$27,000,000,000,000 US debt

[wabuffo]

What is your best guess then to how this plays out in the US and Japan which have a lot of debt but also have their own currency?  Seems like could be a big cost at some point.

 

I'm probably going to disappoint you but I see a lot of one-timers here.  For one, tax revenues during the pandemic have collapsed.  The biggest source of Federal Tax revenues is employment income.  It brings in around $3t per year - give or take.  Here is a graph of federal employment withholding taxes on a rolling 2-week basis using the Daily US Treasury statements (here - I'm trying to simulate a rolling 2-week national payroll account extrapolated for an annualized number).

 

 

The US is running about $500B+ below where it should be in annualized federal tax revenues for employment payroll withheld at source.  Hopefully that should normalize at some point. 

 

In addition, most of the spending programs are one-time in nature and are ending at the Federal level.  While there may be continued support - it looks like it will be at lower levels.  So the history of these kinds of one-time shocks or national economic crises (like wars) are that the deficit blows out for a few years but then reverts back to its normal run rate after a few years' time.

 

Is it possible that this time is different?  Are politicians starting to buy into the MMT prescriptions of running much larger deficits has very little cost?  I have no idea.  But as I said, I am getting a bit nervous that MMT is being bandied about as an excuse to really blow out Federal spending.  That's new - but I don't know if its really gone mainstream yet.

 

wabuffo

 

[Cigarbutt]

^Debt like greed is good, with the proviso that it needs to be kept to a reasonable level and for productive purposes. The definition of those terms is up to you.

The 1997-2001 period is interesting (this was the time i started to take investing seriously) indeed but it's not clear if it's worth it to discuss what is correlation or causation and the direction of such?

What is amazing though is the idea that cycles no longer exist. When we entered the macro-prudential era, the initial terms of the contract implied to alternate between using a margin of safety and then building it (which IMO is the second-best approach) but this MMT thing just feels like a huge free lunch.

[wabuffo]

The 1997-2001 period is interesting (this was the time i started to take investing seriously) indeed but it's not clear if it's worth it to discuss what is correlation or causation and the direction of such?

 

"For the federal government to run a surplus, the private sector must run a deficit (ie, borrow to consume or spend)." 

 

wabuffo

[rb]

Wabuffo,

 

What is your best guess then to how this plays out in the US and Japan which have a lot of debt but also have their own currency?

Seems like could be a big cost at some point.

 

Is the most likely scenario high inflation and higher taxes with lower spending in the US and Japan?

 

Also,  how about historically all the countries that defaulted - Russia and Argentina in the last 25 years.  They had their own currency but issued dollar debt but still defaulted - couldn't one have argued that they could of just issued more debt in local currency and paid back their USD bonds rather than defaulted?  I wonder if it is sometimes easier politically to just default and screw the creditors.

 

It is interesting to me that if there is debt to the public of 200% of GDP then at "normal interest rates of say 5% interest costs alone would be 10% of GDP which is much harder to keep paying.

As Wabuffo explained earlier.

 

In respect to countries that issue debt in their own currency the debt is not so much real. It is money that a country owes itself. Japan is actually the best case of this because they haven't been running trade deficits. So those JGB that are the private sectors' assets. So what actually happens will be dictated by the private sector's appetite for assets and the associated impact with that. In your example you mention interest rates of 5%. OK, well what's gonna cause rates to go to 5% and what effect will that have on the economy?

 

To take your Japan example, if you're gonna have debt paydown in Japan that will be because the population will want to hold less bonds. That means that they will take that money and spend it which will actually be a very positive thing for Japan.

 

In the case of Russia, they had to borrow in foreign currency. That's because they were running current account deficits. Basically they were a net importer of goods. These goods were sold in rubbles. But none of the exporters wanted to hold rubbles so they exchanged them for dollars. Then the Russian government had to go out there and borrow those dollars in order to balance the equation.

[Cigarbutt]

The 1997-2001 period is interesting (this was the time i started to take investing seriously) indeed but it's not clear if it's worth it to discuss what is correlation or causation and the direction of such?

"For the federal government to run a surplus, the private sector must run a deficit (ie, borrow to consume or spend)." 

wabuffo

Thanks for the reminder related to elementary school math. :)

However, even if the saving, investing or withholding have to achieve mathematical balance, this balancing does not mean net saving or, equivalently, net investment will be high, zero or negative. The balance also does not mediate what is consumed and what is invested. The following graph shows the mathematical balance always at zero but does not help to understand what has happened to the saving or investing trend:

(picture borrowed from O'Shaughnessy Asset Management and saved for an example here)

 

If you want to apply this 'concept' to what has recently happened with Covid (just an acceleration of deeply entrenched secular trends IMO), look at the following:

(picture borrowed from Goldman Sachs and saved for an example here)

The government borrowed to compensate for the recipients in need of saving (the areas under the negative and positive curves do match quite nicely) and to support consumption and now we're back to where we were before (long term trend line) except that debt to GDP has risen to a significant degree.

It's hard to swallow that the recent rise in debt is "not so much real". ???

[LongHaul]

Thanks for the response Wabuffo and others.  I think the big deficit is more than one time and has been higher in the last 12 years than historically.

 

https://www.brookings.edu/policy2020/votervital/how-worried-should-you-be-about-the-federal-deficit-and-debt/#:~:text=For%20fiscal%20year%202019%2C%20which,GDP)%20in%20the%20previous%20year.

 

I have no idea what might make interest rates rise.  I really don't understand how they have been this low.

But I think it is very likely that they go up over the next decade to at least over the inflation rate.  Especially if you get a big country that defaults and scares the crap out of the

other democracies.  Historically financial panics came about when big banks or borrowers went under.

 

700 year low of interest rates.

https://www.visualcapitalist.com/700-year-decline-of-interest-rates/

[Spekulatius]

Thanks for the response Wabuffo and others.  I think the big deficit is more than one time and has been higher in the last 12 years than historically.

 

https://www.brookings.edu/policy2020/votervital/how-worried-should-you-be-about-the-federal-deficit-and-debt/#:~:text=For%20fiscal%20year%202019%2C%20which,GDP)%20in%20the%20previous%20year.

