Sitting on the Money
An investigation into the role of the bank-lending channel and its
theoretical and practical effectiveness in recent quantitative easing
programs
Bachelor Thesis
University of Amsterdam
Absract
This thesis examines quantitative easing and the role that the bank-lending channel has played in recent programs. Little research has yet been done into this specific channel. Research has shown a small but significant increase in bank-lending growth in Japan and no increase in the United Kingdom. Research concerning the Federal Reserve’s programs has yet to be done. The efficacy of the bank-lending channel is one of many concerns that economists have about quantitative easing. This thesis looks into the programs implemented in Japan, the US and the UK and concludes that the different strategies that were applied resulted in different outcomes, though generally a small positive effect has been acknowledged. As for the effectiveness of the bank-lending channel: studies suggest that it is either small or non-existent. This thesis finds that the effect of the bank-lending channel is constrained, especially during crises, by pressure to deleverage banks balance sheets and flightiness of deposits. Furthermore it finds that the efficacy of the bank-lending channel might be dependent on the implementation strategy of the quantitative easing program and it gives suggestions for implementation strategies that might actively engage the bank-lending channel.
Date: June 11
th2015
Program:BSc. Bèta-Gamma
Track: Major Economics
Name: M. Ruben de Roos
Student Number: 6056490
Supervisor: Egle Jakucionyte
Statement of Originality
This document is written by Ruben de Roos who declares to take full
responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and
that no sources other than those mentioned in the text and its references have
been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision
of completion of the work, not the contents.
Content
1. Introduction p.4 2. Quantitative easing: How it works p.6 2.1 The policy-‐signalling channel p.6 2.2 The expectations channel p.7 2.3 The portfolio-‐balancing channel p.7 2.4 The bank-‐lending channel p.8 2.5 The liquidity channel p.8 3 The concerns about quantitative easing p.8 3.1 Capital flight p.9 3.2 Excessive inflation p.9 3.3 Exit strategy p.9 3.4 Unequal distribution of wealth effects p.9 3.5 The inefficacy of the bank-‐lending channel p.10 4 Past implementations of quantitative easing policies p.10 4.1 Quantitative Easing Policy in Japan p.11 4.2 Quantitative Easing Policy in the United Kingdom p.12 4.3 Quantitative Easing Policy in the United States p.13 5 Historical efficacy of the bank-‐lending channel p.14 5.1 The bank-‐lending channel in Japan p.15 5.2 The bank lending channel in the United Kingdom p.15 6 Analysis p.17 7 Conclusion p.19 8 References p.20 9 Appendix p.22
1. Introduction
On January 22th 2015, the director of the European Central Bank (ECB) Mario Draghi announced that the ECB would commence with a large asset-‐purchasing program. This program would entail the purchasing of assets, including bonds issued by national governments and EU institutions such as the European commission, amounting to €60 billion per month. It will continue until the end of September 2016 and will total €1.1 trillion euro’s. This type of program is known as the unconventional monetary policy called quantitative easing (QE). This policy is aimed at (in the European context) stimulating inflation (warding off deflation) and “kick-‐starting” the economy. How does this policy help resolve these issues and why would the ECB consider this to be a valid policy choice?
QE, although unconventional, has had several test runs in the past two decades. It was first implemented in Japan by the Bank of Japan (BoJ) in 2001 due too Japans long-‐ standing issue with deflation. After that is was adopted in the United States in 2008 and in the United Kingdom in 2009. The effectiveness of these policies is still in dispute: many economists are doubtful about some of the short-‐ and long-‐term effects of QE policy.
One of these concerns is whether bank lending contributes to the positive effects of QE or not. The bank-‐lending channel is a transmission channel of QE that entails banks using the extra liquidity injected by QE to expand their loans. Thus I pose the question in this thesis: What has been the role of the bank-‐lending channel in recent QE programs and what were its constraints? To answer this question we look into research done on this topic concerning the recent ventures in Japan, the UK and the US.
It should first be noted that little research has been done specifically on the topic of the bank-‐lending channel in QE programs. This can be partially attributed to the fact that these programs have only just recently been implemented1. However it is also possible,
1 As commented by Joyce et al. (2012):“There is inevitably uncertainty over the size and duration of
the effect and the precise channel through which it works, although this is perhaps to be expected given the relatively short historical period we have at our disposal and the extreme nature of the recession which makes thinking about counterfactuals challenging.”
as we are able to surmise from the literature, that the bank-‐lending channel as a transmission mechanism within QE has, a priori, been deemed either non-‐existent or of little consequence. This is also the estimation of Joyce and Spaltro in their paper Quantitative Easing and bank lending: a panel data approach (Joyce and Spaltro, 2014). In their conclusion they state: “To our knowledge, this is the first paper to try and quantify the effect of QE on bank lending in the UK using a bank-‐level panel dataset.”2
This brings us to a second issue that we encounter when researching the bank-‐lending channel: To accurately study changes in bank lending due to QE, bank-‐level data is needed to perform a proper regression. The problem is that this kind of data is only accessible to central bank researchers and thus not publically available. This might also partially explain why little research has yet been done in the bank-‐lending channel. Since the bank-‐lending channel is one of the “key transmission channels” (Benford et al., 2009) of QE, it seems advisable that more research should be done into this topic.
