"Controlling" as an academic discipline: the development of management accounting and management...

Introduction and objectives

For over 30 years, management accounting and management control research in German-speaking countries, have developed in a specific way. Due to language barriers and the self-chosen isolation of most German-language researchers, who were content with their relatively large domestic market, the development of Germanlanguage research in this field (labeled controlling by German-speaking researchers and practitioners) is recognized and understood only to a limited extent. This is unfortunate, as a better understanding of the discipline and its specific development within the language area would make it easier to assess current publications by German-speaking researchers and anticipate the future development of management accounting and management control research in German-speaking countries This article therefore aims to review the development of controlling as an academic discipline in Germany, Austria and the German-speaking parts of Switzerland, as evidenced in the development of chairs at universities and journal publications as well as citations in literature.

In these countries, management accounting and control issues have been dealt with under the label controlling, a term borrowed from the English language. The term became popular in Germany only in the early 1970s It was introduced mainly as a marketing ploy by Albrecht Deyhle, a popular trainer in controlling techniques, and to start with, it was just a fashionable name for cost accounting, reporting and budgeting. Due to increasingly dynamic markets, diversification strategies and other changes in the environment (Coenenberg & Baum, 1987; Mayer, 1990a; Horváth, 2003), the scope of controlling started to encompass planning and management control processes at an operative as well as a strategic level (Becker & Messner, 2005; Wagenhofer, 2006).

In part, this development was also a response to the fact that cost accounting was often not considered a proper academic discipline but rather a prerequisite to other fields in business administration (Schneider, 1991). As a consequence, there has been a continuing debate about its definition and scope in German literature, with academics arguing in favor of a scientific programme or a framework to entrench controlling as a scientific discipline of its own (Küpper, Weber & Zünd, 1990; Weber & Schäffer, 2001; Weber & Hirsch, 2002; Scherni & Pietsch, 2004). In this context some researchers have increased the scope of controlling to cover overall managerial coordination of managerial accounting, planning and control, organizational structure and the firm's incentive system (Schmidt, 1986; Küpper, 1987; Küpper, Weber & Zünd, 1990; see also: Küpper & Mattesich, 2005). Controllers in practice also took interest in this discussion, as a narrow definition of controlling as cost accounting or internal accounting increasingly proved inadequate due to the changes in the environment mentioned above. In addition, the wider competencies and responsibilities attributed to them by academia proved helpful to legitimize their influence within compames. Despite this discussion, there is little disagreement in academia and practice on the core issues of controlling. They overlap widely with what is internationally considered management accounting and management control (Becker & Messner, 2005; Wagenhofer, 2006).

The article is mainly set in the "archive", using published articles and an empirical survey to describe and explain the development of controlling as an academic discipline in German-speaking countries Eight explorative research questions on the development of the discipline are answered. The analysis covers three perspectives that complement each other: relevant chairs at German-speaking universities (input), German-language publications (output) and citations (outcome) on the subject.

Chairs:

1. How many chairs of controlling exist and what is their focus?

2. How has the number of controlling chairs developed over time?

Publications:

3. Has the number of articles on controlling published in journals increased over the course of time?

4. What are the central topics of articles on controlling and how have these changed over time?

5. How has the share of research methods changed over the course of time?

6. How has the share of different types of authors changed over time?

Citations:

7. Which sources are cited most often?

8. Which journals are cited most often?

The questions have been formulated to guide the reader's attention to aspects that we, the authors, deem to be particularly relevant. We are aware of the fact that this nanows down the focus, but feel this is tenable regarding the current level of knowledge that exists on the development of the discipline.

While a large number of studies have been conducted to analyse English-language publications in the field of accounting (Klemstine & Maher, 1983; Beattie & Ryan, 1989; Brown, 1996; Zeff, 1996; Shields, 1997; Hussey & Cottingham, 1999; Carnegie & Potter, 2000; Cottingham & Hussey, 2000; Scapens & Bromwich, 2001; Anderson, 2002; Bjørnenak & Mitchell, 2002; Lukka & Granlund, 2002; Carnegie et al., 2003; Cannona, 2004), to the best of our knowledge, the only such studies in German-speaking countries are a publication analysis by Küpper (1993) on internal cost accounting and publication analyses by Hess et al. (2005) and Wagenhofer (2006), both for the short period from 1998-2004. Our own study is in part based on recent German-language publications by Binder and Schäffer (2005a, b) and Schäffer et al. (2006), but extends this research in several respects: first, the article takes a different perspective and aims to describe and explain the development of controlling to an international audience. To this end, it builds on some of the analyses presented in the German-language articles mentioned earlier, while modifying others and adding new ones In addition, relevant characteristics of the Germanlanguage context are described to give the audience mentioned earlier a deeper understanding of the development of controlling.

To conduct an analysis of controlling in German-speaking countries at the present moment makes particular sense, as a new generation of researchers is currently taking over. These young academics increasingly attend international conferences and publish in international journals, whereas most academics of the "first generation" published almost exclusively in German periodicals

Chair analysis

Procedure and methods used

Universities in Germany, Austria and Switzerland are traditionally organized into academic chairs, rather than the departments which are the case at most international universities. This means that there is usually only one tenured professor for each discipline, such as controlling or marketing, who is not part of a broader department and enjoys a high degree of autonomy. He or she is responsible for research and teaching activities in the faculty and is supported by assistant professors and a number of research assistants. Therefore, we feel that the development of chairs is a good indicator for the development of controlling as an academic discipline in German-speaking countries.

In this section, we show how many chairs explicitly include the word controlling and what their focus is Furthermore, we will see how the number of controlling chairs has developed over time. The study was conducted as follows: we first identified all chairs at German-speaking universities in Germany, Austria and Switzerland that explicitly use the word controlling in their title. Based on our own Internet research and the results of the study by Hirsch (2003), we identified 72 chairs (as of October 2004). Before sending a survey to the tenured professors, we compiled key data on these chairs using the Directory of University Professors of Management and additional research on the university homepages In February 2004, we sent 70 active university professors and seven with emeritus status a table in electronic form, asking them to update or complete the researched data. Based on initial feedback, we followed up this survey both in writing and by telephone. By early October 2004, we had received answers from all university professors we had written to, giving us a 100 per cent response rate.

