• No results found

Effect of the outright monetary transactions announcement : GIIPS vs. non-GIIPS

N/A
N/A
Protected

Academic year: 2021

Share "Effect of the outright monetary transactions announcement : GIIPS vs. non-GIIPS"

Copied!
22
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

Effect of the Outright Monetary

Transactions announcement:

GIIPS vs. non-GIIPS

Bachelor Thesis

Joris Jongerius 10324038

Supervisor: Damiaan Chen

(2)

Statement of Originality

This document is written by Student Joris Jongerius 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 for the contents.

(3)

Table of contents

Introduction 4

Literature Review 6 Data and Empirical Model 11

Results 14

Discussion and Conclusion 17

Reference List 19

(4)

1. Introduction

Up to July 2012, the European Central Bank (ECB) has used three different monetary mechanisms to support the European Union member states and banks in financial distress. These were the Covered Bonds Programmes (CBP), the Long Term Refinancing Operations (LTRO) and the Securities Market Programmes (SMP) (Siekmann, 2015, p2).

Despite these measures, the interest rates for the bonds of various euro area member states rose sharply in July 2012. Speculations were growing that southern European countries, Spain, Italy, Portugal and especially Greece, would exit the Euro. According to Darvas, this would lead to increased capital outflows from southern Eurozone members, fragmentation of credit markets across national borders and lower consumer and investment demand. All these consequences would negatively impacting output, employment, bank balance sheets and public finances (Darvas, 2012, p. 1).

While most policies during the crisis have aimed to solve problems concerning public finances, a new program has been announced in August 2012 to decrease the interest rates for the government bonds. The Governing Council of the ECB said that the Central Bank was going to intervene in the secondary markets for government bonds. To be exact, on August 2, 2012, the President of the ECB, Mario Draghi, said that ‘’the ECB may undertake outright open market operations’’. On September 6, 2012, the ECB announced three technical features of the Outright Monetary Transactions (OMT)

program. First, the purchases would concentrate on bonds with remaining maturities up to three years. Second, no quantitative limits would be considered at the beginning of the program. And third, bond purchases would be conditional, so not every country is directly eligible to get this support (Altavilla, Giannone & Lenza, 2014, p. 4).

After almost three years since its announcement, none of the euro area countries has asked to activate the OMT program yet. Existing literature focused on the financial and macroeconomic effects of the OMT announcements on particular countries. This study will go further and investigate the effects of the announcement on the 10-year government bond yields of Greece, Italy, Ireland, Portugal and Spain (the ‘GIIPS’

countries) and the non-GIIPS countries. After that, these two outcomes will be compared with each other to see if there are different reactions on the announcement. This way the success and implications of the policy will be measured. Doing this, it will give an answer to the central research question: does the announcement of the OMT program have a positive effect on the 10-year government bond yields?

(5)

This thesis proceeds as follows. Section 2 begins with a comprehensive literature review about the Outright Monetary Transactions program. Section 3 describes the methodology behind this study. It describes the used dataset, model and variables. Section 4 contains the results of this study. Finally, section 5 consists of the conclusion and the discussion.

(6)

2. Literature Review

In the literature review the theory behind the Outright Monetary Transactions program will be explained. After this, evidence from previous studies on the subject will be presented.

2.1 Theory on the OMT program

The Outright Monetary Transactions program is a monetary mechanism by the

European Central Bank. The ECB purchases bonds issued by Eurozone member states, so-called outright transactions, in the secondary, government bond market. This will bring down the market interest rates faced by countries subject to speculation that they might leave the Euro. The decision to introduce this policy has another major objective. The initiation of the policy aims to guarantee proper transmission of the ECB its interest rates to the euro area economy and the singleness of its monetary policy more broadly . The potential purchases of government bonds are seen to improve the confidence of investors in the Euro. This way they will be encouraged to eventually buy bonds in the normal market.

