Chapter

Chapter 5. Monetary Operations

Author(s):
Tobias Adrian, Douglas Laxton, and Maurice Obstfeld
Published Date:
April 2018
Share
  • ShareShare
Show Summary Details
Author(s)
Darryl King and Tommaso Mancini-Griffoli

The authors gratefully acknowledge comments from Chikako Baba, Ivan Luis de Oliveira Lima, Kelly Eckhold, Claney Lattie, Romain Veyrune, Mariam El Hamiani Khatat, and Nils Maehle

Monetary operations that stabilize and align short-term market rates with the policy rate reduce liquidity risks, assisting banks with their liquidity management and pricing policies. When short-term rates are stable, discrete changes in the policy rate will have greater impact on banks’ pricing behavior as they will have more confidence that changes in the structure of interest rates will be sustained. —IMF (2015).

Prior chapters have focused on the design of institutions, the policy-making process, the construction of a consistent macroeconomic forecast, and the actions and words used to buttress transparency and credibility. This chapter focuses on implementation—the daily actions taken to translate policy decisions into tangible changes in economic incentives that underpin the effectiveness of monetary policy.

The operational framework of monetary policy comprises the operating target and monetary instruments (Figure 5.1).1 Its purpose is to do the following:

  • Ensure that the operating target of monetary policy (for example, a short-term money market interest rate) is consistently achieved
  • Strengthen the credibility of the policy signal
  • Ensure predictable and consistent liquidity conditions to market players
  • Support development of financial markets, thereby strengthening the transmission of policy signals.

Figure 5.1.The Monetary Policy Framework

Source: Authors’ construction.

The Building Blocks

The Operating Target

The operating target signals the central bank’s monetary policy stance and can be specified in various ways: as an interest rate, an exchange rate, or the growth rate of reserve money. In countries where markets and the financial sector are well developed, economies are sufficiently large and diversified, and central banks are independent from government, operational targets are usually defined in terms of interest rates. That is the focus of this chapter, in line with this book’s overall purpose of exploring the frontiers of monetary policy. The alternative operating targets are briefly discussed in Box 5.1.

To continually align short-term interest rates with a desired policy stance, operational targets are generally specified in one of two ways:

  • Explicitly, on the basis of an announced target for a market rate (or a combination of market rates). Traditionally, central banks have targeted an unsecured interbank rate because the interbank market was the most actively traded of the short-term money markets. However, unsecured interbank activity decreased since the global financial crisis with increased collateralization of short-term lending. Some central banks such as in Brazil target a secured rate, while others use a combination of rates—for example, repo and unsecured rates in Canada.
  • Implicitly, on the basis of the rate attached to a central bank instrument. Some central banks define the marginal price of the liquidity that they offer or they accept as the policy rate. Transmission to short-term money market rates and beyond is usually ensured by arbitrage.

Operating targets are generally short-term rates—seven days or less—for four reasons.2 First, central banks have more control over short-term rates, given their monopoly over bank reserves. Second, longer rates require central banks to carry more duration or credit risk. Third, targeting longer rates may induce more short-term rate volatility due to shocks to term premiums. Fourth, holding longer-term rates fixed in between policy meetings could complicate communications and be financially costly when policy rate changes are needed and widely expected.

Box 5.1.Exchange Rate and Reserve Money Operating Targets

Most countries on the frontier of monetary policy have adopted some form of inflation-forecast targeting. In particular, most have a clear inflation target and a well-developed inflation forecast to guide policymaking. The operating target is commonly defined as an interest rate, managed with regular open market operations.

But in some cases, countries may adopt a different operational framework, while still targeting inflation. Two alternative frameworks stand out.

