The Effects of Unconventional Monetary Policy on the European
Banking Sector
W.J. Oude Lenferink
S2560534
+31 (0) 6 41 79 22 06
University of Groningen
Faculty of Economics and Business
MSc. Finance
Supervisor: Dr. G.T.J. Zwart
January 2016
Abstract
In this paper I examine the effects of unconventional monetary policy on the European banking sector. The effects are measured by regressing abnormal stock return and abnormal changes in Distance to Default on a monetary surprise measure. The analysis covers all of the unconventional monetary policy related announcements made by the European Central Bank from the end of 2008 until 2015. The overall results suggest that unconventional monetary policy announcements by the ECB had a positive effect on both the stock market and the market-based measure of the soundness of the European banking sector.
2 1. Introduction
Since the global financial crisis that began in August 2007, the European Central Bank (ECB) faced a difficult task of restoring monetary transmission to support the economy during these exceptional circumstances. Following the financial crisis, the Eurozone was further hit by the sovereign debt crisis that erupted in 2009. During normal economic circumstances, the ECB uses three main monetary policy instruments (open market operations, the minimum reserve system and the standing facilities).The problem with conventional monetary tools is that they become limited in their usefulness in periods of exceptional economic circumstances and may no longer be effective in achieving their goals. The financial crisis and the subsequent sovereign debt crisis disturbed the traditional transmission mechanisms of standard monetary policy and the ECB was no longer able to achieve its goals through conventional monetary policy instruments. In order to restore the monetary transmission mechanisms and support the economy, the ECB implemented several unconventional monetary policies to achieve its goals.
The unconventional measures taken by the ECB have been targeted mainly at the banking sector because of the important role this sector plays in the transmission of monetary policy and the financing of the economy. The first non-standard measures of the ECB include fixed rate full allotment tender procedures that gave institutions unlimited access to central bank liquidity, extension of the maturity of liquidity provision in order to encourage banks to continue providing credit to the economy, currency swap agreements to support foreign currency funding and the easing of collateral requirements.
In July 2009, the ECB launched the first covered bond purchase programme (CBPP1). In November 2011 and October 2014, the respectively second and third covered bond purchase programme was announced. The goal of the three covered bond purchase programmes was to recover liquidity in covered bond market which is a primary source of funding for banks. CBPP1 and CBPP2 are terminated, but CBPP3 is still operational and will last until June 2016.
3
As mentioned before the ECB had to introduce the above mentioned non-standard policy measure because interest rates were approaching the zero lower bound, which will cause the standard measures to be ineffective in achieving it goals. The main objective of these non-standard measures is to stimulate the economy and lower medium- to longer-term interest rates. From previous studies we know that lowering the interest rate increases bank risk-taking behavior.
4
2. Literature review
The implementation of unconventional monetary policy instruments by the ECB was the most significant shift in the practice of monetary policy in recent years. In order to evaluate the effects of unconventional monetary policy on the banking sector a thorough understanding of the different transmission mechanisms of unconventional monetary policy is essential. The transmission mechanism of standard interest rate policy is well documented in the current literature. However, the implementation of unconventional policies spurred a debate concerning their numerous direct and indirect effects. The aim of this section is to outline the 3 different transmission mechanisms of unconventional monetary policy which are identified in the recent literature, for example Joyce et. Al. (2012) and Krishnamurthy & Vissing-Jorgensen (2011). Firstly I will briefly describe the non-standard measures adopted by the ECB. Secondly the three theoretical transmission mechanisms will be discussed, thirdly I will describe the theoretical impact of the non-standard measures on the banking sector. Finally an overview of the current literature and the formulation of the hypothesis.
2.1 Non-standard measures
During normal economic circumstances the ECB base their monetary policy on the control of short term interest rates that influence the economy through various monetary policy transmission mechanisms. However, the financial turmoil that started with the financial crisis in late 2007 revealed that the usual mechanisms were ineffective in achieving its goals during exceptional circumstances. This section will provide a brief overview of the nonstandard measures adopted by the ECB. (ECB, 2011)
2.1.1 Enhanced Credit Support
5 2.1.2 Covered Bond Purchase Programme
On 7 May 2009 the ECB announced its first Covered Bond Purchase Programme (CBPP1). The objective of this first purchase programme is to purchase Euro denominated covered bonds issued in both the primary and the secondary markets in order to support financial segments that are important for the funding of banks and support the credit supply in the economy through credit institutions. The amount of purchases under CBPP1 was aimed at EUR 60 billion. On 30 June 2010 the ECB terminated CBPP1 because the total amount of purchases has reached the EUR 60 billion target. When tension in the financial market started to increase due to the deepening of the sovereign debt crisis the ECB announced a second covered bond purchase programme (CBPP2) of EUR 40 billion. CBPP2 ended on 31 October 2012, and the total purchases under CBPP2 reached to an amount of EUR 16.4 billion. Under a third covered bond purchase programme (CBPP3) the ECB started buying covered bonds again on 20 October 2014. CBPP3 is intended to last for at least two years and will be conducted alongside the ABSPP and the PSPP. The monthly purchases of these 3 programmes combined are targeted at EUR 60 billion each month.
2.1.3 Securities Market Programme
In order to ensure depth and liquidity in those market segments that are dysfunctional due to the sovereign debt crisis the ECB launched the Securities Markets Programme (SMP) in May 2010 (ECB, 2011). The malfunctioning of several securities market threaten the transmission mechanism of conventional monetary policy. Under the SMP the ECB started to intervene in secondary government bond markets. The initial programme focused on Greek, Irish and Spanish government bonds, and from August 2011 the programme was extended to Italian and Spanish government bonds. Special liquidity-absorbing operations were organized to ensure that liquidity conditions were not affected. On 6 September 2012 the SMP was terminated and the existing securities in SMP portfolio will be held to maturity1
2.1.4 Outright Monetary Transactions
In September 2012 the SMP was terminated and replaced by the Outright Monetary Transactions (OMT) programme. The essence of the SMP and the OMT programme are the same; both programmes allow the ECB to intervene in the secondary government bond market. There is, however one difference between the OMT and SMP. Under the OMT programme countries are only eligible for OMTs when they are subject to austerity programmes. Furthermore, the programme has no ex ante
6
quantitative limits and will last for as long as necessary. Also in this case the liquidity created through OMT is sterilized by liquidity-absorbing operations.
2.1.5 Asset-Backed Securities Purchase Programme
In the press release of 19 September 2014 the ECB announced that it will purchase a broad portfolio of simple and transparent Asset Backed Securities (ABS) with underlying assets that contain claims against the non-financial private sector in the EU (ECB, 2011). The operational details of the Asset-Backed Securities Purchase Programme (ABSPP) were announced on 2 October and the first purchase of ABS began on 21 November and will last for at least two years. The main objective of the ABSPP is to diversify the funding sources of banks, stimulate the issuance of new securities, enhance the credit support to the real economy and improve the transmission of monetary policy. In contrast to earlier purchase programmes such as SMP and OMT, the created liquidity through the ABSPP is not sterilized. 2.1.6 Expanded Asset Purchase Programme
On 22 January 2015 the ECB announced an Expanded Asset Purchase Program (EAPP), which will add EAPP to the Public Sector Securities Programme (PSPP) to the existing purchase programmes that are conducted in the private sector (ABSPP, CBPP3). Under the PSPP the ECB starts to buy private sector marketable debt instruments in order to inject liquidity into to the banking system and fight deflationary pressure. The ECB intends to continue the monthly purchases until September 2016. The combined amount of purchases under the APP - which include CBPP3, ABSPP and PSPP - is targeted at EUR 60 billion each month.
