L Street – Selgin’s prescription for Money Market reform

Yesterday, I wrote a post on George Selgin’s latest presentation on monetary reform at the Italian think tank the Bruno Leoni Institute. In his presentation George essentially outlines a tree step strategy for the privatisation of the supply of money.

I described these three steps in my post yesterday, but I would like to take a deeper look at the first reform George proposes – reform of the way the US Federal Reserve controls the US money base and more concretely reform of the US money markets.

He outlines these reforms in his paper “L Street: Bagehotian Prescriptions for a 21st-Century Money Market”.

L-Street is of course a wordplay on Lombard Street – the book in which Walter Bagehot outlined the famous proposal for central banks to act as lender of last resort – and on the address of the US Federal Reserve in Washington DC.

George of course is the world’s foremost expert on Free Banking Theory, but in L-Street George takes a more practical approach to the monetary issue and assume that central banks at least for now is with us. That said, George also clearly states that he see money market reform as a step in the direction of Free Banking. However, this in no way central to the understanding of George’s proposals and I think that George’s reform proposals have a lot of merit on its own and should be of interest also to those you are not in favour of abolishing central banks.

Here is the abstract from George’s paper:

“In Lombard Street Walter Bagehot offered some second-best suggestions, informed by the crisis of 1866, for reforming the Bank of England’s conduct during financial crises. Here I respond to the crisis of 2008 by proposing changes, in the spirit of Bagehot’s own, to the Federal Reserve’s operating framework. These changes are aimed at reducing the Fed’s interference with the efficient allocation of credit, as well as its temptation to treat certain financial institutions as Too Big to Fail, during crises. More fundamentally, they seek to ground Fed operations more firmly in the rule of law, and to thereby make them less subject to the whims of committees, by allowing a fixed but flexible operating framework to serve the Fed’s needs during financial crises as well as in normal times.”

The failure of Fed’s operating framework

George’s starting point is a critique the Federal Reserve’s reliance on so-called primary dealers in its general operating framework. I find his discussion of the role of the primary dealers in events of 2008 very interesting and insightful and his discussion clearly illustrates that structural flaws in Fed’s operating framework seriously hampered the Fed’s role of an effective lender of last reserve. Furthermore, it gives a “micro perspective” on some of the issues discussed by market monetarists likely Scott Sumner who normally tend to have a more macro perspective on the crisis. These “explanations” in my views complement each other perfectly well.

Here is George:

“Further consideration suggests, however, that the apparent need for direct lending during crises stems, not from the inadequacy of open market operations as such, but from the inadequacy of the Fed’s particular rules and procedures for conducting such operations, including its reliance upon the primary dealer system. In particular the Fed, by depending upon a small set of primary dealers, and on two clearing banks, for its open-market operations, risks a breakdown in the monetary transmission mechanism when these agents themselves become troubled. Consequently the Fed may be compelled, not merely to engage in direct lending, but also to depart from Bagehot’s principles by bailing out insolvent firms when their failure threatens to cause a breakdown in its operating framework. The Fed’s reliance upon primary dealers and tri-party repos thus contributes to the notion of the “Systemically Important Financial Institution” (SIFI), official recognition of which, according to former Kansas City Fed President Thomas Hoenig (2011), poses a serious threat to the future of capitalism.”

I think George’s analysis fundamentally is right and it is also a good explanation why the Federal Reserve has been so preoccupied with saving (investment) banks rather than focusing on ensuring a stable nominal anchor for the US economy.

Market Monetarists including myself have tended to blame Fed chairman Bernanke’s Creditist views for the Fed’s intense focus on financial intermediation rather than on for example increasing nominal GDP. However, George’s discussion of the Fed’s operating framework also shows that the overreliance on the primary dealers in the day-to-day (and emergency) conduct of monetary policy basically “forced” the Fed in the direction of the Creditist position. Furthermore, as the system relied heavily on the primary dealers (and still do) the Fed basically had no other choice than to help bail out these institutions if it wanted to maintain its basic operating framework.

