Jaakko Huotari Monetary policy announcements and stock market reactions Evidence from the Finnish stock markets in 2000-2022 Vaasa 2024 School of Accounting and Finance Master’s thesis in Finance Master’s Degree Programme in Finance 2 UNIVERSITY OF VAASA School of Accounting and Finance Author: Jaakko Huotari Title of the Thesis: Monetary policy announcements and stock market reactions – Ev- idence from the Finnish stock markets in 2000-2022 Degree: Master of Science in Economics and Business Administration Programme: Master’s Degree Programme in Finance Supervisor: Sami Vähämaa Year: 2024 Sivumäärä: 64 ABSTRACT: After the global financial crisis in 2008, the monetary policy of central banks has changed dra- matically, and especially in recent years, unconventional monetary policy actions have played a more central role in the monetary policy pursued by central banks around the world. Quantita- tive easing and historically low interest rates have directed capital to riskier investments such as the stock market. This thesis examines the relationship between the European Central Bank monetary policy and the Finnish stock market, focusing especially on the short-term effects that monetary policy announcements have on the Finnish stock market. Daily stock index returns are used as a measure for reactions. The research has been conducted as an event study, which examines the effect of individual events on changes in the share price. The event window is three days around monetary policy announcements, the day before, the day of the announce- ment, and the day after the announcement. Each monetary policy announcement is treated as its own event, and for each announcement, abnormal return from what is expected is calculated. The research material needed for the study has been compiled manually by going through all the monetary policy announcements of the European Central Bank from 2000 to 2022, and the monetary policy announcements have been divided into four categories: no changes in interest rates, interest rate increases, interest rate decreases, and unconventional monetary policy ac- tions taken in the announcement. In addition to this, the data set contains 5778 daily observa- tions of the OMXHPI price index of the Helsinki Stock Exchange. Based on the data, abnormal returns are calculated for each of the three days, and cumulative abnormal returns are calcu- lated for the entire event window. The statistical significance of the abnormal returns obtained from each announcement is tested with a t-test. The statistical significance of average abnormal returns is tested over the entire data period through the examination of cumulative returns of individual days and the event window. This is done by using the average abnormal return in a t- test, and the average cumulative abnormal return is analyzed in the same way. Corresponding average abnormal returns are also calculated for market sentiments (bull and bear market) and monetary policy direction cycles (tightening and loosening monetary policy) to determine the effects of different market sentiments and monetary policy direction trends on the stock index. The findings of this thesis show that during the three-day event window, the cumulative abnor- mal returns are statistically significant, while the individual days during the event window are not found to result in statistically significant abnormal returns. Market sentiment was found to be a significant factor in how the markets react to monetary policy announcements. The direc- tion of monetary policy is also a significant factor in explaining market reactions; loosening mon- etary policy creates significant abnormal returns, while this was not found for tightening policy. Significant evidence of abnormal returns resulting from certain types of monetary policy actions during the event window was not found. KEYWORDS: central banks, monetary policy, stock prices, interest policy, announcements 3 VAASAN YLIOPISTO Laskentatoimen ja Rahoituksen akateeminen yksikkö Tekijä: Jaakko Huotari Tutkielman nimi: Monetary policy announcements and stock market reactions – Ev- idence from the Finnish stock markets in 2000-2022 Tutkinto: Kauppatieteiden maisteri Oppiaine: Rahoituksen maisteriohjelma Työn ohjaaja: Sami Vähämaa Valmistumisvuosi: 2024 Sivumäärä: 64 TIIVISTELMÄ: Vuoden 2008 globaalin finanssikriisin jälkeen keskuspankkien rahapolitiikka on muuttunut dra- maattisesti ja erityisesti viime vuosien aikana epätavanomaiset rahapoliittiset toimet ovat näy- telleet keskeisempää roolia keskuspankkien harjoittamassa rahapolitiikassa. Määrällinen elvytys ja historiallisen alhaiset korkotasot ovat ohjanneet pääomia riskialttiimpiin sijoituksiin kuten osakemarkkinoille. Tässä tutkielmassa tutkitaan Euroopan keskuspankin rahapolitiikan ja Suo- men osakemarkkinoiden suhdetta, keskittyen erityisesti lyhytaikaisiin vaikutuksiin, joita rahapo- liittisilla ilmoituksilla on Suomen osakemarkkinoille. Reaktioiden mittarina käytetään päivittäisiä osakeindeksin tuottoja. Tutkimus on toteutettu tapahtumatutkimuksena (engl. Event study) jolla tutkitaan yksittäisten tapahtumien vaikutusta osakkeiden hinnan muutoksiin. Tapahtu- maikkuna on kolme päivää rahapoliittisten ilmoitusten ympärillä, päivä ennen, ilmoituspäivä sekä ilmoituksen jälkeinen päivä. Jokainen rahapoliittinen ilmoitus käsitellään omana tapahtu- manaan ja jokaiselle ilmoitukselle lasketaan odotetusta tuotosta poikkeava tuotto. Tutkimusta varten tarvittava tutkimusaineisto on koostettu manuaalisesti läpikäymällä kaikki Euroopan Keskuspankin rahapoliittiset ilmoitukset vuosien 2000–2022 ajalta ja rahapoliittiset ilmoitukset on jaettu neljään kategoriaan: ei muutoksia koroissa, koron nosto, koron lasku ja epätavanomaiset rahapoliittiset toimet, joita ilmoituksella on ilmoitettu. Tämän lisäksi aineisto sisältää 5778 päivittäistä havaintoa Helsingin pörssin hintaindeksi OMXHPI:stä. Aineiston poh- jalta jokaiselle kolmelle päivälle lasketaan odotuksesta poikkeavat tuotot ja koko tapahtumaik- kunalle lasketaan kumulatiivinen poikkeava tuotto. Jokaisesta ilmoituksesta saatava poikkeava tuoton tilastollinen merkitsevyys testataan t-testillä. Yksittäisten päivien sekä tapahtumaikku- nan kumulatiiviset tuotot testataan myös keskimääräistä poikkeavaa tuottoa hyödyntäen siten, että koko aineiston ajalta laskettujen poikkeavien tuottojen keskiarvon tilastollinen merkit- sevyys testataan t-testillä ja keskimääräinen kumulatiivinen poikkeava tuotto testataan samoin. Vastaavat keskimääräiset poikkeavat tuotot lasketaan myös markkinasentimenteille (härkä ja karhumarkkina) sekä rahapolitiikan suunnan sykleille (tiukentava ja löysentävä rahapolitiikka) jotta eri markkinasentimenttien ja rahapolitiikan suunnan trendin vaikutuksia voidaan selvittää. Tutkimuksen tuloksista huomataan, että käsitellyn kolmen päivän aikaikkunan kumulatiiviset tuotot poikkeavat tilastollisesti merkittävästi odotetusta tuotosta, mutta yksittäisten päivien tuotot eivät poikkea merkittävästi odotuksista. Markkinasentimentin huomattiin olevan merkit- tävä tekijä siinä, miten markkinat reagoivat rahapoliittisiin ilmoituksiin. Myös rahapolitiikan suunta on merkittävä tekijä markkinareaktioiden selittäjänä; löysentävä rahapolitiikka tuottaa merkittäviä poikkeavia tuottoja, kun tiukentavan politiikan osalta tätä ei havaittu. Todisteita tie- tynlaisten rahapoliittisten toimien aiheuttamista poikkeavista tuotoista kolmen päivän aikaikku- nan aikana ei löydetty. KEYWORDS: central banks, monetary policy, stock prices, interest policy, announcements 4 Contents 1. Introduction 7 1.1 Purpose of the Study 9 1.2 Research question 10 1.3 Development of Hypotheses 11 1.4 Contribution 12 2 Theoretical background 14 2.1 Central Banks 14 2.1.1 European Central Bank 18 2.1.2 Monetary policy decisions 19 2.2 Stock markets and monetary policy 19 2.3 Market efficiency 20 2.4 Monetary policy tools 22 3 Literature review 25 4 Data and research methodology 32 4.1 OMXH PI Index 34 4.2 Methodology 35 4.3 Issues with Daily Data 37 4.4 Event study 37 4.5 Measuring the abnormal returns 39 4.6 Statistical testing 42 4.7 Testing procedure 42 5 Empirical Analysis 44 5.1 Results of all announcements from 2000 to 2022 44 5.2 The effect of different announcements 47 5.3 Investor sentiment 50 5.4 Tightening cycles and loosening cycles of monetary policy 52 6 Conclusions 54 5 7 Future research 57 References 59 6 List of Pictures Picture 1. Simplified channels for traditional monetary policy tools. ............................ 16 Picture 2. Three forms of market efficiency ................................................................... 22 Picture 3: Event study timeline ....................................................................................... 36 List of Figures Figure 1: Historical price levels of the OMXH PI Index (Nasdaq, 2023) .......................... 35 Figure 2: Distribution of cumulative abnormal returns in significance level over the event window period. ............................................................................................................... 46 List of Tables Table 1: Descriptive Statistics ......................................................................................... 33 Table 2. Average abnormal and cumulative abnormal returns of all announcements .. 44 Table 3. Different announcements and reactions. ......................................................... 47 Table 4. Event window significance, % of total announcement of the same kind. ........ 49 Table 5. Cumulative abnormal returns and abnormal returns bull & bear market ........ 51 Table 6. Tightening and loosening cycles of monetary policy ........................................ 52 7 1. Introduction Central Bank monetary policy decisions and actions have been one very topical subject in the stock markets since the financial crisis in 2008. The monetary policy we see today might seem radical to the past economists and policymakers. Large-scale security pur- chase programs and policy frameworks, which are new, remain a source of confusion for investors as well as ordinary citizens. (Bernanke, 2022) After the financial crisis, different sorts of unconventional monetary policies have moved the markets and we have wit- nessed historically low interest rates, even negative rates in some areas. According to Baumann, Lodge & Miescu (2024), the support provided by central bank policies acted a significant role in the GDP growth during the financial crisis. For a long time, the zero lower bound was seen as a limitation of conventional monetary policy, but as we have seen in recent years, this has been tackled with what we know as unconventional actions, and negative market interest and deposit rates have been the reality. In the 2020s we have witnessed massive actions to support the economy in the form of quantitative eas- ing (QE) and very recently as inflation started to rise, quantitative tightening. Covid-19, for example, caused a massive decline in economic activity, which could have led to recession, with more radical consequences than we saw. The main reason why we did not end in a more radical decline in the economy is the monetary policy which was conducted by central banks all over the world. This has been shown as an effective way of boosting the outcome of the economy and returns of stock markets by Feldkircher, Huber, and Pfarrhofer (2021). Hartley, Jimenéz, and Rebucci (2022) focused on quantita- tive easing and bond yields, finding that quantitative easing has not lost its effectiveness in the developed markets. We can say that central banks are playing a valuable role in the economy. There has been extensive examination of causality with monetary policy decisions and changes in financial markets. It is quite commonly known that unexpected changes in monetary policy have a larger impact on stock prices, as well as all the non- anticipated information. 