A corpus-assisted comparative genre analysis of corporate earnings calls between Korean and native-English speakers

English for Specific Purposes 32 (2013) 170–185

Contents lists available at SciVerse ScienceDirect

English for Specific Purposes journal homepage: www.elsevier.com/locate/esp

A corpus-assisted comparative genre analysis of corporate earnings calls between Korean and native-English speakers Hyeyoung Cho a, Hyunsook Yoon b,⇑ a b

Department of TESOL, Graduate School, Hankuk University of Foreign Studies, 107 Imun-Ro, Dongdaemun-Gu, Seoul 130-791, Republic of Korea Department of English Education, Hankuk University of Foreign Studies, 107 Imun-Ro, Dongdaemun-Gu, Seoul 130-791, Republic of Korea

a r t i c l e

i n f o

Article history: Available online 25 May 2013 Keywords: Earnings call Business English Corpus linguistics Keyword analysis Genre analysis

a b s t r a c t This study explores the value of earnings call transcripts as authentic resources for the investigation of English for Specific Purposes (ESP) in an international corporate context. Earnings calls are a company’s periodic reporting of financial results through teleconferencing with prospective investors or potential business partners. As English has emerged as the lingua franca of international business, corporate earnings calls would seem to be a natural arena for the study of the fluency of business English among non-native speakers. In this study, the authors argue for the pedagogical benefits of comparing the macro-textual and micro-textual distinctiveness of Korean participants’ use of English during earnings calls relative to their native English-speaking counterparts. The findings suggest that Koreans have low-level genre awareness of earnings calls, which causes confusion or misunderstanding for native English-speaking participants during the calls. This study highlights the importance of promoting awareness of professional genre specificity in teaching business English, which may lead to improved communication between personnel of international corporations. Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction With the globalization of information and communication technologies, earnings calls have become an important vehicle for financial disclosure in international business. In fact Crawford Camiciottoli (2010a) situates earnings calls within a financial disclosure genre chain that consists of earnings call announcements, earnings releases, the actual earnings calls, quarterly reports, and the annual reports. Specifically, an earnings call is ‘‘a conference call organized by companies through an external teleconferencing service to report periodic economic results’’ (Crawford Camiciottoli, 2010a, p. 345). It provides a quarterly overview of earnings as related to business operations, a forecast for the upcoming business cycle, and the chances for financial analysts to interact with corporate representatives during a question-and-answer session. Of particular interest to scholars of English for Special Purposes (ESP), this latter feature brings prospective business partners together for an opportune dialogue that can be captured as a source of authentic business conversation at the international level (e.g. Crawford Camiciottoli, 2009). Crawford Camiciottoli (2006) presciently suggested the pedagogical potential of examining and exploiting earnings calls transcripts for ESP practitioners, e.g. ‘‘from a teaching perspective, it would be worthwhile to compare the language used by native vs. non-native speakers of English who participate in corporate earnings calls’’ (p. 130). Building on Crawford Camiciottoli’s suggestion, this study utilizes the earnings calls of non-native speakers in order to compile distinct Korean and native corpora for a comparative analysis. The analysis will occur on two planes: first, at the

⇑ Corresponding author. Tel.: +82 2 2173 3978; fax: +82 2 2173 3359. E-mail addresses: [email protected] (H. Cho), [email protected] (H. Yoon). 0889-4906/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.esp.2013.03.001

H. Cho, H. Yoon / English for Specific Purposes 32 (2013) 170–185

171

macro-textual level to investigate the structural characteristics of the Korean corpus in comparison to the native counterpart; second, at the micro-textual level to identify the distinctive linguistic features in the Korean corpus. This approach is designed to understand the level of non-native speakers’ awareness of structural and linguistic specificity of earnings calls with the overall intent to help them communicate more clearly and effectively as professionals in their own fields. 2. Literature review 2.1. The use of language in financial releases as impression management strategy In recent decades, the outbreak of high-profile business scandals involving companies, such as Enron and WorldCom, has drawn scholarly scrutiny of the use of language in financial releases. Such studies continue to be relevant in the wake of the 2008 worldwide collapse of financial markets, which launched the ‘‘Great Recession,’’ the worst economic crisis since the 1930s. These studies have seized upon corporate communications as objects of research in order to explore their intent for promoting transparency and fair trade in the market place. These communications run the gamut of sub-genres ranging from annual reports, CEO statements and corporate press releases to earnings calls. Studies of the communications of financial releases during economic downturns have discovered that companies have circumvented transparency by using language to manage the impressions of their weak performances. Rutherford’s (2005) examination of annual report narratives suggested the existence of the Pollyanna effect phenomenon, or ‘‘language biased toward the positive’’ (p. 349) even for poorly performing companies. According to Rutherford’s analysis, poorly performing companies mentioned profits more frequently than their financial underperformances. Furthermore, the proportion of positive to negative words in annual reports were similar regardless of the profitability or financial adversity of the company, which lent additional evidence that companies manipulated language to put their best feet forward on muddy terrain. In fact, Schleicher and Walker (2010) found that companies use annual reports proactively to manage impressions that averted expectations of bad news. In their study, firms expecting a decline in performance used language to generate a positive tone in the outlook. Firms adeptly recast negative statements into positive assertions, which contributed to the upbeat – although suspect – quality of their reports. CEO statements, written to inform ‘‘a company’s philosophy, especially about its principles and values, and [to] create a particular image of a company’’ (Raab, 2009, p. 4), are another form of financial communication that can be cogent objects of linguistic analysis. As unaudited disclosures, CEO statements are not considered to carry the empirical weight of other standard financial reports. However, Smith and Taffler (2000) were able to distinguish between bankrupt and financially viable firms by carefully analyzing themes and keywords that expressed the intended narratives of chairmen’s statements. Following that line of enquiry, Clatworthy and Jones (2006) argued that CEO statements of ‘‘unprofitable companies focus less on key financial indicators, use fewer quantitative results, fewer personal references and more passive sentences, and focus more on the future’’ (p. 506). The gist of these studies indicates that the language used by a company can be more instructive of its financial condition than profit-and-loss figures given on the balance sheet. Corporate press releases have also provided rich data that circumscribe the use of language for modifying financial disclosure. It was Catenaccio (2008, p. 9) who defined press releases as ‘‘short pieces of writing issued by companies or institutions to communicate newsworthy information [emphasis added]’’ to the general public as well as to the journalists. Drawing on the work of Fairclough (1992), Bhatia (1993, 2004), as well as McLaren and Guraˇu (2005), and Catenaccio (2008) identified, the press release as a hybrid genre serving both informative and promotional purposes, which adds nuance to the definition that it should be ‘‘newsworthy.’’ In order to narrate the company’s positive self-image, the corporate press release focuses on four central themes, i.e. sales, product, financial, and organizational stories (Pallas, 2004). Analyzing corporate press releases from companies listed on the Swedish stock exchange, Pallas (2004) revealed that sales and product stories were mainly concerned with promoting corporate strength and growth, while financial and organizational stories aimed at presenting a corporate image as a forward-looking, well-managed and prosperous enterprise. McLaren and Guraˇu’s (2005) corpus-based analysis of biotechnology firms’ press releases discovered a standard pattern or move, which was announcement, elaboration, CEO’s comments, contact details and editor’s note. Catenaccio (2008) proposed a refined move structure of the genre which holds strong potential for pedagogical applications in ESP. 2.2. Exploring genre specificity of earnings calls in ESP This study of earnings call transcripts occurs within the context of accounting and financial studies (e.g. Brown, Hillegeist, & Lo, 2004;Frankel, Johnson, & Skinner, 1999;Tasker, 1998), which demonstrated that the interactive and real-time nature of earnings calls has a direct impact on the market. The nature of these studies begs for subsequent investigation on the linguistic and structural distinctiveness of earnings calls as a genre of financial disclosure. It was Crawford Camiciottoli (2006, 2009, 2010a, 2010b) who pioneered a corpus-based approach to explore genre specificity of native speakers’ earnings calls in US businesses. Her study in 2006 examined the rhetorical strategies of different types of call participants, which laid the groundwork for more sophisticated studies of the macrostructural features of earnings calls. Later, Crawford Camiciottoli (2010a) constructed a corpus of 20 earnings call transcripts in order to refine the macrostructural findings into a more viable and inclusive framework. The macrostructure of native speakers’ earnings calls was constituted of these seven moves: opening, introduction, financial overview, financial details, wrap-up, question and answer (Q&A) sessions, and the closing.