The remainder of this thesis is structured as follows. First I will describe the conceptual workings of QE, thus establishning a theoretical framework wherein we can view the bank-‐lending channel. I will then address some of the concerns held by economists, with the bank-‐lending channel among them. This will give us some perspective on the problem of the bank-‐lending channel. I then take a look at the historical implementation of different QE programs and the role that the bank-‐lending channel has played. Finally I will analyse these programs and theories and give some suggestions on how to engage the bank-‐lending channel more actively.
2 “So far, the vast majority of research on QE has focused on its impact on economic growth and
financial markets, while the effect of QE on bank lending has received much less attention. This relative neglect reflects the fact that policymakers in the United Kingdom and elsewhere expected QE to affect demand mainly through its impact on asset prices, while the effect on bank lending was expected to be small because of banks’ incentives to deleverage and reduce the overall size of their balance sheets.”(Joyce and Spaltro, 2014)
2. Quantitative Easing: How it works
In order to supply a proper context for the so-‐called bank-‐lending channel (Kaysap and Stein, 1994) we need to establish what place it has within the QE mechanism.
To stimulate a stable economic growth, central banks use their policies to maintain a low and stable inflation of around 2%. In normal times (in the economic sense) when inflation is too low, central banks lower the interest rate for which they lend money to commercial banks: the instrument rate (in the UK: the bank rate, in the US: the federal funds rate). This will in turn raise the demand and also the supply of money, thus raising the inflation level. However, in times of economic distress, such as the recent global credit crisis or Japan’s “Lost Decade”, this policy mechanism may turn out to be ineffective. When inflation is too low and the bank rate is already near zero any further adjustment to the bank rate will not yield higher inflation. This is known in economic literature as the liquidity trap (Krugman, 2000). To escape this liquidity trap central banks can turn to QE: QE consists of large asset purchases by central banks over an extended period of time.3 By doing so, the central banks directly increase the money
supply of the economy, thus raising expected inflation.
How do these purchases increase the money demand? Several researches identify different transmission channels. Benford et al. (2009), in an article for the Bank of England’s Quartarly Bulletin, identify the portfolio-‐balancing channel, the bank-‐lending channel and the expectations channel as the three “key channels” through which QE operates. Bell et al. (2012), also for the Bank of England, additionally identify the policy-‐ signalling channel and the liquidity channel. We discuss these transmission channels here.
2.1 The policy-‐signalling channel
These policy and purchasing announcements are made explicitly and emphasize the central banks commitment to their targets. This is a quote from Mario Draghi’s announcement: “They [purchases of €60 billion per month] are intended to be carried out until end-‐September 2016 and will in any case be conducted until we see a sustained
3 The specific types of assets purchased differ per program, depending on the specific targets of
the central banks. However the programs usually contain low-‐risk long-‐term assets. This is one of the main differences between QE and normal Open Market Purchases (Blinder, 2010).
adjustment in the path of inflation.” Draghi is hereby signalling that the ECB is committed to this policy and the inflation target. This kind of signalling can change expectations that people have about the markets.
This satisfies two of Svensson’s (2003) three elements for “the optimal way to escape from a liquidity trap and deflation”, which he formulated based on literary research into the liquidity trap and QE. “(1) an explicit central bank commitment to a higher future price level; (2) a concrete action that demonstrates central bank’s commitment, induces expectations of a higher future price level and jump-‐starts the economy;”(Svensson, 2003). The third element mentioned by Svensson (2003) will be discussed later during our analysis of the concerns.
2.2 The expectations channel
In the article by Bell et al. (2012) the expectations channel is identified. It is not clear in what sense this channel essentially differs from the policy-‐signalling channel. The article mentions a consumer confidence boost due to changed expectations about the economic future because of the QE program (Bell et al., 2012). It does not elaborate as to what makes it different from the policy-‐signalling channel and so the distinguishment remains vague.