8 While price stability is the main mandate for most central banks, the fluctuations of prices are not so easily controlled, and it requires a large toolbox so the complex dynam- ics of price development can be kept smooth. The central bank of the United States, the Federal Reserve, also serves another mandate, which is maximum employment. As this thesis focuses on the European Central Bank, the maximum employment mandate is not discussed deeply. The strategy of the ECB and Federal Reserve is mostly built on statistics that are gathered in the Euro area or the US, and published at some point in time, which is later than the actual timing of the data. This makes the monetary policy usually come with a little delay compared to the markets where prices are formed. This requires flex- ibility and forecasting from central banks and their policies. The Federal Reserve’s mon- etary policy has been described as forecast targeting (Svensson, 2020) and the ECB‘s strategy as forward-looking (Hartmann & Smets, 2018) which means the same goal; be- ing ahead to maintain low volatility and uncertainty in economies. Central banks’ announcements are closely followed and monitored by financial markets, the general public, the media, and investors (Abel, Bernanke, & Croushore 2014). Econ- omists and analysts are all the time analyzing and forecasting the movements in markets and stock prices. One thing that is frequently measured is the effect of some economic event or other kind of event on stock prices. One relevant question is, what happens to the stock prices when central banks are publishing an announcement on their monetary policy actions, in other words, how do stock markets react to this new information? An interesting aspect of central bank announcements is, that the timing of these announce- ments is usually well known in advance and therefore tried to price in as well as possible. Do the forecasts given by economists and analysts guide the market to price efficiently the upcoming actions taken by central banks? The theory suggests that only surprises will affect market prices. Common sense says that if the markets are efficient towards interest rate policies, prices should change approximately the amount that the change in discount factor presumes. I.e., the future cash flows are discounted with a new discount rate when central banks raise or decrease the interest rates. If this is not the case, then it seems that the markets 9 are reflecting also other information, like expectations, etc. which might not have fac- tual fundamentals, and inefficiencies might appear. To measure the impact of some- thing on the financial markets, economists as well as academics often employ a meth- odology known as event study. The objective of this thesis is to investigate the relationship between European Central Bank monetary policy announcements and the market returns in the Finnish stock mar- ket, the Helsinki Stock Exchange. The study aims to analyze the impact of the announce- ments on stock market returns and identify statistically significant abnormal returns as- sociated with monetary policy decision announcements around the announcement. To examine the impacts of the announcement, this study employs a traditional event study method on how the monetary policy decisions by central banks are affecting the Finnish stock market. The study is conducted to examine if the monetary policy an- nouncements are creating systematically, and on average significant abnormal returns in the Finnish stock market by the given event window, and testing the significance of the abnormal returns. 1.1 Purpose of the Study As said, this study examines monetary policy announcements by the European Central Bank, the results these announcements create in the stock market, and the significance of these reactions. The study is conducted by forming a data set with all monetary policy announcements by the ECB from 2000 to 2022, with information on actions taken in the announcement. The actions are divided into four categories; no changes in rates (if the announcement did not include a change in any of the three main rates), interest rate raise, interest rate decrease, and whether the announcement included unconventional monetary policy actions. The purpose is not to analyze the qualities of the decisions but to examine the reaction that is seen in the markets on the day before the announcement, announcement day, and day after the announcement. As the qualities of the decisions 10 are not analyzed more in-depth, these four categories are simply yes or no questions in the data set. Many studies have been conducted to measure the volatility or trading activity in the markets at the time of the announcement. Vähämaa & Äijö (2011) examined how the Federal Reserve’s monetary policy decisions affected the implied volatility of the S&P500 index. A similar approach would have been suitable for this thesis, but as the Helsinki stock exchange is a rather small market, the implied volatility for Finnish stock markets is not calculated as it is for the S&P500. This makes the defining of surprise a little bit hard, which the implied volatility would have expressed. The interest in this study is to examine the pre-event day, event day, and post-event day returns of the stock markets, not the trading volumes or volatility. With this 3-day event window, it can be tested, if the market movements are stronger around the announce- ment date or if the strongest abnormal returns are seen on the actual announcement day. This event window provides a short time horizon view on the effectiveness of mon- etary policy and market efficiency towards these announcements from the stock market perspective. 1.2 Research question The research question is the following: to what extent do European Central Bank mone- tary policy announcements impact Finnish stock markets on the announcement day and around the announcement date? The primary objective of this study is to assess the stock market efficiency, concerning the actions undertaken by European Central Bank. Each monetary policy announcement is treated as an independent event. The purpose is to assess if the different characteristics of announcement contents have different out- comes in stock markets. The main idea is to focus on the specific date when the an- nouncement is made and on the contents of the announcement itself, not analyze the possible speeches or interviews given by representatives of the European Central Bank. 11 After all, the purpose is to understand how the markets are reacting to new information that is given in the announcement release. The period, which will be used in this study includes announcements from the beginning of 2000 to the end of 2022. By taking over 20 years to the scope, the possible stronger effects of the unconventional actions we have seen lately can be eliminated. With over 20 years of observations, it is also easier to identify if the reactions differ in times of recession and strong economic environments. In the analysis, this aspect is analyzed by examining the cycles of monetary policy and the reactions seen during tightening cycles and loosening cycles. 1.3 Development of Hypotheses This subchapter discusses the hypotheses of the thesis. Based on the existing literature, discussed further in Chapter 3, and the purpose of this study the following hypotheses have been formed: H1a: The European Central Bank’s monetary policy announcements have a significant impact on the Finnish stock market on the days surrounding the announcement and an- nouncement day. H1b: The European Central Bank’s monetary policy announcements have a significant impact on the Finnish stock market on the three-day event window measured by cumu- lative abnormal returns. H2: Announcement day creates the highest abnormal returns in the event window. H3: Investor sentiment has a significant effect on how markets react to monetary policy announcements. 12 H4: The direction of the monetary policy cycle is a significant factor in the stock market reaction toward monetary policy announcements. These hypotheses help to test the relationship between Finnish stock markets and Cen- tral Bank monetary policy while considering also how the stage of the economy will af- fect the results. 1.4 Contribution This study examines the market reactions using stock market returns as a measurement. There are numerous studies conducted with volatility as a measure for the reactions. This study gives insight into how the returns will be distributed between pre-announce- ment day, announcement day, and post-announcement day in the Helsinki stock ex- change. To understand how the content of the announcement affects markets, the four different characteristics are also analyzed individually. On the other words, the cumulative aver- age abnormal returns and average abnormal returns are calculated as well for all an- nouncements with different characteristics. The findings of this study help interpret the reactions seen in the Helsinki stock exchange and examine how efficiently markets are pricing the new information. It is also examined if the abnormal returns are on average highest on the event day or the day before or after, this helps to understand how markets are anticipating the an- nouncements. The analysis of abnormal returns can also help to understand the effect of expectations and confidence about the future on the reaction seen toward the mon- etary policy announcements. As the timing of most monetary policy announcements is known in advance, this is a good way to measure how efficient the expectations of the markets, and forecasts of the markets are toward these announcements. The analysis of bull- and bear market sentiments and different directions of monetary policy gives a 13 wide perspective on the relationship and helps also to understand how efficient markets see the monetary policy actions from the financial markets perspective. 14 2 Theoretical background This section discusses the theoretical background related to the topic of this study and the basic causalities between the stock market and central bank monetary policy. The first chapter discusses shortly the history and development of central banking and the main tasks of central banks in general. The functions of the European Central Bank will be discussed, as it is at the center of this thesis. 2.1 Central Banks The early stage of central banking reaches as far back as the seventeenth century, i.e. to the 1600s. The first institution that could be described as a central bank is the Swedish Riksbank, founded in 1668. Some years later the Bank of England was established. What was in common with these first central banks, was the purpose which they were founded for; to purchase government debt. Later, other central banks were founded also to deal with the disorder in the context of monetary. The earliest central banks issued notes to serve as currency on which they often had a monopoly. The major change, when it comes to the goals of central banks and their focus areas, happened after the First World War. The interest of central banks started to move towards employment and price levels. (Bordo, 2007) The actions of central banks and the central banks in general, can still nowadays be a little blurred, although most central banks do argue in favor of transparent central bank- ing. The business of central banks is technical, and it requires a very high skill set to do successfully, which leads to situation where the actions cannot be easily described or explained, and understanding the actions requires a prerequisite about how the econ- omy works and what is the purpose that central bankers try to fulfill. (Singleton, 2011) As it is seen, central banks have a long history, and they are playing a crucial role in the economy. In today’s economy, they are a public institution that has the power to control 15 the money supply, meaning the actual amount of money in circulation. Some countries have their own central banks, for example, the Bank of England or the Sveriges Riksbank, and some areas, like the euro area have their own central bank, which is the responsible authority for the euro area. Setting the interest rates is one of the main tools central bankers have. Interest rates are the ‘’cost of money’’ and by setting the interest rates, central banks can make an impact on the market rates, by which commercial banks loan money forward to individuals, companies, etc. A central bank is therefore not a commer- cial bank, and individual people cannot open accounts in central banks. (ECB, 2015) One of the most important outcomes of the Central Bank monetary policy is the effect seen in the market interest rates. When central banks raise rates, they control the risk-free rate level, which is crucial in the valuation of different asset classes. Therefore, the role is instrumental also from the perspective of an equity investor and the stock market in general. The following picture expresses how central banks can lend money and who can take loans from central banks: 16 Picture 1. Simplified channels for traditional monetary policy tools. The role of the European Central Bank in the European economy is stated clearly on the front page of their website: ‘’We work to keep prices stable in the euro area’’ (ECB, 2023). This means keeping inflation under control, contributing to keep safety in the banking system, and keeping financial infrastructure running smoothly. The price stability being the priority, the ECB uses monetary policy as a tool to impact inflation and possibly heating economy. In recent years the slowing economy has also been something that ECB has backed. The central bank makes decisions on monetary policy every six weeks, determining what should be done to meet the 2% annual inflation target. The meetings are followed by a press conference where the decisions are dis- cussed by the president of the ECB. The monetary policy decisions are usually included with information on changes in interest rates and possible unconventional actions. Lately, we have seen a turn in the inflation rates as inflation has been the highest in 30 years in the euro area. European Central Bank has reacted strongly to the increasing Central Bank All Banks Other financial institutions, finance companies, insurance offices, credit unions 17 prices by increasing the interest rates. This has been a major driver in the markets since early 2022. This is a textbook example of the actions that will be conducted to keep prices stable. Monetary policy is generally controlling and changing the amount of domestic currency in circulation to influence nominal exchange, interest rates, and domestic liquidity con- ditions. Movements in interest rates and exchange rates are affecting the behavior of macroeconomic variables widely. There have been disagreements in the past on how monetary policy should be conducted. Today, controlling inflation is seen as the foremost end goal of monetary policy. (Makin, 2002, p. 97) Nowadays European Central Bank has set the target level at 2%, as it can be achieved most efficiently. In recent years, as there has been a global pandemic situation which interrupted the normal economic behavior of individuals, central banks have employed unconventional monetary policies to sup- port the economy. Three employed unconventional policies by ECB after the financial crisis are the OMT program (Outright Monetary Transactions), negative policy rates, and corporate bond purchases. (Ferrando, Popov & Udell. 