  • A managed exchange rate framework. In some economies with exceptionally large shares of imports in the consumption basket, domestic inflation is heavily influenced by exchange rate shocks, or shocks to the price of foreign goods. In response, some countries have chosen to manage the exchange rate to stabilize prices. The Monetary Authority of Singapore (MAS), for instance, operates a managed float regime for the Singapore dollar. The operational target comprises a sloping band for the exchange rate, defined against a basket of currencies and reset periodically to ensure that it remains consistent with economic fundamentals and that it ultimately delivers price stability. The instrument of policy is foreign exchange intervention, although MAS acknowledges it does not control interest rates or the money supply. MAS has delivered consistently low and stable inflation (below 2 percent on average) since the framework’s adoption in 1981, but there are few other successful examples of this type of arrangement. Few countries are as open to trade as Singapore (where the annual value of both exports and imports is well above 100 percent of GDP) or meet the other stringent conditions necessary for success: foreign reserves must be plentiful, fiscal policy must be especially predictable and responsible, shocks to capital flows must be relatively subdued, and business cycles must be well aligned with the country or countries against which the currency is managed.
  • Monetary targeting. The operational target is the growth rate of reserve money, and the instrument is open market operations (commonly, the intermediate target is the growth rate of a broad monetary aggregate). Countries that adopt such a framework typically face one or two main constraints: very limited market development and financial sector penetration and a high degree of fiscal dominance (severe political pressure to keep interest rates low). The first constraint undermines the impact of an interest rate signal (although ensuring stable and predictable interest rates is likely an important catalyst for market development in the first place). And the second constraint limits the ability of the central bank to freely set interest rates. Often, however, frameworks targeting reserve money growth are transitory. As markets develop, changes in money velocity undermine control of inflation and the central bank establishes greater independence; as this occurs operational frameworks typically shift to targeting interest rates (see IMF [2015] for a discussion). The Bank of Tanzania, for instance, targets reserve money in its operations (and broad money as the intermediate target). It has achieved stable inflation between 2013 and 2017, close to its announced inflation objective of 5 percent. However, rigidly adhering to reserve money targets has resulted in volatile interest rates, and the Bank of Tanzania is now moving to an interest-rate-based operational framework with inflation forecasts as the intermediate target.

In the presence of other types of constraints central banks may rethink the entire monetary policy framework, including the objective and intermediate targets. For instance, countries with little institutional capacity may choose a pegged exchange rate regime, a choice that may be facilitated by a relatively closed capital account. Highly dollarized economies also face the choice between a currency board, an exchange rate peg, or, under some conditions, a strict inflation-targeting regime in the hope of rebuilding trust in the domestic currency. The presence of a high degree of administered prices does not necessarily call into question the appropriateness of inflation-forecast targeting, but it may reduce the effectiveness of monetary policy transmission.

Monetary Policy Instruments

The traditional monetary policy instruments used outside of crisis situations are well known: reserve requirements, open market operations, and standing facilities.

Generally, a central bank’s ability to guide market conditions comes from its monopoly over the supply of bank reserves, as illustrated in Figure 5.2, and, where relevant, its ability to influence the exchange rate by buying and selling foreign exchange with market participants.

Figure 5.2.A Stylized Central Bank Balance Sheet

Source: Authors’ construction.

That said, reserve requirements no longer play an important role in monetary policy implementation. This is true not only in most emerging market and advanced economies, but also in many low-income countries. Such requirements were previously used to help control monetary aggregates, as in the United States from the 1930s until the 1970s.3 More recently, as operational frameworks shifted to target interest rates, reserve requirements have mostly been used—if at all—as microprudential or liquidity-management tools. Canada, New Zealand, and Norway, for instance, have removed reserve requirements altogether.

Open market operations involve the purchase or sale of marketable instruments either through outright or repurchase (repo) transactions or the issuance of deposits. The central bank seeks to adjust the supply or the price of bank reserves, depending on the operating target, in response to forecasts of banking system liquidity.4

Standing facilities are instruments priced at fixed rates. They are available on demand at the discretion of central banks’ counterparties that are used to set bounds on overnight interest rates because no bank would lend funds at a rate lower than the central bank’s floor, whereas banks would not normally borrow at a higher rate than offered by the central bank’s ceiling.5 The width of the corridor matters, particularly when the operating target is set in the middle. If the corridor is too narrow, banks may have little incentive to trade; if it is too wide, rates may be more volatile.

