The Quantitative Easing Experience

(1)

The Quantitative Easing Experience

- A comparative study of unconventional monetary policies and their influence in the US and Europe -

Type of assignment: Master’s Thesis Name of Supervisor: Søren Ulrik Plesner Hand-in Date: 26^th August 2016

Number of pages and characters: 78 pages/ 168,768 characters Program: Applied Economics & Finance

Name: Martje Marieke Schmidt Copenhagen Business School 2016

(2)

1

Executive Summary

In the face of weak growth and severe deflationary risks the European Central Bank (ECB) decided to introduce a large-scale quantitative easing (QE) program in March 2015. The ECB thus follows the Japanese, UK, and US central bank in using unconventional monetary policies to stimulate the economy when interest rates are close to zero. This study especially focuses on the US experience with QE and analyses how these findings can be applied to the ECB’s QE program.

The primary objective of this thesis is to examine the US QE program and its influence, to draw a comparison to the European program, and establish pivotal success criteria of QE. The study is based on an extensive literature review of both qualitative and quantitative research. The comparative analysis also uses information and recent data gathered from the respective central banks.

The US and Euro area differ largely in their institutional and financial systems, and the QE program designs have been tailored to the needs of their economies. In the light of empirical evidence and the author’s view, there are a few key lessons that can be learned from the US program: Open-ended QE with a sizeable effect on the central bank’s balance sheet that is tied to the policy goal signals a credible commitment and has been most powerful in affecting long-term rates. Furthermore, managing inflation expectations involve clear and consistent communication to the public about the monetary policy actions.

The main transmission mechanisms of QE in the US were the portfolio balance channel and the signaling channel. In Europe, the bank lending and exchange rate channel could play a pivotal role. A unique aspect of the ECB’s QE program is its negative interest rate policy, which should encourage bank lending and further supports the exchange rate channel through currency depreciation. The impact on inflation will ultimately depend on the banks’ willingness to lend and whether the ECB’s accommodative stance successfully restores confidence among market participants. This increases the importance of credible fiscal and structural policies that support a healthy banking system and economic growth.

(3)

2

1. Introduction

1.1. Monetary policy responses to the global financial crisis

The weak economic situation in the Euro area with severe deflationary risks forced the European Central Bank (ECB) to introduce a large-scale quantitative easing (QE) program. The program started in March 2015 and will be conducted until March 2017, or in any case until medium-term inflation is consistent with the ECB’s price stability mandate of levels close to, but below 2%. The ECB thereby follows other major central banks in using unconventional monetary policies to stimulate their economies by altering government and corporate bond yields (Claeys & Leandro, 2016).

The financial crisis that followed after the bankruptcy of Lehman Brother's in September 2008 left most financial markets dysfunctional, with falling GDP levels and short-term interest rates close to zero. Central banks and governments worldwide faced similar challenges posed by the global crisis, but their monetary policy responses differed, reflecting the structure of their respective financial sector and economy. Initially, major central banks' policies focused on providing liquidity and restoring stability in the financial markets, but their focus soon shifted towards stimulating real growth and preventing disinflation (Fawley & Neely, 2013).

The depth of the global financial crisis and the subsequent recessions called for unprecedented policy responses by fiscal and monetary authorities. In normal times, conventional monetary policy acts by setting a target for short-term interest rates and adjusting the supply of central bank reserves to achieve the target through open market operations (Smaghi, 2009). Thus, the composition and size of the central bank balance sheet results passively from steering the short-term interest rates consistent with their monetary policy stance (ECB, 2015d). Pre-crisis conventional policies had been reliable in providing monetary stimulus during recessions, stabilizing inflation, and ensuring the functioning of the money market. In the aftermath of the global financial crisis, however, conventional monetary policy proved ineffective in achieving the central banks' policy mandates. Usually, central banks cut their target for policy rates in order to stimulate the economy. But the severity of the shock meant that in many countries, central banks needed to bring their interest rates close to zero (Smaghi, 2009).

(6)

5

Furthermore, as a result of the rising mistrust among market participants, the relationship between changes in official interest rates and market interest rates broke down (Joyce, Miles, Scott, &

Vayanos, 2012). Once interest rates hit the zero lower bound (ZLB) and monetary transmission processes were significantly impaired, central banks had to resort to unconventional monetary policy measures.

Unconventional monetary policy takes many forms and has been defined as policies that directly target the cost and availability of external finance to banks, households and non-financial companies (Smaghi, 2009). They differ from conventional policy measures as they aim at expanding the central banks' balance sheet and attempt to influence interest rates other than the traditional short-term rates. Specifically, QE policies which form the topic of this thesis, focus on targeting the quantity of bank reserves (Joyce et al., 2012). Classical QE usually involves purchases of financial assets such as long-term government bonds, which raises the prices of those assets and lowers yields. Asset purchases result in an increase in central bank reserves held by the banking system and ultimately increase base money. By injecting money into the economy, in return for other assets, it increases the liquidity of private sector balance sheets which in turn boosts nominal spending and influences

inflation rates (Smaghi, 2009).

In the aftermath of the financial crisis, a number of central banks resorted to unconventional

monetary policies in an effort to influence economic activity and remove deflationary pressures. With short-term rates close to the effective lower bound in the United States, Unites Kingdom, Japan, and the Euro area, the policy target shifted from policy rates to the level of excess reserves. Each central banks' QE program reflected the structure of its respective economy, with bond markets playing a pivotal role in the US and UK while banks were more important in Europe and Japan. The monetary policy responses of the Federal Reserve Bank (Fed) and Bank of England (BOE) thus differed from those of the Bank of Japan (BOJ) and ECB, by concentrating on bond purchases rather than lending directly to banks (Fawley & Neely, 2013).

(7)

6

The BOJ was the first central bank to introduce QE prior to the global financial crisis in 2001 when it dealt with an extended period of extremely weak economic conditions. The BOJ initially increased its target for bank reserves from ¥4 trillion to ¥5 trillion as an attempt to bring the overnight call rate from 0.15% to zero and fight deflation. By 2004, the target had been raised to ¥35 trillion and involved both public and private debt purchases. In 2006, the program ended and was widely regarded as a failure because of its small size. However, the Japanese experience would provide an example for subsequent QE programs, with important lessons learned.

Probably the most prominent QE program in terms of scale and importance was the Federal Reserve's large-scale asset purchases (LSAPs) starting in 2008, which included debt of government-sponsored enterprises (GSEs), mortgage-backed securities (MBS), and long-term Treasuries. The Federal Open Market Committee (FOMC) also made increasing use of forward guidance policies about the future path of key policy rates and the size of its asset purchases. Overall, the Fed’s balance sheet increased to about $4.5 trillion, about 26% of its GDP. The BOE followed closely after the Fed's first measures by announcing plans to purchase assets in March 2009. The program involved purchases of private asset and government securities and started with a target of £50 billion, which was incrementally raised to

£375 billion. Lastly, the BOJ reintroduced QE policies in 2008 which included outright asset purchases amounting to ¥101 trillion by the end of 2012. In April 2013, the BOJ introduced a massive

Quantitative and Qualitative Monetary Easing (QQME) program with the aim of doubling the size of its monetary base in two years, mainly through purchases of government bonds (Gros, Alcidi, &

Groen, 2015).