2.3 The portfolio-‐balancing channel
This channel operates through the asset purchases made by recipients of QE purchases. When the central bank purchases assets, the prices of these assets go up and the yields go down, thus reducing the cost of borrowing. Also, asset owners will become wealthier due to these price increases. This should encourage them to spend more. When asset holders sell their assets to the central bank, they have an excess money balance, which they use to reinvest in other assets so as to regain the same risk-‐profile in their portfolio as before (assuming that their preferences are still the same and they wish to retain that profile). This is a continuing chain of buyers and sellers until finally all asset prices have risen and the amount of money held is in balance with the portfolios. Since yields are dropping, companies and households will likely want to invest in assets with a higher risk in order to retain their risk profile. Furthermore, because of the extra money injected into the economy, money is more readily available and thus becomes cheaper to obtain. Because there is now less danger to run out of liquidity, this might encourage
investors to hold more illiquid assets in their portfolio (Benford et al. 2009). According to Butt et al. (2014), in their empirical research into the bank-‐lending channel, the QE program in the United Kingdom worked mainly through this channel, as they ascertained through the variance levels of bank deposits.4
2.4 The bank-‐lending channel
QE programs will, whether by direct purchases or indirect purchases, increase the reserves of commercial banks. Therefore, there is an increase in the commercial banks liquidity. Banks need to retain a certain balance of liquid stock in order to meet payment demands of their normal dealings. If the bank has more liquidity than needed for its balance, it can use this extra liquidity to extend more loans than it would have done before. These additional loans should meet the demand for money and thus increase consumption and investment (Benford et al, 2009).
2.5 The liquidity channel
The liquidity channel, mentioned by the BoE report, refers to the effect that when central banks inject liquidity into the markets, thus making it cheaper to obtain, this will actively encourage trading because “Asset prices may consequently increase as a result of lower illiquidity premia.”(Bell et al., 2012). Again this seems to be a part of the aforementioned portfolio-‐balancing channel, wherein investors will reinvest in assets with different risk-‐premia due to changing risks.
An overview of the transmission mechanisms of QE, as viewed by the Bank of England, is graphically represented in the Figure 1, which can be found in the appendix.
3. The concerns about quantitative easing
Mario Draghi’s announcement of the European QE program was preceded by a long period of heated debates between the ECB council members. This is because there are still some concerns about QE. In this segment I discuss the main concerns : capital flight, excessive inflation, exit strategy and the bank-‐lending channel.
3.1 Capital flight
There is a possibility that, when liquidity is increased by central banks, investors and financial institutes have the option to reinvest and banks have the option to expand their loans, they might not choose to invest in assets or loan to agents within the country of the central bank (or in Europe’s case, within the European Union). Instead, considering the fragile economy of the homeland, they might use their liquidity to invest abroad in more stable countries. According to Michael Hudson of the Levy Economics Institute of Bard College, this effect did in fact occur during the QE program of the Federal Reserve (Hudson, 2010).
3.2 Excessive inflation
QE might backfire: The goal is to increase inflation, but when the money supply is so vastly expanded there is a danger that this goal might be overshot and lead to hyperinflation (Meaning and Zhu, 2011). “Because with the excess reserves in banks, a massive increase in the money supply through significant increase in their lending or investing could create a galloping inflationary scenario.”(Mahajan, 2015)5
3.3 Exit strategy
When the QE program has been successful and inflation rates are up again, how does a central bank reduce its balance sheet (which is now hugely inflated) without disrupting the market? This is a common concern raised also by Joyce et al. (2012). Would we have fixed the burst of one bubble just by creating another? This is also tied with the concern of whether the “purchasing of government bonds are helping to contribute to unsustainable levels of government debt.” (Joyce et al., 2012). This is where I arrive at Svensson’s (2003) third element of the optimal way to escape the liquidity trap: “(3) an exit strategy that specifies when and how to get back to normal.”(Svensson, 2003).6
3.4 Unequal distribution of wealth effects
Because QE focuses on lowering yields and raising asset prices, this might not be equally beneficial for everyone. Specifically it might even have adversely affected some
5 Mahajan (2015) also comments that this risk mostly concerns less developed economies: “So in
case of developing economy, where inflation remains a major concern quantitative easing type expansionary monetary policy should be adopted with utmost care.”(Mahajan, 2015)
6 Under the leadership of Ben Bernanke the Federal Reserve has devised a seven-‐point exit
strategy. This plan also contains some unorthodox and innovative techniques to avoid market disruption.(Blinder, 2010)
households who do not have large asset investments and thus could not benefit from the increased value of those assets. They may however suffer from a lowered return on their saving deposits. These findings are based on an empirical study done for the Bank of England by Bell et al. (2012). Thus the bank-‐lending channel, which does not work through asset prices, may more effectively target households (who do not hold many assets).
3.5 The inefficacy of the bank-‐lending channel
The recent implementation strategies of QE policies give evidence that there is little faith in the efficacy of the bank-‐channel. Economists theorize that in times of severe economic distress, there is a large amount of pressure on banks to deleverage their balance sheets. Deleveraging means that the relative amount of debt on the balance sheet must be reduced (or the relative amount of capital must be increased). Banks might therefore use the additional liquidity to pay of their debts instead of expanding their lending. “This focus on the importance of bank’s balance sheets is consistent with the literature on the so-‐called bank capital channel, which suggests that capital can be an important driver of banks’ lending decisions particularly in periods of market stress.” 7
(Joyce and Spaltro, 2014).