2.2 Transmission mechanisms of unconventional monetary policy
The implementation of unconventional policies spurred a debate concerning their numerous direct and indirect effects. The aim of this section is to outline the 3 different transmission mechanisms of unconventional monetary policy which are identified in the recent literature, for example Joyce et. Al. (2012) and Krishnamurthy & Vissing-Jorgensen (2011).
2.2.1 The Signaling Channel
7
expectation of the path of future policy rates. Through the expectations channel of the term structure long term interest rates will fall. According to Chodorow-Reich (2014), the decline of the long term interest rate has different implications for the real economy and the soundness of financial institutions. These implications will be discussed later.
2.2.2 The Liquidity channel
The second transmission channel of unconventional monetary policy is the liquidity channel. By undertaking large asset purchasing programs, such as the covered bond purchase programme, asset backed securities programme and public sector purchase programme, the central bank increases the liquidity in the hands of investors. This additional liquidity is assumed to induce banks to extend new credit to corporations and households.
Joyce et al. (2011) argues that the presence of the central bank in the market as a significant buyer of assets may improve market functioning and, as a result of increased liquidity and improved market function, the premium for illiquidity will decline. The increased level of liquidity and improved market conditions may have positive effects on asset prices and eventually prevent commercial banks from credit rationing or fire sales of assets.
2.2.3 The Portfolio Balance Channel
The third and probably most important transmission mechanism of unconventional monetary policy is The Portfolio Balance Channel. This channel operates only with asset purchase programmes such as CBPP, SMP and ABSPP. The idea behind this channel is that a change in the balance sheet composition of the central bank may alter the decisions of economic agents. According to Joyce et al (2014) purchases of financial assets financed by central bank money initially increase broad money holdings and push up asset prices, as those who have sold assets to the central bank rebalance their portfolios into riskier assets. This then stimulates expenditure by increasing wealth and lowering borrowing costs for households and companies.
2.2.4 Implications for the financial sector
8
decrease in the risk-free interest rate will cause the hurdle rate to decline. Chodorow-Reich (2014) argues that the change in the hurdle rate alters the total project risk in the economy because a change in the risk-free rate implies that newly viable projects have either lower risk or lower expected returns. For instance, if newly viable projects have mostly higher risk, the risk in the economy will rise, On the other hand, if the expected return and variance is lower the risk in the economy will decline. Since the banking sector is exposed to the risk in the economy via the assets on the balance sheet of individual banks, a change in the risk of the economy will have impact on the volatility of the individual bank’s assets.
Second, a change in the interest rate environment may spur the search for more return “Reaching for Yield”. In an environment where interest rates are declining, managers can get dissatisfied with the low returns. In order to restore the average return, managers can shift credit supply towards more risky projects, increase leverage or shift investment into different asset classes. This poses a classic principle agent problem in which incentives of managers may not align with the objectives of share- and debt-holders (Jensen and Meckling 1976; Rajan 2005). Reaching for yield may increase the volatility of the bank’s assets when the risk appetite of managers change in order to restore average return.
Third, announcements made by the ECB change the expectations of businesses and households through the signaling channel. If businesses and households are more optimistic about the future of the economy, they are more likely spend money and make new investment, which will increase aggregate demand. The same holds for low interest rates, low interest rates, lowers the costs of capital and will increase spending and investment. According to Chodorow-Reich (2014), the value of legacy assets held by financial institutions will raise because of this general equilibrium effect. The increase in spending and investment rise the profits and lowers the default probability of non-financial institutions, which in turn results in an increase in legacy asset value.
9 2.3 Related literature
The introduction of unconventional monetary policy has raised a powerful interest among academics. The majority of the current literature mainly relates to the impact of unconventional monetary policy on core macroeconomic variables.
Doh (2010) examined whether the Federal Reserve’s Large Scale Asset Programs lowered long-term interest rates. He finds that by changing the supply of long term bonds, the large scale purchases of bonds decrease the term premia.
Peersman (2011) used a structural VAR model to identify disturbances at bank credit volumes caused by non-standard monetary policy actions. He find that non-standard measures such as an increase in the monetary base or the size of the central bank balance sheet has a “hump-shaped effect” on economic activity and a permanent impact on consumer prices
Falagiarda & Reitz (2013) used an event study and a GARCH framework to study the effects of unconventional monetary policy-related announcements on the sovereign risk of Italy. They used long term bond futures to measure the unexpected part of ECB announcements. The results indicate that unconventional monetary policy-related announcements substantially reduced Italian long term government bond yield spread relative to Germany.
Despite the interest in the effects on macroeconomic variables, there is not much literature that relates to the impact of unconventional monetary policy on the soundness of the European banking sector. From studies conducted on the relation between the interest rate environment and bank risk taking, during conventional monetary policy periods, we know that a change in the interest rate affects the soundness of banks. Buch et al. (2014) state that as a result of a decrease in the interest rate, banks soften their lending standards and change their balance toward more risk taking activities.
Jiménez et al. (2014) studied the Impact of the stance and path of monetary policy on the level of credit risk of individual banks loans and lending standard. They find that bank risk-taking increases when interest rates are lower prior to loan origination and that, in this way monetary policy affects the quality distribution of borrowers in the banks’ loan portfolios.
10
examined the effects of unconventional monetary policy on financial institutions by employing a high frequency event study. He found that the introduction of unconventional monetary policy by the FED had a strong stabilizing impact on banks and especially life insurance companies.
Lambert & Ueda (2014) used an event study approach to measure the effects of unconventional monetary policy on the soundness of the banking sectors in the UK, US, and the Eurozone. They found little supporting evidence that unconventional monetary policy benefits banks. They do not detect clear effects of monetary easing on bank stock valuation, but find a positive effect for medium-term bank credit risk.
Rogers et al. (2014) studied the effect of unconventional monetary policy by the Federal Reserve, Bank of England, European Central Bank and Bank of Japan on stock prices. They find that the expansionary monetary policy shock significantly raises domestic stock prices in the US, UK and the Eurozone. In this paper, I propose to study the effects of unconventional monetary policy on the European banking sector using an event study approach. The idea is to construct a monetary policy surprise measure which captures the surprise component of each non-standard measure-related announcement made by the ECB over the period between the end of 2008 and the beginning of 2015. To gauge the effects on the European banking sector, I regress daily changes in stock return and credit risk on a measure of monetary policy surprise. I use a daily Distance to Default indicator as a measure of credit risk. Given the theoretical implications and the results of Chodorow-Reich (2014), I expect that the introduction of the non-standard measures benefit the European banking sector in terms of overall credit risk, therefore I formulate the following hypothesis:
H1: The European banking sector benefits from the introduction of unconventional monetary policy.
