A New Operating Framework for the Federal Reserve (Joseph Gagnon and Brian Sack)

Commenter AH pointed to an FT Alphaville post that in turn pointed to this paper by Joseph Gagnon and Brian Sack:


The paper is excellent and certainly deserves a complete read. This is a highly technical topic, not only with respect to bank reserves and reserve management regimes, but also as it relates to the Treasury repo market and the particular way in which the authors propose to incorporate a very active Federal Reserve reverse repo facility into their proposed operating framework for future monetary operations. There is no getting around the complexity. In an effort to assist and to encourage a reading of the full paper, I’ve attempted to summarize it in précis form by direct quotation of a sequence of selected key paragraphs, connected by comments of my own. My general comments on the proposal then follow.

A Summary of the Proposal

The authors recommend a change in the Fed’s operating framework in which:

a) The policy interest rate target changes from the fed funds rate to a rate which will be administered in dual form as the interest rate paid on Fed reverse repurchase transactions (RRP rate) and a matching interest rate paid on bank reserves (IOR).

b) Because of the scope of the intended repo facility, more non-bank counterparties will be included in RRP transactions

c) The eventual balance sheet would include a permanent tranche of excess liquidity in the form of both reserves and reverse repurchase agreements, paying interest respectively at the same administered rate. The RRP rate will be maintained at the IOR level through “full allotment” auctions.

They explain it this way:

“The Fed should set the interest rate at which it will offer overnight reverse repurchase agreements as its policy instrument and that it should maintain the interest rate paid on bank reserves at the same level. Under our proposal, all banks and many other financial institutions would have an unlimited ability to invest at the Fed at the specified interest rate. All other interest rates, including the federal funds rate, would be determined in the market, presumably with the risk-free interest rate set by the Fed exerting a powerful influence on them… the Fed should not shrink its balance sheet all the way back to a size that would have been considered normal prior to the global financial crisis but should instead leave a larger amount of liquidity in the financial system on a permanent basis.”

It seems the financial crisis was the original motivation for the idea of supplying unlimited reverse repo transactions to the market in the context of a permanent operating framework. In the crisis, those supplying investment dealers with repo financing began to reject lower quality collateral for purposes of securing that funding. Dealers experienced a financing crunch because they couldn’t supply high quality Treasury collateral to secure their financing. High quality collateral became scarce relative to demand for it. And those supplying the funding were willing to accept lower rates for the right kind of collateral. General collateral repo rates dropped well below the fed funds rate. And even after credit conditions improved to a more normal stance, another issue resulted in a continuing dislocation in interest rate spread relationships:

“The federal funds rate and other short-term interest rates were expected to remain relatively close to the IOR rate. One reason to expect this pattern is that banks can perform an arbitrage by borrowing in the federal funds market and holding the resulting funds as reserves at the Fed… In practice, though, bank funding rates and other overnight interest rates have traded notably below the IOR rate.”

This interest rate spread dislocation was driven in part by the existence of different arrangements for interest paid on balances held at the Fed by two different categories of participants:

“This pattern reflects the fact that some large lenders in the federal funds market are not eligible to receive interest on reserves (mainly government-sponsored enterprises, or GSEs). Banks are willing to borrow from these entities and hold the proceeds as reserves, performing the arbitrage noted above, but they require a yield spread to do so because they view the associated increases in their balance sheets as costly in terms of required regulatory capital and internal oversight. In addition, banks have to pay a fee to the Federal Deposit Insurance Corporation (FDIC) related to the size of their balance sheets, which directly reduces the return on this activity by 10 to 15 basis points on average.”

And many participants such as money market funds were not allowed to hold balances at the Fed:

“It is possible that this gap would have been larger if the federal funds rate had not been constrained from below by the zero bound on nominal interest rates. Moreover, other types of short-term interest rates, such as the rate on an overnight RP for general Treasury collateral (Treasury RP rate), also traded well below the IOR rate over this period. As with the federal funds rate, this pattern reflects the fact that key participants in this market are not eligible to receive interest on reserves from the Fed and hence have to accept a lower return in the market.”

The authors see these aberrant spread relationships as a problem:

“The presence of these gaps suggests that the ability of the Fed to control a broad range of market interest rates by setting the IOR rate alone cannot be taken for granted.”