 

I have no idea what might make interest rates rise.  I really don't understand how they have been this low.

But I think it is very likely that they go up over the next decade to at least over the inflation rate.  Especially if you get a big country that defaults and scares the crap out of the

other democracies.  Historically financial panics came about when big banks or borrowers went under.

 

700 year low of interest rates.

https://www.visualcapitalist.com/700-year-decline-of-interest-rates/

 

Interest rates were low in the 50’s and 60’s too, especially considering there economic growth. The spending from the Vietnam war was one factor increasing inflation and interest rates.

 

This is a Fiat System and it is build on trust. Once trust wanes, all bets are off. The Fed will lose control of the narrative, the USD May sell off, inflation will rise. Once the cat is out of the bag, it will be very difficult to get it back in. last time this happened we needed a Volker to ale some hard decision. We will see what happens next time.

[scorpioncapital]

And how did Volker erase decades and years of mistrust? I mean in countries like Greece or Argentina you have serial default for their entire history, one after the other but far enough apart that old incidents are forgotten. I would say a short memory does more to clear distrust (temporarily) than any action by a harsh central banker. Even still you would expect average rates to be permanently higher the more such incidents exist.

[wabuffo]

I have no idea what might make interest rates rise.  I really don't understand how they have been this low.

 

They are low because the rate paid on US Treasuries at issuance is at the discretion of the US Federal government.  It can be zero forever, if that's what the Treasury decides. 

 

That's because:

a) the US Treasury spends first, creating new reserves in the banking system's accounts at the Federal Reserve. (this creates new financial assets for private sector)

b) then the US Treasury comes in and removes these excess bank reserves via borrowing (conducts an asset swap with private sector for previously created financial assets). 

 

In true monetary and economic terms, its not really borrowing and using that word confuses everyone. US Treasury debt issuance is a hygiene activity for bank reserves - otherwise they would accumulate and drown the banking system.  TINA - there is no alternative.  In fact, the private sector prefers a Treasury asset over a bank reserve because bank reserves are illiquid and trapped at the Fed.  Treasuries are used as collateral because there is a shortage of credit-risk-free collateral around the world.  So even in a world of zero rates on bank reserves and zero rates across the entire yield curve, there would still be a preference for US Treasury debt due to its liquidity and safety as collateral.

 

The only real constraint is not borrowing capacity, then, because its not really borrowing.  The real constraint from deficit spending is currency debasement.  If there is enough currency debasement such that there is obvious inflation, then what will the Fed and US Treasury do?  That's the big question.  I don't have an answer - but the Fed is hinting that they will tolerate it and keep rates low for a long time to compensate for undershooting inflation in the past.  There do not appear to be any Volcker's at the Fed.  And I say all this while believing that the Fed is not as powerful as everyone believes (the real 800-lb gorilla of monetary policy is the US Treasury).

 

In addition to there being no Volckers (who I think is overrated frankly), I would add that there are no Greenspans either (who I think was the best Fed Chairman ever).  Despite the grief Greenspan gets, he always kept an eye on the gold price (as well as other commodities) and used it to guide Fed policy.  There's some quantitative proof for this.  Greenspan kept the USD more stable in price vs gold for longer than any other Chairman of the Fed.  To this day I'm convinced that Greenspan kept his target at $350/oz (10x the old gold price peg pre-Nixon closing the gold window in 1971).  The average price of gold during his long tenure averaged pretty much at this target price (with a +/- $50 band). 

 

1) Here's a table I once made up (and just updated for Powell) to show how Fed Chairman since 1971 have done vs the price of gold.  Greenspan shines.  (so does Yellen actually - though she wasn't there long enough to really tell).

 

 

2) Here's an academic study by George Selgin (I recommend reading his stuff if you are interested at all in central bank operations) that also hypothesizes that Greenspan followed gold in his rate decisions. [not sure if this link works - so I will attach a pdf of this academic paper as well]

 

https://www.semanticscholar.org/paper/The-Price-of-Gold-and-Monetary-Policy-Lastrapes-Selgin/c5d577dd46e898781bfeff704436a483b6ac72db?p2df

 

But I think it is very likely that they go up over the next decade to at least over the inflation rate.  Especially if you get a big country that defaults and scares the crap out of the other democracies.  Historically financial panics came about when big banks or borrowers went under.

 

Well every reserve currency default is different.  But it is interesting to examine how the UK lost its reserve currency status after World War I.  In its case, there were two reasons

1) it spent heavily to fund its war effort during World War I and eventually lost its military pre-eminence to a rising United States whose entry into WWI was the difference-maker in winning that war.

2) it borrowed heavily to buy supplies and armaments from the US to fight WWI - but its borrowings were largely in USD rather than GBP.  The UK also made a mistake trying to re-peg its currency to its pre-WWI gold price unleashing UK deflation.

 

So perhaps those are the two red flags for to watch for reserve currency loss:

a) getting eclipsed as the most powerful military in the world, and

b) and being forced to conduct sovereign borrowing in a currency that is not your own.

 

At present, there does not appear to be any clear and present danger to the United States' reserve currency status under either condition a) or b).  Finally, I would add another competitive advantage that the United States has.  Its geography:

https://worldview.stratfor.com/article/geopolitics-united-states-part-1-inevitable-empire

 

wabuffo

The_Price_of_Gold_and_Monetary_Policy.pdf

[Munger_Disciple]

They are low because the rate paid on US Treasuries at issuance is at the discretion of the US Federal government.  It can be zero forever, if that's what the Treasury decides.

 

wabuffo,

 

By this do you mean that the Federal Reserve is a subsidiary of the US treasury? I thought the Fed set the interest rates (used to be ST only, now possibly the whole yield curve).

 

-MD

[wabuffo]

By this do you mean that the Federal Reserve is a subsidiary of the US treasury?

 

Munger_Disciple - the first part of your question opens up a very interesting thread.  Who owns the Federal Reserve System?  Of course, the simple answer is that its conduct is regulated by Congress.  The Fed is a product of the 1913 Federal Reserve Act and the US Treasury is the arm of the Executive Branch that manages the public purse as per the laws passed by Congress (taxation, spending, debt limits, etc).