According to Benford et al. (2009) the bank-‐lending channel is one of the key transmissions channels in QE. As such it can be viewed as part of the foundation that determines the success of a QE policy. Additionally, if it does work, it would probably lessen the concern of unequal distribution of wealth effects. That is why this thesis is focused on this particular concern.
4. Past Implementations of Quantitative Easing Policies
In this section I take a look at how the QE policies were implemented in Japan, the United Kingdom and the United States8
. Although all policies consisted of expanding the central banks balance sheets, there are some important differences between the implementation
7 Joyce and Spaltro refer to Kaysap and Stein (1994), van den Heuvel (2002), Jimenez et al.
(2010) and Gambacorta and Marquez-‐Ibanez (2011).
8 Although no literature has been found on the bank-lending channel in the US, I do add a description
of the implementation process in the US because it will broaden our understanding of the practical possibilities and limitations of QE, which may also affect our understanding of the bank-lending channel.
strategies. This will give us a sense of the scale and focus of the different programs, which might help us understand the resulting effects of the bank-lending channel.
4.1 Quantitative Easing Policy in Japan
In the 1990’s, Japan was struck by low economic activity and consumer price deflation after the asset-price bubble burst in 1991. This decade of economic depression is known as the “Lost Decade”. To combat this deflation, the Bank of Japan reduced its instrumental rate. By 1999 this rate was reduced to zero, yet it did not manage to stop deflation (Bowman et al., 2011). The Japanese economy was stuck in the liquidity trap. Then, in 2001, the global IT bubble burst. The Japanese economy had still not recovered from the Lost Decade. So the BoJ decided to use unconventional policy measures to boost the economy and launched the - “..the only relevant historical precursor.. [to the Federal Reserves Program]”(Blinder, 2010) - Quantitative Easing Policy (QEP).
The QEP was three-pronged: (1) The BoJ changed its main target from the instrument rate to the current account balances (CABs) held by financial institutions at the BoJ. (2) To do this the BoJ heavily increased its purchases of government bonds (including long-term JGBs) and some other assets. (3) The BoJ signalled that it was committed to “maintain the QEP until the core CPI {…} stopped declining” (Bowman et al., 2011). From 2001 to 2006 the BoJ expanded its holdings of Japanese Government Bonds (JGBs) by 63 trillion yen, which amounted to 4 percent of the GDP (Gagnon et al., 2011)9
.
According to Gagnon et al. (2011) it was the BoJ’s intention for the QEP program to operate mainly through the bank-lending channel. Gagnon et al. (2011) claim that the BoJ’s purchases were not necessarily focused on long-‐term JGBs: “…the BoJ did not claim that purchases of JGBs would reduce longer-‐term interest rates. Rather, JGBs were viewed as an appropriate and convenient asset for the BOJ to buy in order to supply banks with liquidity.” Because of this attitude, BoJ’s holdings of JGBs were skewed more toward short-‐term bonds. This view is also supported by Joyce et al (2012). However, this contradicts an article by Blinder (2010) in which he states that, in contrast to the Federal Reserve, the BoJ focused on reducing term-premiums mainly by purchasing long-term
9 Martin and Milas (2012) also support this claim that the QEP was focused on short-‐term money
JGBs.10
The literature’s judgment about whether the QEP was successful or not remains vague. Many articles mention that the QEP did have some positive effects on the Japanese economy, but that these were small and did not succeed in recovering Japans economy. “When gauging the effects of the QEP broadly, the results show limited effects on raising aggregate demand and prices” (Ugai, 2006).11
Girardin and Moussa (2011) suggest that for the QE to be effective, it needs to be maintained for a longer period than the QEP in Japan.
4.2 Quantitative Easing in the United Kingdom
After the global financial crises hit in 2008, the Bank of England followed the example of the Japan and the United States and announced its own QE program in 2009. It made purchases of UK government bonds, known as gilts, through the Asset Purchasing Facility, which was created on January 30th
2009 as “a legally separate entity run by the Bank of England but with indemnity assurances from the Treasury for any possible losses.”12 (Breedon et al. 2012).
During the period of March 2009 through January 2010 the BoE expanded its balance sheet, mainly by purchasing gilts with a maturity of over 3 years for 200 billion pounds. These purchases amount to 14 percent of the GDP and resulted in a 300 percent growth of the BoE balance sheet (Joyce et al., 2011). In 2011 the BoE did another round of QE amounting to 175 billion pounds by the end of 2012.