172

H. Cho, H. Yoon / English for Specific Purposes 32 (2013) 170–185

In addition, Crawford Camiciottoli (2010a) proposed a genre chain, as identified in the introduction of this paper, which provided the foundation for understanding the calls within a framework of financial disclosure. Among the seven moves of earnings calls, Crawford Camiciottoli (2009) brought the Q&A session under the microscope. In the examination of textual metadiscourse used by call participants, the research suggested that corporate representatives employed rhetorical strategies to make the interpretation of information as positive as possible while analysts used them to make their questions indirect, which was not a gesture of politeness, but a tactic to extract as much information as possible from the answer. In a focus on linguistic choices in earnings calls, Crawford Camiciottoli (2010b) compared the different uses of logical connectives, such as contrastive, additive, and resultative connectives, in earnings presentations and earnings releases. The former were oral presentations during the earnings call, whereas the latter were internet-published written counterparts. The overall frequency showed that the connectives were more frequently used in the presentation corpus than in the releases corpus. The qualitative analysis indicated that different types of discourse connectives served a very similar function that highlighted the positive results, while downplaying any negative information in both corpora. Although the study was purely linguistic, it raised questions about the conflation of cultural nuance in the discourse. It was apt to suggest that the optimistic trend of the discourse reflected the overall world view of US companies, which were the major objects of Crawford Camiciottoli’s studies. These insights prepared the ground for further studies germane to the international field of ESP. Specifically, we want to expand the work on genre specificity on earnings calls to non-native speakers, particularly to the personnel of Korean multinational corporations. Such a study requires a comprehensive approach in order to ensure the study is of maximum use pedagogically. Being comprehensive, we aim to compare both the linguistic and structural distinctiveness between non-native and native English speakers. In this way, we will be able to analyze a comparable body of corpora that will point to genre specificity, which can help raise the awareness of the distinctions in teaching business English to non-native speakers. 2.3. Investigation of Asian business discourse This study works toward alleviating the paucity of literature regarding financial communications in Korea and more generally to that of Asia. Investigations of financial communications have focused largely on Western business concerns, i.e. the United States and some European countries. The few studies of Asian business discourse which exist have focused on the dominant economic player in the region, i.e. Japan (e.g. Kameda, 2001; Marriott, 1997; Miller, 1994, 2000; Yamada, 1997), and the emerging economy of China (e.g. Bilbow, 1995, 1997, 2002; Spencer-Oatey & Xing, 2003, 2005; Yeung, 1997; Zhu, 2005). These studies engender a cultural perspective that defies dominance of written and spoken norms and conventions within the business genres. For example, Yeung (1997) revealed a unique Chinese business culture in examining business correspondence of Chinese and Westerners. The study tested the independent impact of Brown and Levinson’s (1987) three factors on the choice of politeness expressions in making requests, i.e. social distance, relative power, and imposition (cf. Bargiela-Chiappini & Harris, 1996; Holtgraves & Yang, 1992). Yeung (1997) reported a significant value of imposition in English data and highlighted the significance of on-record appropriateness in Chinese business requests, i.e. on-record requests should be avoided when one is not in a social position to make them. After making a request, it was equally important to express gratitude by extending commitment to reciprocate in the future. Zhu (2005) examined written business discourse of sales letters, invitations, and business faxes written by Chinese and native English speakers. Zhu’s sociocognitive perspective was able to identify the emotional approaches in rhetorical strategies and move structures in texts written by Chinese people, which would be shared with other Asian business cultures heavily influenced by Confucian values, such as Korea, and possibly Japan. In the area of spoken business discourse, Bilbow and Yeung (1998) reported that Western interviewers negatively evaluated vagueness of language in contrast to their Chinese counterparts who tolerated more circumlocution, or indirectness. Bilbow (1997) argued that in business meetings, Chinese interlocutors tended to perceive circumlocution as authoritative while Westerners considered it manipulative. Along with the Asian Business Discourse(s) project, whose findings consisted of a collection of essays (Bargiela-Chiappini & Gotti, 2005) and a special issue of the Journal of Asian Pacific Communication (Bargiela-Chiappini, 2005, 2006), recent attention to business discourse has been expanded to other parts of the region, such as Malaysia (Hadina & Rafik-Galea, 2005; Nair-Venugopal, 2000, 2001), India (Kaul, 2003; Kaul, Ansari, & Rai, 2005), Vietnam (Chew Chye Lay, 2005), and Korea (Jung, 2005a, 2005b, 2012). Jung’s work (2005a, 2005b) is valuable as it investigated Korean companies’ internal (e-mail messages) and external (formal letters) business correspondence, which enables English native speakers to better understand the unique rhetoric structure and strategies in Korean business communication. Jung (2005a) reported on business strategies of expressing power and politeness, such as solidarity enforcement, conflict avoidance, and off-record strategies in Korean business correspondence. Also, Jung (2005b) investigated the rhetorical structure of Korean correspondence which showed a three-part-pattern: –Ki (reason), –Sung (certain positive face threatening acts), –Kyul (request). Jung’s latest investigation (2012) of three Korean companies’ CEO messages on official websites discovered inconsistencies between the work orientation projected by the CEO messages and the official key values of that company. Jung’s studies are extremely relevant to the pedagogical issues in ESP as he highlights the business culture and identifies the rhetorical devices and linguistic choices in Korean business texts.