Central banks must decide whether to set the policy rate in the middle or near the floor of the corridor. Having the rate in the middle often requires commercial banks to borrow, on net, from the central bank each day (a structural liquidity deficit), whereas setting the rate at the bottom is consistent with having banks investing, on net, with the central bank (a structural liquidity surplus). The central bank can structure its balance sheet to fit either environment by adjusting its longer-term assets (such as government securities and foreign assets) and liabilities (long-term, liquidity-absorbing repos and the reserve requirement, for example).

The rates at which the central bank conducts open market operations with its counterparties are transmitted to short-term money market rates. Counterparties are thus typically chosen because they are active in financial intermediation, usually banks and securities dealers. In economies with large and diverse financial sectors, it may be impossible to deal with all such entities; the central bank instead typically appoints a subset of the most active entities as counterparties, often called primary dealers. However, no one group of participants should have a distinct advantage over others, and primary dealers must be subjected to greater supervisory scrutiny to ensure that transmission of the rates is not impaired when shocks materialize.

The choice of counterparties must be adapted to the structure of the financial sector and the operational framework. For example, before the global financial crisis, the Federal Reserve Bank of New York—the implementation arm of the Federal Reserve System—dealt with about 20 primary dealers. After the global financial crisis, when structural liquidity in the banking system had increased dramatically, the Federal Reserve’s list of counterparties expanded to include about 150. This happened because not all account holders were eligible to receive remuneration on their balances at the announced rate, the interest on excess reserves (commonly known as the IOER), thereby weakening the Federal Reserve’s control over interest rates in the environment of abundant liquidity. To increase its control over interest rates, the Federal Reserve introduced a second rate attached to an overnight reverse repo facility (known as the ONRRP) and made it available to a broader range of entities upon application, subject to certain conditions. These entities were mainly money market mutual funds.

Policies governing the types of collateral accepted for open market operations and at the standing lending facility also vary across central banks and may affect monetary policy transmission. The choice of which collateral to accept may depend on the central bank’s preference for liquid assets, the availability of various types of collateral, and development objectives. Some central banks accept only highly liquid securities, whereas others accept a very wide range of assets including loans. Risk-mitigation measures (haircuts and margining) are used to contain risks within prescribed limits and to keep price distortions across asset classes to a minimum.

Four Basic Designs

Interest-rate-based operational frameworks can be designed in four general ways: (1) a mid-corridor system targeting a market rate, (2) a mid-corridor system with the policy rate attached to a central bank instrument, (3) a floor system with bank reserves remunerated at the policy rate, and (4) a tiered-floor system with bank reserves remunerated at the policy rate up to a set limit, with the balance remunerated at a lower rate. There are, however, many variations and individual frameworks do not always fit neatly into any one of these categories.

Mid-Corridor System Targeting a Market Rate

Examples: Brazil, Canada, Chile, Czech Republic, Mexico, Sweden

This approach involves offering just enough liquidity to the banking system to satisfy the operational target, usually defined as a market rate (Figure 5.3). To achieve a rate in the middle of the corridor the probabilities of the system being long (downward pressure on rates) and short (upward pressure on rates) must be symmetrically distributed around zero. As a result, the system requires accurate high-frequency liquidity forecasts of the supply versus demand for bank reserves.6 Forecasting reserve balances can be a difficult exercise. Shocks are often dominated by movements in the government’s operating account (held at the central bank). Liquidity forecasting errors result in deviations of the market rate from target, especially when the demand curve is steeper, as toward the end of reserve maintenance periods.7 Ideally, shocks can be minimized by having the government provide credible forecasts of its daily cash flows. And when shocks occur, mitigating action is possible, such as late-day fine-tuning operations.

Figure 5.3.Mid-Corridor System Targeting a Market Rate

Source: Authors’ construction.

The design of open market operations may also increase the chances of meeting the operating target. For example, setting the policy rate as the reserve price on such operations may help guide bidding.

