Risk management versus operational action: Basel II in a Swedish context

Management Accounting Research 20 (2009) 53–68

Contents lists available at ScienceDirect

Management Accounting Research journal homepage: www.elsevier.com/locate/mar

Risk management versus operational action: Basel II in a Swedish context Gunnar Wahlström ∗ Department of Business Administration, School of Economics and Management, Lund University, P.O. Box 7080, S-220 07 Lund, Sweden

a r t i c l e

i n f o

Keywords: Basel II Regulation Bank managers

a b s t r a c t The New Basel Capital Accord, Basel II, promotes standards for measurement of financial and operational risk in the banking industry. Its approach to such risk measurement has been severely criticized in the literature, inevitably raising doubts concerning the effectiveness of Basel II. Using data from 25 semi-structured interviews with banking staff in four Swedish banks, the study suggests that Basel II is well established in practice, but there are significant concerns that such measurement of risk may adversely affect banks’ activities. Whilst Basel II is generally supported by banking staff who work directly with risk measurement, its usefulness is questioned by banking staff in operations. This difference between these two groups may be explained in relation to variations in their respective frames of reference. Both groups are inclined to take account of information that meshes well with their existing frames of reference and are thus more inclined to value changes that accord with their own viewpoints. One suggestion for addressing this schism within banks is to encourage a wider debate about the various approaches to implementing Basel II. © 2008 Elsevier Ltd. All rights reserved.

1. Introduction Basel II,1 the New Capital Accord issued by the Basel Committee in 2004 seeks to foster a secure and reliable financial sector (Basel Committee, 2004a), leading to improved stabilization in the security of daily payment systems. Compared to Basel I, its predecessor, Basel II goes further in regulating the measurement of financial and operational risk at individual banking institutions. However, despite its good intentions, there have been criticisms of Basel II’s approach to enhancing banking stability and security (Blejer, 2006; Danielsson et al., 2002; Heid, 2007; Jarrow, 2007). The very notion of risk measurement itself has even been questioned (McGoun, 1992, 1995). The critical literature of risk measurement (cf. Chua, 1996; McGoun, 1992, 1995; Young, 2001) is based on a logi-

∗ Fax: +46 31 786 56 19. E-mail address: [email protected]. 1 Basel II is summarized in Appendix A. 1044-5005/$ – see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.mar.2008.10.002

cal reasoning. In this study the critique of risk measurement is complemented by empirical data from semi-structured interviews with 25 Swedish bank officers involved in the implementation of the Basel II regulations. In the analysis of the interview data, both positive and negative perceptions of Basel II emerged that had implications for how the regulation was implemented. Thus, the aim of this study is to answer the question: How do bank staff perceptions define contextual outcomes within a Basel II implementation setting? Accounting regulation in practice and in management research is a problematic area. Bromwich and Hong (2000) in particular emphasise that regulation raises many difficult questions regarding application. As this is an area that often seems neglected by management research, Bromwich and Hong strongly encouraged accounting researchers to study accounting problems and issues in regulated industries. One approach is to focus on the accounting system. This is the approach these authors took in their detailed analysis of the British Telecom accounting system, the purpose of which was to set charges for interconnection.