After rejecting QE measures similar to those of the US, UK, and Japan for most of the post-financial crisis period, the ECB became the last of the central banks to implement a QE program in 2015, when it was facing weak growth and deflationary pressures. Prior to this, the ECB had undertaken several measures in direct response to the crisis, which mainly focused on the provision of liquidity in support of the banking system and involved two Covered Bond Purchase programs (CBPP), several Long-term Refinancing Operations (LTRO), and a Securities Markets Program (SMP), later replaced by the

(8)

7

Outright Monetary Transactions (OMT) program. However, overall inflation rates had been falling since late 2011 and even turned negative again in 2016. With inflation threatening the ECB’s price stability mandate and key interest rates at the ZLB, the ECB decided to launch a set of unconventional measures to further ease the monetary policy stance. The Public Sector Purchase Program (PSPP) introduced in March 2015 would be part of the expanded Asset Purchase Program (APP), consisting of a third Covered Bond Purchase Program (CBPP3), and the Asset-backed Securities Purchase Program (ABSPP), adding purchases of government bonds and securities from European institutions and national agencies.

Since its implementation, the ECB has made several significant changes to the original design of the QE program to expand the list of eligible assets for purchase (Claeys, Leandro, & Mandra, 2015).

Furthermore, the PSPP has been extended from September 2016 initially, to March 2017, while at the same time increasing monthly purchases from 60 billion to 80 billion euros per month. Recently, the ECB added a Corporate Sector Purchase Program (CSPP), allowing for purchases of Euro area

corporate bonds. The ECB QE policies have been largely controversial as the treaties that founded the EU prohibit monetary financing of a member state, and thus government bond purchases gave rise to a number of political and operational challenges. In order for the ECB to stay within its legal mandate, it had to extend the scope of its operations far beyond what had been envisaged by the Maastricht Treaty (Micossi, 2015).

1.2. Problem statement

There already exists substantial research on the effectiveness of unconventional monetary policies, especially in the US and the UK, QE was found to successfully reduce government and corporate bond yields. For example, Gagnon, Raskin, Remache, and Sack (2011), study the impact of the Fed’s LSAP’s between 2008 and 2010 (QE1) and find economically significant effects on longer-term interest rates.

Joyce, Lasaosa, Stevens, and Tong (2010), estimated that the BOE’s QE program reduced gilt yields by about 100 basis points. Neely (2013), finds that the Fed’s QE1 announcements significantly reduced international long-term bond yields and led to a depreciation of the US dollar.

(9)

8

Despite a large amount of empirical research, there exists less literature that describes and compares the different QE programs. In particular, the author will argue, that the design features of each QE program are largely dependent on underlying economic structures and determine the success of QE.

The Fed’s QE program had encompassed three rounds of LSAP’s and the central bank officially ended its balance sheet expansion with the tapering process in 2014, after major improvements in labor markets, GDP growth, and inflation outlooks. In December 2015, the Fed announced to raise short- term interest rates from the ZLB for the first time since the financial crisis – a decision that would restore confidence in the US economy while most other advanced economies still struggled.

Meanwhile, the ECB announced to further lower its policy rates into negative territory while expanding the size and length of its QE program in an effort to fight deflationary pressures.

The significant divergence between the two central banks and their policy responses set the stage for this thesis. Owing to the success of the Fed's QE program, it serves as a benchmark for evaluating the potential impact of the recently introduced ECB QE program on the economy, which leads to the primary research question:

1. To what extent can the findings from the US experience with QE be applied to the ECB in its attempt to stimulate inflation in Europe through unconventional monetary policy measures?

This research studies QE and its impact on the economies where it has been implemented by

explaining unconventional policy measures taken by the major central banks BOJ, BOE, Fed, and ECB, with a special focus on the latter two. The primary objective of this thesis is to examine the US QE program and its influence, to draw a comparison to the European program and establish pivotal success criteria of QE. Studying the ECB’s QE program is of particular relevance not only because of its recency but also since its unique design features give rise to much ambiguity that leaves room for clarification. As a monetary union consisting of 19 diverse member states, all facing different economic conditions, the ECB faces various challenges in the implementation of a single monetary policy. Given the contrasting institutional structures of the US and the Eurozone, it is especially

(10)

9

interesting to compare the ECB’s QE program to the one of the Fed. The relative importance of the US QE program arises from the size and its positive economic impact, whereas primarily focusing on the US experience gives the opportunity for a more detailed comparison between the Fed’s and ECB’s QE measures. Furthermore, there exists profound research on the US QE program and its effects, while the impact of the ECB’s APP on the economy remains to be assessed. The thesis aims at establishing a clear understanding of QE policies in theory and praxis and drawing a conclusion of the major success criteria for the ECB’s program.

The paper attempts to answer the following sub-questions:

a) What is QE?

b) What was the economic impact of the QE programs by the Fed, BOJ, and BOE?

c) What are the major implications from the Fed’s QE program?

d) How does QE operate under the ECB’s policy mandate?

e) How does the ECB’s QE program differ from the Fed’s QE program?

f) What are the lessons from the US QE and success criteria for the ECB?

1.3. Methodology

The design of the study results from the author’s decision processes about how the research will be conducted and is closely related to the framework of the study. The research design chosen for this thesis takes the form of a descriptive historical research using secondary data sources. A survey of important articles, books, and other sources helps to gain an insight into the topic of QE.

Furthermore, the evaluation of the impact of QE on the economies studied requires a thorough analysis of key developments over time since their implementation. Secondary data has the advantage that it allows for a far larger sample to be analyzed and interpreted in the given time (Saunders, Lewis, & Thornhill, 2009). Evidence about the impact of QE measures on the respective economies stemmed from both former empirical studies and qualitative research.

(11)

10

Due to the current nature of the topic, many media sources and updates from the central bank’s websites were used as they cover more recent developments. Numerous graphs and recent empirical research from those sources were used to show economic implications of the QE programs. These sources were also useful to find information about recent changes to the ECB’s QE program to later highlight parallels and differences in the design features of the QE measures and to consider possible success factors for the ECB. While no independent empirical research was carried out by the author of this thesis, the study concludes the findings from the literature review with a comprehensive

discussion on the QE measures, its effectiveness, and the lessons learned.