The Fed has been testing new instruments in anticipation of its planned transition from its quantitative easing program. One of these is essentially the same RRP instrument that the authors propose for the future operating framework on a more permanent basis:

“In order to enhance its control over short-term interest rates, the Fed has developed and tested two new tools: a term deposit facility (TDF) and a reverse repurchase (RRP) facility that can be implemented for overnight or term transactions… the TDF is available to the same depository institutions that receive interest on reserves. In effect, the facility allows those institutions to “lock up” their reserves for a longer period, presumably in return for a slightly higher interest rate than the expected path of the IOR rate over the term of the deposit. In contrast, the RRP facility is available to a broader set of market participants, including primary dealers, money market funds, and GSEs.”

The federal funds rate would no longer be the key short term rate for setting monetary policy:

“The rate at the RRP facility (which we will simply call the RRP rate) would replace the federal funds rate target as the primary policy instrument of the Fed. The FOMC would ask the Federal Reserve Board to set the IOR rate equal to the RRP rate.”

The proposed system is a customized version of a policy framework that has already been adopted by several central banks:

“Our proposed policy framework falls into a category that is sometimes referred to as a “floor system.” The key attributes of a floor system are a large volume of liquid central bank liabilities, including bank reserves, and control of short-term interest rates through the rates paid by the central bank on its liabilities. Several major central banks are operating floor systems at this time, although those systems rely entirely on interest paid on bank reserves to manage overnight interest rates. Our proposal instead recognizes the importance of the nonbank sector in the creation of credit and the determination of broader financial conditions in US markets, and it therefore extends the floor system to a broader set of firms through the RRP facility.”

The authors foresee the following interest rate relationships developing under the proposed system:

“We expect the interest rate on Treasury bills maturing within a few days to trade at rates modestly below the RRP rate, given that bills can be held by a broader set of market participants (such as pension funds and individual investors); the Treasury RP rate to trade close to or several basis points above the Fed’s RRP rate; the federal funds rate and the overnight eurodollar rate to trade modestly above the Treasury RP rate for highly rated borrowers in normal times, with the spread reflecting the credit risk associated with unsecured transactions; and three-month London interbank offered rate (Libor) to price off of the expected path of the overnight federal funds/eurodollar rate at a spread that is similar to that currently observed. Bank deposit rates and money market mutual fund rates could be lower than the RRP rate because of costs of intermediation.”

They foresee the Fed if necessary expanding the scope of RRP transactions even further in order to set a broad floor on short term interest rates:

“If other short-term rates trade significantly below the RRP rate, the Fed could respond by expanding the list of RRP counterparties to include important investors in the assets whose rates are lower, thereby giving them the option of earning a higher return. By including most money market mutual funds as counterparties, the Fed has already captured the most important class of nonbank investors, and hence we believe that such expansion will not prove necessary. However, there is little reason not to be as expansive as possible, as broader participation should only strengthen the floor and make the proposed framework more effective.”

Moreover, its control of the rate structure through RRP transactions could be complemented by RP transactions (which inject reserves rather than withdrawing them):

“The Fed could consider modifying our proposed framework to incorporate an automatic mechanism for providing reserves if market interest rates are judged to be too far above the RRP rate or too volatile even in normal times. In particular, within our proposed framework, the Fed could also offer a standing RP facility with full allotment at a rate moderately higher than the RRP rate (say 10 basis points above the RRP rate). This facility would automatically inject reserves by lending to financial institutions when market rates became firm relative to the RRP rate. The modified framework would therefore have a symmetric approach, with standing facilities to react to soft market rates (through RRPs) and to firm market rates (through RPs).”

The proposed operating framework is quite aggressive in its prescription for expanded Fed influence in money markets – to the point of positioning the Fed effectively as an investment dealer of last resort as it concerns the repo market. This even includes what is in effect a profit rationale for such active balance sheet management:

“Moreover, our proposed framework should have positive implications for the taxpayer over the longer run by allowing the Fed to maintain a larger balance sheet. Even though the larger balance sheet would result in larger interest payments on reserves and RRPs, those liabilities would be matched by assets that generate positive income. On average, we expect that the Fed would earn more on its assets than it pays on its liabilities, because of a positive risk premium on the assets held.”

