From conventional to unconventional monetary policies

FROM CONVENTIONAL TO UNCONVENTIONAL MONETARY POLICIES: THE CONSEQUENCES OF THE MARKET-MAKER-OF-LAST-RESORT ROLE Gabriel A. Giménez Roche†, Nathalie Janson‡

Abstract: The increasing importance of commercial banks’ noninterest-earning activities rendered conventional monetary policy relatively obsolete in the contemporary economy. It has been a long time since commercial banking was limited to bank lending activities. Indeed, it is now present in practically all segments of the financial markets. Although this opened new outlets to improve performance, it also exposed commercial banking to greater volatility. In order to meet their stability goals, central banks around the world have now to intervene in all those markets where commercial banking has extended its activities. This effectively changed central banking from having a lender-of-last-resort role to that of a market-maker-of-last-resort. This paper provides a survey of contemporary practices in central banking, and exposes the consequences of so-called unconventional monetary policy to the world economy. The main point here is that the emergence of the market-maker-of-last-resort role actually makes the economy weaker instead of strengthening it. JEL Codes: B53, E43, E44, E51, E52, E58 Keywords: monetary policy, qualitative easing, quantitative easing, unconventional policies

† Groupe ESC Troyes en Champagne. Email: [email protected] ‡ Neoma Business School. Email: [email protected]

1

1. Introduction The age of contemporary central banking was unquestionably inaugurated with the creation of the Federal Reserve System in 1913. The predecessors of contemporaneous central banks (henceforth CBs) were financial institutions holding some form of State-granted privilege on national currency issue. Notwithstanding this, they did not officially assume the role of lenderof-last-resort (LoLR), though they did take on that role informally and occasionally (Kindleberger, 1984). Only at the beginning of the twentieth century did central banking take on its distinctive status as LoLR and guardian of monetary stability, features that are described in all contemporary macroeconomics textbooks. Nevertheless, the increasing development and complexity of financial markets have changed the role of central banking. At least in Western developed countries, contemporary central banking has evolved beyond the LoLR role. It now assumes a market-maker-of-lastresort (MMoLR) role, meaning that the subjects of monetary policy are no longer limited to commercial banks (Bastidon Gilles, et al., 2012; Cecchetti & Disyatat, 2010; Mehrling, 2011). Indeed, central banks were dealing directly with financial markets even before the 2008 subprime crisis made quantitative easing commonplace (Borio, 1997; Borio & Disyatat, 2010). Moreover, CBs also deal in return-driven asset management like any other commercial bank (Borio, et al., 2008). CBs even helped trigger the growth of the derivatives market in order to produce a greater diversity of financial products (Baba, et al., 2005; Borio, et al., 2008). Although before the 2008 crisis, central banking tended to target inflation, foreign exchange, and interest by piloting short-term interest rates, since then the situation has definitively changed. Today, central banks have added financial stability as one of their targets

2

alongside traditional ones such as inflation (Joyce, et al., 2012). The LoLR role has thus been redefined: At its core, the objective of the LOLR is to prevent, or at least mitigate, financial instability through the provision of liquidity support either to individual financial institutions or to financial markets. The underlying premise is that shortages of liquidity, by which we mean the inability of an institution to acquire cash or means of payment at low cost, can lead to otherwise preventable failures of institutions that then result in spillover and contagion effects that may ultimately engulf the financial system more broadly with significant implications on the real economy. By signaling its willingness and ability to act decisively, the central bank demonstrates its intention to restore confidence in the system by avoiding “fire sales” of assets and supporting market functioning. (Cecchetti & Disyatat, 2010, pp. 29-30)

This is understandable, since the involvement of investment funds and broker-dealers in creditgranting is undeniable, particularly in the United States but in other countries as well (Adrian & Shin, 2009). Furthermore, post-2000 global finance is increasingly characterized by financial disintermediation resulting in market and institutional overlapping, which positively impacts the growth of over-the-counter markets (Bastidon Gilles, et al., 2012). Given the extensiveness of financial markets today, it is fair to say that the reach of monetary policy is such that central banking is increasingly quasi-fiscal, as in practice, it indirectly subsidizes investment in both goods and financial markets (Buiter, 2010). This paper makes two contributions. First, based on the latest literature on current developments in central banking, we present a survey of both conventional and unconventional CB monetary policies (hereafter, CMPs and UMPs). This survey shows on the one hand that CMPs are more varied and complex than merely outright purchases or sales of bonds in openmarket operations (OMOs). On the other hand, many of what are seen as UMPs are more conventional than initially thought. In fact, we will explain that UMPs are unconventional not so much because of the volume of the operations involved, but because of the markets targeted and 3

the context of the operations in question—i.e., zero-lower bound, disrupted financial markets (Joyce, et al., 2012). Our second contribution is to show that the MMoLR role and its UMPs weaken rather than strengthen the economy. UMPs artificially sustain asset prices to prevent a deleveraging process resulting in a halt of investment. However, this artificial buoying of asset prices widens the gap between market and fundamental asset valuation, while leaving the economy at a cliffhanger. If UMPs unwind, then the artificially-buoyed asset prices collapse forcing companies into deleveraging, if not liquidation. If CBs persist with UMPs, then investment is hesitant in fear of an eventual abandonment of UMPs, which entails an increasing dependence on CB activism.

2. Conventional Policies and the LoLR Role of Central Banks Most modern CBs around the world set their main policy objectives as, in varying degrees, price stability, long-term economic growth, and full employment (Dominguez, 2006). They usually monitor these goals by means of some rule or reaction function, which they feed with information on chosen goals and policy stance indicators—the foremost example of which is Taylor rules, particularly regarding inflation-targeting (Bernanke & Mishkin, 1997; Taylor, 1993). These indicators frequently include the evolution of exchange rates, long-term interest rates, volume of broad money (i.e., base + bank money), asset prices, and so on (Borio, 1997, pp. 12-13). The adoption of rules and reaction functions does not necessarily imply mechanical reactions to variations in the adopted set of indicators. They are just a guideline for policy reaction (Bernanke & Mishkin, 1997; Taylor, 1993; 1999). In this manner, central bankers can always resort to discretionary policy-making while adopting explicit priority rules whenever other intermediary targets conflict (Bernanke & Mishkin, 1997, p. 5).

4

Nevertheless, the aforementioned goals and rules set by central banks apply to their longterm strategy. Moreover, CBs are unable to intervene directly in the policy indicators that monitor this strategy. In fact, the operational implementation of CB monetary policy reveals that the most obvious permanent goal of central banking is not really price stability or economic growth. This goal is the appropriate provision of liquidity to the commercial banking sector—as conceived by Walter Bagehot (1873), the father of contemporary central banking.

2.1 Deposit banking and liquidity types Commercial banking is essentially deposit banking. Deposit institutions can create deposits upon reception of money from their clients or by granting credit in the form of deposits. In both cases, deposits constitute of claims on narrow money, that is, CB issued currency. All commercial bank services are based on deposits. These services can be either interest-earning or noninterestearning activities (DeYoung & Rice, 2004). Noninterest income consists of fees for bank services on deposits (e.g., money transfers, issue of means of payment, account management), as well as income originating from their investment banking (commissions), brokering (spread, fees), and insurance (fees, capital gains) services. In addition to noninterest income, banks earn interest charged on credit they grant as deposits. Consequently, the more credit banks grant, the more deposits are created, which increase both their interest and noninterest income. However, this generates a maturity mismatch in their balance sheets between the term maturity of the credit aspect of the transaction and the potential immediate maturity of the deposit aspect of the same transaction. This implies a fiduciary character to the credit that commercial banks grant, that is, banks create more deposits than they dispose of bank reserves of CB currency to back them.

