Summa: The Great Myth

Listening to an interview of Richard Duncan by David McAlvany we were finally motivated to explore an issue that has nagged at us for quite a while. What disturbs us is that we find ourselves alone in disagreement with the prevalent ‘wisdom’ regarding the Fed’s quantitative easing (QE) policies and their effects on liquidity and markets. This essay will explore our position.

Richard Duncan: A Mainstream View

We begin by extracting some of the points from the Duncan interview, discussing the fallacies either implicit or explicit in them. We do this because his views are generally quite mainstream. The points are (approximate time into the interview in brackets):

  1. (4;40) He states that in 2013, the Fed will have created and pumped about $1 trillion (T) into the economy in the form of fiat money that he calls liquidity (a meaning consistent with our use of the term). This is based on the $85 billion (B) per month of the current QE program. This is the first widely held view and will be the centerpiece of our discussion to follow and which we will show is patently false.
  2. (~13:40) He says that the government deficit borrowing “sucks liquidity” out of the system. We cannot understand how he can have such an egregious misunderstanding of money flow in the economy. Because the government is an actor in the economy, every dollar it borrows it spends. The cash or currency or money with zero maturity (MZMc as we will refer to it hereafter) received from Treasury instrument buyers is distributed as salaries to government employees, used for arms purchases from the defense industry, and distributed through widespread handouts as benefits in entitlement programs, etc. The liquidity in the economy is unchanged by any action of any level of government. The productive return on such liquidity redistribution is another matter that is not germane to our case.
  3. (~15:00) Moving on to foreign trade, an area where he is usually strong, he suggests that the US current account deficit is the second major source of increase of liquidity in the US economy after QE. We have discussed the binary nature of economic transactions before (The Banks Are Not Lending …., also on China’s disposition of Treasury bonds). Point two above represents a failure by Duncan to consider the binary nature of government financing. This point is likewise a similar failure. The $400 B current account deficit forecast for this year he describes as $400 B of liquidity coming into the US. But that $400 B is simply the net outflow of investment and trade dollars from the US into the global market – a $400 B liquidity outflow of USDs. The flow of liquidity of USDs nets to zero! As an observation, he describes this inflow of USDs to purchase Treasury debt as $400 B of liquidity flowing into the economy, contradicting his point in two above that such represents a drain of $400 B from the economy.
  4. (~16:40) At this point he adds the liquidity figures from points 1 and 2 above getting $1.4 T of “liquidity creation”. He subtracts the budget deficit for the year of $700 B to get roughly $700 B of excess liquidity going into other assets such as bonds, stocks and property. He specifically ascribes the rise in all asset prices to the increased liquidity from these two sources. This represents the second generally held point of view. We shall deal with this point below also.

We stopped listening to the interview around the 18:00 mark, not needing any more.

And Others …

A sampling of recent comments on the market “liquidity” from QE follows (emphasis ours). Grant Williams in his Dec. 30 edition of Things That Make You Go Hmmm… writes Consistent amounts of “liquidity” are pumped into the system every month, and things gently float ever higher.

Even Charles Biderman of Trim Tabs and one sharp pencil, said in his Daily Edge video of 2/9/2012:  The Fed has been flooding the economy with 5 trillion dollars of virtually no-cost money since 2008. In a 1/5/2013 article titled Is Fed Policy Responsible for the Rally?, he is quoted as saying: Although I am not suggesting that this is the ONLY reason, it’s worth considering that the main reason for this big move up may have been the Fed’s liquidity. Similarly he quotes one of his staff in a 1/30/2013 article in Forbes – How The Fed Is Helping To Rig The Stock Market – as saying The Fed is exchanging about $4 billion in newly created money every business day for various types of bonds.  All else being equal, the Fed’s bond buying puts more money in investors’ hands to buy other assets, including stocks.

We were handed another example today (20131231) from Zero Hedge in the post: Correcting Some Misconceptions About A New Secular Bull Market. From author Lance Robert’s concluding remarks we read: the current momentum driven markets, fueled by ongoing Federal Reserve interventions … We assume the ‘fueling’ is referring to Fed liquidity.

Finally we note that who has been a staunch proponent of the liquidity mantra said today in Great dollar rally of 2014 as Fukuyama’s History returns in tooth and claw: As the Fed turns off the spigot of dollar liquidity, it will starve the world’s dysfunctional economy of $1 trillion a year of stimulus.

Also read: One Bear at a Time.

One has to understand the nature of Fed transaction – exactly who the Fed deals with – and the composition of its balance sheet in order to see that these commonly held views in emphasis are wrong! And that is what we propose to show.