However, it is also interesting that that the authors don’t completely buy into the idea that QE has negatively affected repo market conditions by withdrawing Treasury collateral from the market, citing the large gross supply of new securities being delivered to the market by large budget deficits. They are merely “open to the idea”. It is difficult to square this qualified view entirely with their more general argument that collateral shortages have driven repo rates down below IOR, thus requiring in their view the aggressive proposal presented in the paper:

“The Fed’s large-scale asset purchases have reduced the amount of liquid debt securities available as collateral for a variety of financial market activities. Some argue that the resulting scarcity of Treasury securities has hindered the efficient functioning of financial markets and the ability of the markets to effectively create credit outside of the banking system. Although we find these concerns to be overstated in light of the large net issuance of Treasury securities in recent years, we are open to the idea that making the Fed’s holdings of Treasury securities available to the market could benefit the functioning of financial markets… By having a standing RRP facility, the Fed would be making its holdings of Treasury securities available to alleviate any shortages. Indeed, if the market were finding Treasury securities to be in short supply, the Treasury RP rate would be expected to decline, all else equal. This would make the Fed’s RRP facility more attractive, allowing the market in aggregate to shift liquidity from bank reserves into Fed RRPs. These transactions would, in effect, make the Fed’s Treasury holdings available to the market, providing some relief for the collateral shortage.”

That series of quotations summarizes the key points in the paper and the proposal.

Comments on the Proposal

This proposal considers the future operating framework for monetary policy from a starting point of a complex setting that includes the following dimensions:

a) Outsized excess banking system reserves due to the implementation of quantitative easing

b) A prolonged zero bound interest rate policy setting

c) Payment of interest on reserves

d) Potentially troubling interest rate spread relationships between interest on reserves, fed funds, and general collateral repo rates

e) A planned long term exit from the quantitative easing policy

f) The likelihood and near certainty of a required increase in policy interest rates prior to the completion of the transition from QE

g) An uncertain “end game” after the completion of the transition from QE, with a permanent framework for monetary operations to be fully defined at that time

The authors map out a future that concludes with a proposed permanent framework that will not be entirely free of the residual QE portfolio until around 2025. That is the time frame that the authors foresee as required for complete exit from quantitative easing. They build in flexibility for adjustment of policy along the way as the Fed gains experience in observing how the markets react to various stages in the transition. The envisaged permanent state incorporates the capacity to respond to future financial crises contingencies with operational capabilities that can meet the challenges of disruptions to normal operations.

The proposal ensures that the Fed has very tight control over two short term interest rates – interest on bank reserves (IOR) and the general collateral overnight reverse repo rate (RRP). The Fed will set an administered interest rate on reserves and supply unlimited full-allotment auction reverse repo at the same rate. The fed funds rate is expected to assume less importance. The Fed will transact in the reverse repo market as a regular matter of interest rate maintenance. Both bank reserves and reverse repos constitute balance sheet funding liability positions for the Fed. The authors estimate this will require an excess liquidity provision when the Fed balance sheet reaches its permanent status of approximately $ 200 billion, which will be apportioned between excess reserves and reverse repo transactions, according to the demand for reverse repo at the administered rate. That $ 200 billion compares to today’s excess reserve position of approximately $ 2.6 trillion.

The authors have given careful consideration to the many risks and contingencies inherent in the full transition to a future Fed operating framework and deal expertly with those dimensions. The proposal is seamless and flexible and looks very good as a coherent and workable design for a new type of Fed monetary operation. But questions arise. The following observations are offered in respect of some of the salient points that underlie this comprehensive proposal:

1. In one sense, bank reserve balances held at the Fed might be viewed as the most liquid asset in world. But a fallacy of composition may be present. If one views the banking system as a whole in juxtaposition with the rest of the world, then it might be said that bank reserves in aggregate constitute an asset with no liquidity at all. For example, the banking system could choose to sell all of its excess treasury bills to the rest of the world. But it could not choose to sell all of its reserves to the rest of the world. It couldn’t sell any reserves in that sense – not without the assistance of depositors exercising their option to convert their bank deposits to central bank currency (which extinguishes reserves) or of the central bank exercising its option to tighten the system by actively withdrawing reserve balances. The banking system itself has no such option. That would seem to be the essence of illiquidity. One could argue that this macro illiquidity of bank reserves is essential to the structural premise for the authors’ proposal. It is the inability of banks to make their reserve balances at the Fed accessible to non-banks that is essential to the argument the authors put forward regarding what they perceive to be counterproductive pricing differentials that result from that macro illiquidity effect.

