Central Banks' Communication Strategy and Content ... - Science Direct

Available online at www.sciencedirect.com

ScienceDirect Procedia - Social and Behavioral Sciences 235 (2016) 618 – 629

12th International Strategic Management Conference, ISMC 2016, 28-30 October 2016, Antalya, Turkey

Central banks’ communication strategy and content analysis of monetary policy statements: The case of Fed, ECB and CBRT Eyup Kahvecia, , Aysun Odabaşb a,b The Central Bank of the Republic of Turkey, Ankara,06050, Turkey

Abstract Prior to the 1990s, it was believed that Central Banks should be in mystery giving away little on the policies they were following. However with time, the importance of transparency and working with, rather than against, markets began to be understood. This has become especially important after central banks began to implement inflation targeting regime and communication strategy has turned out to be a major tool for central banks. In this framework, central banks began to use different communication channels one of which is the monetary policy statements after policy decisions. Our work offers a semantic examination of monetary policy statements from the US Federal Reserve, the European Central Bank and the Central Bank of the Republic of Turkey (Fed, ECB and CBRT) in order to observe the change in tone of monetary policy statements with movements toward greater transparency. In the study, we examine the monetary policy statements of three central banks in the pre-crisis years of 2002 to the subsequent period of 2015 and compare the specifics of the language used both pre- and post-crisis by making use of Diction 7 software. The results show that optimistic tone of Fed has decreased over the period we studied while certainty tone has increased. As for ECB and CBRT, there is no significant tone difference in certainty, optimism and the realism over time. However, for the CBRT’s last two years statements, there is a significant increase of optimistic tone.

2016 The Authors. Published by Elsevier . article under the CC BY-NC-ND license ©©2016 Published by Elsevier Ltd. This is an openLtd access (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of ISMC 2016. Peer-review under responsibility of the organizing committee of ISMC 2016. Keywords: Central bank communication strategy, Monetary policy strategy, Global crisis, Crisis communication, Content analysis, Diction 7

1. Introduction Prior to the 1990s, it was believed that Central Banks should be inscrutable bodies that communicating sparingly and giving away little on the policies they were following. It was thought that monetary policy-makers should speak as little as possible and deliver coded messages when they did so. Over time this has given way to the view that effective monetary policy is possible through transparency and effective communication and later to the idea that the management of expectations is fundamental to monetary policy. Effective communication is essential to expectation management. The post-crisis transition in Central Bank policy from mystery and inscrutability to the era of

Corresponding author. Tel. + 90-312-507-7686

Email address: [email protected]

1877-0428 © 2016 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of ISMC 2016. doi:10.1016/j.sbspro.2016.11.039

Eyup Kahveci and Aysun Odabaş / Procedia - Social and Behavioral Sciences 235 (2016) 618 – 629

transparency and the importance of “forward guidance” is a demonstration of how rapidly and radically communication strategy has changed. Now communication is accepted both in academic literature and in real-world practice as central to the effectiveness of monetary policy. This work offers a semantic examination of monetary policy statements from the US Federal Reserve, the European Central Bank and the Central Bank of the Republic of Turkey (Fed, ECB and CBRT) in the pre-crisis years of 2002 to the subsequent period of 2015 and compares the specifics of the language used both pre- and post-crisis. It attempts to observe the change in tone of monetary policy statements from central banks alongside the recent increase in transparency and to identify whether there has been any shift in the tone used to describe both the current states of monetary policy and the economy and the future. In this article, we study the tone and diction, or word choice, of the monetary policy statements to better understand how the explanations about the economy and the monetary policy stance formed, how they related to the performance of the economy, and how they may have changed with movements toward greater transparency. This study differs from previous work on the subject both in terms of the period and central banks under examination, and in terms of method, by making use of Diction 7 software for content analysis since the content analysis have been historically less used as techniques in economics. This is especially the case with respect to research undertaken inside central banks. We found that optimistic tone of Fed has decreased over time while certainty tone has increased. There is no significant tone difference in certainty, optimism and the realism of ECB and CBRT. However, for the CBRT’s last two years statements there is a significant increase of optimistic tone. 2. The Development of Central Bank Communication Strategy While in 1987 Greenspan stated that “Since I’ve become a central banker, I’ve learned to mumble with great incoherence. If I seem unduly clear to you, you must have misunderstood what I said.”1, he stated just the opposite in 2001: “Openness is an obligation of a central bank in a free and democratic society.” (Greenspan, 2001). This shows how the communication strategy of the central banks has evolved over time. Prior to the 1990s, it was hard to predict monetary policy decisions because of the multiple aims and instruments involved in monetary policy. For this reason, it was thought that the best option was for central banks to refrain from speaking about policy decisions and measures. Without a clear policy framework or a well-defined criterion on which to judge central bank performance, central banks were less open to public accountability (Vayid, 2013; 1). During this period, it was thought that monetary policy-makers should speak as little as possible and deliver coded messages when they did so (Blinder et al., 2008; 2). Furthermore, it was believed that achieving an effective monetary policy necessitated surprising financial markets. Central banking during this period can be defined by three words: mystery, secrecy and opacity. While Mishkin (2004) attributes this situation to the desire of the bureaucracy to maximize power and prestige, and avoid accountability, Kydland and Prescott (1977) and Calvo (1978) stress “Time-Inconsistency Theory” (Mishkin, 2004). This approach sees central bank secrecy as a result of the pressure for excessively accommodative monetary placed on policy-makers by politicians, whose political time horizons are shorter than the optimum time horizon demanded by monetary policy. In other words, politicians seek implementation of short-term popular policies, while monetary policy-makers work to avoid popular policies to focus on steps to achieve long-term stability. According to Chant (2013), the secrecy surrounding monetary policy was the result of fixed exchange rate strategies that amount to “the enemy of openness”. In these terms, the shift among central banks towards floating exchange rates from fixed-rate strategies was equivalent to the lifting of a major obstacle to transparency. Brunner (1981) describes the mystique surrounding central banking in the following terms: “Central Banking thrives on a pervasive impression that it is an esoteric art. (…) The esoteric nature of the art is revealed by an inherent impossibility to articulate its insights in explicit and intelligible words and sentences” (Brunner, 1981; 5).Alan Blinder said in a speech in 1996 to the London School of Economics that… “greater openness might actually improve the efficiency of monetary policy… [because] expectations about future central bank behavior provide the essential link between short rates and long rates. A more open central bank… naturally conditions expectations by providing the markets with more information about its own view of the fundamental factors guiding monetary policy…, thereby creating a virtuous circle. By