H. Cho, H. Yoon / English for Specific Purposes 32 (2013) 170–185

173

In the context of these efforts, there is a clear need to investigate Koreans’ use of English in an international business context in comparison with native English speakers. It is an important time to examine the difference as the Korean government seeks to expand economic growth through the Free Trade Agreement (FTA).1 As never before, Korea’s historic overseas economic expansion will require high levels of language competency of the commercial lingua franca, business English. In an attempt to investigate Koreans’ use of business English for pedagogical implications, this study takes earnings calls as an authentic resource for creating business spoken corpora and examines the macro- and micro-textual distinctiveness of Korean earnings calls in comparison to those of native counterparts. Thus, this ‘‘genre-based corpus study’’ (Flowerdew, 2005, p. 325) of Korean and native corporate earnings calls aims to investigate the following questions in order to find ways to help learners achieve a high standard of fluency in business English: 1. What are the differences in the macrostructure of earnings calls between the Korean corpus and the native corpus? 2. What are the distinctive linguistic features of the Korean corpus in comparison with the native corpus? 3. How can the macro-textual and micro-textual findings be applied to develop ESP pedagogies for business people who are non-native speakers of English? 3. Methodology 3.1. Sources of the corpora Materials for both the native and Korean corpora were collected from three web sites2 that post earnings call transcripts for investors and analysts. The authors chose transcripts from the years 2006 to 2011 in order to examine genre specific language use regardless of the major market fluctuations precipitated by the recent and ongoing global financial crisis. The native corpus was chosen from companies from seven economic sectors listed in the International Standard Industrial Classification of All Economic Activities (ISIC), which held earnings calls on a regular basis. The sectors included manufacturing, energy, wholesale/retail trade, transportation/storage, information/communication, financial/insurance, and education. Three major companies that actively engage in international business were selected from each sector, to attain a representative corpus as well as to investigate the common linguistic features across these sectors. Materials for the Korean corpus were collected only from four sectors because major companies in the areas of wholesale/ retail trade and transportation/storage do not, in practice, release their earnings calls details on the internet. This was unfortunate from a business transparency standpoint, and also because such materials could not be used to develop corpus-based approaches to improve English skills, which may affect these companies’ bottom line. The Korean corpus tends to be skewed toward manufacturing because those companies accounted for 44.6% of total tokens, which corresponds to a high dependency of manufacturing in the Korean economy (KIET e-report, 2009). Bio-sketches posted on the internet confirmed that corporate representatives making presentations and responding to analysts’ questions appeared to be native Koreans, although we could not verify if any individuals had happened to be born to Korean parents overseas and raised outside of Korea. In other cases, we identified a Dutch Vice President (VP) and an American Investor Relations (IR) VP in two of the companies; however, their contributions to the calls were relatively insignificant. Table 1 shows the overall makeup of the native and the Korean corpora. The total number of words in the native corpus is about four times that of the Korean one. 3.2. The macrostructure of earnings calls This research adopted the macrostructural approach aided by Crawford Camiciottoli’s (2010a) framework of earnings calls to examine the structural distinctiveness of Korean earnings materials relative to those of the native equivalents. Based on an examination of 20 earnings call transcripts, Crawford Camiciottoli provided a structural norm of international earnings calls, from which any deviance could generate questions and confusion for native English-speaking participants during the calls (Table 2). The framework included seven moves, each of which consisted of several steps managed by designated speakers. Move 1 (opening) was the official opening and welcoming of the participants to the earnings call, followed by move 2 (introduction), which was the beginning of the earnings presentation, usually mediated by an investor relations officer. In move 3 (financial overview), a CEO or a CFO greeted the participants and commented on the overall financial results. Move 4 (financial details) was a discussion of the financial statement and a forecast of future trends, followed by move 5 (wrapup), which discussed remaining topics and initiated move 6 (Q&A sequences), where investors had the opportunity to participate. Once the Q&A was concluded, the operator might offer the corporate representatives a chance to add any closing remarks they wished before announcing the end of the call by asking the participants to disconnect (move 7). A common step from moves 3 through 5 is to instill confidence, hinting that a goal of the earnings call was to promote a positive com1 In recent years, South Korea and the United States have agreed to an FTA, while talks are in the works regarding the development of a free-trade zone with China and Japan. 2 http://seekingalpha.com, www.morningstar.com, and http://goliath.ecnext.com are websites which provide information on corporate earnings calls and market analysis free of charge.

174

H. Cho, H. Yoon / English for Specific Purposes 32 (2013) 170–185

Table 1 Overview of native corpus and Korean corpus. Industry (code)

Native corpus

Korean corpus

Company

# of words

Company

# of words

Manufacturing (C)

N1 N2 N3

11,572 11,961 9131

32,664 (16%)

K1 K2 K3 K4

7,855 966 7,440 6,812

23,073 (44%)

Energy (electricity, gas, steam and air conditioning supply, D)

N4 N5 N6

7108 8349 7020

22,477 (11%)

K5

3,262 (6%)

3262 (6%)

Wholesale and retail trade (G)

N7 N8 N9

10,901 8503 9358

28,762 (15%)

–

–

–

Transportation and storage (H)

N10 N11 N12

8881 10,690 8669

28,240 (14%)

–

–

–

Information and communication (J)

N13 N14 N15

9237 10,178 8687

28,102 (14%)

K6 K7

8,039 6,328

14,367 (28%)

Financial and insurance activities (K)

N16 N17 N18

8993 1360 12,618

35,214 (18%)

K8 K9

5,839 5,440

11,279 (22%)

Education (P)

N19 N20 N21

12,601 2199 8569

23,369 (12%)

–

–

–

Total

21

198,828 (100%)

9

51,981 (100%)

Table 2 Macrostructure of earnings calls (Crawford Camiciottoli, 2010a). Move/steps

Speaker(s)

1. Opening Welcoming participants Giving technical instructions Passing the floor to first corporate speaker

Teleconference operator

2. Introduction Greeting participants Introducing the speakers Outlining the presentation Stating the Safe Harbor disclaimer Passing the floor to the next corporate speaker

Investor Relations Officer (or: Chief Executive Officer, Chief Financial Officer, Vice President)

3. Financial overview Greeting participants Commenting on overall financial results Instilling confidence Passing floor to next corporate speaker

Chief Executive Officer (or: Co-Chairman/Co-CEO, Chief Financial Officer)

4. Financial details Elaborating on financial statements Forecasting future trends Instilling confidence Passing the floor to the next speaker

Chief Financial Officer (or: Chief Accounting Officer, Treasurer)

5. Wrap-up Discussing remaining topics Instilling confidence Opening Q&A sessions

Chief Executive Office (or Chief Financial Officer, Chief Operating Officer)

6. Q&A sequences (repeated until participating analysts have had opportunity to ask a question)

Teleconference operator + Analyst + Member of Executive Team

7. Closing Announcing the end of the call Inviting participants to disconnect

Teleconference operator

175

H. Cho, H. Yoon / English for Specific Purposes 32 (2013) 170–185

pany image, not just to present an obligatory report of factual data. It is critical for non-native speakers to understand the cultural nuances of the discourse in order to effectively represent their companies in the global economic context. 3.3. Data analysis The examination of macro-textual distinctiveness of the Korean corpus started with compiling the earnings calls corpora of both native English speakers and Koreans. The Korean corpus is examined according to each move of the international earnings calls proposed by Crawford Camiciottoli (2010a). In this study, moves 1 and 7 were excluded from the discussion because teleconference operators who mediated the moves tended to be native speakers and were not germane to the actual business discussions. Based on the examination of moves 2–6, the idiosyncrasies of the Korean corpus were identified in comparison with the native counterpart, followed by the qualitative investigation of each move. In order to investigate the micro-textual distinctiveness, a keyword list of the Korean corpus was created. In preparation, the native and the Korean corpora were cleaned up by removing headers and unreadable symbols, and a wordlist was created for each corpus, which was lemmatized by loading e-lemma designed by Someya (1999). The two lemmatized wordlist files were computed by WordSmith 5.0 (Scott, 2008) to generate a keyword list of the Korean corpus by using the native corpus as a reference. The keyword list included positive and negative keywords, which clearly showed the distinctive linguistic features of the Korean corpus. Non-representative linguistic features of the Korean corpus were removed, for example, proper nouns such as wafer and telecom, or acronyms, e.g. ARPU (average revenue per user). 4. Results and discussion 4.1. Analysis of the Korean corpus macrostructure The moves of relevance to the macrostructural investigation are the financial overview (move 3) and the financial details (move 4). The internal analysis of these moves reveals the small number of steps that mark the transitions between participants during the presentation as is shown in Table 3. 4.1.1. Earnings calls: extent of participation A plausible explanation for the low frequency of turn transitions between participants in moves 2, 3, and 4 in the Korean corpus is the small number of corporate participants, which reduces the need for introducing other speakers during the presentation. In many cases, one Korean speaker took charge of moves 2, 3, and 4, each of which invited different speakers specialized in the given areas in the native corpus as exemplified in Excerpt 1. What is interesting, though, is that the average number of participants of Korean earnings calls is not significantly smaller than that of native earnings calls (Table 4). As we shall discuss later, the Korean participation in earnings calls tends to be restricted to a few individuals relative to their counterparts of the native corpus in which all participants are more likely to interject questions, comments and opinions throughout the calls.