2022) Central banks have also started to give more forward guidance on the future policy directions which is consid- ered as unconventional monetary policy action. When the European Central Bank is pursuing to fulfill its mandate to keep the prices stable, they do this by controlling the three key interest rates. The rates are main refi- nancing operations (MRO), which is providing bulk liquidity into the banking system, de- posit facility rate which banks may use to make overnight deposits in the Eurosystem and rate of marginal lending, which is offering overnight credit to banks from the Eu- rosystem (ECB, 2023). When the ECB is changing interest rates, it will affect to market rates, and this will lead to changes in the discount factors and therefore valuations of companies. This has straight impact on stock markets as willingness to take risk might change when attractive returns are within reach with lower risk level. The valuation method is discussed later to explain the relationship between rate changes and valua- tions. 18 2.1.1 European Central Bank European Central Bank is the central bank for the Euro area. The first steps in forming ECB were taken in 1988 when it was decided to build an Economic and Monetary Union. The reasons behind the decision can be seen today in the Euro area as there is free cap- ital movements and a common monetary authority. Operating of ECB began in 1998 when it started to operate as a custodian with a mandate to maintain price stability. (ECB, 2023) In Finland euro was taken as a currency in 1999 and at that time, the possibility to de- valuate the former Finnish currency ‘’markka’’ started to become history. The European Central Bank’s monetary policy has been at the core of the Finnish economy since then. In recent years, the role of the European Central Bank has become more and more im- portant, also in the security markets. This has also changed the strategy of the European Central Bank. Markets are waiting for central banks to take actions when things are start- ing to look bearish. Until March 2022, we lived for years in an environment where ECB target rates were close to zero and Euribor 12 months was negative. The negative inter- est rate environment started approximately in the first quarter of 2016. Before that, the existence of negative interest rates was argued in a paper by Jarrow, R. (2013) where Jarrow stated that negative interest rates are possible, and the zero lower bound does not exist unlike it was believed before, and on what the unconventional monetary policy tools were previously based on. The mandate of central banks is needed to pursue, no matter what the macroeconomic circumstances are. This includes the fact, that central banks need to prepare for turbu- lent periods and times of instability that may arise from their actions. In times of financial crisis back in 2007-2009, according to Mojon (2010), the ECB had very little it could have done to prevent the financial crisis, which was caused by the fragility of the US economy. As a response to the financial crisis European Central Bank had three major steps it took: 19 Distribution of liquidity to the banking sector, lowering interest rates, and 12-month full- allotment refinancing operations. After the financial crisis, the role of the ECB has grown significantly, or at least the actions they have done have had a larger and more straightforward impact on security markets. This has been conducted in the form of quantitative easing, with operations called asset purchase programs. These operations are in nature buying securities from the markets, with the main focus on bonds. These asset purchasing programs have continued from the aftermath of the financial crisis and the Covid-19 crisis until this day in some form. 2.1.2 Monetary policy decisions The monetary policy in the euro area is formulated by the Governing Council of the Eu- ropean Central Bank. The formulation includes decisions related to key interest rates, supply of reserves in the Euro system, monetary objectives, and establishment of guide- lines for the implementation of mentioned decisions. They also have the responsibility to adopt the guidelines and make decisions to ensure the performance of the tasks en- trusted. The council consists of six members of the Executive Board and governors of the national central banks of the euro area countries. Meetings of the Governing Council are usually set up twice a month in Frankfurt am Main in Germany at ECB’s premises. The economic and monetary developments are assessed, and monetary policy decisions are done every six weeks. When the council has other meetings, they discuss issues related to other responsibilities of the European Central Bank and the Euro system. The deci- sions for monetary policy and other responsibilities are held in different meetings to en- sure the separation of decisions. (ECB, 2023) 2.2 Stock markets and monetary policy One of the key issues and questions in existing literature is how monetary policy deci- sions move the stock markets, and it is often associated with large movements in prices. 20 The value of an individual company is often calculated as the value of its discounted cash flows. This method is one of the most used valuation methods and it is linked to the market rates heavily as the market rates usually are in contact with the discount factor. The discounted cash flow model is expressed as follows: 𝐷𝐶𝐹 = 𝐶𝐹1 (1+𝑟)1 + 𝐶𝐹2 (1+𝑟)2 + 𝐶𝐹𝑛 (1+𝑟)𝑛 (1) Where: 𝐶𝐹1 = Cash flow, year one 𝐶𝐹2= Cash flow, year two 𝐶𝐹𝑛= The cash flow for additional years, until year n 𝑟 = The discount rate. As can be seen, the changes in discount rate have a direct impact to the value of the future cash flows. The further in the future we go, the stronger the impact is. This said the characteristics of an individual company will affect how strong the impact of the rate change is on the valuation done with the DCF model. In growth-stage companies, cash flows are usually further in the future and so-called value companies might generate cash flows constantly and therefore changes in discount factor has a higher impact on the prices of growth companies than value companies. 2.3 Market efficiency When assessing the impact of an event on stock markets, one key theoretical aspect is the efficiency of pricing the new information to prices. According to Fama (1970), stock prices should always reflect and incorporate all the relevant information. This means 21 that all the prices should always be correct, and there is no mispricing in the markets, assuming of course that markets are efficient. New relevant information should there- fore be in the prices immediately and this is at the center of this thesis. This is known as the efficient market hypothesis. When central banks are pursuing to fulfill their responsibility to keep the prices stable, they do this by controlling the three key interest rates. The rates are main refinancing operations (MRO), which provides bulk liquidity into the banking system, deposit facility rate which banks may use to make overnight deposits in the Euro system and rate of marginal lending, which offers overnight credit to banks from Euro system (ECB, 2023). As they are changing interest rates that will affect market rates, this will lead to changes in the discount factors and therefore valuations of companies. This has a direct impact on stock markets as the risk willing might change when returns are available with lower risk. Back in the first paper on efficient capital markets, Fama (1970) divided market efficiency into three forms: weak, semi-strong, and strong form. The weak form is the least rigorous of the three forms. When the weak form is found in the markets, the prices are reflecting the information about the historical prices and therefore move randomly. In a weak form of efficiency, excess returns cannot be earned by following investing strategies that are based on historical prices. Semi-strong form of efficiency implies that assets are reflecting all public information available. This means, that investors with non-public information, known as inside infor- mation, may have an advantage. Anomalies in pricing are found quickly. The strong form of market efficiency means a state in the markets, where prices are re- flecting all information, both public and insider information. This is basically a situation where no additional value-bringing data is available. 22 Picture 2. Three forms of market efficiency 2.4 Monetary policy tools When discussing central banks, it is important to understand the different interest rates that are the traditional tools of monetary policy. Nowadays understanding also the other monetary policy tools is important. In this subchapter, the main interest rates, which the European Central Bank uses, are shortly discussed as well as the other tools. Marginal lending facility Marginal lending facility is the interest rate, that is used when banks borrow from the European Central Bank overnight. To be able to borrow, banks must provide collateral to ECB. Collateral can be for example securities. As it is usual in banking, these collaterals Strong Form: All public and Private information Semi-strong Form: All public information Weak Form: Historical prices 23 are given to guarantee the payback for ECB. (ECB, 2018) This is the rate, which is used to provide overnight credit for banks in the Euro system. Main refinancing operations rate Also known as MRO, is the rate that is used when banks borrow from the European Cen- tral Bank for one week. As well as a marginal lending facility, also for MRO collateral is needed. (ECB, 2018) This is the rate, which is used to provide bulk liquidity to the banking system. Deposit Facility Rate The two interest rates above are rates for banks when they borrow from the central bank. The deposit facility rate is the interest rate when banks make deposits to the central bank. Deposits are overnight deposits. (ECB, 2022) This is the rate for banks to make overnight deposits in the European Central Bank. These are the three main interest rates, that the European Central Bank sets every six weeks to fulfill their mandate of stable prices in the Euro area. In the previous years, as is generally known, there have been negative interest rates. Other monetary policy tools, which are considered mostly unconventional policy tools include the following themes: - Offering central bank loans for banks as they face the need for it. These loans are given against collateral at fixed interest rates. - Negative interest rates - This has been in use for several years in the Euro area and is used to encourage banks to lend with lower rates, so businesses and peo- ple can borrow with lower interest expense, in other words, cheaper. 