1

Alan Greenspan as quoted in the Wall Street Journal, September 22, 1987.

619

620


making itself more predictable to the markets, the central bank makes market reactions to monetary policy more predictable to itself.” (Blinder et al., 2008; 2) Work by Fry et al. in 2000 showed that 74 percent of central banks regarded transparency as either “crucial” or “very important” to the monetary policy framework (Fermo, 2012; 1). Five years after Alan Blinder, Woodford told central bankers at assembled at the Jackson Hole conference that; “successful monetary policy is not so much a matter of effective control of overnight interest rates… as of affecting… the evolution of market expectations... [Therefore,] transparency is valuable for the effective conduct of monetary policy… this view has become increasingly widespread among central bankers over the past decade.” (Blinder et al., 2008) There has been a rapid shift in central bank communication, from 1981 when Brunner was able to say that central banks should not communicate, through 1996 when Blinder said more communication would increase policy effectiveness, to 2001 and Woodford's comments on the centrality of expectations management, and finally to the latest period in which post-crisis central banks have begun to implement policies of “forward guidance” that show the importance of communication (Yilmaz & Kahveci, 2014). It is now accepted in academic literature and in practice that communication has an important role to play in effective monetary policy. 3. Communication Strategy in Inflation-Targeting Throughout the 1980s research increased on the question of how central banks can set clear monetary policy goals and pre-commit with confidence to achieving them. With time, the importance of transparency and working with, rather than against, markets began to be understood (Vayid, 2013; 2). These developments pushed central banks to seek clearer policy frameworks and eventually led to the development of a new monetary policy strategy, the inflationtargeting framework based on a single instrument (the policy rate) and a single goal (price stability). In 1990 the Reserve Bank of New Zealand became the first central bank to implement an inflation-targeting strategy and put a strong emphasis on transparency. Central banks generally became relatively transparent on their monetary policy goals and engaged in sustained and regular communication with the public. This represented an important step in resolving the problem of time inconsistency. The inflation-targeting strategy necessitated giving central banks the task of ensuring price stability and giving them the independence to achieve that duty. That independence brought with it a burden of accountability and caused central banks to become more transparent in their policy actions in order to achieve democratic legitimacy in the eyes of the public (Yilmaz & Kahveci, 2014; 5). What decisions were taken, and the reasons behind the decisions became a major aspect of central bank communication strategy. As a result, communication strategy became a major tool for central banks and efforts were made to increase its effectiveness (Blinder et al, 2008; 4). Under an inflation-targeting strategy, communication becomes crucial in bringing the expectations of economic actors closer in time with the goals announced by the central bank. Once the public understands a central bank's “reaction function” more clearly, their expectations for the course of the monetary policy can be better managed. In fact, just as monetary policy increasingly became “the art of expectation management” (De Haan et al, 2007; 2), so communication strategy became a tool to shape expectations for long-term interest rates, although monetary policy controlled short-term rates (Vayid, 2013; 7). This means that central banks are not simply using interest rates to direct expectations, their ability to shape the economy also depends importantly on managing expectations for the long-term trajectory of interest rates. Alongside accountability, it is also very important for central bank credibility that they use an effective communication strategy to inform the public of monetary policy and that they implement it accordingly. Laidler points out that the value of transparency can quickly be lost, so communication must be an ongoing process. Additionally, if communication is seen as a means to enhance accountability, he says that communication must then be a two-way process (Vayid, 2013; 1). Another reason why inflation targeting demands transparency and effective communication is the fact that a forward-looking posture lies at the root of the policy's emergence. Under inflation targeting strategy, monetary policy is shaped not by current conditions, but by medium-term trends and medium-term goals. Additionally the strategy requires the public to understand that the impacts of monetary policy on the economy and on inflation take place with a lag. As a result, the majority of central banks that implement inflation-targeting strategy take another step towards transparency by periodically publishing an inflation report that shares with the public the aims and limitations of monetary policy, the horizons of the inflation target, the rationale behind that target, how it is to be achieved and reasons for any deviations from the target (Mishkin, 2004; 3).