 

But what exactly is the ownership structure of the Fed?  In what ways is it independent of the US Treasury?  In what ways is it dependent on the US Treasury?

 

Many people mistakenly think that the Federal Reserve System is a private entity because chartered banks must purchase "shares" in their local Regional Federal Reserve Bank in order to become a Federal Reserve "member bank" and have access to the Federal Reserve.  Thus, the conclusion must be that the Federal Reserve System is owned by the chartered member banks and is therefore a quasi-private organization.  But there was a recent Federal Court Case (U.S. v Wells Fargo) that decided this issue.

 

https://law.justia.com/cases/federal/appellate-courts/ca2/18-1746/18-1746-2019-11-21.html

 

In this case, the Federal Court ruled that the "shares" in the regional F.R. banks owned by chartered banks are really debt contracts of the Regional F.R. Banks and that net profits of the Federal Reserve System all belong to the US Treasury (and not the chartered banks through their "shares").  So there you have it -- the US Treasury owns the equity of the Federal Reserve.  In addition, the settlement balances in checking accounts created for the benefit of chartered banks (i.e., bank reserves used for clearing payments in the Federal settlement system) that the Fed creates are also a product of the US Federal government even though they are "unappropriated dollars".  IOW - the chartered banks trade a private sector asset for a public sector asset (i.e, a bank reserve - which in turn is "an asset" of the Fed govt via the US Treasury).  So much for the idea that bank reserves can be "withdrawn" from the Fed to make loans....

 

Basically at the inception of the Federal Reserve, the US Treasury "purchased" a payments system infrastructure (a public good) and provided it to the central bank in exchange for an equity claim equal to 100% ownership (and retaining all "profits" the Fed makes).  The Fed thus becomes the US Treasury's subsidiary (ie, the US Treasury's bank) performing three important functions: a) running the US Treasury's general account for spending and taxing transfers to/from the private sector banks, b) running the national payment clearing system with the chartered banks that are members of the Federal Reserve System, and c) acting as custodian/transfer agent for the US Treasury's debt issuance by recording all transactions and ownership.

 

In an earlier post, I said the Federal government has three types of liabilities:

a) currency in circulation

b) bank reserves

c) US Treasury debt.

 

One can now see that, in effect, these are all, ultimately, liabilities of the US Treasury (some directly, some via the Fed).  Currency is minted (or printed) inside the US Treasury organization and delivered to the Fed when it issues an order to the US Treasury to provide it with currency (that's why the Secretary of the Treasury signs US banknotes and not the Chairman of the Fed).  We saw that the Federal Court basically ruled that even bank reserves created by the Fed are unappropriated "property" of the US Treasury/Federal Government.  And of course © is obviously an obligation of the US Treasury.

 

After its creation and throughout most of its history, the Fed built up its balance sheet by issuing currency to the private sector in exchange for Treasury bills/bonds.  Since the Fed ran a very asset-light balance sheet, the remainder of its balance sheet came from banks freely adding reserves to use the payments system by borrowing them from the Fed in exchange for collateral (more Treasury bills/bonds).  In fact, right up to the GFC crisis, this was the typical Federal Reserve balance sheet - dominated by holdings of US Treasury holdings on the asset side and banknotes (ie, currency in circulation) on the liability side, each accounting for 90% of the respective sides of the Fed balance sheet.

 

All in all - a long way of saying that the only way the Fed is independent is in how it sets the interest rate on reserves.  In all other ways, operationally as well as equity-wise it is dependent on the US Treasury and the US Federal Government and its accounts really should be looked at in consolidation with the US Treasury.  It is no more independent on most of its activities than the Post Office is.

 

wabuffo

[Munger_Disciple]

Thanks wabuffo for such a lucid explanation of the interconnection between the treasury department and the fed. So it seems the only thing fed can do independently of the treasury dept is to set interest rate on bank reserves and treasuries and control what assets can the private sector can own. By purchasing MBS, treasuries, munis and corporate bonds, the fed is effectively taking them off the private sector hands, thus forcing the private sector into riskier assets to get any return.

[Cigarbutt]

Thanks wabuffo for such a lucid explanation of the interconnection between the treasury department and the fed. So it seems the only thing fed can do independently of the treasury dept is to set interest rate on bank reserves and treasuries and control what assets can the private sector can own. By purchasing MBS, treasuries, munis and corporate bonds, the fed is effectively taking them off the private sector hands, thus forcing the private sector into riskier assets to get any return.

What you describe is the hot potato effect on a large scale. By actively suppressing interest rates that were already brought down by deep secular forces, the Fed’s only significant effect (apart from unintended consequences) is to swap the maturity profile of Treasury ‘debt’ issued (bonds, bills to fiat) and the effect of the maturity profile recycling has been, through asset price inflation, to bring all forward returns closer to zero (cash) or closer to what the Fed is paying on excess reserves. This is why an argument can be made today that to-be realized returns from growth stocks like Apple, from old-economy stocks like oil-related and financials, from razor-thin spread corporate bonds and from government bonds to bills have been compressed to a very narrow band.

-----

Active suppression of interest rates (ie as the last example of this secular trend: the Fed absorbed most of the debt issued so far peri-Covid) has made it easier to issue debt (this point should not be controversial). Is this a good thing or is this the ideal choice?

Some perspective from previous CBO ‘forecasts’:

 

 

So, the Treasury has taken advantage of the fact that interest rates on its debt (which is almost yield-curve-controlled at this point) has remained low and in fact gone towards zero (and more?). Retrospectively thinking in terms of alternatives has limited usefulness because of very obvious limitations (cannot change one variable without considering second-order or domino effects), but it is obvious that the Treasury could not effectively ‘function’ now if they had been right on their forecasts of the direction of interest rates (they just followed the consensus and conventional view that growth would revert back to historical levels and more and that this would steepen the yield curve; but growth has not bounced back to historical trends (to the contrary) and the curve is ‘steep’ but the 30-yr rate is at 1.49% (!) at the time of this writing (who thought this would be possible just a few months ago?)).

 

Just for fun though, let’s see the result of this impossible scenario (what CBO predicted during this last 'recovery' in terms of rates with everything else unchanged).