Development of controlling chairs

Question 1: how many chairs of controlling exist and what is their focus?

Our study revealed that there are 92 business faculties (betriebswirtschaftliche Fakultäten) at German-speaking universities and 72 chairs that explicitly include the word controlling in their name. All but three identify controlling-speáñc topics as a focus of their research. In comparison, 95 chairs at German-speaking universities have a reference to marketing, of which 73 are explicitly Chairs of Marketing and 22 have related titles such as Absatz und Handel (Sales and Trade). Taking into account that the universities of Duisburg, Munster and Zurich each have two chairs of controlling, around 25 per cent of the business faculties in Germany, Austria and German-speaking parts of Switzerland do not have a chah that explicitly contains the word controlling in its title. It is notable, however, that chairs that do not explicitly use the term controlling in their title often cover aspects of controlling. Examples include the Lehrstuhl für Betriebliches Rechnungswesen (Chair of Management Accounting, Innsbruck, Albrecht Becker) and the Lehrstuhl für Betriebswirtschaftslehre, insbesondere Organisation und Planung (Chair of Business Administration, in particular Organization and Planning, Hagen, Ewald Scherm).

An analysis of chair titles shows that only 15 per cent of the 72 chairs identified are "purely" controlling chairs.The. remaining 85 per cent are connected to other disciplines (see Figure l).The large number of combinations shows that controlling can be associated with various aspects of business administration and is therefore a discipline with a classical integrating function. The most frequent combination in our study was with accounting (Unternehmensrechnung/Rechnungswesen, 36%) and auditing (Wirtschaftsprüfung, 13%). According to Hirsch, "the combination of the disciplines 'controlling' and 'accounting' in the same chair title indicates how close both disciplines are and where they originate, namely in accounting" (Hirsch, 2003, pp.255).1 Nevertheless, 36 per cent of the chairs analysed in our study are in combination with other disciplines

All controlling chairholders are German, Austrian or Swiss nationals Their average age is 51.7 years. 13 per cent are aged between 30-9 years, 31 per cent are between 40-9 years old, 23 per cent between 50-9 years old and 33 per cent between 60-7 years old. In comparison, a survey carried out by the Federal Statistical Office in 2003 set the average age of 3841 professors of economics at universities and technical universities in Germany at 49.9 years; only 20 per cent were between 60-7 years old.2

This means we can expect a new generation of younger professors over the next few years; indeed this generation switch has already started. This is particularly significant as most controlling experts of the "first generation" were not or hardly known at an international level and almost exclusively published in Germanlanguage journals

Question 2: how has the number of controlling chairs developed over time?

Before looking at changes in the number of chairs of controlling at Germanlanguage universities, we need to establish that the institutional development of controlling goes back to the appointment of Péter Horváth to the newly established Chair of Controlling at the Technical University of Darmstadt in 1973. As early as 1974, a second chair was inaugurated at the University of Duisburg, since held by Michael Wohlgemuth. Before this, chairs with other designations such as General Business Administration (Allgemeine Betriebswirtschaftslehre) or Accounting (Rechnungswesen) partly covered aspects of management accounting.

Until the end of the 1980s there was little change; by 1988 there were just 14 chairs of controlling. Since then, 14 chairs of controlling have been established in former East Germany and 44 in West Germany (see Figure 2). AU of those in the east and around half in the west were not renamed, but set up as new chairs Controlling as an academic discipline therefore also played a key role in the growth of business administration faculties (Betriebswirtschaftliche Fakultäten) at German universities (Binder & Schäffer, 2005a).

There are several explanations for the growth in the number of chairs of controlling, in particular in the last decade of the twentieth century (Weber, 1997; Küpper, 2004). A key factor is certainly the changing role of controlling in the business world (Bramsemann, 1990; Horváth, 1999). An analysis of job ads in the Frankfurter Allgemeine Zeitung - the leading daily newspaper in Germany - carried out by Weber and Kosmider (1991) and Weber and Schäffer (1998) from 1949-94 reveals a considerable increase in the number of positions related to controlling. A study by Küpper, Winckler and Zhang (1990) shows the increasing diffusion of controlling not only in major corporations but also in small and medium-sized companies (for a detailed discussion see also Kosmider, 1994; Weber & Schäffer, 1999).

This obviously had implications for universities A publication in 1987 on career perspectives for students of business administration ranked information technology (IT), human resources (HR) and controlling in the top three positions (Mayer, 1990a); a year later a Commission for Education set up by the Schmalenbach-Gesellschaft proposed controlling as a new area of general business administration.

In addition to universities Fachhochschulen (universities of applied sciences) played an important role in the dissemination of the concept of controlling in German-speaking countries Unlike classical universities, they do not have the right to award doctoral degrees. These universities focus more on practice-oriented courses based on scientific results In addition, professors at universities of applied sciences have a greater teaching load, which allows them less time for scientific work compared to their colleagues from classical universities In 2004, approximately 83,000 students were enrolled in business administration courses at universities of applied sciences compared to approximately 79,000 at classical universities.

As early as 1971, Elmar Mayer set up a working group of academics and practitioners in the field of controlling at the University of Applied Sciences in Cologne (Mayer, 1990b), the AWW Köln (Arbeitsgemeinschaft Wirtschaftswissenschaft und Wirtschaftspraxis im Controlling und Rechnungswesen). Three years later he also established a Chah of Business Administration, Controlling and Accounting (Betriebswirtschaftslehre, Controlling und Rechnungswesen). Other universities of applied sciences integrated the concept early on and some of their professors (e.g., Bramsemann in Bielefeld, Ebert in Nürtingen) gained a high reputation in the 1970s and early 1980s By the end of 1989, as many as 34 out of 40 accounting chairs at universities of applied sciences included controlling in their title (Mayer, 1990a). As the number of chairs at classical universities increased and discipline became more institutionalized, the visibility of the work of professors at universities of applied sciences and their importance for the dissemination of the discipline diminished.