621


Transparency, openness and communication have gained importance for inflation-targeting central banks. As time went on, they became approved not only by the inflation targeting banks, but by others too, and have become reflected in their policies. 4. Central Bank Communication and Effective Monetary Policy Central Banks’ communication strategy can be examined from two perspectives: the short-term and long-term management of expectations through predictable monetary policy. An effective communication strategy can involve giving signals of monetary policy tendencies prior to short-term monetary policy decisions. In this way market uncertainty before the decision falls, and the movements of markets are shaped by the signals provided. However, new developments that cause revisions to central bank assessments and actions that differ from the initial signals can also cause volatility on markets. For this reason the signaling has to be seen as a conditional undertaking (Issing, 2005, 70). Another way in which communication can increase the effectiveness of monetary policy is to assist with the formation of expectations of future monetary policy actions among financial market participants that are closer to the central banks aims. When the two elements act in harmony, monetary policy becomes more predictable and inspires more confidence. For this reason, the more central banks can use effective communication to provide a useful and clear path of the future trajectory of interest rates to market participants, the more impact that will have over long-term interest rates (Issing, 2005; 70-71). At this point it is important to specify the relationship the expectations hypothesis sees between short-term rates and long-term rates and the role expectations play. The hypothesis was first outlined by Fisher and was later developed by Hicks (1939). The hypothesis says that long-term rates are defined fundamentally by expectations for short-term rates and the risk premium.

(1)

In this formula Rnt is the rate on a bond at time t, Et rt +i is the expectation at time t for the rate at time t+i, and εt is an expression of the risk premium. The formula states that for a long-term bond, the rate on the bond is formed by the expected average of short-term rates and the risk premium. In this framework, monetary policy can both directly affect short-term rates, and also guide long-term rates through the expectations channel for future short-term rates.

Source: Naghadiyev, 2011, 20 Figure 1: Central Banks Communication Objectives

An effective communication strategy is crucial to expectations management both in terms of “creating news” by providing meaningful and necessary information, and also in terms of “reducing noise” by preventing misinformation.

622


Transparency combined with the two elements will allow communication strategy to achieve greater public understanding of central bank policies and increase the effectiveness of the policies through greater trust, as well as increasing confidence in central banks (Blinder et al., 2008; 4). With higher credibility and anchored inflation expectations, central banks can obtain not only lower and more stable inflation, but also less volatility of outcomes. Their performance can increase not only inflation, but also in outcomes (Miskin, 2004; 3). In terms of successful expectations management in central bank communication, the work of Sturm and De Haan (2011) on statements given by the European Central Bank governor at news conferences after a rates decision shows the statements contain information that is of assistance to predicting future policy actions. 5. Central Bank Communication Strategy Following the Global Crisis After the global financial and economic crisis that broke out in 2007-2009, central banks found themselves struggling to guide the economy and manage monetary policy through traditional tools. This period saw the banks seek to extricate their economies from crisis with as little damage as possible through non-traditional monetary policy tools and actions. Additionally communication policies and tools became just as important as these non-traditional implementations. The work of Naghadiyev (2011) on post-crisis communication at central banks shows that one of the factors that deepened the global crisis was the fact that central bank communication was not as effective as desired and pushed central banks to see that communication was just as important as the content of policies. In fact at times, communication could become a policy in its own right, instead of being a tool for monetary policy. In this framework, Nagadiyev argued that closer relations with the public and increased transparency should not be the final aim of communication and that the time had come for central banks to think about the institutionalization of communication (Nagadiyev, 2011, 4-8). Thus communication became more than just a means to transparency or accountability. Analysis indicates that increased transparency and reduced uncertainty as part of an effective communication strategy has an impact on the volatility of financial markets. Better understanding by the markets of monetary and financial stability policies leads to reduced uncertainty in the financial sector and influences the expectations and decisions of market players. If the markets have a clearer understanding of how the central banks will react to new data, the reaction of asset prices and bond yield can be made more in line with the aims of monetary policy. In addition, when financial markets are better informed and more homogeneous in the process of forming expectations, the growing transparency accelerates the harmonization of private sector expectations and reduces volatility (Knutter et al., 2011; 6).

