairness, I switched out of GLD in 2014 after deficit-to-gdp ratio fell under 4%). Look - I always prefer to own businesses than a shiny metal so not a fair comparison. And people should only invest in things with which they are comfortable. Inflation protection to me means protection from currency debasement. So right now, for the cash portion of my portfolio, I prefer GLD to cash. Reasonable people can disagree on this, of course, but still respect the opinion of others because there are many paths to investment success/capital preservation. wabuffo

Recently, numbers have indeed been huge as the Fed has been absorbing (indirectly...) what the Treasury has been issuing. The % is now above 20%. It's not really above 20%. One must net out the intra-governmental asset/liability accounts, IMHO. It's like intra-company receivable/payable accounts between a holdco/opco structure. (ie, the Fed is a 'subsidiary' of the US Treasury). More importantly, both sit on the opposite side of the private sector. As at June 10th, 2020, ....The Treasury owes the Fed $4.15t (because the Fed holds its debt - an asset of the Fed), ....but the Fed also owes the US Treasury $1.5t (because the Treasury has a large deposit in its account at the Fed - this is a liability of the Fed). In the good ol' days, one could ignore the US Treasury's deposit account which typically maxed out at $5B. But today it sits at $1.5T (an expansion of 300X!). Here's an updated chart (vs the FRED chart that I posted that goes til Q3, 2019). Still way below the historical average despite the Fed owning over $4t of Treasury debt. Again - my point isn't to dismiss what's going with the US Treasury's spending and debt issuance. It's only to place the Fed properly in its context as a disinterested spectator rather than active participant. In fact, the very large balance in the Treasury general account is causing havoc with the Fed's balance sheet and future bank reserve forecasts. wabuffo

It seems to me that if the Fed buys a bunch of bonds and the Treasury puts that money in the economy, Congress expressly forbids the US Treasury from selling its debt directly to the Federal Reserve - it must always sell its debt to private markets. The way it works is: 1) The US Treasury sells bills/bonds to the private sector. 2) The Federal Reserve as part of its monetary operations swaps bank reserves for Treasuries. The argument you hear is that the Fed is increasingly "monetizing" the Treasury's debt by buying ever larger amounts. If debt monetization by the Fed was truly happening, I would expect to see the Fed's share percentage increasing -- but it isn't. I think people just like to throw around big numbers without context. Now - one could argue about the size of current US Treasury spending. But that is a separate issue and doesn't involve the Federal Reserve and its monetary operations/asset-based lending. wabuffo

Do you see inflation as a result of current Fed actions? Or future Fed/Treasury actions? Definitely tipped to inflation and solely because of the US Treasury because of its net spending -- which creates new private sector deposits. The Fed controls the price of money but not the supply since it can only do asset swaps (Treasuries in exchange for bank reserves, for example). Here's a long run chart of gold vs deficit-to-gdp ratios. You can see that the price of gold is a measuring stick that is sensitive to currency debasement when deficits become large vs gdp. wabuffo

I should clarify that I actually agree that there is a risk of currency debasement - but not because of the Fed. It will be a function of US Treasury stimulus spending. wabuffo

It seems to me that the FED's 4bn per day QE is way bigger than the prior 3 QEs. Narrator: "Its not" The Fed's share of total US Treasury Debt is not significantly higher than the past going all the way back to 1970. There is no "monetization" going on. wabuffo

Have you read much from Scott Sumner or George Selgin? Scott Sumner - no George Selgin - yes, I have his book Floored! His main point seems to be that the Fed should abandon the floor system with all these excess reserves and go back to old corridor system with minimal reserves. I basically agree with him on that. I really like his critiques of current Fed policy and monetary management. wabuffo

