Why monetarism matters

Monetarism can be viewed as a way of explaining changes in macroeconomic aggregates by looking at shifts in the supply and demand for money.  Monetarists like to sharply distinguish between money (the medium of account) and credit (loans of the medium of account.)

Tyler Cowen recently linked to an article by Haelim Park and Patrick van Horn, which analyzes the role of the 1936-37 increase in reserve requirements.  Here’s the abstract:

We analyze the impact of contractionary monetary policy through increases in reserve requirements on bank lending. We compare the lending behavior of banks that were subject to the requirement increases in 1936-37, Federal Reserve member banks, to a group of banks that were not subject to the reserve increase, Federal Reserve nonmember banks. After implementing the difference-in-difference estimators, we find that the increases in reserve requirements did not create financing constraints for member banks and lead them to reduce lending. Therefore, the actions of the Federal Reserve concerning the required reserve ratios cannot be blamed for instigating the economic downturn of 1937-38.

Let me first emphasize that I don’t have any problem with the overall article, which I haven’t even read.  Elsewhere I’ve argued that the reserve requirement increases probably did not play a major role in the 1937-38 recession.  But I’d like to quibble with the word “Therefore”, used in the final sentence of the abstract.  The fact that the reserve requirement increases did not reduce lending in directly affected banks by more than in other banks does not imply that the reserve requirement increases were not contractionary.

Let’s review why reserve requirements matter.  A higher reserve requirement may lead to more demand for bank reserves, and hence more demand for base money. An increase in [demand for] base money is, ceteris paribus, a contractionary monetary policy. It reduces prices and NGDP.  Of course ceteris is often not paribus, and in many cases monetary policy actions that look contractionary do not have contractionary effects, perhaps because they are viewed as temporary, as was most likely the case in 1936-37.  Or because banks held large quantities of excess reserves, and hence an increase in the reserve requirement did not increase the total demand for reserves. Or perhaps because it was expected to be offset by some other monetary action, such as a change in the base.

Even with all of the preceding caveats, I think it’s quite possible that the reserve requirement increases were at least slightly contractionary in effect.  But even if they were contractionary, I’d expect roughly the same impact on bank lending for those banks that face reserve requirements, as for those that do not.  To see why, imagine that reserve requirements applied to retailers, not banks.  The Fed required retailers to hold cash balances equal to at least 10 days sales.  Now the Fed suddenly raised the required ratio to 20 days sales.  What would happen to bank lending?

To simplify the exercise let’s assume that the Fed doesn’t change the monetary base.  The retailers must scramble to get the extra cash reserves from the fed funds market.  That causes the fed funds rate to rise.  Banks pass on the higher wholesale cost of credit with higher interest rates on loans.  So bank lending declines, even though (by assumption) the reserve requirement doesn’t apply to any banks, only retailers.

Here’s another channel, more in line with monetarist thinking.  The increase in the retailers’ demand for cash reserves has the effect of reducing base velocity. Because (by assumption) the base is fixed, this reduces NGDP (and perhaps RGDP.) As banks see NGDP decline, they conclude that there are fewer good loan prospects, and hence curtail lending.

Note that in both cases, the interest rate channel and the base money demand channel, the effect of higher reserve requirements on retailers is to indirectly reduce bank lending, even though the higher reserve requirements for retailers have no direct effect on banks.  

The endogenous money crowd often argues that reserves don’t constrain bank lending. At the level of individual banks they are correct, although they often draw incorrect conclusions from this observation, especially at the macroeconomic level, where the overall money supply may not be endogenous.  For instance, in 1936-37 the base was partly constrained by the size of the monetary gold stock. Nonetheless, it’s true that if a bank sees a good loan opportunity, it can make the loan and worry about getting reserves later, perhaps by selling off bonds, or perhaps by borrowing in the fed funds market.

