Deeds versus words: De jure and de facto exchange rate regimes in Central and Eastern European countries

Transcription

1 UNIVERSITEIT GENT FACULTEIT ECONOMIE EN BEDRIJFSKUNDE ACADEMIEJAAR Deeds versus words: De jure and de facto exchange rate regimes in Central and Eastern European countries Masterproef voorgedragen tot het bekomen van de graad van Master of Science in de Toegepaste Economische Wetenschappen: Handelsingenieur Jolien Tortelboom onder leiding van Prof. Michael Frömmel

2

3 UNIVERSITEIT GENT FACULTEIT ECONOMIE EN BEDRIJFSKUNDE ACADEMIEJAAR Deeds versus words: De jure and de facto exchange rate regimes in Central and Eastern European countries Masterproef voorgedragen tot het bekomen van de graad van Master of Science in de Toegepaste Economische Wetenschappen: Handelsingenieur Jolien Tortelboom onder leiding van Prof. Michael Frömmel

4 Confidentiality clause Permission The undersigned declares that the content of this paper may be consulted and/or reproduced, if acknowledgement is given. Jolien Tortelboom I

5 Preface This thesis concludes my five years at Ghent University and is written in order to obtain my master s degree in Business Engineering with a major in Finance. I have chosen this subject because the world of finance and especially macro-economic structures has always fascinated me. But most of all, it allowed me to have a good variation between mathematical reasoning and challenging theoretical clarifications. I would not be able to write this thesis with the help of a few people. First of all, I would like to thank my promoter professor Michael Frömmel for providing me the right guidelines and constructive feedback. I would also like to thank Xing Han for teaching me how to work with the program DataStream, without his assistance I would not be able to collect the required data. Furthermore, I m very grateful for the help of Ms. Franziska Schobert who afforded me valuable information which gave this thesis an added value. A special thank you to Gert Elaut for his support. Without his help and good insights this thesis would not have become was it is today. At last, I would like to thank my family and friends for their support and necessary distraction when I needed it the most during my 5 years of studying. Jolien Tortelboom II

6 Table of Contents Confidentiality clause... I Preface... II Table of Contents... III List of Abbreviations and Symbols... IV List of Tables... VI 0 Introduction Description of the official exchange rate regimes Literature on the fear of floating Monetary policy interventions Two-corner hypothesis Literature on the methodology Inferring flexibility Inferring weights Synthesis of the two techniques Empirical Analysis Data and Method Results Extension Conclusions References... VIII Appendix 1: Data... Appendix 1.1 Appendix 2: Tables... Appendix 2.1 Appendix 3: Programs... Appendix 3.1 Appendix 4: Dutch Summary... Appendix 4.1 III

7 List of Abbreviations and Symbols Abbreviations: AvM2 BSI CBR CEEC CHF CIS CNB CZK DEM DEP ECB Ecu EMP EMU Average value of M2 Band width Bank of Slovenia Constant term Central Bank Russian Republic Central and Eastern European Countries Swiss Franc Commonwealth of Independent States Czech National Bank Czech koruna Rate of crawl Deutsche Mark Standard deviation of the rate of Depreciation Exchange rate for a certain country (time series) European Central Bank European Currency Unit Exchange Market Pressure Economic/European Monetary Union ERM II Exchange Rate Mechanism EUR FF FLEX G7 GBP GMM HUF Euro Fear of Floating or French Franc Factor analysis exchange rate Flexibility Group of seven most advanced economies British Pound Sterling Generalized Method of Moments Hungarian forint IV

8 i IFS IMF IT M2 N NBH NBP NBR NBS Obs. OLS PLZ Interest rate Interest rate (time series) International Financial Statistics International Monetary Fund Inflation targeting Secondary Monetary supply Amount of foreign currencies National Bank of Hungary National Bank of Poland National Bank of Romania National Bank of Slovakia Observations Ordinary Least Squares Polish zloty ² R Squared Res RES RMB ROL RUR RVEI RVER SDR SIT SKK Absolute change in reserves Relative change in reserves Reserves Standard deviation of the stock of reserves Chinese yuan Romanian leu Russian ruble Relative Volatilities of the Exchange rate and Interest rates Relative Volatilities of the Exchange rate and Reserves Special Drawing Right Slovenian tolar Slovak koruna Exchange rate for certain country (time series) V

9 Std USD Standard deviation Time variable United States Dollar Weight against currency Symbols: Log Weight against the EMP component Error term (time series) Error term (time series) Exchange rate volatility Volatility of exchange rate changes Volatility of reserves Natural logarithm Euro $ US dollar Yen Change in value List of Tables Table 1: Czech Republic: official monetary and exchange rate regime... 3 Table 2: Hungary: official monetary and exchange rate regime... 5 Table 3: Poland: official monetary and exchange rate regime... 6 Table 4: Romania: official monetary and exchange rate regime... 6 Table 5: Russian Republic: official monetary and exchange rate regime... 8 Table 6: Slovakia: official monetary and exchange rate regime... 9 Table 7: Slovenia: official monetary and exchange rate regime Table 8: Estimation results Czech Republic without adding up constraint Table 9: Estimation results Hungary without adding up constraint Table 10: Estimation results Poland without adding up constraint Table 11: Estimation results Romania without adding up constraint VI