5

The idea of a money multiplier of credit then comes to mind. But one should not think that this multiplier implies a sequential causality between a bank’s money reserves and deposit multiplication. Indeed, a bank does not need to wait for actual money deposits to flow into its vaults to grant credit: In fact, the level of reserves hardly figures in banks’ lending decisions. The amount of credit outstanding is determined by banks’ willingness to supply loans, based on perceived risk–return trade-offs, and by the demand for those loans. The aggregate availability of bank reserves does not constrain the expansion directly. (Borio & Disyatat, 2010, p. 77)1

Indeed, the more developed an economy’s money market, the easier it is for banks to find liquidity ex post deposit creation. However, in addition to the maturity mismatch and fiduciary problems, depositors’ payment operations drain their deposits and exert liquidity pressure on banks. This entails constant bank reserve needs so the banks can maintain their liquidity vis-à-vis their depositors. This need is not always satisfied by interbank loans, or REPOs (repurchase agreement operations), thus triggering CB involvement. In fact, three types of liquidity are the object of CB bank operations: central bank liquidity, market liquidity, and funding liquidity (Cecchetti & Disyatat, 2010, pp. 30-31). Central bank liquidity refers to bank reserves held at the CB. These reserves are more or less consumed by the regular operational needs of depositors draining the bank’s reserves. Commercial banks need to maintain a certain level of bank reserves at all times to meet their depositors’ operational needs while also maintaining their level of credit creation.

1

This assessment is echoed in the speech of Vítor Constâncio, vice-president of the ECB, in 2011, when commenting on the sequence between loan creation and funding for that creation: “In reality the sequence works more in the opposite direction with banks taking first their credit decisions and then looking for the necessary funding and reserves of central bank money” (Constâncio, 2011). Also based on this view, Goodhart (1995) goes as far as to state that instead of a “credit multiplier,” it would be more appropriate to speak of a “credit divisor.”

6

Market liquidity concerns the stable marketability of large volumes of assets. In other words, the market for an asset is liquid when it can absorb large movements in the supply of and demand for that asset at any time, and without substantial price volatility. This is fundamentally linked to funding liquidity, which implies a bank’s ability to raise large quantities of cash via asset sales or borrowing. If market liquidity refers to the marketability of assets, funding liquidity in its turn refers to a bank’s reputation and ability to trade marketable assets—and hence overlaps with a bank’s solvency. It is essentially a bank’s capability to find market counterparties. Conventional shortages of CB liquidity occur whenever interbank and money market operations are insufficient to provide liquidity reserves for the overnight needs of commercial banks. The CB uses two instruments to deal with this problem: OMOs and standing facilities (see Table 1). The latter are overnight operations directed at specific institutions for their operational daily needs or whenever there is a distribution problem with CB liquidity—that is, when the market is well provided in CB liquidity but some over-accumulate and others under-accumulate. They usually—though not always—involve collateralized borrowing and lending of CB liquidity (Cecchetti & Disyatat, 2010; Lenza, et al., 2010). OMOs are undertaken at regular though more extended intervals than standing facilities, usually weekly. Consequently, they have longer maturities than standing facilities, ranging from one week to one month. They correct less urgent disequilibria that build up in the money market as a whole during that interval (cf. Lenza, et al., 2010, p. 304). Since OMOs aim at sustaining the dynamism of money markets, CBs trade in the open market with any willing financial institution inclined to trade with liquidities in the interbank market (Borio, 1997; Borio & Nelson, 2008). Outright sales and purchases of assets are rare for CMPs, since the objective is simply to adjust the provision of CB reserves in the

7

market (Cecchetti & Disyatat, 2010). Given the scheduled and less urgent character of OMOs, REPOs are discounted at the main refinancing rate, which is lower than the discount, or marginal lending, rate of standing facilities. Table 1: Central bank monetary policy operations:

Other instruments of monetary policy regulating the reserves market (see Table 2) include setting the level of reserve requirements, if any; the choice of eligible counterparties for CB operations; and eligible collateral securities for those same operations (Borio, 1997; Borio & Disyatat, 2010; Borio & Nelson, 2008). The level of reserve requirements directly affects the dependence of the banking sector on the bank reserves market. The greater the requirement, the more likely banks are to resort to the bank reserves market and vice-versa. The choice of eligible collateral for CB refinance operations—that is, OMOs and standing facilities—influences the market liquidity of the concerned securities but also the risk exposure of the CB’s counterparties. An eligible security is a prime choice for CB refinance and is therefore considerably sought after by the CB’s counterparties. Moreover, CBs incite higher risk exposure by accepting riskier securities as 8

eligible, and vice-versa. Finally, the choice of counterparties signals how much the CB wants to influence both market and funding liquidity. The greater the number of counterparties apart from commercial banks, the greater the competition for CB liquidity, but also the better the refinance possibilities for commercial banks in the money market. Table 2: Central bank monetary policy tools

2.2 Liquidity shortages and CMPs CMPs target CB liquidity shortages, as these are mostly concentrated on the bank reserves market (i.e., interbank market). Indeed, the power of the CB in this market is at its strongest, because of its absolute monopoly of bank reserve supplies. Commercial banks need to keep a minimum level of bank reserves at all times to cover their daily operational drains. Since this minimum level sustains the fiduciary deposits previously created by the commercial banks, they do not lend those reserves and avoid overextending fiduciary credit without additional reserves.

9

Therefore, their demand for that minimum level of bank reserves is practically inelastic (Borio & Disyatat, 2010; Friedman & Kuttner, 2010). The CMPs aimed at avoiding operational CB liquidity shortages at the end of the day consist of standing facilities (Cecchetti & Disyatat, 2010). These take the form of REPOs targeting the interbank overnight rate by signaling a policy rate level (Borio & Disyatat, 2010; Disyatat, 2008). Given the inelasticity of demand for bank reserves at the minimum level necessary to cover operational drains, CBs can set that policy rate at any level they want, low or high, as long as they supply just that minimum level. CBs thus can set their policy rate within a wide range. The floor rate of this range is the bank-reserve remuneration rate —if any—and the ceiling rate is the maximum rate banks can pay given their solvency. This means that a given level of the policy interest rate can co-exist with different levels of bank reserves circulating in the market, and vice-versa (Borio & Disyatat, 2010; Disyatat, 2008; Friedman & Kuttner, 2010). Both large and small supplies of bank reserves can be compatible with either low or high levels of interest rate. Nevertheless, holding reserves at the CB entails an opportunity cost for banks: the market remuneration they could have obtained for those funds (Borio & Disyatat, 2010). This is particularly true when bank reserves are not remunerated or are remunerated at a lower level than the policy rate. In this case, commercial banks inject to the interbank market any excess supply of central bank reserves above the minimum needed, instead of keeping them as deposits at the CB. Thus, the market rate is driven down to the remuneration rate and hence below the policy rate. The demand for bank reserves becomes very elastic. On the other hand, an insufficient supply of bank reserves by the CB results in skyrocketing interest rates.