This framework incentivizes banks to manage daily liquidity because being long or short incurs a penalty equal to half the width of the corridor. Interbank activity stimulates market development and monitoring, with some holding the view that transmission is strengthened.8

The Bank of Mexico’s framework illustrates this option well. Through an agreement with the Ministry of Finance, all government receipts and payments are known one day in advance. This virtually eliminates liquidity shocks given that transactions in the foreign exchange market are also known in advance (usually because settlement occurs two days after the transaction is agreed) and that the demand for currency in circulation is stable although risk premium changes can still affect demand for reserves and thus overnight rates. The Bank of Mexico conducts two daily operations, with cut-off rates set at the policy rate. The last of these takes place at the end of the day to account for any error in the forecasted demand for currency. Because market participants are confident that the central bank will provide the necessary liquidity, bank excess reserves are almost always zero.

Mid-Corridor System with the Policy Rate Attached to a Central Bank Instrument

Examples: The euro area (before 2008), Indonesia

Central banks can attach the policy rate to a monetary policy instrument and then rely on arbitrage to pull other short-term rates close to it. The fixed-rate instrument is generally offered on a full allotment basis to avoid market distortions that may arise from trying to control both price and quantity (Figure 5.4).

Figure 5.4.Mid-Corridor System with the Policy Rate Attached to a Central Bank Instrument

Source: Authors’ construction.

Under one version of this system the central bank engineers a liquidity deficit (although it can also be used with a liquidity surplus) forcing banks to borrow from a seven-day repo facility each week. Banks must forecast their own liquidity needs, which they reveal at the weekly auction.

In some cases, if trading activity is low, market segmentation and coordination failures may result in overbidding and produce short-term volatility in the interest rate around the policy target.9 Publishing liquidity forecasts and allowing banks to meet reserve requirements over sufficiently long periods (four to six weeks) can reduce the incentives to overbid and can encourage market activity. Furthermore, the central bank can conduct intraweek fine-tuning operations, although doing this frequently would resemble the mid-corridor system targeting a market rate, as discussed above.

Floor System with Excess Reserves Remunerated at the Policy Rate

Examples: United Kingdom and European Central Bank (after 2008)

Central banks can set the policy rate at the floor of the corridor and supply sufficient reserves to keep rates at or close to that floor (Figure 5.5).10 This “floor system” became popular during the global financial crisis as central banks increased reserves much beyond what banks would require for payments and precautionary purposes.11 Norges Bank first operated a floor system in the mid-1990s.

Figure 5.5.Floor System with Excess Reserves Remunerated at the Policy Rate

Source: Authors’ construction.

The floor system breaks the link between the price and the quantity of bank reserves (for further discussion, see Goodfriend [2002]). Within the corridor and up to a certain amount, the demand for reserves is downward sloping; banks will agree to hold more reserves for precautionary reasons if the opportunity cost of holding them—the difference between the interbank rate and the rate on the deposit facility—is lower. However, when the interbank rate equals the deposit rate, the opportunity cost of holding reserves is zero.12 The central bank can therefore increase the amount of reserves without the need to change the interest rate or cause it to move. The central bank thereby controls both the price and the quantity of reserves, as illustrated in Figure 5.5: the demand curve is flat to the right of where it meets the floor.

Ideally, the central bank would supply liquidity close to the point where the demand curve meets the floor in order to retain a degree of liquidity risk and thus spur some trading of reserve balances among banks. If supply is far to the right of this point (as when central banks engage in quantitative easing), there is little incentive to trade.

A Tiered-Floor System

Examples: Norway, New Zealand

Some central banks are concerned about the reduced level of interbank activity that results from floor systems. To attenuate such effects and to incentivize banks to manage their liquidity more actively, some central banks remunerate reserves at the policy rate up to a limit, while offering a lower rate on the balance (Figure 5.6). The limit is based on an assessment of each bank’s precautionary demand for reserves. In New Zealand, the limit is based on payment flows, the volume of liquid assets, and other balance sheet metrics. Reserves exceeding the quota are remunerated at 100 basis points below the policy rate.13 Furthermore, to implement this framework, the Reserve Bank of New Zealand had to ensure sufficient liquidity in the banking system. Because of a shortage of high-quality liquid assets, the central bank resorted to cross-currency swaps.

