M2 Money supply growing at 28.4%

[LearningMachine]

Have folks been paying attention to the M2 Money supply?

 

Since March 2, 2020, it has increased 23.7% (annualized rate of 28.4%) from $15.513 Trillion to $19.197 Trillion, much faster then it grew during money printing days of GFC.

 

Maybe we'll get high inflation and high interest rates this time.

 

Source: https://fred.stlouisfed.org/series/M2

 

 

[JRM]

I think I posted a link to this interview in a different thread, but I think its an excellent discussion of the topic:

 

https://podcasts.apple.com/us/podcast/the-end-game-ep-6-lacy-hunt/id1508585135?i=1000487560045

 

Lacy Hunt has some very good insights as to why M2 alone doesn't account for CPI inflation, and what could change that.

[wabuffo]

I think of of money supply this way:

Private sector assets = Public sector (i.e., Federal govt) liabilities. 

 

So money is basically a liability of the Federal government and an asset of the private sector.  As I've indicated in other posts, the Federal government offers the private sector three forms of "money":

a) currency in circulation

b) reserve balances with the Federal Reserve banks

c) US Treasury securities

 

In my view, the traditional measures of money supply - M1, M2, MZM are simply imperfect measures of private sector assets that can be more easily measured by looking at their mirror image of Federal government liabilities of "moneyness".

 

Here's a FRED chart that puts it together.  The formula is: (currency in circulation + reserve balances at the Fed + US Treasury debt held by the public - US Treasury debt held by the Fed).  As of this Dec. 30th, 2020 - the total money liabilities of the Federal government stood at $22t.  That liability has grown 18.9% since March 4, 2020.  While that sounds impressive - this measure has been growing at +10.8% per year since mid-2008.  [click on chart for full-screen viewing]

 

 

Ok - so what is the effect of all this?  My own theory is that it is already having an effect - just not in CPI measures, but in debasement of the currency which manifests itself in asset inflation.  The canary in the coal mine for me is gold which has been very sensitive to this "money" growth.  Here's two examples from this dataset - one when money "supply" fell significantly and one when money "supply" increased significantly.

 

The first is from 1997-2001 when the US actually ran a Federal surplus (taxes exceeded spending).  This caused "money supply" to fall for one of the few times in our recent history (-4% per year over this period).  Look what happened to gold - it fell too.  In addition, the stock market had a three-year funk (2000-2002) as it finally started to feel the cumulative deflationary effect at the beginning of 2000.

 

 

The next chart is after the GFC, when spending ramped up in response to the crisis, "money supply" increased 17% per year from mid-2008 to the end of 2012.  Gold responded to this as well.  Now the relationship between money supply growth and asset inflation (or gold) isn't linear or perfect so its not a perfect "hard and fast" rule.  But I think the general relationship makes sense to me as the supply of new gold mined every year is around 1.8% of the above-ground gold inventory.  Gold's monetary attribute is stability since it grows very slowly.  This is also what Bitcoin is trying to do - grow supply at 2% per year (like gold).

 

 

My guess is that the reason gold is jumping again since early December is because it is starting to "feel" the effect of this second round of stimulus that has begun this week and will start to appear in the US Treasury spending numbers in January.

 

FWIW,

wabuffo

[LearningMachine]

My own theory is that it is already having an effect - just not in CPI measures

 

I agree with this. So far, the goods and services that CPI measures have not gone up in price.  Part of it is probably that folks impacting the marginal/supply demand of those goods and services have not been getting the increased money supply.

 

With minimum wage going up, I think those folks will start getting a part of the increased money supply as well, and they will then compete with each other on the same limited goods and services to raise their prices.

 

Regarding asset inflation so far - yes, part of it could be due to money supply, but I think for a lot of assets, e.g. real estate, it has been due to the lower interest rates.  With inflation, will come high interest rates, which will be negative for CRE.

[Guest cherzeca]

@wabuffo

 

"Private sector assets = Public sector (i.e., Federal govt) liabilities."

 

how about public sector assets? how much is the Grand Canyon (etc) worth?