Figure 2: Barclays 2008 Survey of Central Bank Communication


Knutter et al. (2011) examine the impact of central bank communication tools on financial markets and asset prices and show that the most effective in terms of influencing markets are the statements and press conferences that follow rate decisions. Those tools are used by market players to form expectations for the course of short- and long-term interest rates and to increase predictability particularly at times of uncertainty. The tools also appear to have a general positive impact on asset prices (Knutter et al., 2011; 10-15). On the other hand, according to Barclays 2008 survey on Central Bank Communication conducted in 2008 just before the collapse of Lehman Brothers which is the starting point of the global crisis, the most important tools are statements that follow changes to policy, or no changes to policy. The record of votes is third, the meeting minutes fourth, the tapes fifth, press conferences sixth, and research papers last (Figure 2). A similar survey in 2014 showed the minutes of monetary policy committee meetings and press conferences at the top of the list and the votes of individual members last (Figure 3).

Figure 3: Barclays 2014 Survey of Central Bank Communication

When we compare the two surveys of Barclays, we see that the information dissemination mechanisms that participants see as important have changed with the crisis. The 2008 survey reveals 77 percent of respondents calling “statements issued after a policy change” as indispensable, placing these announcements at the top of the list. When we come to 2014, top of the list is “meeting minutes”. In 2008 41 percent called such minutes essential. By 2014 that has risen to 56 percent. On the other hand, while the importance attributed to press conferences and meeting transcripts rose post-crisis, the number who saw member votes as important declined. Looking at the two surveys together, we can conclude that meeting minutes and press conferences, direct means of monetary policy communication, have risen in importance. In the framework of increasing central bank transparency, one issue that remains the subject of debate particularly for inflation-targeting banks is whether or not to publish how members of decision-making boards vote on monetary policy issues. It is argued that announcing differences of opinion among members could give mixed signals to the public, upset markets and prevent members from openly expressing their views (Fermo, 2012; 4). On the other hand, the case is made that publishing voting records increases accountability and encourages members to come to meetings better prepared (Kedan, 2014; 94). It can be argued that it is not a very effective tool, since the Barclays survey of 2014 placed “how members voted” as last on the list of importance. Looking at the central banks of developed nations, the United States, United Kingdom, Sweden and Japan all publish voting records and minutes of meetings after a set period. The European Central bank publishes minutes after 30 years, and Central Bank of Norway after 12 years. Canada, Switzerland and New Zealand do not publish their minutes (Kedan, 2014; 99). Alongside these shifts, the post-crisis period also brought about discussion of the effectiveness of monetary policy tools. The discussion focused on the effectiveness of policy rates and whether there was a need for quantitative tools. With short-term interest rates lowered as far as they could go (effective lower bound), central banks began to implement unconventional measures and asset-purchase programs as part of quantitative easing that aimed to stimulate the economy, contribute to financial stability and manage expectations. As part of this, central banks increasingly