Publication analysis

Procedure and methods used

Publication analysis is a method classified under bibliometry. It is concerned with the quantitative analysis of publications, authors and institutions such as libraries using statistical methods One of its key objectives is to make empirical statements on the proliferation and the dissemination of specific topics in journals and to determine any patterns.

We have limited our publication analysis to articles in academic and practitioner journals This can be problematic, as book publications are often an important outlet, especially in the early years of a discipline or for practitioners looking to publish. It is also notable that books play a more important role in German-speaking academia than in many other countries One reason for this is the fact that dissertations and habilitation theses are usually published as books (Wagenhofer, 2006).3 In our study, however, we have decided not to include book publications as this would prevent us from drawing empirically sound conclusions on the degree of penetration and the significance of controlling in German-language management literature. This is in line with an international convention that assesses scientific merit primarily on the basis of publications in recognized journals (Kalaitzidakis et al, 1999; Hennig-Thurau et al., 2003); all publication and citation analyses known to us are limited to articles in journals Therefore, broadening the scope to include other publications could lead to a certain distortion of the results - not to mention the problem of how to cope with the huge amount of data.

Our analysis encompasses articles from academic and practitioner journals in Germany, Austria and German-speaking parts of Switzerland from 1970-2003. The start of our investigations in 1970 coincides with the emergence of the German term controlling in industry and science (Matschke & Kolf, 1980; Bornemann, 1985; Vahs, 1990; Habersam, 1997).

The academic journals included in the publication analysis were selected based on a ranking of business journals put together by members of the German Academic Association of Business Research (Verband der Hochschullehrer für Betriebswirtschaft) in Germany, Austria and Switzerland, which - unlike international rankings - covers all relevant periodicals in the German-speaking world (Hennig-Thurau et al., 2003). As the leading German-language business periodicals do not exclusively address individual fields such as management accounting, we looked at the rubric for general business administration (Allgemeine Betriebswirtschaftslehre) in the ranking and chose the leading journals in that category: Schmalenbachs Zeitschrift für betriebswirtschaftliche Forschung (ZfbF) including the English-language Schmalenbachs Business Review (sbr),4 Zeitschrift für Betriebswirtschaft (ZfB) and Die Betriebswirtschaft (DBW).5 Due to their role in the development of the discipline in the eyes of experts and because of their sufficiently high ranking, Die Unternehmung - Schweizerische Zeitschrift für betriebswirtschaftliche Forschung und Praxis, Zeitschrift für Planung und Unternehmenssteuerung (ZP) and the journal Betriebswirtschaftliche Forschung und Praxis (BFuP) were also included. AU these Journals use anonymous referee processes with the exception of BFuP, whose yearly six issues are devoted to topics determined by the editors who commission articles BFuP publishes only a few unsolicited articles, which are subject to a normal review process However, it is the only German-based business administration journal that is included in the Social Science Citation Index. The ZfB and the ZfbF also publish special issues in addition to their regular issues They are usually devoted to specific topics, and some of the special issues include articles by invited authors seen over the entire period, it transpires that ZfB and ZfbF are the key academic journals for issues related to controlling.The number of articles on controlling published in these journals varies considerably from year to year, mainly as special issues are published on specific topics at inegular intervals In comparison, however, the journals Die Betriebswirtschaft and Die Unternehmung have published the fewest articles on controlling.

Four practitioner journals were chosen based on their role in promoting the discipline in the eyes of experts and their acceptance in industry: (1) Kostenrechnungspraxis (krp), renamed Zeitschrift für Controlling & Management (ZfCM) in 2003, (2) Controlling - Zeitschrift für erfolgsorientierte Unternehmenssteuerung, (3) Controller Magazin and (4) Der Betrieb: The Zeitschrift für Controlling & Management dates back to 1956. In 2003, the old title Kostenrechnungspraxis (Cost Accounting Practice) was dropped in favour of a title that incorporated the word controlling as the publishing company and the new editor Jürgen Weber found that the original title no longer adequately represented the content of the journal and did not sound attractive enough to the majority of controUers The journal Controlling was founded in 1989 by Peter Horváth and Thomas Reichmann and is usually considered to be the main competitor to ZfCM. While Zeitschrift ßr Controlling & Management and Controlling are probably the highest-regarded practitioner journals, the Controller Magazin (since 1976) has the highest print run and is distributed free of charge to the members of the Controlling Verein, the largest German organization of controlling practitioners6 The publisher is Albrecht Deyhle, who has already been mentioned earlier. Finally, Der Betrieb (since 1948) is broader in scope than the other practitioner journals and also encompasses general management and tax issues.

With the exception of Controller Magazin and Der Betrieb, all practitioner journals are edited by university professors of management accounting and related fields The majority of articles contain conceptual analyses, descriptive case studies and practical (proposed) solutions to designing and implementing management accounting and management control techniques. However, the share of articles based on research and empirical methods is larger than in comparable practitioner journals in the USA and in the UK. In this context it is also interesting to note that ZfCM recently created separate sections for research papers and practice papers, with a double-blind review of the former to attract more high quality research papers.

At this point, it should be said that a publication analysis in German-speaking countries involves looking through all the selected periodicals and recording all the relevant articles by hand - an exceedingly time-consuming process Large literature databases have been kept for international journals since the beginning of the 1960s, but articles from German business magazines have not yet been entered systematically and no equivalent databases exist in German-speaking countries.

We only included in our analysis articles from the selected journals that fulfilled at least one of the two following requirements: (1) The term controlling or controller was explicitly included in the title, the abstract or the executive summary; (2) The name of a controlling instrument was explicitly mentioned in the title, the abstract or the executive summary and explicitly put in the context of controlling in the main body of the text. Our definition of controlling instruments is based on the predominant typology in German-speaking countries as defined by Küpper (2001), who classifies a series of instruments as isolated or comprehensive controlling instruments.