Table 3 Macrostructure of Korean corpus from Move 2 to 6. Move

Company Steps

K1

K2

K3

K4

K5

K6

K7

K8

K9

Total

2. Introduction

Greeting participants Introducing the team of speakers Outlining the presentation Stating the Safe Harbor disclaimer Passing the floor to the next corporate speaker

O O O O O

O O O X X

O O O O O

O X X O X

O X O X O

O X O O O

O O O O O

O X O X X

X O X O O

8 5 7 6 6

3. Financial overview

Greeting participants

O

O

O

X

O

O

O

X

O

7

Commenting on overall financial results Instilling confidence Passing the floor to the next corporate speaker

O O X

O O X

O X O

O X X

O X X

O O X

O X X

O X X

O O O

9 4 2

Elaborating on financial statements

O

O

O

O

X

X

X

O

O

6

Forecasting future trends Instilling confidence Passing the floor to the next speaker

O O X

O X X

O O X

O O X

X X X

X X X

X X X

X X X

O O X

5 4 0

5. Wrap-up

Discussing remaining topics Instilling confidence Opening the Q&A sessions

O O O

X X O

O O O

O O X

X X O

X O X

X O O

X O O

X X O

3 6 7

6. Q&A sequences

(repeated until all participating analysts have had the opportunity to ask a question)

O

O

O

O

O

O

O

O

O

9

4. Financial details

176

H. Cho, H. Yoon / English for Specific Purposes 32 (2013) 170–185

Excerpt 1. Example of frequent turn transition in the native corpus (C14) (Note. Letters in brackets indicate names of the speakers which were erased to maintain anonymity).

Table 4 Number of corporate participants in both corpora. Participants

Korean corpus

Native corpus

Chief Executive Officer (CEO chairman or president) Vice President (VP) Chief Financial Officer (CFO) Others (mostly investor relations directors) Total Average (# of total participants/# of earnings call)

3 4 5 13 25 2.8

29 9 20 17 75 3.6

The qualitative examination of Korean earnings call transcripts reveals that some participants remain silent from moves 2 to 5 (the presentation) and become active in move 6 (Q&A session) as in K1, K5, and K8. Even though one representative moderates the Q&A session, additional participants tend to supply details to satisfy analysts’ attention to accuracy. Moreover, some participants (e.g. the head of IR of K2, CTO and CMO of K7, CEO of K8) make no appearance throughout the call, even though they are introduced as participants at the beginning of the presentation. This is in stark contrast to the native corpus in which usually more than three corporate representatives actively engage in the presentation as well as in the Q&A session. In terms of the participants’ profiles, Korean earnings calls involve a high number of Investor Relations (IR) team members (Table 4). Korean executives tend to hand the proceedings over to the IR team, compared to the native corpus in which highprofile executives (such as CEO, VP, and CFO) actively participate. This tendency can be explained by several variables. One could be the age of high-level Korean executives. They may have come through the corporate ranks when English interaction was not particularly important for career advancement. A second factor could be the culturally hierarchical nature of Asian businesses in which the top leaders tend to play a distant, aloof and paternalistic role to their junior executives. A third factor could be the educational background of younger executives who may have been able to avail themselves of recent opportunities, educational or otherwise, in English-speaking countries; hence, they are familiar with more recent colloquial patterns of the language. Fourth, due to web-based forms of international business communication, e.g. email, webconferencing, and telephoning, it is the tech-savvy junior executives who interact with international counterparts on a frequent basis. In fact, the globalized nature of international communications makes such frequent contact not only possible, but also essential in order to survive in a global economy. Finally, junior executives are more likely to be active consumers of Western media and entertainment genres, which offer continuous exposure and updates on new forms of English. 4.1.2. Efforts to instill confidence in the corpora Instilling confidence is the only common feature found in moves 3, 4 and 5 of Crawford Camiciottoli’s (2010a) framework, which suggests that the purpose of earnings calls is not only to release financial details, but to leave positive impressions of the company in order to instill confidence and to encourage investment. However, the few occurrences of instilling confidence steps in the Korean corpus indicates that Koreans tend to focus on delivering simple facts without embellishing these details to develop the confidence of potential investors. In contrast, native speakers use the language to avoid giving any negative impressions to analysts as shown in Excerpt 2.

H. Cho, H. Yoon / English for Specific Purposes 32 (2013) 170–185

177

Excerpt 2. Differences in efforts to instill confidence between Korean and native representatives (Note. REP stands for the corporate representatives of a company hosting the earnings call who make the presentation and take questions from analysts during the Q&A session.).

Excerpt 3. Misunderstanding in the Q&A session of Korean corpus I. Note. AN stands for analysts who participate in the call and OP⁄⁄ stands for a teleconference operator who mediates turn taking of the Q&A session.

This excerpt, which came from the wrap-up (move 5) of two financial companies’ earnings calls, is indicative of the critical difference in the occurrence of instilling confidence steps between Korean and native speakers. The differences are even more remarkable given that the earnings calls occurred in the context of a sudden and dramatic fall in the global financial market, which would seem to be a time for pessimism. As he openly reports bad news in the underlined sentences in Excerpt 2, the Korean speaker seems to be indifferent to, or unaware of, the need for any effort to offset the negative atmosphere, while the native counterpart keeps highlighting their ability to navigate through the difficulty. Such differences in attitude will affect the corporate image, which consequently changes the market evaluation of the company. A possible factor playing a role here is the ‘‘cling onto’’ (Granger, 1998, p. 156) feature of Korean speakers, given their lack of fluency to articulate the nuances required for various social, and in this case economic, contexts. 4.1.3. Miscommunications in Q&A The Q&A session is an exceptional opportunity to capture unscripted English interaction between non-native speakers and their native counterparts, thus providing examples of authentic business communication between them. The Q&A is characterized by spontaneous expressions and instant responses in which non-native corporate representatives display a limited understanding of the intended meanings of the questions. This becomes apparent in pragmatically inappropriate responses of non-native speakers to their native-speaker colleagues. Unable to elicit the sought-after information, the native speaker makes a perfunctory acknowledgement and ends the discussion. It is clear, then, that these episodes represent missed opportunities for the Korean speaker to tell the full story of the company’s earning potential, and hence may affect future potential investment in the firm. Excerpt 3 is an example in which the native speaker’s intended meaning eludes the Korean counterpart. The question in lines 4 and 5 concerns whether the company’s future plans will involve either developing new markets or else growing the share of the existing ones. The Korean representative, however, takes the question too literally and repeats his answer twice as shown in lines 6 and 8, as if to avoid discussing the vulnerability in the marketing plan. As there seems to be no hope of getting the answer he seeks, the analyst ends the line of inquiry (line 9). The Korean answer may be culturally nuanced in an attempt to save face of what he perceives as a threatening situation. Saving face is a response appropriate to Asian cultures based on Confucian values. On the other hand, one can imagine that the analyst may find the response defensive, or else infer a low fluency level of the Korean representative that could hamper full development of the business relationship.

178

H. Cho, H. Yoon / English for Specific Purposes 32 (2013) 170–185

Excerpt 4. Misunderstanding in the Q&A session of Korean corpus II.