Selgin’s prescription: Abolish the Primary Dealer System

George’s solution to the problem of overreliance of the primary dealer system is simply to do away with it. Here is George:

“The most obvious operating system reform suggested by the crisis is to replace the primary dealer system with one in which numerous financial firms, and perhaps even some nonfinancial firm, take part in the Fed’s open market operations.

There are good reasons for the Fed to dispense with its primary dealer system even putting aside the dangers of relying upon it during crises. “In central banking terms,” as Chris Giles and Gillian Tett (2008) observe, despite its long pedigree the Fed’s primary dealer system “is decidedly old-fashioned,” having, as Bob Eisenbeis (2009: 2) explains, “evolved prior to the advent of electronics and computerization of the bid and auction process when institutions relied upon messengers to transmit paper bids to the [System Open Market] Desk.” Today, Eisenbeis goes on to observe, there’s no reason why a much larger number of qualified firms “could not take part in the daily Open Market transaction process through the System’s electronic bidding process.””

George continues:

“The Shadow Financial Regulatory Committee, of which Eisenbeis is a member, has recommended that the Fed take advantage of modern technology to adopt an approach similar to that of the ECB, which routinely conducts open-market operations “with more than 500 counterparties throughout the Euro Zone,” and which might deal with more than twice as many. Doing so, the committee maintains, “would increase the efficiency of the SOMA transaction process, lower costs, reduce dependence upon a geographically concentrated set of counter parties, and enhance the monetary policy transmission process” (Shadow Financial Regulatory Committee 2009). Electronic trading could also preserve the anonymity of firms seeking funds from the Fed.

Such improvements, it bears noting, would supply a rationale for doing away with the primary dealer system even if primary dealers’ soundness were never in doubt.

So far as outright open-market purchases are concerned there is no reason at all for the Fed to restrict the number of its counterparties, even by limiting participation in open-market operations to financial firms, since it doesn’t expose itself to counterparty risk in making outright purchases. The only risk it takes on is that connected with depreciation of the securities it acquires, which is of course a function, not of the counterparties it deals with, but of the securities it chooses to buy.”

I think George’s position makes a lot of sense and to me the main point in terms of the conduct of monetary policy must be that the central bank is as little dependent on any individual institution so the central bank will not be taken “hostage” of a single institution. That is effectively what happened during late 2008 and 2009.

The Fed as the market maker of last resort
Furthermore, Selgin suggests to get rid of the Federal Reserve’s “Treasuries only” policy for open–market operations and instead “broadening of the set of securities used in its temporary, if not in its permanent, open-market operations”.

Here is George:

“In particular, there are good reasons for having the Fed engage in temporary purchases of some of the private market securities it has traditionally accepted as collateral for discount window loans, provided that it subjects those securities to “haircuts” sufficient to both protect it against potential credit risk, while otherwise adhering to the classical rule of supplying credit only on relatively stiff terms.”

This proposal is similar to what Willem Buiter and Anne Seibert have called “the central bank as market marker of last resort”.

While I have a lot sympathy for this proposal I also think it is very important to stress that if the central bank acts as market maker of last resort then that does not mean that the central bank should try to manipulate the relative prices of financial assets. Sometimes it seems like especially the Fed has been trying to do exactly that. What George is suggesting is rather that the Federal Reserve expands the type of collateral that it will accept in the conduct of market operations. Here the condition for sufficient haircuts of course is key. Central banks should not be in the business of doing subsidized lending.

To avoid that open market operations lead to a distortion of relatively prices of financial assets Selgin suggests an auction system for liquidity:

“The procedure I have in mind, if only in the crudest of outlines, combines a set of reverse (single price) auctions for particular securities, with prospective counterparties allowed to submit multiple (but mutually-exclusive) bids involving different securities, with a multiple-yield (discriminatory price) auction that determines which securities the central bank actually ends up purchasing. The Fed would first have to decide what security types are eligible, favoring those for which holdings are sufficiently dispersed to provide for competitive bidding, and (to further discourage adverse selection) indicating maximum values of total and individual security purchases that it is prepared to make from a single participants. The list of such securities could be compiled, and regularly updated, using reports regularly submitted by prospective counterparties as one requirement for eligibility. Next the Fed would announce the total value of an intended purchase, along with reference prices (reflecting risk-based “valuation haircuts”) for particular securities. It would then hold simultaneous reverse (single-price) auctions, with descending prices expressed as reference-price percentages, for each security type, allowing individual counterparties to take part in any or all; but before actually completing any purchases it would first rank offers from the various auctions according to the percentage of the security reference prices involved, and then accept offers starting with those entailing the lowest percentage, progressing to higher percentages until the predetermined aggregate purchase is completed.”