2. The authors’ proposal essentially bundles two proposals into one – the choice of a floor system for central bank reserve management and the modification of that system to incorporate non-bank participation through the full allotment reverse repo mechanism.

3. It is possible the authors may have underestimated the potential for IOR to act as an effective overnight anchor for the rate structure of the entire financial system. This is not certain, but it is possible. It is reasonable to assume that the Fed will begin to increase interest rates from the zero bound only when as a necessary condition it perceives that some threshold of stability in credit markets has been reached – i.e. post financial crisis in full. From a credit risk standpoint, we know that reserves are risk free. Therefore, it is logical for commercial banks to price asset credit risk – i.e. the interest rate differential required to accept assets of a certain risk – as the interest rate increment over the IOR. Just as there is a cost to funding new assets at the margin, there is an opportunity cost of holding excess reserves in the same context. That is the expected return on equity that is foregone by not acquiring a risky asset with an acceptable risk adjusted rate of return. Other things equal, the acquisition of that asset will be paid for with bank reserves. Even when loans create deposits, it is reasonable to assume that the deposit will leave the lending bank, following the borrower’s use of funds elsewhere in the financial system. Banks obviously also require sufficient capital to make such a decision to acquire a risky asset. In normal times, without the excess liquidity overhang of quantitative easing, a commercial bank will use a straightforward cost of funds benchmark. This may be a yield curve that begins with the overnight fed funds rate or a similar CD rate. But in times of outsized excess reserves where asset liquidity is forced onto the banking system, the cost of funds benchmark may instead be interpreted as the interest rate that is foregone on the reserve balances that can be expected to leave the bank when the asset is acquired. Thus, the anchor point for the effective cost of funds curve will shift from the liability side of the balance sheet to the asset side. None of this is to deny the fact that the expansion of risky assets requires capital. And it does not require excess reserves. But outsized excess reserves are still an asset with an interest rate and a corresponding effect on bank interest margins. And even though the supply of reserves for the system is closed as a whole, banks in competition recognize the alternatives that arise from acquiring assets that may displace reserves from their own bank to a competitor bank. Therefore, IOR will serve the same purpose as overnight fed funds previously might have in acting as a benchmark indicator for appropriate rates to set or bid on both asset and liability pricing. The scope of this perspective is enormous. From a funding perspective, this includes trillions of dollars of deposits, bonds, and equity funding. And it include the vast majority of commercial bank assets that are quite disengaged from the general collateral repo market. And all of this asset-liability activity requires disciplined pricing – either administered or market – and pricing requires a reference point, which at the front end of the yield curve may be IOR for the reasons given. And note importantly that none of this core commercial bank funding is repo funding, and nor can the vast bulk of it be replaced by general collateral repo funding, simply because of the nature of the core assets that are being funded. Therefore, there is no necessary natural connection between the pricing of this funding and that of general collateral repo funding, simply because substitutability is generally inapplicable.

4. Given these trillions of dollars in non-repo funding and assets, it is not clear why the repo market cannot be left to clear at its own pricing relative to IOR. In other words, it is not clear why the repo market cannot be left to price in the same way as Treasury bill and Treasury bond prices are left to price according to the market’s determination. Yes, the repo market is several trillion dollars in size. But core commercial bank balance sheets are much bigger than this in total. And risk markets are much bigger than the government market. Given the relative minor size of repo balances on commercial bank balance sheets and the requirement to take in a much greater size of other types of funding, IOR as set independently of a reverse repo rate may be the more natural reference point to start the yield curve for balance sheet pricing.

5. It is uncertain what future spreads would be between repo funding and IOR, if repo funding costs were left to float instead of being fixed as in the proposal. But if it is the case that IOR can be the overnight pricing anchor for a general lift in commercial bank asset and liability pricing, the degree of overall tightness in the system by comparison would be a mixture of the tightness of commercial bank pricing and that of repo funding, Treasury markets, and all other markets affected by either of these sources. To the degree that the Fed has to learn through experience the new spread relationship between IOR and market priced repo funding, the general lift in rates as indexed through IOR might have to be higher than the case where IOR and reverse repo are both equal administered rates in effect. But if the assumption prior to any interest rate tightening by the Fed is that the credit markets are working with reasonable efficiency, why not let the markets determine that spread?