The older monetarists of the 1950s-1980s may have erred in arguing that we should target M2 growth at a steady rate.  But let’s not throw the baby out with the bathwater.  The older monetarists had important insights on the need to focus on money, not credit.  Monetary policy is about shifts in the supply and demand for base money, it’s not about bank lending. By ignoring these insights the profession badly misjudged the nature of the Global Financial Crisis, which was fundamentally due to a failure of monetary policy, not banking policy.

PS.  I just returned from 10 days in Costa Rica—did I miss anything important? Last night the Miami airport was an insane madhouse, with the entire TSA/Customs/Passport control regime reaching new heights of incompetence.  Four long lines and over two hours to change places, even with us literally running between torture chambers.  Every day I get a bit more libertarian, as I get mugged by reality.

PPS.  The original post omitted the phrase “demand for”



27 Responses to “Why monetarism matters”

  1. Gravatar of E. Harding E. Harding
    4. August 2015 at 13:59

    That’s a pretty unique way of looking at things.
    And I don’t think you missed anything important. I mean, Windows 10 is out, and some lion has been killed in Africa. But that’s about it.

  2. Gravatar of ssumner ssumner
    4. August 2015 at 14:03

    I mentioned the lion in my newest Econlog post–looks like I didn’t miss anything.

  3. Gravatar of marcus nunes marcus nunes
    4. August 2015 at 14:56

    Doug Irwin “took care” of the RR event in his paper:

    “The recession of 1937-8 is often cited as illustrating the dangers of withdrawing fiscal and monetary stimulus too early in a weak recovery. Yet our understanding of this severe downturn is incomplete: existing studies find that changes in fiscal policy were small in comparison to the magnitude of the downturn and that higher reserve requirements were not binding on banks. This article focuses on a neglected change in monetary policy, the sterilization of gold inflows during 1937 and finds that it exerted a powerful contractionary force during this period. The transmission of this monetary shock to the real economy appears to have worked through lower asset (equity) prices and higher interest rates.”

  4. Gravatar of benjamin cole benjamin cole
    4. August 2015 at 15:27

    Excellent blogging.

    Regarding the TSA: I guess it’s forgotten today that in the 1960s, at the very height of the Cold War—in fact, during the Cuban Missile Crisis in 1962—a person could board an airplane in the United States without any ID. You had an airline ticket, like a bus ticket. That’s when we had a bona fide military opponent.

    Today because we face an “enemy” that has not an army navy or airforce, we are told we must spend hundreds of billions of dollars every year on a TSA and other national security apparatus, and have every e-mail and phone call recorded.

    BTW, 180,000 Americans have been murdered by terrorists since 9/11. Those terrorists are known as drunk drivers.

  5. Gravatar of Major.Freedom Major.Freedom
    4. August 2015 at 16:00

    The primary characteristic of money is actually medium of exchange. The fact that whatever is money is a medium of exchange is subsequently then adopted as a medium of account.

    Accounting in dollars arose because dollars were first, both temporally and logically, a medium of exchange.

    Accounting cannot take place without prices, and prices are exchange ratios.

    The easiest way to understand that money is primarily a medium of exchange is imagining what would be necessary if dollars ceased to be a medium of exchange, and what would be necessary if dollars ceased to be a medium of account.

    If dollars ceased to be a medium of exchange, it would be impossible to account for goods in dollars. Prices are needed, and prices are exchanges.

    If on the other hand dollars ceased to be a medium of account, it would not be impossible to exchange dollars for goods. In fact, that is what took place in history. Exchanges took place before accounting as a discipline arose. Accounting arose because of a need to analyze and tabulate past exchanges, not vice versa.

    Thus, if money is to “be” one thing primarily, it is a medium of exchange.


    Regarding the “sharp” division between money (the medium of exchange) and credit (loans of the medium of exchange), one of the most common errors in epistemology made by Sumner is the fallacy “If we don’t pay attention to X, then X is not implied or affective or necessary in what we advocate.”