-Typical historical range of net interest payments on debt as % of federal outlays: 7-15%  now: about 9%

-Typical historical range of interest (in fed outlays) as % GDP: 1.2-3.1%  now: about 1.75%

-Average interest rate on US gov. marketable debt securities: 2000: 6.41%  2007: 4.93%  now: 1.66% (!)

 

So assuming 2007 to 2000 avg rates apply now, this would mean:

-net int. as % of outlays: 27-35% (!)

-interest as % of GDP: 5.2-6.8% (!)

 

Obviously this could not have happened without something (structural reform, focus on productive growth) but the message is that this easing has put the US on a path that looks like a debt trap (Canada and others look the same). There is hope to get out of this pattern by spontaneous reversion to the mean and to grow out of it but it seems that this is ‘easy’ (or even MMT-like) magical thinking that is over and above what typical animal spirits could accomplish under present circumstances. This set up of easing rates looks more and more like a liquidity trap self-fulfilling prophecy.

 

[wabuffo]

Active suppression of interest rates (ie as the last example of this secular trend: the Fed absorbed most of the debt issued so far peri-Covid) has made it easier to issue debt (this point should not be controversial). Is this a good thing or is this the ideal choice?

 

I think its controversial because I believe its wrong.  What if the conventional wisdom has the causation totally backwards? 

 

What if the reality is that the US Treasury issues debt to help out the Fed, (instead of the Fed buying Treasury debt to help the US Treasury "borrow")? 

 

wabuffo

[Cigarbutt]

^i'd say this thread is unlikely to get traction if we start to argue about the mechanics of the Treasury Account at the Fed. :)

Your perspective, as always, is thought-provoking.

 

Your perspective seems to include that the Treasury is in the driver's seat (which i agree with), so what is the Treasury trying to achieve here? To help the Fed?

During and around the Great Depression and WW2 (extreme example but valid and maybe we're in a war, an unusual war in which we don't see (or don't want to see) the enemy), the Treasury needed to expand and to pay for unplanned expenditures so they entered into an agreement with the Fed to keep rates low. So, was the Treasury into government-sponsored social- and war-related expenditures or into Fed's reserves-management help?

A major objection to the Fed help point of view is that the Treasury took advantage of a flight-to-safety mood at whatever level of interest rates to issue debt enough to increase reserves (through central bank purchases as a side effect) AND to raise the Treasury Account (in cash or printed money) to unusually high levels. This cash reflects the planned expenditures (have you read the Cares Act and what it means for federal outlays going forward?), will eventually enter the system but is meant essentially to match the output gap and some of that will become reserves as part of the Fed's plan to keep rates low for as far as the eye can see. If the Treasury's intent was to help the Fed, why would it issue debt to the extent that a lot of it ended up parked at the Fed?

Let's say you're the CFO of a bankrupt airline in an emerging country and the IATA, somehow, becomes in charge of the terms of DIP financing and decides to keep the terms very loose. Isn't it possible then that the distressed airline would 'secure' too much debt for its ultimate productive capacity? The market 'works' because it does (when it's allowed to). Or would the airline's goal be to help IATA in search of reserves?

[wabuffo]

so what is the Treasury trying to achieve here? To help the Fed?

 

I think you already know my perspective.  Borrowing by the US Treasury is a reserve balance mop-up operation to absorb bank reserves that accumulate in the banking sector when the US Treasury net spends.  There was a ready, fire, aim approach to the crisis by both the Fed and the Treasury - so in this crisis, there was a simultaneous Treasury debt buying operation by the Fed (asset swap that increases bank reserves) + massive deficit spending via the Treasury through the CARES Act (creates new bank deposits in the banking sector but also increases bank reserves).

 

The banking sector was getting the asset side of its balance sheet massively converted to deposits at the Fed earning zero.  I'm sure there were alarms raised by the big banks by early April (especially with the big April federal tax season also postponed that would've drained some reserves).  But let's back up and look at the numbers.

 

I took a look at the US Treasury Daily Statements from March 11th (the Tom-Hanks-has-the-virus/NBA-cancels-a-game night that unleashed the all-hell-breaks-loose financial panic) to today (I'm using 9/30 - the latest Fed H.4.1 report o/s).  Here are the numbers.  The US Treasury spent $2t more than it received.  But it issued $1.4t more in debt than it needed just to maintain its settlement balance at the Fed at a constant amount (000s). 

 

US Treasury total expenditures less total receipts (ex. Public debt issues/redemptions:)($2,083,088)

US Treasury public debt issues less public debt receipts:                                           $3,492,430

Net Change in US Treasury general account balance at the Federal Reserve:           (A)$1,409,342

 

We can foot these numbers with the change in its federal reserve account deposit balance as reported by the Fed.

 

US Treasury general account balance at 9/30/20:                                                       $1,781,679

US Treasury general account balance at 3/11/20:                                                         $372,337

Net Change in US Treasury general account balance at the Federal Reserve:           (A)$1,409,342

 

So why did the US Treasury borrow so much more than it needed?  Since the GFC, it typically runs its account balance at around $400b +/-.  I believe it was to help the banking system by soaking up $1.4t in settlement balances that would've remained in the banking sector.  When the US Treasury issues a $1000 bond, it moves $1000 from the banks' reserve balances at the Fed to its own account.  (asset swap).  I drew up a week-by-week chart to compare what bank reserves would've looked like if the US Treasury kept its balance at $400b every week vs what they actually came in at.  The orange line is the actual weekly reserve levels while the blue line represents where reserves would've been if the US Treasury didn't help the Fed out by building up its general account (even if it didn't need the "money").  IOW, the Treasury's general account would remained at $400b every week - and borrowing would match net spending.  Remember the commercial banking sector has ~$20t in total assets - so bank reserves w/o the Treasury's help would've exceeded 20% of all assets (earning basically nothing).