A Eurozone member state can get the OMT support if it meets certain conditions. The most important condition is that the member state has asked support from the European Financial Stability Facility/European Stabilisation Mechanism (EFSF/ESM) sovereign support program. One form of support from the EFSF/ESM is through the OMT program. To get this support, member states have to make agreements with the European Commission about how the country is going to recover from the economic situation. The agreements made between these two parties are recorded in the so -called

Memorandum of Understanding. The second condition is that the member state will fully

respect and will follow these agreements all the time; otherwise no purchases will happen through the OMT program. Additionally, the ECB reserves the right to mandate the International Monetary Fund (IMF) to participate in designing and monitoring of such programs.

The third and final condition is that the member state has managed to regain complete access to private lending markets. Only when the member state meets all these conditions, the OMT purchases can start. According to the definition of the ECB, a

(7)

has succeeded to issue a new government bond series with a 10-year maturity (Algemene Rekenkamer, 2015).

During the press conference on September 6, 2012, Mario Draghi announced the main features of the program. First, are were no ex-ante quantitative limits set on the size of the OMT. This means that the ECB has not predetermined how much money they want to use on the program. Transactions will be focussed on sovereign bonds with a maturity of between one and three years. Second, the ECB is accepting the same treatment as private or other creditors (pari passu). To be clearer, if in a hypothetical situation a country defaults, its central bank would also be bankrupt if it is pari passu to private bondholders. Third, market values and aggregate holdings in the OMT program are published weekly. The average duration and breakdown by the specific countries are published monthly. Under these specifications the market is expected to be fully aware of the actions taken by the ECB at all levels with respect to the program. Finally, the liquidity created through OMT is fully sterilized. This is the point where the OMT program is different from Quantitative Easing (QE). With QE, new money is printed to buy bonds from countries. Therefore, the total money supply increases. With the OMT program the total amount of money in the market is staying at the same level. This is called fully sterilised. The amount of money withdrawn from the markets through buying bonds by the ECB is returned into the market by selling other securities such as Treasury Bills.

The OMT program is different from earlier programs such as the Securities Markets Programme (SMP). The SMP had never the credibility to provide liquidity in a necessary case, because of its restricted resources. The OMT program does not have this problem, because the ECB said that the program would have no ex-ante quantitative limits on the potential sovereign bond purchases. The other key difference is the conditionality requirement: the euro system would be treated as pari passu with other creditors for bonds purchased through OMT.

2.2 Empirical evidence from previous studies

Existing literature has extensively focused on the financial effects of the OMT

announcement, but only on particular countries. Altavilla, Giannone and Lenza aim to quantify the financial and macroeconomic effect of the OMT announcement on Germany, France, Italy and Spain (2014, p. 2). Even though none of these euro area countries has

(8)

asked to activate the OMT program, Altavilla et al. suggest that asset prices such as bond prices should have, at least in part, incorporated the information publicly ava ilable to market participants. Due to the changes in financial prices, private agents may alter their behaviour, potentially affecting the rest of the economy.

In the first stage of their research, Altavilla et al. forecast the financial effect of the OMT announcement. In the second stage they forecast the macroeconomic effect of the OMT announcement. Only the first stage of their research is important for this paper. Their study looked at the daily data on bond yields (for the sample period January 2007 to February 2013) and conducted an event study to isolate the effects of the

announcements on financial prices. This has been done by regressing sovereign bond yields on an event dummy. The event dummy takes the value of one for the date of the OMT announcement and zero for other days. The outcome of this event study is that the OMT announcement had significant impacts on bond yields of Italy and Spain. The 2-year bond yield in Italy declined with 1.75%, while the Spanish 2-2-year bond yield declined with 2.09%. The bond yields of Germany and France were not significantly impacted (Altavilla et al., 2014, p. 13). They also investigated the effect of the OMT announcement on the 10-year government bond yields through the same method. The same conclusions were drawn for the 10-year government bond yields, although the effects were lower. For Italy the 10-year government bond yield declined with 0.69% whereas the 10-year government bond yield for Spain declined with 0.97%. The

announcement did not have a significant effect on the French and German government bond yields (Altavilla et al., 2014, p. 13).

Nagel, Krishnamurthy & Vissing-Jorgensen also investigated the effects of the OMT program on bond yields (2013). They focused mainly on Spain, Italy and Portugal. According to Nagel et al., there are different channels that can influence the bond yields. First, policy agents may take announcements by the ECB about government bond purchases as a more accommodative position of the ECB. This can have a direct effect in reducing government bond yields (Nagel et al., 2013, p. 7). Second, longer-term bond yields may reflect duration risk premia. If the ECB affects macroeconomic outcomes due to their announcements, it may affect duration risk premia and hence lower longer bo nd rates. Finally, suppose the underlying reason of a high bond yield of a country is the risk of insolvency. Then, if the ECB uses its OMT program, the policies will lower bond yields because the ECB increases the repayments of bondholders. More broadly, any channel

(9)

that lowers sovereign bond yields should lower default risk, which in turn lowers bond yields further (Nagel et al., 2014, pp. 7-8).

According to Nagel et al., the announcement of the OMT program by the ECB has been successful in lowering government bond yields. The channel that had the most influence in reducing the yield was the sovereign default risk channel (Nagel et al., 2014, p. 26). The effect of the OMT announcement, based on research by Nagel et al., on Italy was a reduction of 2.01% on the 2-year bond yield and 0.83% reduction on the 10-year bond yield. For Spain, it was a reduction of 2.48% on the 2-year bond yield and 1.29% on the 10-year bond yield. And for Portugal, it was a reduction of 0.74% on the 2 -year bond yield and 1.18% on the 10-year bond yield (Nagel et al, 2015, p. 26).

Hristov, Hülsewig, Siemsen and Wollmerhäuser investigated the potential effectiveness of the ECB its OMT program to restore the monetary transmission mechanism (2014). To do this, they used autoregressive models with time varying parameters. The regression has been done to examine the reaction of bank lending rates to movements in government bond rates. The regression has been done on several euro area periphery countries over the period 2003-2013. The set of countries includes Ireland, Italy, Portugal and Spain, because these are the countries that are expected to ask for support from the OMT program. According to Hristov et al. a major importance in the context of the OMT program is the strength and stability of the relationship between these interest rates (Hristov et al., 2014, p. 5).

The results of the regression are giving serious doubts on the potential effectiveness of the OMT program, as only a minor reaction of bank lending rates to movements in sovereign bond rates has been found. Before the outbreak of the government bond market crisis in 2010, the bank lending rates in the euro area

periphery countries already reacted only slightly to the changes in sovereign bond rates. But after the outbreak their response to changes in sovereign bond rates has been reduced significantly thereafter. Hristov et al. state that the theoretical framework of the OMT program, namely the opinion that the high bank lending rates are mostly

determined by the government bond rates, is hardly supported empirically.

Szczerbowicz also did research on the unconventional monetary policies. One of the policies he includes, is the OMT program. He measures the effects of the

unconventional policy on sovereign spreads in periphery Eurozone countries. According to Szczerbowicz, the periphery countries are Greece, Ireland, Italy, Portugal and Spain

(10)

(the so-called GIIPS countries). The spread is calculated for all periphery countries apart as the difference between the year government bond yield of Germany and the 10-year government bond yield of the periphery country. The regression relies on dummy variables to make clear when the announcements were made (Szczerbowicz, 2015)

According to Szczerbowicz, the announcement of the OMT program has a significant impact on the 10-year government bond yield spread. The response of the Spanish spread to the OMT was the strongest. The spread decreased with 0.59% at the 1% significance level. The impact on Italian and Portuguese spreads was significant at the 5% level. The spread of Italy dropped with -0.31% whereas the Portuguese spread dropped with -0.54%. The Irish spread was significant only at the 10% level and had a decrease of -0.24%. The Greek spread had no significant impact (Szczerbowicz, 2015)

The next section of the current research presents the data employed and the main empirical model used for the purpose of investigating what the effect was of the OMT announcement.

(11)

3. Data and Empirical Model

In this section the different variables used in the model are explained. All the data is gathered from the ECB and from the Organisation for Economic Co-operation and Development (OECD). The frequency for all the used data is quarterly. The countries used for this study are the countries within the Eurozone. This is done because the ECB is responsible for the monetary policy within the Eurozone and the ECB announced the OMT program. The Eurozone is the part of the European Union that has implemented the Euro as their national currency. Although the Eurozone currently counts nineteen countries, only the countries that joined the Eurozone from the beginning are added into the regression. These countries are Austria, Belgium, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, the Netherlands, Portugal and Spain. The other seven

countries joined the Eurozone at certain points in the period of interest and this would not give a fair view in the results. The period of interest covers the time frame from 1 January 2006 till 1 April 2015. The data has been found for 37 quarters in total. This is the case for every country that is in the regression. This way the financial crisis and the most actual data are included in the analysis.

To measure the effect of the OMT announcement on the 10-year government bond yields, this analysis uses three variables and two dummy variables. The first variable in the regression is inflation. The main goal of the ECB is to maintain price stability. To see what the effect of the main goal of the ECB is on government bond yields, inflation is added. The ECB uses the Harmonised Index of Consumer Prices (HICP) to measure consumer price inflation. The data for the percentage ch ange, which is quarter on quarter, has been acquired from the ECB. The HICP is not seasonally adjusted.

The second variable that was added to forecast the change in government bond yields is the percentage change in the Gross Domestic Product (GDP). The data was directly available from the website of the OECD. Data has been found for the twelve countries and for every quarter year. Every observation is the percentage change with respect to the previous quarter. The GDP growth is not seasonally adjusted.

The third variable is the yield of 10-year government bonds. The data was directly available from the website of the OECD. Every observation is the percentage change with respect to the previous quarter. The variable will have a lag value of one quarter to prevent it from autocorrelation.

(12)

Furthermore, there are two dummy variables used in the regression. The first dummy that has been used is the announcement of the OMT program. This is to see what the effect of the OMT program is. The ECB announced the OMT program on 16

September 2012. So for the third quarter of 2012 the value of the dummy is 1. For all the other quarters in the regression, the dummy has the value 0.

The other dummy that is used in the regression is for the period the countries of interest are in the financial crisis. Although there is no common view when the financial crisis actually begun, according to the ECB it started when the Lehman Brothers went bankrupt on 15 September 2008. Till today there are still financial problems in Greece, which means the financial crisis has not been solved yet. This means that the dummy will have the value of 0 for the first quarter of 2006 till the second quarter of 2008. From the third quarter of 2008 till the first quarter of 2015 the dummy changes to 1 and the Eurozone countries are considered to be in a financial crisis.

The main empirical model for the purpose of the current research uses a panel data regression in order to test the relationship between the 10-year government bonds yield, GDP growth, inflation rate, the effect of the financial crisis and the effect of the OMT announcement. In the first regression all the twelve countries are added into the model to see the total effect. After that, the regression is split up in two parts. The first part will be the regression for the non-GIIPS countries. These are in total seven

countries. No changes will be made with respect to the model. The second part is

regression for the GIIPS countries. These are in total five countries and again no changes will be made with respect to the model. The regression is split up because this study wants to investigate whether there is a difference between the GIIPS and the non-GIIPS countries. The general model is given below.

𝑌𝑖 ,𝑡 = 𝛽0+ 𝛽1∗ 𝐺𝐷𝑃𝑖,𝑡 + 𝛽2∗ 𝐼𝑛𝑓𝑖,𝑡 + 𝛽3∗ 𝐵𝑌𝑖,𝑡−1+ 𝛽4∗ 𝑂𝑀𝑇𝑡+ 𝛽5∗ 𝐶𝑅𝑡+ 𝜀𝑖,𝑡+ 𝐹𝐸/𝑅𝐸

The dependent variable Y is the 10-year government bond yield. 𝛽0 is a constant. 𝛽1and 𝛽2 are the coefficients for the growth of GDP and for the change in inflation respectively. 𝛽3 is the coefficient for the change in the 10-year government bond yield. It is the same bond yield as the dependent variable, although the moment in the timespan is different. It has a lag of one quarter. 𝛽4 and 𝛽5 are the coefficients for the dummy variables. FE/RE is added to include the Fixed Effects or the Random Effects. It is always possible to

(13)

include the Fixed Effects, but the Random Effects are giving a better result. To see whether to add the Fixed Effects or the Random Effects, the so -called Hausman test has to be done. If the p-value is greater than 0.05, Random Effects are added instead of Fixed Effects.

𝛽1 and 𝛽2 are expected to be positive coefficients for the non-GIIPS countries as the GDP growth and the inflation were positive during most of the time in these

countries. The opposite is true for the GIIPS countries, so here the expectation is that the two variables have a negative coefficient. When GDP grows, inflation will be stimulated and when inflation rises, bond prices tend to drop. If bond prices drop, bond yields are rising. Therefore, positive GDP growth and inflation are expected to have a positive effect on the bond yield. According to the data set bond yields were increasing overall, so 𝛽3is expected to be positive. For 𝛽4 the expectation is that the coefficient is negative, because the idea behind the OMT program was to lower the government bond yields for both non-GIIPS and GIIPS countries. 𝛽5 is expected to have a positive coefficient because during the financial crisis bond yields were highly increasing. This is the case for all the countries in the dataset, although the effect of the crisis on the GIIPS countries is presumably higher than on the non-GIIPS countries.

The next section of the current research presents the outcomes of the regressions and comments on the findings.

(14)

4. Results

The main regression results according to the afore-mentioned specifications are

presented in the next table. Significance was tested at the 5% level. The appendix of this paper contains the full regression tables according to the three different regressions that are done. The Hausman tests are also included in the appendix. The results from the Hausman tests show that all the three regressions are allowed to use the Random Effects (p>0.05). This means that the outcomes are based on the described model with Random Effects and not Fixed Effects. The outcomes of the regressions are presented below in table 1.

Table 1 - Effect of the OMT announcement on bond yields (Dependent variable: bond yield)

Independent Eurozone-12 non-GIIPS GIIPS

variables Regression (1) Regression (2) Regression (3) GDP Growth -0.68548 -0.09042 -0.86499

(0.48768) (0.70767) (0.68031) Inflation -0.64893 -1.17583 -0.43216

(0.66998) (1.27154) (0.77605) Bond Yield (1-quarter 0.55959*** 0.53169*** 0.56656*** lagged value) (0.04946) (0.07145) (0.06964) OMT -0.02011 0.07474* -0.15581*** (0.03407) (0.04452) (0.05137) Crisis -0.03693*** -0.05180*** -0.01560 (0.01400) (0.01840) (0.02105) Constant 0.01090 0.01175 0.00863 (0.01259) (0.01747) (0.01840) R-squared 0.2885 0.2647 0.3546

Note: Parenthetical figures show std. error values; * denotes significant at 1% ** denotes significant at 5% *** denotes significant at 10%.

(15)

The first regression has been done on all the twelve countries within the dataset. The variables bond yield and OMT have the expected signs. Bond yield has a positive coefficient, whereas OMT has a negative coefficient. What stands out, are the negative coefficients of both GDP growth and inflation, as one would expect that GDP growth and inflation would contribute positively to an increase in government bond yields. Although this is contrary to the expectations, it could be explained by the fact that both variables are not significant at the 10% level. This means that within this dataset and model, there is not enough evidence to conclude that government bond yields depend on th ese

variables. The bond yield is significant at the 1% level and this was expected. This is due to the fact that the bond yield is the dependent variable as well as one of the

independent variables, although with a time lag. The dummy variable for the fina ncial crisis is also significant at the 1% level, although the coefficient is negative in contrast to what is expected. The dummy variable of the OMT announcement is not significant, which means that in the total Eurozone the announcement had not the desir ed effect. It had no significant effect on the 10-year bond yields.