The focus of the program was on the portfolio-balancing channel, according to studies by Butt et al. (2014), Joyce and Spaltro (2014), Benford et al. (2009). As such the program consisted of long-term gilt purchases owned by non-bank financial institutes (e.g. pension funds, insurers). The Monetary Policy Committee (MPC) expected that the bank-lending channel would be ineffective and thus buying assets or gilts directly from banks would not result in an expansion of lending and of the money supply, because of the pressure on banks to deleverage their balance sheets. (Joyce et al. 2011).
10 In any case, the article done by Gagnon et al. (2011) seems to have more validity on this issue,
as they are able to reference different sources supporting their claim, while Blinder (2010) does not.
11 Ugai (2006) also concludes that the biggest effect came through the expectations channel:
“This suggests that, in conducting monetary policy with due recognition of the zero bound constraint on interest rates, communication from the central bank to the private sector about monetary policy is critical in manifesting policy effects.”
However, before QE was implemented in 2009, the BoE had already applied several policies to mitigate the dire economic situation, including the Special Liquidity Scheme of 2008. In this scheme the BoE purchased assets directly from banks (financed by issuing Treasury Bills, so strictly speaking there was no monetary expansion, because the money put in to the economy by purchasing assets was taken out of the economy by selling treasury bills) to provide them with extra liquidity. Combined with a loose regulation concerning bank reserve balances held at the BoE (banks were allowed to set their own target within a certain range) this actually did lead to some monetary expansion. “Thus, in some sense, a small measure of QE occurred over this period since some of the increase in the Bank of England’s assets was financed by an unusually large expansion of the monetary base.”(Breedon et al., 2012). If we indeed view this initial period as part of QE, one might argue that by directly providing banks with liquidity the bank-lending channel could have been more actively engaged.
Overall, the effectiveness of the BoE’s QE policy is still disputed. In their empirical research titled “The impact of QE on the UK economy – some supportive monetarist arithmic” Bridges and Thomas (2012) conclude that “the increase in the broad money supply induced by QE is likely to have had a positive effect on GDP and inflation”. However, Lyonett and Werner (2012), in their literary and empirical research into the QE program, found, that the QE policy was not significantly different from normal monetary policy (with the exception of the high media-profile of the announcements). They found that it was not helpful concerning the recovery of the economy. They further state that the BoE should have focused more directly on “the growth of bank credit for GDP-transactions” (Lyonett and Werner, 2012). They suggest that credit is the most important factor for economic growth. Thus maybe it was not wise of the BoE to negate the bank-lending channel.
4.3 Quantitative Easing in the United States
When the Lehman Brothers went bankrupt and the credit crisis unfolded, the Federal Reserve was quick to implement its own QE program. In the initial announcement, however, the Federal Reserves chairman Ben Bernanke called the program “credit-easing”, perhaps to differentiate it from the Bank of Japans program (which might not have had the best of reputations) (Blinder, 2010). As the crisis was developing in 2007, the Federal Reserve had already begun decreasing their instrument rate from 5.25 percent to 0.25 percent in 2008 (Gagnon et al., 2011).
QE1 consisted of 1.25 trillion dollars in Mortgage Backed Securities (MBS) purchases, 1.75 trillion dollars in agency-debt securities, and 300 billion dollars in long-term treasury bills. QE2 consisted of 600 billion dollars in long-term treasury bills. QE3 was an open-ended program, as there was no announcement as to when the program would be ended. It consisted of 40 billion dollar a month purchases of MBSs. In addition, the Federal Reserve started a Maturity Extension Program (known as Operation Twist) in which they purchased long-term treasury bonds financed by selling short-term bonds. This operation amounted to 400 billion dollars but, because it was financed by short-term bonds, it did not expand the Federal Reserve’s balance sheet. However, when the program stopped in December 2012 (due to a lack of short-term bonds), the Federal Reserve revised and extended the program to 45 billion dollars a month (open-ended), without it being financed by short-term bonds anymore (Rosengren, 2015). These programs together expanded the Federal Reserve’s balance sheet with about 6 trillion dollars. According to Gagnon et al. (2011), halfway trough QE2 these purchases already amounted to 12 percent of the GDP.
Although the implemented policies started of in a haphazard manner13
, the QE policy of the Federal Reserve ended up being the longest and largest in our historical review. As we can see by the purchase choices (mainly MBSs and agency-debt purchases in order to lower risk in that market, and less so long-term bonds purchases which should reduce long term interest rates), the Federal Reserve concentrated its efforts on risk-premia and not on term-premia (Blinder, 2010). Rosengren (2015) observes that the United States have managed to raise inflation rates slightly with their QE policies. They are not, however, back to pre-crisis levels. Furthermore, the exit-strategy will ultimately determine whether the programs will be successful.