The nature of this research design makes it very dependent on the data sources used for explanations in the analysis. The research questions have been addressed by analyzing mostly quantitative

secondary data and making use of relevant theories and academic research where applicable. Relying on secondary data and sources required thorough selection by the researcher to identify relevant and valuable sources to make correct inferences. To assess relevance and value of the literature, the author focused on the recency of the research, theory robustness, and the context in relation to the thesis (Saunders et al., 2009). To ensure the validity of the information, the number of citations of the relevant literature in the databases served as a main quality indicator. All data used such as charts, tables or graphs are sourced from internationally renowned sources such as Eurostat and the central banks of the countries in question. The same applies to latest economic data on exchange rates, yields, and GDP, to ensure the highest amount of credibility when analyzing the data. Furthermore, explicit limitations in regards to the scope of the topic are essential for the thesis not to become too extensive and trivial.

1.4. Topic and literature delimitations

The topic of QE lies within a large field of empirical findings regarding national and international effects of the respective programs. It was, therefore, crucial to limit the scope of the subject to

achieve a stronger outcome of the thesis. The novelty and the controversies revolving around the ECB QE program, together with a large debate about the overall effectiveness of QE make it an especially

(12)

11

interesting topic to study. The other central banks chosen represent major economies which depict crucial differences in their economic structures and experiences with QE. The BOJ, often credited with inventing QE, faces similar deflationary pressures as the ECB. The UK was the first European country to implement QE measures, while the US QE program is probably most prominent because of its scale and effectiveness. The context of the Fed’s monetary policy measures is sufficiently different from that of the ECB to allow for a relevant discussion. While academics have already conducted

substantial research on QE policies, it is most profound for the US, which is why the thesis has a large focus on the US QE experience. The existing literature allows for a deeper analysis of QE and at the same time builds the foundation to understand the potential impact of the QE program in Europe.

Because of the novelty of the ECB’s QE program, there is only limited empirical data available; the PSPP officially started on March 9, 2015 and has been augmented several times since then. The thesis mostly focuses on unconventional policy measures taken in response to the financial crisis in 2008 and will conduct a preliminary assessment of some economic indicators in the Eurozone observed until May 2016. Since asset purchases will be conducted at least until March 2017 and because of considerable time lags in the implementation of monetary policies, a full evaluation of the influences on long-term growth and inflation, the ultimate goal of QE, is not possible yet. The study concludes with implications of QE policies and an outlook on unconventional monetary policy. Exit strategies of QE will not be covered in this paper, as the focus is set on the ongoing QE programs and their impact.

1.5. Structure of the thesis

The thesis is divided into six chapters, which are structured as follows: The first chapter gives an introduction to the topic of QE and explains the thought process behind the author’s research. The second chapter addresses the definition of unconventional monetary policies, with a focus on how and when QE is applied. The chapter also includes a discussion of monetary policy theory in relation to QE. After a brief overview of the Japanese and the UK QE experience, Chapter 3 describes the US QE program in more detail, followed by a thorough empirical review of its economic impact. Chapter 4 describes the ECB monetary policy under its policy mandate after the financial crisis and QE

measures at present. Chapter 5 examines the differences between the Fed and ECB policies and

(13)

12

discusses the major implications from the US QE experience and success criteria of QE. The closing chapter consists of a conclusion of the results and finishes with an outlook on longer-term

implications of the measures in the light of modern economic theory of unconventional policies.

Figure 1 Structure of the thesis

(14)

13

2. Monetary policy

2.1. Unconventional vs. conventional monetary policy

The main objective of monetary policy is to influence financial and macroeconomic conditions in order to achieve the central banks' policy mandate. Conventional monetary policy typically acts by signaling the desired policy, setting a target for the overnight interest rate in the interbank money market and adjusting the supply of central bank money through open market operations. Within this operating framework, central banks can steer the interbank interest rate by injecting reserves into the banking system in response to banks' demand and thereby pursue its monetary policy objective. The size and composition of a central bank's balance sheet are primarily determined by exogenous factors such as government deposits, reserve requirements, and public demand for capital. This being said, in

"normal times" the central bank is not involved in direct lending to the private sector or the

government. Further, does it not conduct outright purchases of government bonds, corporate debt, or other types of debt instrument (Smaghi, 2009).

In abnormal times, however, conventional monetary policy tools may not be sufficient in meeting the central bank's objective. This might occur when an economic shock is powerful enough to bring the nominal interest rate close to zero. At the ZLB, further increases in the money supply have no effect on long-term interest rates. Monetary stimulus can then only be achieved by resorting to

unconventional monetary policy tools, namely by guiding the medium to long-term interest rate expectations, by changing the composition of the central bank's balance sheet, and by expanding the size of the central bank's balance sheet. These policies have in common that they affect financial conditions beyond short-term interest rates and instead focus on the long-term interest rate (Smaghi, 2009).

Unconventional measures may also be needed when the policy rate is above zero, that is, if the transmission mechanism of monetary policy is severely impaired. Central banks have responded by lowering the short-term nominal interest rate even further than usually, and implemented policy measures that directly impact the transmission process (Smaghi, 2009).

Policy makers can influence real long-term interest rates by affecting expectations which work

(15)

14

through several channels. If the public expects a higher future inflation, an increase in inflation expectations causes real interest rates to decrease, even while the nominal interest rate is close to the ZLB. Moreover, central banks can directly impact expectations about future interest rates by committing to zero interest rates for a significant period, and thereby flatten the yield curve. Such a strategy, often referred to as "signaling", could also prevent expected inflation from falling, keeping real interest rates from raising, which would otherwise reduce spending (Smaghi, 2009).

Alternatively, central banks can influence interest rates by affecting market conditions of assets at various maturities such as government bonds, corporate debt, commercial paper or foreign assets.

The policy termed QE aims at affecting the level of the longer term interest rate of financial assets, mainly "risk-free" government bonds. The size and composition of the central banks' balance sheet can be affected by directly buying the relevant assets or indirectly. Direct QE usually involves purchases of longer-term government bonds from commercial banks in exchange for central bank money. Since sovereign yields can determine prices of riskier privately issued securities, central bank purchases of long-term government bonds will cause yields on privately issued securities to fall along with the yields on government bonds. The fall in long-term interest rates supports price stability by encouraging investments and aggregate demand. Since this implies that central banks directly hold the assets, until maturity or resale, they also hold the risk on their balance sheet (Smaghi, 2009).

Some central banks also resorted to indirect QE/credit easing measures. The aim is to increase the balance sheet through central bank creation of reserves by lending capital to banks at longer maturities, against collateral which includes assets of significantly impaired markets. The policy affects the yield curve over the committed timeframe of the monetary policy operations, such as operations conducted at a fixed rate, full allotment. This implies an indirect increase in the monetary base based on the demand of the banking system for excess reserves (Smaghi, 2009).