This resulted in 2,529 articles (see Figure 3), which were then categorized and subjected to a quantitative and descriptive content analysis. To ensure that the articles were analysed and classified in a meaningful and stringent way based on the set criteria, we carried out an investigator triangulation (i.e., using different investigators to reduce the subjective influence of individuals and validate the findings) (Flick, 2003). In a first step, articles published between 1990 and 2003 were analysed regarding their relevance to controlling by one of the authors and a further researcher in parallel and categorized according to their content using classification criteria. In the next step, the investigators discussed the few discrepancies in detail to find a way to apply the criteria and allocate the articles to one category. The remaining articles were analysed based on the results.

In addition, we carried out an analysis of management accounting articles in English-language journals written by authors affiliated to an institution in German-speaking countries. Our selection of accounting journals follows Wagenhofer (2006) and includes nine leading international journals: Accounting, Organizations and Society (AOS), Contemporary Accounting Research (CAR), European Accounting Review (EAR), Journal of Accounting and Economics (JAE), Journal of Accounting Research (JAR), Journal of Management Accounting Research (JMAR), Management Accounting Research (MAR), Review of Accounting Studies (RASt), and The Accounting Review (TAR). All publications in a journal exclusively devoted to management accounting research (such as JMAR and MAR) were considered to be relevant to the scope of the study. For journals with a broader focus, we selected topics based on our understanding of what management accounting encompasses and using the criteria for articles in German-language journals.

Publication Activity

Question 3: has the number of articles on controlling published in journals increased over the course of time?

The absolute number,7 and the relative share,8 of publications on controlling gives us a first indication of the significance of the discipline. If we look at the absolute number of academic articles on controlling in German-language journals and the accumulated number of pages, we see that the number of articles on controlling published in academic and practitioner journals rose visibly towards the end of the 1980s (see Figures 4 and 5). In the 1990s, the number of articles published more than doubled on average. One explanation for this increase in published articles and the growing interest in the discipline could be the rise in the number of controlling chairs at universities in German-speaking countries during the same period, as revealed by our analysis of university chairs On the other hand, it could also be explained by changes in the business world mentioned in the introduction and the wider circulation of academic and practitioner journals following German reunification.

An analysis of articles in German-language practitioner journals shows a similar trend. The emergence of the journal Controlling in 1989 played a key role in the rise in the number of publications in the early 1990s Around 45 per cent of all practitioner articles recorded since its foundation have been published in this journal. The arrival and the growing circulation of this journal, which was launched into the market in the wake of Controller Magazin, Kostenrechnungspraxis and others, indicates that there was sufficient demand for an additional journal in the field. This meant that the large number of controlling chairs at universities now had a wide range of practitioner journals for their publishing and research activities They were an increasingly important medium for many academics, as many of the traditional academic periodicals could only publish a limited number of articles related to controlling.

An analysis of relative developments confirms the results: the share of publications on controlling in the 1970s and 1980s in academic journals was around six per cent on average (see Figure 6). This fell to around four per cent in the mid1980s, which leads us to surmise that controlling issues met with less interest at the time or there were insufficient scientific findings to publish. In addition, in the mid1980s, there was an intensive discussion on the Fourth, Seventh and Eighth EU Directives pertaining to the Bilanzrichtlinien-Gesetz (Accounting Directive Law),9 so that professors of chairs with combined titles in particular may have focused more on topics related to external accounting during this time. In the early 1990s, the share of publications on controlling doubled to around 12 per cent on average.

In leading international accounting journals, only 25 management accounting research articles were published between 1970 and 2003 by authors affiliated with an institution in German-speaking countries (see Table 1). Prior to 1994, only one single article was published in a leading international journal (AOS) by a German author. Since then, there has been a steady flow of one to five articles per year. The European-based journals EAR (with 10 articles) and MAR (with 8 articles) are the most-chosen publication outlets for German-speaking researchers. Three of the commonly considered top-tier international journals (CAR, JAE and TAR) have not published any articles by authors affiliated with a German-based institution.

Contents

To get a general idea of the central topics covered in the articles and how these have changed, we classified all articles according to their content. This classification was based on work by Shields (1997) and Scapens and Bromwich (2001) and compared with classifications of the most important textbooks on controlling in Germany, Austria and German-speaking parts of Switzerland to validate and hone the results (Eschenbach, 1996; Küpper, 2001; Reichmann, 2001; Weber, 2002; Horváth, 2003). As all forms of classification have their limitations, in cases where several topics were addressed in one article, we had to make a judgment on which topic the article should be allocated to. Furthermore, when classifying articles by topic, some classification criteria will inevitably take precedence over others In our classification, controlling instruments and management aspects play a key role. Topics such as strategic versus operative controlling or function and industry-specific controlling, on the other hand, take a back seat. Articles from these areas that met the specified criteria were recorded and allocated to the individual categories depending on the focus of the article. Table 2 provides a list of the individual categories and the number of recorded articles.

Question 4: what are the central topics of articles on controlling and how have these changed over time?

Controlling instruments are the central topic in over half of the analysed articles in academic and practitioner journals This result confirms normative statements by German controlling experts, who consider instruments to be decisive for the subject (Remmel, 1991; Reichmann, 1996; Küpper, 2001; Horváth, 2003). An investigation of central themes in German-language standard textbooks on controlling by Wall (2002) shows a similar picture. It is notable that monitoring features more predominantly in academic journals Practitioner journals, on the other hand, have a stronger emphasis not only on controlling applications, but also on aspects relating to changes in controlling.

Looking at changes in the content of articles over time, the share of articles on cost-accounting, budgeting and planning has dropped, leaving room for other controlling topics (see Figure 7). In both academic and practitioner journals articles on cost accounting and planning dominate until the mid-1980s. In the early years of its existence, controlling was obviously seen as just a new label for cost accounting, budgeting and corporate planning. Nevertheless, as time went by it evolved to encompass aspects such as the application of cost information or the design of controlling systems.