 

 

I would also point out that this crisis is a little different than WWII or even the GFC.  During WWII, the US was still under the gold standard - so the reserve currency was gold (and not the USD).  During the GFC there were also aggressive responses by both the US Treasury and the Fed -- but in opposite directions.  The Fed was running an interest rate target with no excess reserves.  In the financial panic of 2008, fed funds rates were spiking - so the Fed was selling Treasuries (not buying them) - the so-called "Treasuries for trash" program.  In fact, here too, the US Treasury issued debt to help out the Fed but in a different way.  In a one-off, double-secret probation letter kind of deal, the US Treasury sold ~$560b of short-term bills directly to the Fed (Fed is only supposed to buy in the open market).  The reason was that there was a real worry that the Fed was going to run out of "bullets" (ie Treasuries) on its balance sheet because of Treasuries-for-trash.

 

This is also the point in my ramble where I take my obligatory swipe at the bureaucratic stupidity of the Fed's responses to external stimuli.  The whole point of the Fed's aggressive buying of Treasury assets was to increase its share vs the private sector's share (so as to reduce "supply").  But because the Fed needed to be bailed out by the US Treasury as noted above, the Fed actually owns a smaller percentage of US Treasury net assets than it did before the crisis started.  Doh!

 

wabuffo

[Cigarbutt]

^In simple terms, the Treasury increased the cash held in its General Account by 1,4T, which, by definition, drained reserves in the banking system by the same amount. The question is why?

Was it planned as a primary purpose or was it only a side effect?

It's possible that 'cooperative' discussions occurred, especially during the last days of March (the repo and reverse repo market showed remarkable swings) and the content of those discussions may never reach the ordinary citizen but there are several problems with the need to drain reserves hypothesis:

-Just like many private parties, including many corporations, available credit lines were used to build a cash position based on a very uncertain outcome, in terms of future cashflows necessary to get through the episode. Many corporations will adjust and have and will use cash to repay at least part of the credit line and the Treasury simply needs to lower debt issue for a while and spend the cash in the Account for its own general corporate-like purposes. The next few quarters will help to find out.

-Cash in the Fed Treasury General Account has gone up since the GFC but it was well below 100B when excess reserves reached their pre-Covid maximum in mid 2014 (excess reserves at 2.6 to 2.7T). There was no need to drain reserves then. We live now in an abundant excess reserves world. A new normal low point was reached somewhere around 2019 but maximum levels are only tied to theoretical constraints and Japan has showed that excess reserves can reach essentially the GDP level of the country.

-The Account can be seen as a checking account with a complicated overdraft procedure so it's normal for the Treasury to raise its cash level in this account in correlation to its relentlessly growing needs and as a function of the uncertainty it is facing.

 

We can expect variations in this Account (up and down) in accordance with timing issues but i maintain that we ain't seen nothing yet in terms of debt issues over time by the Federal government.

[Cigarbutt]

^Follow-up note

 

The order of the day: mischaracterized levels of uncertainty and conventional acceptance that massive monetary and fiscal support will continue.

During Q3, the Treasury borrowed 454B and slightly increased the Treasury cash balance at the Fed by 60B (to 1782B). For Q4, they expect to borrow 617B (599B less than announced last August) and expect a closing balance down to 800B. As always, this remains to be seen and will depend on where the wind blows. The numbers will help to explain if the main goal of the Treasury is to use the Fed General Account as a checking\deposit account or to soak up reserves in order to help banks (banking system).

https://home.treasury.gov/news/press-releases/sm1172

 

In a separate release, the Treasury described that they plan to issue longer-dated tenors.

 

The Fed is likely to continue and perhaps increase its asset purchase programs and it’s expected that excess reserves deposited at the Fed will grow or, at least, be maintained at present levels.

Opinion: it’s likely that the correlation that has recently happened between growing excess reserves and growing Treasury cash balance at the Fed will break down.

 

-Additional note about a potential misconception #1

In this thread (and at least another), it’s been submitted that the Treasury is the primary ‘creator’ of money. This is playing with words to a certain degree but is not valid on a fundamental level. When the Treasury credits a private account with money, it also (pretty much simultaneously for practical purposes) records a matching liability which means that the net result is a positive equity event for the private market in exchange for a matching negative equity event on the public ledger. The money ‘created’ is fundamentally a temporary form of asset swap between market participants and is fundamentally based on the idea that the debt held by somebody will be paid back eventually by somebody else. Even if most of this debt money can be rolled over, it is an asset that is matched by a liability. How the link between the two can be stretched is the question. It’s like an elastic band. The energy required to stretch is not constant and it can snap back. The Fed, by acquiring the debt issued by the Treasury and by keeping interest lower than otherwise, have ‘helped’ to decrease the tension in the elastic band but the iron laws of balance do eventually apply. IMO, the ‘real’ money creation is from the banks, through the fractional reserve system. Credit growth is bound to get jammed in higher-debt-levels environments and more debt does not seem to be constructive for the ‘real’ economy.

 

 

-Additional note about a potential misconception #2

It’s been mentioned that excess reserves have been detrimental to banks implying also that they had to hold higher amounts (proportion) of low yielding ‘cash’. I think the higher proportion of cash holdings is a fact but the higher cash holdings have not happened at the expense of other (potentially) higher yielding assets; it simply represents the money that the government has ‘injected’ in the private system as a result of record debt issuance.

Following the data from 2019 to 2020 so far:

H.8 releases; assets and liabilities of large domestically chartered commercial banks in US, seasonally adjusted; similar patterns exist when including small and foreign banks with a slight difference for larger banks for which the increase in cash assets as a percentage of increase in total assets was slightly higher.

 

                                          Sept2019    March2020    June 2020    Sept2020    Oct212020

Total Assets (B)                      10059          10862            11717          11691          11728

Cash Assets (B)                      755              1005              1729            1582            1675

%CA/TA                                  7.5                9.3                14.8            13.5              14.3

 

TA, over the period, increased by 1669B, more than the increase of CA (920B).

So the total assets have increased at the banks’ level and the increase in cash assets has not happened at the expense of other assets although it’s becoming clear that holding a higher proportion of cash assets and a higher proportion of government debt with ultra-low yields (+\- asset swap operations of the Fed) will tend to lower the return on capital measure for banks. So, there does not appear to be a role for reserves soaking in order to help the Fed or the banks.