10

An insufficient supply of bank reserves can be easily remedied by the CB. If the commercial banks’ need for CB liquidity pressures the interbank market, the CB can simply increase its supply until it reaches the minimum level required by commercial banks. An excess supply, however, demands a more roundabout operation. The CB sterilizes the excess supply either by supplying remunerated reserves at an attractive rate or by selling some of its assets to resorb the excess, thus contracting its balance sheet (Borio & Nelson, 2008). From the above, we may derive a few preliminary conclusions. First, OMOs in CMPs, contrary to what the textbooks tell us, are far less important than standing facilities for daily piloting of the market interest rate. OMOs do not target the urgent daily operational needs of specific banks but correct market disequilibria at more extended intervals, including other financial institutions that participate in the money market. In any case, central banks always prefer REPOs to outright operations in CMPs, whether they consist of OMOs or standing facilities. This is consistent with the goal of long-term price stability, so CBs do not need their balance sheet to increase continuously. Second, standing facilities follow commercial banks’ creation of deposits. REPOs are short-term operations (e.g., one night, one week, one month), and represent a cost to commercial banks (i.e., the marginal lending or discount rate). Moreover, as mentioned earlier, holding bank reserves entails an opportunity cost. Therefore, commercial banks are unlikely to incur these costs before securing debtor-depositors. Finally, given the inelasticity of the demand for bank reserves when CB liquidity shortages are involved, the CB’s balance sheet needs not vary significantly in size to accommodate CMPs (Borio & Disyatat, 2010; Disyatat, 2008; Friedman & Kuttner, 2010). The fact that standing facilities and OMOs are undertaken through short-term REPOs only reinforces

11

this point. In this manner, falls in the policy rate need not translate into monetary expansions and price inflation, and vice-versa. The limitations of CMPs appear when they involve funding market liquidity shortages. Funding liquidity shortages involve a lack of willing counterparties in the interbank market due to the perceived insolvency of a commercial bank. If the actual solvency situation of the bank does not corroborate that perception, then, a standing facility operation at the end of the day might suffice to address a one-off funding liquidity problem. But, in the case of a more acute and generalized liquidity shortage—that is, beyond end of day and involving more than one bank— CBs prefer a longer term emergency standing facility operation (Cecchetti & Disyatat, 2010). Insolvency becomes thus secondary to systemic risk in the list of the CBs’ priorities. The fact is, as Adrian & Shin (2009) point out, that banks are not limited to interbank refinancing, but can also look for REPOs with broker-dealers and funds such as MMMFs (Money Market Mutual Funds) and investment banks. It is therefore extremely important to maintain funding liquidity to sustain the supply provided by broker-dealers and mutual funds. If these institutional players are negatively affected by failing banks, their balance sheets will be compromised, resulting in the risk of a market liquidity shortage (Adrian & Shin, 2009). When funding liquidity shortages entail market liquidity shortages, CMPs prove to be insufficient (Cecchetti & Disyatat, 2010). Short-term operations are no longer feasible because of insolvency. Lending funds at a cost would only add more strain to already over-indebted banks. Furthermore, as market liquidity shortages are actually generalized funding liquidity shortages, the CB will be forced to expand its list of eligible collateral and counterparties.

3. UMPs and the MMoLR Role of Central Banks

12

3.1 Noninterest bank activities and the necessity of UMPs CMPs were conceived as a remedy for liquidity shortages originating from interest-earning bank lending and its ancillary noninterest-earning activities (i.e., means of payment services). The LoLR role does not encompass the other noninterest-earning activities of commercial banks— namely, investment banking, brokering, and insurance—since they are not ancillary to bank lending. However, this neglects the fact that they are indirectly connected to bank lending through deposits. Losses in these noninterest-earning activities can jeopardize deposits and consume bank reserves. This situation leads CBs to intervene in markets other than the bank reserves markets (Mehrling, 2011, p. 26ff.). The 2008 financial turmoil showed that CMPs are insufficient during crises precisely because of the above. In such a situation, eligible collateral might be lacking, or worse still, quality collateral becomes extremely scarce in all kinds of market transactions (Cecchetti & Disyatat, 2010). Indeed, as the crisis unfolds it rapidly corrodes asset value. This increasingly exposes institutions to acute liquidity shortages as their collateral loses value. Moreover, the noninterest-earning activities of commercial banks have been growing ever since the 1980s (DeYoung & Rice, 2004, pp. 34-35), further exposing these banks to market liquidity shortages. If market liquidity crashes, then these activities suffer losses that exert additional pressure on the banks’ liquidity needs as their capital becomes insufficient to cover their losses. In 2008, these losses were amplified by the growth of another source of noninterest income that is equally sensitive to market liquidity: shadow banking (Noeth & Sengupta, 2011; Pozsar, et al., 2010).2 Meanwhile, liquidity-rich institutions reassess market and counterparty risks, and abstain from

“A shadow bank is an institution or bank-sponsored special-purpose vehicle that has persuaded its customers that its liabilities can be redeemed de facto at par without delay (or can be traded as if they will be executed at par without fail at maturity) even though they are not formally protected by government guarantees.” (Kane, 2012, p. 2) 2

13

supplying a failing market with their surpluses. Their withdrawal reinforces the downward spiral and transforms an acute funding liquidity crisis into a generalized market liquidity one. However, many of the financial and non-financial institutions in need are not eligible for CB refinancing. This is equally true of the noninterest-earning activities of commercial banks, whether shadowy or not, which are not usually covered by CB assistance and government deposit insurance. If the CB is to fulfill its role, it must reactivate entire markets beyond the bank reserves market and not just specific institutions. Therefore, it must assume the role of financial intermediary (Cecchetti & Disyatat, 2010, p. 35; Mehrling, 2011, p. 80ff.). It will radically alter its balance sheet in order to liquefy markets, different kinds of assets, and institutions. The CB might even provide borrowers and investors with direct funding to achieve its goals. The above portrays the situation in which UMPs become an alternative to CMPs. The latter aim at temporary, one-off refinancing of specific institutions. Standing facilities target minor CB liquidity problems of specific banks, while OMOs correct distribution disequilibria in the bank reserves market. Market liquidity is not a main goal because CMPs supposedly prevent markets from becoming illiquid by tending to CB and funding liquidity shortages. The CB rarely needs to change the composition and size of its balance sheet significantly. Counterparty and collateral eligibility is limited to actors in the bank reserves market and its underlying instruments.

3.2 Qualitative and quantitative easing UMPs are characterized by three ways of using CB tools: 1) Considerable changes in the composition and size of the CB balance sheet, 2) Enlarged eligibility of counterparties and collateral, and 3) broader monetary policy goals (Lenza, et al., 2010, p. 299).

14

The first aspect is perhaps the most visible one. All CB policies are balance sheet policies. As such, there is always a change in either the composition or size of the CB’s current account balance. However, a change in composition need not translate into a change in the size of the balance sheet. The CB can sterilize injections of liquidity from its refinance facilities that it considers excessive by selling other assets, thus rebalancing its asset composition. Alternatively, it can compensate monetary contractions in the bank reserves market by acquiring other assets in other markets. If no sterilization or compensation takes place, then the change in composition also entails a change in the size of the CB’s balance sheet. As long as these balance sheet modifications concern the bank reserves market, they are still considered conventional. A CB is expected to discount interbank instruments and liquefy the bank reserves market, however massive those actions are. Things become unconventional when the CB discounts instruments beyond the bank reserves markets (cf. Borio & Disyatat, 2010, p. 59). This takes us to the second aspect of UMPs. During a crisis, losses might be greater than the reserve liquidity that eligible financial institutions can raise by means of their stock of eligible assets. If these losses spill over to other institutions that provide refinance or collateral to banks (e.g., money-market mutual funds, insurance companies, investment banks), then the economy is clearly threatened by an acute funding liquidity shortage. If the CB is to live up to its LoLR role, it must broaden the eligibility of both collateral and counterparties. By broadening collateral and counterparty eligibility, the CB accepts to rescue other markets beyond the bank reserves market (Lenza, et al., 2010). This has particular consequences on the composition and size of the CB’s balance sheet. When extending discount to instruments from the money, capital, and derivative markets, the CB increases its risk-exposure and deteriorates the quality of its balance sheet. Instead of holding