623

624


turned to forward guidance as an important part of the communication aspect of their unconventional policies (Vayid, 2013; 18-19). With central banks turning to unconventional policies after the global financial crisis, it became harder for the public to anticipate the likely path of monetary policy, so the banks also took novel steps in terms of communication (Woodford, 2013; 96.) Forward guidance is a communication instrument by which central banks convey their monetary policy orientation going forward, conditional on their assessment of the economic outlook. (Praet, 2013). Praet also defines forward guidance as “a form of advance communication about future policy orientations”. Central banks turned to this kind of guidance as a means to fight against the difficulties of carrying out monetary policies in extraordinary times. Central banks used forward guidance not just because the lower bound had been reached, but because it was reached in the wake of a global financial crisis (Woodford; 2012, 2). In fact, the post-crisis guidance as part of unconventional tools used by central banks took the form of “extraordinary” or “enhanced” forward guidance (Vayid, 2013; 25). This meant providing greater guidance on the future path of interest rates and for many central banks it meant stating how long they planned to hold rates at the lower bound. For example, one of the first central banks to use extraordinary forward guidance was the Bank of Canada, which in April 2009 said it intended to keep the policy rate at the lower bound until the end of the second quarter of 2010, depending on the inflation outlook. Such conditional guidance was effective in shifting market expectations for the policy rate and as a result precluded the need for other unconventional policy measures (in particular quantitative easing) (Vayid, 2013, 28). Carney (2013) divides extraordinary forward guidance into three stages. In the early years of the crisis (end-2008 and early 2009) phrases such as “some time, considerable time, extended period” were used to express how long rates would remain low (open-ended or qualitative guidance). Carney calls this “first generation forward guidance”. The Bank of Canada statement referred to above from April 2009 gives the second quarter of 2010 as a clear date (date-contingent guidance). Carney sees this statement was the leader of “the second generation of forward guidance”. In December 2012, the US Federal Reserve said that rates would not rise until unemployment fell to a specified level (state or data-contingent guidance) and thus began “the third stage of forward guidance” (Carney, 2013; 16). 6. Measuring the Effectiveness of Language in Central Bank Communication Central Banks’ post-meeting statements constitute one of the key vehicles through which the banks communicate their policy actions, their assessment of the economy, and their thinking about future policy. However, determining what the central banks find relevant to policy discussions and how these discussions might have changed over time can be challenging. Since central banks release carefully constructed statements and meeting minutes to the public, applying text-mining techniques to those statements can help quantify this information to provide a rich analytical resource reflecting real-time economic and financial analysis. Although widely applied in other fields such as political science, business management, communication and marketing, text mining and content analysis have been historically less used as techniques in economics. This is especially the case with respect to research undertaken inside central banks. However, central banks can use these techniques to reach and analyze a range of data sources. This enables an assessment of monetary and financial stability and can assist the banks to perform their jobs more efficiently. Additionally, text mining and particularly content analysis may be attractive to central banks because they can not quantitatively analyze these data by other means (Bholat et al., 2015). Policymakers and academics have revealed both the pros and cons that are associated with opening up the internal debate within a central bank to public scrutiny. On the positive side, it is posited that transparency makes policymakers more accountable, and gives them reasons to work harder and behave in more ethical ways. Other observers, however, counter that excessive transparency can impede open discussion on central bank decision-making bodies (Hansen et al., 2014). Bailey & Schonhardt-Bailey (2008) and Schonhardt-Bailey (2013) also used text analysis to examine Fed FOMC transcripts. They focus on the arguments and rhetorical techniques used by policymakers (using evaluations from the “Alceste" software package) during three selected time periods (1979-1981, 1991-1993, and 1997-1999). They are not alone in applying algorithmic techniques to central bank text outputs. Romer & Romer (2004) examined such documents to extract a narrative-based examination of the impact of monetary policy shocks (Hansen et al., 2014) Fligstein, Brundage and Schultz (2014) applied LDA to Fed minutes from the years 2000 to 2007. They examine the documents within the framework of the sociology field of “sense-making”, finding consistencies with the theory.


They go on to argue that standard models can blind macroeconomists to the ramifications of their decisions on sectors such as housing, financial markets and the macroeconomy. In addition, Hansen et al., (2014) applied LDA to a dataset from the Greenspan era (1987-2006) and developed a range of measures of communication patterns at the meeting-section-speaker level. They went on to feed these measures into a difference-in-differences regression framework to observe the ways in which individual incentives can change with levels of transparency (Hansen et al., 2014). Cannon (2015) analyzed the FOMC transcripts from 1977 to 2009 by measuring the tone of FOMC deliberations, exploring the differences across speakers and examining how the tone of the discussions relates to a measure of economic activity. The results show that the tone of the FOMC’s discussion varies by speaker class, and that Bank Presidents contribute to the discussion in significantly different ways relative to Governors or Federal Reserve staff members. In addition, basic sentiment analysis shows the tone measure for the Committee discussions is strongly related to real economic activity, but that the relationship varies by speaker class. Finally, the analysis confirms the findings of other research that FOMC discourse shifted measurably after the decision to publish the transcripts in 1993 with both the tone and expression of the discussions changing measurably in the latter period. Another research into FOMC statement analysis was done by Meade and Acosta (2015). They used techniques developed in natural language processing (NLP, also referred to as computational linguistics) to study how the content of the FOMC's post-meeting statement has changed from May 1999--when the Committee began releasing statements consistently--through December 2014. They found that the semantic content of the statements from one FOMC meeting to the next is less similar and more variable than would appear at first glance. 7. Sample and Data Collection This research utilizes the Diction 7 software to analyze the semantic content of monetary policy texts from the Federal Reserve Board of the US (Fed) and the European Central Bank (ECB) from 2001 to 2015, and the Central Bank of the Republic of Turkey (CBRT) from 2002 to 2015and reveal how it has developed over time. Diction 7 is a computer-based program that searches a passage for five general lexical features as well as thirty-five sub-features to determine the tone of a verbal message. The software conducts its searches via a 10,000-word corpus and up to thirty user-created custom dictionaries that allow researchers to determine topical or negative words for particular research objectives. In this study we compiled a set of monetary policy statements from the Fed and CBRT, as these statements constitute one of the key vehicles through which these banks communicate their policy actions, their assessment of the economy, and their thinking about future policy. On the other hand, ECB does not include its policy actions, its assessment of the economy, or its thinking about future policy in its monetary policy decision statements. Such ECB statements just give interest rates. Thus, for the case of the ECB, we used the introductory statements given at press conferences after policy meetings since they provide detailed information about the economy, policy actions and future policy. All these statement were compiled through the ECB website. We used three master variables out of five that we believe that are important in terms of transparency and managing expectations through central bank communication: certainty, optimism and realism. The two master variables are defined below (Diction 7, 2015): x Certainty - Language indicating resoluteness, inflexibility, and completeness and a tendency to speak ex cathedra. x Optimism - Language endorsing some person, group, concept or event, or highlighting their positive entailments. Detailed information about the master variables is given in Appendix A. 8. Analysis and Results Results of the analyzed statements are given in Table 1 and in Table 2. There is heterogeneity across the three central banks in terms of the level of detail contained in the statements of policy meetings that are put on the public record. The average number of word character per statement is the same for the ECB and the CBRT. During the precrisis period of 2001 to August 2008, the optimism score averaged 50,64, while in the following period it was 49,27. On the other hand, the certainty score shifted from 43,76 to 48,84 (Table 1). Table 1: Optimism and Certainty Scores of Statements FED