The discussion in Excerpt 4 exemplifies that the Korean representative’s lack of linguistic ability causes a serious disjuncture in communication with his counterpart. Asking the reasons for a cost reduction (lines 7 and 8), an analyst is confused by the interjection of ‘‘fleet cost’’ (i.e. the corporate spending on transportation) in line 9, because it does not seem to be related to a unit of cost reduction. Despite a lengthy explanation from lines 11 through 19, which is not related to transportation spending, the analyst remains confused despite asking for clarification (lines 20–22) and continues to feel that the answer is not sufficiently clear (line 24). 4.2. Keyword analysis Table 5 shows positive and negative keywords drawn from the Korean corpus by using the native corpus as a reference. In preparing for the keyword analysis, the authors had to consider how it would be affected by the high inclusion of manufacturing companies in the Korean corpus. Therefore, we screened out any industry-specific items from the analysis by checking the keyword plot. Among the positive keywords, subscriber only appears in information and technology companies, while deposit and loan are mostly used in the financial sector. The use of TV and inch is concentrated in the manufacturing industry, and trillion and KWR (South Korean won) are frequent due to aspects related to the Korean currency. The exclusion of the industry-specific items finds two conspicuous verbs – decrease and will – that represent the linguistic distinctiveness of the Korean corpus in general. The two verbs will be discussed in detail along with the negative keywords in the following sections. 4.2.1. Overuse of decrease In order to explain the high keyness of decrease, this research examines the frequency of downgrading expressions, i.e. denoting downward changes in business such as loss and decline, in both corpora. For each item, the raw frequency is normalized by the parameter of token per 10,000 words (pttw) for purposes of conducting a more reliable comparison. Table 6 shows that the downgrading expressions are more frequently used in the Korean corpus than in the native one. As the transcripts for both corpora were collected across different business sectors within the last six years, the high frequency of downgrading expressions in the Korean corpus is unlikely to be the sole result of an economic downturn or the consequence of financial distress in a certain Korean business sector. When delivering less-than-positive results during the calls, it seems that Koreans tend to overuse downgrading expressions with little or no effort to mitigate the negative tone. In contrast, native speakers attempt to moderate the negative impact of the language. This is clear in Excerpt 5 from the financial overview (move 3), which illustrates Koreans’ repetitive use of downgrading expressions in delivering negative financial performance (as underlined). In similar circumstances, native speakers use language to temper the financial underperformance. Native speakers quantify the revenues straightaway, indicating that the company can perform at a good level, and seemingly as an afterthought, they mention the percentage decrease from the prior quarter. It is a sort of good news then bad news approach. Second, native speakers explain the disappointing results in terms of a difficult business climate, thus justifying the results and shielding the company from any adverse judgment of its management. Third, native speakers provide details of

179

H. Cho, H. Yoon / English for Specific Purposes 32 (2013) 170–185 Table 5 Positive and negative keywords: Korean corpus. No.

Positive keyword

Freq.a

%

RC freq.b

RC%

Keyness

1 2 3 4 5 6 7 8 9 10

subscriber trillion deposit TV decrease inch KRW loan regard will

101 86 70 58 83 43 41 115 97 515

0.19 0.17 0.13 0.11 0.16 0.08 0.08 0.22 0.19 0.99

7 1 11 4 32 – – 90 69 1100

– – – – 0.02 – – 0.05 0.03 0.55

269.47 260.35 161.15 154.81 140.20 135.38 129.08 122.78 112.09 110.71

Negative keyword

Freq.

%

RC freq.

RC%

Keyness

over thing drive think get

57 11 12 118 32

0.11 0.02 0.02 0.23 0.06

544 267 288 1002 477

0.27 0.13 0.14 0.50 0.24

55.27 66.09 70.84 82.97 83.41

No. 1 2 3 4 5 a b

Freq. means raw frequency of a keyword. RC is short for reference corpus, i.e. Native corpus of this study.

Table 6 Frequency of downgrading expressions in both corpora. Words

Korean corpus

Native corpus

Normalized freq. raw freq.

Normalized freq. raw freq.

decrease low (lower) down loss decline reduce cut fall minus downturn drop downward lose diminish squeeze shrink drawdown slowdown

15.97 15.20 14.81 9.81 8.66 8.46 2.50 2.12 1.73 1.54 1.35 1.35 1.15 0.58 0.19 0.19 0.19 0.19

83 79 77 51 45 44 13 11 9 8 7 7 6 3 1 1 1 1

1.61 14.13 14.79 4.02 4.78 3.22 0.91 2.36 0.75 0.50 1.31 0.15 1.11 0.20 0.25 0.15 0 0.35

32 281 294 80 95 64 18 47 15 10 26 3 22 4 5 3 0 7

Total

91.57

476

54.52

1084

Excerpt 5. The use of downgrading expressions in Korean and native corpora.

180

H. Cho, H. Yoon / English for Specific Purposes 32 (2013) 170–185

Table 7 Top 5 downgrading expressions in the Korean corpus. Semantic prosody

decrease low down loss decline

Korean corpus

Native corpus

Number (%)

Amount/ size (%)

Level/ degree (%)

Wealth/ value (%)

Others (%)

Number (%)

Amount/ size (%)

Level/ degree (%)

Wealth/ value (%)

Others (%)

46 7 14 18 40

13 14 3 2 9

23 47 19 4 18

18 32 39 76 33

– – 25 – –

70 4 29 15 23

7 18 10 6 10

10 43 25 3 15

13 35 13 76 52

– – 23 – –

Excerpt 6. The use of decrease in Korean and native corpora.

Excerpt 7. The use of decline in the Korean corpus.

positive actions the company is taking to address the challenging business environment. In other words, regardless of the reported shortcomings, the company is being proactive in identifying ways to turn around its performance. In their overuse of downgrading words, it is striking that Koreans seem to prefer decrease to its alternatives. This pattern aligns with the findings of Granger (1998, p. 156) that non-native speakers tend to overuse or ‘‘cling onto’’ certain expressions. Specifically, it indicates the limited lexical dexterity of non-native speakers, i.e. the lack of sufficient lexical knowledge for choosing viable and nuanced alternatives. To identify the semantic differences of downgrading words between the native and the Korean corpora, the authors categorized the definitions of the words by using Collins’ COBUILD Advanced Dictionary (2008). The category of ‘number’ includes cases where a downgrading word is used with an exact number. When the word is used with words of measurable mass, such as sales, volume, and capacity, it is categorized as ‘amount/size.’ The use of the word with pressure, risk, ratio, and usage is categorized as ‘level/degree,’ whereas the category of ‘wealth/value’ includes a case where a downgrading word is used with words of financial value such as earnings, revenue, and assets. The last category of ‘others’ includes cases where the word is used as a part of phrasal verbs with little relevance to a downgrading meaning (e.g. write down, jot down, break down). Occurrences of the downgrading expressions were counted to calculate the percentage of each category as shown in Table 7. In terms of low and loss, both corpora show similar patterns, while the use of decrease, down, and decline present obvious semantic differences between the two corpora. The native corpus displays a strong tendency for decrease to be used with an exact number, while Koreans frequently use the word to describe the downgrading of level/degree, wealth/value, and amount/size. The wider spectrum of usage seems to be closely related to the high keyness of decrease in the Korean corpus relative to the native corpus as exemplified in Excerpt 6. In Excerpt 6, the Korean speaker overuses decrease instead of better alternatives, such as decline and down, while the native speaker uses decrease with exact numbers such as 150 and 240 basis points (in lines 2 and 3) and in line with the morphological pair, i.e. increase (as underlined in lines 1 and 5). Similarly, the use of down and decline in the Korean corpus is at odds with the way in which these are used in the native corpus. Down is frequently used in the native corpus with an exact number and expressions related to the level; yet, Koreans tend to use it to explain the loss in wealth/value. As for the use of

181

H. Cho, H. Yoon / English for Specific Purposes 32 (2013) 170–185 Table 8 Future tense: frequencies in native and Korean corpora. Korean corpus Normalized freq. will (‘ll) be going to Total

Native corpus Raw req.

Normalized freq.

Raw freq.

99.07 9.81

515 51

55.32 15.74

1100 313

108.89

566

71.07

1413

Excerpt 8. The use of will in the Korean corpus.

Table 9 Frequency of modal verbs in native and Korean corpora. Type of modal

Korean corpus Normalized freq.

will would can could should may have to might must shall Total

Native corpus Raw freq.

Normalized freq.

Raw freq.