    He has made a serious blunder in this respect before, when a couple of years ago (it was so profoundly obtuse I will never forget it) he claimed that because the Fed was not consciously tracking M2 or M3 during the 1920s, it means we cannot blame excessive monetary supply inflation for the unsustainable boom of the 1920s that culminated in a sudden “need” in 1929 for the Fed to drastically inflate even more.

    His error seems to not have been corrected.

  6. Gravatar of ssumner ssumner
    4. August 2015 at 16:55

    Marcus, Yes, Doug has done very good work on that episode.

    Ben, Yes, I recall those golden days of travel.

  7. Gravatar of Ray Lopez Ray Lopez
    4. August 2015 at 17:11

    This paper was discussed by Sumner’s apt pupil Sadowski in the comments section of the blog post “Is it time for the Fed to cut rates”. Here is what that this Photoshopped PhD says: “@ Mark A. Sadowski who says: “to my knowledge no market monetarist has ever blamed the 1937 recession on the raising of reserve requirements”

    Imagine that. See my comments there for the rebuttal.

  8. Gravatar of Airman Spry Shark Airman Spry Shark
    4. August 2015 at 17:24

    “An increase in base money is, ceteris paribus, a contractionary monetary policy.”

    Should that be “…in demand for base money…”, or am I completely lost?

  9. Gravatar of ssumner ssumner
    4. August 2015 at 19:00

    Ray, Which one has blamed rising res. req. for the recession?

    Airman, Thanks, that was a bad typo.

  10. Gravatar of Nick Rowe Nick Rowe
    5. August 2015 at 04:34

    ” imagine that reserve requirements applied to retailers, not banks.”

    Nice thought experiment.

    This post is an argument for general equilibrium as opposed to partial equilibrium analysis of money.


  11. Gravatar of ssumner ssumner
    5. August 2015 at 04:41

    Thanks Nick, I wasn’t sure that I didn’t miss something obvious, until I was reassured by your comment.

  12. Gravatar of TravisV TravisV
    5. August 2015 at 05:17

    ???? Ambrose Evans Pritchard: “Monetary expansion in Europe, America and China all point to stronger growth this year, signalling another leg to the global expansion” http://www.telegraph.co.uk/finance/economics/11782568/Day-of-reckoning-postponed-as-global-recovery-builds.html

    Meanwhile, inflation expectations are falling: https://twitter.com/SoberLook/status/628731544769830912/photo/1

  13. Gravatar of TravisV TravisV
    5. August 2015 at 05:25

    Oh, no, Noah Smith:

    “Austrian Economics Might Explain China’s Turmoil”


  14. Gravatar of dtoh dtoh
    5. August 2015 at 05:53


    As I keep saying the transmission mechanism works through an increased or decreased exchange of financial assets for real goods and services by the non-banking sector. An increase in the exchange will always be expansionary; a decrease will be contractionary.

    A change in the amount of financial assets exchanged for goods and services can be effected either through asset prices or through NGDP expectations.

    One of the reasons, there is confusion about the effect of Fed policy is that, the Fed has a couple of tools (asset purchases and reserve levels) that are often used at cross purposes.

    Also you should think about registering for Global Entry. It takes a lot of the hassles out of immigration, customs and TSA. If you travel overseas with any frequency, it’s definitely worth getting.

  15. Gravatar of TravisV TravisV
    5. August 2015 at 06:23

    VIX falls to 2015 lows: http://www.zerohedge.com/news/2015-08-05/vix-collapses-2015-lows

  16. Gravatar of collin collin
    5. August 2015 at 08:22

    You missed nothing important but it appears Donald Trump is not burning out…It might be time to take him more seriously winning the nomination.

    Otherwise, China stock market panic has ended.

  17. Gravatar of Ray Lopez Ray Lopez
    5. August 2015 at 08:31

    @Sumner, who asks me a question. Here is the answer I gave Sadowski, redacted for clarity:

    @ Mark A. Sadowski who says: “to my knowledge no market monetarist has ever blamed the 1937 recession on the raising of reserve requirements” –

    Do you know former Georgia U prof and monetarist Timberlake, who G. Selgin replaced? Timberlake wrote a whole book on the Fed Reserve and here is what he wrote in 1999.