With regard to the stock market, I expect a positive relation between the introduction of non-standard measures and the reaction in the stock market.
11
3. Data and methodology
In order to measure the effects of unconventional monetary policy on the European banking sector, I first examine the stock market reaction on days when the ECB makes announcements that relate to unconventional monetary programmes. Since the main goal of this paper is to examine the influence of non-standard measures on the soundness of the European banking sector, I have to come up with a variable that measures the credit/systematic risk in the European banking sector. Previous studies conducted on this topic uses spreads on Credit Default Swaps to measure credit/systematic risk. The credit risk measure that I will use is a daily time series of aggregated Distance to Default (DtD) for the 50 biggest banks in the EMU. In order to construct a variable that only captures the surprise component of ECB announcements, I will build upon the idea of Wright (2012) who uses changes in government bond futures prices to obtain yield changes. What follows is a description of the data and the different methodologies that I use to gather the above-mentioned variables.
3.1 ECB Announcements
The first step in the process is constructing an overview that contains information about ECB announcements which relate to unconventional monetary policy. The ECB uses monthly press conferences to inform financial markets and the general public about their monetary policy decisions. After each press conference, a transcript of the press conference is published on the website of the ECB. From a previous study, conducted by Falagiarda & Reitz (2013), I used an already existing list which contains all the ECB announcements from 2008 until 2012. By consulting the database of transcripts from the period after 2012, I extended the list of announcements up to 2015. The final list of unconventional monetary policy-related announcements consists of 51 press releases over the period from late 2008 until the beginning of 2015, see table 8 of the appendix. In the period between 2008 and 2015, the ECB used different instruments to combat the financial turmoil, therefore I classified the announcements in different instrument related categories:
ECS: announcements under the ECS category relate to liquidity injecting instruments such as maturity extensions, availability of central bank liquidity, currency swap arrangements, and collateral extensions.
12 3.2 Monetary Policy Surprise Measure
To assess the effects of unconventional monetary policy on the soundness of the banking sector, I regress daily abnormal changes in aggregate DtD on a monetary policy surprise variable on all announcements dates during the end of 2008 until the beginning of 2015. There are two major drawbacks in assessing the impact of (unconventional) monetary policy. The first drawback is that part of the decisions made by the ECB are already anticipated by the market at the time of the announcement. The consequence of this drawback is that I can only measure the surprise component of the announcements made by the ECB. The second problem relates to the nature of unconventional monetary policy. During conventional policy periods, one can measure the monetary policy surprise component relatively easy by taking the difference between the government bond futures rate before the announcement and the announced policy rate (Kuttner, 2001). This method becomes useless in unconventional policy periods because announcements concerning unconventional measures do not involve target rates. For the construction of a monetary policy surprise measurement, I build upon the idea of Wright (2012). In this paper, quotes on the front contract of medium- to long-term government bond futures are used to construct a surprise measure by computing the first principal component of the yield changes. The yield changes are constructed as returns on the futures contract divided by the duration of the cheapest-to-deliver security in the deliverable basket. The returns on the futures contract is measured with intraday data; a window of 15 minutes before each announcement to 1 hour and 45 minutes afterwards is used. Because I do not have access to intraday data, I have to use a slightly different approach. For the medium to long term yield I use 2 different quotes on front government bond futures contracts. The medium term yield is represented by the Euro-Schatz Futures where the underlying asset for the contract is a German debt securities with a time to maturity of 1.75 to 2.25 years. The long-term yield is represented by the Euro-Bund Futures where the underlying asset is a German debt securities with a time to maturity of 8.5 to 10.5 years. Data on front contract quotes and cheapest-to-deliver securities is derived from Datastream. Daily yield changes are constructed by using the method of Wright (2012). Firstly simple returns on both medium- and long-term contracts are calculated. For the construction of daily yield changes, I use the properties of the duration measure. From the properties of duration we know that a small change in the yield ∆𝑦 changes the bond price by ∆𝐵.
∆𝐵
𝐵 = −𝐷∆𝑦 (3.1)
13 ∆𝑦 =∆𝑏
𝐷 (3.2)
Where ∆𝑏 is the change in the bond price relative to the bond price ∆𝐵𝐵 . To measure the overall change in the medium- to long-term yield from both yield changes, I use a Principal Component Analysis (PCA). A PCA is a statistical procedure which converts a set of possibly correlated observations into a set of uncorrelated variables; principal components. I use the first principal component (PC1), which accounts for 83.6% of the variance in the yield changes of both maturities (see table 1). The average daily change in the yield of both maturities is constructed by multiplying each yield change by its eigenvector component (see table 2).
F1
F2
Eigenvalue 1.673 0.327
Variability (%) 83.630 16.370
Cumulative % 83.630 100.000 Table 1: Eigenvalue of PC1 and PC2
F1
Euroschatz 0.707
Eurobund -0.707 Table 2: Eigenvectors
14 3.3 Stock market reaction
To measure the reaction of stock markets on all ECB announcement days, I use traditional event study methodologies as described in MacKinley (1997). According to MacKinley (1997), event studies are based on the assumption that; “Given rationality in the marketplace, the effects of an event will be reflected immediately in security prices.” Conducting an event study requires the calculating of abnormal returns on event days. The abnormal return is calculated as the actual return on the event day minus the normal return of the firm during the estimation window. The estimation window relates to a specific time period prior to the event date, usually 250 trading days. To test the statistical significance of the observed abnormal returns on event days, several test statistics can be used depending on the type of event study.
The usual starting point for an event study is the identification of the event(s) of interest and the selection of firms that will be included in the study. The events of interest that I use to assess the stock market reaction is defined by the list of ECB announcements, which is described in section 3.1. The total list of events consist of 51 unconventional monetary policy related announcements ranging from the end of 2008 until the beginning of 2015. Because I am interested in the overall reaction of the European banking sector, the event study will include the 50 biggest banks in the European Monetary Union. The relative size of each bank is measured by its market capitalization. In order to reduce the interval in stock data observations, I will use daily observations of opening and closing prices. The use of opening and closing prices is justified because all the announcements of the ECB are made during trading hours. The stock data is derived from Datastream. To measure normal returns in the estimation window, I will use the constant mean return model instead of a more sophisticated statistical or economic model. According Brown & Warner (1980), the result of the constant mean return model is similar to those of more sophisticated models. Abnormal returns based on the constant mean return model are calculated by:
𝐴𝑅𝑖𝑡 = 𝑅𝑖𝑡− 𝑅𝑖 (3.3) Where 𝐴𝑅𝑖𝑡 is the abnormal return of security 𝑖 on time 𝑡, 𝑅𝑖𝑡 is the actual return of security 𝑖 on time
𝑡, and 𝑅𝑖 is the average return in the estimation window. The estimation window is set to 250 trading
days prior to the event date. On the event date, the 𝐴𝑅 for each of the 50 banks is calculated and averaged to obtain the average abnormal return (AAR).