99.07 26.74 21.55 17.12 8.85 8.66 7.12 2.31 1.54 0.77

515 139 112 89 46 45 37 12 8 4

55.32 28.37 25.05 11.87 10.41 8.00 3.42 4.78 0.40 0.15

1100 564 498 236 207 159 68 95 8 3

193.72

1007

147.77

2938

decline, it often refers to shrinking in wealth/value in the native corpus, whereas it is frequently used with exact numbers in the Korean corpus as shown in Excerpt 7. The qualitative examination of downgrading expressions in the Korean corpus indicates speakers do not avoid lexical repetition by using synonyms. As exemplified in Excerpts 6 and 7, Koreans repeat the same words – decrease and decline respectively – in adjacent sentences. Based on the investigation of semantic differences of downgrading expressions between the two corpora, it is apparent that Koreans overuse downgrading expressions, particularly decrease. 4.2.2. High keyness of will The analysis of high keyness of will in the Korean corpus tests two possibilities: that it is more frequently used as the future tense, and a preference for will over other modal verbs. In order to examine the first possibility, the authors compared the frequency of the future tense between the native and the Korean corpora. Table 8 suggests that Koreans use the future tense more frequently than native speakers in general (108.89 vs. 71.07). More significantly, Koreans prefer will to be going to as a future tense expression. However, we refer to Celce-Murcia and Larsen-Freeman (1999) who explained that ‘‘simple future with will is used for future predictions and spontaneous decision when the person has control over the action. On the other hand, be going to is used for future prediction in more informal tone, future intentions based on prior decision, and future certainty based on current condition or present evidence’’ (p. 127). It appears that Koreans do not seem to be aware of such differences and stick to some familiar expressions (will in this case) as exemplified in Excerpt 8. In other words, they seem to lack a level of fluency by which they can appropriately choose a future tense expression relative to the situation they wish to communicate. Koreans’ overuse of will also suggests that their preference extends to the use of will as a modal auxiliary, which requires an examination of frequency of modal verbs in both corpora. Table 9 shows that Koreans use will more frequently than do native speakers. The difference in the normalized frequency of other modal verbs, such as would, can, and may, between the native and the Korean corpora is less than for will, which suggests that Koreans tend to overuse will in comparison to other modal verbs. Based on the findings, it seems to be a fair conclusion that Koreans prefer will as a future tense expression as well as a modal verb.

182

H. Cho, H. Yoon / English for Specific Purposes 32 (2013) 170–185

Excerpt 9. The use of I think in the native corpus.

Table 10 Frequency of items expressing caution in both corpora. Items

probably seem would may might perhaps possible(ly)

Korean corpus

Native corpus

Normalized freq.

Raw freq.

Normalized freq.

Raw freq.

3.46 6.16 26.74 8.66 2.31 0.77 4.04

18 32 515 45 12 4 21

0.1 2.72 28.37 8.00 4.78 1.26 1.51

2 54 564 159 95 25 30

4.2.3. Negative keywords of the Korean corpus The negative keyword list includes informal and spoken expressions such as get, thing, and think (or I think). In particular, think is considered as a feature of spoken language because it can function as a discourse marker. Discourse markers are ‘‘members of a functional class of verbal (and non-verbal) devices which provide contextual coordinates for ongoing talk’’ (Schiffrin, 1987, p. 41). In a general conversation, speakers use discourse markers such as you know, I mean, and then, and you know what I mean (O’Keeffe, McCarthy, & Carter, 2007). Think and I think also function as discourse markers as demonstrated by several researchers (e.g. Aijmer, 1997; Kärkkäinen, 2003). In particular, Aijmer (1997) stated that I think ‘‘has developed a number of new functions as a response to the demands of planning and interaction with the hearer which may in their turn become conventionalized’’ (p. 40). In order to examine the use of think by native speakers, we conducted an analysis of think clusters or bundles using WordSmith. The result indicates a high frequency of think clusters such as I think, I think we, and I think it is. A semantic examination reveals that I think is used to mark the cautiousness of one’s utterance, not to indicate the status of engaging in thought. In his study of pragmatics, Yule (1996) explained that ‘‘cautious notes expressed about how an utterance is to be taken’’ (p. 130) suggest that the speaker is hedging, or show that the speaker is conscious of maxims of a cooperative principle of conversation, i.e. quality, quantity, relation, and manner (see Grice, 1975). Excerpt 9 exemplifies the use of I think as a discourse marker in the native corpus. At the beginning of each sentence in lines 1, 4, and 7, the phrase I think functions as hedge, which relates to the quality maxim (i.e. the accuracy of the main statement). The use of a discourse marker to express cautiousness is understandable in earnings calls, where there is a premium on distinguishing between certainty and qualified opinion about circumstances affecting the bottom line. Therefore, the participants use the discourse marker to express as accurately as possible the extent of their knowledge related to the nature of the information regarding a particular situation. In order to examine the possible use of alternative discourse markers to show caution in the Korean corpus, this study compares the normalized frequency of seven items which are commonly used to express caution, i.e. probably, seem, would, may, might, perhaps, possible(ly), between the two corpora as shown in Table 10. Probably, seem, may, and possible(ly) are more frequent in the Korean corpus. However, the gap in the normalized frequency between the two corpora seems to be minimal (less than five), which makes it difficult to conclude that Koreans express caution by employing these particular hedges more than native speakers do. Instead, the high keyness of think as a negative keyword in the Korean corpus strongly suggests that Koreans have a low awareness of expressing caution during the calls in the same way as native speakers. 5. Conclusion 5.1. Major findings and discussion Accounting and finance studies of earnings calls have focused on examining the impact of language in terms of trading volume, analysts’ forecasting, and information asymmetries (e.g. Bowen, Davis, & Matsumoto, 2002; Brown et al., 2004;