    The Reserve Requirement Debacle of 1935-1938 – By Richard H. Timberlake, Jr. | Posted: Tue. June 1, 1999 Also published in The Freeman – “The Fed Board, under the urging of the Fed Bank of New York, raised reserve requirements by 50 percent in August 1936…The magnitude of this change is understood best when it is viewed in equivalent dollar value to an open-market sale of government securities. As Friedman and Schwartz point out, $3 billion “amounted to nearly one-quarter of total high-powered money.” …since the horrendous blunders in their implementation during the 1930s, the Federal Reserve has used them less and less for monetary control purposes…Even if not used, however, legal reserve requirements should be abolished completely so that the Federal Reserve Board could not blunder again into the monetary catastrophe it fostered in the 1930s.”

  18. Gravatar of Tom Brown Tom Brown
    5. August 2015 at 09:44

    Scott, O/T: Any thoughts on Romer’s recent posts?

    Have you ever encountered someone in the profession with more Stigler conviction than Feynman integrity? If so, you obviously lived to tell about it. ;^)

  19. Gravatar of TravisV TravisV
    5. August 2015 at 10:13

    Excellent reporting in the WSJ, feels an awful lot like late 2008:

    “China’s Response to Stock Rout Exposes Regulatory Disarray”

    “I want strong measures to rescue the market,” ordered the usually mild-mannered premier, according to the officials. At the meeting were those in charge of China’s central bank, the country’s securities and banking regulatory agencies and the Finance Ministry.


  20. Gravatar of TravisV TravisV
    5. August 2015 at 12:57

    I’m impressed that stocks increased in the face of this news:

    “The Odds of a September Rate Hike Have Surged in the Last Two Daus”


  21. Gravatar of Major.Freedom Major.Freedom
    5. August 2015 at 16:02

    “Monetarism can be viewed as a way of explaining changes in macroeconomic aggregates by looking at shifts in the supply and demand for money.”

    This incidentally is why Monetarism cannot in fact be used to explain what Monetarists habitually advocate!

    Never reason from a spending change.

    Since the supply of money is monopolized by the fascist-lite state institution called the Federal Reserve System, the only activity that comes even remotely close to market activity is the demand for money. Not actually market driven of course, because the demand for money is in part influenced and determined by the supply of money, which we know is socialist, not capitalist.

    Monetarism cannot be used to explain the causes of the periodic, widespread, and acute increases in the demand for money that we have observed in history, such as the period 2007-2009. Much like Keynesianism, it forces its adherents to deviate from economic science. To the Keynesians, it is the unscientific, inexplicable, almost religious-like “animal spirits”. Monetarism, since it is also incomplete in this way, uses similar conceptualizations.

    Couched of course in an epistemological inversion that claims the cause of an event is not what actually took place, but on what did not take place that allegedly would have otherwise quelled and overcome the effects of the “animal spirits.” Tiptoeing of course around the fact that the advocated solution is itself the primary cause for the effect having been repeatedly observed in the first place, all the while building up allocation errors making future events more pronounced than previous events.

    So we are told that the volatile swing in the demand for money that occurred 2007-2009 has no reason, or at least no reason that Monetarism can be used to address (because, of course, it would result in Monetarism self-detonating even in the minds of the Monetarists).

    Thus it is the same old same old socialist dogma that capitalism fails and it is up to the good government doctor, run of course by people, to save people from themselves. The trillions of dollars in costs, and the priceless value of human life, are to be sacrificed in maintaining the state so that we can live with a centralized counterfeiting institution, made legal by the state pf course, so that what? So that fixed income contracts, such as wage contracts, no matter how ludicrous they are in the aggregate, no matter how discoordinated labor is relative to all other labor, no matter if capital is to be consumed on net, no matter if people are poorer in the long run, no matter if wars overseas and domestic terrorism (from the state) results in hundreds of thousands of people not just losing their jobs, but their actual lives, all this we are told is a case where “the benefits” somehow can be summed up so as to “exceed the costs.”?