54

G. Wahlström / Management Accounting Research 20 (2009) 53–68

Another approach, the one taken by this study, is to focus on users’ opinions of regulation rather than on the accounting system. This study is also distinct from the Bromwich and Hong study in that it presents empirical material from another highly regulated industry, in this case, the banking sector. It is claimed that “accounting is what accountants do” (Young, 2006, p. 581). In Basel II incentives exist to measure risk and to use the information acquired in internal decision making (Basel Committee, 2004b). Accountants are involved with building up such information systems and processing the resulting information for internal decisions. Accountants also use Basel II to report to supervisory authorities and to inform the market of the levels of bank risk. Thus accountants and Basel II have become closely interrelated. Like its predecessor, Basel I, it is anticipated that Basel II will be used worldwide (Basel Committee, 2004b). In comparison with Basel I, however, it is expected that Basel II will reward banks that measure risk levels2 and use this information in their daily business to lower their required capital reserves. With lower capital reserves, more capital is freed up for investment, resulting in increased profits. Banking staff from the four largest Swedish banks who were interviewed in this study expressed both positive and negative opinions of Basel II. The most frequently stated positive opinions were that the measurement of risk is deeply rooted in bank practice, that there are great advantages to measuring risk using the models some banks have already developed and were allowed to use, and that Basel II creates more efficient internal systems and better measurements for control of risk. Amongst the most commonly expressed negative opinions, interviewees stated their resistance to allowing quantitative risk modelling to unduly influence daily practice, their concern that theoretically derived models do not actually work in practice, their fear of increased centralisation, and their belief that the good practice formulations in Basel II were altogether too vague. Additionally, interviewees felt that the implementation of Basel II required excessively large amounts of resources. Clearly, the positive opinions were expressed by the category of interviewees, including risk managers and project leaders, who worked directly with risk management and assessment. Conversely, most negative opinions were expressed by interviewees engaged primarily in operational tasks. The differences in the groups can be explained in terms of frames of reference. Consistent with typical human behaviour as shown in prior research, these interviewees were most likely to accept changes that supported their pre-existing beliefs (Bateson, 1972; Hedberg and Jönsson, 1978; Jönsson and Lundin, 1977; Young, 2003). An additional intent of this study is to contribute to the critical literature on accounting numbers. This literature criticizes the role of the researcher in society and calls for the academic community to take a greater role in improv-

2 The author does not claim Basel II is only a statistical tool since Basel II also requires qualitative information. In comparison with Basel I, Basel II emphasises and rewards measurement of risks.

ing practice (Briloff, 1993; Lee, 1995, 2006; Sikka et al., 1995; Tinker, 2002; Tinker and Carter, 2002; West, 2003; Willmott et al., 1993). The paper is structured as follows. The Basel Committee and the rationale underpinning Basel II are summarized, followed by a critique of its measurement of risk approach and then by an outline of the method. Thereafter, the paper presents and analyzes the interviewees’ opinions. The research findings are then presented and discussed. The paper concludes with a proposed solution for overcoming the problems that potentially arise owing to conflicting bank officers’ opinions of Basel II. Finally, the paper makes suggestions for future research in the area. 2. The Basel Committee and the rationale underpinning Basel II 2.1. The Basel Committee and the Basel Accords Financial crises create demands that ultimately lead to changes in regulation (Flesher and Flesher, 1986; Merino and Neimark, 1982; Tinker, 1997). After a period of turmoil on the financial markets, most notably due to the failure of the small German bank, BankHaus Herstatt, the Basel Committee on Banking Supervision (hereafter, the Basel Committee) was established in 1974 (Basel Committee, 2004b). The largest economies in the world, as represented by their central banks or other banking authorities, are amongst the Basel Committee members, including Canada, Japan, the United Kingdom, Germany, the United States and France. With the appointment of individuals well-known in the international banking community to the chairmanship of the Basel Committee, the credibility of the Basel II work has been greatly enhanced. Currently the Basel Committee Chairman is Mr. Nout Wellink, President of Netherlands Bank, who succeeded Mr. Jaime Caruana, Governor of the Bank of Spain, in 2006. The objective of the Basel Committee is to improve the understanding and the quality of banking supervision worldwide (Basel Committee, 2004a, 2006a,b). However, the Basel Committee has no formal, supranational supervisory authority; instead, it formulates broad supervisory standards and guidelines that may then be implemented by national authorities. In this way, recognition of differences in national banking systems is taken into account. Thus the likelihood of acceptance of Basel II regulations is increased since they can be modified to fit users’ national banking systems (Basel Committee, 2004a, 2006b). In recent years the Basel Committee has focused on the capital requirements that banks need to protect against various types of financial and operational risk (Basel Committee, 2004b). Beginning in the 1980s, the Basel Committee became concerned about the capital ratios of large, international banks. Because of this concern, there was strong support amongst the Basel Committee members for strengthening the stability of the international banking system and for creating the conditions for equal competition amongst banks by recommending similar capital requirements across borders. As a result, in July 1988, the Basel Committee released the first Basel Accord entitled Basel I (also known as Basel 1988) that is used by most banks