15

short-term, highly-liquid securities, the CB will hold now longer term, less liquid ones (Bagus & Schiml, 2009; Bagus & Howden, 2009; Buiter, 2008). Consequently, the maturity of the REPOs undertaken through standard facilities and OMOs must also be extended. The CB thus becomes more exposed to credit risk than before. The change in composition of the CB’s balance sheet resulting from broadening counterparty and collateral eligibility can be termed qualitative easing (Bagus & Schiml, 2009; Buiter, 2008; Farmer, 2012).3 Nevertheless, it is unlikely that a CB will resort solely to composition changes when broadening counterparty and collateral eligibility during a crisis. If transaction losses are greater than liquid collateral availability—a likely scenario when losses accumulate from commercial bank noninterest-earning activities— the CB must obviously increase the size of its balance sheet (at least for a while). This size change qualifies as a UMP because the instruments discounted, the counterparties involved, and the maturities granted by the CB all go beyond the bank reserves market. Consequently, the CB also moves into quantitative easing. Can a CB undertake quantitative easing without resulting in any qualitative easing of its balance sheet? It depends. If the CB increases its balance sheet without modifying the asset composition, then quantitative easing has no qualitative impact. It could increase its balance sheet massively while maintaining a conventional asset composition—that is, highly liquid money market paper. Nevertheless, this is not unconventional in itself, since the CB is simply concentrating on the bank reserves market on a massive scale. However, if the CB first undertakes purely qualitative easing of its balance sheet and only subsequently quantitative easing, then, there is no qualitative impact because balance sheet deterioration had already taken place (Lenza, et al., 2010). In any case, it becomes increasingly difficult for a CB to

3

A change in balance sheet composition involving liquid and short-term securities is not qualitative easing, as risk and term do not increase.

16

maintain quality if quantitative easing continues indefinitely. At a given moment, the marginal quality of the collateral absorbed by the CB must deteriorate. No given quality is in infinite supply. Therefore, quantitative easing becomes a special case of qualitative easing, where the increasing number of poor quality assets weakens the CB’s balance sheet (Bagus & Schiml, 2009, p. 49). The final aspect of UMPs is a corollary of the above two. In order to fulfill its LoLR role, the CB must broaden the goals of its monetary policy beyond the stability of the bank reserves market. In fact, it is not unconventional for CBs to intervene in other markets than the bank reserves market (Disyatat, 2008; Friedman & Kuttner, 2010). These usually involve ancillary operations in support of CMPs (Lenza, et al., 2010). For instance, interventions in the foreign exchange markets might be necessary after a change in the policy interest rate. If the policy rate is set higher than in other countries, then, this might attract bond investors. This influx of foreign investment would appreciate the domestic currency, which could be detrimental to exports. Therefore, the CB would sterilize this appreciation by acquiring foreign denominated instruments. Borio & Disyatat (2010, pp. 60-63) identify three kinds of CB intervention that go beyond the bank reserves market: exchange rate policies, quasi-debt management policies, and credit policies. Exchange rate policies target exchange rates and impact private sector exposure to foreign currencies (e.g., export policy). Quasi-debt management policies target public debt securities and impact private sector holdings of those securities: “A primary intention is to alter the yield on government securities, thereby influencing the cost of funding and asset prices more generally” (Borio & Disyatat, 2010, p. 62). Credit policies target private sector securities and

17

impact the private securities markets by modifying the exposure of the CB’s balance sheet to private sector securities: This can be done by modifying the profile of a given amount of private sector claims held by the central bank or by modifying the composition between private and public sector (central bank and government) claims held by the private sector. Such changes can be implemented in a number of ways, including through modifications of collateral, maturity and counterparty terms on monetary operations, by providing loans or acquiring private sector claims, including equities. (Borio & Disyatat, 2010, p. 63)

Finally, bank reserve policies also influence other markets because of the collateral accepted by the CB as eligible. The shift from conventional to unconventional monetary policies consists then of the use of the above policies primarily to stabilize other markets beyond the bank reserves market. The unconventionality of UMPs is not in the tools used. These are the same tools already available to CBs in more conventional times. The unconventionality is in the new, broader goals of the CB during a crisis. Table 3 summarizes the main differences between CMPs and UMPs.

18

Table 3: Liquidity problems

When CMPs fail, it is because they mainly concentrate on influencing markets by setting interest rates: The depth of the recession in many countries meant that Taylor rules would recommend negative nominal interest rates but market interest rates are effectively bounded by zero (or close to zero) because agents can always hold non-interest bearing cash. With the interest rates that central banks can set at or close to zero, other interest rates or forms of monetary policy needed to be considered. (Joyce, et al., 2012, p. F272)

Another problem limiting the efficiency of CMPs at the zero lower bound is that liquidity needs do not concern an unbalanced distribution of CB liquidities in the system. All three liquidity types—CB, market, and funding—are lacking in the economy. If the CB concentrates solely on its LoLR role, then, the most natural measure would be to reactivate the banks’ lending capacity immediately (Joyce, et al., 2012). This would consist of quantitative easing encompassing bank reserves and quasi-debt management policies (cf. Ugai, 2006). However, this would not be

19

sufficient. The reestablishment of bank liquidity does not translate into the general reestablishment of market liquidity for the economy. Indeed, banks would recover but would not lend to a failing market. They would prefer to keep their newfound liquidity reserves with the CB at the deposit rate. UMPs should thus consist of bypassing banks and directly financing other financial and non-financial institutions. The objective is to return market liquidity to all kinds of markets, not only the bank reserves market. This results in the CB liquefying all sorts of financial markets (Baba, et al., 2005; Bernanke, et al., 2004). Consequently, quantitative easing must include credit policy in order to liquefy other markets. This would amount to what Bernanke (2009) termed credit easing. This means that the CB’s balance sheet changes in composition, integrating less liquid, longer maturity, hence riskier securities from the private sector. Quantitative and qualitative easing thus go hand-in-hand. In bypassing banks, the CB assumes the role of financial intermediary to non-bank markets. This is possible thanks to the CB increasing its balance sheet precisely in order to absorb the failing securities behind the economy’s collapse. The CB creates a market for those securities, thus restoring funding and market liquidities. In other words, it goes beyond its role as a LoLR. It effectively becomes the MMoLR for the whole economy.