ECB

CBRT

Optimism

Certainty

Optimism

Certainty

Optimism

Certainty

Before Crises (average)

50,64

43,76

51,99

46,64

49,69

44,57

After Crises (average)

49,27

48,84

51,28

46,45

50,03

47,24

625

626


Table 2: Central Banks’ Statements Data FED Total Word Characters Word Characters per Statements Number of Statements

ECB

CBRT

241161

1322243

358557

2010

2406

2406

120

168

149

According to the results for the Fed, based on 120 monetary policy statements from January 2001 to December 2015 obtained from its website, the average number of word characters per statement is 2,010. A total of 241,161 word characters are analyzed for the Fed. The first result to draw attention is that while the scores for optimism show a declining trend over the study period, the certainty score rises with time (Figure 4). This result is consistent with Meade and Acosta's (2015) results on the semantic content of the statements.

Figure 4: Fed’s FOMC Statements’ Optimism and Certainty Level

For the ECB, 168 introductory statements from press conferences from January 2001 to December 2015 were obtained from the Bank's website. The average number of words per statement is 7,870. A total of 1,322,243 word characters were analyzed. In the case of the ECB, the average scores obtained from Diction 7 for before and after the crisis are at the same level, and no clear difference in tone can be discerned (Figure 5).

Figure 5: ECB’s Introductory Statements’ Optimism and Certainty Level


For the CBRT, 149 monetary policy statements from January 2002 to December 2015 were obtained from the Bank's website. The average number of word characters per statement is 2,406. A total of 358,557 word characters are analyzed. Like the ECB, the scores obtained from Diction 7 for the CBRT shows no major difference between the preand post-crisis periods. One point of interest is that in the period before the 2008 crisis, particularly the years from 2002 to 2006, the certainty score displays wide volatility. Another intriguing finding is the significant increase in the optimism score after December 2013 (Figure 6).

Figure 6: CBRT’s MPC Statements’ Optimism and Certainty Level

9. Conclusion and Discussion This research into the communication policies of central banks makes use of monetary policy statements from the Fed, ECB and CBRT from 2001/2002 onwards to examine the semantic content before and after the 2008 global crisis, and determine whether there was any shift in the tone in the post-crisis period. Statements from two major banks that have significant impact on the global economy with their decisions and statements, Fed and ECB, were chosen and alongside them we looked at statements from CBRT, one of the most affected emerging economies by the global crisis. Naturally, the Fed’s statements are in English. For the purposes of the research, we examined the English-language versions of statements from the ECB and CBRT. In the case of the Fed, while the scores obtained from Diction 7 show the Bank using more certain expressions about the economy and policy that will be implemented in the framework of the forward guidance, they also draws attention to the fact that it uses less optimistic expressions, stemming from the fact that the economy remained in a period of crisis. The data obtained from Diction 7 concerning the three master variables reveal no great change in the tone for the ECB and CBRT in terms of pre- and post-crisis communication language. There is a slightly increase the tone of certainty for CBRT over the time. In the case of the Fed, there is a steady decline in optimism, particularly from the pre-crisis period to the post-crisis period, while at the same time there is a significant increase in certainty. This increase in certainty in the period of crisis is thought to be a reflection of the forward guidance phrases used in this period that linked quantitative easing to conditional or situational variables such as growth and unemployment. It is interesting here to consider the decline in the optimism score of Fed statements. Central Banks communicate their policy actions, their assessment of the economy, and their thinking about future policy with their post-meeting statements. Here, the language used in expressing future policy evaluations is crucial to expectations management. (i.e. De Haan et al. (2007)’s definition of “the art of expectations management”). The decline in Fed optimism, particularly after the global crisis, means we can say that the Fed was less optimistic in that period. The central assumption here is that if central banks are more optimistic in their statements or evaluations (in terms of expectation management), expectations will accordingly be better. The Fed score shows the reverse of this. While we might