H. Cho, H. Yoon / English for Specific Purposes 32 (2013) 170–185

183

Frankel et al., 1999). More recently, researchers such as Crawford Camiciottoli (2006, 2009, 2010a, 2010b) initiated the investigation of earnings calls as a genre in ESP; however, the pedagogical implications of this body of work are in need of development. To promote financial transparency, earnings calls require a high level of language fluency in order to accurately communicate between the parties. Lack of fluent communication skills undercuts the basic premise of the earnings call, and can seriously affect the relationships between the parties. Combined with a comparative analysis of native and non-native speakers’ performance, the investigation of earnings calls can suggest pedagogical strategies for the effective teaching of business English. In the long run, this approach can help to even the playing field for historically less-advantaged firms seeking an equitable share of global commerce (Frank, 1966; Freire, 1993). As the world economy has been structured according to enormous asymmetries of power situated in core international centers, i.e. London, New York, Hong Kong, etc, there is an urgent need to mitigate the asymmetries of global economic power for Korean as well as other non-English speaking companies striving to survive in the face of foreign competition. In response to such a need, our comparative analysis presents an examination of the difference between native and Korean earnings calls on macro- and micro-textual levels. Through a comparative move analysis and keyword analysis aided by the WordSmith tool, this study identified distinctive features in the Korean earnings calls which appear to be primarily caused by a low awareness of genre specificity of earnings calls and limited linguistic and pragmatic competence of Korean participants, not by cultural factors, such as an emotional or qing-oriented approach (Zhu, 2005) and tolerance for indirect language (Bilbow & Yeung, 1998). The macrostructural analysis of Korean earnings calls reveals the small number of steps to introduce the next speaker in moves 3 and 4, which is caused by the limited participation in earnings presentations. Largely due to either a lack of linguistic ability or to the uniqueness of the Korean business culture, or a combination of both, Korean corporate executives seem not to actively participate in the calls, instead handing over the job to the investor relations team. Another feature of Korean earnings calls is the absence of language to instill confidence in the company. Considering that the purpose of earnings calls is not only to make available the factual data, but also to promote the company’s image, it is imperative for Korean representatives to be aware of the power of the tone of their language as it relates to the roles of the earnings call. In addition, unlike the examination of the presentation (moves 2–5), the qualitative analysis of the Q&A session (move 6) shows Koreans’ limited pragmatic competence (i.e. ability to understand intended meaning) as well as linguistic competence. Our evidence revealed that Koreans tend to misuse terms, causing confusion for native counterparts, and also misunderstand the intended meaning of questions, which leads to pragmatically inappropriate responses. The keyword analysis shows the linguistic features that are overused and underused in the Korean corpus. Most of all, excessive use of downgrading expressions runs counter to the general characteristics of the financial disclosure, the Pollyanna Effect, which is the phenomenon of strategically using positive words in order to enhance impressions of a company. In financial disclosure, it is a standard practice even for poorly performing companies to accentuate the positive (e.g., Bhatia, 2008, 2010; Rutherford, 2005). The Pollyanna Effect seems to be emblematic of American companies and has yet to enter into Korean business culture. An example would be that Koreans overuse downgrading expressions and prefer decrease to others, without due consideration of the negative impact on their company’s image. In addition, informal and spoken expressions such as get, thing, and be going to (as an alternative to will) are significantly underused. Along the same lines, the high keyness of think in the negative keyword list indicates the low use of I think as a discourse marker in the Korean corpus. Considering the oral nature of the genre, the underuse of spoken expressions in earnings calls would make Koreans sound unnatural and bookish. During the Q&A session, in particular, the lack of this discourse marker as well as the nature of spoken expressions may cause native speaker analysts to feel that Koreans are too direct and sometimes even face-threatening. In such cases, native speakers may be offended and therefore terminate the conversation hastily. The consequences of miscommunication are worrisome in a highly competitive business climate, particularly as national economies seek every advantage in the current downturn. One possible solution for Korean companies might be to employ native speakers of English to participate in the calls. Throughout the presentation and the Q&A sessions, they would mediate the process and manage the discussion when misunderstanding occurs between native and Korean participants. Another solution is to teach Korean speakers to be aware of the international norms of earnings calls and improve their linguistic and pragmatic competence, which would likely prove more valuable in the long term. 5.2. Pedagogical implications and limitations The findings of this genre analysis have several pedagogical implications for business English teaching and learning. First, instructors and learners need to understand the cultural nuances of the language in business communication. In a globalized capitalist system, it is the power of dominant interests that sets the rules. In this case, American companies, self-absorbed by a can-do ideology, are intent upon putting a bright face on what could in reality be a dismal situation, regardless of the transparent premise of the earnings call. Despite the overt duplicity of the situation, it is imperative for Koreans and other nonnative speakers of English to know the implicit rules in order to play the game successfully. During earnings calls, a negative conversational tone may influence the market evaluation of the company, which could result in a financial setback, and if spread over several companies, it could possibly affect the performance of the national economy. When delivering bad news, the speaker needs to offset the negative impact by further elaboration and deliver some positive news, promoting the value of the company to encourage future investment.

184

H. Cho, H. Yoon / English for Specific Purposes 32 (2013) 170–185

Second, business English learners need to increase their awareness of the difference between written and spoken language. Inappropriate use of spoken and written language can negatively affect the image of the speaker as unprofessional and incompetent. In addition, the underuse of spoken (or informal) expressions and discourse markers makes the speech unnatural and bookish, which may increase the undesirable psychological distance between interlocutors. Third, learners need to understand differences in meaning between near-synonyms. Non-native speakers, without due consideration, tend to overuse certain expressions, which can negatively affect messages. Particularly in the business world, where a subtle difference in language may encumber congenial business relationships, nuanced differences between synonymous expressions should be included in the curriculum of business English. Effective business communication involves more than an understanding of profit-and-loss, supply-and-demand, accounting principles and such like; it also requires a cultural understanding of the social and economic context in which deals are made. Fourth, business English learners need to improve their linguistic and pragmatic ability in order to communicate clearly and efficiently in English. Business conversations, such as dialogues between corporate representatives and analysts during earnings calls, carry dense information and are jargon-laden. Learners who are required to engage in such conversation need to develop highly sophisticated linguistic and communicative skills. However, this study is not free from limitations, primarily due to the structural nature of Korean business. The high inclusion of manufacturing companies and chaebols (i.e. Korean conglomerates) in the Korean corpus may affect the result of the keyword analysis. For instance, K1, K4, and K7 are affiliates of a conglomerate in Korea, which push up the rank of the mother company name in the positive keyword list. In addition, the fluency of participants in the Korean corpus may be far above the average of Korean business people because high ranking corporate representatives are more likely to have traveled or studied abroad. In that regard, additional work needs to be conducted with less-fluent non-native speakers of mid-level and smaller Korean concerns, as they are just as firmly integrated into the world of commerce. This is particularly important as these firms may deal with English-speaking firms of commensurate size whose personnel have less experience in communicating with non-native speakers. Nevertheless, the pedagogical suggestions of this study are still valid, as the concentration of certain industries and companies in the Korean corpus was taken into account during the analysis. In addition, even though the fluency of Korean participants in this study is probably above average, there are clear differences in the data across the corpora on both a linguistic and structural level. In other words, there is still plenty of work to do in terms of advanced language instruction, particularly in the business English domain. Despite some limitations, our findings can heighten learner awareness of genre specificity of earnings calls and help them fine-tune their language to clearly communicate as professionals in international business. In addition, the findings can be applied in business English material and curriculum development in order to improve authenticity. It is also hoped that this study has shed more light on the value of earnings calls to enhance ESP pedagogy, which can empower learners of business English to actively engage in international business communication. Acknowledgements This work was supported by Hankuk University of Foreign Studies Research Fund. The authors would like to thank Dr. Stephen Zolvinski for his useful comments on an earlier draft. We are also very grateful to the ESP journal editors and anonymous reviewers for their constructive comments and valuable insights. References Aijmer, K. (1997). I think – An English modal particle. In T. Swan & O. Westvik (Eds.), Modality in Germanic languages: Historical and comparative perspectives (pp. 1–47). Berlin/New York: Mouton de Gruyter. Bargiela-Chiappini, F. (2005). Asian business discourse(s). Special issue of the Journal of Asian Pacific Communication Part, 15(2). Bargiela-Chiappini, F. (2006). Asian business discourse(s). Special issue of the Journal of Asian Pacific Communication Part, 16(1). Bargiela-Chiappini, F., & Gotti, M. (2005). Asian business discourse(s). Bern: Peter Lang. Bargiela-Chiappini, F., & Harris, S. J. (1996). Requests and status in business correspondence. Journal of Pragmatics, 28, 635–662. Bhatia, V. K. (1993). Analyzing genre: Language use in professional settings. London: Longman. Bhatia, V. K. (2004). Worlds of written discourse: A genre-based view. London: Continuum. Bhatia, V. K. (2008). Genre analysis, ESP and professional practice. English for Specific Purposes, 27, 161–174. Bhatia, V. K. (2010). Interdiscursivity in professional communication. Discourse & Communication, 4(1), 32–50. Bilbow, G. (1995). Requesting strategies in the cross-cultural business meeting. Pragmatics, 5(1), 45–55. Bilbow, G. (1997). Spoken discourse in the multicultural workplace in Hon Kong: Applying a model of discourse as ‘impression management’. In F. BargielaChiappini & S. Harris (Eds.), The languages of business. An international perspective (pp. 21–48). Edinburgh: Edinburgh University Press. Bilbow, G. (2002). Commissive speech act use in intercultural business meetings. International Review of Applied Linguistics in Language Teaching, 40(4), 287–303. Bilbow, G., & Yeung, S. (1998). Learning the pragmatics of ‘successful’ impression management in cross-cultural interview. Pragmatics, 8(3), 405–417. Bowen, R. M., Davis, A. K., & Matsumoto, D. A. (2002). Do conference calls affect analysts’ forecasts? The Accounting Review, 77(2), 285–316. Brown, S., Hillegeist, S. A., & Lo, K. (2004). Conference calls and information asymmetry. Journal of Accounting and Economics, 37, 343–366. Brown, P., & Levinson, S. C. (1987). Politeness. Some universals in language usage. Cambridge: Cambridge University Press. Catenaccio, P. (2008). Press releases as a hybrid genre: Addressing the informative/promotional conundrum. Pragmatics, 18(1), 9–31. Celce-Murcia, M., & Larsen-Freeman, D. (1999). The grammar book (2nd ed.). Boston: Heinle & Heinle. Chew Chye Lay, G. (2005). The function of ‘dúoc’ (OK; possible; can) in business communication in Vietnam. Journal of Asian Pacific Communication, 15(2), 229–256. Clatworthy, M., & Jones, M. J. (2006). Differential patterns of textual characteristics and company performance in the chairman’s statement. Accounting, Auditing, and Accountability Journal, 19(4), 493–511. Collins COBUILD Advanced Dictionary, 2008, 6th ed. Heinle & Heinle Publishers.