QE can be deflationary if a central bank insists on paying interest on reserves. I know it's in fashion to say that private banks create most of the money, but this isn't true in a practical sense. The nominal anchor is still the monetary base and that is controlled by the Fed Under normal open market operations (OMOs), the Fed injects 0% yielding reserves in place of say 2% yielding treasuries. This makes a commercial banks balance sheet a bit less risky. Since reserves are "stuck" in the commercial banking system, the only thing these commercial banks can do (as a whole) is make more loans, etc to bring their balance sheet metrics back in line However, when the Fed insists on paying interest on these reserves, now the open market operations are swapping say 2% treasuries for 2% yielding reserves. There is no incentive for banks to "rerisk" their balance sheet and hence the banks have very little reason to make more loans. Of course this will look like "hoarding" of money, etc and result is lower economic (nominal GDP) growth. Jim - interesting take. I think we both agree that reserves are basically inert. Where we disagree is that somehow reserves lead to more lending, if only the Fed were more accomodative (ie, lowering interest rates). I just don't think there's any relationship between the two. More importantly the data backs it up. Bank credit is a function of collateral and regulatory capital requirements. Here's a long run chart of the ratio of bank credit to reserves since early 80s. The chart merely reflects that bank credit grew even as reserves fell during much of the period from the 80s to 2008. Then the Federal Reserve changed its interest-rate targeting and reserve policy and the ratios since 2008 reflect that. One (bank credit) has very little to do with the other (reserves). Its apples-and-rutabagas - no relationship at all. Indeed, some central banks operate just fine with zero reserves (Bank of Canada). Even the traditional reserves-are-10%-of-deposits doesn't show any relationship. It almost looks like reserve balances are independent and disconnected from deposits or bank credit. We will probably have to agree to disagree on this, but in my view, FWIW - reserve balances at the Federal Reserve serve only two purposes: 1) Payment settlement. 2) Meeting reserve requirements (per bank safety and soundness regs). Reserve balances are not used for: 3) Money creation 4) Bank Credit extension. Most of the theory about money multipliers and base money just haven't shown any traction in reality and practice. That's why so much of the expectations of what the Fed can do to stimulate the economy (or slow it down) are bound to fail. It really is the height of arrogance by the Fed officials that they act as if they do. If one wants to stimulate the economy, the only lever is fiscal policy through the US Treasury. Small changes to Fed interest rates have only minor effects. The rest is risk-appetite/risk aversion by the private sector. So for me, the order of influence on the economy is: 1) Risk Appetite 2) Fiscal Policy . . . 3) Fed Policy But I do enjoy reading your stuff (most of the mechanics, I think we agree on). wabuffo

It seems sort of surprising that Dimon doesn’t exactly know the red line in terms of reserves. One would think that the rule for clearing payments and the reserves are crystal clear, Spek - from what I can tell, I think the rules are pretty clear, but what creates uncertainty is volatility in daily payments. I think you have some experience working in manufacturing companies - so you can relate to how manufacturers try to set their inventories of finished goods to meet sales orders. Their supply chain people get demand forecasts from sales, orders on hand plus production schedules and calculate required inventory levels/targets. But of course, demand forecasts can be wrong and production can sometimes come up short vs the schedule. So inventories must have an extra level of safety stock to ensure high service levels to customers are always maintained and no sale is missed. I think of reserve levels the same way. Dimon probably knows that, on average, he needs $60B of reserves to meet daily expected fluctuations, but he feels his bank needs an extra $60B of safety stock. That leaves him with zero available to lend out. When JPM was holding $200-$300B, the $60B + $60B of safety stock left him ample reserves to lend. I think the real wild card is the actions of the US Treasury when it comes in with either big spending or big withdrawals (taxes). Here again, pre-GFC, the Fed and the US Treasury used to have a better system and would coordinate their activities. For one, the Fed was in the fed funds market in order to maintain its interest rate target and so it had to get information from the US Treasury every week on what it was planning to do with spending, taxes, debt issuance/redemption. For another, the US Treasury used to keep account balances at not just the Fed - but also in the commercial banking system. These were called Treasury Tax & Loan Acounts (TTLs). The Treasury's account at the Fed would hold $4-$5B and the TT&L accounts would also hold maybe $20-$50B in balances. Thus, if there was a big tax withdrawal day, the cash would stay at the commercial banks and just move deposits from households/corporations to the Treasury's commercial bank accounts - no initial impact to bank reserves at the Fed. The Fed and the US Treasury would then co-ordinate when the Treasury would withdraw the money from the commercial banking sector so as not to create big ripple effects. Today - the US Treasury holds no commercial banking deposit accounts - the TT&L accounts were phased out after the GFC. Instead, the US Treasury is holding all of its cash in its account at the Fed and every move swings reserves back and forth from bank reserves. Frankly, the Fed and US Treasury had a pretty good system to operate monetary policy before the GFC, now they run a clunky and bloated system that is a bureaucratic nightmare for the banks. wabuffo