The Fed and the Economy

We argue that the Fed is outside the economy. It participates in the economy when it pays its employees and buys toilette paper for its washrooms. But the vastly larger activities of its monetary policies we will argue as being outside the economy. All other entities are economic actors (actors for short) whether they be consumers, corporations, charitable institutions or governments. This is not to say that all of these actors make net positive contributions to the economy. But whether they are productive in the sense of creating wealth or whether they are net destroyers of wealth, they are all part of economic activity which reduces to transfers of MZMc or its electronic equivalent.

All MZMc at any time resides with some actor. Every actor at any time has a fraction of MZMc whether it be change in a pocket or cash in a bank vault. Most actors prefer or are required to maintain a certain cash reserve but the cash that an actor has available beyond this reserve is what he is willing to spend or lend (lending is spending), and he will do so at the appropriate opportunity. This is the economy.

All financial institutions are such actors with the exception of the Fed. They want to play in the economy to make money. The Fed has no such mandate. So MZMc outside the Fed is in the economy and MZM inside the Fed is not. To sharpen this point we distinguish between MZMc deposited with the Fed by depository institutions and unissued currency in Fed vaults which therefore does not appear on the Fed’s balance sheet as a liability which is the case for MZMc.

The cash on deposit with the Fed is mostly in a category called excess reserves. It is important to understand that these deposits are both part of the economy but not in the economy. It is money on the sidelines. It sits as a digital record in the Fed’s computers. A depository institution may withdraw its deposit, presumably at any time. The Fed simply and can only issue the equivalent cash from its vaults. It reduces the institution’s excess reserves by the amount withdrawn and increases the amount of money in circulation by the same amount. This means the total liabilities on its balance sheet are unchanged. At that point we would say this new MZMc has entered the economy and is now part of the practical liquidity in the economy.

The Fed’s balance sheet is what provides the clarity. If a primary dealer (PD) goes to its appropriate Federal Reserve bank and deposits $1 B in cash, this $1 B which appears on the Fed’s balance sheet under liabilities as “Federal Reserve notes” is cancelled and a digital credit appears in that PD’s “Other deposits” line under balance sheet liabilities. The money is available to the PD but it is not available to the economy unless the PD takes it out of the Fed.

But where did the excess reserves come from? The Fed transacts monetary policy through the PDs. If the Fed wishes to affect the liquidity in the economy it can increase it by buying assets – principally Treasuries and Mortgage-Backed Securities – from PDs with new currency (this assumes the PD needs the liquidity to say increase its loan portfolio). To decrease liquidity, it sells the same assets that are on its balance sheet for MZMc and ‘retires’ this currency which  removes it from the economy in the manner of a deposit as described in the last paragraph.

Dissecting the Fed’s Balance Sheet

Up to the point that the Fed began its Large Scale Asset Purchases (LSAP) or QE, the major component of the Fed’s balance sheet (Table 8 of the H.4.1 statistical release) was currency in circulation. In Figure 1, we plot total assets on the Fed’s balance sheet against currency in circulation. At the start of the period for which the data is available, Jan. 2003, the assets are about $45 billion or 6% more than the currency so we add this as a scaling factor to the currency data to create an overlay of the two curves.

Figure 1. Currency in circulation (blue line; billions of dollars) and total Fed balance sheet assets (red line; transformed into billions of dollars) (click to enlarge).

Next, in Figure 2, we look at the total of excess reserves and currency in circulation compared to the total size of the balance sheet. To get this graph we add excess reserves to currency and remove the 45 scaling factor.

Figure 2. Currency in circulation plus Excess reserves (blue line; billions of dollars), total Fed balance sheet assets (red line; transformed into billions of dollars), and difference between the two lines in green (click to enlarge).

Using the data inspection feature of FRED (beta version not shown here) allows us to see that at Nov. 1, 2013, the two curves are about $230 billion apart. This difference is accounted for in accumulated capital and other deposits, particularly the Treasury’s account which fluctuates considerably over time as the government conducts its operations. We would also note that until the arrival of LSAP, most of the Fed’s balance sheet on the liability side consisted of MZMc and excess reserves which if examined are seen to be close to zero. The failure of Lehman Bros. and the resultant market lockup necessitated new and unorthodox interventions including the assumption of toxic assets in the three Maiden Lane programs (we always thought that Maiden Lane was the secluded place where maiden’s were led to be deflowered – most appropriate) and TALF and for which there remain a residue of assets to be seen on the Fed’s balance sheet.

In any case, from this time onward a noticeable perturbation was introduced into the Fed’s balance sheet but the important point is that it is holding steady around $300 B. This is not affecting liquidity except at the margin if at all.