A closer analysis reveals that cost accounting topics dominated in the early and mid-1980s Both types of journal focused on various aspects of contribution analysis The sharp rise of other controlling topics in the mid-1980s in academic journals is mainly due to the debate at the time on applying and implementing controlling. Articles in practitioner journals, on the other hand, increasingly tackled functional aspects of controlling. Authors also wrote about calculation and pricing. During the 1990s, in addition to articles on controlling instruments, changes in controlling, such as lean controlling, played a key role. It is also notable that since the mid-1990s, the number of articles dealing with controlling concepts and theory has risen in practitioner journals. This could be on the one hand due to the limited capacity of academic journals for publishing articles on the definition and scope of controlling, and on the other to the fact that individual academics also started publishing their findings in practitioner journals.

Methods

Question 5: how has the share of research methods changed over the course of time?

In this section, we will take a closer look at the research methods used in the articles. We classified them into two main categories "analytical" and "empirical" following Grochla (1976). Analytical articles are divided into normative/conceptual approaches and formal-analytical methods. We grouped the empirical texts into the following categories based on Kaplan (1986); Shields (1997) and Scapens and Bromwich (2001): practitioner articles/case studies, research case studies, (quantitative) surveys, multivariate analysis methods, document and archive analyses, laboratory experiments and action research.

Traditionally, controlling articles in German-language academic journals were predominantly based on normative and conceptual research and dealt with the structure and theoretical background of the phenomenon controlling. For instance, several frameworks were developed and applied to a number of controlling concepts In addition, in quite a few articles from the 1970s and 1980s, the controlling phenomenon was described more generally since researchers lacked specific knowledge of how controllers work in different industries and functions In later years, descriptions of industry and function-specific manifestations of controlling started to appear and articles began to address more concrete problems In the 1990s, authors of normative and conceptual publications struggled to find the right controlling concept. In the last couple of years, the share of normative and conceptual articles dropped sharply.

Little empirical research has been carried out on controlling in the past compared to other disciplines of business administration in German-speaking countries as a study by Hauschildt (2003) shows. Based on 165 empirical articles in ZfB, ZfbF/sbr and DBW between 1997 and 2000, he comes to the following conclusion: 75 per cent of empirical research published is from the areas of marketing (29%), financial and capital market research (25%) and organizational theory and personnel management (21%). Empirical research on controlling on the other hand is under-represented with a share of just 12 per cent.

Our analysis revealed that the average share of empirical articles published by the mid-1990s was around 20 per cent. This rose to around 28 per cent on average with fluctuations in subsequent years (see Figure 8). The calls for more empirical research on controlling (Küpper, Weber & Zünd, 1990; Küpper, 1993; Brockhoff, 2002) seem to have been heard. Empirical articles tend to concentrate on institutional and formal aspects of controlling and the analysis of individual controlling instruments. Most of the articles are descriptive texts based on quantitative research methods such as questionnaires (for example, definition and scope of controUership and positioning of controlling in the organization) and document analyses (analysis of job ads).

Formal-analytical research has also become more popular in recent years (see also Küpper & Mattessich, 2005). Our analysis revealed that articles based on formal-analytical methods made up around 12 per cent of all articles on average by the mid-1990s Similar to empirical research, this rose to around 30 per cent on average, also with fluctuations, in subsequent years (see Table 2). Traditionally, articles using formal-analytical methods often deal with cost and production theory. For instance, cost theory based on analytical research provided fundamental insights for measuring costs and designing accounting systems In recent years, however, the number of articles based on cost and production theory has taken a downward turn. In contrast, since the early 1990s, articles on controlling in academic journals have started addressing issues related to information asymmetries and incentive systems We found a sharp increase in the number of articles based on the principal agent theory, most of which also use formal-analytical modelling. Today, they constitute almost 20 per cent of all academic publications on controlling. While we agree that this area of research has gained in significance, we cannot confirm Wagenhofer's claim that "analytic economic-based research has ... established itself as the dominating research paradigm ... in Germanic ... journals" (Wagenhofer, 2006).

In practitioner journals (see Figure 9), articles based on normative and conceptual work dominate over the entire period studied. The number of empirical articles has risen steadily since the mid-1970s Initially, most of these examined different aspects of internal accounting in production and trade and use practical examples or case studies In subsequent years, the focus shifted to explorative case studies on a number of controlling topics and specific application examples of controlling concepts and instruments in a variety of contexts By the 1990s, the average share of empirical articles in practitioner journals had grown to approximately 30 per cent.

Contrary to the trend in academic journals, formal-analytical articles in practitioner journals play a diminishing role. This can be explained on the one hand by the fact that accounting issues and therefore articles based on cost and production theory have become less relevant. On the other hand, articles based on principal agent theory are only published in academic journals, mainly because most researchers in this field place greater importance on scientific rigor and publication in academic journals than on direct practical relevance and transferring their findings to practitioners In addition, the increasingly complex models in this field of research are not always easy to communicate to practitioners

Authors

Question 6: how has the share of different types of author changed over time?

As well as studying the content of the articles and the methods used, it is interesting to take a closer look at the authors We classified the authors into the following categories based on a study by Bjørnenak and Mitchell (2002): practitioners, consultants, academics teaching at classical universities, at universities of applied sciences (Fachhochschulen) and at universities of cooperative education (Berufsakademien) as well as combinations of these categories. Articles whose author could not be clearly allocated due to insufficient information were not included in the analysis Authors who work in both academic and practitioner fields are allocated to the academic category. Articles with several co-authors are assigned to all authors involved, whereby all authors are weighted equally.

Articles by academics dominate in academic journals and, since 1990, in practitioner journals too. At the same time, the number of articles by academics has stayed more or less constant in both types of journals (see Figure 10). If we look at the absolute figures, however, there is a sharp increase at the end of the 1980s and the early 1990s, which can be explained by the emergence of new university chairs in the discipline and the establishment of the subject at universities (see Figure 11).

Authors from universities of applied sciences and universities of cooperative education publish comparatively little and mainly in practitioner journals In both absolute and relative terms, the number of practitioner publications by authors of universities of applied sciences has risen in the last two years In contrast the share of articles published by this group of authors in academic journals has remained low.