𝐴𝐴𝑅𝑡 =∑𝑁𝑖=1𝐴𝑅𝑖,𝑡
15
Because I initially use an event window of 1 day, the actual announcement day, the cumulative average abnormal return (CAAR) is equal to the AAR. When the event window is extended to multiple days the CAAR is equal to the sum of the AAR’s in the event window.
𝐶𝐴𝐴𝑅𝑡 = ∑ 𝐴𝐴𝑅𝑁𝑖 𝑡 (3.5) In MacKinley (1997), several significance tests for non-overlapping event windows are discussed. However, a problem arises when the event window is the same for all the companies in the sample. The usual statistical tests assumes that the covariance between the abnormal returns is equal to zero. This assumption allows you to calculate the variance of the CAAR which is used to obtain the level of significance. In this study, all the announcement dates are the same for all banks in the sample which is known as event clustering. The problem with event clustering is that it violates the crucial non-zero correlation assumption. I will use a portfolio approach in order to deal with the violation of the non-zero correlation assumption. In the portfolio approach, the abnormal returns of each individual bank on each specific event will be aggregated into portfolios. The abnormal returns of the different portfolios will be regressed on the monetary surprise measure. For the regression analysis, I will use the Ordinary Least Squares (OLS) method. Figure 1 plots the average abnormal returns (AAR) from day -2 to +2 where 0 is the event day. “MPS NEGATIVE” represents announcement days with a negative surprise value. “MPS POSITIVE” represents announcements days with a positive surprise value.
16 3.4 Daily time series of Distance to Default
To measure the impact of unconventional monetary policy on the soundness of the European banking sector I use a daily time series of aggregated Distance to Default (DtD). The DtD indicator is a measure of credit risk that shows how many standard deviations away a particular bank is from its default point. The default point is reached when the market value of assets is equal to or lower than the book value of debt. The formula to calculate the DtD for each bank is derived from the option pricing model of Black and Scholes.
𝐷𝑡𝐷 =
ln( 𝑉𝑡 𝐷𝑡)+(𝑟𝑓−𝜎𝑡 2 2)𝑇 𝜎𝑡√𝑇 (3.6) 𝑽𝒕: 𝑀𝑎𝑟𝑘𝑒𝑡 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑎𝑠𝑠𝑒𝑡𝑠 𝑎𝑡 𝑡𝑖𝑚𝑒 𝑡 𝑫𝒕: 𝐵𝑜𝑜𝑘 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑑𝑒𝑏𝑡 𝑎𝑡 𝑡𝑖𝑚𝑒 𝑡 𝑹𝒇: 𝑅𝑖𝑠𝑘 𝑓𝑟𝑒𝑒 𝑖𝑛𝑡𝑒𝑟𝑒𝑠𝑡 𝑟𝑎𝑡𝑒 𝝈𝒕: 𝑉𝑜𝑙𝑎𝑡𝑖𝑙𝑖𝑡𝑦 𝑜𝑓 𝑎𝑠𝑠𝑒𝑡𝑠 𝑎𝑡 𝑡𝑖𝑚𝑒 𝑡 𝑻: 𝑀𝑎𝑡𝑢𝑟𝑖𝑡𝑦 𝑜𝑓 𝑑𝑒𝑏𝑡Most of the data that is used as input variable in the DtD formula is directly observable in the market. To obtain daily estimates of the book value of debt 𝐷𝑡, I interpolated the annual total debt reports
from Datastream using cubic spline to yield daily observation. For the risk free rate 𝑟𝑓, I used the
annualized 3 months Euribor which is closely related to the dynamics in the banking sector. The maturity of debt 𝑇 is set to 1 year. The values for the two remaining input variables, market value of assets 𝑉𝑡 and volatility of assets 𝜎𝑡, cannot be observed directly from the market. According to Merton
(1977), equity can be modelled as a call option on the assets of the firm. With a strike price equal to the total book value of the debt, we can use option-pricing theory to derive the market value and volatility of assets from observable equity values.
𝐸 = 𝑉 𝑁(𝑑1) − 𝑒−𝑟𝑇𝐷 𝑁(𝑑2) (3.7)
𝜎𝑒= (𝑉𝐸) 𝑁(𝑑1)𝜎 (3.8) Where E is the market value of equity, calculated by multiplying the number of outstanding shares by current share price, and 𝜎𝑒 is the historical volatility of equity, the historical volatility is estimated as
the annualized standard deviation of equity over the preceding 250 trading days. Since the values for 𝐸 and 𝜎𝑒 can be obtained from the market, the values for 𝑉𝑡 and 𝜎𝑡 can be calculated by simultaneously
17
market average of the DtD measure. In the literature section, I discussed different theoretical implications of unconventional monetary policy. According to Chodorow-Reich (2014), unconventional monetary policy affects the soundness of financial institutions in different ways. In his paper he used the concept of distance to default as a framework for the theoretical implications. Because I constructed a daily distance to default indicator, I am able to shed some light on the evolution of the variables underlying the distance to default measure. Figure 1 shows the evolution of the aggregate distance to default over time.
Figure 1: Evolution of aggregate Distance to Default over time
According to Chodorow-Reich (2014), unconventional monetary policy affects the soundness of financial institutions by altering the volatility of assets 𝜎𝑡, value of assets 𝑉𝑡 and leverage. For an
explanation of the different channels see the literature sector. Chodorow-Reich (2014) argues that the
change in the hurdle rate and the reaching for yield motivation affect the volatility of assets 𝜎𝑡.
Figure 2 shows the evolution of 𝜎𝑡 over time.
18
Since I only observe the evolution of 𝜎𝑡 over time, it is impossible to draw any inferences. But the
overall pattern of 𝜎𝑡 does not seem to point at an increase of 𝜎𝑡 due to a change in the project risk or
the reaching for yield motivation. This is in line with the findings of Chodorow-Reich (2014) as he finds that unconventional monetary policy does not raise the riskiness of banks. The second theoretical implication relates to the value of assets. Chodorow-Reich (2014) argues that the value of legacy as sets, 𝑉𝑡, increases due to general equilibrium effects. The increase in spending and investments, caused
by the decrease in the interest rate, raises the profits and lowers the default probability of non-financial institutions. Figure 3 shows the evolution of 𝑉𝑡 over time.
Figure 3: evolution of 𝑉𝑡 over time.
The last theoretical implication relates to the leverage of financial institutions. The theoretical implications suggest that leverage is affected in two ways. First, changes in interest rates affect the opportunity costs of collateral and second, changes in 𝜎𝑡 and leverage feeds into value at risk models
used by financial institutions. Figure 4 shows the evolution of leverage over time.