4. The MMoLR Role in Current Events 4.1 UMPs and changes in investment incentives The efforts to stabilize the economy using UMPs since 2008 have come at a macroeconomic cost (Gambacorta, et al., 2014). Indeed, the success of the massive outright OMOs necessary to restore the liquidity of internationally connected financial markets depended on the loss absorption capacity of CBs. Transnational collaboration between CBs was therefore a necessity,

20

as no single CB had the capacity to absorb all the losses of the world financial system (Cecchetti & Disyatat, 2010). UMPs immediately impact the exchange rate of the currencies of affected countries. The ability of a CB to maintain its currency’s purchasing power depends on the quality of assets it holds in its balance sheet. The qualitative deterioration of a CB’s balance sheet provoked by UMPs could result in the depreciation of the CB’s currency (Bagus & Howden, 2009; Bagus, 2009). For instance, following the adoption of UMPs by the Bank of England in 2008, sterling consistently lost value against other major currencies such as the euro, the US dollar, and the Japanese yen (Figure 1d). Indeed, it lost more than 14% of its value against these other currencies when British narrow money growth peaked at practically 100% in 2009. Similar currency depreciations can be observed for the euro (UMP adopted in 2012) and for the Japanese yen (Abenomics’ UMP in 2013). Figures 1a through 1d clearly show the depreciation trends of those three leading currencies at the peak of their UMPs. Figures 1a &1b

21

Figures 1c & 1d

Currency depreciation is not always evident. The American dollar actually appreciated against the euro and sterling at the peak of the Fed’s UMP in 2009, as seen in Figure 1d. This is understandable. The Bank of England adopted a more extensive UMP than the Fed in 2009. Moreover, although the European Central Bank did not expand monetary growth to the same extent as the other two, the Eurozone suffers from more market rigidity than its Anglo-Saxon partners, which leaves it more exposed to crises. One can notice, however, that Japan, being the only country among the four with mild monetary growth at the time, experienced a generalized appreciation of the yen, which the inflating dollar could not fight. This indicates that though currency depreciation might not always be visible, its pressure is still present in the economy. Another immediate impact of UMPs is the bypassing of the interbank market. If market liquidity shortages emerge, this is due to the meltdown of asset prices resulting in illiquidity. Normally, a bank facing a failing debtor can compensate the cash flow loss by liquidating the debtor’s assets. However, in the eventuality of market liquidity shortages, this is not possible and 22

the bank must meet its losses with retained earnings, provisions for losses, or by raising additional capital. The last option would hardly be available during a market liquidity shortage. Concerning the other two, since bank credit creation largely surpasses a bank’s capital, the losses will be too much for the bank to cover with its retained earnings or provisions for losses. This explains why the CB intervenes to restore the liquidity of other markets besides the interbank market. The goal is to avoid the build-up of loss in the banks’ balance sheets. However, this generates a dichotomy in the interbank market. Liquidity-short institutions can no longer resort to it, and become totally dependent on continuous CB refinancing to remain afloat. They thus become zombie institutions (Kane, 1987, pp. 78-79). Meanwhile, liquidity-abundant institutions avoid the failing interbank market and prefer to keep their reserves on their CB account. In this way, the interbank market becomes atrophied by the very UMP that is devised to save it from illiquidity (Baba, et al., 2005). The atrophy of the interbank market helps to explain the necessity for massive increases in the CBs’ balance sheets. If markets are made sufficiently liquid, then illiquid assets “become” liquid again. The downside to this liquidity recovery technique is that asset yields are driven down (cf. Joyce, et al., 2012, p. F274). In fact, UMPs aim to restore market liquidity and restart credit by making use of at least three channels: signaling, bank funding, and the “broad portfolio channel.” If CBs are credible enough, then, signaling a commitment to purchase illiquid assets might restore investor confidence on these assets, thus restarting their liquidity (Borio & Disyatat, 2010, p. 70). Nevertheless, it is quite unlikely that a mere signal will be enough in a crisis, hence the necessity of actual UMPs. The massive volume and longer maturity of UMPs keep markets afloat by lowering the yields of long-term assets (Bernanke, et al., 2004). This has two effects. On the one hand, lower yields translate into higher asset prices. This results in a

23

wealth effect that spills over to demand in the economy as predicted by the portfolio channel (Joyce, et al., 2012, p. F278)—assuming that a large share of the population are investing their savings in the financial markets. On the other hand, though the prices of long-term assets rise as their yields fall, this might incite investors to shift from debt to equity investment. In this manner, UMPs’ impact on equity securities also translates into rising prices. The general rise in asset prices results in an improvement of the collateral of debtors, who become attractive for bank credit once more (Borio & Disyatat, 2010, pp. 70-71). As long as banks consider their funding liquidity uncertain, they avoid lending. However, the commitment to sustain massive, longer operations aims to mitigate this uncertainty, to the point that banks begin to lend again, hence resulting in a reactivation of the bank funding channel (Joyce, et al., 2012, p. F281). Finally, buy-and-hold investment gives place to speculative investment as long-term asset yields fall following UMPs, thus resulting in an incentive from risk-averse to risk-seeking investment (Baba, et al., 2005, pp. 18-19; Borio & Zhu, 2012).

4.2 UMPs and inflation Can massive UMPs result in price inflation? This is not a concern from the UMP perspective and experience has shown that it is unlikely. The main goal of UMPs is to avoid severe liquidity shortage leading to a major default and debt deflation with all the negative macroeconomic consequences. The fact that massive purchase programs may create “negative inflation” by preventing deflation from occurring is not seen as a problem; on the contrary, it is seen as an adequate cure. Once the liquidity of the markets has been restored, the reason for continuing UMPs has been to reach a targeted level of inflation and/or unemployment as defined under CMPs (Borio & Disyatat, 2010; Friedman & Kuttner, 2010). Indeed, quantitative and credit

24

easing in Japan, the USA, the UK, and the Eurozone have not led to any major inflationary surges in the consumer goods markets, as shown in Figures 2a through 2d. CPI and PPI in the Eurozone, the US, Britain, and Japan show no significant evolution since 2001, despite considerable liquidity injections. Figures 2a & 2b

25

Figures 2c & 2d

The concern about inflation is usually brought up when discussing the exit strategy (Buiter, 2010). This does not mean, however, that the quantity theory of money no longer holds. Although UMPs monetize debt, as long as banks are not creating any additional credit, UMPs do not necessarily increase the amount of money in the broader sense. It can be seen in Figures 3a through 3d that UMP adoption by the Eurozone (2011-2012), the US (2008-2009), Japan (20012002, 2013-2014), and Britain (2008-2009) was immediately followed by a fall in lending in all countries.

26

Figures 3a & 3b

Figures 3c & 3d

If UMPs are not succeeding in pulling output prices up by reigniting bank lending, they are succeeding in keeping asset prices high. By liquefying financial assets, UMPs prevent their price 27

from falling. This is mostly evident in equity markets. One can observe that following the massive expansion of narrow money in the Eurozone, the United States, Japan, and the United Kingdom, stock indexes recovered almost all they had previously lost. The Euro Stoxx 50 recovered after a major monetary expansion in 2012 (Figure 4a), while the Dow Jones Industrial and the S&P500 recovered amazingly following UMP adoption by the Fed in 2009 (Figure 4b). A similar movement took place in the United Kingdom, though two major expansions were necessary to boost the FTSE100 (Figure 4d). The timing was different in Japan, but the effects were the same (Figure 4c). After major falls in the late 1990s and early 2000s, UMP adoption enabled the Nikkei225 to recover between 2004 and 2006. The current adoption of “Abenomics” has also coincided with a major recovery of the Nikkei225 since 2013. Figures 4a & 4b

28

Figures 4c & 4d

Another visible effect of UMPs is the depression of bond yields, since these are inversely related to bond prices. By preventing bond prices, among other financial assets, from falling, UMPs are effectively forcing down their prices. Figures 4a through 4d show precisely this point, bond yields in all the four countries are stably low at the lower-zero bound. It is particularly interesting to point out the falling spread between short-term and long-term yields.