A further trend is the relative decline in the number of articles by practitioners The absolute figures, however, reveal that the number of articles by practitionersin academic periodicals has remained at a more or less constant level. This is surprising considering the increasing complexity of review processes and the rising number of competing academics within this field. In absolute terms, the number of articles by practitioners in practitioner journals has even risen. A closer look at articles written by practitioners reveals that the topics have become more specialized over the years Articles written in the 1970s and 1980s mostly dealt with general aspects of controlling, such as management accounting, planning, and governance. In later years, articles became more focused, for instance on specific aspects of management accounting or on particular applications of instruments and controlling systems.

The number of articles written by consultants has also grown in relative and absolute terms since the early 1990s, especially in practitioner journals While in the 1980s and early 1990s these addressed general aspects of IT as accounting support, in later years they focused on more specific applications of instruments such as balanced scorecards, activity-based costing, target costing and value-oriented control.

In recent years, articles by a combination of different types of author are featured more and more in practitioner journals and - to a lesser extent - in academic journals The combination of academics with practitioners/consultants in academic journals can often be explained by the fact that doctoral students, for instance, publish the results of their research projects together with their supervisor. At the time the article is published, these students are usually employed in companies and are therefore classified as practitioners or consultants The most common combination of authors in practitioner journals is academics with practitioners or consultants with practitioners. These publications usually deal with specific application examples of controlling instruments or systems.

Citation analysis

Procedure and methods used

A citation analysis investigates the relationship between citing and cited publications. A citation is interpreted as a measure of the significance given to the source or its author in the respective article, regardless of whether the article agrees with or criticizes the source. Although a source or author may be cited even if the author of the article explicitly denies that the cited source is relevant for his own argumentation, cases such as these rarely have an impact when evaluating large volumes of data.

The citation analysis of academic publications on controlling in Germanlanguage journals is based on the 643 articles in academic journals identified in our publication analysis,10 and is divided into two periods: 1970-89 and 1990-2003. This dichotomy is relevant as controlling as a discipline made a considerable leap in the early 1990s, as revealed in the chair and publication analyses described in this article. The purpose of the citation analysis is to answer the following questions: (1) Which sources were cited most often in the period investigated and which impact do the most-cited publications have within controlling research? The citation value is the most important indicator here. It places the frequency of citation in relation to the total number of analysed articles (2) Which journals have the greatest impact on controlling research? Impact is measured by the share of publications from specific journals in relation to the total number of all cited sources.

Citation frequency

Question 7: which sources are cited most often?

The 20 sources cited most often in academic articles on controlling in German-language journals for both periods investigated are shown in Tables 3 and 4. With the exception of the numerous cited textbooks no single source is present in both time-frames This as well as the low average age of cited sources indicates a high rate of fluctuation, reflecting the dynamics of this area of research. The most-cited sources come solely from German-language texts and, with two exceptions, from textbooks and monographs Most texts take as their subject cost accounting or investment appraisal; only Horváth's (2003) work refers explicitly to controlling. The 50 mostcited sources include just three English texts (Lutz & Lutz, 1951; Horngren et al., 1963/2002; Steiner, 1969) and only seven citations refer to articles in journals

In the second period investigated, the most-cited sources are again exclusively textbooks Whereas the most-cited sources in the first timeframe include three accounting textbooks, one investment and financing text (Schneider), one business administration textbook (Gutenberg) and one controlling textbook (Horváth), in the second period under review, three controlling and three accounting textbooks are among the top 6. This could be seen as a first indication that the discipline is developing its own profile. A look at the top 20 reveals a similar picture: Alongside general monographs and standard textbooks, numerous articles can be found that primarily address individual topics and instruments related to controlling (for example, Horváth & Mayer, 1989; Coenenberg & Fischer, 1991; Kaplan & Norton, 1996/1997).

With regard to the share of journal articles and English-language sources, differentiating between two timeframes helps us to detect developments and trends. Citations of articles from journals have gained slightly in significance; indeed, in the period from 1990 to 2003 four of the 20 most-cited sources are journal articles Compared with the citation usage in other disciplines (see Meyer et al., 2006), however, this is very low. Similarly, there is still a limited number of Enghsh-language sources in the second period from 1990-2003. However, whereas the 20 most-cited sources from 1970-89 did not include one single English-language source, the second period reviewed features two (Johnson & Kaplan, 1987; Kaplan & Norton, 1996/1997) - both books - and there are 12 Enghsh-language sources in the top 50. However, in both periods investigated, the most-cited sources do not include a single article from leading international journals on controlling and management accounting. The first source of this kind is only ranked in position 326 for the period from 1970-89 and in position 61 for the period from 1990-2003 (Cooper et al., 1981; Zimmermann, 1979). Neither do the most-cited sources include articles from leading international journals in other relevant related disciplines such as statistics or social psychology, which are often cited in articles on marketing, for example Sources of this kind are only ranked 61st in the period from 1970-89 and 40th in the period from 1990-2003 (hoteUing, 1925; Holmström, 1979). A study on French publications in financial and management accounting for the period from 1994-2003 comes to a similar conclusion in two respects: books also play an important role and articles of related disciplines such as organization or commercial law are rarely cited in French journals However, the majority of citations in France is from English-language publications, and French-language sources only account for 43 per cent of all citations (Charreaux & Schart, 2005). The focus on national sources and the self-referentiality associated with it is therefore somewhat less pronounced in France than in Germany.

Question 8: which journals are cited most often?

Table 5 shows the number of journals cited compared to the total number of all sources over the period investigated. The most-cited journals - Zeitschrift für Betriebswirtschaft (ZfB) and Zeitschrift für betriebswirtschaftliche Forschung (ZfbF) are also the top-tier German-language periodicals according to the VHB ranking. These are foUowed in fourth place for the period from 1990-2003 by Die Betriebswirtschaft (DBW); in the period from 1970-89, however, DBW was only ranked at position nine. This is partly because DBW was discontinued in 1943 and reappeared in 1977 in its 37th year. At the same time, Betriebswirtschaftliche Forschung und Praxis (BFuP), which obviously played a key role in the early years in establishing controlling in Germany as an academic discipline, dropped from third to seventh place.