6. Thus, it is not obvious why the overnight Treasury repo market should need to converge to the same pricing as IOR on the basis of risk characteristics. The usual definition of the risk free rate is a focus on credit risk – on the risk of eventual capital loss. But such a concept does not take into account the macro liquidity difference between bank reserves and government securities, as noted above. The uniqueness of the bank reserve category means that there is no apparent argument for perfect pricing convergence. Even though bank reserves are risk free in the conventional sense while fed funds are not, the unique “macro captive liquidity” characteristic of excess reserves sets them apart. This suggests that bank reserves can be viewed as a substitute overnight pricing benchmark in an environment of outsized excess reserves.

7. The authors’ proposal visualizes full exit from quantitative easing around the year 2025. That’s a long time for the Fed to be able to adjust for experience. The role of full allotment reverse repos, now being tested by the Fed, will naturally be experimental in that context. There’s considerable time to adjust for purposes of both interim management in the winding down of quantitative easing and the eventual desired permanent state for future monetary policy.

8. The current level of $ 2.6 trillion in excess reserves is in effect a by-product of the Fed’s LSAP program – large scale asset purchases. The Fed is quite right in referring to this as LSAP rather than quantitative easing. The objective is not bank reserves per se – it is interest rate transformation. Bank reserves just happen to be the balance sheet consequence of that interest rate strategy. Payment of interest on reserves is a functional requirement for central bank interest rate management. But outsized reserves of the size created by LSAP are functionally redundant in their usual primary role as the commercial banks’ medium of exchange. It is the interest rate on the asset and the effect of that interest rate on interest margins that commercial banks must take into account in assessing other opportunities for interest margin and return on equity management.

9. The current enormous supply of excess reserves naturally reduces the demand for fed funds borrowing. This softness activates an opportunity for a particular kind of systematic reserve arbitrage that has always existed in theory, but which now exists in practice. The authors note the institutional circumstances that have led to the undesirable situation of fed funds trading below interest on reserves. A primary source of this problem is the fact that certain participants – notably the GSEs – maintain accounts at the Fed on which (unlike the banks) they receive no interest. This opens up an arbitrage opportunity whereby these participants can lend fed funds at a rate lower than IOR to banks that do maintain accounts at the Fed. One would think that the Fed balance sheet should be managed in such a way that it avoids creating such an arbitrage opportunity. It would seem that an obvious solution is to pay interest on all balances. The Fed could immediately request this from Congress, the same way in which it requested authority to pay interest on bank reserves. From a political standpoint, this would have the undesirable effect of increasing the deficit, other things equal, but the amount of money is immaterial relative to hundreds of billions of dollars that the Fed has remitted to Treasury during the financial crisis as a result of its expanded balance sheet operations. This would remove potential for this particular type of overt arbitrage due to the different interest rate treatment for two different classes of depositors on the Fed’s balance sheet. And we note that the reverse repo facility will be offered to those GSEs that are presently taking advantage of this arbitrage facility, so the net result in terms of the Fed’s interest rate cost will be no different. The Fed should not have a liability structure that is susceptible to systematic arbitrage. A related problem is that that foreign banks with reserve accounts at the Fed do not pay the same FDIC fees as the domestic banks. This moves the calculus for their break-even arbitrage additionally in their favor when they act as intermediaries between lower cost money market funding and balances at the Fed that pay IOR. That is an institutional difference in the cost of funds that may be more difficult to neutralize.