Figure 5.6.A Tiered-Floor System

Source: Authors’ construction.

Assessing Operational Frameworks

Three categories of criteria help identify the strengths and weaknesses of the various operational frameworks: (1) clarity, robustness, and accuracy in satisfying the operational target; (2) transmission, market development, and activity; and (3) costs and capacity requirements, along with risks to the central bank.

Clarity, Robustness, and Accuracy in Satisfying the Operational Target

To maximize effective transmission, frameworks must be clearly understood by market participants and counterparties, and must be appropriately designed to reflect their levels of financial development.14 The central bank should clearly communicate both the liquidity conditions it considers consistent with its operational target and the terms for participating in open market operations. Operational transparency reduces information asymmetries and aids pricing and transmission to longer rates.

Operational frameworks are generally clearer if they do not need to be adapted to changes in economic circumstances. Some designs may be more robust to changes in market conditions than others. Quantitative easing, for instance, forced a shift from mid-corridor systems to floor systems. But all systems allow for some flexibility through increasing the number of counterparties and broadening the range of eligible collateral if needed, without fundamentally changing the operational framework.

Transmission, Market Development, and Activity

The central bank is the heart of an economy’s financial system. Its operations therefore shape the behavior and functioning of the market and influence policy transmission.

The choice of an operational framework will impact interbank activity. Increased risks and costs of excess or insufficient liquidity to meet end-of-day payment needs will motivate interbank trading. Instrument design also impacts the incentives to trade (for example, the width of the corridor) and the pricing of securities (such as through the supply of high-quality liquid assets). The operational framework may also affect the allocation of credit.15 The smooth functioning of markets can be promoted by specific facilities (for example, securities lending). Last, counterparty and collateral policies can affect the pricing and distribution of liquidity and securities. The question, however, is the extent to which short-term interbank activity is needed for effective transmission.

Costs and Capacity Requirements and Risks to the Central Bank

Operational choices also have an impact on the costs to the central bank. High-frequency liquidity forecasts and daily operations require administrative resources and significant capacity. In addition, the choice of a framework may affect operating costs—for example, when sterilizing a large amount of liquidity. Risks will also depend on the design of operations. Complex procedures increase operational risks, and larger balance sheets induce more interest rate risk, and perhaps credit risk as well. Except in a crisis, central banks generally limit credit risk by holding government securities against their currency liabilities.

Conclusions

The interest-rate-based frameworks discussed earlier can be broadly mapped against the above criteria. Four trends appear when moving from the frameworks on the left toward the frameworks on the right in Figure 5.7 (a more detailed assessment appears in Annex 5.1).

Figure 5.7.Comparing Operational Frameworks

Source: Authors’ construction.

First, clarity, robustness, and accuracy increase. The floor system is the simplest framework, and it remains applicable in both normal and crisis times—even when hitting the effective lower bound and when needing to vastly expand reserves. Moreover, the policy rate can be achieved independent of market structure and conditions, because the central bank controls the remuneration of reserve balances. Conversely, the mid-corridor system targeting a market rate requires precise forecasting of liquidity conditions and calibration of operations to achieve the operating target, tasks that are beyond some central banks.

Second, incentives to trade diminish. Incentives for banks to actively manage liquidity are highest under mid-corridor systems targeting a market rate, when liquidity is scarce and the opportunity costs of being caught with an excess or shortage of liquidity are also high. In contrast, under floor systems, incentives for trading and monitoring in short-term money markets and liquidity risks are reduced.16

Third, operational costs diminish when moving away from the mid-corridor system targeting a market rate, especially because there is less need for liquidity forecasting. This is true even in mid-corridor systems attached to a central bank instrument and more so in floor systems, which do not require accurate liquidity forecasts, frequent operations, or well-functioning markets.

Finally, the larger balance sheet of floor and tiered systems impose higher risks to central banks. Asset price fluctuations affect central banks’ equity positions, and although most central banks do not mark their balance sheets to markets, perceived losses can bring about political scrutiny and undesirable pressures.