5. Historical efficacy of the bank-lending channel
In this segment I will discuss research done into the bank-lending channel and place it within the context of the various implementations. Unfortunately, too my knowledge, there has been no study focused on the bank-lending channel in the United States. However, as noted before, the context of the United States QE implementation may help us understand the boundaries
13 “The early stages of the quantitative easing policy were ad hoc, reactive, and institution based. The
Fed was making things up on the fly, often acquiring assets in the context of rescue operations for specific companies on very short notice (e.g., the Maiden Lane facilities for Bear Stearns and American International Group [AIG].” (Blinder, 2012)
and possibilities of QE and give us a better sense of the viability of the bank-lending channel.
5.1 The bank-lending channel in Japan
Bowman et al. (2011) run a panel data-set regression to study the bank-lending channel during the quantitative easing program (QEP) of the Bank of Japan. In this study the amount of bank lending is regressed on a number of variables, including bank total assets, equity ratio, non-performing loan ratio and the bank type14
. Total bank-lending levels dropped during the QEP (see Figure 2 in the appendix). However, they do find “a robust, positive and statistically significant effect” of the QEP on bank-lending growth, suggesting that “the expansion of reserves associated with QEP likely boosted the flow of credit to the economy” (Bowman et al., 2011).
But this effect was very small and was overshadowed by the fact that banks reduced their liquid holdings in the form of loans held at other banks. According to Bowman et al. (2011) this is possibly because of the perceived risk and the regulatory capital requirements for such loans. The total liquidity of banks was thus reduced despite the liquidity boost from the BoJ.
Their research also shows that weaker banks (banks with lower equity ratios) showed a greater boost in bank lending than stronger banks. They did not find significant differences for bank size. These effects were only significant in the first years of the QEP after which bank health and confidence were restored. As such, the abrupt end of the QEP program did not lead to any deterioration in bank lending. These findings are corroborated by Inoue (2009), who also finds significant bank-lending growth due to the QEP, though he does contradict Bowman et al. (2011) by stating that smaller and healthier banks responded well to the QEP.
5.2 The bank-lending channel in the United Kingdom
As noted before, the Monetary Policy Committee had deemed the bank-lending channel ineffective before implementing QE. They believed that the pressure on banks to deleverage their balance sheets would prohibit the bank-lending channel to come into effect (Joyce et al., 2011). Joyce and Spaltro (2014) and Butt et al. (2014) (in rapid succession) both used panel
14 An interesting variable, which controls for Japanese bank types, the characteristics of which
are beyond the scope of the other control variables. Such a variable is not seen in the research done into bank lending in UK banks. Bowman et al. do not include these variables in their regression results, however they do state that regional banks seemed to have higher loan-‐ growth compared to other types of banks during the QEP period.
dataset regression studies to research the bank-lending channel in the QE program of the Bank of England. Joyce and Spaltro (2014) find a small but significant effect on bank lending growth, although these effects are likely to be even smaller because they assume “a full pass-through from QE deposits”(Joyce and Spaltro, 2014), which means they do not distinguish between different kind of deposits like the study by Butt et al. (2014).
They also found that small banks responded more to the QE program than large banks. They suggest this is because small banks are more liquidity constrained than larger banks and thus have a harder time acquiring liquidity. QE can thus be a real help to them, whereas large banks already have other means of acquiring liquidity. The bank-lending growth effect is also dependent on how well capitalized banks are. When banks have more capital, there is less pressure to deleverage.
In contrast, Butt et al. (2014) conclude that there is no significant increase in bank lending due to QE. By identifying the banks that participate in gilt sales by other financial corporations (OFCs), they show that only a portion of the resulting deposit remains at the bank after the sale, as OFCs use the liquidity to invest and re-balance their portfolio, exactly as described by the portfolio-balancing channel. As such Butt et al. (2014) do not assume a full pass-through of QE to deposits. Butt et al. (2014) put forth the hypotheses that the bank-lending channel is dependent on the type of deposits created: “flighty”(uncertainty about the future amount of the deposit) or “sticky”(certainty about the future amount of the deposit). When the central bank purchases gilts from an OFC, the OFC deposits the cash from the sale at a bank. The bank now has extra and cheap liquidity, which should induce it to increase its lending. If, however, these deposits are to be used for transactions with other OFCs holding deposits at other banks (as will happen when OFCs re-balance their portfolios), then the bank would have to resort to more expensive funding to back up its new lending. That is why, when banks receive deposits that are considered flighty, they will not be induced to expand their lending because the variance of their reserves likely increase and they will not be certain about the amount of liquidity they hold in the future. This suggests that bank-lending channel is rendered infective by the portfolio-balancing channel. They show that bank reserve variance did increase during the QE program and suggest that this might be the reason why bank-lending growth did not occur in the United Kingdom.
6. Analysis
As we have seen, QE is a monetary policy that is not yet very clearly defined. The general idea is clear: expand the money supply by purchasing a large amount of assets thereby increasing expected future inflation rates. Yet it does not define specifically what kind of assets should be purchased, where they should be purchased, how large the total amount of purchases should be, what duration the policy should have and through which channels it should work. This is evident when we look at the different implementations in Japan, the United Kingdom and the United States.