4.3 The MMoLR role and the “credit crunch” However, one of the main goals of UMPs is precisely to restart lending and broad money expansion. The failure in reigniting bank lending could indicate a credit crunch. The decreasing spread between short- and long-term yields contribute to this view, since it becomes costly to finance longer-term loans. This leads banks to adopt more stringent criteria in granting loans— regulators’ pressure notwithstanding. In this way, banks ask for more and better collateral to safeguard loans operations. Nevertheless, CBs had absorbed most of the good collateral early on 29

during the subprime crisis, helping raise its price in the markets. Therefore, the scarcity of good collateral and the lower spread between short- and long-term rates make it more expensive for enterprises to look for bank loans. Indeed, in practically all the regions analyzed, lending has suffered a major blow after 2008. Financial corporations in the Eurozone and the United Kingdom practically ceased their loans investments. Although loans assets are recovering in the United States and Japan, that recovery is way below pre-2008 levels. However, lending activity in the form of investment in debt securities has been much less impacted by the crisis. British financial corporations returned to pre-2004 levels, while debt securities investment in the United States now surpass pre-2008 levels (Figures 5b and 5d)). In the Eurozone, debt securities and equity investments have returned to the levels observed in the early 2000s (Figure 5a). In Japan where QE was used for the first time in the early 2000s, investment in debt securities have been preferred to loans until 2010, when another QE round has been reabsorbing those securities, thus flooding depository corporations with currency and deposits (Figure 5c).

30

Figures 5a & 5b

Figures 5c & 5d

There are three explanations for these movements in the assets of financial institutions. First, the QE-buoyed asset prices added to the increased cost of lending has led to asset value gains being 31

much more attractive to the interest revenue that can be reaped from loans. Second, QE implied the absorption of good collateral from troubled financial institutions during the 2008 crisis. It is only natural to see those institutions rebuilding their collateral holdings. Finally, financial regulation incites banks to prefer safer debt securities—like government debt—over loans, which are considered riskier to bank’s capital. The MMoLR role and its UMPs instead of reigniting lending, actually create all the conditions to make it unattractive. An apparent exit for this situation would be to wait until financial institutions fully rebuild their good collateral holdings and their banks reserves. Once this is done, CBs could turn back to CMPs and raise the interest rates again, thus making lending attractive again. However, this would result in two problems. First, we do not know today how “good” the collateral being rebuilt really is, since its price has been artificially sustained by UMPs. Second, the end UMP-sustained asset prices would trigger another problem, this time with nonfinancial institutions.

4.4 Balance sheet recession and the limits of the MMoLR role The credit crunch argument rests on the idea that enterprises’ demand for loans is not met by financial institutions. It is true that credit creation is actually unencouraged by UMPs meant to incite it. But it is also true that most nonfinancial institutions in the leading Western economies and Japan cannot expand investment either. This explains helps explain why UMPs are failing in reigniting loans. Enterprises are simply not demanding those loans. Figures 6a through 6d describe the loans, debt securities, and equity liabilities of nonfinancial enterprises in the Eurozone, United States, Japan, and United Kingdom. All the economies analyzed depict a massive fall in loan finance when UMPs are adopted: 2009 in the

32

United States and the United Kingdom, 2010 in the Eurozone, and 2002 and 2011 in Japan. In all the Western economies, nonfinancial businesses drastically reduced leverage. British nonfinancial corporations are deleveraging to this day. A similar situation is observed in the Eurozone, though with less intensity. Japan and the United States show another picture that goes against the idea of an ongoing credit crunch. In fact, Japan was already deleveraging ever since crisis in the late 1980s. The Japanese UMPs initiated in 2002 were followed by a major recovery of loan finance in 2005 and 2008. After 2008, however, both Japan and the United States observed deleveraging that lasted until 2011, when loan finance recovered. More recently, loan finance in these two countries has been consistently reaching pre-crisis levels. This is further corroborated by the fact that debt security finance in American corporations actually surpassed pre-2008 levels. Meanwhile in Japan, debt security finance is balanced, which is an improvement to the deleveraging in the early 2000s. Figures 6a & 6b

33

Figures 6c & 6d

However, all the above mentioned economies are suffering from stagnated—or negative in the case of the United States—equity finance in nonfinancial corporate business. The stagnated equity finance in addition to the accentuated deleveraging immediately following the 2008 crisis indicates that far from a credit crunch, these economies are actually facing a demand contraction for loans as pointed out by Koo (2008, pp. 221-252; 2011). This demand contraction for loans can be explained by a balance sheet recession phenomenon. In other words, companies faced massive devaluation of their assets during the crisis and are now using any free cash flow available to pay down debt. It does not matter, how much liquidity is pumped into the economy by CBs, companies will hardly reignite investment unless they fully pay down debt, thus cleaning up their balance sheets. Consequently, demand for loans stagnate and in the absence of regular free cash flows, companies will use their equity reserves to pay down debt. A scenario that is consistent with the liabilities flows of nonfinancial corporates previously described.

34

An examination of the sectoral financial balances of an economy can give enough indications of a balance sheet recession (Koo, 2008, pp. 19-21; 2011, pp. 25-26). Under normal conditions, households are net savers (positive balance) to nonfinancial business (negative balance). Financial business, albeit intermediaries between households and nonfinancial business, also keep positive financial balances since they advance funds to the latter. If the economy runs foreign trade deficits, then it concomitantly runs surpluses in its foreign financial account, and vice-versa. Meanwhile, government ideally maintains a balanced budget. During balance sheet recessions, households’ and financial businesses’ financial balances deteriorate without necessarily turning negative, while the financial balances of nonfinancial businesses go into surplus as they become net savers. If government wants to prevent a significant fall in GDP, it goes into deficit trying to sustain the economy with its expenditures (Koo, 2008, p. 24ff.). Figures 7a through 7d describe the sectoral financial balances of the Eurozone, the United States, Japan, and the United Kingdom. It can be seen that the balance sheet recession threat clearly affects the Eurozone, Japan, and the United Kingdom. Japan already entered the 2000s facing a balance sheet recession that almost turned into a bonanza between 2005 and 2008, with nonfinancial corporations clearly running financial surpluses and paying down debt. Nonfinancial corporations in the Eurozone are running financial surpluses ever since late 2008 and are not yet in the deficit zone. In the United States, the situation turned sour in the 20082010 period, but has improved since. However, this improvement remains to show its force as nonfinancial corporate financial balances have recently neared the surplus zone again—which helps explain the hesitance of the Fed in raising rates and exiting UMPs. The situation is more unstable in the United Kingdom. Like Japan, nonfinancial corporations in the United Kingdom have been running surpluses since the early 2000s, indicating down payments of debt.

35

Households entered deficit territory in 2003, getting out of it in 2008 only to go back downwards as of recently. All economies show governments running deficits, though these deficits have been reduced in the last five to three years. Figures 7a & 7b