Finally, to clinch our argument, in Figure 3 we take the total amount of securities held outright by the Fed including Treasuries, GSE instruments and Mortgage-backed securities and subtract excess reserves and MZMc. The difference rounds to zero ($15 B of a $4 T balance sheet)!

Figure 3. Total securities held outright – excess reserves – currency in circulation (in billions of dollars) (click to enlarge).

This proves that the effect of LSAP, whatever it is, is contained inside of excess reserves and MZMc. The former we have argued is not liquidity and not active in the economy. We will show in the next section that QE has had no appreciable effect if any on currency in circulation.

What’s Up With Cash?

We have shown that almost all the liquidity created by QE remains inside the Fed as excess reserve deposits. It is not active in the economy!!!! To leave the Fed and enter the economy it must do so as MZMc. So the key issue then is has currency in circulation grown abnormally? We can try and do this in a couple of ways. The ideal state of the currency supply is that in a growing economy, the base currency is expanding proportionately so that the liquidity in the economy stays at the same functional (percentage) level. Figure 4. shows GDP in red, MZMc in green and the ratio of MZMc to GDP in blue.

Figure 4. Currency in circulation (green; left scale),  GDP (red; left scale),  and currency/GDP (blue; right scale) (click to enlarge).

From this  graph we see that since 1950 GDP has been growing faster than currency but at a decelerating rate.  Around 1990 this relationship reversed and the rate of currency growth over GDP turned positive (slope of the blue curve). A notable downturn in the rate began in 2003 before it was sharply reversed bu QE1 in Sept. 2008. One can argue that the rate of growth of currency to GDP has been linear since 1990 and QE simply restored the rate to that linearity. It’s really to soon to tell for sure.

What we can conclude is that with the current ratio at the 1955 level, the relationship between currency and GDP does not appear to be abnormal. The effect of QE on the ratio would seem to simply have corrected an anomaly that began in 2003. As a comment, GDP data is reported quarterly and the currency data used is monthly. This creates some ‘noise’ in the ratio.

The curve of currency in Figure 3 appears to be exponential in nature as does that for GDP. An easy way to test this and reveal anomalies is to plot currency on a log scale. This we do in Figure 5.

Figure 5. Currency in circulation on a vertical log scale (click to enlarge).

Using weekly data going back to 1984, the linearity of the curve in this graph tells us that currency growth is a smooth exponential function and that QE has created no major anomaly other than a reversion to mean in the 2009 recession.

It’s All Binary Baby

A theme we harp on is the binary nature of transactions in the economy. Let’s see how QE really works. A PD buys an asset, say a 10-year note at a Treasury auction. In exchange for the note, the PD gives the Treasury an equivalent amount of MZMc. The liquidity in the economy moves around but doesn’t change in quantity (note to Richard Duncan). The PD then sells the note to the Fed for an equivalent credit of MZM in its account with the Fed. Liquidity has been taken out of the economy in the sense that it is sidelined inside the Fed. This is not the same as the Fed selling that same note to a PD and removing the equivalent amount of MZMc from the economy. In this case it is truly gone. One could also say that if the PD had kept the money in the first place it would be sidelined inside the PD.

We looked at why the PD might not have any use for this money (read: The Banks Are Not Lending ….). It can sit in the PD’s vault earning nothing, be transferred to the Treasury for a note earning something, or be deposited with the Fed and earn a fraction of a percent which we suppose on a couple of trillion dollars amounts to something also. We note that had the PD bought the Treasury note and held it but suddenly needed liquidity, it could simple sell the note in the market for an amount of MZMc equivalent to the note’s market value. Liquidity moves around again but doesn’t change in total.

We have seen suggestions that QE has an impact on collateral used by institutions in the economy. Treasuries are recognized as the highest quality of collateral. True MZMc has a higher quality because it has no market risk while a Treasury has a miniscule but finite risk of default. But you would not use MZMc as collateral for a loan of an equivalent amount of MZMc (partly because MZMc is fungible). So selling a Treasury to the Fed does not increase the collateral you have to work with and does not affect the amount of collateral you need (otherwise you would have kept it).

So What Is QE’s Effect on the Economy and Stocks in Particular?

We have argued that the Fed has two tools. It can control short-term interest rates which until ZIRP, could be used to affect economic activity by impacting borrowing decisions of actors. When they reached the zero bound (see: Flash Point: The Fed is Dead!, Flash Point: What We Missed the First Time Around) they had to move out the yield curve via QE  programs. As the Fed  acquires a significant portion of the various securities markets, the effectiveness of QE seems to be decreasing and markets may be beginning to re-assume their control over interest rates.