G. Wahlström / Management Accounting Research 20 (2009) 53–68

around the world (Basel Committee, 2004b). Basel I implemented a framework that called for a minimum capital standard of 8 percent by the end of 1992. According to Cerny (1994), Basel I was at that time the most important regulatory agreement achieved as a result of international negotiations. In June 1999, a proposal for a new capital framework was published by the Basel Committee. After several drafts, refinements and extensive communication with the industry, the new capital framework, the second Basel Accord, Basel II, which develops Basel I’s standards on risk and capital management, was released in June 2004. Since the release of Basel II, the G-10 nations have worked to implement its requirements in their respective countries and many other non-G-10 countries have publicly endorsed the Basel Committee’s work. It thus appears that Basel II has the solid potential to be well accepted by the banking sector worldwide. However, the Basel Committee has repeatedly emphasised that Basel II should not be regarded, in any sense, as a final solution since it will continue to pursue the development and standardisation of international banking regulations (Basel Committee, 2004b). 2.2. Arguments for Basel II The Basel Committee has presented several arguments for Basel II including an argument related to globalisation (Basel Committee, 1999). The consensus was that Basel I did not respond adequately to the forces that globalisation creates. As an example, the Committee referred to the East Asian Financial Crisis of 1997. Although this crisis did not affect the banks from the G-10 countries directly, nevertheless it revealed a changing global context for which Basel I was unable to provide an adequate response. A second argument for Basel II is that it better reflects banks’ underlying risks and better responds to the financial innovations of recent years, such as securitisation (Basel Committee, 2004b). The Basel Committee (1999) describes securitization as a shift of concentration in international portfolios to lower quality assets. Basel I provided relatively weak protection for the financial distress that originates from such securitisation. Third, it is argued that Basel II reflects improvements in risk management practise that are not evident in Basel I. In summary, Basel I was considered too blunt an instrument to meet the demands of the complexity of risk calculation and decision making in contemporary banking practice. For instance, Basel I, it has been claimed, was insufficiently calibrated to the default risk of different borrowers (Basel Committee, 1999, 2001a,b). The aim of Basel II is thus to create a more refined measurement of risk. According to the Basel Committee (1999), the more refined measurement of risk outlined in Basel II can solve a number of problems facing the banking industry. Unlike Basel I, Basel II also takes notice of differences amongst banks. For example, if an individual bank can prove to its national banking supervisor that its risk measuring models are sufficiently advanced, that bank will be rewarded with a lower capital reserves requirement. The benefit to the bank is that, as a result of the reduced capital requirement, its yields will increase. Clearly, banks have an

55

incentive for developing advanced risk measuring models. For its part, the Basel Committee argues that more refined and sophisticated measures of risks will lead to a safer, more secure and more efficient banking systems since the regulatory capital will be much better aligned to banks’ actual degree of risk (Basel Committee, 2001a,b). However, banks that cannot develop their own models to measure risk are obliged to use a general requirement. To conclude, the traditional approach to measuring risk is still strongly supported by the Basel Committee and is enshrined in Basel II. 2.3. The silence surrounding Basel II The Basel Committee has devoted much energy to the development of arguments designed to persuade financial institutions around the globe to embrace Basel II and to promote its universal acceptance. It must be said that the arguments advanced in support of Basel II are persuasive. Indeed, no responsible bank officer could be critical of arguments for a system designed to promote “safer”, “more secure” and “more efficient” banking. In addition, by working through consensus, the Basel Committee has achieved acceptance of its work by its methodical and indeed painstaking process of gathering information widely, listening to interested parties, encouraging dialogue and agreeing to many outside proposals and suggestions for modifications. Furthermore, as the Basel II text shows, the Basel Committee has gained the backing for Basel II from authoritative bodies such as central banks and supervisory authorities around the world as well as from prominent individuals at such organisations. Thus, the Basel Committee has worked to earn the trust of global banking regulators, and, in so doing, has created an environment of well-intentioned and well-supported “good regulation”. Basel II supporters argue that this second Basel Accord allows banks to foresee and plan for future events, thereby creating an expectation of protection and insulation from the harmful effects of banking sector failures (Calem and LaCour-Little, 2004; Pederzoli and Torricelli, 2005; Peuro and Jokivuollo, 2004; VanHoose, 2007). Unable to challenge this argument effectively, Basel II critics limit their objections to specific details of the regulations, for example, the issue of the proper approach to risk measurement. In effect, the Basel II critics cannot diverge too radically from the predominantly positive discourse that accepts the work of the Basel Committee on Basel II. As a result, as Young (2003) argues, the support of respected organisations and prominent individuals combined with the very persuasive arguments in favour of Basel II, with its potential to provide financial security and stability, has silenced more fundamental criticisms. 3. The criticism of the measurement of risk approach 3.1. Assumptions underpinning risk measurement models The criticism of the measurement of risk approach in Basel II is based on the assumptions that underpin