Figure 4: Evolution of leverage over time. Leverage is calculated as (𝑉𝑡𝑉−𝐷)
19 3.5 Cross sectional regression
The main goal of this paper is to study the overall impact of unconventional monetary policy on the banking sector. Besides the overall impact, it also interesting to see whether the impact is different in certain countries. Furthermore, it also interesting to investigate whether the impact of unconventional monetary policy is different for banks with certain characteristics, such as size. I will use a cross sectional regression model to examine the difference in the magnitude of the impact across countries and the difference in the magnitude across bank specific characteristics. The cross sectional analysis will be performed for both the abnormal returns in the stock market and the aggregate distance to default measure. I will use a general cross sectional model which takes the following form:
∆𝐷𝑡𝐷𝑖𝑡 = 𝑎 + 𝛽𝑀𝑃𝑆𝑡 (3.9) Where ∆𝐷𝑡𝐷𝑖𝑡 is the abnormal change in the aggregate Distance to Default for sub sample 𝑖 on time
𝑡, 𝑡 corresponds to ECB announcement days. The abnormal change in Distance to Default is measured in the same way as the abnormal stock returns in equation 3.3. 𝑀𝑃𝑆𝑡 corresponds to the monetary
policy surprise measure on time t. The cross sectional model for the stock market reaction is:
𝐴𝐴𝑅𝑖𝑡 = 𝑎 + 𝛽𝑀𝑃𝑆𝑡 (3.10) Where 𝐴𝐴𝑅𝑖𝑡 is the abnormal return of sub sample 𝑖 on time 𝑡, the abnormal return is calculated as
20
4. Results
4.1 Distance to Default
At first I will discuss the results for the full sample which include the 50 biggest banks in the EMU. The abnormal change in the Distance to Default is regressed on four different announcement samples. 𝑀𝑃𝑆𝑎𝑙𝑙 𝑎𝑛𝑛𝑜𝑢𝑛𝑐𝑒𝑚𝑒𝑛𝑡𝑠 contains all the announcements between the end of 2008 and the
start of 2015. 𝑀𝑃𝑆𝑠𝑖𝑔𝑛. 𝑎𝑛𝑛𝑜𝑢𝑛𝑐𝑒𝑚𝑒𝑛𝑡𝑠 contains all the announcements with surprise values that are
statistically significant. 𝑀𝑃𝑆𝑛𝑒𝑔𝑎𝑡𝑖𝑣𝑒 contains all the announcements with a negative surprise value
and 𝑀𝑃𝑆𝑝𝑜𝑠𝑖𝑡𝑖𝑣𝑒 are all the announcements with a positive surprise value. The results are shown in
table 4. Variables
-5.69* -9.53*** -11.90 -12.46** Std. error (2.87) (2.91) (8.86) (5.45) R2 0.08 0.40 0.07 0.18 N 51 18 25 26
Table 4: Regression result of the full sample. (* 10% significance level ** 5% significance level *** 1% significance level)
The observed coefficients across all samples are highly negative. A negative coefficient indicates that a negative monetary policy surprise, a decrease in the medium- to long-term yield, has a positive effect on the overall risk in the European banking sector. The highly significant coefficient from the significant announcements sample suggest that a 1 basis point decrease in the level of the yield curve will cause the aggregate distance to default to increase with 0.0953 units.
One interesting finding is that the coefficient for the negative surprises sample is lower than the coefficient for the positive surprise sample and that the coefficient for the positive surprise sample is significant whereas the coefficient for the negative surprise sample is not. This could possibly be explained by the fact that markets have certain expectations prior to the announcement day. When the announced measures are not in line with the expectations of the market, an immediate market reaction will occur in order to reflect the new information. A negative monetary policy surprise value is observed when investors are positively surprised by the announcement. A positive monetary policy surprise value is observed when the markets are negatively surprised by the announcement. The fact that I observe a statically significant, negative relation between the change in DtD and a positive monetary policy surprise value suggests that, in terms of distance to default, the magnitude of a negative surprise correction is greater than a positive surprise correction.
The results from the full sample analysis indicates that, since the main objective of the unconventional monetary policy measures is to reduce the medium- to long-term yields, the European banking sector 𝑴𝑷𝑺𝒑𝒐𝒔𝒊𝒕𝒊𝒗𝒆 𝑴𝑷𝑺𝒏𝒆𝒈𝒂𝒕𝒊𝒗𝒆
𝑴𝑷𝑺𝒔𝒊𝒈𝒏. 𝒂𝒏𝒏𝒐𝒖𝒏𝒄𝒆𝒎𝒆𝒏𝒕𝒔 𝑴𝑷𝑺𝒂𝒍𝒍 𝒂𝒏𝒏𝒐𝒖𝒏𝒄𝒆𝒎𝒆𝒏𝒕𝒔
21
benefits, in terms of distance to default, from the non-standard measures introduced by the ECB. These findings are consistent with the result of Chodorow-Reich (2014) who found that the soundness of the United States banking sector benefited from the introduction of unconventional monetary policy by the FED. I now move on to the results of the bank size related subsamples. The results are shown in table 5. Variables
-10.15* -12.02* Std. error (5.55) (6.29) R2 0.11 0.19 n 51 18 -7.85*** -13.94*** Std. error (3.08) (3.87) R2 0.12 0.50 N 51 18 4.64 1.40 Std. error (4.48) (3.60) R2 0.02 0.009 N 51 18
Table 5: Regression results of the bank size related subsamples (* 10% significance level ** 5% significance level *** 1% significance level)
The regression results of the bank size related subsample show highly negative and significant values for the mid and high capitalized banks. The most impressive impact is found for the mid-size institutions. The results suggest that a 1 basis point decrease in the surprise measure of the medium- to long-term yields results in a 0.1394 unit increase in the Distance to Default measure. The results of the low-cap subsample suggest that low capitalized banks do not benefit from unconventional monetary policy measures. A possible explanation for this finding relates to the collateral requirements for monetary policy transactions. It could be that low capitalized banks are not eligible to certain market operations because they do not meet the collateral requirements.2
2 https://www.ecb.europa.eu/paym/coll/html/index.en.html ∆𝑫𝒕𝑫𝒉𝒊𝒈𝒉𝒄𝒂𝒑
∆𝑫𝒕𝑫𝒎𝒊𝒅𝒄𝒂𝒑
22 Variables -10.29 -14.09 Std. error (9.47) (11.54) R2 0.02 0.08 N 51 18 -3.80 -0.67 Std. error (5.35) (6.7) R2 0.01 0.006 N 51 18 4.70 -1.79 Std. error (6.62) (5.82) R2 0.01 0.005 N 51 18 -10.79** -17.91*** Std. error (5.08) (6.20) R2 0.08 0.34 N 51 18 0.98 3.28 Std. error (3.60) (3.32) R2 0.001 0.05 N 51 18 -8.21 -9.35 Std. error (5.98) (5.963) R2 0.037 0.12 N 51 18
Table 6: Regression results of the country related subsamples (* 10% significance level ** 5% significance level *** 1% significance level)
The regression result of the country related subsample show very high negative coefficients for both Italy and France. These results suggest that Italy benefits the most from the unconventional monetary policy measures. Italy is one of Europe’s largest economies, but as a result of the financial crisis and the subsequent debt crisis the public debt of Italy rose to 116 percent of GDP in 2010. This is the second largest debt ratio in Europe; only Greece had a larger debt ratio. This exceptionally high debt ratio causes the interest rate on government bonds to skyrocket. The banking sector is highly exposed to the dynamics in the sovereign bond market for two reasons. First, the banking sector is a major buyer in the domestic sovereign bond market. Second, the sovereign bonds serves as collateral for different kind of financial transactions, including monetary policy transactions. The high exposure to the sovereign bond market in combination with the increased tension in these markets was the reason for the ECB to implement several bond purchase programmes (CBPP1, CBPP2, and CBPP3). The above results suggest that Italy benefits the most from the introduction of unconventional monetary policy. The second highest coefficient, however not significant, is observed for France. If we look at the dynamics in the France banking sector we see that France displays relatively high exposure to the Euro
23
area’s peripheral countries. In a press release on the website of Moody’s, where they motivate a recent downgrade of France government bonds they mention; “France’s exposure to peripheral Europe through its trade linkage and its banking system is disproportionately large”.3 The high exposure in the
France banking sector could be an explanation for the observed highly negative coefficient. According to data from the Bank of International Settlements, French banks are most exposed to Italian debt. From the total amount of claims at the end of 2010 ($866 billion) almost the half ($392,5 billion) was owed to French banks.4
4.2 Stock market reaction
This section describes the results of the stock market reaction. Table 7 displays the regression results for the full banking sample.