627

628


expect, from an expectations management perspective, for the post-crisis optimism score to remain the same or perhaps exceed the pre-crisis period, in fact the opposite happened. It raises the question of whether the economy would have recovered from the crisis with less damage had the Fed issued statements with a higher optimism score? The answer to that question matters not just for the US and the Fed. The Fed's statements have an impact on the entire global economy and matter in particular for many emerging economies. Perhaps the damage of the crisis would have been minimized by a post-crisis Fed that was at least as optimistic, if not more, than prior to the crisis. As is often the case in the social sciences and economics, there is no way to test this proposition. On the other hand, the optimism of CBRT statements significantly increased after December 2013. It is worth considering whether the positive message in these statements helped to offset the negative sentiment that surrounded emerging markets because of the increased interest rates from the Fed tapering process. One limit to the research is the fact that ECB statements are translated into 22 languages, including English. The ECB's native language is not English and since the statement is translated into 22 separate languages the stresses of the tone of expression may not be the same as that of a country such as the Fed that operates in English. The specifics of communication may not be fully transferred into each of the 22 languages. The same is true for CBRT statements. The CBRT's native language is Turkish and publishes its statements translated into English. A further restriction is the fact that the results are based on monetary policy statements. While these are major documents for monetary policy, central banks have varying communication tools and channels. And each of those tools and channels may have their own tone. It is to be expected that central banks will give the message they wish to spread through each channel with the same tone and specifics. Subsequent research could focus on the tone and details of these alternate channels. For instance, inflation reports, financial stability reports, or speeches by governors. By adding additional countries to the dataset, the results found here could be tested for consistency. Lastly, Diction 7 scores from different countries could be studied in relation to macroeconomic variables from those countries through panel data analysis or other data analysis methods. While the idea that communication has become more effective at central banks through the increase in transparency, the example of the Fed suggests that approaching the negative situation of the crisis period with transparency and a decline in optimism may have led to the negative situation lasting longer. Perhaps, rather than being very transparent in times of crisis, central banks could be more effective by returning to their old secretive ways and positively swaying expectations by adding more optimism to their statements.

Appendix A. Diction 7 Master Variables Master Variable

Definition

Formula

Certainty

Language indicating resoluteness, inflexibility, and completeness and a tendency to speak ex cathedra

Optimism

Language endorsing some person, group, concept or event or highlighting their positive entailments. Language featuring movement, change, the implementation of ideas and the avoidance of inertia. Language describing tangible, immediate, recognizable matters that affect people’s everyday lives. Language highlighting the agreed -upon values of a group and rejecting idiosyncratic modes of engagement.

[Tenacity + Leveling + Collectives + Insistence] – [Numerical Terms + Ambivalence + Self Reference + Variety] [Praise + Satisfaction + Inspiration] – [Blame + Hardship + Denial] [Aggression + Accomplishment + Communication + Motion] – [Cognition + Passivity + Embellishment] [Familiarity + Spatial Terms + Temporal Terms + Present Concern + Human Interest + Concreteness] – [Past Concern + Complexity] [Centrality + Cooperation + Rapport] – [Diversity + Exclusion + Liberation]

Activity Realism Commonality

Source: Diction 7 Help Manual, 2015, 5.

References Bholat, D., Hansen, S., Santos, P., & Schonhardt-Bailey, C. (2015). Text Mining for Central Bank. Centre for Central Banking Studies, Handbook, 33. Bailey, A., & Schonhardt-Bailey, C. (2008). Does Deliberation Matter in FOMC Monetary Policymaking? The Volcker Revolution of 1979. Political Analysis, 16, 404–427. Blinder, A.S., Ehrmann, M., Fratzscher, M., de Haan, J., & Jansen, D. (2008). Central Bank Communication and Monetary Policy: A Survey of Theory and Evidence. European Central Bank Working Paper, 898.