24 - Targeted longer-term refinancing operations (TLTROs). This means long-term lending for banks with the condition that banks will lend this money to busi- nesses and people. - Asset purchases – This gives the central bank the possibility to buy private and public financial assets. - Forward guidance – Clear communication on the central bank’s intentions for fu- ture monetary policy. (ECB, 2024) As seen from the possible tools central banks can use, they have a significant ability to impact economic activity, in good times as well as in bad times. This gives the central bank a massive responsibility to keep up its credibility, so markets will function normally even when things start to look downwards and not rely too much on central bank inter- ventions. 25 3 Literature review In this chapter the focus is on the previous studies on central bank’s monetary policy, the relationship of monetary policy and stock markets, and the theories relevant for this the- sis. Methodology and previous literature related to the methods of this thesis is dis- cussed later in Chapter 4. Previous research gives an essential wide perspective on the topic and on the findings of existing literature. Previous literature in this context has found that monetary policy may have a significant impact on the stock market’s returns and volatility. Many of the studies in the field were conducted before the financial crises that we have seen lately. After the financial crisis in 2008, the role of the European Cen- tral Bank has grown significantly, as unconventional tools have become an important tool for the central bank in the Euro area. When looking back in time and the research conducted in the context of monetary policy and stock markets, we can say there has been a large amount of research done. Money supply announcements have been in the interest of researchers long before the Euro- pean Central Bank was founded. Pearce and Roley (1983) found in their paper, where they examined stock reactions to weekly money supply announcements, that stock prices respond only to unanticipated changes in money supply. This is in line with the efficient market hypothesis, and it can be assumed that markets should only have sig- nificant abnormal reactions if the announcement includes unanticipated information. Later Pearce & Roley (1985) tested efficient market hypothesis by identifying unex- pected components of the announcements and stated that surprises move stock prices. They also found that there is only weak evidence that stocks are responding to surprises beyond the announcement day. This is one aspect of this thesis, as the abnormal returns are observed also from the day after the announcement day. This is tested in paper by Thoerbecke (1997) where the findings state monetary policy exerting large effects on ex-ante and ex-post stock returns, and the effects are in line with the DCF model pre- sented earlier, where we saw that further in the future incoming cash flows are affected more heavily by the changes in discount factor. 26 Thoerbecke had results where small firms were more effected by monetary policy than large firms. Thoerbecke’s findings also explains the greater impact on growth companies than value companies we saw in 2022, when interest rates started to rise. Monetary policy also affects firms’ ability to get credit. In the previous years, credit ability of all companies has been quite good, even for smaller and non-profitable companies as in- terest expenses have been so low. After the rapid rise of interest rates, the effects have also been strongest in small-cap companies. One of the first significant papers by Bernanke & Blinder (1992) studied the informative- ness of the Federal fund rate of monetary policy actions. This was found to be a very informative indicator. Stock market reactions to the policy of the Federal Reserve are discussed in a paper by Bernanke and Kuttner (2005). They found that on average, a hypothetical unanticipated 25-basis-point decrease in the Federal Funds rate target is increasing 1% in stock indexes. The close relationship between market interest rates and central bank interest rates has been shown. Vance and Sellon Jr (1995) examined the relationship between central bank monetary policy actions and long-term interest. They found that long-term interest rates seem to move well in advance of policy actions. According to Ellingsen & Söderström (1998), if monetary policy reveals information about the development of the economy, interest rates are moving in the same direction in all maturities to policy innovation. On the other hand, if monetary policy is revealing information about the central bank pref- erences of policy, short rates and long rates move in opposite directions. These studies show a clear relationship between market rates and central bank monetary actions. As we know market rates are crucial in determining the value of a company, determining the risk-free rate often used for example in the capital asset pricing model. Vance and Sellon also made clear points that expectations play a crucial role in the response of long-term rates to monetary actions. 27 The announcements from central banks have been proven to have a significant effect on investor sentiment, in both bear market and bull market (Kurov 2010). Kurov’s findings also show that monetary policy, in times of bear market periods, tends to have a larger effect on stocks that are more sensitive to changes in investor sentiment. This study will test the reactions in different market sentiments, to test whether reactions differ de- pending on the investor sentiment, in other words, if the reaction differs when markets are bullish or if markets are considered to be bearish. Another paper by Kurov discussing the relationship between the stock market and monetary policy was done in 2012. He found that during economic growth cycles, stocks tend to react negatively to announce- ments of higher rates ahead, and in recessions, stocks have positive reactions to signals of monetary tightening. When deciding, whether some new information have been anticipated or can be consid- ered as a surprise, the central bank’s announcement schedule is known in advance and therefore forecasted carefully. The surprise effect of a monetary policy announcement has been researched extensively in the past. Nikkinen & Sahlström (2004) found that the S&P 100 index implied volatility is increasing before a scheduled announcement of mac- roeconomic factors, for example, a Federal Reserve Open Market Committee meeting (FOMC), and after the meeting volatility is decreasing. Bekaert et al (2013) studied the risk aversion of markets and monetary policy. They found that loosening or lax monetary policy decreases risk aversion and uncertainty. This should support the stock market, as the investors are not afraid of the additional risk as much. Ricci (2015) studied the impact of monetary policy announcements on large European banks during the financial crisis. She found that banks tend to be more sensitive towards unconventional policies than interest rate decisions. She also states that the same type of announcement, or action may result in different kinds of impact in the markets, 28 depending on the stage of the crisis. Weaker balance-sheet banks and higher-risk oper- ating banks were more sensitive to monetary policy. According to Jarocinski & Karadi (2018) surprise monetary policy tightening raises inter- est rates while it lowers stock prices. The opposite monetary policy information, loosen- ing, on the other hand, raises both interest rates and stock prices. This statement lines up with the basic DCF model presented in Chapter 2.2; as the risk-free rate decreases, the discount factor is lower. On the other hand, when policy is loosening the market interest rates lower which encourages more risk-taking by investors and therefore sup- ports stock markets. Their findings follow the findings of Bekaert et al (2013). Both, pol- icy loosening and tightening are discussed later in empirical analysis, and CAARs and AARs are calculated to test the effect of policy directions. The quantitative easing, as a method as well as a tool, was clearly described by Claeys, Leandro & Mandra (2015). They showed how the quantitative easing programs are de- signed to work by describing the process of how the Central Bank is easing the economy. The simplest way to describe the QE is through a monthly purchasing program where the central bank will buy corporate bonds, government bonds, and supranational bonds from the markets and add them to the central bank’s balance sheet. This describes easily what is the basics of the tool being used. In this study and its dataset, the unconventional policies are mostly QE programs that are either begun or continued or some kind of in- structions on how the program will be done in the future is given. Fratzscher et al (2016) examined the impact of the ECB’s unconventional actions during and right after the financial crisis. They tested a large range of asset prices and portfolios to see if ECB policies have an effect on asset prices. They found that the European Central Bank’s monetary policy boosted equity prices. They also found a spillover effect on de- veloped markets and emerging markets, as the policy of the ECB had a positive impact on equity markets and market confidence. They found also a relationship between ECB 29 policy and lowering credit risk among banks in the euro area as well as in other G20 countries. Bernanke (2020) talks about overcoming the limits that traditional monetary policy has. He focuses on quantitative easing and forward guidance, which have been the new tools in use by the ECB, Federal Reserve, and other advanced economies central banks. He argues in favor of these new tools, as they have proven to be efficient at easing financial conditions. He also argues that these new tools should become part of the standard toolkit of central banks. These tools are supporting, or at least stabilizing the stock mar- ket in default, as forward guidance reduces uncertainty and quantitative easing pro- grams reallocate money into the stock markets. The unconventional policy by the European Central Bank was discussed by Zabala & Prats (2020) when they tested the effectiveness of unconventional monetary policy on infla- tion and the growth of the economy. They found a slight impact of balance sheet policies on the inflation rate and growth rate of the economy in the Eurozone. These findings support the effectiveness of monetary policy adopted by the European Central Bank to its mandate of price stability. When discussing central banks, the financial and economic disruptions especially in the last couple of years been at the center of the discussion. The COVID-19 pandemic caused the latest significant and fast drop in stock prices and economic activity that required central banks to take quick actions to support the economy. Deng, Xu & Lee (2022) dis- cussed COVID-19-related policies and found that for stock markets, rate cuts were signif- icant actions to support stock markets. The findings align with Jarocinski & Karadi (2018), but in a more specific sample period, as stock prices increased when rates were cut. This although does not apply in the euro area, as the rates were already at zero when the pandemic burst. 30 Blampied & Mahadeo (2023) discussed about uncertainties of monetary policy and stock market uncertainties by testing responses to tightening or easing shocks. They found evidence of monetary policy being a potentially effective tool to reduce uncertainties, through easing policies during recessions and tightening policies during expansion peri- ods. This is discussed later in this thesis when different cycles of monetary policy, tight- ening cycles, and loosening cycles, are tested. Bekaert et. al (2023) discussed risk, monetary policy, and asset prices in the context of the global world. The study highlights the importance of central bank policies and the relationship it has with the stock markets. They found that monetary policy has strong domestic effects on stock market prices in both, the US, and Euro area. There are also international spillover effects in US monetary policy to international stock markets, which is somewhat expected given the importance of the US economy and stock markets globally. An interesting finding is the spillover effect in Europe being economically stronger than the US. This is interesting given the relatively small size of European stock markets. Given this, it can be said that the euro area’s monetary policy is more significant on stock markets in relative terms. The impact of conventional and unconventional monetary policy actions by the Euro- pean Central Bank was recently examined in the context of Swiss stock markets by Fausch & Sutter (2024). They tested the impact of both monetary policy types on the Swiss stock markets and found that the Swiss equity market returns responded significantly to both conventional and unconventional monetary policy. They found that conventional actions only had an impact before the financial crisis. As seen in the previous literature, the relationship between central bank policies and the stock market exists and is significant. This study tries to explain the short-term relation- ship between the Finnish stock market and the European Central Bank. The expectation based on previous literature is that the stock market should only result in significant 31 abnormal results if the announcement included a surprise element that was not already anticipated. From the literature review, the findings of Kurov (2010), Jarocinski & Karadi (2018), and Bernanke (2020) are especially in touch with this thesis, as their findings are in touch with investor sentiment, interest rate changes and the unconventional monetary policies respectively, and which are in the core of this thesis hypothesis testing. The main previ- ous studies related to the methods are discussed later in chapter 4. 