56

G. Wahlström / Management Accounting Research 20 (2009) 53–68

measurement theory. McGoun (1995) argues that if the three assumptions he discusses were revealed and made fully transparent, then the models for measuring risk would soon be abandoned. These three assumptions are explained next. The first of McGoun’s assumptions is that historical experience is a basis for forecasting the future. The problem, of course, with this assumption is the proper identification of the historical conditions, if any, that are sufficiently similar to current conditions to project the future. Indeed, other critics have expressed this objection in several mainstream journals (e.g., Basak and Shapiro, 2001; Blum, 1999; Crockett, 2000; Danielsson, 2002; Danielsson et al., 2002, 2004, 2005; Danielsson and Zigrand, 2008; Morris and Shin, 1999; Persuad, 2000; Power, 2004). Such critics argue that risk modelling is based on a fundamental misunderstanding of the statistical properties of risk. This misunderstanding results because of the basic assumption in statistical risk modelling that the statistical properties during stable times remain the same during times of crisis. However, working on the premise that history repeats itself is a highly problematic, if not flawed, basis for risk modelling. For instance, during stable financial times, some investors buy whilst others sell (i.e. people act individually). By comparison, during times of financial crisis, investors act much more similarly, for example, generally searching for safer investments (Danielsson et al., 2004; Danielsson, 2002; Dunbar, 1999; Jacobs, 1999). In short, as this example suggests, how investors act in one period is not necessarily determinative of how they will act in another (Basak and Shapiro, 2001; Blum, 1999; Collins, 1985; Crockett, 2000; Danielsson, 2002; Danielsson et al., 2002, 2005; Danielsson and Zigrand, 2008; Morris and Shin, 1999; Persuad, 2000; Power, 2004; Young, 2001). Two striking examples of the unreliability of history as a predictor of the future are the Russian financial crisis of 19983 and the worldwide stock market crash of 1987. In mid-1998, most financial institutions used the same models for risk measurement and risk constraints due to regulatory considerations (Danielsson et al., 2004, 2005; Danielsson, 2002; Dunbar, 1999). When the Russian crisis hit, the volatility of the capital markets rose and Russian banks exceeded their risk limits. There was a flight from volatile to stable assets that created a price movement that drained liquidity from the market as financial institutions continued using trading techniques based on similar models for risk measurement and risk limits. Thus the crisis escalated. In the 1987 stock market crash, portfolio insurance was used widely with program hedging strategies for futures contracts to control downside risk (Danielsson, 2002; Jacobs, 1999). The portfolios worked well under the stable pre-crisis conditions when the futures markets functioned. However, when the financial institutions used portfolio insurance to execute similar program

3 One company that achieved notoriety during the Russian crisis was Long Term Capital Fund (LTCF). In this company, measurement of risks was in use, but evidently too much reliance was placed on such information as LTCF later failed (Dunbar, 1999; Furfine, 2006; Jorion, 2000; Young, 2001). Large financial institutions led by the Federal Reserve Bank in New York had to rescue LTCF from liquidation (Greenspan, 1998).