Seems like treating a toddler prone to meltdowns with candy and Ice cream each time when a tantrum is due. Spek - I may agree with your general comment overall - but as I indicated, the Fed is a side-show. Its like the Wizard of Oz - its voice booms over the economy, but it is a meek little man behind the curtain who can't even do his basic job sometimes (as in the Sept repo mess). Its unfortunate that the Fed runs its affairs the way it does - with excess reserves and an interest rate on excess reserves. It is probably the only central bank that does this -- and it is totally unnecessary. So it draws oohs and ahs when its balance sheet expands. As I've said before, I think the Fed could run with zero reserves, no problem. It would then retreat into the background, and economic conspiracy theorists would have to find something else to point to... The real heavy hitter to pay attention to is the US Treasury (and Congress). But they like to avoid the spotlight and push the Fed to take the slings and arrows. wabuffo

The U.S. government literally ran out of cash dollar lenders for their Treasury securities; both domestic and foreign. In reality, the root cause was that the U.S. government ran out of lenders. Foreigners, pensions, insurance companies, retail investors, and finally large banks and hedge funds, simply weren't buying enough Treasuries at that point compared to how many Treasuries the government was issuing, at over $1 trillion annualized. They had no excess dollars from which to lend to the U.S. government at those rates. Vinod - her explanation of the Sept repo crisis is just goofy....and wrong. I don't think she totally understands how any of this works. In order to understand the Sept repo "crisis' - you have to understand that this was the Federal Reserve causing a self-inflicted wound to itself by not keeping its eye on the ball when it comes to its most important function - running the US payment clearing system between banks. US PAYMENT CLEARING SYSTEM: The Federal Reserve manages the US payment clearing and settlement process between federally-chartered banks and thrifts (also includes a few other financial firms but also the US Treasury, of course). Banks have checking accounts at the Fed and those balances (reserves) are used to make sure all inter-bank payments clear and no bank overdraws its account during daylight hours. Despite all the focus on FOMC meetings, etc - this is the single most important reason why the Federal Reserve system exists. So what is the sum total of financial transactions processed in a day, a week, a year? The Fed provides this data for the two major payment clearing systems (Fedwire - which is run by the Fed) and (CHiPS - which is run by ~50 big US banks and the US subs of foreign banks but does the final settlement with the Fed). Here are their transaction volume summaries: FEDWIRE: https://www.frbservices.org/resources/financial-services/wires/volume-value-stats/monthly-stats.html CHIPS: https://www.theclearinghouse.org/-/media/new/tch/documents/payment-systems/chips-volume-and-value.pdf In 2019, Fedwire processed $696t in US payments while CHIPS processed $417t in US payments. That's a total of $1.13 quadrillion in annual financial payments cleared (vs $21t US GDP) There are also a small amount of transactions settled in cash but we can ignore those. That's over $3.1t per day. But I do like to say "$1 Quadrillion Dollars!" ROLE OF BANK RESERVES IN PAYMENT CLEARING: All inter-bank payment clearing happens at the Fed. To participate in this activity, banks have "checking accounts" at the Federal Reserve. The key question is what is the right amount to keep at the Fed given the very high transaction amounts. What governs the amounts banks have to have there is the Federal Reserve itself (including its regulations and practices). Let's use our own experience with a checking account and how we would manage our finances - one with overdraft protection and one without. Since we are all investors, we would prefer to keep the minimum possible in this checking account because there is an opportunity cost to keep large balances vs holding stocks, etc. So on a particular day, we have a large bill payment to make but we are also expecting our bi-weekly pay via direct deposit to come in that day as well. What we don't know is the order in which these transactions will hit our account (plus there could be a surprise outflow we forgot about). If we keep the balance low and the bill payment goes out before the pay comes in, we'll go into a daylight overdraft and get hit with large overdraft fees by the bank. But if we have overdraft protection from the bank, we don't care what the order of transactions is, because the bank will cover us for any short-term negative balances. Banks manage their reserves with the same thought process - do they lend out reserves to other banks for some income or do they play it safe. This is where the Fed, its interest rate management policy and its regulatory rules come in. Pre-GFC, the Federal Reserve would allow banks to run negative reserve balances intra-day (which the Fed would cover) so long as they ended up at the end of each day