10. The important point here is that these types of inefficiencies become exploitable when the supply of reserves is not binding for the fed funds rate. It is the existence of chronic excess bank reserves that has created the pre-conditions for such potential institutional inefficiencies to be exploited and used for interest rate arbitrage purposes. This is not to question the necessity of excess reserves as a by-product of quantitative easing objectives for interest rates in the bond market. But the excess reserves that are now sitting on bank books are for the most part redundant for their primary purpose –which is to make payments in the interbank clearing system. They are mostly useless for that purpose and yet they are now arbitraged – in large part by the GSEs and foreign banks – for the purpose of earning arbitrage spreads. To the degree that these inefficiencies contribute to the interest rate spread aberration now observed under conditions of quantitative easing, a proposal for a permanent floor system would seem to act as a catalyst for carrying over existing institutional inefficiencies that in themselves become part of the rationale for moving to a full allotment administered rate reverse repo facility. By contrast, it is conceivable that a corridor system with tight excess reserve management might mitigate the opportunity for such inefficiencies to take hold of market spread relationships – as apparently was the case prior to the financial crisis. In other words, the full allotment idea solves a problem that might be forestalled to some degree by adopting a corridor reserve system (asymmetric or symmetric) as the future permanent framework – insofar as the exploitation of these inefficiencies is concerned. Furthermore, with a general tightening of interest rates in an environment that is hopefully characterized by healthier credit conditions overall, it should be the case that the repo market at large will not be characterized by a systemic shortage of Treasury collateral and that the Treasury collateral market is being priced efficiently by the financial system on that basis. After all, the only reason for the Fed to stand in as price maker for Treasury collateral is if the private sector is not doing that job efficiently itself. But who is to say that pricing efficiency means that GC overnight repo rate must converge to IOR? In any event, the potential resolution of such institutional inefficiencies might be addressed up front as a precursor to setting forth an optimal design for the new policy framework.

11. The proposal is grounded by the choice of a floor system as the future central bank reserve management regime. The authors have described the full proposal as one that expands that reserve management component to a wider range of participants through the implementation of the full allotment reverse repo facility. Thus, the proposal might be characterized as a core reserve management framework, plus the reverse repo add-on.  Options for the core reserve management framework include the previous pre-2008 “asymmetric corridor” system of the Fed, a standard symmetric corridor system, or a floor system. This raises a fundamental question regarding a proposal that chooses as the core system the one with outsized excess reserves and excess liquidity as a matter of course (the floor system). It is the existence of this excess liquidity that has been associated with slackness in the interbank fed funds market and with some of the institutional interest rate arbitrage that currently exists insofar as reserve balances are concerned. Foreign banks primarily are sourcing lower cost funds from GSEs and other sources and passing that through to reserve holdings at the Fed. And yet the reverse repo add-on is designed in part to remedy this malady that is associated with the choice for the core reserve system. The usefulness of the full allotment reverse repo facility as an extension of the reserve system thus seems dependent upon the selection of the core system as between floor and corridor. While the authors have done an excellent job of explaining how intertwined some of these characteristics are, the proposal might emphasize the logical relationship between these two decisions. Regarding the choice for the foundation reserve regime, apparently studies have been done that suggest a large cushion of reserves in some form assists with the daily interbank clearing process and smooth over the requirement for daylight overdrafts. But does the existing tranche of required reserves not serve the same purpose in the US system?

12. In the area of this particular arbitrage issue, the proposal has a feel of circularity about it. It is anticipated that the end game will include perhaps $ 200 billion of excess reserves and reverse repos as Fed liabilities, on which the same interest rate will be paid. That will constitute an integrated floor system for interest rates. At the same time, the excess reserve component may become superfluous in the event of a return to normal liquidity conditions in financial markets. The history of pre-2008 excess reserves is that a miniscule $ 2 billion or so was required in order for the Fed to maintain the fed funds rate on target. Thus, the integrated floor system envisaged by the proposal has the potential to feed the system with the excess reserves that the GSEs and foreign banks require for purposes of sustaining their arbitrage game. This seems counterproductive. It may also make judging the effectiveness of the evolving framework more difficult during the transition period. That said, while the interbank market is currently underutilized, it seems reasonable to believe that it should slowly regain its usefulness in some proportion to the steady withdrawal of excess reserves from the banking system as the Fed exits quantitative easing.

13. The proposal seems to imply an anticipation that the repo market will not price “correctly” on its own and that the market will been tend to be (interpreted as) structurally short of high quality collateral, with the Fed always on offer of full allotment reverse repo in order to “correct” market pricing up to the IOR rate. But this definition of structurally short collateral is itself a function of the premise that the market is seen to be in collateral balance only when the repo rate converges with IOR. And that in itself is a judgement call on how the market should price repo relative to how the Fed prices IOR. This is circular. And, as noted above, it is not obvious that even with a pricing differential that IOR will not have sufficient weight to lift the entire market structure of interest rates if the Fed chooses to use IOR as the policy rate for tightening and let repo rates float with the market for the most part.