George rightly conclude that if the Fed would act as market maker of last resort as he suggest the Fed would no longer need to act as a traditional lender of last resort through the so-called Discount Window and he therefore suggests that the Discount Window should be abolished. Fine with me, but don’t do it before you are sure that the other reforms suggested are working.

One can conclude that if the Fed moved in the direction of becoming market marker of last resort and got rid of the Discount Window then the Fed could effectively control the money base without any reliance of any particular institution and it would mean a minimum of distortion of relatively prices in the financial markets. This would be a huge step forward in the direction of a more market based monetary system.

Bernanke should have a look at Selgin’s prescriptions

Overall, I think that George Selgin’s L-street reforms make a lot of sense and if implemented likely would make the framework for conduct of US monetary policy a lot more transparent, but most important it would greatly insure against the kind of ad hoc policies measures that the Fed has had to resort to after the outbreak of the crisis in 2008.

In addition to Selgin’s proposal for L-street reform he also proposed NGDP targeting and finally the privatisation of the supply of money based on what he terms a “Quasi-Commodity Money” standard in his Bruno Leoni Institution presentation. As far as I understand George is in the process of finalizing papers on these two topics as well. I am surely looking forward to reading (and maybe reviewing) these papers as well.

PS To my American readers on this Super Bowl Sunday – Football is a sport where you kick the ball with your FEET and you are not allow to pick it up in the hands…anyway enjoy the game.

Update: See a presentation of Selgin’s reform ideas here.

Update (April 6 2012): Scott Sumner also comments on George’s paper here.

Leave a comment


  1. Selgin’s suggestions sound very good. have another idea, though, for how a central bank could manage the base.

    If they need to increase it, they either add money to a treasury account or send out equal amounts to all citizens.

    If they need to decrease it, they could either be allowed to create government debt and sell those on the open market, or they could deduct money from the treasury account (perhaps forcing them to issue more debt),

    This could prevent the central bank from ever having to make any OMO:s at all.

  2. Diego Espinosa

     /  February 5, 2012

    In general all great ideas from Selgin. The problem is an explicit MMOLR promise amounts to a free Fed put on illiquidity issued to all securities holders. This would result in actors having less liquid portfolios, and it would increase systemic risk. The Fed’s 1998 LTCM bail out had a similar effect on the system. Look at the evolution of short term wholesale financing by banks and shadow banks since the early 2000’s. The growing over-reliance on this funding — enabled by the Fed put — eventually became an important cause of the GFC.

    I wonder how Selgin would address the issue of free central bank puts handed out to non-supervised financial firms? Would a mandatory liquidity insurance scheme (with risk-adjusted premiums) work to eliminate the distortions caused by the explicit liquidity put?

  3. Diago, good points…lets hope George drops by and comment on some of these issues.

  4. Vaidas Urba

     /  March 2, 2012

    ECB’s 3-year LTRO is an auction that works much better than Selgin’s auction scheme.

  1. Selgin on Quasi-Commodity Money (Part 1) « The Market Monetarist
  2. Is Market Monetarism just market socialism? « The Market Monetarist
  3. International monetary disorder – how policy mistakes turned the crisis into a global crisis « The Market Monetarist
  4. Selgin’s challenge to the Market Monetarists « The Market Monetarist
  5. Wrap-up: My Free Banking related posts « The Market Monetarist
  6. A five-step plan for Mark Carney | The Market Monetarist
  7. The Casselian-Mundelian view: An overvalued dollar caused the Great Recession | The Market Monetarist

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: