• No results found

Quantitative Easing in Europe : the effect of a QE announcement on stock index prices

N/A
N/A
Protected

Academic year: 2021

Share "Quantitative Easing in Europe : the effect of a QE announcement on stock index prices"

Copied!
23
0
0

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

Hele tekst

(1)

Quantitative Easing in Europe: The effect

of a QE announcement on stock index

prices

Bachelor Thesis: Economics and Finance

University of Amsterdam

Rosalie van der Weide - 10204504

Supervisor: Gabriele Ciminelli

(2)

Statement of originality

This document is written by Rosalie van der Weide, who declares to take full

responsibility for the content of this document. I declare that the text and the

work presented in this document is original and no sources other than mentioned

in the test and its references have been used to creating it. The faculty of

Economics and Business is solely responsible for supervision of completion of

the work, not for the content.

(3)

Abstract

On the 22

nd

of January the ECB announced the beginning of their large-scale

asset purchase programme that started on the 9

th

of March. This unconventional

monetary policy is called quantitative easing and is used to reduce interest rates

and boost economic activity. In this thesis the effect of the announcement of

quantitative easing on the change in bank index compared to the main index is

tested using an event-study methodology. This is tested for six different

European countries with different yields on their 10-year government bonds.

The results show a significant effect of the QE announcement on both the bank

index and the main index implying that indeed investors expected a positive

market reaction. The main result shows that the banking system benefitted more

in countries where banks have a high market value and therefore have a larger

influence on the countries’ economy.

Table of content

1.

Introduction……….4

2.

Literature review……….6

- The channels behind quantitative easing………...6

- The implementation of quantitative easing………9

3.

Methodology………...11

- Hypothesis……… .15

4.

Results……….15

- Internal validity………..18

5.

Conclusion………...19

6.

Bibliography………20

7.

Appendix……….22

(4)

1 Introduction

On the 22nd of January 2015 the European Central Bank announced that it would begin with their asset purchase policy called Public Sector Purchase Program (PSPP). This program started last year on 9 March 2015 and will continue until at least September 2016 and when the Governing Council will detect a recovery of the inflation path (Claeys, Leandro, & Mandra, 2015).

This policy is also called quantitative easing and contains buying bonds worth of 60 billion on a monthly base from banks and other financial institutions by the ECB that are issued by the central governments, agencies and European institutions in the secondary market against central bank money (Claeys, Leandro, & Mandra, 2015). The effect of buying these assets is that these institutions can use the proceeds of selling bonds to either buy new bonds or to create more loans for households or other financial institutes. Furthermore buying bonds will increase their demand and thus their price and will decrease their yields. Therefore investments are stimulated because of the low interest. Due to these effects the goal of PSPP is to stimulate investments, consumption and simultaneously bring the inflation rate back around the targeted 2 per cent.

Japan was the first country that implemented this unconventional monetary policy around the year 2000. Since the beginning of the financial crisis, quantitative easing has been also used by the Federal Reserve and the Bank of England to stabilize the economy. In the United Stated QE had a significant effect on lowering the nominal interest rates (Krishnamurthy & Vissing-Jorgensen, 2011). Also in the United Kingdom there are several indications that their policy has worked to recover the economy in times of economic distress by lowering the nominal gilt yield (Joyce , Lasaosa, Stevens, & Tong, 2011).

There is still much debate about the effectiveness of quantitative easing in Europe. Supporters of the programme think that quantitative Easing will increase the economic activity because it helped the US, UK and Japan in stabilizing their economy in times of severe crisis. On the contrary opponents think that the programme does not have a real effect on the economy because the Eurozone is still involved in a macro-economic crisis. This crisis is fourfold and consists of a growth crisis, the debt crisis, the banking crisis and a crisis of

(5)

confidence. Due to these problems in Europe, using solely quantitative easing as a policy would not be sufficient to get Europe out of these crises (Fratzscher, et al., 2016).

In Europe banks play a central role in the financial system. The financial crisis has shown that banks were affected the most (Claessens & van Horen , 2014). This mainly due to the fact that banks are highly globally integrated with other financial institutes. Quantitative easing has an direct impact on the balance sheet of banks This is because buying large amounts of assets from banks and non-banks with central bank money, increases the price of assets (Joyce , Lasaosa, Stevens, & Tong, 2011). Banks mostly hold or sell assets, such as stocks and bonds. Therefore banks will benefit if those asset will increase in value. Furthermore, banks receive cash for selling their assets to the central bank. This will increase the money holdings of a bank’s balance sheet. Banks will use this money to lend to borrowers. Though these effects are not directly measurable since QE works through different channels. Also it is difficult to test the contribution of PSPP because of the simultaneous influences of economic events and developments (Joyce , Lasaosa, Stevens, & Tong, 2011). Therefore it is interesting to analyse the part of QE that can be tested, namely the announcement of the beginning of PSPP on the 22th of January. Stock market indexes are highly sensible for macro-economic announcements. A positive macro-economic signal that could lead to an increase in economic activity, will increase investor’s believes of future growth. Prospect of growth could lead to more investment opportunities and therefore demand for assets and their price will increase (Li & Zuliu, 1998). The use of QE in the US and the UK has showed that the announcements had a positive effect on asset prices. Since the countries in Europe are highly diverse in economic and political environment, it is interesting to test if this announcement has the same positive impact.

In this thesis I will use an event-study approach to examine the effect of the announcement of PSPP on the 22nd of January on the stock prices of banks in six European countries. I will test this using an OLS regression technique. First I will choose six European countries which all have a different financial situation depending on the interest rates of government bonds. As mentioned above, there are many converse opinions about the working of QE in Europe. One of these opinions states that European countries differ too much economically and politically. Due to these differences, the impact of the QE announcement will affect each country differently. Countries with a underdeveloped banking system could therefore have a minor price impact on stocks than countries that are higher developed. Since QE will decrease interest rates to around zero, and asset prices and interest rates are directly related, I will expect that countries with higher interest rates before the QE announcement

(6)

have a larger price impact compared to countries with lower interest rates. This is because a larger drop in rates will give a larger price impact.

After this I will construct two regressions for each country to test if banks benefit more from the announcement compared to the country index. Finally I will compare the result of the six countries to see if there is a significant difference between these countries due to the impact of QE announcements on different states of the economy. I expect that banks will benefit more from the QE announcement. Since central banks will buy assets directly from other banks, prices will increase and the value of the bank that holds those assets will increase as well. Investors are interested in gaining the highest returns on assets and therefore they will prefer holding assets with higher value. Banks are large holders of assets such as bonds and assets, compared to other companies and therefore they will experience a large increase in market value.

The results show that the announcement on the 22nd of January, 2015 had a positive effect on the returns of banks and the overall economy. This indicates that investors expected a future growth. Also it shows that in countries with a more stable economy, the price change was heavier compared to instable economies. Furthermore in four of the six countries, the change in prices was higher for the banking sector compared to the main economy. In these countries, banks are large financial institutes with a high market share. This implies that larger banks have a heavier impact on stock prices compared to countries with relatively smaller banks.

This thesis will be divided in five parts. Part 2 is a complete description about the existing literature on quantitative easing and its channels. The last part of section 2 will describe the implementation of QE in other countries. Part 3 is a full explanation about the method used to test the research question. Part 4 describes the results that came from the regressions. Part 5 concentrates on the conclusion.

2 Literature review

2.1 The channels behind Quantitative Easing

A standard economical assumption is that monetary policy has an effect on short-term rates. Central banks perform monetary policy by purchasing and selling short-term debt securities to influence the nominal short-term interest rates (Fawley & Neely, 2013). Furthermore CB’s can influence the monetary base by purchasing public bonds and lending money to banks and other financial institutes. There are two types of monetary policy, expansionary and

(7)

contractionary. When a central bank conducts an expansionary monetary policy it increases the money supply. With a contractionary policy the CB decreases the money supply. A main result of increasing the money supply is a decreasing interest rate. A low interest rate encourages households to spend more money instead of saving money. Additionally, borrowing is getting attractive due to the low interest payments, so this will stimulate agents to invest. Furthermore, interest rates are negatively related to asset prices. The prices of certain assets such as bonds and stocks tend to increase when interest rates are decreasing. In general, investors are interested in getting the highest return on their investments. When risk-free interest rates are falling below the coupon rate because of the monetary expansion, investors will have an increase in demand for bonds due to the higher expected return. Eventually when demand for bonds increases, the price will increase as well (Cox, Ingersoll, & Ross, 1985). When investors are searching for investments opportunities in stocks, they will require a certain compensation for the risk that they take above the risk-free interest rate. The risk premium is the return of a portfolio minus the risk-free rate. When the risk-free rate is falling, the risk premium will increase (Sharpe, 1964). This higher compensation for taking risk will attract investors. Ultimately, higher demand for stocks will increase their prices.

When interest rates are low, investors and households can always choose to hold currency instead of deposit it in a bank. This is the reason why nominal interest rates cannot go below zero. When interest rates are zero, conventional monetary policy loses its effectiveness (Fawley & Neely, 2013). Due to this limitation central banks started looking for other policy tools. In the late 1990’s the bank of Japan started developing unconventional measures in order to boost economic activity when interest rates are zero (Shirakawa, 2009).

The most used form of unconventional monetary policy is quantitative easing. This policy is conducted by purchasing large amounts of assets. This large increase in assets results in an expansion of a central banks’ balance sheet (Joyce , Miles , Scott, & Vayanos, 2012). Central banks buy assets from other banks with central bank money. This means an increase in the money reserves of those banks. Ideally, when banks hold more money, they get stimulated to lend more to households and enterprises. The effect of QE is not directly detectible because it works through different channels. The transmission mechanism of quantitative easing can be divided in four channels: price channel, signalling channel, exchange rate channel and portfolio rebalance channel.

The price channel has a direct effect on the prices of government bonds and stocks. In perfect capital markets asset prices only depend on risk-adjusted expected return. The market security line shows the relation between the systematic risk and expected return. This relation

(8)

will help to price the individual underlying security (Sharpe, 1964). If the policy of the CB buying assets does not change the adjusted expected return ultimately, the price will not change. In reality, capital markets are not perfect and there is a demand for holding government bonds and because they are riskless compared to holding other assets. Therefore, buying government bonds will increase its price because it decreases the availability to the public of these bonds. Hence, when the supply of bonds decreases, the price of bonds will increase.

According to Krishnamurthy & Vissing-Jorgensen (2011) unconventional monetary policy could only have a positive effect on the interest rate when central banks are credible committed to this policy. Therefore credibility can be achieved through the programme of buying larges amounts of long-term assets by the central bank. This is called the signalling channel and has an effect on the interest rates of the bond market. This channel works through the expectations hypothesis. The hypothesis states that the long-term interest rate is influenced only by current and expected short-term interest rates. Credibility of a policy will increase certainty about the future (Cox, Ingersoll, & Ross, 1985). Thus, when the central bank credible commit to quantitative easing, investors expect the interest rate to be low in the future. According to the expectations theory, the long-term interest rate is the average of the short term interest rates expected over the life of a long term bond. So when investors expect low short term rates, the long-term rate will also be low. Additionally, bond prices and interest rates are negatively related, so when interest rates are decreasing, the prices of bonds will increase equivalently. Therefore, the signalling channel has an effect on the prices of assets.

With the growing globalization thoughout the world, the effect of monetary policy on the exchange rate has become more important (Mishkin, 2001). If quantitative easing affects the short-term interests rates as discussed in the previous sub-paragraphe, investors will find it more attractive to invest their capital in foreign countries with higher interest rates compared to the domestic country. As a result of this capital outflow, the domestic currency will depreciate compared to the foreign currency (Pilbeam, 2013). This depreciation will lead to cheaper domestic goods and expensive foreign good, so therefore net export will increase. Eventually, an increase in net export will cause an increase in aggregate spending.

In contrast to the previous effect, the exchange rate effect could have a negative effect on the balance sheet of financial and non-financial companies. When central banks conduct a monetary expansion, the exchange rate will increase and the domestic currency will depreciate. When banks have their debt denominated in foreign currencies, a deterioration of

(9)

the currency means an increase in the value of their liabilities. Eventually the banks’ capital value will decrease and lending will be cut back (Mishkin, 2001).

It is therefore very important to distinguish between different countries and economic character. Industrialized economies normally do not have their debts denominated in foreign currencies so a monetary expansion is likely to have a positive effect on aggregate demand.

Finally, when CB’s buys assets of non-banking institutes in exchange of money, the central bank will enforce the rebalancing of portfolio from government to corporate bonds because banks sell government bonds to the central bank. (Tong, Joyce, & Woods, 2011). This channel is called the portfolio rebalance channel. When the central bank buys assets, the prices will increase and thus the value of the sellers’ money holders increase as well. Assuming that assets and money are not perfect substitutes, a change in the amount of assets will simultaneously result in a change in expected return (Joyce, Lasaosa, Stevens, & Tong, 2011). Due to these changes, investors will try to rebalance their portfolios by looking for better assets. Due to this channel, their will be constantly an exchange of excess money and assets. This mechanism will increase the price of assets and thus will decrease their yields. Lower interest rates will increase the spending behaviour of households and investors.

These four channels all have a direct and indirect effect on the interest rate and therefore on the prices of those underlying assets. Due to their direct effects on the prices of assets, the price channel and the signalling channel are the most important to discuss in this thesis. The direct effect on banks will be that due to buying large amounts of assets by the CB, prices of the stocks will increase and ultimately the value of holding these assets will increase. Therefore banks will experience a increase in the value of their stock. Also, the direct effect of the signalling channel could be that investors will expect low short-term interest rates, that could be interpreted as a future growth of the economy. Therefore these investors will increase their investments in order to make profit in the future. The demand for assets will increase and thus their price will increase. The portfolio balance and the exchange rate channel will also influence asset prices and therefore are important to mention. But in contrary to the price and signalling channel, these channels have a large indirect effect on asset prices and so its very difficult to capture these impacts.

2.2 The implementation of Quantitative Easing.

On March 19, 2001, Japan was the first country who conducted the use of quantitative easing in order to increase the stagnating economy. Their programme consists of three pillars. The first pillar was an increase in the current account balance held by the commercial banks. This

(10)

meant a change in its operating target from the overnight call rate to the outstanding balances. (Shiratsuka, 2010). The second pillar included to carry on with the first pillar until the consumer price index inflation increased to around zero. Finally, the last pillar included the increase of the amount of the BOJ’s long-term Japanese government bonds. The effects of the nearly six year QE programme were limited on output and inflation. This was mainly due to the failure of the transmission channel between the financial and non-financial markets. On the contrary to the non-financial markets, QE did have a positive impact on the financial markets. The credit spread declined significantly after the start of quantitative easing (Shiratsuka, 2010). Furthermore in the period between 2000 and 2009 there was an increase in the liquidity of banks due to promoting of bank lending. Together with the decrease in the credit spread, this means that QE programme helped to enhance the economy. Finally the overall two effects of the programme of the bank of Japan are that there were some significant declines in interest rates. Secondly, the programme has helped very weak banks to recover from the economic downturn and therefore it increased the risk-tolerance in the Japanese banking system (Spiegel, 2006).

The second country that implemented quantitative easing is the Federal Reserve. After the fall of the Lehman brothers, the Fed implemented a policy that would spur economic activity. Krishnamurthy & Vissing-Jorgensen (2011) evaluated the effects of purchasing long-term Treasuries and other long-long-term bonds on the interest rates. They divided the total period in two sections; QE1 (2008-2009) and QE2 (2010-2011). For both of the outcomes of QE1 and QE2 is that the announcements of quantitative easing had a strong and significant effect on the decrease of yields. Additionally this was the result of the signalling effect. Signalling channel is also called macro/policy news channel and covers the expectations of economic agents on the future state of the economy and the future monetary policy decisions. Another outcome of QE1 and QE2 is the incease in inflation expectations and a decrease in inflation uncertainty.

Research has also been done on the effects of the Bank of England’s unconventional monetary policy. As well as for the Fed, the BoE implemented their Quantitative Easing program after the start of the global financial crisis (Joyce , Lasaosa, Stevens, & Tong, 2011). Their main focus was on enhancing the balance sheets of banks by purchasing private and public assets. Joyce et al. (2011), investigated this policy and found that gilt yields were significantly lower due to quantitative easing. They also found that most asset prices recovered from 2009. Finally the overall conclusion that can be drawn for both the Fed and the BoE is that quantitative easing has helped to decrease the yield terms and increase the

(11)

economic activity. However, the effectiveness of QE can only be examined by their macro economic impact because the direct impact on the real economy is still uncertain (Joyce , Lasaosa, Stevens, & Tong, 2011).

3 Methodology and dataset

The effect of quantitative easing announcements on the change in index of banks compared to the overall market index in Europe can be measured by an even-study analysis. This section will describe the methodology and data source that has been used to perform the regression. The last subparagraph describes the hypothesis that has been formed as a result of the existing literature.

First, to answer the thesis question I will search for six different European countries with different yields on their 10-year governmental bonds. On these six countries I will perform two regressions with data of relevant announcements that could influence the stock prices conducted from the period of the 1st of January 2014, until the 31th of December 2015. The first regression will test the impact of the quantitative easing announcement of 22 January 2015 on the country’s bank index. The second regression will test the same impact of the announcement on the main index. After this I will check if there is a significant different in impact between these two indexes to answer my thesis question.

In this research there will be six different countries examined based on different interest rates on 10-year government bonds on the 16th of January 2015. As mentioned above, the effect of QE is that interest rates will decrease around zero. Since banks are holders of government bonds, a decreasing yield would lead to increasing prices. The higher the initial interest rate, the more room it has to decrease to zero, and thus the larger the price impact. The date is relevant because it is the week before the announcement of QE and therefore the interest rate is not affected yet by this economic shock. Furthermore, due to the difference in interest rates between countries, it is interesting to see which country is most impacted by the quantitative easing announcement. The interest rates of the 10-year government bonds are collected from Bloomberg and investing.com and are present in table 1.

(12)

Table 1.

Assuming the price channel will increase the price of assets, it is interesting to test whether the impact of this announcement is bigger in the banking market compared to the overall market. Financial institutes and other companies that are holders of assets will also be affected by the price increase. To test this I collected the index data of each country’s overall market and banking market from Datastream. The six countries each have different national indexes and these indexes are presented in table 2.

Table 2.

For each country, I will conduct two regressions. The dependent variable of the first regression will be the daily change in bank index between the period of the 1st of January 2014 and the 31th of December 2015. The dependent variable of the second regression will be the change in main index, taking the same period as in regression 1. To calculate the change in indexes I will use the capital gain formula to calculate the return of stocks.

𝑆𝑡𝑜𝑐𝑘 𝑅𝑒𝑡𝑢𝑟𝑛 = 𝑃𝑡− 𝑃𝑡−1 𝑃𝑡−1

To test the effect of the announcement of 22th of January I will make this independent variable a dummy variable that takes the value of 1 on the day of the announcement and a zero on the remaining days. Since stock prices are only trading five days of the week, the amount of testing days in my regression will be 522. Macro-economic news is an important variable that impacts the interest rates and thus the prices of assets (Krishnamurthy & Vissing-Jorgensen, 2011). Therefore solely focussing on the announcement of QE as an independent variable could lead to omitted variable bias. Additionally, because this news plays an important factor in influencing asset prices, I will

Country Interest rate 10-years government bonds (in %)

Germany 0,454 France 0,543 Belgium 0,604 Spain 1,502 Italy 1,658 Portugal 2,210

Germany France Belgium Spain Italy Portugal

Bank .TRXFLDDEPBANK .TRXFLDFRPBANK .TRXFLDBEPF12 .TRXFLDESPBANK .TRXFLDITPBANK .TRXFLDPTPBANK Country .TRXFLDDEP .TRXFLDFRP .TRXFLDBEP .TRXFLDESP .TRXFLDITP .TRXFLDPTP

(13)

include announcements of ECB meetings, European inflation, European GDP, European employment, US GDP and US non-farm payroll. These announcements will capture the surprise effect of these market releases. These surprise effects will account as control variables that will isolate the effect of the QE announcement from other announcement that might affect financial markets. This is calculated by the actual rate – the forecasted rate. These releases show how the economy has behaved in the last future. With the data of these other macro-economic announcements I can test if the QE announcement has an impact on returns. The announcement data and market expectations are found in forexfactory.com. With all these variables I can set up the following models for each country.

∆𝐵𝑎𝑛𝑘 𝐼𝑛𝑑𝑒𝑥𝑡 = 𝛽0+ 𝛽1𝑄𝐸𝑡+ 𝛽2 𝐸𝐶𝐵𝑡+ 𝛽3𝐺𝐷𝑃 𝑡+ 𝛽4𝑈𝑁𝐸𝑀𝑡+ 𝛽5𝐼𝑁𝐹𝐿𝑡+ 𝛽6 𝐺𝐷𝑃𝑈𝑆𝑡+

𝛽7𝑈𝑆𝑃𝐴𝑌 + 𝜀𝑡

∆𝑀𝑎𝑖𝑛 𝐼𝑛𝑑𝑒𝑥 𝑡 = 𝛽0+ 𝛽1𝑄𝐸𝑡+ 𝛽2 𝐸𝐶𝐵𝑡+ 𝛽3𝐺𝐷𝑃 𝑡+ 𝛽4𝑈𝑁𝐸𝑀𝑡+ 𝛽5𝐼𝑁𝐹𝐿𝑡+ 𝛽6 𝐺𝐷𝑃𝑈𝑆𝑡+

𝛽7𝑈𝑆𝑃𝐴𝑌 + 𝜀𝑡

𝑄𝐸 tests the effect of the announcement of quantitative easing on the 22th of January 2015. This is a dummy variable that will have a value of 1 on the announcement day and a zero on every other testing day.

𝐸𝐶𝐵 is also a dummy variable and tests the impact of the announcements of the ECB. By including this variable there will be distinction made between the impact of normal ECB meetings and the QE announcement.

𝐺𝐷𝑃 measures the announcements on the European GDP. This variable is a

surprise variable and is calculated by subtracting the forecasted level of GDP from the actual level of GDP. When the actual level of GDP is higher than forecasted the economy has done better than expected. This forecast is an indicator of economic health and reacts fast to market conditions. Changes in this forecast could be a sign of a change in economic activity.

Additionally, when this announcement is released, it has a strong impact on the financial market. The index is reliable because of its large sample size and its correlation with the Eurozone’s economic health (Felbermayr, 2016). Several researches has been done between the correlation of GDP growth and stock returns and there are some contradicting results. Siegel (1998) stated that here was not a significant effect between GDP growth and stock returns because expected GDP growth is already incorporated in stock prices. He found out that due to this discounting the future returns would decrease. In contrast, Li and Zuliu (1998)

(14)

shows that the expectation of economic growth can influence the investment behaviour of economic agents. Therefore the expectation of an increase in GDP can in turn lead to an increase in the demand of bonds and stocks.

𝑈𝑁𝐸𝑀 measures the impact of the announcements on unemployment rates of the ECB. This variable is also a surprise variable and is calculated by subtracting the forecasted unemployment level from the actual unemployment level. Furthermore, it measures the percentage of the total workforce that is unemployed and is seeking for employment. This announcement tends to have a toned impact on the economy. Although this variable has a less severe impact on the economy, it is still important to incorporate because the unemployment rate is a good measurement of the health of the overall economy because the spending behaviour of households and firms and related with the labour market.

𝐼𝑁𝐹𝐿 is a surprise variable that tests the impact of the expectation of future

inflation rates in Europe. This is calculated by subtracting the forecasted CPI percentage from the actual CPI percentage. This is measured by the Consumer Price Index Flash Estimate of the ECB. The CPI is an indicator of inflation because its measures increases in consumer prices. Taylor (1979) states that the announcement of future policy objectives can create a direct stabilization effect. When uncertainty about the future is decreased, this in turn can lead to reduced incentives for inflationary wage and price bias.

𝐺𝐷𝑃𝑈𝑆 is a surprise variable that measures the impact of the Fed’s announcements of the GDP growth of the US. This variable is calculated the same way as 𝐺𝐷𝑃. Since the US has the largest world economy in the world it is important to include this economic shock in the regression. In 2015 the total trade between the US and Europe was 619,660 billion euros. This makes the US the largest country that import goods from Europe. Thus this would indicate that Europe is dependent of the US’s economic health. Smith & Devereux,

(1994) emphasized that highly intregrated markets will influence each others asset prices due to spillovers.

In addition to 𝐺𝐷𝑃𝑈𝑆 I include the 𝑈𝑆𝑃𝐴𝑌 variable as a surprise variable as well. The announcements that have the most impact on the financial markets of both the US and Europe are the US releases on payrolls (Goldberg & Leonard, 2003). The NonFarm Payroll Reports give inside in the quantity of umemployment for the US workforce. These statistics are a good measurement for the economic health of the US and have a direct impact on the stock market, the exchange rate and the value of gold. Goldberg & Leonard (2003) found out that the release of these NonFarm Payroll reports have a significant impact on German yields

(15)

within a hour of its announcement. Since Europe and the US are highly interdepent, the announcement of these report would have a large influence on the financial market of Europe.

3.1 Hypothesis

In this section two hypothesis will be formed about the effect of the quantitative easing announcement.

The effect of the announcement of quantitative easing on the 22nd of January has a positive effect on the stock indexes of banks and the main index of each country.

As mentioned earlier, positive macro-economic announcements can affect the expectations of economic agents. (Li & Zuliu, 1998). The announcement of QE can be seen as an sign of future growth, because of the low interest rates and therefore the increase in economic activity. Due to the asset purchases the price will increase and their yield will decrease. This implies an expected positive price change on stock indexes.

Countries with higher interest rates on their 10-year government bonds will benefit more from the announcements than countries will with a lower yield.

One main tool of QE is that interest rates will decrease to approximately or around zero as a result of this operation. As discussed in the literature review, changes in interest rates will inmediately change the value of assets such as stocks and bonds. So the higher the initial interest rate, the heavier the price change will be after those rates drop to zero. For the six countries tested this implies that Portugal should have the highest price impact. Germany had the lowest initial interest rate and therefore should have the lowest price impact.

4. Results

In this section I will explain the results of my OLS regressions. Due to the possibility of autocorrelation, heteroskedasticity-robust standard errors are used to test. The results are tabulated in table 3 and 4. The first row after each control variable shows the coeficient and the second row in brackets are the standard errors. The significance of the coeficients are signed with *,** or ***.

(16)

Table 3. The determinants of the change in bank stock index prices.

Table 3. decribes the outcomes of the six regressions of the bank index of the six countries. The R2 is very low in every model, but the model could still predict some changes due to the significant coeficients. The announcement effect of the quantitative easing programme is very significant in every country tested. This means that economic agents did expected a positive future state of the economy. Furthermore, the positive coeficients indicate that these outcomes are consistent with the hypothesis that the announcement of QE will positively increase the returns of the banking index. The country with the largest price impact is France (2,92%). This means that on the day of the announcement, the french banks’ return increased by 2,92%. Initialy France had an bond yield of 0,543%. This indicated a low yield and a high impact on prices which contradicts with the expectations. The same conclusion can be drawn for Germany which had an increase in stock return of 2,69% on the announcement day and a bond yield of 0,454%. These contradictions could be explained by the fact that France and Germany have the largest gross domestic product of Europe. Additionally, of the top 4 banks of Europe, three of them are located in France and Germany. Therefore these two countries can be seen as stable economies to invest in. Investors are more attractive to stable economies because these will reduce uncertainty. Ultimately, less uncertainty about the future state of the economy will lead to a decrease of risk in investments (Schneider & Frey, 1985). Therefore investors will be more interesting in buying stock of large banks that offer less risk in times

(17)

of expected future growth. Surprisingly, Portugal had the highest initial yield and the lowest impact on the bank returns. This could also be explained by the fact that Portugal has high unemployment rates and govermental debt that will decrease the expectation of increasing economic activity. Besides, banks in Portugal are perceived to be very weak due to low banking conditions. As mentioned before, investors will choose more stable and economic healthy banks to buy stock because this reduces their own risk. In the appendix the tables are presented with Europe’s top 25 banks and companies in 2015.

Table 4. The determinants of the change in the overall stock index prices

Table 4. shows the results of the second regression on the main index of the six countries. In this table all the coeficients of the quantitative easing announcements are significant. Also here investors had positive market expectations. This outcome is consistent with the expectation that quantitative easing has a positive effect on the prices of stocks of the overall economy. Of all the regression the country in which the announcement had the largest impact is Italy (2,48%). This country had an initial bond yield of 1,658% and therefore is consistent with the hypthosis that changes in higher yields will imply higher price changes. France and Germany still have the second and third largest impact so this can also be explained by the fact that in those two countries very large companies with high market values are operating. Li and Zuliu (1998) showed that size of a company and profitability are positively related. Large companies have a higher market share and therefore are more protected against

(18)

economical influences. Small companies tend to be more sensible for fluctuations in the economy, because they are more exposed to several risks such as changes in the risk premium. As with the bank return, the QE announcement does not have a big impact on the returns of the economy of Portugal. The research of Li and Zuliu (1998) could explain the low coeficient of Portugal due to the country’s large amount of small companies. Another characteristic of a small company is that it is highly bank financially dependent. Due to the weak banking system in Portugal, the companies that are lending from the banks are also very risky for investors.

To answer the thesis question if banks benefitted more from the QE announcement than the overall economy, there need to be an comparison between the two coeficients of each country. Surprisingly, not in every country did the banking sector benefitted more from the QE announcement. In Germany, France, Spain and Italy the bank returns where higher than the returns of the economy. These outcomes are consistent with the expactations that banks will get higher returns on their stocks. In Belgium and Portugal the bank returns are lower than the overall return. The explanation for Belgium and Portugal could be that those two countries have a small number of large banks compared to Germany, France, Italy and Spain. Also in both of the countries, the market value of those banks are smaller than the market value of large companies implying that the announcement effect has a bigger impact on the non-banking sector due to their larger market value.

4.1 Internal validity

Looking at the causal relation in the regression I can conclude that indeed the QE announcement have a significant effect on stock prices. However it is very difficult to give conclusive answers due to the fact that QE works through many direct and indirect channels that have certain time lags that affect the economy contineously. The real affect of the QE announcement could therefore have not yet been captured completely because of the relatively short testing date from 1 January 2014 untill 31 december 2015. Also investors’ expectations are very hard to measure and to forecast because human behaviour changes all the time. Futhermore the Ordinary Least Squares regression is a very basic econometric estimation technique. The model has a lot of assumptions and thus can limite the outcomes if these assumptions do not hold. A suggestion would be using a statistic technique that is more advanced in estimating certain economic impacts.

(19)

5. Conclusion

Unconventional monetary policy has been succesfully used before by the BoJ, BoE and the Fed to enhance economic activity in times of crisis. On the 22nd of January the ECB also announced that it would start with an unconventional monetary policy called quantitative easing. This paper analysed the effect of the quantitative easing announcement on the 22nd of January 2015 on the change in index prices of stocks of banks compared to the change in prices of the main index of the economy in Germany, France, Belgium, Italy, Spain and Portugal. These countries are selected based on their different 10-year governemnt bond yield. The coeficients of the regressions performed are estimated with OLS. Based on existing literature two hypothesis were formed. Due to the announcement, investors will get a positive expectation about the future’s state of the economy and therefore the announcement will have a positive effect on stock returns. The results show that all the coeficients are significant and positive related the the stock indexes in both the banking and the non-banking sector. This outcome corresponds with existing literature and therefore could be said that the QE announcement of 22 January, 2015 increased the returns on banks and other companies of all the six countries tested. The second hypothesis assumed that prices of countries with higher yields on their corporate bonds would change more due to the QE announcement. The outcome did not give evidence that this hypothesis can be accepted. Finally, the comparison between the difference in impact between bank return and the overall main index return gave some surprising results. Out of the 6 countries tested, Gernany, France,Italy and Spain had a higher bank return compared to the return of the country’s index. In all those countries there are large banks operating, so there overall economy depend heavily on those banks. On the contrary, Belgium and Portugal both showed larger returns on their country’s index. These two countries have less large banks compared to the other countries and this could explain the smaller impact. Additionally, these outcomes suggets that European countries are very diverse and that large economical differences exists. Due to these differences the impact of the QE announcement can result in outcomes that were not expected.

(20)

6. Bibliography

Claessens, S., & van Horen , N. (2014). The impact of the Global financial crisis on banking

globalization. IMF working paper.

Claeys, G., Leandro, A., & Mandra, A. (2015). European Central Bank Quantitative Easing:

The detailed manual. Bruegel .

Cox, J. C., Ingersoll, J. E., & Ross, S. A. (1985). A theory of the term structure of interest

rates (Vol. 53). Wiley: The econometric society.

Fawley, B. W., & Neely, C. J. (2013, januari ). Four stories of Quantitative Easing . Federal

Reserve Bank of st. Louis Review, pp. 55-88.

Felbermayr, G. (2016). Economic Analysis of TTIP. IFO working paper.

Fratzscher, M., Gropp, R., Kotz, H.-H., Krahnen, J., Odendahl, C., Weder di Mauro, B., et al. (2016, april 10). Retrieved juni 2, 2016, from Bruegel :

www.bruegel.org/2016/04/mere-criticism-of-the-ecb-is-no-solution/

Goldberg, L., & Leonard, D. (2003). What moves sovereign bond markets? The effects of Economic news on US and german yields. Federal Reserve Bank of New York, 2-7. Joyce , M. A., Lasaosa, A., Stevens, I., & Tong, M. (2011, september). The financial Impact

of Quantitative Easing in the United Kingdom. Bank of England, pp. 43-44.

Joyce , M., Miles , D., Scott, A., & Vayanos, D. (2012, november ). quantitative easing and unconventional monetary policy - an introduction. The economic journal, pp. 272-290. Krishnamurthy , A., & Vissing-Jorgensen, A. (2011, october ). The effects of Quantitative

easing on interest rates: Channels and Implications for Policy. The National Bureau of

Economic Research, pp. 36-37.

Li, L., & Zuliu, H. F. (1998). Responses of the stock market to macroeconomic

announcements across economic states . Working paper of the International Monetary

Fund.

Mishkin, F. S. (2001, December). The transmission mechanism and the role of asset prices in monetary policy . National Bureau of Economic Research .

Pilbeam, K. (2013). International Finance. Basingstoke: Palgrave MacMillan . Schneider, F., & Frey, S. B. (1985). Economic and Political Determinants of Foreign

(21)

Sharpe, W. F. (1964). Capital Asset Prices: A theory of market equilibrium under conditions

of risk (Vol. 19). Wiley.

Shirakawa, M. (2009). Unconventional monetary policy - Central banks : facing the

challenges and learning the lessons. Bank of Japan.

Shiratsuka, S. (2010). Size and Composition of the Central Bank balance sheet: revisiting Japan's experience of the quantitative easing policy. Monetary and economic studies. Siegel, J. J. (1998). Stocks for the long run. Mcgraw-Hill.

Smith, M. B., & Devereux, W. G. (1994). International risk sharing and economic growth. Osaka University: Wiley.

Spiegel, M. M. (2006). Did quantitative easing by the bank of Japan work? Economic

research.

Taylor, J. B. (1979). Estimation and control of a macroeconomic model with rational expectations. 47(3), 1267-1286.

Tong, M., Joyce, M., & Woods, R. (2011). The United Kingdom's quantitative easing policy:

(22)

7. Appendix

Table 5. Top 25 banks in Europe Source : relbank.com

Bank Country Assets (dec 2015) in billions

HSBC holdings UK 2409.66

BNP Paribas France 2180.45

Credit Agricole Group France 1857.57

Deutsche Bank Germany 1781.29

Barclays PLC UK 1659.77

Banco Santander Spain 1465.44

Societe Generale France 1459.02

Groupe BPCE France 1275.49

Royal Bank of Scotland UK 1208.37

Lloyds Banking Group UK 1195.45

UBS AG Switzerland 954.27

Unicredit S.p.a Italy 940.796

ING Group Netherlands 920.389

Credit Suisse Group Switzerland 830.774

BBVA Spain 820.134

Credit Mutual France 808.906

Intensa SanPaolo Italy 739.679

Rabobank Group Netherlands 732.985

Nordea Bank Sweden 707.284

Standard Chartered PLC UK 640.483

Commerzbank AG Germany 582.389

KFW Group Germany 549.979

Danske Bank Denmark 482.561

DZ bank AG Germany 446.479

(23)

Table 6. Top 25 companies in Europe Source: forbes.com

Company Country Market Value (dec 2015) in billions

Novartis Netherlands 272.6

Nestle Switzerland 247.3

Roche Holding Switzerland 240.4

Anheuser Busch Germany 204.6

Royal Dutch Shell Netherlands 195.4

HSBC Holding UK 167.8

Novo Nordisk Norway 147

Sanofi France 136

Unilever Netherlands 129.1

Bayer Germany 126.4

Volkswagen Germany 126

BP UK 120.8

Total S.A. France 120.2

Actavis Ireland 116.7

GlaxoSmithKline UK 114.1

Santander Group Spain 109.4

Loreal Group France 106.6

Inditex Spain 103.4

Daimler AG Germany 103.3

British American Tobacco UK 99.6

Siemens Germany 97.7

BASF Germany 93.5

SAP Germany 90.2

Vodafone UK 88

Referenties

GERELATEERDE DOCUMENTEN

Closing Material Loops at various scales Reuse of resources – materials, energy, and waste Circular City Spatial Planning & Neighbourhood Development Urban

[r]

We additionally expected the above described effects to be most pronounced in patients with epileptic focus or surgery in the speech-dominant hemisphere, since our primary

When we considered the impact of obesity on life expectancy in the 26 European countries, we found that without obesity, the increase in e 0 between 1975 and 2012 would have

The two-continua model, the buffering effects of well-being in general, and the adaptive impact of specific positive characteristics such as positive emotions (Garland et

(1.) The blue arc indicates the stator current at rated load that reflects a constant apparent power where the real and reactive power can vary according to

For women, we found that this indirect effect of gender identification was also contingent upon the gender ratio (i.e., numerical dominance) in the direct work environment. That

En materia económica, los invernaderos de alta tecnología reportan mayores rendimientos en los cultivos hortícolas, alta eficiencia en el uso de insumos, baja demanda de agua, así