14. So how important is it that the reverse repo rate trade at the IOR rate? The financial crisis included a run to high quality collateral which resulted in an observable plunge in GC repo rates. The Fed took action to alleviate that collateral shortage as a matter of responding to a dramatic deterioration in credit markets. The question becomes, why does the Fed have to support this market as part of a permanent operating framework which presumably will exhibit more “normal” credit market conditions in the future? Why should the Fed assume a structural position in the repo market that is permanently skewed to being short of collateral? Why should it support this particular interest rate in a future permanent operating framework, apart from those unusual times when it needs to spring into emergency contingency mode?

15. We’ve also noted above that the authors themselves seem to have doubts about the evidence of a collateral shortage in markets more recently, given the increased supply of bills and bonds from expanded government deficits. This is slightly difficult to square with a proposal for a permanent facility of what amounts to price support for a particular market.

16. The paper makes reference to a minor point that seems slightly misleading or ambiguous – at least in the way that it reads. This point is not important for what the authors have put together by way of a very constructive proposal, but it is interesting nevertheless:

“Our proposed operating framework relies on the fact that the traditional analysis of the money multiplier based on so-called high-powered money (central bank liabilities) is defunct when central bank liabilities pay interest. In our view, central banks always achieved macroeconomic stabilization and low inflation by managing interest rates. Formerly, the path to controlling interest rates was the supply of non-interest-bearing currency and bank reserves. Now central banks can control interest rates directly through the rates they pay on their liabilities. This new framework severs the link between the size of a central bank’s balance sheet and inflation.”

The traditional money multiplier analysis is of course defunct in all cases, whether or not central bank liabilities pay interest, which in turn depends only on the reserve regime and how the targeted interest rate is maintained. This idea connects nicely with an interesting exchange about a year ago between Steve Randy Waldman and Paul Krugman. Regarding this subject, I disagree with the notion that for a given general level of interest rates, the economics of currency (i.e. central bank banknotes) depends in any way on the choice of the reserve regime. This aspect has nothing to do with whether the system is an asymmetric corridor, symmetric corridor, or floor. Only the targeted interest rate level matters. There is a propensity for economists to conflate the two different components of the monetary base – banks reserves and currency – even though they have very different economic characteristics. That difference has nothing to do with whether bank reserves are compensated or not compensated, which itself is a function of the choice of reserve regime. It is the target interest rate that matters, not the reserve regime and its particular IOR arrangements.

Having cobbled together this post rather quickly as an initial reaction to the authors’ proposal, I have more questions than answers at this point. The proposal is fascinating, and while the questions here reflect some push back on that proposed framework, I’d be happy to be persuaded that it’s the right way to go.


View all posts by

Leave a Reply

36 Comments on "A New Operating Framework for the Federal Reserve (Joseph Gagnon and Brian Sack)"

Notify of

3 years 4 months ago

Wow, monster post. I read the paper and agree that it’s good. I need to finish reading your comments.

I think the elephant in the room is how exactly repo rates are linked with the interbank rate. On a brief skim, it seems many of your comments revolve around this issue. The answer is still an open question in my opinion. There are many different (basic) theories out there, but I’m not sure any of them satisfy a rigorous understanding of funding markets.

The proposal mostly boils down to this for me –

“2. The authors’ proposal essentially bundles two proposals into one – the choice of a floor system for central bank reserve management and the modification of that system to incorporate non-bank participation through the full allotment reverse repo mechanism.”

with the motivation being that in the modern world, funding rates seem to be keying off the repo rate in addition to the FFR (perhaps more so the former?). The proposal is sort of a sure-fire way to make sure the Fed maintains control over short-term rates by using a catch-all approach. But as (I think) you suggest, perhaps it’s still possible to ensure that one rate follows the other. My sense is there is a lot of uncertainty around this, so the proposal attempts to get around that uncertainty by expanding the Fed’s reach (although the authors don’t really admit this).

(Part of my motivation for delving into the specifics of the Keister/Woodford/Bindseil models we’ve briefly discussed was to launch into an investigation of FFR vs. repo rates.)

3 years 4 months ago

Definitely expanding the Fed’s reach in terms of price setting.

Big change.