Another term often discussed in this context is credit easing, which is differentiated from QE. Credit easing directly addresses liquidity shortages and involves purchases of private sector debt and securities of impaired markets such as commercial paper, corporate bonds and asset-backed

(16)

15

securities. Thus, credit easing affects the risk spread across assets and is intended to reduce specific interest rates and restore market functions. While QE comprises any policy that increases the central banks liabilities at the ZLB, credit easing can also be conducted when short-term nominal interest rates are above zero. When financed by the creation of central bank reserves and conducted at a large enough scale, credit easing measures can have considerable effects on real variables. Credit easing may in this case also have a "quantitative" impact on the central bank balance sheet, making it difficult to distinguish credit easing from quantitative easing (ECB, 2015d; Fawley & Neely, 2013).

All of the above thus classify as unconventional monetary policy measures and have been implemented by central banks to various degrees, depending on institutional characteristics and country-specific issues. This paper particularly considers QE as a financial crisis response, which is commonly associated with using communication policies to shape public expectations about the future course of interest rates, increasing the size- and changing the composition of the central bank's balance sheet when interest rates are at the effective ZLB.

2.2. Liquidity trap: the conceptual foundation of QE

Usually, when a recession hits the economy, the central bank cuts the interest rate in order to stimulate spending and achieve low and stable inflation. However, it cannot move its target lower than zero percent, since negative interest rates would imply that banks get a higher return from hoarding cash than from lending it. Thus, following the financial crisis, central banks had to resort to unconventional measures as the nominal interest rate targeted by central banks effectively hit the ZLB (Smaghi, 2009). The optimal policy responses when the policy rate hits the ZLB have been widely discussed, especially among Keynes and his successors.

The special case of a liquidity trap can be analyzed in a standard IS/LM-Model. Since the LM-curve is flat under these conditions, expansionary monetary policy is ineffective in stimulating economic activity, as money and bonds become close substitutes. When the interest rate cannot be lowered due to the binding ZLB, the central bank operations through the interest rate channel are not effective either. When the economy falls into a liquidity trap, a persistent recession can lead to a deflationary spiral as it causes real interest rates to rise (Blinder, 2010).

(17)

16

Economic research suggests that monetary policy is far from powerless when interest rates hit zero, as a flattening of the yield curve and lower risk premiums can boost aggregate demand. The liquidity trap puts a floor on nominal interest rates, but not necessarily on real interest rates. In New

Keynesian models, monetary policy at the ZLB is therefore seen to be most effective through

increased inflation expectations, which lower long-term real interest rates and stimulate investment and spending (Blinder, 2010).

Woodford (2012) suggests that forward guidance, clear communication about the expected path of the future interest rate, can provide policy accommodation when the ZLB is a binding constraint. The central bank’s commitment, if credible, to keep policy rates low for an extended period of time can effectively increase inflation expectations. Woodford (2012) further argues, that most effects from balance sheet policies stem from the signal they provide and thus cannot substitute the central bank commitment about future policy. Two challenges arise from forward guidance however: time-

inconsistency and clear communication. Central banks have an incentive to raise rates after the crisis to avoid the rise in inflation, which limits its credibility. Furthermore, if the public does not

understand the central bank's policy path and/or has different expectations about the future of the monetary policy path, forward guidance may be ineffective (Williams, 2012).

Recent studies have found empirical evidence that QE can affect the yields of various financial assets and be an effective tool to help overcome the ZLB (Bhattarai, Eggertsson, & Gafarov, 2014; Hamilton

& Wu, 2011). Most empirical studies on QE have used event studies for their analysis and estimates differ across studies depending on the response window around policy announcements applied.

Bhattarai et al. (2014) show that QE acts as a signaling tool, where open market operations shorten the duration of outstanding government debt, which in turn affects the incentives of central banks to keep real interest rates low in the future. Thus, QE makes promises of future expansionary policy

"credible", as central banks want to avoid balance sheet losses. In a liquidity trap, QE can be effective to fight deflation and output gaps, by lowering future expected real interest rates and thereby real long-term interest rates.

(18)

17 2.3. Transmission channels and evidence

The ultimate goal of QE is to stimulate nominal spending and inflation, which can work through a number of transmission channels (Figure 2). There is a broad consensus that monetary policy through the balance sheet is mainly transmitted via the signaling channel and the portfolio balance channel.

But also the bank lending channel, the wealth channel, and the exchange rate channel play an important role in the transmission process (Joyce, Tong, & Woods, 2011).

Figure 2 Transmission channels of QE based on Federal Reserve Bank of New York (2015)

The interest rate important for stimulating investment and consumption is the long-term expected real interest rate which is determined by three components: average expected short-term interest rates, a term, and/or risk premium, and expected inflation (Duprat, 2015b). While the signaling channel mainly affects expected policy rates, the portfolio balance channel works through the reduction of term premia and overall risk premia. Lower interest rates stimulate investments and bank lending, lower exchange rates and lead to higher aggregate demand and inflation (Joyce et al., 2012).

(19)

18

Through the signaling channel, QE measures can help to keep inflation expectations at an optimal level by reinforcing the low path of short-term interest rates. When the central bank conducts asset purchases, especially with sizeable effects on its balance sheet, it is signaling a strong commitment to add monetary stimulus. The aim to meet an inflation target may further signal to the public that policy rates remain low for longer than expected. Central bank communication about their monetary policy stance is a crucial part of the transmission mechanism and can provide additional stimulus or help to align public expectations (ECB, 2015d). Increases in inflation expectations result in lower real long-term rates and support investment and consumption. Forward guidance can further reduce uncertainty, leading to lower interest rate volatility and through this channel also lower risk premia (Filardo & Hofmann, 2014).

Via the portfolio balance channel, QE lowers yields on a broad range of assets. Portfolio balance effects are often divided into scarcity or local supply effects and duration risk (Joyce et al., 2012).

Central bank asset purchases of long-term private or public debt, lead to a scarcity of those assets, which will increase prices and put downward pressure on long-term interest rates. When the liquidity received is not considered a perfect substitute for the assets sold, investors will try to rebalance their portfolios and replace the received short-term central bank deposits with riskier assets. In order for this to happen, the assets purchased by the central bank must differ largely in their risk characteristics in terms of liquidity, duration and credit risk. When the central bank buys riskless and liquid assets with long maturity such as long-term government bonds against reserves, they take the duration risk out of the investors’ portfolios. Investors will replace the assets purchased with securities of similar characteristics, leading to an increase in asset prices until a new equilibrium is reached which will, in turn, lower yields and the costs of external financing (ECB, 2015e).

The credit channel influences the credit supply and is divided into the balance sheet channel (borrowers balance sheet) and the bank lending channel (supply of loans by banks) (Al-Eyd &

Berkmen, 2013). Central bank asset purchases result in a rise in bank reserves held at the central bank and at the same time an increase in customer deposits. The rising asset prices and higher level of liquidity could encourage the banking sector to increase the supply of loans (Joyce et al., 2011).