Apart from affecting interest rates the only other tool the Fed actually has is control over investor psychology. We have shown that there has been no inflow of new liquidity into markets and particularly stocks. A comment we recently read (and unfortunately did not save) was to the effect that markets were no longer being driven by fundamentals but by investor psychology. The source noted that if there’s bad news the market goes up believing that the Fed will have to continue the illusory insertion of liquidity into the market. If there’s good news then positive investor sentiment takes over and moves the market up on fundamentals.

We suggest that near record amounts of margin ($424 B in December) must play a role. Flow of funds out of other assets must play a role and returning retail investors backed by savings and debt must play a role. And the positive psychology of the mythical Fed liquidity sustains the Ponzi drive.

A Note on Repos and Reverse Repos

As we were finishing this another comment appeared on Zero Hedge in the post: WTF Chart Of The Day: Fed Soaks Up Record $200 Billion In Year End Excess Liquidity. The topic was reverse repos (RRs), something we studied years ago. While people get excited over the size of the operations, they neglect to note that so far these are short duration – as short as overnight. Hence they appear as temporary open market operations (TOMO) and are transitory on the Fed’s balance sheet if they make it at all into the weekly H.4.1 statistical release. We note that the term of operations of the last 20 reverse repos (RRs) has varied from 1 to 3 days and the amounts have grown steadily larger, from as low as $3 B to the last at $197 B. This smacks of Fed experimentation – “let’s throw this on the fire and see what happens.” From the Zero Hedge article:

And yes, nobody actually ever had to sell anything to hand over the fungible electronic cash equivalents to the Fed because… the magic of repo and shadow banking rehypothecation of claims. Remember: $2.5 trillion in excess deposits serve as dry powder to chase risk higher purely in the form of initial margin on marginable securities like the E-Mini, and no money every actually changes hands.

First of all, we suspect that the prop desks of the PDs do not have ready access to the excess reserve accounts. Even if they did, as we have explained the withdrawal of deposits for margin purposes can only happen through the actual “printing” of currency and this as we have shown is not happening to date – at least not in any size we can measure.

As for the effect of an RR on the Fed’s balance sheet, the PD transfers funds – electronically is fine – and receives the corresponding Treasury purchased to its account, again a virtual transaction. The Fed decreases its assets by the amount of the Treasury and decreases its liabilities by subtracting the currency equivalent from the PD’s reserve account. Since this is a temporary  operation, no physical property need ever change hands.

In short, the putative use of reserves as the Zero Hedge author proposes is impractical and certainly not through a reverse repo, and is simply not happening.

Also note that repos and RRs are initiated by the Fed. A PD simply can’t ring up the New York Fed Trading Desk and say give me $x through an RR. It is done daily at a specified time as an auction by the New York Fed for implementing short-term monetary policy.

Putting It Together

QE has injected no significant liquidity if any into the economy. Check!

Postscript 20140106

Today, John Hussman wrote in his weekly letter titled Confidence Abounds:

I continue to believe that quantitative easing has no mechanistic relationship to the economy or the financial markets beyond creating a purely psychological discomfort with zero interest rates, and encouraging a reach for yield in speculative assets that has already set reckless extremes in median stock valuations, margin debt, and “covenant-lite” lending.

Then again on 20140120, John wrote in Superstition Ain’t the Way:

We are increasingly moving away from a “fractional reserve” banking system, as the relationship between reserve creation and lending has collapsed. Idle bank reserves now exceed the amount of demand deposits in the U.S. banking system by more than two-to-one, and exceed 25% of all deposits in the U.S. banking system. Keep in mind that these reserves don’t “go into” the stock market (every buyer of stocks is matched by a seller who gets the cash). Rather, these reserves may change owners, but stay in the banking system [inside the FED – our comment] in aggregate, depressing short-term interest rates, and resulting in a pool of zero-interest deposits that change hands from one uncomfortable holder to another.

Back in December 2011 in the essay Understanding Money: Part 3 – Measures of the Money Supply, in the section titled Leverage and Liquidity In the Money Supply, we displayed a graph that showed the leverage in the various measures of the money supply. The leverage in M1, the measure of demand deposits has dropped 50% since 1959 whereas M2 and MZM remain about the same. This supports Hussman’s opening remark. The remainder of his comments confirm our thesis above.

Thank you John. We have great respect for your work and rest in the comfort of confirmation put so eloquently and simply.

Powered by WordPress | Designed by: photography charlottesville va | Thanks to ppc software, penny auction and larry goins