Variables
-5.35*** -5.20** -7.99** -5.38
Std. error (1.418) (2.12) (3.44) (3.50)
R2 0.22 0.27 0.19 0.08
N 51 18 25 26
Table 7: Abnormal return regression results of the full sample. (* 10% significance level ** 5% significance level *** 1% significance level)
The regression results of the full banking sample show highly negative and statistical significant coefficients. These results suggest that unconventional monetary policy-related announcements have a positive effect on stock prices. These results are in line with the findings of Roger et al. (2014) who also find that expansionary monetary policy shocks significantly raise domestic stock prices. If we look at the difference between the impact of a positive and negative monetary surprise value, we see that the coefficient for the negative monetary surprise sample is larger than the coefficient for the positive monetary surprise sample. Furthermore, the coefficient for the negative monetary surprise sample is statistically significant, whereas the coefficient for the positive monetary surprise sample is not. This suggest that the impact, in terms of stock prices, of an announcement with a negative surprise value is more pronounced than the impact of an announcement with a positive surprise value. Now I turn to bank size related subsamples. The results are listed in table 8.
24
Table 8: Abnormal return regression result for the bank size related subsample (* 10% significance level ** 5% significance level *** 1% significance level)
The regression results of the bank size related subsample shows highly negative and significant coefficients for both the high- and mid-cap subsamples. The coefficient of the high cap subsample suggests that a 1 basis point surprise decrease in the surprise measurement of the medium- to long-term yield corresponds to an increase of 0.0769 in the abnormal returns on announcement days. For the mid-cap sub-sample, a 1 basis point decrease corresponds to an increase of 0.0518 in the abnormal return on announcements dates. The minimal effect on the low cap subsample are in line with the results of the DtD measure.
In table 9 the results of the country specific subsamples are displayed. The results of the country specific subsamples shows highly negative and statistically significant coefficients for Italy, France and Spain. The observed coefficients for Italy and Spain are most both significant at the 1 percent level. If we compare the country specific results between the stock market reaction and the change in Distance to Default, we can conclude that France and Italy seem to benefit the most from the introduction of unconventional monetary policy.
25 Variables -5.547*** -5.40*** Std. error (1.32) (1.80) R2 0.26 0.35 N 51 18 -3.94*** -3.26 Std. error (1.29) (2.04) R2 0.15 0.13 N 51 18 -1.44 0.83 Std. error (2.93) (2.76) R2 0.004 0.005 N 51 18 -7.38*** -7.38*** Std. error (1.67) (2.26) R2 0.28 0.39 N 51 18 -3.50** -3.24* Std. error (1.46) (1.85) R2 0.10 0.16 n 51 18 -6.46*** -6.82** Std. error (1.68) (2.68) R2 0.23 0.28 N 51 18
Table 9: Regression results of the country specific subsample (* 10% significance level ** 5% significance level *** 1% significance level)
4.3 Robustness checks
In this section I will perform two additional robustness checks to investigate the robustness of the main results. First, because OLS regression models are highly sensitive to outliers it is important to check whether the regression results are robust against outliers in the observations. An outlier is defined as an observation which is not in line with the pattern of other observations in the regression. The full sample results from both the stock market and the distance to default regressions are checked for outliers. The observations of the distance to default regression contains one outlier. The observations of the stock market regression contains three outliers. To check whether the results are robust against outliers, I removed the observations which are classified as outliers. Table 10 displays the results of the outlier adjusted regression.
26 Variables 𝑴𝑷𝑺𝒂𝒍𝒍 𝒂𝒏𝒏𝒐𝒖𝒏𝒄𝒆𝒎𝒆𝒏𝒕𝒔 -6.22*** Std. error (2.48) R2 0.12 N 50 -7.38*** Std. error (1.195) R2 0.45 N 48
Table 10: results of the outlier adjusted regression (*** 1% significance level)
The outlier adjusted regression shows an improvement of the coefficient and the level of significance for both the distance to default and stock market reaction. To conclude, the removal of outliers leads to an improvement of the results, therefore we can conclude that the main results are robust against outliers. The second robustness check relates to the monetary policy surprise measure. For the main results I use a monetary policy surprise which is derived from both the change in the medium- and the long-term yield on German government bonds. As a robustness check, it is interesting to see whether the results are different when the monetary policy surprise variable is derived from one particular yield maturity. To empirically test the impact of changes in the monetary policy surprise variable, I constructed two different monetary policy surprise variables. 𝑴𝑷𝑺𝑬𝒖𝒓𝒐𝒔𝒄𝒉𝒂𝒕𝒛 is constructed from the
abnormal changes in the medium-term yield. Changes in the medium-term yield are derived from daily price changes in Euro-Schatz futures. The procedure for calculating yield changes from changes in futures prices is described in section 3.2. 𝑴𝑷𝑺𝑬𝒖𝒓𝒐𝒃𝒖𝒏𝒅 is the monetary surprise variable derived from
the abnormal changes in the long-term yield, changes in the long-term yield are derived from daily changes in the Eurobund futures price. Table 11 displays the regression results for the different surprise measures. Variables
𝑴𝑷𝑺
𝑬𝒖𝒓𝒐𝒃𝒖𝒏𝒅𝑴𝑷𝑺
𝑬𝒖𝒓𝒐𝒔𝒄𝒉𝒂𝒕𝒛 ∆𝑫𝒕𝑫𝒂𝒍𝒍 𝒃𝒂𝒏𝒌𝒔 62.71 -3.86* Std. error (42.68) (2.03) R2 0.04 0.06 N 51 51 𝑨𝑨𝑹𝒂𝒍𝒍 𝒃𝒂𝒏𝒌𝒔 26.78 -3.79*** Std. error (23.21) (0.99) R2 0.026 0.22 N 51 51Table 11: regression results for different monetary policy surprise variables. (* 10% significance level ** 5% significance level *** 1% significance level)
∆𝑫𝒕𝑫𝒂𝒍𝒍 𝒃𝒂𝒏𝒌𝒔
27
As mentioned before, the objective of unconventional monetary policy is to influence the medium- to longer-term yield. Table 11 show a very interesting finding, the results suggesting that the changes in distance to default measure and the stock market reaction are mainly driven by the changes in the medium-term yield. One possible explanation for the observed results relates to the nature of banking. Traditional banking, which depends on maturity transformation, profit from a decrease in the interest rate as long as long-term interest rates remain stable. To conclude, the second robustness check suggests that the observed results in the previous sections are mainly driven by the change in the medium-term yield.