[wabuffo]

how about public sector assets? how much is the Grand Canyon (etc) worth?

 

i was talking about the US monetary system - you know, more mundane stuff. Accounting ledger debits and credits....

 

I don't do the big picture stuff like you do. My cognitive capacity is quite limited.  8)

 

 

wabuffo

[Morgan]

Perhaps I don’t understand M2 completely, but the price of everything I bought in 2020 for my company (plumbing and electric supplies, lumber, tools, etc) has gone up by at least 10%. Frequently 20% - 100% increases in prices. Lumber in particular has been completely insane. This is anecdotal of course, but everything is getting more expensive!

[gary17]

that’s because everybody is renovating their homes during covid lockdowns

[mattee2264]

 

I think what is different this time is the money supply growth is enabling enormous budget deficits. That has a much more direct impact on aggregate demand and as capacity is still somewhat constrained I'm expecting quite significant inflation later this year (that will ease slightly as encouraged by strong demand businesses are incentivized to ramp up)

[wabuffo]

I think what is different this time is the money supply growth is enabling enormous budget deficits.

 

Is it really different this time, though?  The numbers look big, but as always I think its important to put them in context.

 

During the response to the GFC, the US hit a peak deficit-to-GDP ratio on a trailing-twelve month basis during Sep Q 2009 of 12.58%. 

During this crisis, the peak deficit-to-GDP ratio on a trailing-twelve month basis was during the June Q 2020 of 14.8%.  The deficit continued to go up on a TTM basis for the Sep Q 2020 but so did GDP such that the ratio for Sep Q 2020 actually fell slightly to 13.31% of GDP.

 

So yes - the budget deficits are a bit bigger, but not enormously so, when place in the context of the size of the US economy. 

 

Did we have inflation after 2009?  I didn't see much of it.  Did we have currency debasement?  That we sure did - gold took off and hit a peak in 2012 but declined after that as deficit-to-gdp ratios came back down to the single-digit range percentages.

 

wabuffo

[JRM]

Isn't there an argument to be made that the GDP growth was from the increase in debt?

[Vish_ram]

Increase in money supply should be viewed in relation to velocity of money. The velocity has been dramatically decreasing.

 

Given the backdrop of dramatic productivity gains due to technology, decreasing population growth, greater consumption of services compared to goods, dramatic capacity expansion in commodity production (during first decade of this century),

we are having lower real rates, lower fed funds rate, lower inflation, lower velocity & increasing money supply. The gold bulls have totally misread the situation and are purely engaged in mental masturbation.

 

The so called money printing hasn't increased inflation. It has only gone to support existing Treasury's mandatory spending.

 

Think of a couple with several kids (school tuition, books, clothing spending), supporting their parents (social security, medicare).. Now if wife is unemployed & husband works part time, then their rich uncle (Fed) uses his credit card to loan some money to this couple. The couple is not splurging on anything, just maintaining their usual spending without missing a beat.

[LearningMachine]

The velocity has been dramatically decreasing.

 

So far.  An interesting article on why velocity has been decreasing so far by the Federal Reserve Bank of St. Louis: https://www.stlouisfed.org/on-the-economy/2014/september/what-does-money-velocity-tell-us-about-low-inflation-in-the-us

 

The answer lies in the private sector’s dramatic increase in their willingness to hoard money instead of spend it.

...

And why then would people suddenly decide to hoard money instead of spend it? A possible answer lies in the combination of two issues:

 

A glooming economy after the financial crisis

The dramatic decrease in interest rates that has forced investors to readjust their portfolios toward liquid money and away from interest-bearing assets such as government bonds

 

I wonder if it is a question of "when" people might stop hoarding money not a question of "if".

 

 

[wabuffo]

I wonder if it is a question of "when" people might stop hoarding money not a question of "if".