 

-Final note

By issuing a record amount of debt, the Treasury has run ahead of its needs and has increased its cash balance at the Fed. Doing the latter is a temporary equivalent opposite of what the Fed is doing with its quantitative easing program. The Fed virtually prints cash (liability) in exchange for bonds (asset). When the Treasury parks cash at the Fed, it becomes a cash deposit with, from the Fed’s accounting point of view, the deposit as a liability and cash as an asset. It is ironic to see this apparent conflict as the notions of collaboration and independence are being redefined. But again, IMO, the increase in the cash balance is temporary and will be related to cyclical trends contrary to rising debt levels which are more secular.

 

Disclosure: After having spent some time on this, i must admit that my level of confusion has only increased about today's markets.

 

[wabuffo]

<b><i>The numbers will help to explain if the main goal of the Treasury is to use the Fed General Account as a checking\deposit account or to soak up reserves in order to help banks (banking system).</b></i>

 

<b><i>Opinion: it’s likely that the correlation that has recently happened between growing excess reserves and growing Treasury cash balance at the Fed will break down.  </b></i>

 

Agree to disagree.  You know where I stand on this.  The US Treasury set the same goal for Q3 ($800B at the TGA) and didn't even try to hit it.  I don’t believe them.  The Fed asked them to increase the size of their TGA account to help the banks.  Here's what reserves would've looked like if the US Treasury kept its TGA at its historical $400B average target (click on charts to view full-size).  Bank reserves would've grown to over $4.2t (or over 20% of total bank assets) which is a real problem when total assets only grew by $1.5t.

 

                             

 

After having spent some time on this, i must admit that my level of confusion has only increased about today's markets. ...When the Treasury credits a private account with money, it also (pretty much simultaneously for practical purposes) records a matching liability

 

Its actually pretty simple as you've stated it.  What is money?  Is it a unit of account or is it actually always a fiat government liability?  I think the latter.  As I've indicated elsewhere, you need to consolidate the accounts of the US Treasury and the Fed on one side and the private sector on the other.  The Feds (Treasury and Reserve) have three types of liabilities (with balance sheet amount as at 10/28):  The US govt gives you a choice on how you want to hold your govt asset (in effect you can have it as cash, demand deposit or time deposit).

1) Currency in circulation  ($2,042.7b)

2) Bank Reserves ($2,947.3b)

3) Treasury Debt, net held by the public less any Federal Reserve holdings. ($16,546.7b)

 

Here's a chart for this year that shows the growth in total of all three added together (these are the private sector's asset holdings of Fed govt liabilities).  The real rapid growth happened in March-May.

 

                               

 

IMO, the ‘real’ money creation is from the banks, through the fractional reserve system.

 

Go back to the three types of Federal govt liabilities above.  Total bank credit since the crisis has only grown by $818b.  So how did total holdings of Fed govt liabilities grow by almost $3t?  Where did that "money" come from?  As agents of the Federal Reserve banking system, banks can create some money by following the regulations they operate under, but the vast majority of money (in terms of money as a Federal govt liability) is created by the US Treasury.  Go look at the Q2, 2020 involving the US Treasury, the CARES Act EIP debit card program and Meta Financial (CASH) to see a micro- real-world example of how this works (no lending involved there, yet CASH doubled its assets in one quarter).

 

wabuffo

 

p.s your bank cash assets don't make sense to me.  Total reserves (i.e, cash assets) are nearly $3t now. 

[Cigarbutt]

...

Agree to disagree.  You know where I stand on this.  The US Treasury set the same goal for Q3 ($800B at the TGA) and didn't even try to hit it.  I don’t believe them.  The Fed asked them to increase the size of their TGA account to help the banks.  Here's what reserves would've looked like if the US Treasury kept its TGA at its historical $400B average target (click on charts to view full-size).  Bank reserves would've grown to over $4.2t (or over 20% of total bank assets) which is a real problem when total assets only grew by $1.5t.

wabuffo

p.s your bank cash assets don't make sense to me.  Total reserves (i.e, cash assets) are nearly $3t now.

Also, from a previous post:

"The banking sector was getting the asset side of its balance sheet massively converted to deposits at the Fed earning zero."

---)

Interesting and helpful as always and this is a reply, not to attack you, but to attack the foundations upon which you stand. :)

The data from the H.8 release, as mentioned, is for large domestically chartered commercial banks in the US. There is a reference below for the specific data recently released.

For all banks, seasonally adjusted, the picture is similar and you get your magic 3T number:

 

                                          Sept2019    March2020    June 2020    Sept2020    Oct212020

Total Assets (B)                      17524          18919            20200          20171          20228

Cash Assets (B)                      1635            2177            3058            2932            3027

%CA/TA                                  9.3              11.5              15.1            14.5            15.0

 

TA, over the period, increased by 2704B, more than the increase of CA (1392B). So, i come to a similar conclusion. If you follow the % of cash assets over total assets in US banks since 2009, two things stand out: 1) the total assets apart from cash assets have grown quite consistently (as a function of economic activity and government borrowing) and 2) cash assets % have varied (swap operation with bonds) as a function of quantitative easing, some tightening and then MORE easing. When the government credits bank accounts, they are creating new assets in the private banking system and, through quantitative easing, this newly 'created' money ends up as excess reserves so that both cash assets AND total assets held by banks increase by the same amount. For example, let's say for fun that i'm able to enter the Eccles Building and reach the Ultimate Computer and transfer all the money (assume 1782B and October 21 numbers)) from the Treasury General Account into bank accounts of private individuals (distribution by helicopter would also work to some degree for the example but it would take longer and would be more 'messy'). TA become 22010B and cash assets become 4809B. CA over TA becomes 21.9%. This would make the return on assets number go lower but where is the 'massive conversion' or the 'real problem'?

https://www.federalreserve.gov/releases/h8/current/default.htm

See footnote 18 for the standard definition of cash assets

PS i'm spending time on large (black-box) banks these days but will eventually try to understand Meta Financial, at some point.

[wabuffo]

Interesting and helpful as always and this is a reply, not to attack you, but to attack the foundations upon which you stand

 

Sorry CB - I'm kind of dense so I don't really understand what you are trying to say (attack?).  Is your point about MMT?  Is it about bank reserves? 