H. Cho, H. Yoon / English for Specific Purposes 32 (2013) 170–185

185

Crawford Camiciottoli, B. (2006). Rhetorical strategies of company executives and investment analysts: Textual metadiscourse in corporate earnings calls. In V. K. Bhatia & M. Gotti (Eds.), Explorations in specialized genres (pp. 115–133). Bern: Peter Lang. Crawford Camiciottoli, B. (2009). ‘‘Just wondering if you could comment on that’’: Indirect requests for information in corporate earnings calls. Text & Talk, 29(6), 661–681. Crawford Camiciottoli, B. (2010a). Earnings calls: Exploring an emerging financial reporting genre. Discourse & Communication, 49, 343–359. Crawford Camiciottoli, B. (2010b). Discourse connectives in genres of financial disclosure: Earnings presentations vs earnings releases. Journal of Pragmatics, 42(3), 650–663. Fairclough, N. (1992). Discourse and social change. Cambridge: Polity Press. Flowerdew, L. (2005). An integration of corpus-based and genre-based approaches to text analysis in EAP/ESP: Countering criticisms against corpus-based methodologies. English for Specific Purposes, 24(3), 321–332. Frank, A. (1966). The development of underdevelopment. Monthly Review, 18(4), 17–31. Frankel, R., Johnson, M., & Skinner, D. (1999). An empirical examination of conference calls as a voluntary disclosure medium. Journal of Accounting Research, 37(1), 133–150. Freire, P. (1993). Pedagogy of the oppressed. New York: Continuum. Granger, S. (1998). The computer learner corpus: A versatile new source of data for SLA research. In S. Granger (Ed.), Learner English on computer (pp. 3–17). London: Longman. Grice, H. P. (1975). Logic and conversation. In P. Cole & J. Morgan (Eds.). Syntax and semantics (Vol. 3, pp. 41–58). New York: Academic Press. Hadina, H., & Rafik-Galea, S. (2005). Communicating at the workplace: Insights into Malaysian electronic business discourse(s). In F. Bargiela-Chiappini & M. Gotti (Eds.), Asian business discourse(s) (pp. 121–143). Bern: Peter Lang. Holtgraves, T., & Yang, J. (1992). Interpersonal underpinnings of request strategies: General principles and differences due to culture and gender. Journal of Personality and Social Psychology, 62(2), 246–256. Jung, Y. (2005a). Power and politeness in Korean business correspondence. In F. Bargiela-Chiappini & M. Gotti (Eds.), Asian business discourse(s) (pp. 291–312). Bern: Peter Lang. Jung, Y. (2005b). The rhetorical structure of Korean business writing. In P. Gillaerts & M. Gotti (Eds.), Genre variation in business letters (pp. 347–368). Bern: Peter Lang. Jung, Y. (2012). Work orientation in Korean CEO’s on-line greetings. Online Journal of Communication and Media Technologies, 2(2), 153–181. Kameda, M. (2001). The implications of language style in business communication: Focus on English versus Japanese. Corporate Communications: An International Journal, 6(3), 144–149. Kärkkäinen, E. (2003). Epistemic stance in English conversation: A description of its interactional functions, with a focus on I think. Amsterdam: John Benjamins. Kaul, A. (2003). Talking up: Study of upward influence strategies. In Proceedings of the 68th annual convention of the association for business communication. Retrieved 01.10.12. Kaul, A., Ansari, M. A., & Rai, H. (2005). Gender, affect and upward influence. IIMA working papers. Retrieved 01.10.12. Marriott, H. (1997). Australian–Japanese business interaction: Some features of language and cultural contact. In F. Bargiela-Chiappini & S. Harris (Eds.), The languages of business. An international perspective (pp. 49–71). Edinburgh: Edinburgh University Press. McLaren, Y., & Guraˇu, C. (2005). Characterising the genre of the corporate press release. LSP & Professional Communication, 5(1), 10–30. Miller, L. (1994). Japanese and American indirectness. Journal of Asian Pacific Communication, 5(1–2), 1–19. Miller, L. (2000). Negative assessment in Japanese–American workplace interaction. In H. Spencer-Oatey (Ed.), Culturally speaking. Managing rapport through talk across cultures (pp. 240–254). London: Continuum. Nair-Venugopal, S. (2000). Language choice and communication in Malaysian business. Bangi: Penerbit Universit Kabangsaan Malasia. Nair-Venugopal, S. (2001). The sociolinguistics of choice in Malaysian business settings. International Journal of the Sociology and Language, 152, 21–52. O’Keeffe, A., McCarthy, M., & Carter, R. (2007). From corpus to classroom: Language use and language teaching. New York: Cambridge University Press. Pallas, J. (2004). Talking organizations: A narrative analysis of corporate press releases – an institutional view. In Paper presented at the 4th EURAM conference, St. Andrews, Scotland. Raab, A. (2009). Approaching ESP texts: Genre analysis of CEO statements. GRIN Verlag. Rutherford, B. A. (2005). Genre analysis of corporate annual report narratives: A corpus-linguistics based approach. Journal of Business Communication, 42(4), 349–378. Schiffrin, D. (1987). Discourse markers. Cambridge: Cambridge University Press. Schleicher, T., & Walker, M. (2010). Bias in the tone of forward-looking narratives. Accounting and Business Research, 40(4), 371–390. Scott, M. (2008). WordSmith tools (version 5). Liverpool: Lexical Analysis Software Ltd.. Smith, M., & Taffler, R. J. (2000). The chairman’s statement: A content analysis of discretionary narrative disclosures. Accounting, Auditing & Accountability Journal, 13(5), 624–646. Someya, Y. (1999). A corpus-based study of lexical and grammatical features of written business English. Masters dissertation, Dept. of Language and Information Sciences, University of Tokyo. Spencer-Oatey, H., & Xing, J. (2003). Managing rapport in intercultural business interactions: A comparison of two Chinese–British welcome meetings. Journal of Intercultural Studies, 24(1), 33–46. Spencer-Oatey, H., & Xing, J. (2005). Managing talk and non-talk in intercultural interactions: Insights from two Chinese–British business meetings. Multilingua, 24(1–2), 55–74. Tasker, S. (1998). Bridging the information gap: Quarterly conference calls as a medium for voluntary disclosure. Review of Accounting Studies, 3(1/2), 137–167. Yamada, H. (1997). Different games different rules. Why Americans and Japanese misunderstand each other. Oxford: OUP. Yeung, L. (1997). Confrontation or resolution management: Discourse strategies for dealing with conflict in participative decision-making. Journal of Applied Management Studies, 6(1), 63–75. Yule, G. (1996). Pragmatics. Oxford: Oxford University Press. Zhu, Y. (2005). Written communication across cultures: A sociocognitive perspective on business genres. Amsterdam: John Benjamins.