How central banks decide to balance these trade-offs depends on country circumstances and constraints. The importance of decisions about clarity and simplicity, for instance, are functions of the central bank’s ability to communicate, and of the sophistication of counterparties.

The importance of trading incentives will likely depend on the relative priority of financial market development and the extent to which interbank activity matters for transmission (Potter 2016). Some views have changed since the global financial crisis, at least in countries where markets are well-developed. Active short-term interbank markets may be less important for transmission and interbank monitoring than previously thought.17 However, further investigation is necessary to reach firmer conclusions, and data are still being collected as central banks begin normalizing rates.

Market development is likely to be a relatively more important objective in lesser developed markets, favoring approaches that incentivize interbank activity, such as mid-corridor systems. An analogous situation is that when central banks step back from foreign exchange intervention, participants are incentivized to actively manage and distribute risks, including through product innovation, such as new hedging instruments.

There are competing views on the appropriate size of central banks’ balance sheets. Buiter and others (2017) emphasize that the optimal size is “unknown and probably unknowable.” Bindseil (2016) favors a small balance sheet to signal well-functioning markets but does not account for the possibility that markets may not function well or that balance sheets may be large for other policy or legacy reasons. Greenwood, Hanson, and Stein (2016) argue for retaining a large balance sheet to control financial stability risks outside of the banking system.18 The political economy risks associated with a large balance sheet may also be a relevant consideration in some jurisdictions.

Other constraints will affect central banks’ choice of framework. For example, central banks may be forbidden from remunerating reserve holdings of all entities or from issuing securities. There also could be legal restrictions on the use of other instruments (such as reserve requirements) that undermine central banks’ ability to provide a buffer against liquidity shocks. And the availability of liquid government securities could limit the size of a central bank’s balance sheet. Finally, weak transmission is not necessarily a reason for changing operational frameworks. Transmission may be affected by external factors, such as uncompetitive banking systems with high nonperforming-loan ratios and poor creditor rights, or information asymmetries leading to segmented money markets. Nevertheless, appropriate design, communication, and conduct of monetary operations are prerequisites for effective policy transmission.

Annex 5.1
Table 5.1.1.Assessing the Operational Frameworks
Mid-Corridor Targeting a Market RateMid-Corridor Policy Rate Attached to a CB Instrument (such as a seven-day repo)Tiered FloorFloor
Other Operational FeaturesAveraged Reserve Requirement (daily operations)Averaged Reserve Requirement (ad hoc intraweek operations)Quotas Set on Payments Activity and Balance Sheet MetricsLiquidity Supplied Close to the Amount Required to Keep Rates at the Floor
Clarity, Robustness, and Accuracy in Satisfying the Operational Target
Achieves the operational target (where liquidity forecasting is difficult or markets do not function smoothly)1233
Is durable across different financial states of the world1123
Transmission, Market Development, and Activity
Provides incentives to trade in the short-term money market3221
Has minimal impact on credit allocation3321
Costs and Capacity Constraints, and Risks to the Central Bank
Minimizes the operational costs (liquidity forecasts and frequency of operations)1223
Minimizes financial risks (credit and interest rate risks)3321
Source: Authors’ construction.Notes: 1 = not supportive, 2 = neutral or unclear, and 3 = broadly supportive. CB = central bank.
Source: Authors’ construction.Notes: 1 = not supportive, 2 = neutral or unclear, and 3 = broadly supportive. CB = central bank.
References

    BeaupainRenaud and AlainDurré. 2016. “Excess Liquidity and the Money Market in the Euro Area.” Journal of Macroeconomics47 (2016) 3344.

    • Search Google Scholar
    • Export Citation

    BechMortenL.ElizabethKlee and ViktorsStebunovs. 2012. “Arbitrage, Liquidity and Exit: The Repo and Federal Funds Markets Before, During, and Emerging from the Financial Crisis.” Finance and Economics Discussion Series 2012–21. Board of Governors of the Federal Reserve System.

    • Search Google Scholar
    • Export Citation

    BernhardsenTom and ArneKloster. 2010. “Liquidity Management System: Floor or Corridor?Norges Bank Staff Memo 4/2010Oslo.