(20)

19

The wealth channel which is often linked to the confidence channel, works through the effect of higher asset prices. Higher asset prices increase the net wealth of the holder, while lower yields mean lower borrowing costs for households and companies, which stimulates spending. If QE is believed to improve the economic outlook and inflation rates, rising inflation expectations result in lower real interest rates and the boost in consumer confidence supports spending. The increased confidence can translate into lower risk premia which leads to higher asset prices (Joyce et al., 2011).

Through the exchange rate channel, QE affects the expectations of inflation and exchange rates.

Lower interest rates and higher inflation should result in a currency depreciation. The effect can also be attributed to the portfolio balance channel, if rebalancing involves purchases of foreign assets in search for yields. This should then put upward pressure on the prices of those assets and lead to a depreciation of the domestic currency (Blot, Creel, Hubert, & Landondance, 2015).

Studies have tried to evaluate which transmission channels of QE are most important. Krishnamurthy and Vissing-Jorgensen, (2011) test a number of different transmission mechanisms of QE such as the duration risk, the safety premium, default risk, and the liquidity and inflation channel. They suggest that QE in the US mainly worked via the signaling channel and a reduction of the safety premium, which lowered yields on safe assets. They find no evidence in support of the duration risk channel.

Christensen and Rudebusch, (2012) analyze the declines in US and UK government bond yields and find that the relative importance of signaling and portfolio balance channels of QE depend on market specific structures and central bank communication strategy. They suggest that the Fed's QE mainly worked through the signaling channel, whereas the portfolio balance channel was more important in the UK.

2.4. Natural rate of interest: why are interest rates so low?

Over a century ago, Wicksell (1898) discussed the difference between the observed market interest rate and the unobservable ‘natural' rate of interest which determines an equilibrium between optimal savings and investment at full employment. The natural rate of interest is the real, or inflation-

adjusted, interest rate that leads to stable prices. When the natural rate is above the market interest

(21)

20

rate, capital accumulation, and inflation increases; contrary, when the natural rate of interest is below the market rate, capital accumulation decreases while deflation occurs (Duprat, 2015b).

Woodford (2003) later revisits the concept of the natural interest rate in the light of contemporary economics. Using a modern New Keynesian model, he shows how the natural rate of interest responds to economic shocks or changes in consumer preferences. Furthermore, he explains that central banks can conduct monetary policy through the application of a Taylor Rule, which links the policy rate to measures of economic activity and prices, and thereby lead the economy toward the natural rate and price stability.

The natural rate of interest has since become a benchmark for monetary policy as central banks can stimulate or slow the economy by setting their key rates below or above the natural rate. However, because the natural rate of interest rate is unobservable and changes along with economic trends over time, it cannot be measured directly. Moreover, estimates are highly uncertain as accurate models are still to be found (Duprat, 2015b).

Economic theory suggests that the natural rate of interest has been declining over the last decade as savings have outweighed investments. Proponents of secular stagnation argue that because of changes in demographics and deleveraging the natural rate of interest is even negative for most advanced economies (Duprat, 2015b). For instance, Eggertsson and Krugman (2012) show that a large deleveraging shock can drive down the natural rate of interest. In contrast, Bernanke (2015) argues in favor of the global savings glut hypothesis, that excess savings has to do with unfavorable exchange rates and fiscal policies. In his view, Asia and the Eurozone are currently the main contributors to this trend. Whatever the underlying economic fundamentals, both theories conclude that excess global savings have put downward pressure on market rates (Duprat, 2015a).

Until recently, academic thought agreed: interest rates cannot be set below zero, otherwise holding cash becomes more attractive than investing in bonds. At the zero bound, central banks must therefore resort to unconventional measures such as QE, to prevent deflation and stimulate the economy. The discussion about the ZLB returned when some central banks – including the ECB and

(22)

21

BOJ – set their key policy rates at negative levels, making it costly for banks to hold excess reserves. In fact, the Taylor Rule fell deeply below the ZLB for most economies after the financial crisis, suggesting that central banks should set negative policy rates. This would imply that the ZLB no longer binds (Dupor, 2015; Duprat, 2015b).

The effects of negative rates are mostly comparable to those of very low but positive rates. Negative deposit rates should encourage banks to buy alternative assets, which would by increasing the prices of those assets, put downward pressure on yields and borrowing costs. When real interest rates are kept below the neutral level, negative nominal rates can stimulate consumption and investment. They also boost exports, as declines in domestic interest rates lead to a depreciation of the currency and prevent capital inflows to safe-haven currencies (Worldbank, 2015).

Due to the costs of carry for currencies, the actual lower bound is not zero, but slightly negative. The reason is that there are other benefits of keeping money in a bank than earning interest, such as the insurance against theft or physical destruction. Furthermore, large amounts of currency are

inconvenient to use for big transactions and are expensive to safeguard. Thus, the cost of holding cash is what actually defines the effective lower bound on the policy rate. If central banks were to lower interest rates too far, there is a risk of significant cash holdings by large sectors of the economy (Duprat, 2015b).

2.5. Modern monetary policy theory and QE

It follows that central banks have cut their key policy rates to near zero in the attempt to stimulate the economy by following the natural rate of interest down. However, because of the severity of the shock, interest rates might not have been able to fall far enough to reach the natural level (Duprat, 2015b). Eggertsson and Krugman (2012) show that a large deleveraging shock can push the economy into a liquidity trap, where besides ultra-low interest rates, nominal rates remain too high, given the inflation expectations. The main challenge of central banks in a liquidity trap situation is therefore to raise inflation expectations. It is believed, that by targeting the amount of excess reserves and increasing the monetary base, QE can be effective.

(23)

22

But while QE has proven effective empirically, showing theoretically that QE has real economic impact is more difficult. In a New-Keynesian model, Krugman (2014) shows that in a liquidity trap situation, an increase in the money supply does not lead to an equivalent increase in the price level when future money supply is unchanged. The problems with QE is that first, a large expansion of the monetary base under QE is inherently temporary. This is because central banks explicitly committed to not allow higher inflation than their long-run targets and signaled a normalization of the size of their balance sheets by communicating potential exit strategies. And second, for QE to increase demand at the ZLB, monetary base growth needs to be permanent. From a Wicksellian view, this would imply a persistent increase in the price level in the long-run which raises inflation expectations and thus lower real interest rates to their market clearing level (Cohen-Setton, 2015; Woodford, 2012).

Woodford (2012) gives two suggestions how central banks can stimulate the economy at the ZLB in his famous Jackson Hole paper. He argues, that central banks’ signals to return monetary bases back to levels seen before QE programs contradict the intentions of QE and lead to a weakening of inflation expectations. For QE to be effective, central banks would need to promise a permanent increase in reserves and allow a temporary rise in inflation. He suggests that central banks should target GDP levels instead of inflation and thus commit to returning nominal GDP to the trend path that it had prior to the crisis. This would ensure that monetary easing is to remain for longer than usual Taylor Rules that link interest rates to GDP growth and inflation would imply, and at the same time make sure that the central bank does not accept inflation rates above their target.