trading strategies during the crash, the futures markaced on such information as LTCF latet ceased working properly, and the crisis escalated. When the program strategies were abandoned, the market recovered quickly. The lesson of these two stories is that it cannot always be taken for granted that risk can accurately be measured, controlled and managed in a future period using models and methods that are successful in a previous period (Chua, 1996; Collins, 1985; Danielsson et al., 2004, 2005, 2004; McGoun, 1992, 1995; Power, 2004; Young, 2001). The second assumption related to risk measurement that McGoun criticizes is the tendency to neglect the “law of large numbers”. This is an assumption based on the notion that the mean, or expected value, will occur in the long run, with the consequence that the standard deviation, expressed as the element of risk, will become obsolete. Finally, and third, McGoun points out the problem with the assumption of estimation. Investors who must choose between two investments, both with a mean yield of 15% and a standard deviation of 10%, might still prefer one investment to the other. Because of expectations about future events, an investor may see a greater “risk” with one investment than with another. Consequently, in order to understand why a certain investment is chosen and not another, we need to understand each investor’s reasoning; and this, of course, is impossible. In order to be effective for the purposes of day-to-day operational control and management, the assumptions that underpin risk measurement models must describe reality. If they do not, such models risk breaking down due to the discrepancy between their underpinning assumptions and the de facto, situational reality of the here and now (Basak and Shapiro, 2001; Blum, 1999; Crockett, 2000; Danielsson, 2002; Danielsson et al., 2002, 2004, 2005; Danielsson and Zigrand, 2008; Morris and Shin, 1999; Persuad, 2000; Power, 2004). When such discrepancies occur, numbers are unlikely to provide any guidance in decision making, except, in the best cases, as points of departure for initiating discussion. Thus, given that employing such models has only marginal usefulness, it is arguable that management time might be more profitably spent on other tasks. In the extreme, it has even been explicitly and categorically stated that the models for quantifying risk have reached a theoretical dead end (Boyle, 2001; McGoun, 1992, 1995). These models are still honoured and used, however, because both theoreticians and practitioners have vested interests in the abilities of models to reduce the complexity of asset valuations, thus making them suitable for a wide range of purposes in accounting. 3.2. Numbers and the financial sector Numbers and calculations are strongly rooted in the financial sector (Blum, 1999; Basak and Shapiro, 2001; Crockett, 2000; Danielsson et al., 2004, 2005; Chua, 1996; McGoun, 1992, 1995; Morris and Shin, 1999; Pederzoli and Torricelli, 2005; Persuad, 2000; Peuro and Jokivuollo, 2004; Porter, 1995; Power, 2004) and are able to summarize complex relationships using ratios and other quantitative measures. Typically, such measurements may easily be read and understood by informed users. However, moral and

G. Wahlström / Management Accounting Research 20 (2009) 53–68

ethical aspects of financial and accounting activities are sometimes obscured by such calculations. Interestingly, as Porter points out, when numbers were first introduced as the basis for decision making, statistical modelling experts were more reluctant to use them than other finance professionals because numbers were seen as too blunt. Nowadays, generally convinced of the superiority of financial models, fewer people challenge the use of numbers. If most finance professionals are convinced of the superiority of a particular approach, especially if they have invested time and reputation in developing models, it will be adopted, perhaps irrespective of the overall cost to society if the approach proves inadequate or faulty. Chua (1996) explains that the reason for the dominance of the quantitative paradigm is fundamentally that being responsible for measuring risk is a respected, even envied, assignment in many and various levels of society. For example, the widely held perception is that an investment banker’s work is more challenging, as well as more glamorous and more remunerative, than that of the run-of-the-mill accountant or bookkeeper. University academicians who publish their quantitative, deductive research in peer-reviewed, scientific journals teach their students to appreciate quantitative reckoning. When these students enter practice they are keen to apply this theoretical knowledge to practical accounting matters. According to Chua (1996), the tradition of calculation is dominant not because of its inherent superiority, but because people have been persuaded of its superiority: “So it could be that a calculative tradition in research and education thrives not because of its technical efficacy but because of its embeddedness within a particular set of contradictory social relations” (p. 144). In the critical literature, numbers have sometimes been described as “appealing”. This appeal derives from the perception of visibility created by numbers, resulting in the notion that facility with numbers leads to the exercise of control (Broadbent and Laughlin, 1994; Carnegie and Napier, 1996; Funnell, 1998). Thus modes of control can be constructed that are geared to specific activities and resources within organisations (Hopwood, 1983; Jönsson, 1996; Munro, 1995). Thus, the manager who desires to exercise control by making responsible and justifiable operational decisions finds that numbers provide an appealing tool. With its emphasis on measuring risk that is the cornerstone of Basel II, its calculations automatically lead to the generation of numbers. In turn, this means that the methodology promoted by Basel II goes hand-in-glove with the conviction that number crunching is required to exercise control and to make fact-based and defensible decisions. 3.3. Numbers and researchers Yet there is a body of literature that is critical of an over-reliance on numbers. Such criticism offers a variety of perspectives on the inadequacy of numbers, one of which directly concerns the role of researchers in society (Briloff, 1993; Lee, 2006; Sikka et al., 1995; Tinker, 2002; Tinker and Carter, 2002; West, 2003; Willmott et al., 1993). This