It is worth noting the large number of practitioner journals cited. In the period from 1970-89 the most-cited practitioner journals were Der Betrieb (DB), Kostenrechnungspraxis (krp), Harvard Business Review (HBR) and Controller Magazin (CM). From 1990-2003, Kostenrechnungspraxis (since 2003: Zeitschrift für Controlling und Management) was the most-cited practitioner journal, while Controller Magazin dropped to place 23 with just 50 citations. A new addition is the journal Controlling founded in 1990. Another tenacious contender in the top 15 is Wirtschaftswissenschaftliches Studium (WiSt), a journal that primarily targets students, which even rose to position 11 in the second period. The analysis of French publications from 1994-2003 mentioned earlier comes to a similar result and notes that articles from practitioner journals - in particular the Revue Francaise de Comptabilité - play an important role: they are cited almost as often as articles from academic journals (Charreaux & Schatt, 2005).

The number of international journals cited remains low, although between 1990 and 2003, six of the 15 most-cited journals were international, namely the Harvard Business Review, Accounting Review, Management Science, Strategic Management Journal,Joumal of Accounting Research and Accounting Organizations and Society. In the period from 1970-89, only three Enghsh-language journals were among the top 15.

Discussion

The aim of this article was to review the development of controlling as an academic discipline in Germany, Austria and (parts of) Switzerland as shown in the development of chairs at universities and in analyses of publications and citations in the relevant literature.

Our chair analysis revealed that the discipline is still in its early years in academic circles and only became established in the majority of business administration faculties in Germany, Austria and Switzerland at the beginning of the 1990s when the number of controlling chairs started to increase significantly. On the other hand, a new generation of academic researchers and tenured professors will take over in the next few years This is significant as most academics of the "first generation" were not or hardly active at an international level and published almost exclusively in German. The new generation, however, participates more in international conferences and publishes in English-language journals.

The publication analysis revealed that the number of publications on controlling in German-speaking journals has risen sharply and the market share of controlling in leading German academic journals has grown. Analogous to the discussion surrounding the definition and scope of controlling mentioned in the introduction to this article, the content of articles published has grown beyond cost accounting, budgeting and planning. However, German controlling is and remains driven by practical applications and instruments, even in academic publications More than 50 per cent of the academic and practitioner articles address controlling instruments; in addition, most of the authors are academics, even those who publish primarily in practitioner journals

Despite its practical orientation, there are clear signs that controlling has become an established discipline of business administration at German-speaking universities One indication of this is the relative increase in the number of publications in leading German-language academic journals since the early 1990s Looking at the methods used, we found that fewer articles used normative and conceptual research approaches while empirical and formal-analytical methods were more popular. This evolution towards a more balanced mix of methods can be considered as an indicator of the discipline's increasing maturity. However, there is still considerable potential to be gained both from transferring theoretical know-how from other disciplines (Schneider, 1991; Kieser, 2003; Becker et al., 2003) and ensuring the quality of methods in controlling research (Homburg & Klarmann, 2003). Furthermore, it has to be emphasized that the debate in German-speaking countries is largely isolated from the international community, as revealed by the small number of international contributions by German-speaking authors (and German contributions by English-speaking authors).

The citation analysis also indicates that controlling is increasingly becoming an established discipline of business administration at a national level. The most frequently cited sources in the second period investigated, for instance, include fewer publications from the area of investment and business administration and more original controlling sources Furthermore, more English sources and journal articles are cited in the second period investigated than in the first period. This can be interpreted as a sign that the discipline has started to adapt to the conventions of business administration, although both figures are still low compared to other disciplines This leads us to the conclusion that, despite the obvious efforts of some researchers, the German-speaking controlling community is still largely isolated by its own choice, as reflected in the publication and citation analyses. As well as a lack of English-language sources, few publications from other associated disciplines are cited while a large share of citations comes from textbooks and practitioner journals This shows how important teaching and practical applications of controlling are but does not militate in favour of the discipline becoming institutionalized and the German-speaking community being integrated internationally. The picture for German-speaking countries is therefore similar to that painted for France in the aforementioned study by Charreaux & Schatt (2005, p.26): a focus on domestic markets and a certain self-referentiality (the latter being rather more pronounced in Germany), little impact at an international level and the domination of practitioner journals and textbooks.11

To gain a better understanding of why the publication and citation patterns of controlling researchers have developed in this way, we need to compare what observers in Germany see as the prevailing characteristics of the discipline with our findings. Controlling is considered a comparatively young discipline of business administration that is traditionally highly practice-oriented (Albach & Weber, 1991; Horváth, 1999; Kieser, 2003), but also self-referential (Homburg, 2001; Kieser, 2003) and for the most part not integrated into the international community (Kieser, 2003). They also point out that more and more German-speaking researchers of controlling are adapting to the changing rules in the market place, which should on the one hand facilitate integration into the international research community and, on the other, reduce the focus on business practice.

If we assume that this assessment is correct, it explains the results of our analyses to some extent. The diversification over time of topics related to controlling as revealed by our publication, citation and co-citation analyses is typical for the evolution of a young discipline (Kieser, 1996; Sabel, 1999). Compared with authors who focus on subjects like planning or general business administration, those who write mainly about controlling have gained in significance. This shows that the discipline is in the process of maturing. Only the importance of books in relation to journal articles does not completely fit in with this explanation. While the large number of cited books in the first period is what one would expect from a young discipline in its early years, the relative share of cited books in the second period investigated only declines slightly.

Other results can be explained by the fact that controlling is a highly practice and application-oriented discipline. Both the publication analysis and the journal-based citation analysis revealed the important role played by practitioner journals. This is typical for a discipline that has grown out of controlling practice and sees itself in close interaction with the business world, addressing problems as they arise and offering practical solutions The drop in the number of citations from practitioner journals that one would expect as researchers adapt more to the international community has only materialized to a limited extent so far. We have to assume at this point that the change in citation patterns has not taken place to the extent anticipated as defined earlier.

