Recent Posts

Tuesday, August 11, 2020

Never Reason From A Change In Monetary Agggregates

I have been looking at questions on the Economics Stack Exchange, and many of them followed a depressing pattern: where is the inflation that results from the change in monetary aggregates? This is a common concern that has been popping up since the end of the Financial Crisis (even earlier in the context of Japan). The prescription is straightforward: one should never reason from a change in a monetary aggregate. The reason behind this is that monetary instruments are financial assets, and the size of the stock held depends upon the portfolio preferences of the private sector, which should be expected to be unstable.

This helps confirm that the message from Abolish Money (From Economics)! has not caught hold yet in the public imagination.

Chart: M1 growth, nominal GDP growth, inflation.

The chart above summarises the pitiful nature of money growth as an indicator in recent decades. We see massive accelerations and the occasional deceleration, while nominal GDP and (core) inflation chugged along at sedate paces (with a bit of excitement in 2020, but nothing like the jump in the money supply).

From a post-Keynesian perspective, this is exactly what one should expect.

The first thing to note is the belief that one needs a stock of money to transact is obviously not in touch with modern reality. I spent most of my career as a financial analyst with a maximum of about $400 in narrow money holdings at the end of the day: all my inflows were immediately swept into a money market fund or trading account, and I used my credit card for any expenses that went beyond my $400 bi-weekly "fun money" budget. Meanwhile, I was a home and car owner, etc. My cash aversion may have been extreme, but it demonstrates that end-of-day narrow money balances are not required to function in a modern capitalist system.

Instead, monetary instruments are components of portfolios, and the stock of those balances depends upon sentiment about portfolio construction. The Keynesian argument is that money balances are held to hedge against fundamental uncertainty: they can always be liquidated near par in order to meet cash outflows.

(This leads into my personal example: what about my broad money holdings, e.g., including money market funds? Those end-of-day holdings would depend upon my animal spirits. For example, if I were stupidly bearish about equities -- which only went up during that period -- my money market fund holdings would jump higher.)

In the developed countries, the private sector has a large stock of financial assets, and there is considerable financial innovation. This allows for extreme rapidity in the restructuring of portfolios, hence monetary aggregate growth should be jumpy. Having central bankers mucking around with their balance sheets adds to this instability. As such, the portfolio weighting is moving much faster than nominal incomes -- explaining the figure above.

Unless regulators put the shadow banking system out of business -- a big ask -- no amount of playing with the definitions of monetary aggregates will allow them to offer useful leading information. Money numbers reflect outcomes of economic developments, it is a mistake to believe monetary aggregates will cause changes in the economy.

(c) Brian Romanchuk 2020


  1. So you are saying Egmont Kakarot Handtke was right this whole time.

    1. Yes, he deserves the “Nobel” prize for his amazing work.

  2. re: "The chart above summarises the pitiful nature of money growth as an indicator in recent decades"

    Some people don’t know money from mud pie, M2 from liquid assets (cash equivalents or near money substitutes), etc.

    Sweep accounts exerted their maximum impact in the 1990's. Sweeps have become so unimportant that the BOG ceased publishing the data on, May 2, 2012. The low rates of the GFC ended many swaptions, retail and institutional sweep accounts.

    And Brian Romanchuk’s example is like saying that if one person can do it, all people can do it. Not so, viz., the money paradox.

    Alfred Marshall's cash balances ("Money, Debt and Economic Activity" (2nd ed.; New York: Prentice-Hall 1953), p. 197 His "Money Paradox"

    "If the public considers its real balances excessive or deficient, forces will be set in motion which will alter the value of the cash holdings of the public, but not necessarily in the fashion desired by the public. For example, if the public on balance considers the real worth of its cash balances deficient, this will bring about an increase in the demand for money and a decrease in its supply.

    I.e., paradoxically, it destroys money velocity.

    It is the same argument presented by Princeton Professor Dr. Lester V. Chandler, Ph.D., Economics Yale.

    Chandler’s 1961 theoretical explanation was:

    “that monetary policy has as an objective a certain level of spending for N-gDp and that a growth in interest-bearing deposits in the payment’s system involves a decrease in the demand for money balances, and that this shift will be reflected in an offsetting increase in the velocity of the remaining transaction’s deposits”.

    The saturation of DD Vt according to Professor Dr. Marshall D. Ketchum, Ph.D. Chicago, Economics:

    "It seems to be quite obvious that over time the “demand for money” cannot continue to shift to the left as people buildup their savings deposits; if it did, the time would come when there would be no demand for money at all”.

    Thus, and therefore money velocity, the S-Curve” dynamic damage (sigmoid function) plateaued by the first half of 1981..

    Milton Friedman, in the Journal of Political Economy: “The Lag in Effect of Monetary Policy”
    Vol. 69, No. 5 (Oct., 1961), pp. 447-466 said:

    “This would mean that effective monetary action requires an ability to forecast a year ahead, not an easy requirement in the present state of our knowledge.”

    Which lead Friedman to conclude:

    “The central empirical finding in dispute is my conclusion that monetary actions affect economic conditions only after a lag that is both long and variable”

    Not so. I cracked the code in July 1979. The distributed lag effect of money flows have been mathematical constants for over 100 years.

    Monetary Flows { M*Vt } 1921-1996

    And we knew this already:

    In 1931 a commission was established on Member Bank Reserve Requirements. The commission completed their recommendations after a 7 year inquiry on Feb. 5, 1938. The study was entitled "Member Bank Reserve Requirements -- Analysis of Committee Proposal" its 2nd proposal: "Requirements against debits to deposits"

    After a 45 year hiatus, this research paper was "declassified" on March 23, 1983. By the time this paper was "declassified", Nobel Laureate Dr. Milton Friedman had declared RRs to be a "tax" [sic].

    However, as of March 26, 2020, we can now expect the FED to lose control of the money stock.

    ---Michel de Nostredame

    1. I cannot respond to your point because I cannot tell what it is, and since I can see no logical structure to your statements, it is too painful to try to infer what you are trying to express.

      If you want to be taken seriously, you need to think about the structure of what you are writing, and use the minimum number of words to get there.

  3. What does it tell you when the sum of all financial assets (not just "monetary" assets) is very high vs trend GDP? (as compared to the historical ratio).

    1. Given that the “duration” of aggregated financial assets is fairly long, that tells us that discount rates have fallen, and/or the population is wealthier.

    2. This gets confusing: wealth is a stream of cash flows divided by a discount rate. Presumably a falling discount rate signals lower growth expectations and thus lower cash flow growth. Shouldn't these have a somewhat offsetting effect?

    3. Nominal growth rates are not much lower in recent years than in the early 1990s, but discount rates are much lower. Look at the chart of the 10-year Treasury yield since the early 1980s, and one would expect higher valuations.

    4. So there is little (no?) link between long-term bond yields and nominal growth expectations?

    5. There’s short-term correlations. I certainly have written some variation of the statement that the main driver of bond yields is growth expectations. But that is a cyclical story. If we look at levels, the spread between them varies considerably over time. That explains the drift in wealth/income ratios.

    6. This presents an entirely different dilemma. There is a natural incentive to add (lots) of leverage when rates are below NGDP expectations (we've seen this in practice). This then limits the ability for rates to rise in the future without sparking some sort of crisis. Perhaps this is why we seem to be stuck in a boom-bust pattern.


Note: Posts are manually moderated, with a varying delay. Some disappear.

The comment section here is largely dead. My Substack or Twitter are better places to have a conversation.

Given that this is largely a backup way to reach me, I am going to reject posts that annoy me. Please post lengthy essays elsewhere.