 

From a monetary system perspective:

 

1) If the only thing happening is the Fed is buying Treasuries in order to increase the size of its balance sheet (and the US Treasury is running a balanced budget).  Then, yes, the Fed will convert the US commercial banking sector's assets into a greater percentage of cash assets (reserves) while total assets more or less stay constant because the Fed is swapping a bank asset for another bank asset and not creating any new asset.

 

2) If the only thing happening is the US Treasury deficit spending while the Fed stands pat (rolls over its Treasuries but doesn't add or subtract from its pile).  Then the US commercial banking sector's cash AND total assets will grow together (ie - a new asset will be created - a reserve balance; as well as a new liability - a bank deposit).  This was the Meta example.

 

3) When both are happening - then both the cash assets will grow slightly faster than total assets.

 

If this is what you are arguing, I have no dispute with this and the data would show the same thing.  This is much of what has been happening since March (unless you are Wells Fargo and are under a hard asset cap due to being in the Fed's penalty box).

 

wabuffo

[Cigarbutt]

Interesting and helpful as always and this is a reply, not to attack you, but to attack the foundations upon which you stand

Sorry CB - I'm kind of dense so I don't really understand what you are trying to say (attack?).  Is your point about MMT?  Is it about bank reserves? 

From a monetary system perspective:

...

wabuffo

An argument is like a (mental) fight and learning is often proportional to the intensity. The comments are a personal (in my direction) reminder that, when engaging with you, the "opponent" from whom i'm learning is clearly superior in many respects. Arguing is great and i don't know many people who are interested by the raw data released by the Federal Reserve. :)

-----

Your last post helps to put things in perspective.

From a pure accounting point of view it's hard to argue that receiving money is a bad thing. So, in relation to the spirit of this thread, the important underlying questions are: 1-how the liability is funded and 2-how the asset is invested.

 

Let's say Berkshire Hathaway tomorrow receives 150B from the Treasury. The outcome, in large part, depends on how the new capital will be invested. i have a feeling (we may again disagree here) that Mr. Buffett would keep the funds in cash or equivalents (he may increase the investment in Barrick Gold) . Regulations may also hinder how the funds would be redeployed and may, for the more extreme asset cap example, prompt Mr. Buffett to simply return the money, something that Wells Fargo cannot do. The big issue i continue to have about quantitative easing is that it's simply an asset swap between assets that have similar yields and is being done in an environment requiring banks to hold more capital and to be more liquid. Keeping interest rates low has not (IMO) resulted in productive investments into the real economy (see velocity of money trends, whatever that means) and low rates have facilitated government borrowing (see G-7 countries' debt levels vs GDP in a previous reply). Is this cyclical or something else is the question. For my part, i see the asset cap as a relative positive for Wells Fargo as now may not be a good time to grow and later may offer a better environment for growth. Of course, i could be wrong.

 

It looks like the next few months will be interesting, will offer food for thought and further opportunities to learn. i would offer the opinion that the 27T number may look completely irrelevant.

[wabuffo]

i would offer the opinion that the 27T number may look completely irrelevant.

Thanks for your thoughts, CB.

 

Can we also agree that the $27T number is not just irrelevant, its plain wrong?  As of yesterday's Fed H.4.1 report (plus US Treasury's daily statement for the same day), the total liabilities of the Federal government (currency in circulation + bank reserves + Treasury debt held by the public less that held by the Fed) = $21.648T. 

 

The private sector has decided it wants to hold the portion of its wealth ($21.648T) that is in US Federal Govt assets ("money") in the following mix:

a) cash  (9.5%)                        [currency in circulation]

b) demand deposit (13.8%)      [bank reserves]

c) time deposit (76.8%)            [Treasury bills and bonds]

 

If our total wealth doesn't change, but the Fed switches some our wealth from time deposits to demand deposits, why does the Fed think we will spend more? (ie, more consumption and investment).  If one thinks about it that way, it shows how ridiculous the Fed's policy is.  Plus, as I've shown before, lowering interest rates hurts savers more than it helps borrowers - so savers save even more, further hurting consumption.  The savings from interest rates may not even hold.  When the Fed buys Treasuries, it is swapping a long-term fixed interest rate (interest rate on a, say, 30-year Treasury bond) for a short term variable rate (interest rate on excess reserves).  While in the short-run that might lower the interest rate the Govt pays (and I'm including the Fed here as part of the consolidated Federal govt), in the long run it could actually cost more if rates rise.

 

Its crazy, stupid policy.

 

wabuffo

[LongHaul]

so what is the Treasury trying to achieve here? To help the Fed?

 

I think you already know my perspective.  Borrowing by the US Treasury is a reserve balance mop-up operation to absorb bank reserves that accumulate in the banking sector when the US Treasury net spends.  There was a ready, fire, aim approach to the crisis by both the Fed and the Treasury - so in this crisis, there was a simultaneous Treasury debt buying operation by the Fed (asset swap that increases bank reserves) + massive deficit spending via the Treasury through the CARES Act (creates new bank deposits in the banking sector but also increases bank reserves).

 

The banking sector was getting the asset side of its balance sheet massively converted to deposits at the Fed earning zero.  I'm sure there were alarms raised by the big banks by early April (especially with the big April federal tax season also postponed that would've drained some reserves).  But let's back up and look at the numbers.

 

I took a look at the US Treasury Daily Statements from March 11th (the Tom-Hanks-has-the-virus/NBA-cancels-a-game night that unleashed the all-hell-breaks-loose financial panic) to today (I'm using 9/30 - the latest Fed H.4.1 report o/s).  Here are the numbers.  The US Treasury spent $2t more than it received.  But it issued $1.4t more in debt than it needed just to maintain its settlement balance at the Fed at a constant amount (000s). 

 

US Treasury total expenditures less total receipts (ex. Public debt issues/redemptions:)($2,083,088)

US Treasury public debt issues less public debt receipts:                                           $3,492,430

Net Change in US Treasury general account balance at the Federal Reserve:           (A)$1,409,342

 

We can foot these numbers with the change in its federal reserve account deposit balance as reported by the Fed.