32 4 Data and research methodology This chapter discusses the data and methodology used to conduct this thesis. The nec- essary data for this thesis is easy to find, as the data includes the daily index points from the Helsinki stock exchange and the information from ECB monetary policy releases. The index in this study is the Helsinki Stock Exchange Price Index (OMXHPI). With this data the necessary calculations for returns, standard deviations, abnormal returns, expected normal returns and cumulative abnormal returns can be calculated. Data for the returns is collected from the Nasdaq Nordic website and the European Central Bank release ar- chive. The data used in the empirical analysis consists of the daily returns from the Finnish stock market main index, the dates of ECB monetary policy announcements, and the actions stated in the announcement. This information is freely available on the central bank’s websites. The longer the period in which the research is conducted, the more there needs to go through the historical releases of monetary policy announcements. For this study, all the announcements are gone through manually and the data set is formed by hand to see what the actions have been in the announcement. Since the purpose is not to study directly what the actions have been, i.e. what the an- nouncements said precisely, the actions are divided into four categories: no changes in rates, interest rate raise, interest rate decline, and information of unconventional mon- etary policy actions (yes or no). The unconventional actions, as well as the size of the interest rate changes, are not analyzed deeply, or considered in the analysis. It is im- portant to note that the unconventional actions in the announcements are not exactly new actions conducted by the central bank but rather if the announcement included for example some sort of guidance on asset purchase programs or other unconventional actions, it is noted as unconventional actions in the data set. If the announcement included information that had an impact on the ongoing uncon- ventional actions, for example, change in the amount of asset purchase program or 33 change in the length of it, it is noted as unconventional actions taken. The purpose is to see what the market reaction is when the announcement is given and to recognize the difference in reaction between different policies. The sample period in the study spans from 2.1.2000 to 30.12.2022. During the sample period European Central Bank has given 268 monetary policy announcements in total, but which three were given on Finland’s Independence Day when markets are closed in Finland, and for this reason are left out of the analysis. Rates were changed 43 times and unconventional actions were included in the total of 63 announcements. Rates remain unchanged in 222 announcements. The sample period includes 5 778 daily observations of the Helsinki stock exchange price index. The returns have been calculated as daily logarithmic returns. For the assessment of the significance of each announcement, there is an expected return needed. The expected returns and how they are calculated are explained in detail later. Descriptive statistics for the data set and real returns of the event windows are repre- sented below in Table 1. Table 1: Descriptive Statistics No. Of Obs Mean Median St.Dev Min Max ECB announcements 265 Daily obs. of OMXH_PI 5778 No changes in rates 222 Unconventional actions 63 Interest rate raise 20 Interest rate decline 23 Prior event day returns -0,0459 0,0603 1,8700 -7,8606 5,8082 Event day returns 0,1427 0,0784 2,4770 -10,7876 14,5631 Post-event day returns -0,0066 0,0975 1,8838 -5,4717 5,0251 34 Table 1 gives an overall summary of the statistics regarding the data set. In the table, the event window daily returns are real returns. As we see from the table, In the event win- dow, on average daily returns are positive only on the event day. 4.1 OMXH PI Index The index used in this study is the Helsinki Stock Exchange price index, known as OMX Helsinki PI. PI means price index. This index tracks the stock performance of all listed companies in the Helsinki stock exchange, without including the effect of dividends. It includes all companies in different business areas and different sizes. Helsinki stock ex- change has also a few other indices: OMXH25 & OMXH15 which follows 25 and 15 most traded stocks respectively, and OMXH GI which includes the effect of dividends in the development. For smaller companies, First North Finland PI provides the possibility to become publicly traded, even with a smaller market capitalization. To assess the effect on prices, and measure the effect of announcements, the price index is most suitable. It is notable that in the early 2000s one company, Nokia, had a signifi- cantly large weight on the index. This said some of the movements can be highly corre- lated with the sensitivity of one company to monetary policy. 35 Figure 1: Historical price levels of the OMXH PI Index (Nasdaq, 2023) 4.2 Methodology This thesis is built on a quantitative approach to examine whether the announcements have a significant effect on the Finnish stock index. The method in this thesis follows the method described by Brown &Warner (1985). With this approach, all the events can be handled as individual events, and measure the market level abnormal returns of each event. This can be achieved by statistically comparing the difference between pre-, event day, and post-event day returns and cumulative returns with the estimation period re- turn (expected return). If there are large abnormal returns, which are statistically signif- icant, the null hypothesis can be rejected and the market has seen the event, i.e. the announcement important or containing non-anticipated information. 0 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 1 .3 .0 0 1 .3 .0 1 1 .3 .0 2 1 .3 .0 3 1 .3 .0 4 1 .3 .0 5 1 .3 .0 6 1 .3 .0 7 1 .3 .0 8 1 .3 .0 9 1 .3 .1 0 1 .3 .1 1 1 .3 .1 2 1 .3 .1 3 1 .3 .1 4 1 .3 .1 5 1 .3 .1 6 1 .3 .1 7 1 .3 .1 8 1 .3 .1 9 1 .3 .2 0 1 .3 .2 1 1 .3 .2 2 Historical development of OMXH PI Index 36 Picture 3: Event study timeline Where: t= trading days relative to event day To extract monetary policy announcements’ effect on the stock index, I use a very tradi- tional abnormal return approach. For the event study, I use a 3-day event window in- cluding the day before the announcement, the announcement day, and the day after the announcement. This approach lets us measure how the markets are reacting on the an- nouncement day but also to see if there is delay or anticipation towards the announce- ment in the market. This will also link to the efficient market hypothesis, which assumes that all relevant information is priced in. To calculate abnormal returns, the expected returns need to be determined. There are different methods to do this. In this study, we have observations of a market index, and therefore applying some traditional expected return model such as the capital asset pric- ing model is not a good choice for determining the expected return. The expected return for the index is calculated as an average return from the index for a specific time. As we are going through many events, which some might be very close to each other, deter- mining the estimation window is crucial. As the efficient market hypothesis assumes, new information is priced in markets effi- ciently and rapidly, we can assume that market movements will be concentrated around the event. The daily data might give us some noise in the market, which we want to 37 reduce as much as possible. To do that the estimation period is set to be 5 days and the estimation period is set to be after every event so the new information given in the an- nouncement will be known in the markets on the estimation period. The expected return is discussed later more carefully. This kind of abnormal return approach is widely used in the literature. This approach gives insight which is examined by looking at every an- nouncement individually. 4.3 Issues with Daily Data When dealing with daily data, there are a couple of important issues that need to be considered. According to Brown and Warner (1985), using daily data might have the fol- lowing problems: Non-normality issues Daily observations of stock returns might differ substantially from normality. There is evidence of daily observations being fat-tailed relative to normal distribution. This is not similarly observed when using monthly data. Estimation of variance Estimation of the variance for mean excess return is an important for testing statistical significance. Brown and Warner discuss about an issue that is shown by Beaver (1968) and Patell & Wolfson (1979), they showed that variances of returns are increasing in the days around the events. 4.4 Event study Event study is an important methodology in finance and economics. It has become an important tool for testing market efficiency after Fama, Fisher, Jensen & Roll (1969) pub- lished their somewhat seminal paper. They investigated stock price adjustments to new 38 information, for example, earnings announcements or some macroeconomic indicators. The theory of efficient markets is most famously stated by Eugene Fama (1970). After these papers, there has been conducted many relevant papers about event studies and using event study methodology. Brown and Warner (1985) wrote a paper about us- ing daily stock returns in event studies and Campbell and Wasley (1993) discussed the issues of using daily NASDAQ security returns in event studies. Both studies have simi- larities in the data set and in this thesis, daily data from the Finnish stock market is being used. In 1997 Campbell, Lo & MacKinlay stated modifications for the basic methodology, one of them being using daily return data instead of monthly data. They also stated that the methods to estimate the abnormal return and their significance have become more sophisticated. MacKinlay (1997) did a study on Event studies, where he discussed the usefulness of the method in the context of finance and economics. The study is a good starting point when looking at the basics of the method and how to conduct the study. Some important mod- els are gone through in the study. Kothari and Warner (2007) stated in their report that between 1974 and 2000, major journals which they listed being five, have published over 560 articles containing event study results. This said as those are the ones that were published in the five major jour- nals, the number of published papers might be significantly higher if we consider all the journals in the context of finance and accounting. Some widely known purposes for the event study method have been published in papers that have investigated the effects of terrorist attacks on stock prices (Chen & Siems. 2004), stock price reaction to environmental performance (Yamaguchi. 2008), Aggarwal, Akhigbe, and Mohanty (2012) discussed on the effect of oil price shock to asset prices. 39 In this study, all the announcements are divided into their own events. From every event, the abnormal returns will be calculated for three days, as well as the cumulative abnor- mal returns and standard deviation of the abnormal returns. These derive one test sta- tistic, t-statistic which is employed to examine if the announcement has statistically sig- nificant impact on the three-day event period. Identification of the correct event date is essential in event studies. If the date for an- nouncement is known precisely, one day event period that is including the announce- ment date only is usually the best choice. However, in practice, it might not be possible to know the exact time when the new information is reaching the investors. This causes a problem estimating the event window; if it is too short, it might not include the time when investors truly learn the event, and if it is too long there might be also other factors affecting the price, rather than just the event of interest. 