    BindseilUlrich. 2016. “Evaluating Monetary Policy Operational Frameworks.” Paper prepared for the Federal Reserve Bank of Kansas City’s Jackson Hole Conference.

    • Search Google Scholar
    • Export Citation

    BuiterWillemEbrahimRahbariKim D.Jensen and CesarRojas. 2017. “Why Does the Fed (or any Central Bank) Wish to Shrink Its Balance Sheet?Citi Research Multi-Asset Global Global Economics View August 9 2017.

    • Search Google Scholar
    • Export Citation

    ChristensenJensH. E. and James M.Gillan. 2017. “Does Quantitative Easing Affect Market Liquidity?Federal Reserve Bank of San Francisco Working Paper.

    • Search Google Scholar
    • Export Citation

    Committee on the Global Financial System Markets Committee (CGFS). 2015. “Regulatory Change and Monetary Policy.” CGFS Papers 54.

    Federal Reserve. 2016. “Minutes of the Federal Open Market Committee November 1–2, 2016.” Board of Governors of the Federal Reserve System.

    • Search Google Scholar
    • Export Citation

    GoodfriendMarvin. 2002. “Interest on Reserves and Monetary Policy.” FRBNY Economic Policy Review (May).

    GraySimon. 2011. “Central Bank Balances and Reserve Requirements.” IMF Working Paper 11/36International Monetary FundWashington, DC.

    • Search Google Scholar
    • Export Citation

    GreenwoodRobinSamuel G.Hanson and Jeremy C.Stein. 2016. “The Federal Reserve’s Balance Sheet as a Financial-Stability Tool.” 2016 Economic Policy Symposium Proceedings. Jackson Hole: Federal Reserve Bank of Kansas City.

    • Search Google Scholar
    • Export Citation

    International Monetary Fund (IMF). 2015. “Evolving Monetary Policy Frameworks in Low-Income and Other Developing Countries.” IMF Policy PaperOctober.

    • Search Google Scholar
    • Export Citation

    PotterSimon. 2016. “Discussion of ‘Evaluating Monetary Policy Operational Frameworks by Ulric Bindseil’.” Remarks at the Economic Symposium at Jackson Hole Wyoming.

    • Search Google Scholar
    • Export Citation

    VeyruneRomainGuidoDella Valle and ShaoyuGuo. Forthcoming. “Relationship between Short-term Interest Rates and Excess Reserves in the Euro-system: A Logistic Approach.” IMF Working PaperInternational Monetary FundWashington, DC.

    • Search Google Scholar
    • Export Citation
1

For the purposes of this section the term instrument means monetary instrument and is distinguished from the reference in other chapters where instrument refers to a policy instrument in a broader macroeconomic sense.

2

The Swiss National Bank is often cited as using an operating target longer than seven days. It sets a 1 percentage point range (currently –1.25 to –0.25) on the three-month Swiss-franc London interbank offered rate, and generally aims to keep the rate in the middle of the range. In managing this rate, the Swiss National Bank operates in the short-term repo market.

3

In the United States, reserve requirements were initially used as a prudential tool to provide backing to the notes issued by private banks. Later refinements that permitted banks to meet reserve requirements on average over a given period (called the maintenance period) facilitated liquidity management in the face of liquidity shocks; see Gray (2011).

4

Open market operations also serve purposes other than the fine-tuning of liquidity conditions. Transactions of longer-term assets are used to manage trends in bank reserves arising from changes in the demand for currency in circulation, the amount of net foreign assets held, and other assets and liabilities on the central bank balance sheet.

5

A central bank may not always be able to cap interbank rates at the ceiling of the interest rate corridor because banks may have insufficient eligible collateral or may fear stigma attached to borrowing from the central bank. In such cases, banks may prefer to pay higher rates to demonstrate continued market access, even though doing so may still send a negative signal. The situation in the United States is slightly different than described above; the federal funds rate does trade below the interest rate that the Federal Reserve pays depository institutions (namely, the interest on excess reserves—IOER) because not all entities that trade in the federal funds market are eligible to receive the IOER.