with a positive reserve balance. If some bank looked like it was going to be short, it would try to borrow reserves from banks that would be over. This might cause interest rates for Fed funds (reserve balances) to increase. If this was threatening to make the Fed funds rate to rise above the target of the Fed, the Fed would come in with a repo operation - supplying reserves for Treasuries to get the interest rate back to target. This Fed interest rate targetting is called a corridor system (ie, the interest rate on fed funds stays within a band target set by the Fed and the Fed intervenes when it threatens to break above or below the band - thus the "corridor") But back then, the total excess reserves at the Fed were less than $10B - to clear daily payments of $2-3t per day. Just like us with overdraft protection at the bank, the banks had daily overdraft protection at the Fed and ran with extremely low reserve balances. After the GFC, the Fed changed the rules for banks. Daylight overdrafts were no longer permitted (the Fed would cover in an emergency because the Fed must ensure no payment ever fails to clear, but basically the bank CEO would lose his/her job), In addition, there were other rules imposed - there were "living will" requirements that forced banks to hold reserves to cover a certain number of days of payment clearing, etc.. In addition, the Fed expanded its balance sheet to force reserves to expand in size. (remember its the Fed, and not the banks, that control the size of total bank reserves). Now the Fed changed its interest rate management strategy. It no longer needed a corridor, since there were more reserves than banks needed - it went to a "floor" system and paid an interest rate on excess reserves. Fed repo operations basically ceased. The problem for the Fed is no one really knew where the 'floor' was anymore. I mean when you have trillions in excess reserves - how can any payments not be cleared? JPM MORGAN CHASE & RESERVES/REPO: JPMorgan is the biggest and most important bank in the US. It also has the largest amount of bank reserves since the GFC and thus is, at the margin, the lender of last resort to other banks who need reserves. Here is a chart that shows quarterly bank reserves for JPM since the GFC (these are from the FFIEC's quarterly bank Call Reports). So a few things to note about this chart. Prior to the GFC, JPM managed its payment clearing at the Fed with only $2-4B in reserves. Payment clearing volumes have not changed much over that time - total payments grow in line with GDP growth. Then when the Fed started doing all of its QE, JPM's reserves grew to a peak of $450B (over 100x pre-GFC) in March, 2015. As the Fed began to shrink its balance sheet in 2016-2019, total bank reserves began to shrink (and JPMs reserves shrink too). But where is the redline? Especially under the new regulations? At what point do individual banks like JPM think - I don't feel that I have any excess reserves to lend because of the new rules and regulations? Well - it turns out that point was reached in mid-Sept 2019! All it took was three days (Sept 13, 16, 17th, 2019) of large tax payments by the US private sector to the US Treasury (in excess of Treasury daily spending) that moved a net $125B from bank reserves to the US Treasury's general account at the Fed (data from the US Treasury Daily Statements for that period). On Sept. 11, 2019 - total bank reserves per the Fed H.4.1 report stood at $1.458T. Normally (125/1458 =) a 8.6% short-term reduction of bank reserves shouldn't push the payment system from Defcon 5 to Defcon 1 in the space of a few days. But unbeknownst to the Federal Reserve, the regulations and rules imposed on the banks limited their ability to deploy their excess reserves to help each other square up at the end of each of those days. JPM's reserves stood at ~$119b at June 30 and Sep 30, 2019 (which were their lowest levels since the GFC). The signs were there, though, at previous quarter-ends there were very brief spikes in fed funds rates but nowhere near what happened on Monday and Tuesday Sept 16 and 17. Here's JPM's Jamie Dimon on his Sept 30 earnings conference call: So there you have it. This was a bureaucratic and regulatory FUBAR that forced the Fed to resume its repo operations (and thus go back to a 'corridor' system from a 'floor' system in terms of controlling the fed funds rate). Nothing more or less than that. It wasn't because the US Treasury was issuing too much debt -- quite the contrary, the cause was actually the opposite - i.e, the US Treasury running a small 3-day surplus. I'm not a conspiracy theorist - but as the biggest bank, did JPM force the Fed's hand here? Was Jamie Dimon chafing at holding too much cash at the Fed and wanted a little loosening of the regulations? I would never want to accuse Jamie Dimon of anything here, but nor would I want to play poker with him. But if I had to bet between the bureaucrats at the Fed vs Jamie Dimon, I know who I would bet on. Once again, sorry for the long-winded response but my wife always jokes that she doesn't want to ask me the time, because I go into explaining how the watch is made. wabuffo