Woodford (2012) further argues that the only policy that would create inflation is "helicopter money".

The concept was first introduced by Milton Friedman (1969) and picked up later by the Chairman of the Fed, Ben Bernanke. Helicopter money implies that central banks print money and distribute the cash directly to consumers to increase aggregate demand and inflation. The policy is differentiated from QE, as it is directly distributed to households rather than banks and since it is not easily reversible, it is perceived as a permanent increase in money supply. Bernanke (2002) suggested its implementation via transfers to households and businesses through a tax cut, combined with incremental purchases of government debt and finance the tax cut via money creation. Woodford

(24)

23

(2012) proposes a bond-financed fiscal transfer. The immediate transfer to households would make the policy as simulative as helicopter money while preserving the separation between monetary and fiscal policy. The crucial point is for people to understand that the fiscal stimulus is permanently financed with central bank money, so no concerns about Ricardian equivalence costs of an increase in government debt burden in the future arise (Reichlin, Turner, & Woodford, 2013).

Yet, there remains a tension between theory and empirical evidence on the effectiveness of QE at the ZLB. The research suggests that asset purchases alone cannot help overcome the zero-lower bound or liquidity trap situation but rather that QE and forward guidance reinforce each other. Asset purchases help align market expectations with central banks’ future monetary policy stance and may signal a strong commitment by “putting money where the mouth is”. Nevertheless, in order for the policy measures to be credible, they need to be perceived as sufficiently costly and permanent (Woodford, 2012). Unconventional policy measures are still not fully understood as they do not fit well into classic monetary theory. Specifically, because they are unprecedented, the differences in the

implementation of QE programs across central banks can provide guidance about the optimal design features and their effects. In the following, the paper aims to shed some light on QE and its efficacy with a brief summary of the different QE programs and empirical literature on this topic.

(25)

24

3. Quantitative easing implementation and evidence

The aim of this chapter is to give an introduction to the central banks’ monetary policy responses to the financial crisis and the influence on their respective economies. The short overview of the QE programs in Japan and UK is followed by an extensive review of the Fed’s QE measures to outline major differences between countries. The initial purpose of the programs was to ease distressed markets, but the challenges posed by the financial crisis soon shifted the focus to financial stability and targeting inflation. The relevance of the BOJ’s program arises from its longevity, and facing similar deflationary pressures as the ECB, as well as being a bank-centric economy. The BOE was the first European institution to implement QE, which just like the Fed, mostly focused on bond purchases reflecting the importance of bond markets.

3.1. Bank of Japan

The first monetary balance sheet expansion, now known as QE, was initiated by the BOJ on March 19th, 2001. Japan had been faced with low economic growth, prolonged deflation, and a significant amount of non-performing loans in the banking system left since the Asian financial crisis in 1997. In response to an economic downturn after the IT bubble burst, the BOJ adopted a new monetary easing framework, a quantitative easing policy (QEP) as well as forward guidance. QEP consisted of three parts: a commitment to keep low interest rates until inflation rates stabilized; an increase in the balance sheet and a change in the composition of the central banks’ balance sheet through purchases of government debt (Rogers, Scotti, & Wright, 2014).

The BOJ changed its main operating target from the uncollateralized overnight rate to the outstanding balance of the current account balances (CAB) held by financial institutions at BOJ. Initially, the target for bank reserves was increased from ¥4 trillion to ¥5 trillion, with the aim of bringing the overnight call rate to zero (Fawley & Neely, 2013). The BOJ stated that if necessary, the amount of long-term government bond purchases would be increased up to the ceiling of the outstanding balance issued to ensure liquidity (Shiratsuka, 2010). By 2004, the BOJ had progressively raised its target for bank reserves to ¥30 to ¥35 trillion in response to the decline in economic activity, while at the same time purchasing public and private debt (Fawley & Neely, 2013).

(26)

25

In order to meet the CAB target, the BOJ further increased the outright purchases of long-term government bonds (JGB’s) from ¥400 billion per month initially to ¥1,200 billion per month, starting October 2002. The extensive liquidity provision under QEP led the uncollateralized overnight call rate to fall to 0,001% and in 2005, the CPI inflation turned positive. With all three objectives met, the BOJ decided to terminate the QEP on March 9, 2006, and return the operating target of money market operations to its main policy instrument of the uncollateralized overnight call rate, while maintaining the policy rate at zero (Shiratsuka, 2010).

Following the bankruptcy of the Lehman Brothers and the resulting financial market turmoil, the BOJ announced special funds-supplying operations (SFSOs) on December 2, 2008, which entailed lending unlimited amount of capital to commercial banks at a near zero rate. The SFSOs offered 3-month loans to banks at the overnight call rate at 0,3% which was later reduced to 0,1%, with the amount of available collateral being the only limit to the size of the loans. On December 19, 2008, the BOJ increased its monthly purchases of JGBs for the first time since 2006 and introduced a new program to purchase corporate financial instruments. Until then, the BOJ had mainly focused on bank loans to specifically support its struggling banking sector after the financial crisis reflecting the bank-centric structure of the financial system (Fawley & Neely, 2013).

In October 2010, the BOJ announced a comprehensive monetary easing program, the Asset Purchase Program (APP) consisting of three pillars: lowering the target for the uncollateralized overnight call rate to 0-0.10%, a commitment to maintain zero interest rates until the inflation target was reached, and establishing the APP. The aim of the APP was to reduce term and risk premia through purchases of government securities, commercial paper, corporate bonds, exchange-traded funds (ETFs), and Japanese real estate investment trusts (J-REITs). The BOJ purchased a wide array of private assets and not only public debt to reduce the spread between private and sovereign debt yields. The BOJ initially set the size at ¥35 trillion but increased it to ¥101 trillion by December 2012 (Fawley & Neely, 2013).

Following in April of 2013 the qualitative and quantitative easing was introduced which increased purchases of JGBs from ¥20 to ¥50 trillion a year and extended the average maturity from three to

(27)

26

seven years. Further, for the next two years, the BOJ committed to extending the monetary base of

¥60-70 trillion per year while aiming at reaching the inflation target of 2% (Rogers et al., 2014).

In a surprise announcement in October 2014, Japanese policymakers again expanded the QE program, raising the purchases of government bonds to ¥80 trillion to avoid a return to deflation (Monaghan &

Wearden, 2014). Most recently, the BOJ introduced negative rates on required reserves as an attempt to keep the economy from falling back into stagnation (Nakamichi, Fujikawa, & Warnock, 2016).