36

Figures 7c & 7d

The recent and apparent recoveries in the United States and Japan can lead to the conclusion that a persistent and unhesitant use of UMPs might keep away the danger of balance sheet recession, but this is illusory. For instance, it is possible that the households’ financial surplus observed in the Eurozone, the United States, and Japan since 2008 are actually the fruit of artificiallysustained asset prices. These same artificially-sustained asset prices could be behind nonfinancial corporate recovery. If the prices of their assets are maintained at a high level, then nonfinancial corporations might estimate that deleveraging is not necessary and proceed with investment. But it is unlikely that such a situation can persist in the long-run. Enterprises are not oblivious to the fact that asset prices are being artificially sustained by UMPs and know that asset prices will probably fall once these UMPs are unwinded. This would explain why American and Japanese recoveries are hesitant. The balance sheet recession scenario notwithstanding, there is a possibility for UMPs to succeed in lifting up the economy. If CBs definitively assume their MMoLR roles and engage 37

into perennial UMPs, then enterprises would see asset-buoying as permanent. Of course, the problem with this option is that it would eventually delve into a hyperinflationary situation, thus leading to another problem entirely. The MMoLR role is therefore at a crossroads. If it persists for long, then hyperinflation looms at the horizon. If it retrocedes into the classic LoLR role, the balance sheet recession will fully set in or even transform into a full-fledged liquidation recession. The alternative to the MMoLR role would be that of a massively active fiscal policy buoying nonfinancial investment, as Koo (2008) preconizes. However, the growing State debt overhang in Japan and the Western economies make this uncertain. Indeed, it is true that government expenditures can buoy private nonfinancial investment the time enterprises fully clean up their balance sheets. But it is also true that this debt-fueled demand must eventually translate into more taxation later. Koo (2008) sustains that economic recovery would improve tax revenues without any need for raising tax rates. This would be true only if the non-subsidized enterprises grow at a greater rate than the public debt service. However, if the MMoLR is abandoned, recovery will come with higher interest rates, which will raise the public debt service. One could advance that government expenditures could decrease as the economy recovers. This could work smoothly if the subsidized enterprises are not too big and dependent on government expenditures. Nevertheless, this seems rather unlikely given the large magnitude of the government expenditures involved in fiscal policy substitution of the MMoLR.

4.5 UMPs and the emerging economies A more global consequence of zero-lower bound monetary policy is the increase of volatility in commodity markets, which affects emerging countries negatively. By driving interest rates

38

down, UMPs in developed countries exacerbate interest rate differentials with emerging market countries (McKinnon & Liu, 2013). This creates an incentive for investors interested in yield gains to shuffle their portfolios and reorient their financial capital toward emerging markets, where yields are higher and default risk probability had been improving in the first half of the 2000s. Moreover, many emerging countries are also sources of highly valued commodities. These commodities gain even more value thanks to UMP-boosted actors looking for refuge assets. In this way, emerging markets see strong appreciation pressure on their currencies, which threatens the competitiveness of their products. In order to counter this, their CBs intervene massively in foreign exchange markets in order to stabilize their currency. But this entails monetary inflation, which exerts inflationary pressure on domestic costs. As financial crises prove difficult to tame, world economies clearly enter recession, productive investment stagnates and the demand for commodities falls. Emerging countries lose a great deal of their commodity market revenues but the inflated money is still in circulation, generating price inflation and speculative bubbles. Emerging countries become trapped by both UMPs in the developed world and their own lack of monetary control.4 Furthermore, unwinding UMPs could present a disastrous effect on emerging countries precisely because of the inversion in interest rate differentials between developed countries unwinding their UMPs and emerging countries. As yields increase again in the developed world, investors might abandon emerging markets preferring investments in better rated debt securities with improved returns. This could result in geopolitical problems for UMP-running developed countries because of the political fragility of many emerging markets. A situation that would be

4

The use by developed countries of monetary policies that reduce balance sheet quality tends to generate a wave of bubbles across frontiers because of what Schnabl & Hoffmann (2008) termed “vagabonding liquidity.” The excessive injection of liquidity in major markets generates major price and yield disparities in international financial markets. In this way, a crisis in one country can originate in another country’s monetary policy.

39

best avoided given that those emerging markets are the main sources of much sought after commodities such as base minerals and rare earth materials.

5. Conclusion: Economic recovery or push back? The irony of UMPs is that their success in restoring market liquidity can hardly be deemed a success. The main mechanism of UMPs is to sustain the growth of securities’ markets through continuous liquidity injections. This creates an incentive for investors to prefer portfolio investment to productive investment. Moreover, given the low yields resulting from UMPs, banks can hardly earn money on loans. This situation worsens over time, as banks’ portfolios need to be renewed at lower rates of interest—unless they can lend on a floating basis. Consequently, investments in the private sector migrate from the production side to financial markets. Investments in the “real economy” come to a halt. Therefore, in order to compensate for the lack of private investment in the production and restart the economy, the government spends and runs growing deficits. Debt accumulation is not much of a problem, as UMPs provide plenty of liquidity at low interest. This effectively leads to a crowding out of private investment in favor of public investments (Hoffmann & Schnabl, 2011). Therefore, the success of UMPs widens the gap between financial and productive markets. Under normal circumstances, the performance of financial assets should reflect the underlying productive performance of their issuers. The stagnation of private investment and the rise of public expenditures demonstrate that the attractiveness of financial markets is over dependent on the continuous presence of UMPs, and this dependence explains why the Fed, for instance, is so slow to end its zero rate policy. How can CBs unwind UMPs without risking a recession? Simply reducing the CB’s balance sheet through sales of assets would result in a massive fall in asset prices as market

40

participants would be incapable of fully absorbing the CB’s balance sheet immediately (Buiter, 2010, p. 25). This would undo the positive impact of UMPs on the market’s liquidity. Moreover, such an isolated move by one CB could be resisted by others equally engaged in UMPs. The worsening of market liquidity conditions would raise the costs of their own UMPs to prohibitive levels. Consequently, downsizing a CB’s balance sheet would necessarily require transnational concertation between CBs. Another source of resistance to the downsizing of the CB balance sheet is deficit-running governments. Indebted governments benefit from low interest rate policies, since such rates help them to fund their deficit as public expenditure replaces private investment. Low interest rates help them to service their debt even as this debt grows (Hoffmann & Schnabl, 2011, p. 401). When ending their UMPs, CBs will probably sell back public debt instruments to market investors. Consequently, their prices will fall resulting in higher yields. Further issuances of public debt would have to be done at higher yields, which would significantly increase debt-servicing costs if the debt were to be rolled over. Another solution for the CB’s would be to hold the assets purchased through UMPs until maturity—especially the most illiquid ones—as an exit strategy, thus limiting the negative impact on markets (Buiter, 2010, p. 24). This strategy may require time to have a significant impact on the CB’s size, and the time required depends on the maturity structure of the portfolio. Nevertheless, this strategy assumes that banks are not lending massively their excess reserves that have so far been parked at CBs. Moreover, consistent with that policy, the government must run surpluses. For the moment, only the Eurozone aims at this solution, under German leadership. Meanwhile, the United States and Japan show no sign of coordinating their fiscal policy with the unwinding of UMPs. This solution requires close coordination between the government, the CB and the private sector. Indeed, banks and companies must also decrease

41

leverage and clean their balance sheets of bad assets to unwind their UMPs successfully. Commercial banks have already eliminated many “bad assets,” which were very often purchased by CBs. But the new challenge faced by the exit strategy is that UMPs have blurred the differences between “good” and “bad” assets since massive purchase programs including bad assets lead to an increase in the price of all assets. Under these circumstances, it is very difficult to discriminate between bad and good assets, since their prices increase in the same manner. How is one supposed to know whether the assets rising in price that remain in circulation are good or bad? Which assets will maintain their price, or at least mitigate the price fall, and which will not? It is clear that the performance of financial markets must regress. In any case, UMPs are just stalling the liquidation of the accumulation of poor investments made in previous economic booms. Instead of correcting disequilibria in financial and productive markets, UMPs actually generate more volatility, widening gaps across markets, and making financial institutions over-dependent on CB interventions. Meanwhile, production cannot restart on any solid basis because investors are too attentive for any changes in monetary policy. Long-term planning is sacrificed for short-term opportunism. The longer it takes to correct asset prices, the more divergences accumulate in the markets. But even the eventuality of safely unwinding UMPs cannot eliminate another risk. If the economy stabilizes and a promising investment context emerges, then credit creation will resume. However, this resumption of credit creation will take place in an economy highly charged with CB liquidity. This means that credit creation will be much more strongly supported. The recovery would just reignite another boombust cycle, but on a much greater scale. Consequently, a safe withdrawal from UMPs does not depend solely on their unwinding but also on institutional reforms to commercial banking itself.