 

The problem is that one of the largest components in the calculation of "velocity" in the monetary base is reserves.  I can't seem to access my reserve account at the Fed so I guess I'm one of the hoarders.  8)

 

"Velocity" started to fall when the Fed expanded its balance sheet in 2008 by forcing the US commercial banking sector to hold over $3 trillion of reserves when they used to hold less than $5 billion pre-2008.  The quantity of bank reserves is a policy decision controlled by the Fed (banks as a total sector don't have any choice in the quantity of reserves the industry must hold on deposit at the Fed).  Reserves are "frozen" assets that do nothing but sit there to help clear payments.

 

If one is actually trying to measure velocity - I think you could look to the Federal payments systems (Fedwire, CHiPs) and measure the number of turns vs something like GDP.  I haven't updated this in awhile, but this is a chart I drew up comparing these two things ($ value of payment transfers & GDP)

 

Currently, the Federal Reserve payment systems settle over $1.1 quadrillion in payments per year.  Annual US GDP is $21 trillion - give or take.  So every $1 of GDP requires over $60 of payments to circulate on average through the US economy.  This doesn't include payments in cash - but they are small relative to non-cash so I ignore them.  Here's the historical chart. 

 

 

The ratio bobs around a bit but is up in 2020, FWIW - if my little home-grown "velocity" ratio actually tells us anything useful.

 

wabuffo

 

[LearningMachine]

"Velocity" started to fall when the Fed expanded its balance sheet in 2008 by forcing the US commercial banking sector to hold over $3 trillion of reserves when they used to hold less than $5 billion pre-2008.  The quantity of bank reserves is a policy decision controlled by the Fed (banks as a total sector don't have any choice in the quantity of reserves the industry must hold on deposit at the Fed). 

 

As announced on March 15, 2020, the Board reduced reserve requirement ratios to zero percent effective March 26, 2020.  This action eliminated reserve requirements for all depository institutions.

 

Source: https://www.federalreserve.gov/monetarypolicy/reservereq.htm

[wabuffo]

This action eliminated reserve requirements for all depository institutions.

 

This is related to REQUIRED RESERVES and was already essentially obsolete as a concept.  TOTAL RESERVES today are made up of REQUIRED RESERVES + EXCESS RESERVES.  Required reserves stopped being important after the GFC and the expansion of the Fed's balance sheet.  Look at this table - nothing's really changed just because for accounting purposes required reserves are marked at zero -- the numbers just shift over to the excess reserves column now. 

 

https://www.federalreserve.gov/releases/h3/current/default.htm

 

It sounds counter-intuitive, but reserves are a function of the Fed deciding how big a balance sheet it wants to carry.  Banks are just along for the ride as passengers collectively.  They have no choice. 

 

Don't believe me.  Let's hear from a former Fed governor (ex-NY Fed President Bill Dudley):

https://www.bloomberg.com/opinion/articles/2020-01-29/fed-s-repo-response-isn-t-fueling-the-stock-market

 

when the Fed buys T-bills and increases the amount of reserves in the banking system, that liquidity can’t go elsewhere. It can move from bank to bank as households and businesses shift where they hold their bank balances. The only exception is if bank customers decide to increase their holdings of currency. But if they do that, that reduces the amount of excess reserves in the banking system.

 

The Fed’s T-bill purchases substitute a bank reserve (essentially equivalent to a one-day T-bill) for a slightly longer risk-free asset (a T-bill) that the Fed now holds in its portfolio. But that’s it. There are no funds created to purchase equities.

 

Dudley was talking to the issue that many market participants were blaming the Fed restarting its repo program in late Sept. 2019 as causing "liquidity to flood into the stock market" - but focus on the mechanics he's describing. 

1) The Fed increases reserves by buying stuff (strictly a swap of assets with a bank)

2) Those reserves go nowhere and stay in the Fed’s clearing accounts for the banks.

 

So why did the Fed cut the required reserves to zero if banks can't actually change the total reserve levels held by the commercial banking system?  I believe there are calculations embedded within that reserve requirement that are affected by the types of assets banks hold on their balance sheet.  By setting it to zero, it was a very quick-and-dirty deregulation act for banks to free them up to change their asset mix.  Perhaps, the Fed hoped banks could more easily alter their asset mix towards riskier assets and help them better respond to the economic hardships businesses were facing because of the pandemic-related shutdowns.  I don't think it did much.