3.1.1. Empirical evidence

Empirical evidence on the effect of the BOJ’s unconventional monetary measures suggests that the impact on output and inflation were rather small. In a comprehensive survey on QE policies research, Ugai (2006) suggests that QE led to a reduction in yields but its impact on output and inflation was rather limited. Okina and Shiratsuka (2004) examined the effects of the policy commitment of maintaining zero interest rates by looking at the policy duration effect and find that the policy was highly effective in stabilizing market expectations about the future path of short-term interest rates and flattening the yield curve. However, they also acknowledged that the monetary easing effects failed to be transmitted to the whole economy and could not reverse deflationary expectations in financial markets. These disappointing results have largely been attributed to a dysfunctional banking sector which impaired the transmission channels. Schenkelberg, Heike, Watzka (2011) confirm that while QE was successful in stimulating real activity in the short-run, it did not have any effect on inflation. Since they do not find significant decreases in long-term yields or exchange rates, they suggest that the monetary policy measures did not have a direct quantity effect such as the portfolio rebalancing channel.

There is limited evidence on the impact of the latest QE implementation. However, recent financial and macroeconomic developments suggest they are rather small. After some major improvements in inflation following the announcements in 2014, the outlook worsened again in 2015. Despite its bold easing measures and negative interest rate policy, inflation expectations remain weak (Gros et al., 2015; Harding, 2016)

(28)

27 3.2. Bank of England

In response to the global financial crisis, the Bank of England (BOE) implemented both conventional and unconventional monetary policy easing. Between October 2008 and March 2009, the BOE’s Monetary Policy Committee (MPC) lowered its policy rate from 5% to 0,5%, its effective lower bound.

To meet the inflation target of 2%, the MPC had also initiated the Asset Purchase Facility in January of 2009, with the aim of easing specific credit conditions through private asset purchases and to provide monetary stimulus through QE (Rogers et al., 2014). The first purchases were limited to £50 billion in private assets and corporate bonds and were intended to increase liquidity by making corporate credit more available. Since the asset purchases were financed and matched directly by the issuance of short-term government securities (gilts), the BOE’s monetary base did not increase initially and can thus not be considered QE. Similarly, the BOE purchased corporate bonds through a reverse auction.

The measures did however not have the desired effect on the economy; asset holdings were only 6%

of the ceiling of £50 billion (Fawley & Neely, 2013).

In March 2009, the BOE decided that it would need to ease monetary conditions further announcing a

£75 billion APF to purchase public sector assets with a residual maturity of five to five to 25 years. The size of the APF was expanded from £75 billion to £200 billion in February 2010, to meet a level of demand consistent with the inflation target (Rogers et al., 2014). This time, the BOE financed the APF purchases with central bank reserves thereby increasing the monetary base (Fawley & Neely, 2013).

Almost two years later, the MPC resumed increasing the QE target gradually to £275 billion in October 2011, reaching £375 billion in July 2012, in fear of missing the inflation target. In addition, the BOE authorized purchases of private assets financed by the Treasury up to £10 billion, which however have not been reached until now. In August 2013, a forward guidance structure was introduced with the aim of clearly communicating the monetary policy stance, thereby providing confidence for households and businesses, and also to meet the unemployment threshold of 7% (Rogers et al., 2014). There have been no further asset purchases since 2012, but any funds in relation to purchased bonds maturing have been reinvested and the total QE stock of £375 billion will remain at least until the desired inflation target is achieved (BOE, 2016).

(29)

28 3.2.1. Empirical evidence

There is considerable evidence in support of the efficacy of the unconventional policy measures in the UK. Interbank borrowing rates fell sharply in 2008, government and corporate bond yields decreased, capital market borrowing conditions improved, and economic growth returned in late 2009 (Joyce et al., 2012). Using event study analysis, Meier (2009) finds that the initial QE announcement led to a decline in gilt yields by 35-60 basis points, while Joyce et al. (2011) estimate that the medium-to-long- term gilt yields decreased by 100 basis points overall. They further report a fall in corporate bond yields and similar effects on the sterling exchange rate. Glick and Leduc (2011) and Meaning and Zhu (2011) on the other hand, find a smaller effect on gilt yields around 50 basis points, which may reflect the shorter time window of 1-day used in order to measure the announcement effects, in contrast to a 2-day window used by Joyce et al. (2011). Using empirical dynamic term structure models,

Christensen and Rudebusch (2012) find that the gilt yield declines on seven key UK QE announcement dates were driven by declines in term premia, implying a predomination of the portfolio balance channel.

Despite the challenges involved in estimating the wider macroeconomic effects of unconventional monetary policies, growing literature has started to investigate the impacts using structural VAR models (Joyce et al., 2012). Using time-varying parameter structural VAR, Baumeister and Benati (2010) find that in the absence of QE, output growth would have fallen to -12% and that inflation would have been negative at -4%. Using a similar approach Kapetanios, Mumtaz, Stevens, and Theodoridis (2012) construct counterfactual scenarios assuming that the QE programs reduced gilt spreads and find that with QE the level of real GDP was 1,5% higher and inflation 1,25% higher than without QE. However, the impact of QE in the model might have captured other effects as well (Kapetanios et al., 2012).

3.3. US Federal Reserve Bank

Since the onset of the financial crisis in 2007, the Federal Reserve introduced several unconventional policy measures to stabilize financial markets. The Fed follows a clear dual mandate as its monetary policy objective: maximum employment and stable prices. The dual mandate decisively guided the Fed’s policy decisions during the depression.

(30)

29

The Federal Open Market Committee (FOMC) began cutting the federal funds rate in September of 2007, which was revised down in April 2008 from 5,25% to 2% as the crisis unfolded. As part of several emergency interventions, the Fed started selling its holding of Treasuries and buying less- liquid assets to provide more liquidity. Furthermore, the Treasury began to deposit excess funds – borrowed from the Fed – which enabled the Fed to purchase more assets without increasing bank reserves. With the bankruptcy of the Lehman Brothers by late 2008, the FOMC brought the federal funds rate close to zero, its effective lower bound (Blinder, 2010).

As a result, conventional monetary policy was ineffective, and the Federal Reserve turned to LSAP’s as well as forward guidance about the future path of monetary policy. On November 25, 2008, the Federal Reserve announced purchases of government-sponsored enterprise (GSE) debt of $100 billion and the issuance of $500 billion worth of Mortgage-backed securities (MBS). These were soon to be increased by another $100 billion in GSE debt, $750 billion in MBS, and $300 billion in long-term Treasury securities in March 2009. The first round of LSAPs, commonly referred to as “QE1”, was mostly intended to support the struggling housing credit markets but also to improve financial market conditions in general (Fawley & Neely, 2013).