6

The rationale here also applies when targeting a secured or composite rate.

7

Reserve averaging allows banks to meet their reserve target over the maintenance period rather than daily. Banks therefore do not need to respond immediately to liquidity shocks during the maintenance period thereby resulting in a fatter demand curve.

8

Where markets do not function smoothly there may be a wide range in traded rates, with perhaps the central bank intermediating between surplus and deficit banks (through the standing facilities). Calibrating monetary operations in such circumstances is challenging.

10

Conceptually, the central bank can also operate a ceiling system under which there is a structural shortage of liquidity that forces banks to borrow from the lending facility each day. Such a system, once common, is rarely used now because it imposes liquidity risks that may undermine financial stability, given the possibility that banks may need to meet their payment obligations when high-quality collateral may be in short supply. One potential benefit of this system is that it may overcome the reluctance of banks to borrow from the central bank because of stigma, which can exacerbate problems in otherwise sound banks. Systematically forcing the banking system to borrow from the central bank each day may alleviate such stigma.

11

The Federal Reserve was granted the legal right to remunerate reserves from October 2008 upon passage of the Financial Services Regulatory Relief Act of 2006. Initially this provision was to take effect in October 2011 but that date was brought forward to October 2008 as the global financial crisis unfolded.

12

In reality, the interbank rate may never quite reach the floor of the corridor given that deposits at the central bank are risk-free whereas placements with banks involve some credit and operational risks.

13

Bindseil (2016) models this approach, showing that smaller quotas (smaller amounts remunerated at the policy rate) increase money market turnover and interest rate volatility.

14

Central bank operations are generally straightforward, but on occasion financial options or algorithms have been used (for example, Bank of England collateralized operations).

15

For instance, the Federal Reserve focused on the allocation of credit. The Federal Open Market Committee’s September 2014 Policy Normalization Principles and Plans said, “The Committee intends that the Federal Reserve will, in the longer run, hold no more securities than necessary to implement monetary policy efficiently and effectively, and that it will hold primarily Treasury securities, thereby minimizing the effect of Federal Reserve holdings on the allocation of credit across sectors of the economy.”

16

In addition to changes in operational frameworks, changes in banking regulation have undermined activity in short-term money markets. The liquidity coverage ratio requires banks to hold high-quality liquid assets against net outflows within a 30-day window; interbank borrowing falls into this category. A ratio limiting bank leverage (the supplementary leverage ratio) also discourages money market activity (CGFS 2015).

17

On monitoring, Bernhardsen and Kloster (2010) investigate the question of interbank monitoring, and note that most interbank activity is at very short maturities, so lenders only had incentives to consider counterparty solvency over a few days, while an assessment of solvency over a longer period is more relevant to financial stability. On transmission, evidence points to continued effectiveness, though estimates are subject to significant uncertainty. Federal Open Market Committee participants suggest the floor system run by the Federal Reserve since the global financial crisis is “effective in enabling interest rate control across a wide range of circumstances” (Federal Reserve 2016). However, Bech, Klee, and Stebunovs (2012) find some slippage in transmission; they conclude that “pass-through from the federal funds rate to the repo deteriorated somewhat during the zero lower bound period.” Beaupain and Durre (2016) and Christensen and Gillan (2017) actually find that market liquidity increased in recent periods of excess liquidity, both in euro area money markets, and United States’ bond markets, but only as a result of active large scale asset purchases by the central bank. As CGFS (2015) emphasizes, though, empirical work to date has had trouble dissociating the effects of excess liquidity on transmission from those of regulation, bank specific features, and risk premiums associated with the crisis.

18

Greenwood, Hanson, and Stein (2016) argue for the Fed to use its balance sheet by supplying safe assets (that is, overnight reverse repurchase transactions) to lean against private sector maturity transformation. Reducing the scarcity of safe assets also reduces the incentives for financial intermediaries to fund on a short-term basis. They note that by impacting market-determined spreads on interest rates, regulated banks and unregulated shadow banks are affected. Such a measure they say, “gets into all the cracks.”

    Other Resources Citing This Publication