 

US Treasury general account balance at 9/30/20:                                                       $1,781,679

US Treasury general account balance at 3/11/20:                                                         $372,337

Net Change in US Treasury general account balance at the Federal Reserve:           (A)$1,409,342

 

So why did the US Treasury borrow so much more than it needed?  Since the GFC, it typically runs its account balance at around $400b +/-.  I believe it was to help the banking system by soaking up $1.4t in settlement balances that would've remained in the banking sector.  When the US Treasury issues a $1000 bond, it moves $1000 from the banks' reserve balances at the Fed to its own account.  (asset swap).  I drew up a week-by-week chart to compare what bank reserves would've looked like if the US Treasury kept its balance at $400b every week vs what they actually came in at.  The orange line is the actual weekly reserve levels while the blue line represents where reserves would've been if the US Treasury didn't help the Fed out by building up its general account (even if it didn't need the "money").  IOW, the Treasury's general account would remained at $400b every week - and borrowing would match net spending.  Remember the commercial banking sector has ~$20t in total assets - so bank reserves w/o the Treasury's help would've exceeded 20% of all assets (earning basically nothing).

 

 

I would also point out that this crisis is a little different than WWII or even the GFC.  During WWII, the US was still under the gold standard - so the reserve currency was gold (and not the USD).  During the GFC there were also aggressive responses by both the US Treasury and the Fed -- but in opposite directions.  The Fed was running an interest rate target with no excess reserves.  In the financial panic of 2008, fed funds rates were spiking - so the Fed was selling Treasuries (not buying them) - the so-called "Treasuries for trash" program.  In fact, here too, the US Treasury issued debt to help out the Fed but in a different way.  In a one-off, double-secret probation letter kind of deal, the US Treasury sold ~$560b of short-term bills directly to the Fed (Fed is only supposed to buy in the open market).  The reason was that there was a real worry that the Fed was going to run out of "bullets" (ie Treasuries) on its balance sheet because of Treasuries-for-trash.

 

This is also the point in my ramble where I take my obligatory swipe at the bureaucratic stupidity of the Fed's responses to external stimuli.  The whole point of the Fed's aggressive buying of Treasury assets was to increase its share vs the private sector's share (so as to reduce "supply").  But because the Fed needed to be bailed out by the US Treasury as noted above, the Fed actually owns a smaller percentage of US Treasury net assets than it did before the crisis started.  Doh!

 

wabuffo

 

 

"In fact, here too, the US Treasury issued debt to help out the Fed but in a different way.  In a one-off, double-secret probation letter kind of deal, the US Treasury sold ~$560b of short-term bills directly to the Fed (Fed is only supposed to buy in the open market)."

 

Was that legal?  More importantly is that why it is super unlikely the US Treasury should not default because the Fed can just payoff their debt?

Seems like this could be highly inflationary if done routinely.

 

 

https://www.federalreserve.gov/faqs/money_12851.htm

"Why doesn't the Federal Reserve just buy Treasury securities directly from the U.S. Treasury?

The Federal Reserve Act specifies that the Federal Reserve may buy and sell Treasury securities only in the "open market." The Federal Reserve meets this statutory requirement by conducting its purchases and sales of securities chiefly through transactions with a group of major financial firms--so-called primary dealers--that have an established trading relationship with the Federal Reserve Bank of New York (FRBNY). These transactions are commonly referred to as open market operations and are the main tool through which the Federal Reserve adjusts its holdings of securities. Conducting transactions in the open market, rather than directly with the Treasury, supports the independence of the central bank in the conduct of monetary policy. Most of the Treasury securities that the Federal Reserve has purchased have been "old" securities that were issued by the Treasury some time ago. The prices for new Treasury securities are set by private market demand and supply conditions through Treasury auctions."

[wabuffo]

Was that legal? 

 

Well they did it and issued a press release saying they were doing it.

https://www.treasury.gov/press-center/press-releases/pages/hp1144.aspx

The Treasury Department announced today the initiation of a temporary Supplementary Financing Program at the request of the Federal Reserve.  The program will consist of a series of Treasury bills, apart from Treasury's current borrowing program, which will provide cash for use in the Federal Reserve initiatives.

 

More importantly is that why it is super unlikely the US Treasury should not default because the Fed can just payoff their debt?    Seems like this could be highly inflationary if done routinely.

 

LH -- follow your thinking.  There's nothing being "paid off" - its just a swapping of liabilities.  But the liabilities are still real.  They are just financed differently.

 

If the Fed buys debt directly from the US Treasury what does it change?  If the US Treasury swaps $100b in 30 year T-Bonds with the Fed, and then spends it, here's what happens:

 

Fed swaps T-Bonds for a "deposit balance at Treasury's General Account" with the US Treasury

1) Federal Reserve asset:      +$100B T-Bonds

    Federal Reserve liabilities:  +$100B  Tsy General Acct balance

 

Treasury spends the $100B which gets deposited in the US banking system

2) Federal Reserve liability:    -$100B  Tsy General Acct balance

    Federal Reserve liability:    +$100B  Bank Reserve Balances

 

So, net, net - we get $100B of spending which would've been funded by the US Treasury issuing $100B in new 30-year Treasury Bonds (a liability of the US Treasury) but instead we still have $100B of spending but instead it's funded by bank reserves (a liability of the Federal Reserve).

 

The Federal govt (on a consolidated basis) still has "debt" of $100B.  Is it better that it is funded by debt at the Federal Reserve than at the US Treasury?  Is it even cheaper to do it this way?  Perhaps in the short-term, the rate on excess reserves is lower than the fixed rate on a 30-year Treasury, but that short-term rate is variable whereas the rate on the 30-year is fixed.  Will it continue to be cheaper in 5 years, 10-years, 30-years from now vs the fixed current yield from issuing a 30-year Treasury bond today?

 

Inflation can always occur due to excess spending, but "monetization" of Federal debt is a nonsensical issue.  If someone starts hand-waving about "monetization of debt", you should just ignore them.  Its like someone trying to talk sports and saying "I think Tom Brady is going to hit 30 home runs for the Tampa Bay Lightning this year" -- you hear that and you just snicker.  The core issue is deficit spending levels vs GDP, not "how its financed". 

 

Hope this helps.

 

wabuffo