At the end of 2010, faced with the risk of disinflation, the Fed resumed its asset purchases with

reinvestment arrangements of the principal payments on LSAP assets into longer term Treasuries. The Fed’s second LSAP program, QE2, comprised purchases of $600 billion in long-term Treasury securities and lasted until June 2011 (Rogers et al., 2014).

Renewed fears of a recession in the fall of 2011 forced the Fed to introduce a third round of

government bond purchases, which got known as “Operation Twist” since the Fed sold $400 billion in short-term assets while buying $400 billion in long-term assets at the same time. The program aimed at reducing long-term interest rates relative to short-term interest rates and did not expand the monetary base because the purchases were offset by the short-term asset sales. Even after an extension of the Operation Twist with purchases up to $267 billion in long-term Treasuries in June 2012, economic growth remained sluggish, and the US labor market stagnated (Fawley & Neely, 2013).

(31)

30

In September 2012, the FOMC introduced a third QE program (QE3) which included purchases of at least $40 billion MBS per month initially. Starting December 2012, long-term Treasuries were bought at a monthly rate of $45 billion and were no longer funded with sales from short-term Treasuries. QE3 was state-contingent and open-ended, meaning that the program would last at least until

improvements in labor market conditions were achieved. Until the fall of 2014, the Fed had maintained the level of securities holdings from the LSAPs by reinvesting principal payments from maturing assets. The Fed tremendously expanded its monetary base with the four distinct QE programs to over $4.4 trillion at the end of October 2014, comprising $1.567 trillion in long-term government bonds, $1.41 trillion in MBS, $175 million in GSE (Fawley & Neely, 2013).

The FOMC also provided forward guidance about how long it would keep the target funds rate at low levels with the aim of steering long-term interest rates. In addition to its calendar-based guidance on the future path of the policy rate, the Fed used threshold-based forward guidance, which provided information about the economic conditions that would ensure an unwinding of the monetary easing.

In January 2012, the FOMC expressed its long-run inflation target of 2% and the desired rate of unemployment around 5.2-6%, while specifying how its key policy rate could be used to achieve its dual mandate. In a clear statement in December of 2012, the FOMC announced it would keep the federal funds rate exceptionally low until long-run goals of maximum employment and the inflation target of 2% were met (Rogers et al., 2014).

After seven years of near-zero interest rates, the Fed announced in December of 2015, that it would start a gradual tightening path. The federal funds rate was raised up to range between 0.25% and 0.50%. However, the Fed left its strategy open for adjustments in case the economic outlook worsens (Fed, 2015).

3.3.1. Empirical evidence

The Fed’s balance sheet expansion officially ended with the completion of the tapering process in October 2014, but it will still take time until normalization is achieved and thus until a full evaluation of the impact of unconventional policy measures is possible. However, it is possible to assess which features of the QE program were effective and to what extent it helped to realize the monetary policy

(32)

31

goals (Rosengren, 2015). A number of studies have examined the effectiveness of large-scale asset purchases and forward guidance policies at the ZLB, which will be discussed in more detail in the following.

3.3.1.1. Transmission channels

QE in the US was believed to work through the portfolio balance channel and signaling channel. A number of empirical studies use lower-frequency data to investigate the portfolio balance effect by modeling the relationship between long-term Treasury yields, term premia, and the maturity structure of the public’s holding of Treasury debt (Gagnon et al., 2011; Hamilton & Wu, 2011;

Krishnamurthy & Vissing-Jorgensen, 2011). Gagnon et al. (2011), estimate that QE1 reduced the term premium by about 52 basis points and the 10-year Treasury yield by about 82 basis points. Hamilton and Wu (2011) estimate a three-factor term structure model to predict the effects of changes in the maturity structure of US debt from asset purchases and find similar but smaller effects than Gagnon et al. (2011). D’Amico and King (2013) find that QE1 and QE2 operated through scarcity and duration channels, but find no evidence of a signaling channel. Krishnamurthy and Vissing-Jorgensen (2011) on the other hand, find no evidence in support of the duration risk channel but conclude that QE effects on safe and risky assets were likely transmitted through the signaling channel. Also, Bauer and Rudebusch (2014) argue that previous studies wrongly attributed changes in term premia solely to portfolio balance effects and find that signaling effects for the LSAP’s were economically and statistically significant.

3.3.1.2. Forward guidance policy

The Fed used both calendar-based and threshold-based forward guidance in the past. Calendar-based forward guidance informs about the intended path of it policy rate over a certain period, while

threshold-based forward guidance is contingent on certain economic conditions and recognizes that the policy rate adjusts with changes in those factors (Campbell, Fisher, Evans, & Justiniano, 2012).

Williams (2012) argues, that the statement on August 9, 2011, where the FOMC announced it

“anticipates that economic conditions... are likely to warrant exceptionally low levels for the federal funds rate at least through mid-2013” successfully shifted market expectations. The introduction of

(33)

32

calendar-based forward guidance led to a fall in Treasury yields by one- and two-tenths of a percentage point, which was suggested a considerable drop (Williams, 2012).

A few studies have tried to assess the impact of forward guidance policies when the interest rate is at the effective lower bound. For instance, Campbell et al. (2012) show that surprises associated with FOMC policy announcement substantially influence Treasury bond rates, corporate borrowing rates, and private macroeconomic forecasts. Moessner (2013) studies the impact of explicit FOMC policy guidance on short- to medium-term interest rates and finds that forward guidance successfully flattened the yield curve for both Eurodollar futures and US Treasuries, by managing expectations about future monetary policy. While in theory, threshold-based forward guidance can, by providing more clarity about the conditionality, lead to increased credibility of the central bank, it is also more complex and therefore difficult to convey. The empirical evidence confirms that conditional forward guidance had less impact on futures rates and long-term bond yields (Filardo & Hofmann, 2014).

A recent study by Gavin, Keen, Richter, and Throckmorton (2015) examines the Fed’s forward guidance via news shocks to the monetary policy rule in a New Keynesian model at the ZLB. They conclude that forward guidance can stimulate economic activity but only if households believe the economy will recover; otherwise, interest rates will remain near zero. Thus, the current economic outlook likely influenced the effects of the FOMC communications.

The empirical results suggest that the Fed's forward guidance about future monetary policy

successfully reduced long-term interest rates. However, it remains unclear, to what extent the decline was due to the Fed’s communication and how much was due to simultaneous changes in economic conditions. Also, the magnitude of large-scale asset purchases potentially facilitated the credibility and thus the effectiveness of the Fed’s forward guidance. Since the policies were mostly conducted at the same time, it is difficult to isolate their market impacts (Engen, Laubach, & Reifschneider, 2015).

3.3.1.3. Impact on government bond rates

Several event studies have tried to evaluate the effects of the QE programs using a narrow time window around the policy announcements. There exists considerable evidence from QE1, suggesting that the program reduced 10-year government bond yields by about 100 basis points; Krishnamurthy

The Quantitative Easing Experience