Born, B., Ehrmann, M., & Fratzscher, M. (2013). Central Bank Communication on Financial Stability. The Economic Journal, 124, 701–734 (June). Brunner, K. (1981). The Art of Central Banking. Center for Research in Government Policy and Business Working Paper, Graduate School of Management, University of Rochester, 81(6). Callow, J., Hayes, S., Maki, D., & Ungerleider, L. (2009). How Should Central Banks Communicate? Barclays Capital Research. Cannon, S. (2015). Sentiment of the FOMC: Unscripted. Federal Reserve Bank of Kansas City Economic Review (Fourth Quarter) Carney, M. (2013). Monetary Policy After the Fall. Eric J. Hanson Memorial Lecture, University of Alberta, Alberta (May). Chant, J. (2013). The Bank of Canada: Moving Towards Transparency. Bank of Canada Review (Spring). De Haan, J., Eijffinger, S. C. W., & Rybinski, K. (2007) Central Bank Transparency and Central Bank Communication: Editorial Introduction. European Journal of Political Economy, 23, 1-8. Diction 7 Help Manual (2015). http://www.dictionsoftware.com/download.php?file=wp-content/uploads/2015/07/DICTION-7.1Manual.pdf Fermo, L. B. (2012). Central Bank Communication and the BSP’s Monetary Policy Statements. Banko Central NG Pilipinas Economic Newsletter, 12 (5). Fernandez, D., & Goradia, K. (2014). 2014 Survey of Central Bank Communication: The Gap Narrows but the Fed Still Sets the Standard, Barclays FICC Research. Fligstein, N., Brundage, J. S., & Schultz, M. (2014). Why the Federal Reserve Failed to See the Financial Crisis of 2008: The Role of “Macroeconomics” as a Sense Making and Cultural Frame. IRLE Working Paper, 111 -14. http://irle.berkeley.edu/workingpapers/111-14.pdf Fry, M., Julius, D. A., Mahadeva, L., Rogers S., & Sterne, G. (2000). Key Issues in the Choice of Monetary Policy Framework. In: L. Mahadevaand G. Sterne (eds.), MonetaryPolicy Framework in a Global Context, Routledge, London, United Kingdom. Greenspan, A. (2001). Transparency in Monetary Policy. In: The Federal Reserve Bank of St. Louis, Economic Policy Conference. St. Louis, Missouri. Hansen, S., McMahon, M., & Prat, A. (2014). Transparency and Deliberation within the FOMC: a Computational Linguistics Approach. CEP Discussion Paper, 1276. Issing, O. (2005). Communication, Transparency, Accountability: Monetary Policy in the Twenty-First Century. Federal Reserve Bank of St. Louis Review, 87(2, Part 1), 65-83. Kedan, D., & Stuart, R. (2014). Central Bank Communications: A Comparative Study. Central bank of Ireland Quarterly Bulletin, 2 (April). Knütter, R.,Mohr, B., & Wagner, H. (2011) The Effects of Central Bank Communication on Financial Stability: A Systematization of the Empirical Evidence. University of Hagen Discussion Paper, 463. Meade, E., & Acosta, M. (2015). Vanging on Every Word: Semantic Analysis of the FOMC's Postmeeting Statement. FEDS Notes. Mishkin, F. (2004). Can Central Bank Transparency Go Too Far? NBER Working Paper, 10829. Naghdaliyev, N. S. (2011). Central Banks’ Communication in the Post-Crisis Period. The Harriman Institute, Columbia University. Praet, P. (2013). Forward Guidance and the ECB, VoxEU.org column published on 6 August 2013. http://voxeu.org/article/forward-guidance-andecb. Romer, C. D., & Romer, D. H. (2004). A New Measure of Monetary Shocks: Derivation and Implications. The American Economic Review, 94(4),1055-1084. Schonhardt-Bailey, C. (2013). Deliberating American Monetary Policy: A Textual Analysis. Cambridge, MA: The MIT Press. Siklos, P. L. (2014). Communications Challenges for MultiǦ Tasking Central Banks: Evidence, Implications. International Finance, 17(1), 77–98. Sturm, J. E., & De Haan, J. (2011). Does Central Bank Communication Really Lead to Better Forecasts of Policy Decisions? New Evidence Based on a Taylor Rule Model for the ECB. Review of World Economics, 147, 41-58. Subbarao, D. (2011). Dilemmas in Central Bank Communication, Second Business Standard Annual Lecture, New Delhi. Vasko, D. (2012). Central Bank Transparency and Financial Stability: Measurement, Determinants and Effects. Master Thesis, Charles University, Prague. Vayid, I. (2013). Central Bank Communications Before, During and After the Crisis: From Open Market Operations to Open-Mouth Poicy. Bank of Canada Working Paper, 2013(41). Woodford, M. (2012). Methods of Policy Accommodation at the Interest-Rate Lower Bound. In Federal Reserve Bank of Kansas City Economic Symposium on The Changing Policy Landscape, Jackson Hole, WY. Woodford, M. (2013). Forward Guidance by Inflation-Targeting Central Banks. Sveriges Riskbank Economic Review, 2013(3) Special Issue, 81120. Yılmaz, C. B., & Kahveci, E. (2014). Merkez Bankası İletişiminde Yeni Bir Araç: Sözle Yönlendirme (ForwardGuidance), Ülke Örnekleri ve Türkiye Uygulamaları. Bankacılar Dergisi, 88, 5-26.

629