The low number of English sources cited can again be explained by the fact that controlling as a discipline is not integrated into the international research community to a large extent: historically, the reasons for this are complex but at the same time self-evident. On the one hand, it must be said that the language and access barriers were very high for a long time. While in the 1980s, it was not easy to get hold of English-language periodicals in Germany, Austria and German-speaking parts of Switzerland, the same journals are today only a mouse-click away and are stocked in their printed version in most libraries

However, as a brief glance at other non-English speaking countries shows, this is hardly an adequate explanation. On the one hand, the German-language home market was and is comparatively large and highly developed in the area of management accounting (see for example, Küpper & Mattessich, 2005). On the other hand, the first generation of researchers and university professors in controlling was relatively autonomous. They were initially preoccupied with legitimizing the discipline within the national business administration community and satisfying practitioners' need for problem solutions, which often led to them carrying out extensive consulting work. Finally, the fact that German faculties are organized into chairs and not departments did not contribute to reducing the autonomy of university professors or encouraging them to look beyond their own chairs

In the interim, the pressure to adapt to international rules has increased. Whereas the decision to appoint somebody used to depend on a whole plethora of factors, nowadays the number of publications in renowned journals alone is decisive, whereby international periodicals are ranked higher. Even in the ranking of journals by the German Academic Association of Business Research (Hochschullehrerverband), not one German journal is ranked A or even A+. This means that researchers are forced to look to international journals Furthermore, contrary to earher practice, the new salary scale for professors employed by the state includes a variable part, which in most cases depends heavily on publications in renowned and in particular international journals Viewed dynamically, such changes in the material framework could also mean changes in the self-perception and perceived norms of the community.

Due to the changes outlined earlier, we would expect more English citations to rise beyond the comparatively modest changes we observed in our study.

In conclusion, most of the results of our study relating to the structure and development of controlling are coherent. However, in some analyses, publication and citation patterns have not changed or are changing much more slowly than one would expect. There are three possible explanations for this: first, many of the older generation of researchers have not changed their citation patterns, or only to a limited extent. This is conceivable if we assume that researchers usually form relevant thought patterns and customs in their early years and after that get set in their ways A second explanation is that German-speaking researchers that look to the international or English-speaking community no longer publish exclusively in German periodicals This would result in the total number of controlling researchers interested in the leading German-language publications diminishing as fewer younger researchers use this medium for their publications A third reason might be the period investigated: possibly, the period selected to observe the changes in patterns was not sufficient. This is confirmed by the fact that the new generation of researchers in controlling chairs has not yet fully taken over.

It is difficult to predict future developments in controlling research in Germanspeaking countries Basically, three scenarios are conceivable: first, German-language publications continue to move toward the citation patterns common in the Englishspeaking community as more and more authors adapt and the younger generation of researchers and academics takes over. This would result in more citations of Englishlanguage and journal sources and fewer book-based and practical sources We would also expect an increase in the number of citations from leading international A and B journals and periodicals from neighbouring disciplines As a result, the Germanspeaking community and the international research community would grow closer together.

In the second scenario, only part of the German-speaking community will adapt to the citation patterns used in English-language management accounting research in future. This would mean that the number of authors publishing articles on controlling in German academic journals would further decline and that their citation usage would not change or only change very slowly. As a result, we would have one group of authors using primarily national citation patterns and another that aligns mainly with the international community.

In a third scenario, it is conceivable that the trend towards a differentiation of citation patterns might continue even in German academic journals This would mean that more journal-specific communities would spring up with their own citation patterns For communities that are geared to English-language accounting research, the line taken by ZfbF/sbr of gradually changing the publishing language to English seems to be the only strategy, as long as national distinctions of controlling do not exist or become less important (Shields, 1997). As a result, scenario 3 would be an integral part of scenario 2.

Following the hypothesis that developments in the content of controlling literature are primarily influenced by developments in the business world, we can assume that topics will continue to change at a fast pace. It is possible to predict future topics from current developments and from prognoses for controlling in practice. On a wider scale, examples of this are changes in the economic environment as a result of internationalization or globalization or changes in accounting and corporate governance (Paetzmann, 2005; Weber & Schäffer, 2006). If researchers and academics in German-speaking countries adapt more to developments in methods and theory of the international research community, however, the outcome may be different. A follow-on study is needed to clarify this

Finally, we would like to point out that the analyses described in this article have their limitations First, the evolution of a discipline cannot be defined just by studying chah, publication and citation analyses However, these analyses provide important groundwork. They give an overview of how controlling has developed and allow researchers to challenge normative statements (for example, on the significance of formal-analytical research in German-speaking countries). Finally, future research can build on the analyses One possibility might, for example, be to analyse the particular quest for identity that many controlling academics have engaged in on three different, but intenelated levels: the quest for a distinct Germanic identity of management accounting research in the international scientific discourse, the quest for identity of controlling research within German business administration and the quest for identity of researchers within the Germanlanguage controlling community itself. The latter has resulted in different schools of controlling. While our publication analysis revealed the symptoms and consequences of this quest for identity, namely the fact that international literature on management accounting is not taken into account and a traditionally high amount of conceptual contributions, future research could discuss the underlying reasons for such a quest, for example the link between academic work and corporate practice, which also became clear in this article, the normative tradition of Germanspeaking business administration in general and institutional pressures that foster the quest for identity.

Second, a comparison of our results with studies on English-language publications is complicated by the fact that some authors use different classifications (e.g., Beattie & Ryan, 1989; Brown, 1996; Zeff, 1996; Hussey & Cottingham, 1999; Cottingham & Hussey, 2000) in their publication analyses while some restrict their analysis to one journal (e.g., Scapens & Bromwich, 2001) or to single instruments (for example Bjørnenak & Mitchell, 2002). Similarly, a comparison of citation analyses is made more difficult as authors use a variety of research questions (for example, Beattie & Ryan, 1989), restrict their analysis to practitioner journals (for example, Cottingham & Hussey, 2000) and/or use different time frames (for example, Brown 1996). This means that we were only able to interpret the results to a limited extent and correlate them with English-language literature. For the earlier reasons, the groundwork presented in this article should be complemented with further research at a national and international level in order to develop a more comprehensive literature on "Comparative International Accounting History" (Carnegie & Napier, 1996).