 

wabuffo

 

 

[LearningMachine]

It sounds counter-intuitive, but reserves are a function of the Fed deciding how big a balance sheet it wants to carry.  Banks are just along for the ride as passengers collectively.  They have no choice. 

 

Wabuffo, I'd like to understand deeper what you're saying regarding these two events.

 

#1. Fed buys Treasures, T-Bills and Mortgage Backed Securities in the amount of $X.

#2. Banks collectively put $X of their customer deposits (which are part of M2 money supply) into Fed Reserves

 

Are you saying that #1 effectively forces #2 to happen because banks don't have anywhere else safe to put their customer deposits?

[wabuffo]

Are you saying that #1 effectively forces #2 to happen because banks don't have anywhere else safe to put their customer deposits?

 

No - I am saying the two are linked in a payment flow.  The Fed's main job is to manage US large-scale payments clearing.  They do this by using reserve accounts.  All nationally-chartered banks have a "chequing account" at the Fed.  So when the Fed decides it wants to expand its balance sheet by buying a Treasury bond from JP Morgan Chase, this is the payment flow that happens.

 

 

This is very simplified and very high-level.  The tables show what happens to the assets and liabilities of the Fed and the assets and liabilities of JP Morgan Chase when the Fed buys a bond from JPM.  In the first table, JPM owns only $10m of US Treasury Bonds so its total assets are also $10m.  (This is obviously not what the entire balance sheet of JPM looks like).  The Fed at this point has no assets and no liabilities.  (again - not what the Fed balance sheet looks like).

 

Then the Fed buys $10m of US Treasury bonds from JPM.  The Fed makes an electronic accounting entry in JPM's reserve account at the Fed for $10m and takes possession of the bond.  Now the balance sheets have changed.  The Fed expanded its balance sheet and now owns $10m of Treasury Bonds but also has a liability of $10m via JPM's reserve account at the Fed.  JPM's balance sheet hasn't changed in total assets.  It still has the same $10m of total assets.  The difference is that instead of US Treasury bonds, its assets are on deposit at the Fed with an account balance of $10m. 

 

The issue is that these reserve accounts at the Fed can only be used to clear payments between JPM and other federally-chartered banks or between JPM and the US Treasury (ie, tax payments going to the US Treasury, spending from the US Treasury going to JPM).  See!  its a closed-loop system.  In aggregate the reserve balances can shift between banks but for the entire banking sector the total reserves at the Fed can't change unless/until the Fed changes them by selling assets back to the banks.

 

LearningMachine - I hope that makes sense.  You can't think of reserves as something banks can "withdraw" from the Fed, in aggregate they can't.  It is Fed decisions to buy assets (and expand its balance sheet) or sell assets (and reduce its balance sheet) that determine how much reserves that banks have stuck in these accounts.

 

If you want a slightly more complicated flow involving the US Treasury and MetaBank (CASH) and the CARES Act EIP debit card program for stimulus payments to the "unbanked" - I went over that payment flow last summer (a real world example).  Its shows how MetaBank's reserve account exploded when the US Treasury issued the order to the Fed to transfer money from the US Treasury's reserve account at the Fed to Metabank's.

https://www.cornerofberkshireandfairfax.ca/forum/general-discussion/how-can-the-fed-unlimited-qe-be-deflationary/msg425263/#msg425263

 

wabuffo

[Cigarbutt]

^wabuffo's answer is excellent. There's a visual complement:

In my limited understanding, velocity has been going down for a while and M1 (and M2) don't include reserves (required and excess).

[LearningMachine]

Thanks wabuffo for the very clear answer.  It helps me understand your perspective better. 

 

For this perspective to hold, all of the following have to be true:

* #1. JPM has to be required to sell its treasury bonds to the Fed

* #2. JPM's reserve account with the Fed can only be used to clear payments between JPM and other federally-chartered banks or between JPM and the US Treasury

* #3. JPM is not allowed to withdraw from its reserve fund at the Fed

* #4. Even when the asset bought by Fed from JPM matures, JPM is not allowed to withdraw from its reserve fund at the Fed

 

I can understand banks can be induced to do #1 if Fed is paying the highest price, but it is not required right?