5. Conclusion
The implementation of unconventional monetary policy instruments by the ECB was the most significant shift in the practice of monetary policy in the recent years. Despite the interest in the effects of unconventional monetary policy on macroeconomic variables, there is not much literature that relates to the impact of unconventional monetary policy on the soundness of the European banking sector. In this paper, I tried to contribute to the literature by studying the effects of
unconventional monetary policy on the European banking sector. I used an event study approach to gauge the effects of unconventional monetary policy. For all unconventional monetary policy-related announcements, I constructed a monetary policy surprise measure. I regressed these surprise measures on abnormal stock and abnormal changes in aggregate distance to default on all
28
Most of the monetary policy-related literature uses intra-day data to capture the effect of monetary policy decisions in a tight window. Because I do not have access to intraday data I used daily
observations of the market variables. Although it is very common to use daily observations in an event study, the use of daily observations can be classified as a limitation of this study. A
recommendation for future research that I would like to make is to compare the results of the impact of unconventional monetary policy with countries outside the European Monetary Union.
References
Brown, S., & Warner, J. (1980). Measuring security price performance. Journal of financial economics,
8(3), 205-258.
Brunnermeier, M., & Sannikov, Y. (2014). International credit flows, pecuniary externalities, and capital controls.
Buch, C., Eickmeier, S., & Prieto, E. (2014). In search for yield? Survey-based evidence on bank risk taking. Journal of Economic Dynamics and Control, 12-30.
Chodorow-Reich, G. (2014). Effects of unconventional monetary policy on financial institutions.
Brookings Papers On Economic Activity, 155-227.
Doh, T. (2010). The efficacy of large scale asset purchases at the zero lower bound. Economic Review, 5-34.
ECB. (2011). The Monetary Policy of the ECB. Frankfurt am Main: European Central Bank.
English, W., Van den Heuvel, S., & Egon, Z. (2012). Interest Rate Risk and Bank Equity Valuations.
Finance and Economics Discussion Series No. 2012-26.
Falagiarda, M., & Reitz, S. (2013). Announcements of ECB unconventional programs: Implications for the sovereign spreads of stressed euro area countries. Journal of International Money &
Finance 53, 276-295.
Jensen, M., & Meckling, W. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of financial economics, 305-360.
Jiménez, G., Ongena, S., Peydró, J.-L., & Saurina, J. (2014). Hazardous Times for Monetary Policy: What do Twenty-Three Million Bank Loans Say About the Effects of Monetary Policy on Credit Risk-Taking. Econometrica, 4963-505.
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, 113-161.
Joyce, M., Miles, D., Scott, A., & Vayanos, D. (2012). Quantitative Easing and Unconventional Monetary Policy - An Introduction. The Economic Journal, 271-28.
29
Krishnamurthy, A., & Vissing-Jorgensen, A. (2011). The Effects of Quantitative Easing on Interest Rates: Channels and Implications for Policy. Brookings Papers On Economic Activity (2), 215-265.
Kuttner, K. (2001). Monetary policy surprises and interest rates: Evidence from the Fed funds futures market. Journal of monetary economics, 47(3), 523-544.
Lambert, F., & Ueda, K. (2014). The Effects of Unconventional Monetary Policies on Bank Soundness (No. 14-152). International Monetary Fund.
MacKinlay, A. (1997). Event studies in economics and finance. Journal of Economic Literature, 13-39. Merton, R. (1977). An analytic derivation of the cost of deposit insurance and loan guarantees an
application of modern option pricing theory. Journal of Banking & Finance, 1(1), 3-11. Peersman, G. (2011). Macroeconomics Effects of Unconventional Monetary Policy in the Euro Area.
Working Paper Series 1397. European Central Bank.
Rajan, R. (2005). Has financial developments made the world riskier? . Economic Policy Symposium, 313-369.
Rogers, J., Scotti, C., & Wright, J. (2014). Evaluating asset-market effects of unconventional monetary policy: a multi-country review. Economic Policy, 749-799.
30 Appendix
Name: Country Market cap
1 Banco Santander Spain 67761
2 BNP Paribas France 64453
3 Intesa Sanpaolo Italy 49470
4 ING Groep Netherlands 49182
5 Banco Bilbao Spain 47863
6 Allied Irish Banks Ireland 38211
7 UniCredit SpA Italy 33357
8 Deutsche Bank Germany 32471
9 Société Générale France 31950
10 Crédit Agricole France 27074
11 KBC Group Belgium 23559
12 Caixabank Spain 20215
13 Natixis France 15347
14 Commerzbank Germany 11622
15 Bank of Ireland Ireland 11178
16 Erste Group Bank Germany 11061
17 Banco de Sabadell Spain 8732
18 Deutsche Postbank Germany 7739
19 Mediobanca Italy 7629
20 Crédit Industriel et Commercial France 6970
21 Banco Popular Espanol Spain 6958
22 Bankinter Spain 5959
23 Unione di Banche Italiane Italy 5722
24 Banco Popolare Italy 4683
25 Banca Popolare di Milano Italy 3944
26 Banca popolare dell'Emilia Romagna Italy 3552
27 Raiffeisen Bank International Austria 3421
28 Banca Generali Italy 2931
29 Banco Comercial Português Portugal 2893
30 HSBC Trinkaus & Burkhardt Germany 2471
31 Credito Emiliano SpA Italy 2019
32 Aareal Bank Germany 1904
33 Bank of Cyprus Cyprus 1588
34 Banco BPI Portugal 1485
35 Wüstenrot & Württembergische Germany 1476
36 Banca Piccolo Credito Valtellinese Italy 1304
37 National Bank of Greece Greece 1237
38 Alpha Bank Greece 1200
39 Banca Carige Italy 1176
40 DVB Bank Germany 1145
41 Banca Ifis Italy 1133
42 Van Lanschot Netherlands 829
43 Crédit Agricole d'Ile-de-France France 620
44 BKS Bank Germany 582
45 Piraeus Bank Greece 439
46 Crédit Agricole Brie Picardie France 385
47 IKB Deutsche Industriebank AG Germany 380
48 Eurobank Ergasias Greece 294
49 Crédit Agricole Nord de France France 272
50 Bank of Greece Greece 181
31
France Name: Country
1 BNP Paribas France
9 Société Générale France
10 Crédit Agricole France
13 Natixis France
20 Crédit Industriel et
Commercial
France
43 Crédit Agricole
d'Ile-de-France
France
46 Crédit Agricole Brie
Picardie
France
49 Crédit Agricole Nord
de France
France
Table 2: Banks included in the French subsample
Germany Name: Country
8 Deutsche Bank Germany
14 Commerzbank Germany
16 Erste Group Bank Germany
18 Deutsche Postbank Germany
30 HSBC Trinkaus &
Burkhardt
Germany
32 Aareal Bank Germany
35 Wüstenrot & Württembergische Germany 40 DVB Bank Germany 44 BKS Bank Germany 47 IKB Deutsche Industriebank AG Germany
Table 3: Banks included in the German subsample
Greece Name: Country
37 National Bank of Greece Greece
38 Alpha Bank Greece
45 Piraeus Bank Greece
48 Eurobank Ergasias Greece
50 Bank of Greece Greece
Table 4: Banks included in the Greek subsample
Italy Name: Country
3 Intesa Sanpaolo Italy
7 UniCredit SpA Italy
19 Mediobanca Italy
23 Unione di Banche
Italiane
Italy
24 Banco Popolare Italy
25 Banca Popolare di Milano Italy 26 Banca popolare dell'Emilia Romagna Italy
28 Banca Generali Italy
31 Credito Emiliano SpA Italy
36 Banca Piccolo Credito
Valtellinese
Italy
39 Banca Carige Italy
41 Banca Ifis Italy
Spain Name: Country
1 Banco Santander Spain
5 Banco Bilbao Spain
12 Caixabank Spain
17 Banco de Sabadell Spain
21 Banco Popular Espanol Spain
22 Bankinter Spain
Portugal Name: Country
29 Banco Comercial
Português
Portugal
34 Banco BPI Portugal
Table 5: Banks included in the Italian sub sample
32
Date
Event Type Description Monetary
surprise
12/18/2008 PR ECS The GC decided that the main refinancing operations will continue to be carried out through a fixed rate tender procedure with full allotment for as long as needed
-0.490 12/19/2008 PR ECS The GC decided to continue conducting US dollar liquidity providing operations -0.419 02/03/2009 PR ECS The GC decided to extend the liquidity swap arrangements with the FED -0.784 03/05/2009 PC ECS The GC decided to continue the fixed rate tender procedure with full allotment for all main refinancing operations,
special term refinancing operations and supplementary and regular longer-term refinancing operations for as long as needed.