42

6. References Adrian, T. & Shin, H. S., 2009. Money, liquidity, and monetary policy. Federal Reserve Bank of New York Staff Report No. 360. Baba, N. et al., 2005. Japan's deflation, problems in the financial system and monetary policy. BIS Working Papers No. 188. Bagehot, W., 1873. Lombard Street: A Description of the Money Market. London: s.n. Bagus, P., 2009. The quality of money. The Quarterly Journal of Austrian Economics, 12(4), pp. 22-45. Bagus, P. & Howden, D., 2009. Qualitative easing in support of a tumbling financial system: A look at the eurosystem's recent balance sheet policies. Economic Affairs, 29(4), pp. 60-65. Bagus, P. & Schiml, M. H., 2009. New modes of monetary policy: Qualitative easing by the FED. Economic Affairs, 29(2), pp. 46-49. Baker, M. & Wurgler, J., 2002. Market timing and capital structure. The Journal of Finance, 57(1), pp. 1-32. Bastidon Gilles, C., Huchet, N. & Gilles, P., 2012. Amplification effects and unconventional monetary policies. Theoretical and Applied Economics, 19(2), pp. 13-30. Bernanke, B. S., 2009. The crisis and the policy response. Speech given at the Stamp Lecture, London School of Economics, London, January 13. Bernanke, B. S. & Mishkin, F. S., 1997. Inflation targeting: A new framework for monetary policy?. NBER Working Paper No. 5893. Bernanke, B. S., Reinhart, V. R. & Sack, B. P., 2004. Monetary policy alternatives at the zero bound: An empirical assessment. Brookings Papers on Economic Activity, 2004(2), pp. 1-78. Borio, C., 1997. The implementation of monetary policy in industrial countries: A survey. BIS Economic Papers No. 47, pp. 1-148. Borio, C. & Disyatat, P., 2010. Unconventional monetary policies: An appraisal. The Manchester School, 78(S1), pp. 53-89. Borio, C., Galati, G. & Heath, A., 2008. FX reserve management: Trends and challenges. BIS Papers No. 40. Borio, C. & Nelson, W., 2008. Monetary operations and the financial turmoil. BIS Quarterly Review, March, pp. 31-46. Borio, C. & Zhu, H., 2012. Capital regulation, risk-taking, and monetary policy: A missing link in the transmission mechanism?. Journal of Financial Stability, 8(4), pp. 236-251. 43

Buiter, W., 2008. Quantitative easing and qualitative easing: a terminological and taxonomic proposal. Financial Times, William Buiter's mavercon blog, 9 December. Buiter, W. H., 2010. Reversing unconventional monetary policy: Technical and political considerations. In: M. Balling, J. M. Berk & M. Strauss-Kahn, eds. The Quest for Stability: The Macro View. Vienna: Larcier, pp. 23-44. Cecchetti, S. G. & Disyatat, P., 2010. Central bank tools and liquidity shortages. FRBNY Economic Policy Review, August, pp. 29-42. Constâncio, V., 2011. Challenges to monetary policy in 2012. Speech given by Vítor Constâncio, Vice-President of the ECB, at the 26th International Conference on Interest Rates, Frankfurt-amMain, 8 December 2011.. [Online] Available at: http://www.ecb.europa.eu/press/key/date/2011/html/sp111208.en.html [Accessed 6 February 2015]. DeYoung, R. & Rice, T., 2004. How do banks make money? A variety of business strategies. Economic Perspectives - Federal Reserve Bank of Chicago, 28(4), pp. 52-68. DeYoung, R. & Rice, T., 2004. How do banks make money? The fallacies of fee income. Economic Perspectives - Federal Reserve Bank of Chicago, 28(4), pp. 34-51. Disyatat, P., 2008. Monetary policy implementation: Misconceptions and their consequences. BIS Working Papers No. 269, pp. 1-28. Dominguez, K. M. E., 2006. The European Central Bank, the Euro, and global financial markets. Journal of Economic Perspectives, 20(4), pp. 67-88. Farmer, R. E. A., 2012. Qualitative easing: How it works and why it matters. National Bureau of Econnomic Research , Volume w18421, pp. 1-32. Friedman, B. M. & Kuttner, K. N., 2010. Implementation of central bank policy: How central banks set interest rates?. National Bureau of Economic Research Working Paper Series, Issue w16165. Gambacorta, L., Hofmann, B. & Peersman, G., 2014. The effectiveness of unconventional monetary policy at the zero lower bound: A cross-country analysis. Journal of Money, Credit, and Banking, 46(4), pp. 615-642. Goodhart, C. A. E., 1995. The Central Bank and the Financial System. Cambridge(MA): MIT Press. Hoffmann, A. & Schnabl, G., 2011. A vicious cycle of manias, crises and asymmetric policy responses - An overinvestment view. The World Economy, 34(3), pp. 382-403. Joyce, M., Miles, D., Scott, A. & Vayanos, D., 2012. Quantiative easing and unconventional monetary policy: An introduction. The Economic Journal, 122(564), pp. F271-F288.

44

Kane, E. J., 1987. Dangers of capital forbearance: The case of the FSLIC and "zombie" S&Ls. Contemporary Policy Issues, 5(1), pp. 77-83. Kane, E. J., 2012. The inevitability of shadow banking. Presented at the Federal Reserve Bank of Atlanta 2012 Financial Market Conference Financial Reform: The Devil's in the Details, Atlanta Georgia, April 10. Kindleberger, C. P., 1984. A Financial History of Western Europe. London: George Allen & Unwin. Koo, R. C., 2008. The Holy Grail of Macroeconomics: Lessons from Japan's Great Recession. Singapore: Wiley. Koo, R. C., 2011. The world in balance sheet recession: Causes, cures, and politics. Real-World Economics Review, 58(12), pp. 19-37. Lenza, M., Pill, H. & Reichlin, L., 2010. Monetary policy in exceptional times. Economic Policy, 25(62), pp. 295-339. McKinnon, R. & Liu, Z., 2013. Zero interest rates in the United States provoke world monetary instability and constrict the US economy. Review of International Economics, 21(1), pp. 49-56. Mehrling, P., 2011. The New Lombard Street: How the Fed Became the Dealer of Last Resort. New Jersey: Princeton University Press. Noeth, B. J. & Sengupta, R., 2011. Is shadow banking really banking?. The Regional Economist, Volume 10, pp. 8-13. Pozsar, Z., Tobias, A., Ashcraft, A. & Boesky, H., 2010. Shadow Banking. Staff Report, Federal Reserve Bank of New York, No. 458. Schnabl, G. & Hoffmann, A., 2008. Monetary policy, vagabonding liquidity and bursting bubbles in new and emerging markets: An overinvestment view. World Economy, 31(9), pp. 1226-1252. Taylor, J. B., 1993. Discretion versus policy rules in practice. Carnegie-Rochester Conference Series on Public Policy, Volume 39, pp. 195-214. Taylor, J. B., 1999. A historical analysis of monetary policy rules. In: J. B. Taylor, ed. Monetary Policy Rules. Chicago(IL): The University of Chicago Press, pp. 319-348. Ugai, H., 2006. Effects of the quantitative easing policy: A survey of empirical analyses. Bank of Japan Working Paper Series, Issue 06-E-10, pp. 1-63.

45