The second regression has been done on all the non-GIIPS countries that joined the Eurozone at the beginning in 2001. These countries are Austria, Belgium, Finland, France, Germany, Luxembourg and the Netherlands. In regression (2) no big changes occurred with respect to the variables GDP growth and inflation. Both coefficients are still negative in contrary to what was expected. The coefficient of GDP growth increased to a less negative effect, but the coefficient of inflation decreased to a more negative effect. Both coefficients are still not significant at the 10% level though. As expected, bond yields are still significant at the 1% level, although the effect decreased. The dummy variable for the financial crisis is also still significant at the 1% level. In

regression (2) it has a more negative effect than in regression (1). The dummy variable of the OMT announcement becomes significant in the second regression at the 10% level. This means that there is some evidence that the announcement of the OMT

program did have effect on the bond yields, although this was not the case in regression (1). Totally unexpected is that the announcement of the OMT program has an increasing effect on the government bond yields in the non-GIIPS countries.

The third and last regression has been done on the GIIPS countries. These are Greece, Ireland, Italy, Portugal and Spain. Like in regression (1) and (2), the coefficients of GDP growth and inflation are still negative, which are not in line with the

(16)

expectations. The variables in regression (3) are not significant at the 10% level either, so there is still not enough evidence that the variables are having effect on the

government bond yields. The government bond yield itself is still significant at the 1% level and has for all three regressions a significant impact, which is expected. What stands out is the dummy variable for the financial crisis. The dummy in regression (1) and (2) is significant at the 1% level, but in regression (3) the dummy variable is no longer significant. This means that the crisis did not have a significant impact on the government bond yields of the GIIPS countries. This was not expected before the regressions were done. On the other hand, the dummy variable for the OMT

announcement becomes significant at the 1% level in regression (3), meaning that there is enough evidence that the announcement did have an effect. This is in line with the objectives of the ECB. Although there is not enough evidence that the OMT

announcement had effect on the total Eurozone, it did have effect on the two groups of countries that were used in the second and third regression.

The following part of this research presents a summary of the paper and addresses the main shortcomings, which could be implemented by future researches.

(17)

5. Discussion and conclusion

The purpose of this study was to examine the effect of the OMT announcement by the ECB on the 10-year government bond yields within the Eurozone. Based on the results of the regressions the conclusion can be drawn that the OMT announcement by the ECB had no significant effect on the Eurozone in total. The announcement had a small

significant effect on the non-GIIPS countries, although the effect was increasing the bond yield, whereas the goal set by the ECB was to decrease the yield. The announcement had a greater effect on the GIIPS countries and the effect was in the direction desired by the ECB. This was also one of the conclusions of the paper discussed by Altavilla et al. They found that for Spain and Italy the effect of the OMT announcement had a significant decreasing effect on the government bond yields. This is in line with the results of this study. On the other hand, Altavilla et al. did not find a significant effect for France and Germany, whereas this study found a small significant effect on the non-GIIPS countries, which France and Germany are part of. Of course it could be that the other countries in the regression caused the significant effect. This is something that could be researched by new studies. In conclusion, Altavilla et al. found partly different results on the effect of the OMT announcement, for both 2-year and 10-year government bond yields.

Nagel et al. did focus on the 10-year government bond yields, but only on three of the five GIIPS countries. They analyzed the effect of the OMT announcement on Italy, Portugal and Spain. Their conclusion was that the 10-year government bond yield of Italy dropped with 0.83%, the bond yield of Portugal with 1.18% and the bond yield of Spain with 1.29%. According to Nagel et al., these are significant effects. Although this study gives no exact number about the drop in government bond yields, it has the sa me conclusion as Nagel et al. have. The OMT announcement had effect on the government bond yields of the GIIPS countries.

Nevertheless, it should be noted that the current paper might suffer from several shortcomings, which contribute to the outcomes. First of all, the variables GDP growth and inflation used to measure the change in government bond yields did not have a significant outcome in the three regressions. So there is not enough evidence to see the effect of these variables on government bond yields. Other variables should be included to increase the explanatory power of the model.

Additionally, the period investigated, although inclusive of 37 quarters and 12 countries, might be too narrow to show the relationship between the independent

(18)

variables and dependent variable. For example, the financial crisis in the Eurozone has not ended yet. To get a better effect on the measurement of the financial crisis the timespan of the model has to increase. If the timespan of the model increases, also more countries from the Eurozone can be added to the regression model, because the

countries joined the Eurozone less recently. This will increase the total observations and presumably also the explanatory power of the model.

Moreover, the statistical method used for the purpose of this study was a panel data regression, which might be too limited and simplistic. So extending the statistical procedures and tests is a necessary future task.

(19)

6. Reference List

Algemene Rekenkamer. (2015). Europese Centrale Bank (ECB). Retrieved from: http://www.rekenkamer.nl/Publicaties/Dossiers/E/EU_governance_bij_bestrijding_eco nomische_en_financiële_crisis/Financiële_stabiliteitsinstrumenten/Overzichten_financië le_instrumenten/Europese_Centrale_Bank_ECB

Altavilla C., Giannone D., & Lenza M. (2014). The Financial and Macroeconomic Effects of OMT Announcements. ECB Working Paper No. 1707.

Darvas Z. (2012). The ECB’s Magic Wand. Intereconomics, 266-267

Hristov N., Hulsewig O., Siemsen T. & Wollmershauser T. (2014). The potential effectiveness of the ECB’s OMT program in restoring monetary transmission. Retrieved from: http://www.sfm.econ.uni-muenchen.de/forschung/hhsw_2014.pdf

Nagel S., Krishnamurthy A. & Vissing-Jorgensen A. (2014). ECB Policies involving Government Bond Purchases: Impact and Channels. Retrieved from:

https://people.stanford.edu/akris/sites/default/files/ecbbonds.pdf

Siekmann, H. (2015). The Legality of Outright Monetary Transactions (OMT) of the European System of Central Banks. Institute for Monetary and Financial Stability,

Working Paper Series No. 90

Szczerbowicz, U. (2015). The ECB's Unconventional Monetary Policies: Have they lowered market borrowing costs for banks and governments? RIETI

(20)

7. Appendix

Regression (1) for the twelve Eurozone countries

(21)

Regression (2) for the non-GIIPS countries

(22)

Regression (3) for the GIIPS countries

Referenties

GERELATEERDE DOCUMENTEN

We add nuance to this finding by show- ing that, for discounts (both product-specific price discounts and order coupons) at lower levels, relatively high discount expec- tations

As Altides (2013) argues, national organisations gain easier access to their members and as the national media, in the absence of European mass media, inform people on the

The  last  two  chapters  have  highlighted  the  relationship  between  social  interactions   and  aspiration  formation  of  British  Bangladeshi  young  people.

As early as the takeover of Falluja, Obama’s Iraq policy was heavily criticized by Times reporters and the rise of ISIL was entirely his fault, because he didn’t leave ‘even a

However, the non-serious trailer also (unintentionally) provided an imposed goal (i.e., ‘designed to calm and soothe’). Moreover, no significant differences were found between

Independent variables: CFO = cash flows from operation, NDAC = non-discretionary accruals, DAC = discretionary accruals, INDEP = proportion of independent board, AUDCOM = proportion

The governance assessment tool has by now been used in research projects in fifteen countries for assessing the support for water projects and on top of that also in six countries

catalyzed glycerol esterification of free fatty acids in waste cooking oil for biodiesel 9. production, European Journal of Lipid Science and Technology,