    I wonder how long it will take for Monetarists to ask a person what they want for themselves first, rather than presume and then sic the state’s rapist murderous goons on them if they want something else for themselves that differs from what the Monetarists want for them.

    Am I to live the rest of my life, are all individuals to live the rest of their lives, with the Monetarists not having the decency to at least ask? Or do they actually believe that reading a textbook makes them know better for others than others do for themselves such that violence against them is justified?

    This blog is an insane asylum, filled with true believers who are insane.

    At least I am smart enough not to fall for that insanity.

  22. Gravatar of Ray Lopez Ray Lopez
    5. August 2015 at 17:09

    @MF – see the TravisV link to Noah Smith* for a rebuttal to your inference that only Keynesians believe in “animal spirits”. Since economics is in fact non-linear, ‘animal spirits’ is a good shorthand phrase to describe such discontinuities and ‘jumps’.

    * key passage: N. Smith: “It has never been very clear exactly why malinvestment causes an economic hangover. Why don’t businesses just cut their losses and immediately start investing in something more useful, as soon as they realize that they’re doing the wrong thing? Austrian theory has never been particularly clear on that (and its notorious refusal to use precise mathematical models certainly doesn’t help). But at least Austrianism embraces the possibility that businesses might make big, systematic mistakes. “

  23. Gravatar of James Alexander James Alexander
    6. August 2015 at 00:09

    “… the profession badly misjudged the nature of the Global Financial Crisis, which was fundamentally due to a failure of monetary policy, not banking policy.”
    This message is still lost on most of the profession and almost all of the politicians and the public. Why? Because banking failures are so visible, so scary and bankers so often greedy, egomaniacal and imprudent. Still, one should never reason from a banking crisis.

  24. Gravatar of ssumner ssumner
    6. August 2015 at 09:57

    Thanks for the tip dtoh. I was actually in the TSA pre-check line, and it was still a fiasco. Is global entry different from TSA precheck?

    Thanks Travis, those are very helpful.

    Collin, No it is not now and never will be “time to take him seriously.” He’s a clown with 0.00% chance of winning the nomination. I’m more likely to do so.

    Ray, Timberlake is a market monetarist? Who knew?

    Tom, I can think of a famous blogger who shall not be named. Yes, I’ll do a post on that when I have more time.

    James, That’s right.

  25. Gravatar of dtoh dtoh
    6. August 2015 at 14:47

    Global entry will get your through customs immigration faster.

  26. Gravatar of ssumner ssumner
    7. August 2015 at 05:10

    Thanks dtoh.

  27. Gravatar of flow5 flow5
    8. August 2015 at 08:40

    Contrary to the pundits, the incipient power of the RRP facility is indeed awesome. Yellen just barely kept her job and staved off an imminent recession (reversing normalization principles). But recent reporting weeks have reversed part of this early trend.


    “The Committee intends to reduce the Federal Reserve’s securities holdings in a gradual and predictable manner”

    The RRP facility was critical to the Fed’s success. It allowed the injection of reserves (repurchase of securities loaned out), to have an immediate positive impact upon economic activity (which, could be either expansive or contractive – according to the FOMC’s needs). These TOMOs to have a dollar-for-dollar impact on the money stock (as transactions are concentrated when the FRB-NY’s “trading desk’s” sales and purchases are primarily with non-bank counterparties).


    “When the Desk conducts an overnight RRP, as in the current ON RRP exercise, it is selling an asset held in the System Open Market Account (SOMA) with an agreement to buy it back on the next business day. This leaves the SOMA portfolio the same size, as securities sold temporarily under repurchase agreements continue to be shown as assets held by the SOMA in accordance with generally accepted accounting principles, but the transaction changes some of the liabilities on the Federal Reserve’s balance sheet from deposits to reverse repos while the trade is outstanding.”

    This was a surreptitious policy change (not a transparent, up-front, announcement).

Leave a Reply