10 Table 12: Estimation results Russia without adding up constraint Table 13: Estimation results Slovakia without adding up constraint Table 14: Estimation results Slovenia without adding up constraint VII

11 0 Introduction The last years, researchers are becoming aware of the existence of dissimilarities between de jure and de facto exchange rate regimes. The de jure exchange rate regime is the official announced exchange rate regime that the national government of a country is claiming to be following, and the de facto exchange rate regime is the regime it is following in reality. Especially, small open economies often don t do what they are saying (i.e. deeds vs. words), because they have not yet found a stable economy and are characterized by a high passthrough-effect. Therefore, they intervene in their policies to find a balance between sinking the inflation and remaining competitive relative to other countries. In this work, we examine seven Central and Eastern European countries by comparing their official exchange rate regimes with the de facto exchange rate regimes we estimate, by using the methodology based on the work of Frankel (2009). The countries are Czech Republic, Hungary, Poland, Romania, the Russian Republic, Slovakia and Slovenia. These observed countries are fascinating because they all are small open economies and are on their way to the Economic Monetary Union (EMU), except for the Russian Republic that is no member of the European Union. Since there exist a great amount of variants of implicit or intermediate exchange rate regimes, we use a methodology that covers the flexibility criterion next to the more fixed anchors. To accomplish this double-sided approach, the Exchange Market Pressure variable is created which estimates the degree of flexibility at the same time as the currency weights. The goal of this work is to find evidence of differences (or similarities) between de jure and de facto exchange rate regimes, so we can say with certainty which kind of exchange rate regime a country is applying. This is essential, because before we can make an evaluation whether a country is following the optimal exchange rate regime, we first must identify which kind of regime it is actually applying. I have chosen this subject, because it is an issue that is very present and could be of use to future researches to find possible explanations to market trends. As the result of a greater awareness of this deviation, the International Monetary Fund (IMF) attaches more importance to the correct representations of the actual exchange rate regimes applied by countries. Moreover, the effect of the decisions the authority of a country makes concerning the 1

12 exchange rate policy on macro-economic structures is important. Therefore, I found that my work has a strong purpose by clarifying the correct regimes. This stimulating subject also provided me to exert and enhance my mathematical and econometric insights, on top of the also challenging- theoretical descriptions. The remainder of the text is as follows: section one provides an overview of the different countries concerning their official monetary policy and exchange rate regime based on existing literature, official classifications of the IMF or publications of the national banks. We start with this part because it is important to identify the official exchange rate regimes, with which we can compare the de facto exchange rate regimes. The purpose of section two, the literature on the fear of floating, is to give the reader an impression on what has been done in the field until now and list some possible explanations for the frequent adjustments that countries execute to their exchange rate policies. This paragraph starts with the explanations of monetary policy interventions and subsequently elaborates on the two-corner hypothesis. The third section, the literature on the methodology, is divided into three parts. First, the different methodologies described in the literature are rendered on identifying the true flexibility that a currency is applying (i.e. inferring flexibility). Secondly, the methodology and some extensions on the weight inference technique is clarified on defining the true basket a country is adopting, when following a more tight exchange rate policy. Lastly, we go deeper into the clarification of our implemented methodology we will take on in this work, based on the literature which is constructed on a synthesis of the two techniques explained earlier. In the fifth section, the empirical analysis, we provide more information on the accumulation of the data and practiced method. Furthermore, we give an overview of the results in tables and our interpretation per country. To conclude this paragraph, we add an extension to the method by adding one extra variable, the change of the interest rate, and provide the results. In the fifth section, a conclusion is given about the findings in this work and some possible enhancements. 2

13 1 Description of the official exchange rate regimes We will concentrate on six Central and Eastern European Countries (Visegrád group) that recently became members of the European Union and 1 CEEC country (Russian Republic) that is not a member of the European Union. Slovenia, Poland, Hungary, Slovakia and the Czech Republic became members in 2004, while Romania became a member in Slovenia and Slovakia are already applying the euro as their currency, in 2007 and 2009 respectively. In this section we will list the official monetary policy and exchange rate regime of the seven countries appointed above and make a conclusion about this overview. This paragraph is mostly inspired by the work of Frömmel and Van Gysegem (2011). The exchange rates of the CEEC countries have experienced an interesting evolution in the past 20 years. The Visegrád group which became a member of the European Union in the years 2004 and 2007, has the obligation to enter the European Monetary Union. These countries must meet the five Maastricht criteria to adopt the European Union s single currency. One of the criteria is the stability of the exchange rate inside the European Exchange Rate Mechanism (ERM II) two years prior to entry. As a result, the exchange rate regime chosen by the countries is an important subject. Though the CEEC countries all started from a comparable situation -the disbanding of the Soviet Union in December and have the common long-term goal of joining the Monetary Union, they all opted for a different exchange rate policy (Frömmel& Van Gysegem, 2011). The Czech Republic followed a strategy of exchange rate and monetary targeting. This strategy was successful in the early 1990s, but thereafter increasing capital inflows made this strategy unmanageable. In March 1996, they decided to widen the exchange rate band from +/- 0.5% to +/- 7.5%. In May 1997, the Czech National Bank (CNB) opted to switch to a managed float. Because the CNB considered the demand for money to be too unstable to use a monetary aggregate as an intermediate target, they changed the monetary policy to a strategy of inflation targeting. Nevertheless, during this period of inflation targeting, the CNB kept on intervening in the foreign exchange market. Source: (19/02/2014) Table 1: Czech Republic: official monetary and exchange rate regime 1 Source: (19/02/2014) 3

14 Czech Republic Official Exchange Official Monetary Policy Arrangements Time interval Type Time interval Type Exchange rate and Basket peg, 65% DEM, 01/01/ /02/1996 monetary targeting 35% USD, Band +/-0.5% Net inflation targeting (a) 01/03/ /05/1997 Band: +/-7.5% Target band for headline, consumer price inflation targeting 27/05/ Managed float Point target for headline inflation Source: Frömmel, Garabedian and Schobert (2011), Frömmel and Van Gysegem (2011), IMF: Annual Report of Exchange rate Arrangements and Restrictions (2012), website Czech Republic National Bank (a) Headline inflation minus regulated prices and changes in indirect taxes. In Hungary, the monetary policy focused on managing the nominal exchange rate until 2001 within a fairly narrow crawling band. During 2001, the National Bank of Hungary (NBH) widened the band from +/- 2.25% to +/-15%. In the summer of 2001, there was a shift from a crawling band to a horizontal band to reach the criteria of the Exchange Rate Mechanism (ERM II). At the same time they adopted inflation targeting (one percent tolerance band) as a monetary policy. Hungary kept a pegged exchange rate longer than other comparable countries, this is until February

15 Table 2: Hungary: official monetary and exchange rate regime Hungary Official Exchange Official Monetary Policy Arrangements Time interval Type Time interval Type Crawling peg, 70% Ecu, Exchange rate targeting 01/01/ /12/ % USD, Band: +/-2.25% Inflation targeting (CPI annual average)(yearend targets) 01/01/ /12/ % DEM, 30% USD 01/01/ /04/ % EUR /05/ /09/2001 Band: +/- 15% Inflation targeting Horizontal Peg to EUR, (medium-term targets) 01/10/ /02/2008 Band: +/- 15% 26/02/ Managed float Source: Frömmel, Garabedian and Schobert (2011), Frömmel and Van Gysegem (2011), IMF: Annual Report of Exchange rate Arrangements and Restrictions (2012), website Central Bank of Hungary The National bank of Poland (NBP) tried to combine money and exchange rate targets until 1998 by following a crawling devaluation of the zloty against a basket of currencies (USD, DEM, GBP, FRF and CHF). It enabled a reduction in inflation but it did not result in both intermediate targets being met in full. In the beginning of 1995, Poland experimented with targeting money growth, interest rate, and the monetary base. Next, it widened his band from +/- 1% to +/- 10% in In the last quarter of 1998, direct inflation targeting was introduced (formally introduced at the start of 1999) as the result of increased financial market integration. The Polish zloty became freely floating in The exchange rate didn t play a major role in monetary policy until 2010 as the NBP did not intervene on the foreign exchange market. After the NBP tried to talk the zloty down to no avail by more than a month, in April 2010 the NBP was forced to intervene, as the strong zloty was considered to be a threat for the economic growth in Poland. 5

16 Table 3: Poland: official monetary and exchange rate regime Poland Official Exchange Official Monetary Policy Arrangements Time interval Type Time interval Type Crawling peg, 45% USD, Exchange rate and 35% DEM, 10% GBP, 5% 01/01/ /05/1995 monetary targeting FRF, 5% CHF, Band: +/- 1% 16/05/ /02/1998 Band: +/- 7% Inflation targeting (End 25/02/ /12/1998 Band: +/-10% of the year CPI inflation) 01/01/ /04/ % USD, 55% EUR 12/04/ Free float Source: Frömmel, Garabedian and Schobert (2011), Frömmel and Van Gysegem (2011), IMF: Annual Report of Exchange rate Arrangements and Restrictions (2012), Annual Report NBP (2012) While the National bank of Romania (NBR) has not made any official commitments to a certain monetary policy strategy, statements of officials from the NBR openly express that their monetary policy is aimed at a certain exchange rate path to maintain external competitiveness. In discussions with the IMF, Romanian authorities have mentioned that they use inflation targeting and exchange rate targeting as monetary policy instruments for preserving a sustainable external position (IMF 2003). Table 4: Romania: official monetary and exchange rate regime Romania Official Exchange Official Monetary Policy Arrangements Time interval Type Time interval Type /2005 No official commitments to a monetary policy strategy 01/01/ Managed float 8/ Exchange rate and inflation targeting Source: Frömmel, Garabedian and Schobert (2011), Frömmel and Van Gysegem (2011), IMF: Annual Report of Exchange rate Arrangements and Restrictions (2012), website National Bank of Romania 6

17 Russia is one of the CIS countries (Commonwealth of Independent States). The CIS is an association of countries that were former Soviet Republics, formed during the breakup of the Soviet Union. In the beginning of the 1990s, Russia chose a conventional peg because it could use the exchange rate as a nominal anchor for macroeconomic stabilization. The CIS countries had experienced strong inflation and subsequently made price stabilization their main objective. Since 1999, Russia opted for a flexible regime, i.e. a managed floating exchange rate regime. It was intended to gradually decrease the influence of the Central Bank of Russia on the exchange rate dynamics and to create possibility for the shift to a floating exchange rate regime. In 2005, Russia introduced a dual-currency basket as the operational indicator of its exchange rate policy. As the scale of external operations of the Russian economy nominated in euro increased and the euro segment of domestic foreign exchange market expanded, the weight against the euro in the dual-currency basket was gradually growing. Hence, in 2007 the weight against the euro was augmented to 45% and the weight against the USD was set with a remaining 55%. In the end of 2008 and beginning of 2009, the Russian economy faced external shock of a large scale induced by sharp changes of the situation on global financial and commodities markets. Therefore, Russia modified its exchange rate policy framework. In February 2009, Russia shifted towards a floating band with a fixed width of 2 ruble. Since then the width of the floating operational band was gradually increased to ensure shift to a more flexible exchange rate. In 2010, the Bank announced that the fixed band for the ruble value of the dual-currency basket was abandoned and shifted towards a managed floating regime. 2 2 Source: website Central Bank of Russia, < < (3/05/2014) 7

18 Table 5: Russian Republic: official monetary and exchange rate regime Russian Republic Official Exchange Official Monetary Policy Arrangements Time interval Type Time interval Type Inflation targeting Crawling bands (a) Exchange rate targeting 01/ /2005 Managed floating 2001 Fund supported or other Basket peg, 90% USD, 02/ /2007 monetary program 10% EUR (b) Other (no explicitly Basket peg, 55% USD, 02/ /2009 stated nominal anchor) 45% EUR Exchange rate targeting (against a composite) 02/ /2010 Crawling band Other (no explicitly stated nominal anchor) 10/2010-present Managed floating Source: (Dabrowski, 2013), IMF: Annual Report of Exchange rate Arrangements and Restrictions (2012), Annual Report CBR (2012), website Central Bank of Russia (a) Crawling pegs combined with bands of more than +/- 1 per cent (b) The weights in the basket evolved gradually during these years: 02/ /2005: 90% USD/10% EUR, 03/ /2005: 8%0 USD/20% EUR, 05/ /2005: 70% USD/30% EUR, 08/ /2005: 65% USD/35% EUR, 12/ /2007: 60% USD/40% EUR From the beginning Slovakia followed an exchange rate targeting policy combined with a monetary targeting policy (tracking a broad money aggregate, namely M2). In 1997, Slovakia widened its exchange rate band from +/-1,5% to +/-7%. In 1998, the National Bank of Slovakia (NBS) decided to reform it to a managed float. From 1998 until 2008 the NBS has used informal inflation targeting as an instrument for monetary policy. Nevertheless the exchange rate is officially floating; the exchange rate still plays an important role for the National Bank of Slovakia. Since 01/01/2009 the Slovak Republic is participating in the euro area. 8

19 Table 6: Slovakia: official monetary and exchange rate regime Slovakia Official Exchange Official Monetary Policy Arrangements Time interval Type Time interval Type Exchange rate and Basket peg, 60% DEM, 01/01/ /12/1996 monetary targeting 40% USD, Band : +/-1.5% Informal inflation and exchange rate targeting 01/01/ /09/1998 Band : +/- 7% 11/ /2008 ERM-II 01/10/ /11/2005 Managed float 25/11/ /12/2008 ERM-II 2009-present Euro system Euro area member (free 01/01/2009-present floating) Source: Frömmel, Garabedian and Schobert (2011), Frömmel and Van Gysegem (2011), IMF: Annual Report of Exchange rate Arrangements and Restrictions (2012), website National Bank of Slovakia The official policy of Slovenia was monetary targeting until switching to a two-pillar strategy comparable to that of the European Central Bank in The bank of Slovenia states that it pursues the core aim of monetary policy, namely price stability, by simultaneously modifying the quantity of money in circulation and the exchange rate. In order to adjust interest rates and the exchange rate interdependently (Annual Reports, Bank of Slovenia, Annual Report 2002, p 23). Frömmel and Schobert (2006) state that it would not need to adjust exchange rate developments in reaction to capital flows, if the Bank only wants to track price stability as its primary objective in a managed floating exchange rate regime. These statements exposes implicit dual objectives, namely internal price stability and exchange rate targeting. Slovenia applied the euro as its currency since 01/01/

20 Table 7: Slovenia: official monetary and exchange rate regime Slovenia Official Exchange Official Monetary Policy Arrangements Time interval Type Time interval Type Base money targeting 01/01/ /06/2004 Managed float 1996 Base money and M1- targeting 27/06/ /12/2006 ERM-II M3-targeting Exchange rate and Since 01/01/2007- monetary targeting present 06/ /2006 ERM-II Euro area member 2007-present Euro system Source: Frömmel, Garabedian and Schobert (2011), Frömmel and Van Gysegem (2011), IMF: Annual Report of Exchange rate Arrangements and Restrictions (2012), Annual Report BSI (2012) As Frömmel and Van Gysegem (2011) mention, we can conclude from this overview of the de jure exchange rate and monetary policy regimes for the different countries, that the countries mostly opted for a fixed exchange rate as a stabilization strategy after the dissolution of the Soviet Union in December Only Romania and Slovenia adopted a more flexible regime in the beginning. Frömmel and Van Gysegem cite two reasons put forth by different authors in explaining why countries are not willing to let their currency float freely. The first reason could be that small open economies are very sensitive to exchange rate movements; for that reason monetary authorities are obliged to take the exchange rate into account in their monetary policy (Ball L., 1999). Subsequently, emerging and transition countries, such as the Central and Eastern European countries from above, do not yet have developed financial markets. Therefore they don t allow domestic firms to borrow in their home currency and their debt is chosen in foreign currency. As a result they will have the tendency to peg their exchange rates as Hausmann, Panizza and Stein (2001) argue. But Frömmel and Van Gysegem (2011) also mention the argument of Obstfeld and Rogoff (1995) that it is not always easy to constrain to a fixed official exchange rate. As credibility is very difficult to develop and uphold for a government, a pegged exchange rate can ensure to lose a governments credibility and be very costly when it does not meet its promises. Due to the large global capital markets any weakness of a country concerning its commitment to a fixed exchange rate is enlarged and does not give a country the chance to do something about it. (Obstfeld & Rogoff, 1995). In emerging and 10

21 transition countries the political backing for the necessary but unpopular measures to defend the peg may be very low. Moreover, under an officially floating regime, adjustments of the exchange rate are less visible to the public and less costly than devaluations under official peg. At the end of the 90s many countries choose a more floating regime, except for Romania and Slovenia that opted for a managed float from in the beginning. The development that most of the countries followed, i.e. first opting for a pegged exchange rate in the beginning after the transition and afterwards following a more flexible regime, can be explained by the two-corner hypothesis (infra, p. 13). We must stress the fact that the overview above is a de jure list of the exchange rate regimes of the CEE C s. This means that the regime they actually practice can differ from what they are officially saying they will follow (Frömmel & Van Gysegem, 2011). An explanation for this tendency will be given in the next paragraph. 11

22 2 Literature on the fear of floating Member-countries of the International Monetary Fund (IMF) are asked to publish their official exchange rate in the IMF s Annual Report on Exchange Rate Arrangements and Exchange Restrictions. We have established these official regimes in the previous paragraph for the seven emerging countries we will examine in this piece. In general there exist four regimes that a country can follow: fixed or peg, limited flexibility, managed floating and independently floating regime. This categorization can be very misleading. Cases where countries diverge from their official announced exchange rate regimes are very common in emerging countries. This is also illustrated by the path of the studied countries exchange rate regimes in this work. Reinhart and Rogoff (2004) point out that this issue is of great importance to have a correct view on the performance of different regimes. The authors emphasize that due to the differences between de jure and de facto exchange rate regimes, wrong conclusions have been drawn in the past, when considering the official history of exchange rates (Reinhart & Rogoff, 2004). This paragraph represents the theoretical background which is necessary to understand more the writings and analysis in this piece. Furthermore, this paragraph gives an overview of possible explanations from the literature why official currency regimes can differ from the actual practiced currency regimes. We will start by discussing different monetary policies and afterwards amplify on the two-corner hypothesis. 2.1 Monetary policy interventions The occurrence of fear of floating can be a possible reason for the fact that countries actually follow a different regime than they are saying they will follow. This term has been extensively used in the literature and can be simply explained as countries that say that they allow their currency to float freely but in practice do not, in other words: de facto pegging. To go further on this theme, it is relevant to emphasize the distinction between developed and emerging countries. Countries, particularly emerging countries, are afraid of large currency swings in response to shocks, as Calvo and Reinhart (2002) discuss. In this case the monetary policy of a country will respond to these exchange market pressures and influence its currency by using the interest rate or an intervention policy. To test whether the official announced exchange rate corresponds with the applied exchange rate by the countries, Calvo and Reinhart (2002) compare the behavior of exchange rates with changes in foreign exchange reserves and interest 12

23 rates. 3 The authors cite various reasons why countries are not so willing to let their currency follow much variations, but then focus in a simple model on the combination of lack of credibility, high pass-through from exchange rates to prices and inflation targeting. Their findings have led to an answer to the question why there were much unsuccessful studies in finding out important trends in economic growth, inflation or volatility, while applying official classifications of exchange rates (Calvo & Reinhart, 2002). There exists a trend that the intervention in the domestic interest rate is preferred to indirectly interfere with the exchange rate, as also Ball and Reyes (2004) discuss. But likewise adapting international reserves or intervening in the foreign exchange market are possible methods. To clarify the monetary actions of a country, it is important to distinguish policies as a result of a fear of floating (FF) from an inflation targeting regime (IT). There exists an empirical distinction between FF and IT but one must take the other into account to analyze a country s regime. Misclassification of a country s monetary policy could cause a loss of credibility and a higher risk premium (Ball & Reyes, 2004). A short and clear definition for IT is given by Pourroy (2012) as a monetary framework whose nominal anchor is based on price stability (Pourroy, 2012). The exchange rate has an indirect impact on price behavior. Thus, movements in the exchange rate will be taken indirectly into account when the monetary policy wants to play down inflation. This shows that countries are not eager to let their currency float without any control, because the nominal exchange rate has an influence on inflation and the real exchange rate has an effect on the wealth of domestic citizens and the allocation of resources (Fischer, 2001). Therefore, a stable exchange rate gives fragile central banks the opportunity to increase their credibility with the public of their commitment of price stability (von Hagen & Zhou, 2005). Furthermore, there are still other arguments that give an explanation to the phenomenon of fear of floating. Evidence for the reluctance of emerging countries to float freely is given by Lahiri and Végh (2001). The authors analyze developing countries, which are registered to have a floating framework and undergo larger shocks, have lower exchange rate variability and higher reserve variability in comparison with developed countries, which also have a freely 3 In the literature on the methodology -section in this work (infra, p. 16) we will call the changes in the exchange rate, reserves and interest rates the Calvo and Reinhart criteria. 13

24 floating currency. The difference between fear of floating and the opposite phenomenon, i.e. fear of pegging, is handled by Von Hagen and Zhou (2004). They explain fear of pegging by a country that officially declared a fixed rate but changes the currency so frequently that it is not possible anymore for outsiders to differentiate it from a floating regime. Fear of pegging is typical for countries that want to have capital control and dispose of a more advanced financial system (von Hagen & Zhou, 2004). 2.2 Two-corner hypothesis Because of the increase in international capital mobility, adaptable pegs are costly to preserve. Fischer (2001) states the evolution that takes place in the 1990s -after several crises- of the decline of developing market economies that execute intermediate exchange rate arrangements, according to the official classification of the IMF. The trend of economies that are setting their boarders open for international capital flows, is to move either to a more fixed exchange rate or greater exchange rate flexibility. This is called the hypothesis of the vanishing intermediate regimes, bipolar view or two-corner hypothesis. The lack of an explanation for this trend or analytical rationale is stressed by Frankel, Fajnzylber, Schmukler and Servén (2001). Fischer (2001) provides a simple answer: The reason is that soft peg systems have not proved viable over any lengthy period, especially for countries integrated or integrating into the international capital markets. (Fischer, 2001, p. 9-10), but cannot explain why intermediate regimes don t operate successfully in the long term. To further analyze vanishing intermediate regimes the author takes into account the difference between countries open for international capital flows and countries that are not open. Because, in accordance with Fischer (2001), the statement of the bipolar view must be reconsidered. A country that disposes of capital mobility is not an adequate candidate for operating soft exchange rate pegs but still has the choice between different flexible rate arrangements and it is foreseen that the monetary policy of a country will intervene. This last notion is in accordance with the work above on the fear of floating (Calvo & Reinhart, 2002). For countries with a lack of international capital flows all existing exchange rate regimes are sustainable but countries with great capital mobility are confined to very fixed pegs or floating rates (including managed floats). Therefore this nuance makes clear that the two-corner hypothesis does not exclude every exchange rate arrangement but the two corners (Fischer, 2001). 14

25 It is valuable to point out the concept of impossible trinity as a possible alternative reason for the disappearance of intermediate regimes. This concept refers to the impossibility of an economy to have a fixed exchange rate, as well as capital mobility i.e. financial market integration, and monetary independence. One must give up one of the three (Frankel, Fajnzylber, Schmukler, & Servén, 2001) (Fischer, 2001). A soft peg can be seen as a result from a country, open for capital flows, that tries to attain having a monetary policy directed to national goals and a fixed exchange rate. This attempt is the cause of many conflicts between both goals and is not efficient. But this still leads to various questions about the concept and it is not correct to presume the truth of the impossible trinity without having critical doubts (Fischer, 2001). A correct representation of the actual exerted exchange rate regimes is important to make credibility and transparency possible for the market participants. Frankel et al. (2001) discuss the importance of central banks to create credibility through verifiability. In a world of high capital mobility, emerging countries are obliged to the corner point regimes, i.e. the extremes of free floating exchange rates or hard pegs. The authors state that the bipolar view is caused by the necessity of credibility with the public. Easily verifiable regimes, i.e. corner point regimes, can enable market agents to observe better the government s actions. As a result, this can reduce uncertainty and increase consumption and investments. The notion of verifiability can be seen as a justification for the two-corner hypothesis. Moreover it gives the object of this paper a defined purpose and stresses the importance of the credibility of market participants in the announcement of the authority that the published exchange rate policy will be the one they truly operate (Frankel et al., 2001). Unlike the arguments about the corner point regimes mentioned above, Masson (2001) concludes that the trend of exchange rate regimes that undergo only transitions towards the two poles can be rejected in the long term. With the use of estimated transition matrices that modify the official classification, he determines that intermediate regimes will continue to exist among the exchange rate regimes (Masson, 2001). Moreover Calvo and Reinhart (2002) conclude that it is not possible to generalize the hollowing-out hypothesis. Their findings present that the middle is not disappeared but this tendency can be explained by the fact that earlier studies have focused only on the official exchange rate regimes classifications, which leads to wrong conclusions. Besides the authors state the following: In sum, economic theory provides us with well-defined distinctions 15

26 between fixed and flexible exchange rate regimes, but we are not aware of any criteria that allow us to discriminate as to when a managed float starts to look like soft peg. (Calvo and Reinhart, 2002, p. 405). In the literature a lot of evidence has been found that de facto en de jure exchange rate regimes commonly differ in transition economies. The question remains why there exists a real difference between the official announced exchange rate and the actual practiced currency in transition economies. Lahiri and Végh (2001) suggest that the existence of different de jure and de facto rates can be explained by the trade-off between the cost of foreign exchange market intervention and real output losses due to exchange rate volatility (Lahiri & Végh, 2001). Moreover dissynchronization in regime adjustments is mentioned by von Hagen and Zhou (2005) and is related with regime discrepancies. Because changing the announced exchange rate implicates costs for the government of the country (e.g. loss of credibility), compared with a costless adjustment of a de facto regime that is set by the authority without any commitment. Therefore official regimes are changed less frequently (but are larger) than de facto regimes. Besides a reduction in costs, the authority uses the de facto rates to deal earlier and more quickly with macroeconomic developments than when they would adapt their currencies by official announcements. In this work the authors provide us empirical evidence for these discrepancies. Their discrete choice model shows that authorities are tempted to change their reported exchange rate when the chosen regime is not effective and they favor a more flexible (fear of pegging) or more rigid (fear of floating) regime. Additionally, because countries value their reputation and they fear becoming exposed to speculative attacks, de facto rates are more appealing (von Hagen & Zhou, 2005). 16

27 3 Literature on the methodology In recent literature, authors have acknowledged the importance of identifying de facto exchange rate regimes so that they can classify exchange rate activities in an accurate way. As a result there exist a lot of methodologies used by different researchers to take this issue into account. Furthermore, because of the numerous variants of exchange rate regimes that exist between a simple peg and a pure floating currency, we need to have available a method that considers both the flexibility criteria as well as the consequences of a tight basket peg at the same time. For this reason we first discuss the inferring flexibility method, thereafter the weight-inference technique and cogently we provide a synthesis of the two techniques. Lastly, we will go deeply into the method we will practice in this work for the study of the exchange rate regimes for the seven Central and Eastern European countries. 3.1 Inferring flexibility In this section, methodologies on identifying the true flexibility of a currency are listed and they are appropriate for countries following a more flexible exchange rate arrangement. Schnabl (2004) discusses the accuracy of the IMF classifications of the exchange rates for Central and Eastern European countries. He concludes that the de facto stabilization against the euro is much more present than the IMF de jure classifications indicate for Central and Eastern Europe. First, in Schnabl s attempt to test the de facto exchange rate stabilization, the author uses monthly data (low-frequency exchange rate stability) for the Calvo and Reinhart criteria (Calvo & Reinhart, 2002). These criteria are based on the development of changes of the exchange rate in association with policy variables, i.e. changes of official foreign reserves and changes in nominal short-term interest rates by the central bank of a certain country. Secondly, Schnabl uses daily data (high-frequency exchange rate stability) that could give evidence to the daily attempts of central banks to adjust the exchange rate in their favor. Moreover the author measures the exchange variability against possible anchor currencies such as the dollar or the euro. To quantify his results the author sets a standard on exchange rate stability by making use of probability limits. With the help of these results of high or low probability, conclusions about the policy of the government can be appointed. Here too, Schnabl (2004) follows the approach of Calvo and Reinhart (2002), what Calvo and Reinhart (2002) claim is more prompt than 17

28 making conclusions based on the interpretation of standard deviations, because it is possible to elude distortions caused by outliers. Schnabl (2004) uses the standard deviations as a secure second indicator (Schnabl, 2004). In the work of Hausmann, Panizza and Stein (2001), the authors similarly examine the link between exchange rates and monetary policy. Moreover they focus on the management of exchange rates under a floating regime and their differences in their behavior among a broad sample of countries. The first factor discussed is the level of reserves relative to M2 4 each country applies. A large cushion of reserves makes it possible to manage the exchange rate. The authors display the percent of reserves in comparison with the M2 of every country. The bigger the percent of reserves, the more this can be an indication for exchange rate management by using the stock of reserves. Furthermore, the paper observes two variables to point out the degree to which economic authorities let their currency float freely. The first variable is RVER, which stands for the Relative Volatilities of the Exchange rate and Reserves to provide an answer to the question in which expanse countries try to stabilize their exchange rates by intervening in the foreign exchange market. Here for they adopt relative volatilities because comparing exchange rate volatilities may not offer us an entirely overview of the willingness of a country to secure its parity. The exchange rate of a certain country can be more unstable because it is the victim of larger external shocks than the exchange rate of another certain country, even if both countries are trying to stabilize their exchange rate through their monetary policy. The same issue can be applied for the reserves of each country. Therefore, it is better to use relative figures than absolute ones. The formula for RVER according to Hausmann et al. (2001) which acts as a measure of the extent in which countries intervene in the foreign exchange markets by using their reserves, is given by: = () (!" $%&') (1) 4 M2 stands for the secondary monetary supply. It consist of M1 and close substitutes of M1. M2 is an important economic indicator to make predictions concerning inflation. 18

29 The measure of exchange rate volatility is rendered by the standard deviation of the rate of depreciation (DEP). 5 As a measure of the volatility of reserves, they use the standard deviation of the stock of reserves (RES), normalized by the dollar value of the stock of broad money (M2). The authors apply the average value of M2 (AvM2) over the period under consideration, in order to prevent changes in the exchange rate from affecting the measured volatility of reserves through the dollar value of M2. Subsequently, the monetary authorities have another tool at their disposal to intervene in the flotation of their currency than intervening in the foreign exchange market, i.e. tightening or loosening their monetary policy. The second indicator consists of the Relative Volatility of Exchange rates and Interest rates. For this, the authors use the formula: ( = () () (2) For the indicator of the interest rate (), the authors again use the standard deviation of the rate of depreciation for the same reason as already mentioned above. Another and last indicator is observed by Hausmann et al. (2001) in addition to the three already mentioned indicators for exchange rate flexibility. They use factor analysis to figure an overall index of exchange rate flexibility (FLEX). The authors describe the correlation with the reserves over M2 ratio and the indices RVER and RVEI (Hausmann, Panizza, & Stein, 2001). Similarly to the work of Calvo and Reinhart (2002), Schnabl (2004) and Hausmann et al. (2001), Levy-Yeyati and Sturzenegger (2005) construct a classification scheme to discern the true de facto exchange rate of a country. The authors define three classification variables: exchange rate volatility, volatility of exchange rate changes and the volatility of international reserves. They do not take the interest rate into account for several reasons. They explain that the interest rate policy is implicitly considered in their analysis for small open economies through the changes that unsterilized reserve flows infer on monetary aggregates. The authors decide that given the frequency of their data, they can abstract from the interest rate variable without 5 Hausmann et al. (2001) prefer to use the standard deviation of the rate of depreciation as a measure to the volatility of the exchange rate. For the reason that it is possible that an implicit objective for the intervention of a country may be to achieve a crawling peg with a fairly constant rate of crawl, rather than simply to keep the exchange rate at a certain level. 19

30 losing any significance on their results. With the help of the actions of these three variables they can define the regime of a country that it is actually following at any point in time. A fixed exchange rate regime is associated with changes in international reserves aimed at reducing the volatility in the nominal exchange rate, and flexible regimes are associated by substantial volatility in nominal rates with relatively stable reserves. A crawling peg can be explained by active interventions that keep the exchange rate along a certain path, with the help of stable incremental changes in the nominal exchange rates (i.e. low volatility in the rate of change of the exchange rate). Lastly, a dirty float is clarified by a relatively high volatility across all variables, with interventions only partially smoothing exchange rate fluctuations. The authors state different clarifications for the differences between de jure and de facto exchange rate regimes that came up in the literature. 6 A first reason could be the end of the Bretton Woods system that was followed by a trend of applying floating regimes by many countries. Secondly, the hollowing-out hypothesis could be a possible explanation and thirdly the fear of floating phenomenon (supra, p. 11). Levy-Yeyati and Sturzenegger (2005) plot the Exchange rate volatility ( ) as the average of the absolute monthly percentage changes in the nominal exchange rate during a calendar year. The volatility of exchange rate changes ( ) is given by the standard deviation of the monthly percentage changes in the exchange rate. The volatility of reserves ( ) is more complex to measure. The authors use the formula (3) to estimate, as closely as possible, the change in reserves that reflects intervention in the foreign exchange market.the net reserves in dollar are: = )*+,-$./)*+, /4-+135*%.7*.. (3) In equation (3) stands for the price of the dollar in terms of the local currency. Their measure of volatility is the average of the absolute monthly change in, i.e. the average of the absolute monthly change in net dollar international reserves relative to the monetary 6 Levy-Yeyati and Sturzenegger (2005) call these clarifications stylized facts. 20

31 base in the previous month, also in dollars. Levy-Yeyati and Sturzenegger (2005) plot the equation as =!. /!.89 &*-1+:; =! &*-1+:; (4) The authors define the classification variables for 183 countries that reported their exchange rate regime to the IMF for the period The authors use cluster analysis to form groups of relevant exchange rate regimes based on the similarity of the behavior of the three classification variables, independently of the official regime indicated by the country that is assigned to this group. With their classification scheme they want to give the future work of others a starting point, which relies on facts rather than on legal characteristics of the regime of a country. The authors believe that this de facto classification can clarify stylized facts reported in previous literature. When they reconsider these stylized facts with the help of their de facto classification scheme, they conclude that de facto pegs have remained stable throughout the last decade, although there are a lot that apply the phenomenon of hidden pegs (i.e. official fixed regimes that move away from their commitment without admitting it). This is in contrary with the post-bretton Woods reasoning. Secondly, they can recover the tendency of the vanishing intermediate regimes but conclude that this does not apply for countries with limited access to capital markets. Finally, they discover that pure floats are associated with relatively minor nominal exchange rate volatility and that there is a recent upturn in the amount of de jure floats. Nevertheless, at the same time there is also an upturn in the amount of de facto dirty floats (i.e. fear of floating) (Levy-Yeyati& Sturzenegger, 2005). An alternative essential work on classifications schemes is that of Reinhart and Rogoff (2004). The conclusions of Levy-Yeyati and Sturzenegger correspond mainly with the results of Reinhart and Rogoff (2004). Reinhart en Rogoff (2004) employ in their working paper an extensive database on market-determined parallel exchange rates with the use of a natural classification algorithm. The authors assign a difference between the official or standard categorization and the true underlying monetary regime with the help of this new classification scheme. They illustrate that the market-based exchange rate is a better indicator of the monetary policy position of a certain country than the official rate. The authors use monthly data on official and market-determined exchange rates for the period , unlike the official classification that assumes the same regime for the whole year. Their algorithm depends on a broad variety 21