1.483*
03/19/2009 PR ECS The GC decided to continue conducting US dollar liquidity providing operations 2.654* 04/06/2009 PR ECS The GC decided to establish a temporary reciprocal currency arrangement, swap line, with the FED 0.188 05/07/2009 PC,PR OT The GC decided to proceed with the ECS. In particular, the GC decided to purchase euro-denominated covered bonds
issued in the euro area, and to conduct liquidity-providing longer-term refinancing operations with a maturity of one year
-1.579
06/04/2009 PC OT The GC decided upon the technical modalities of the CBPP1 -0.874
06/25/2009 PR ECS The GC decided to extend the liquidity swap arrangements with the Fed 0.131 09/24/2009 PR ECS The GC decided to continue conducting US dollar liquidity-providing operations 0.075 12/03/2009 PC ECS The GC decided to continue conducting its main refinancing operations as fixed rate tender procedures with full
allotment for as long as is needed, and to enhance the provision of longer-term refinancing operations
-0.253 03/04/2010 PC ECS The GC decided to continue conducting its main refinancing operations as fixed rate tender procedures with full
allotment for as long as is needed, and to return to variable rate tender procedures in the regular 3-month longer-term refinancing operations
0.466
05/10/2010 PR OT,ECS The GC decided to proceed with the SMP, to reactivate the temporary liquidity swap lines with the Fed, to adopt a fixed-rate tender procedure with full allotment in the regular 3-month longer-term refinancing operations, and to conduct new special longer-term refinancing operations
-2.143*
06/10/2010 PC ECS The GC decided to adopt a fixed rate tender procedure with full allotment in the regular 3-month longer-term refinancing operations
-0.752 09/02/2010 PC ECS The GC decided to continue to conduct its main refinancing operations as fixed rate tender procedures with full allotment
for as long as necessary, and to conduct 3-month longer-term
refinancing operations as fixed rate tender procedures with full allotment
-0.673
12/02/2010 PC ECS The GC decided to continue to conduct its main refinancing operations as fixed rate tender procedures with full allotment for as long as necessary, and to conduct 3-month longer-term
refinancing operations as fixed rate tender procedures with full allotment
-0.663
33
03/03/2011 PC ECS The GC decided to continue to conduct its main refinancing operations as fixed rate tender procedures with full allotment for as long as necessary, and to conduct 3-month longer-term
refinancing operations as fixed rate tender procedures with full allotment
-1.518*
06/09/2011 PC ECS The GC decided to continue to conduct its main refinancing operations as fixed rate tender procedures with full allotment for as long as necessary, and to conduct 3-month longer-term
refinancing operations as fixed rate tender procedures with full allotment
0.577
06/29/2011 PR ECS The GC decided to extend the liquidity swap arrangements with the Fed 0.112 08/04/2011 PC ECS The GC decided to continue conducting its main refinancing operations as fixed rate tender procedures with full
allotment for as long as necessary, to conduct 3-month longer-term refinancing operations as fixed rate tender procedures with full allotment, and to conduct a liquidity-providing supplementary longer-term refinancing operation with a maturity of six months as a fixed rate tender procedure with full allotment
1.458*
08/08/2011 PR ECS,OT The GC decided to actively implement its Securities Markets Programme for Italy and Spain 1.029 08/25/2011 PR ECS The GC decided to extend the liquidity swap arrangement with the Bank of England, 0.421 09/15/2011 PR ECS The GC decided to conduct three US dollar liquidity-providing operations in coordination with other central banks -1.248 10/06/2011 PC ECS,OT The GC decided to continue conducting its main refinancing operations as fixed rate tender procedures with full
allotment for as long as necessary, to conduct 3-month longer-term re-
financing operations as fixed rate tender procedures with full allotment, to conduct two liquidity-providing supplementary longer-term refinancing operation with a maturity of twelve
and thirteen months as a fixed rate tender procedure with full allotment, and to launch a new covered bond purchase program (CBPP2)
-1.675*
11/03/2011 PR OT The GC decided upon the technical modalities of CBPP2 -0.896
11/30/2011 PR ECS The GC decided in cooperation with other central banks the establishment of a temporary network of reciprocal swap lines,
0.879 12/08/2011 PC ECS The GC decided to conduct two longer-term refinancing operations with a maturity of three years and to increase
collateral availability
1.423 02/09/2012 PC ECS The GC approved specific national eligibility criteria and risk control measures for the temporary acceptance in a number
of countries of additional credit claims as collateral in Euro system credit operations.
-0.823 06/06/2012 PC ECS The GC decided to continue to conduct its main refinancing operations as fixed rate tender procedures with full allotment
for as long as necessary, and to conduct 3-month longer-term
refinancing operations as fixed rate tender procedures with full allotment
-1.625*
06/22/2012 PR ECS The GC took further measures to increase collateral availability for counterparties -0.762* 07/26/2012 SP OT Draghi’s London speech (“. . . the ECB is ready to do whatever it takes to preserve the euro.”) -0.819* 08/02/2012 PC OT The GC announced that may undertake outright open market operations of a size adequate to reach its objective.
Markets disappointed for lack of details about OMT