 

Regarding #2, #3 and #4, any chance you would have a link to the actual regulation that has language stating each of these?

[LearningMachine]

^wabuffo's answer is excellent. There's a visual complement:

In my limited understanding, velocity has been going down for a while and M1 (and M2) don't include reserves (required and excess).

 

Thanks Cigarbutt for the visual complement :-).

 

Federal Reserve Bank of New York agrees with you at https://www.newyorkfed.org/aboutthefed/fedpoint/fed49.html:

The narrowest measure, M1, is restricted to the most liquid forms of money; it consists of currency in the hands of the public; travelers checks; demand deposits, and other deposits against which checks can be written. M2 includes M1, plus savings accounts, time deposits of under $100,000, and balances in retail money market mutual funds.

 

I think what wabuffo seems to be saying is that part of the deposit accounts at banks are somehow required to be held at the Federal Reserves such that customers can't access their deposits??  Trying to understand wabuffo's point deeper and get authoritative sources with questions in post above.

[bizaro86]

I think the consensus appears to be that the Fed expanding its balance sheet (aka printing money) doesn't cause inflation. That doesn't seem intuitive to me, but Wabuffo's points seem logical and well thought out. It also matches with recent history from an empirical stand point. As there has been significant QE but no significant inflation.

 

I'd like to invert this with a question (selfishly to aid my own understanding). What would cause inflation? We know inflation is possible because it has happened in the past.

 

Would it require the velocity of money to go up? What if the fed stopped balancing its assets and liabilities and just started writing cheques to the treasury with no offset. I believe that would effectively expand the money supply?

[Sunrider]

Think of it like this

 

You magically get the Fed’s money printing press (well computer these days) and print yourself 10tn real dollars, i.e. there is no dispute that they are real valid can be spent etc.

 

Does that cause inflation? No, not as usually defined. It will only drive up the price level if

A - you start spending them on stuff (and ideally faster than stuff can be supplied) or

B - People think you are going to spend them and start raising / paying higher prices in anticipation.

 

Same thing here, just putting newly created money into fed reserves for the banks doesn’t create inflation as long as people don’t adjust their expectations or the money actually does go start cycling in the economy. What does that mean? The banks would have to convert those reserves into loans to customers, who then spend the money. So far it seems they only have loaned to financial investors and thus financial assets have inflated... and the public has not (yet?) begun to believe that prices might rise.

 

(There’s a technical point in the above in that you can argue that the creation of these new dollars alone makes all existing dollars worth less in terms of real goods ... a bit like it used to be when the dollar was pegged to gold - then it was more obvious as the quantity of money was (nearly) fixed. ... it just seems that people generally don’t think about it as an exchange of two things anymore and so have not adjusted their price expectations).

 

What Wabuffo and Hunt are also saying, I think, is that if the fed liabilities become legal tender, the. All bets are off, because at the moment the money being printed basically just turns extant debt from one form (government treasuries) into another (fed deposits). That shortens the maturity and should push banks to seek higher yields on these assets.

[JRM]

What Wabuffo and Hunt are also saying, I think, is that if the fed liabilities become legal tender, the. All bets are off, because at the moment the money being printed basically just turns extant debt from one form (government treasuries) into another (fed deposits). That shortens the maturity and should push banks to seek higher yields on these assets.

 

That's the key.  Up to this point quantitative easing has been bolstering bank reserves.  The banks are still the gate keeper for lending the money out.  When money is lent out it increases the M1 money supply, but is also deflationary in the sense that it must be paid back.  Helicopter money, like PPP loans that don't need to be repaid and stimulus checks, should kickstart CPI inflation. 

[mattee2264]

 

Yeah I think large budget deficits accompanied by a recovery of private sector spending could do the trick with higher commodity prices and supply constraints adding some fuel to the fire.