As a pioneer of QE, Japan unfolded a program that was relatively small and focused on the bank-lending channel, though it was mostly effective through expectations.
The United States, with the largest and longest succession of QE policies, initially focussed more on supporting certain industries (MBSs and agency-debt securities) but later turned more to providing general liquidity through long-term treasury bond purchases and then switching back to MBSs again.
The United Kingdom’s policy was more streamlined, in the sense that it focussed almost exclusively on long-term gilt purchases from OFCs and worked mainly through the portfolio-balancing channel. It was also the shortest program, though not the smallest.
Furthermore, leading up to and during these programs the central banks have taken other monetary measures that might also be seen as part of QE (e.g. the Special Liquidity Scheme). This further confuses our definition of QE, making it harder to isolate effects caused by QE.
The effectiveness of these programs is still debated. It seems clear, however, that QE is not as effective as we would wish it to be and “must be seen as a symptomatic treatment; it must be accompanied with a dramatic restructuring in the financial framework.” (Girardin and Moussa, 2011): The problems that caused the liquidity trap are not solved by QE. In order to solve the root of the problem other measures must be taken.
Concerning the bank-lending channel the research tells us that the effectiveness has either been little or none. But since implementation of QE is so diverse it seems probable that the effectiveness of the bank-lending channel can depend on the strategy chosen: The BoE did not have faith in the bank-lending channel and designed a strategy that in effect excluded it from
occurring, as we have seen in the article by Butt et al. (2014). They showed that the portfolio-balancing channel actually frustrated the bank-lending channel because of flighty deposits and the increased variance of bank reserves.
Knowing this, should central banks form their implementation strategies in such a way that they do engage the bank-lending channel? Lyonett and Werner (2012) would suggest that they should as they remind us of the importance of credit availability for economic recovery: “As credit for GDP-transactions is found to have highest significance in explaining economic growth, policy-makers need to consider the methods that may influence this variable.”
(Lyonett and Werner, 2012)
So what might we suggest to improve the effectiveness of this channel? To reduce flightiness, central banks should make QE purchases at commercial banks directly. This results in a direct increase in their reserves instead of an increase in flighty deposits, which happens when the central bank makes purchases at OFCs. We might also focus our resources on smaller and weaker banks, as it will be more effective there because those will be most in need of cheap, extra liquidity. Thus purchases of the same amount will yield more bank-lending growth when made solely at small and weak banks than when they are also made at large and strong banks. The total amount of liquidity injected must also be very large in order for the effect to be sizeable, as we have seen in Japan and the United Kingdom (Bowman et al., 2011; Joyce and Spaltro, 2014).
However, the portfolio-balancing channel has not been entirely ineffective in reviving the economy, as in the case of the United Kingdom. We might therefor consider that a central bank would want to use (at least) both channels to implement QE. In order to let QE operate effectively through multiple channels, I would suggest a multi-facetted program, with specialized purchases for each channel, is needed. This program should be large, long and have a clear exit strategy (Bowman et al., 2011; Girardin and Moussa, 2011; Svensson, 2003).
The QE rounds in the United States seem to approach these criteria, as they boast an expansive multi-facetted program focused on multiple channels. Because the Federal Reserve bought part of its assets directly from banks, it supplied liquidity directly, which may have led to bank-lending growth. The purchasing of long-term treasury bonds from banks and OFCs may have led to portfolio re-balancing, resulting in higher asset prices.
7. Conclusion
In this thesis I have examined QE and the role that the bank-lending channel has played in recent programs. The efficacy of the bank-lending channel is one of the main concerns that economists have about QE. As it is the only concern which relates to the functioning of QE rather than to its effects, it seems important that more research should be done into the bank-lending channel. The programs in Japan, the United States and the United Kingdom have been implemented with different strategies, which have resulted in different outcomes. As for the effectiveness of the bank-lending channel: studies suggest that it is either small or non-existent. (Unfortunately, there were no studies available for the Federal Reserve’s QE program.) Several studies find that the effect of the bank-lending channel is constrained, especially during crises, by pressure to deleverage banks balance sheets and flightiness of OFCs deposits. We suggest that the efficacy of the bank-lending channel might be dependent on the implementation strategy of the QE program. To actively engage the bank-lending channel, central banks should: (1) Purchase assets directly from banks so as to avoid flighty deposits, (2) focus their purchases on smaller and weaker banks who are more likely to be liquidity constrained and (3) increase the size of their program for larger effects. Additionally I suggest that in order to effectively engage all transmission channels a central bank needs to implement a large, long and multifaceted program with a clear exit strategy.