4.5 Measuring the abnormal returns The measurement of abnormal returns requires the measure of expected normal return. The normal return in this study is measured as an average return of 5 days, ten days before the event window begins. This methodology is known as mean adjusted return. As there have been changes in the pace of monetary policy announcements in the esti- mation period 2000-2022, especially in the early 2000s the expected returns are mostly estimated from 5 days average, beginning right after event window have closed. The logarithmic daily returns are calculated as follows: 𝐷𝑎𝑖𝑙𝑦 𝑟𝑒𝑡𝑢𝑟𝑛 𝑅𝑖𝑡 = 100 ∗ 𝑙𝑜𝑔 𝑃𝑡+1 𝑃𝑡 (2) Where: 𝑅𝑖𝑡 = the daily return of the index 𝑖 at time 𝑡 40 𝑃𝑡= the value of the index at time 𝑡 To be able to assess the event’s impact, a measure of abnormal returns for index 𝑖 is calculated as follows: 𝐴𝑅𝑖𝑡 = 𝑅𝑖𝑡 − 𝐸(𝑅𝑖𝑡) (3) Where: 𝐴𝑅𝑖𝑡 = Abnormal return 𝑅𝑖𝑡= The actual return of index 𝑖 in time 𝑡 𝐸 (𝑅𝑖𝑡) = The average return of the share index in time t In this thesis, the expected return 𝐸 (𝑅𝑖𝑡) is calculated as a mean of the index returns in time period of five days, 10 days before the event window begins. As the events might be very close to each other, in cases where the estimation period would overlap with the previous event window, the expected returns are calculated from the five-day period starting right after the event window. By doing this, we can estimate the expected re- turns with the previous event priced in the expected return. Without this, the expected returns would include possibly larger impacts caused by the announcement given in the previous event window. It is also wanted to have a break between the estimation win- dow and the next event window, so the possible anticipation effect of upcoming events can be reduced as much as possible. This is important, so the expected return is not affected by the previous event. With the short estimation period, the dynamic changes over time can be captured in the expected returns and when markets are more volatile, it is also captured in the expected return. Mathematically the expected return is expressed as follows: 𝑃𝑟𝑒 𝑒𝑣𝑒𝑛𝑡 𝑒𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑟𝑒𝑡𝑢𝑟𝑛 𝑅?̅? = 1 5 ∑ 𝑅𝑖𝑡 −11 𝑡=−15 (4) 41 The date of the actual event is t=+1 and the event window start one day before; the pre- event day is t=0. The model to calculate the expected return is known as mean adjusted return model. The expected return is calculated over 5 days from t=-15 to t=-11. The abnormal returns are calculated for the event window, which includes 3 days. The cumulative abnormal returns are the measure of the abnormal returns of the whole event window and is calculated as follows: 𝐶𝐴𝑅𝑖𝑡 = ∑ 𝐴𝑅𝑖𝑡 𝑡+1 𝑛=𝑡−1 (5) Where: 𝐶𝐴𝑅𝑖𝑡 = sum of abnormal returns of index 𝑖 in time 𝑡 𝐴𝑅𝑖𝑡 = Abnormal return of index 𝑖 at time 𝑡 In the formula t-1 and t+1 denote the beginning and ending of the event window, with respect to the event day. The average abnormal returns, AARs are calculated as follows: 𝐴𝐴𝑅𝑡 = 1 𝑁 ∑ 𝐴𝑅𝑖𝑡 (6) 𝑁 𝑖=1 Where: 𝑁= presents the number of abnormal returns in the sample 𝐴𝑅𝑖𝑡= abnormal returns of index 𝑖 in time 𝑡 And Cumulative average abnormal returns are calculated as follows: 42 𝐶𝐴𝐴𝑅 = 1 𝑁 ∑ 𝐶𝐴𝑅𝑖𝑡 𝑁 𝑖=1 (7) Where: 𝐶𝐴𝑅𝑖𝑡= Cumulative abnormal returns of the event windows. 𝑁= presents the number of cumulative abnormal returns in the sample 4.6 Statistical testing When testing the statistical significance of abnormal returns, there are different tests for evaluating the statistical significance. The most widely employed tests are probably t- tests, z-tests, and F-tests. These have been derived by Patell (1976), and Boehmer, Musu- meci, and Poulsen (1991). In 1976 Patell proposed a test statistic, where the event win- dow’s abnormal returns are standardized by the standard deviation of the specific esti- mate window. 4.7 Testing procedure To test the significance of the abnormal returns, in this thesis t-test is applied. The t-test is one of the most suitable procedures for testing the significance. The mathematical formula which is applied in this data set to get the t value is presented below. 𝑡 = 𝐶𝑢𝑚𝑢𝑙𝑎𝑡𝑖𝑣𝑒 𝐴𝑏𝑛𝑜𝑟𝑚𝑎𝑙 𝑅𝑒𝑡𝑢𝑟𝑛𝑠 (𝐶𝐴𝑅) 𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝐷𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛 𝑜𝑓 𝐶𝑢𝑚𝑢𝑙𝑎𝑡𝑖𝑣𝑒 𝐴𝑏𝑛𝑜𝑟𝑚𝑎𝑙 𝑅𝑒𝑡𝑢𝑟𝑛𝑠 𝑜𝑓 𝑡ℎ𝑒 𝑒𝑣𝑒𝑛𝑡 The t-statistic is employed to test the null hypothesis of abnormal returns and cumula- tive abnormal returns: 43 For the average abnormal returns (AARs) and cumulative average abnormal returns (CAARs), the t-test is employed. For the mean returns, the t-test is calculated as follows: 𝑡 = ?̅? 𝑠/√𝑁 (8) Where: ?̅? = the mean of ARs and CARs s = standard deviation of the observations N = number of observations (no. of abnormal returns and events) 44 5 Empirical Analysis The following chapter goes through the empirical analysis and findings of the study. First part discusses the overall results of all the events and are presented in Table 2. Second subchapter discusses the effect of different announcement types, and the third subchap- ter tests the effects of investor sentiment in respect of the market reactions. The last subchapter discusses how the different cycles of monetary policy will impact the markets. The empirical analysis includes the ECB monetary policy announcements from the be- ginning of 2000 until the end of 2022. There are three announcements in total, which are left out of the analysis as they have been given on Finland’s Independence Day, and therefore the reaction differs from all the other events as the markets are closed on the event day. All the monetary policy announcements are handled as independent events and for all events cumulative abnormal returns are calculated. 5.1 Results of all announcements from 2000 to 2022 This subchapter discusses the findings of all announcements given between 2000 and 2022. Table 2. Average abnormal and cumulative abnormal returns of all announcements ECB announcements 2000–2022 CAARevent window 0,4519 t-statistic 1,7696* AARpre event day 0,0681 t-statistic 0,5730 AARevent Day 0,2770 t-statistic 1,7080 AARPost Event Day 0,1067 t-statistic 0,8443 Number of announcements 265 *= significant at 10% level 45 As we can see in Table 2, most market reactions seen in Finnish stock markets are some- what reflecting the efficient market hypothesis. However, with the chosen event window, the positive cumulative abnormal return of 0.45% implicates a significant abnormal re- turn at the 10% level. This said, we can see that reactions are more likely to cumulate than on average create significant abnormal returns in one day. From the individual days in the event window, the event day itself creates the most significant abnormal returns. This implies, that there is more movement in the market on the announcement day than on the days surrounding it. The movement although is not statistically significant and can be said that on average there is no surprise in the announcement and markets tend to behave efficiently towards monetary policy announcements. There is not enough evidence to reject the null hypothesis of hypothesis 1a as the ab- normal returns associated with the individual days in the event window are not statisti- cally significant. There is moderately significant evidence to reject the null hypothesis of hypothesis 1b and the cumulative abnormal returns associated with the three-day event window are statistically significant. In Table 2 we see that event day does generate the highest abnormal returns, but there is no significant evidence to reject the null hypoth- esis and therefore hypothesis 2 cannot be confirmed. In Figure 2 below we see how the different levels of significance have been distributed in all the announcements. Over 60% of the announcements do not create significant abnormal returns over the event window. However, the number of significant announce- ments is still rather high, as it is almost a third of all announcements. It is notable that on the given event window there are almost identical amounts of announcements sig- nificant at 5% and 10% and less at 1%. This given, only 4,5% of the results of the an- nouncement are significant at 1% and therefore markets seem to forecast the announce- ments rather well. 46 As the central bank is acting reactively on the economic environment, the announce- ments are also given in times of crises or other significant events with effects on the markets, and therefore saying the reaction in the markets is caused only by the central bank announcement is not sensible. Figure 2: Distribution of cumulative abnormal returns in significance level over the event window period. Where: *, **, *** indicate the level of significance at 10%, 5%, and 1% respectively. The announcements in Figure 2 also include an error term for non-announcement-re- lated matters that have an impact on markets. One example of this kind of matter is when 9/11 took place and central banks gave guidance on how they will support the economy. Markets were not only reacting to the announcement but to the aftermath of the whole catastrophe. The distribution of significance levels supports markets being ef- ficient towards the announcements known in advance and is in line with the overall ex- pectation of only surprises resulting in significant abnormal returns. 67,17% 14,34% 13,96% 4,53% 0,00% 10,00% 20,00% 30,00% 40,00% 50,00% 60,00% 70,00% 80,00% % not significant % of * % ** % *** Distribution of announcements significance in percentages 47 5.2 The effect of different announcements This subchapter discusses how the results of different announcement characteristics re- sult in stock markets. In other words, how the actual information in the announcement affects the market reaction. The AARs and CAARs are calculated by taking all the an- nouncements with the same information, i.e. if the announcement included an interest rate raise and unconventional actions the abnormal return of that announcement is in- cluded in both. Table 3. Different announcements and reactions. AAR-1 AAR0 AAR+1 CAAREvent window No changes in rates 0,0975 0,2444 0,0763 0,4182 Unconventional actions 0,2894 0,0793 -0,2090 0,1597 Interest rate raise -0,0128 0,2458 0,4853 0,7183 Interest rate decline -0,1544 0,6405 0,0328 0,5189 From Table 3, we can see the interesting distribution of how average abnormal returns are highest in the event day when there are no changes in the rates and when interest rates are declining. This is an expected reaction, as most of the announcements do not include a change of interest rates, and therefore can be said that this is the most com- mon, and most anticipated announcement. The interest rate decline and the immediate reaction to it is in line with Jarocinski & Karadi (2018) and the expectation that declining interest rates are supporting higher returns in stocks. On the other hand, an interest rate raise seems to make the market think longer on the direction, as the post-announcement day is creating the highest average abnormal 48 returns on interest rate raise. The anticipation towards interest rate changes is negative, no matter if markets are waiting for rising or declining rates. This indicates that the fore- casts might be slightly more negative when it is likely to see a change in rates and the correction is seen more rapidly in the case of lowering interest rates. The unconventional actions seem to make markets excited, as they have been at least for the last couple of years a big speculation before the meetings. Markets seem to fore- cast these more optimistically than any other announcements. This is also a very logical reaction, as the buyback programs for example can re-direct huge amounts of money from bond markets into the stock markets, but also support the overall economy. The reaction seen on the event day is positive on average, but markets seem to be a little too optimistic towards the unconventional actions, as there is a correction seen on the post- announcement day, which is negative. It is still important to point out, that none of the reactions seen are statistically significant at 10% or lower, but more likely a trend that can be revealed by using average abnormal returns as a measure. So significant evidence of mispricing and false forecasts cannot be found. Therefore, the hypothesis of one specific announcement characteristic creating significant abnormal returns cannot be rejected. Notable is that cumulative abnormal returns are positive in all announcement types and are highest when interest rates are raised. This is an inter- esting finding, especially as the expected results should, at least based on the previous findings to be somewhat the other way around as higher interest rate levels do not sup- port the stock market similarly to lower interest rates. However, the findings are not statistically significant, so therefore hard evidence to support the reaction in the Finnish stock market cannot be proved and the results might be caused by too pessimistic fore- casts about the announcements. 49 Table 4. Event window significance, % of total announcement of the same kind. No changes in rates Unconventional Interest raise Interest decrease Significant at 10% level 10,36% 11,11% 20,00% 0,00% No. of observations 23 7 4 0 Significant at 5% level 19,37% 22,22% 20,00% 10,00% No. of observations 43 14 4 2 Significant at 1% level 13,51% 15,87% 10,00% 0,00% No. of observations 30 10 2 0 No. of announcements 222 63 20 23 Significant CARs % of total announcements 43,24% 49,21% 50,00% 8,70% From Table 4 we see the percentages that different announcements resulted in signifi- cant abnormal returns at given significance levels, measured with the cumulative abnor- mal returns of the event window. The highest abnormal returns resulted when interest rates rose, and almost as high a percentage of announcements resulted in significant abnormal returns when announcements included unconventional actions. Given that most of the announcements included no changes in interest rates, and it is probable that in the 222 announcements, there were other factors also moving the market, at least in some of the announcements, not only the given announcement of the central bank. The overall results are in line with Bernanke (2020), who stated that unconventional pol- icies are supportive and effective for the economy, and the stock market seems to react in that way. The results show that stock markets have in almost half of the announce- ments resulted in significant abnormal returns when unconventional actions were taken. The strong reaction towards rising interest seems to support the hypothesis of higher interest rates lowering stock prices and once interest has been increased, stocks are rap- idly evaluated for new prices. 50 5.3 Investor sentiment This subchapter discusses the monetary policy effects on different market sentiments. According to Kurov (2010), there is no perfect measurement of investor sentiment. This thesis follows the common rule of thumb of defining bear markets; a 20% decrease in stock prices from their last high marks a so-called bear market. Other times, as the mar- kets tend to be more upward than downward in the long run, are defined as bull markets. In total, there are five periods, which are considered as bear markets. Their maturity varies and the focus is on longer periods of downward trend periods. One exception is the Covid-19 crash, which lasted for a relatively short period but was actually a very quick transition from one of the longest bull market periods in history to a bear market. Also, as mentioned in the beginning, central banks acted as major players in creating trust in the economy during the pandemic period. The periods when markets are bearish and the profit from this period is calculated by dividing the lowest point (where the bear market is also considered to end) by the high- est point of the market (the point where the bear market is considered to begin). This gives the time window for sentiment and the overall decrease can also be seen. The five periods of bear markets in this thesis are as follows: 1. 28.4.2000-10.3.2003 – a total decrease from the highest -74%, 2. 7.11.2007-6.3.2009 – a total decrease from highest -68% 3. 9.2.2011-4.6.2012 – a total decrease from highest -39% 4. 11.2.2020-18.3.2020 – a total decrease from highest -36% 5. 6.9.2021-29.9.2022 – a total decrease from highest -27% Other periods are considered bull markets. The only exception is the time frame from 30.9.2022-30.12.2022 because at this time market did not achieve the 20% increase 51 which is considered here as the definition of the bull market. This period had 2 ECB an- nouncements which were left out of the calculations to make analysis more consistent. Table 5. Cumulative abnormal returns and abnormal returns bull & bear market Bull market Bear market CAAR(event window) 0,1558 1,0098 t-statistic 0,6342 1,8013* AAR(pre event day) 0,0267 0,1260 t-statistic 0,2397 0,4717 AAR(event Day) -0,0146 0,8216 t-statistic -0,104 2,207** AAR(Post Event Day) 0,1437 0,0622 t-statistic 1,4526 0,2130 *= statistically significant at 10%, **= statistically significant at 5% In Table 5 we see the cumulative abnormal returns for the event window and individual daily abnormal returns of the event window. The results of the analysis are partially in line with Kurov (2010). Bear markets tend to result in more significant abnormal returns for the three-day event window, but also on event day. This meaning that monetary pol- icy decisions seem to support stock markets more in a bear market, while the investor sentiment is negative. This is a logical outcome as central banks are pursuing their policy by supporting the economy during difficult times. Most and largest of the unconven- tional actions by central banks are seen in times of unclear future forecasts, for example in the aftermath of the financial crisis or the Covid-19 pandemic. The results also show that stock markets are not anticipating the monetary policy actions correctly during harder times. This is not necessarily market inefficiency, but more likely the actions undertaken by central banks are hard to forecast and some of the actions are not very often done. There is enough evidence to support the hypothesis that investor 52 sentiment has a significant effect on how markets react to monetary policy announce- ments. The results differ significantly, with bear market reactions being significant for event day and through the event window, while the bull market does not result in signif- icant abnormal returns. 5.4 Tightening cycles and loosening cycles of monetary policy This subchapter of the empirical part discusses the different cycles of monetary policy and the effects the cycles have on stock markets. Loosening cycles and tightening cycles are in this study considered as follows: On the loosening cycle, interest rates are lowered and on the tightening cycle interest rates are raised. There are in total 5 different cycles when the rates are trending downwards or upwards. Mostly the policy that ECB has prac- ticed has been loosening. This means, that most of the time central bank has kept the rates low or lowered them versus the previous announcement. According to Jarocinski & Karadi (2018) loosening policy should support stock markets. The options in lower-risk level assets are not profitable, and money is transferred to more risky assets. In Table 6 we can see how the different cycles are affecting to stock market index when measured with CAARs and AARs. Table 6. Tightening and loosening cycles of monetary policy Loosening cycle Tightening cycle CAAR(event window) 0,5846 0,5316 t-statistic 2,3412** 0,7963 AAR(pre event day) 0,2504 -0,4299 t-statistic 2,2951** 1,3307 AAR(event Day) 0,2927 0,2344 t-statistic 1,9974** 0,5132 AAR(Post Event Day) 0,0415 0,2849 53 t-statistic 0,3242 0,8974 Total No. of observations 194 71 ** = statistically significant at 5% As we see from the results, the cycles of tightening and loosening policy do have a sig- nificant impact on stock prices. This is in line with Jarocinski & Karadi (2018), as the loos- ening monetary policy should support the stock market. This finding also supports Blampied & Mahadeo (2023), as in the easing cycles the monetary policy supports stock markets significantly, at least on a short time horizon. It seems that on loosening cycles markets either see monetary policy as surprising or encourage investors to carry the ad- ditional risk when holding stock positions which is shown by Bekaert et al (2013). Tightening, i.e. increasing interest rates on the other hand does not result in significant abnormal returns in the stock market. The rising interest rates do not support risk-taking holding stocks similarly, as new bonds begin to yield better. This might also be caused by the forecast of increasing interest rates. If the expectations of higher interest rates are anticipated, it will be efficiently priced in the stock markets. The evidence supports the hypothesis that the direction of monetary policy is a significant factor in the stock market reaction around the announcement, and there is strong evidence to reject the null hy- pothesis. 54 6 Conclusions This study focuses on the stock market reactions to central bank monetary policy an- nouncements. Particularly the focus is on European Central Bank monetary policy an- nouncements and Finnish stock markets. The main purpose is to understand better how the Finnish stock markets are behaving around the Central Bank’s monetary policy ac- tions. In the study, a very traditional event study approach was applied to test different announcement characteristics, and how different market sentiment and different direc- tions of monetary policy affects on stock returns. This study extends the prior literature in the context of central banks and stock market relationships, especially related to the Finnish stock market. As the main idea is to observe a very short period around the an- nouncements, this study does not account for the possible effects that can be seen in a longer time horizon. The empirical findings of this study show that Finnish stock markets tend to anticipate the monetary policy announcement moderately well, and therefore can be said that markets tend to be efficient towards monetary policy actions. The three-day event win- dow results in moderately statistically significant positive abnormal returns and individ- ual days in the event window do not differ significantly from what is expected. It is nota- ble that the average abnormal return seen in all the three days around the announce- ment were positive. As the findings are not statistically significant, the economic significance is based on too many random factors, and therefore solid evidence was not found to make predictions in the future direction. One aspect of these results might be the direction of the mone- tary policy in the sample period, but also the direction of monetary policy conducted by the ECB in its whole history, which has been mostly loosening, or interest rates have been historically low, even negative. Investor sentiment was found to have slightly different results. Markets tend to react similarly measured by cumulative abnormal returns, but on average, on the event day 55 reaction is statically more significant. The findings of investor sentiment’s effect on the reaction is in line with the findings of Kurov (2010). In bear markets stock market returns differ from expected significantly versus on bull markets. Loosening cycles of monetary policy were found to be significant while abnormal returns in tightening cycles were not significant. The average abnormal returns associated with loosening monetary policy as well as tightening policy were mostly positive, which is similar to the abnormal returns in the total amount of announcements. This study examines only the short-term effects of monetary policy decisions on stock returns and does not reveal any longer-term effects there might be. The short-term ef- fects are important to understand, as markets tend to react on surprising, or other eco- nomically significant new information strongly. To understand the longer-term effects of monetary policy on stock markets, more research is needed to conduct. This study gives a starting point for understanding the short-term reactions in Finnish stock markets and adds into the literature of studies conducted on the relationship between Finnish stock markets and macroeconomic events. When it comes to the effects of monetary policy on the stock markets, policymakers, especially in the central banks, need to understand the consequences of their actions in the stock markets. If the markets become too reliant on central banks, it may have seri- ous consequences when it comes to functional financial markets and of course efficient markets. The effects of being too reliant on the Central Banks will also be seen in the real economy as the economic machine is not working properly, not only in the financial mar- kets. As there has been several crises in the near past, this is something that needs to be researched over time, in different economic conditions and different investment envi- ronments. An understanding of the effects on stock markets can also be helpful in times of crises when financial markets need to be kept functioning. The effectiveness of central bank 56 policies has been shown to be good, but the economic crises, for example, are usually not copies of the previous crises, and therefore the research is important over time. 57 7 Future research This chapter briefly discusses the future research ideas on the relationship between stock markets and monetary policy conducted by central banks, mainly focusing on the Finnish stock markets. Future research is necessary, as the relationship between the Hel- sinki stock exchange and monetary policy is not widely examined. As Finland is a part of the European Union and part of the euro system, it is important to understand the ef- fects that the European Central Banks’s monetary policy has on its financial markets. As this study helps to understand how efficiently markets predict actions and how the monetary policy announcements result in equity prices, future research ideas in this con- text would be a comprehensive study on the volatility of the Finnish stock markets during monetary policy announcements and testing if the volatility follows the same pattern of the individual day’s abnormal returns founded in this thesis. The data for volatility in the Finnish stock market is not available similarly to some bigger markets, and for example a good measurement of surprises, implied volatility is not calculated. As said earlier, a similar study method to Vähämaa & Äijö (2011), where the implied volatility is employed would give valuable insight into the monetary policy surprises and reactions to the sur- prises. Although Finnish stock markets are probably mostly affected by the monetary policy conducted by the European Central Bank, a study that examines the spillover effect from the monetary policy conducted by the US Federal Reserve would give valuable insight into the relationship also to monetary policy conducted by other central banks than ECB. This could help to understand the dependencies between the Finnish economy and the US, and also the dependencies between the Finnish financial markets and US monetary policy actions. It is important to examine how the monetary policy affects the smaller market areas such as Finland. In the case of Finnish stock markets, this is crucial, as the monetary policy is practiced mostly based on what is happening in the European economy, not just in 58 Finland. It is also valuable to understand better how the actions by central banks are forecasted in the smaller market areas where analyst coverage is smaller and test if the results differ significantly from the bigger market areas. As mentioned earlier, this study focuses only on a short time horizon and the effects that can be seen immediately after a monetary policy announcement. Examining the effects on a longer time horizon can reveal a better understanding on how much the monetary policy cycles impact stock markets and how effective the monetary policy is in the con- text of equity markets. It would also be interesting to see if the findings are similar to the shorter time period. 59 References Abel, A. & Bernanke, B. & Croushore, D. (2014). Macroeconomics. Pearson, 534,553. Aggarwal, R., Akhigbe, A., & Mohanty, S. K. (2012). Oil price shocks and transportation firm asset prices. Energy Economics, 34(5), 1370-1379. https://doi.org/10.1016/j.eneco.2012.05.001 Armitage, S. (1995). Event study methods and evidence on their performance. Journal of Economic Surveys, 9(1), 25-52. https://doi.org/10.1111/j.1467- 6419.1995.tb00109.x Ball, R. & Brown, P. (1968). An empirical evaluation of accounting income numbers. Journal of Accounting Research 6, 159–177. https://doi.org/10.2307/2490232 Baumann, U., Lodge, D., & Miescu, M. S. (2024). Global growth on life support? The contributions of fiscal and monetary policy since the global financial crisis. In- ternational Journal of Finance and Economics, 29(1), 76-90. https://doi.org/10.1002/ijfe.2675 Bekaert, G., Hoerova, M., & Lo Duca, M. (2013). Risk, uncertainty, and monetary policy. Journal of Monetary Economics, 60(7), 771-788. https://doi.org/10.1016/j.jmo- neco.2013.06.003 Bekaert, G., Hoerova, M., & Xu, N. R. (2023, November 27). Risk, monetary policy, and asset prices in a global world. ECB Working Paper Series. https://www.ecb.eu- ropa.eu/pub/pdf/scpwps/ecb.wp2879~06c76dacd5.en.pdf Benninga, S., & Czaczkes, B. (2014). Financial Modeling. MIT Press. Bernanke, B. S. (2022). 21st century monetary policy – the Federal Reserve from the great inflation to COVID-19. WW Norton & Co Bernanke, B. S., & Kuttner, K. N. (2005). What explains the stock market's reaction to Federal Reserve policy? The Journal of Finance (New York), 60(3), 1221-1257. https://doi.org/10.1111/j.1540-6261.2005.00760.x Bernanke, B. S. (2020). The new tools of monetary policy. The American Economic Re- view, 110(4), 943-983. https://doi.org/10.1257/aer.110.4.943 60 Binder, J. J. (1998). The event study methodology since 1969. Review of Quantitative Finance and Accounting, 11(2), 111-137. https://doi.org/10.1023/A:1008295500105 Blampied, N., & Mahadeo, S. M. R. (2023). Uncertainties under monetary tightening and easing shocks and different market states. Finance Research Letters, 55, 103834. https://doi.org/10.1016/j.frl.2023.103834 Brown, S. J. & Warner, J. B. (1985). Using daily stock returns: the case of event studies. Journal of Financial Economics 14, 3–31. https://doi.org/10.1016/0304- 405X(85)90042-X Bordo, M. D. (2007). A brief history of central banks. Federal Reserve Bank of Cleve- land: Economic Commentary. https://www.clevelandfed.org/publications/eco- nomic-commentary/2007/ec-2007120 Campbell, J. Y., Lo, A. W. & MacKinlay, A. C. (1997). The econometrics of financial mar- kets. Princeton, NJ: Princeton University Press. Campbell, C. J. & Wasley C. E. (1993). Measuring security performance using daily NASDAQ returns. Journal of Financial Economics 33, 73–92. https://doi.org/10.1016/0304-405X(93)90025-7 Charest, G. (1978). Dividend information, stock returns and market efficiency. Journal of Financial Economics 6, 265–296.https://doi.org/10.1016/0304- 405X(78)90033-8 Chen, A. H., & Siems, T. F. (2004). The effects of terrorism on global capital markets. European Journal of Political Economy, 20(2), 349-366. https://doi.org/10.1016/j.ejpoleco.2003.12.005 Claeys, G., Leandro, A. & Mandra, A. (2015). European Central Bank quantitative eas- ing: the detailed manual. https://www.jstor.org/stable/resrep50956 Corrado, C. J. (2010). Event studies: a methodology review. Accounting & Finance 51, 207–234. https://doi.org/10.1111/j.1467-629X.2010.00375.x Dyckman, T., Philbrick, D. & Stephan, J. (1984). A comparison of event study methodol- ogies. Journal of Accounting Research 22, 1-33. https://doi.org/10.2307/2490855 61 Ellingsen, T & Söderström, U. (1998). Monetary policy and market interest rates. Sve- riges Riksbank Working Paper Series, No. 56. https://www.econstor.eu/bit- stream/10419/83043/1/768631203.pdf European Central Bank. What is a central bank? (2015) ECB website, retrieved 2024-01- 15 from: https://www.ecb.europa.eu/ecb/educational/explainers/tell- me/html/what-is-a-central-bank.en.html European Central Bank. (2018). What is the marginal lending facility rate? ECB website, Retrieved 2024-02-12 from https://www.ecb.europa.eu/ecb/educational/ex- plainers/tell-me/html/marginal_lending_facility_rate.en.html European Central Bank. (2018). What is the main refinancing operations rate? ECB website, Retrieved 2024-02-12 from https://www.ecb.europa.eu/ecb/educa- tional/explainers/tell-me/html/mro.en.html European Central Bank. (2022). What is the deposit facility rate? ECB website, Re- trieved 2024-02-12 from https://www.ecb.europa.eu/ecb/educational/explain- ers/tell-me/html/what-is-the-deposit-facility-rate.en.html Fama, E. F., Fischer, L., Jensen, M. & Roll, R. (1969). The adjustment of stock prices to new information. International Economic Review 10, 1–21. https://doi.org/10.2307/2525569 Fausch, J., & Sutter, D. (2024). Monetary policy spillovers: The impact of ECB conven- tional and unconventional monetary policies on the Swiss stock market. Applied Economics Letters, 31(2), 122-127. https://doi.org/10.1080/13504851.2022.2128173 Ferrando, A., Popov, A., & Udell, G. F. (2022). Unconventional monetary policy, funding expectations, and firm decisions. European Economic Review, 149, 104268. https://doi.org/10.1016/j.euroecorev.2022.104268 Feldkircher, M., Huber, F., & Pfarrhofer, M. (2021). Measuring the effectiveness of US monetary policy during the COVID-19 recession. Scottish Journal of Political Economy, 68(3), 287-297. https://doi.org/10.1111/sjpe.12275 62 Fratzscher, M., Duca, M. L., & Straub, R. (2016). ECB unconventional monetary policy: market impact and international spillovers. IMF Economic Review, 64(1), 36-74. https://doi.org/10.1057/imfer.2016.5 Goodhart, C. (1988). The Evolution of Central Banks. The MIT Press. Goodhart, C. A. E. (2011). The changing role of central banks. Financial History Review, 18(2), 135-154. https://doi.org/10.1017/S0968565011000096 Hartley, J. S., Jiménez, D., & Rebucci, A. (2020). An event study of COVID-19 central bank quantitative easing in advanced and emerging economies. NBER Working Paper Series. https://doi.org/10.3386/w27339 Hartmann, P., & Smets, F. (2018). The European Central Bank’s monetary policy during its first 20 years. Brookings Papers on Economic Activity, 2018(2), 1-118. https://doi.org/10.1353/eca.2018.0026 Iacoviello, M., & Minetti, R. (2003). Financial liberalization and the sensitivity of house prices to monetary policy: theory and evidence. The Manchester School, 71(1), 20-34. https://doi.org/10.1111/1467-9957.00332 Jarociński, M., & Karadi, P. (2020). Deconstructing monetary policy surprises— the role of information shocks. American Economic Journal. Macroeconomics, 12(2), 1- 43. https://doi.org/10.1257/mac.20180090 Jarrow, R. A. (2013). The zero-lower bound on interest rates: myth or reality? Finance Research Letters, 10(4), 151-156. https://doi.org/10.1016/j.frl.2013.08.003 Kurov, A. (2010). Investor sentiment and the stock market’s reaction to monetary pol- icy. Journal of Banking & Finance, 34(1), 139-149. https://doi.org/10.1016/j.jbankfin.2009.07.010 Kurov, A. (2012). What determines the stock market's reaction to monetary policy statements? Review of Financial Economics, 21(4), 175-187. https://doi.org/10.1016/j.rfe.2012.06.010 Makin, A. (2002). International Macroeconomics. Pearson Education Limited. Maynes, E. & Rumsey, J. (1993). Conducting event studies with thinly traded stocks. Journal of Banking and Finance 17, 145–157. https://doi.org/10.1016/0378- 4266(93)90085-R 63 Mishkin, F. S. (2000). What should central banks do? Review - Federal Reserve Bank of St. Louis, 82(6), 1. https://doi.org/10.20955/r.82.1-14 Mojon, B. (2010). The 2007–2009 financial crisis and the European Central Bank. Open Economies Review, 21(1), 175-182. https://doi.org/10.1007/s11079-009-9151-7 Patell, J. A. (1976). Corporate forecasts of earnings per share and stock price behavior: Empirical test. Journal of Accounting Research 14, 246–276. https://doi.org/10.2307/2490543 Pearce, D. K., & Roley, V. V. (1983). The reaction of stock prices to unanticipated changes in money: a note. The Journal of Finance (New York), 38(4), 1323-1333. https://doi.org/10.1111/j.1540-6261.1983.tb02303.x Pearce, D. K., & Roley, V. V. (1985). Stock prices and economic news. The Journal of Business (Chicago, Ill.), 58(1), 49-67. https://doi.org/10.1086/296282 Ricci, O. (2015). The impact of monetary policy announcements on the stock price of large European banks during the financial crisis. Journal of Banking & Finance, 52, 245-255. https://doi.org/10.1016/j.jbankfin.2014.07.001 Roley, V. V., & Sellon, G. H. (1995). Monetary policy actions and long-term interest rates. Federal Reserve Bank of Kansas City Economic Quarterly, 80(4), 77-89. https://EconPapers.repec.org/RePEc:fip:fedker:y:1995:i:qiv:p:73-89:n:v.80no.4 Singleton, J. (2011) Central Banking in the Twentieth Century. Cambridge University Press, vol. 18(3) https://doi.org/10.1017/CBO9780511782053 Svensson, L. E. (2020). Monetary policy strategies for the Federal Reserve. Interna- tional Journal of Central Banking, 16(1), 133-193. http://dx.doi.org/10.3386/w26657 Thorbecke, W. (1997). On stock market returns and monetary policy. The Journal of Fi- nance (New York), 52(2), 635-654. https://doi.org/10.1111/j.1540- 6261.1997.tb04816.x Vähämaa, S., & Äijö, J. (2011). The Fed's policy decisions and implied volatility. The Journal of Futures Markets, 31(10), 995-1010. https://doi.org/10.1002/fut.20503 64 Wang, S., & Mayes, D. G. (2012). Monetary policy announcements and stock reactions: an international comparison. The North American Journal of Economics and Fi- nance, 23(2), 145-164. https://doi.org/10.1016/j.najef.2012.02.002 Yamaguchi, K. (2008). Reexamination of stock price reaction to environmental perfor- mance: A GARCH application. Ecological Economics, 68(1), 345-352. https://doi.org/10.1016/j.ecolecon.2008.04.004 Zabala, J. A., & Prats, M. A. (2020). The unconventional monetary policy of the Europe- an Central Bank: Effectiveness and transmission analysis. World Economy, 43(3), 794-809. https://doi.org/10.1111/twec.12880