Suppose trade deficit is zero and the govt runs a fiscal deficit equal to 10% of GDP every year. As soon the treasury issues debt, the fed immediately buys all the debt by printing money. So in this situation by using wabuffo's analysis, private domestic savings =10% of GDP every year. So to increase domestic private savings, govt can simply run massive deficits and the fed can just monetize the govt debt. Naturally something is wrong with this scenario. Munger_Disciple, nothing is wrong with this scenario. The Fed in this hypothetical scenario is not monetizing govt debt. It is removing excess reserves that would pile up in the banking sector and drive interest rates to zero. The US Treasury does this too when it issues Treasury debt. You can get this if you do a debit/credit accounting of these transactions. Let's just rename the fiscal deficit as $1t instead of 10% of GDP and let's make it happen in one single transaction: 1) The US Treasury buys $1t of goods and labor via its Treasury general account at the Federal Reserve. The funds flow through a)the Fed, then b)the banks, to c)the private sector. Federal Reserve: US Treasury General Acct liability: -$1t Private Sector Bank Reserves liability: +$1t Commercial Banks: Private Sector Bank Reserves asset: +$1t Private Sector Bank Deposits liability: +$1t Private Sector: Bank Deposit assets: +$1t Private Sector savings equity: +$1t You can see how this transaction creates new bank deposits/financial assets out of thin air (think about the recent mailings/direct deposits of checks to every US taxpayer as part of the CARES Act). But the effect of no US Treasury debt issuance is that bank reserves at the Fed = the amount of US Treasury debt issuance that should've been issued but wasn't. These reserves are stuck on bank balance sheets since they are inert and trapped in a deposit at the Federal Reserve because they can only move between accounts at the Fed (and US Treasury). This drives all interest rates to zero because there are more reserves than are required to meet regulatory requirements and payment clearing. So there are two options to remove the excess reserves that have piled up in the banking system: a) the Federal Reserve can remove the excess reserves via a reverse-repo operation (trading US T-bills/bonds on its b/s to banks for their excess reserves), and/or b) the US Treasury can issue Treasury debt to remove the bank reserves and give the private sector a savings vehicle beyond its deposits in checking accounts. Note that these activities (a and b) are bank reserve maintenance functions. They both set govt risk-free interest rates - the Fed for short-term rates and the US Treasury for the rest of the yield curve. So let's look at the accounting/funds flow of both of these options. 2) The Federal Reserve (in order to keep its interest rate at target) performs a $1t reverse-repo operation with the banks (assume that in previous years the Treasury issued debt so that there is more than enough Treasury debt supply on the Fed's balance sheet from previous years' repo operations): Federal Reserve repo: US Treasury Debt asset: -$1t Private Sector Bank Reserves liability: -$1t Commercial Banks: Private Sector Bank Reserves asset: -$1t Private Sector Bank Tsy Bonds asset: +$1t 3) The other option is the more normal one - the US Treasury issues debt and this activity also removes the excess bank reserves via the private sector exchanging its cash for a new type of savings vehicle with higher interest income (say, a 30-year bond). Federal Reserve: US Treasury General Acct liability: +$1t Private Sector Bank Reserves liability: -$1t Commercial Banks: Private Sector Bank Reserves asset: -$1t Private Sector Bank Deposits liability: -$1t Private Sector: US Treasury bills/bonds asset: +$1t Bank Deposit asset: -$1t These transactions also demonstrate that operation 1) is the only one that increases private sector financial assets. Operations 2) and 3) only swap assets and leave everyone where they were before with no greater savings (just the form of the savings changes). The bottom line is I think we all are learning that the Fed Govt via US Treasury deficit spending has more fiscal capacity than anyone truly realized. This is also partly because the foreign sector is also demanding USD assets. There is always a risk of inflation if the US Treasury tries to commandeer too much of the private sector goods and labor. We might be testing that limit in 2020. wabuffo