8. References
Bell, V., Joyce, M., Liu, Z., & Young, C. (2012). The Distributional Effects of Asset Purchases. Bank of England Quarterly Bulletin, Q3.
Benford, J., Berry, S., Nikolov, K., Young, C., & Robson, M. (2009). Quantitative easing. Bank of England Quarterly Bulletin, 49(2), 90-100.
Blinder, A. S. (2010). Quantitative Easing: Entrance and Exit Strategies (Digest Summary). Federal Reserve Bank of St. Louis Review, 92(6), 465-479.
Bowman, D., Cai, F., Davies, S., & Kamin, S. (2011). Quantitative easing and bank lending: evidence from Japan.
Breedon, Francis, Jagjit S. Chadha, and Alex Waters. "The financial market impact of UK quantitative easing." Oxford Review of Economic Policy 28.4 (2012): 702-728.
Bridges, J., & Thomas, R. (2012). The impact of QE on the UK economy–some supportive monetarist arithmetic.
Butt, N., Domit, S., McLeay, M., Thomas, R., & Kirkham, L. (2012). What can the money data tell us about the impact of QE?. Bank of England Quarterly Bulletin, Q4.
Butt, N., Churm, R., McMahon, M. F., Morotz, A., & Schanz, J. F. (2014). QE and the bank lending channel in the United Kingdom.
Gagnon, J., Raskin, M., Remache, J., & Sack, B. (2011). The financial market effects of the Federal Reserve’s large-scale asset purchases. International Journal of Central Banking, 7(1), 3-43.
Gambacorta, L. and Marques-Ibanez, D. 2011. "The bank lending channel: lessons from the crisis", Economic Policy, Vol. 26 (66), 135-82.
Girardin, E., & Moussa, Z. (2011). Quantitative easing works: Lessons from the unique experience in Japan 2001–2006. Journal of International Financial Markets, Institutions and Money, 21(4), 461-495.
Hudson, M. (2010). US 'Quantitative Easing' Is Fracturing the Global Economy. Bard College Levy Economics Institute Working Paper, (639).
Inoue, H. (2009). Bank Lending Channel during the Quantitative Monetary Easing Policy
Period (No. 09J004). Osaka School of International Public Policy, Osaka University.
Ito, T., & Mishkin, F. S. (2006). Two decades of Japanese monetary policy and the deflation problem. In Monetary Policy with Very Low Inflation in the Pacific Rim, NBER-EASE, Volume 15 (pp. 131-202). University of Chicago Press.
Jimenez, G., Ongena, S., Peydro, L., and Saurina, J. 2010. "Credit supply - Identifying balance-sheet channels with loan applications and granted loans", European Central Bank, Working Paper Series, No. 1179.
Joyce, M., Lasaosa, A., Stevens, I., & Tong, M. (2011). The financial market impact of quantitative easing in the United Kingdom. International Journal of Central Banking, 7(3), 113-161.
Joyce, Michael, Matthew Tong, and Robert Woods. "The United Kingdom’s quantitative easing policy: design, operation and impact." Bank of England Quarterly Bulletin (2011).
Joyce, Michael, et al. "Quantitative Easing and Unconventional Monetary Policy–an Introduction*." The Economic Journal 122.564 (2012): F271-F288.
Joyce, M., & Spaltro, M. (2014). Quantitative easing and bank lending: a panel data approach.
Kashyap, A. K., & Stein, J. C. (1994). Monetary policy and bank lending. In Monetary policy (pp. 221-261). The University of Chicago Press.
Krugman, P. (2000). Thinking about the liquidity trap. Journal of the Japanese and International Economies, 14(4), 221-237.
Lyonnet, V., & Werner, R. (2012). Lessons from the Bank of England on ‘quantitative easing’ and other ‘unconventional’ monetary policies. International Review of Financial Analysis, 25, 94-105.
Mahajan, A. (2015). Quantitative Easing: A Blessing or Curse. Advances in Management and Applied Economics, 5(2), 77.
Martin, C., & Milas, C. (2012). Quantitative easing: a sceptical survey. Oxford Review of Economic Policy, 28(4), 750-764.
Meaning, J., & Zhu, F. (2011). The impact of recent central bank asset purchase programmes. BIS Quarterly Review, December.
Rosengren, E. S. (2015). Lessons from the US experience with quantitative easing (No. 94).
Svensson, L. E. (2003). Escaping from a liquidity trap and deflation: The foolproof way and others (No. w10195). National Bureau of Economic Research.
Ugai, H. (2007). Effects of the quantitative easing policy: A survey of empirical analyses. Monetary and Economic Studies-Bank of Japan, 25(1), 1.
Van den Heuvel, S. J. (2002). The bank capital channel of monetary policy. The Wharton School, University of Pennsylvania, mimeo.
9. Appendix
Figure 1. (Benford et al., 2009)
Figure 2. (Bowman et al., 2011)