57

literature has specifically called for the academic community to take a greater role in promoting and facilitating improvements in practice. The argument is made that the academics’ role in society has changed (Lee, 2006) and that often university education has little practical application to everyday practice (Briloff, 1993; Chua, 1996). At the beginning of the last century, practitioners with deep, real world experience taught accounting classes (Canning, 1929). Moreover, until the late 1960s, accounting academics continued in this practical tradition by conducting deductive research and providing normative descriptions of accounting activities (Fleming et al., 1990, 1991). However, generally speaking, from the early 1970s onwards, accounting academics have withdrawn from this pragmatic approach to the subject, preferring to concentrate on two other related areas of interest (Fleming et al., 2000), as described next. One area of interest for academics today is research into objective observations of accounting, without offering any solutions to problems encountered (e.g., Watts and Zimmerman, 1986). Generally, this research concentrates on accounting numbers and capital market activity, measured by share price and volume movements. The claim is that such research thus investigates the use, and not the usefulness, of accounting numbers (Lee, 2006; Sterling, 1990). Critics of this research approach argue that this tradition has neither the prospect, nor the intention, of changing reporting practice (Bricker et al., 2003; Sterling, 1993). The second area of recent research interest for academics, which has been equally criticised, is charged with being inward looking, and, as Lee (2006) claims, aims more to impress fellow academics (and tenure committees) than to influence practitioners. Using The Accounting Review, the American Accounting Association’s premier journal, as an example, Lee concludes that the research published in such journals is unlikely to influence practice. Instead, the objective of these publications is to confirm scientific standards in design, methodology and presentation, rather than to promote the criterion of practical usefulness. The result is that whilst research increasingly satisfies demanding standards of academic enquiry, it fails to challenge practice. The question that thus needs to be addressed concerns how such a narrow and one-dimensional research approach can be reappraised and revitalised. One feasible way may be to return to practice and, together with users and practitioners, initiate discussions designed at finding solutions to ‘real’ problems. In management accounting research at least there is evidence of a movement to relate the research more directly to practice. Although it is beyond the scope of this study to review all the literature that has been produced on this topic, a few examples of this research are representative of the new direction. Consider, for instance, the criticism of the lost relevance of standard cost accounting technologies (e.g., Cooper, 1990; Johnson and Kaplan, 1987; Jönsson, 1998). Or note the increased relevance of the case studies technique in research (Hopwood, 1983; Tomkins and Groves, 1983; Scapens, 1990). Or take account of the criticism of mainstream statistical techniques (AAA, 1977; Chua, 1986; McGoun, 1992, 1995) that make longitudinal studies a counterpart (Bhimani, 1993; Dent, 1991).

58

G. Wahlström / Management Accounting Research 20 (2009) 53–68

The evidence of this refocus on the practical in accounting research, growing out of a sense of a loss in relevancy, calls into question established research norms and practices. For purposes of this study, this movement has implications for the practical implementation of Basel II that depends on its ability to relate its theoretical justifications to managers’ actual work. This makes an investigation of bank officers’ perceptions of Basel II implementation highly relevant. 4. Research methods This paper adopts a social constructivist approach that views knowledge as arising from and created during interactionism, as opposed to existing independently (Berger and Luckmann, 1966; Jönsson, 1998; Young, 2006). Berger and Luckman argue that when people interact, they do so with an understanding that their respective perceptions are related. As people act with this understanding, their knowledge of reality is reinforced. Their perceived reality is created when new experiences confirm or invalidate previous predictions. Thus, as people create knowledge, it becomes part of their objective, socially constructed reality. With respect to Basel II, then, the issue is not only the matter of measuring risk using rigorous, statistical models, but also the issue of how the regulation itself is perceived by those who use it, the bank managers. According to social constructivism, the results of research are always situationally dependent. Furthermore, there is the explicit recognition that the reality we experience today may very easily be different in the future (Young, 2006). The aim of risk measurement is to forecast the future. However, the fundamental assumptions that underpin the theories of risk measurement are all too often treated as “givens” and are rarely reassessed in the light of the prevailing social and economic circumstances (McGoun, 1995). Accounting studies that are grounded in interactionist theory question such “givens” and challenge the latent, and often unarticulated, assumptions that underpin the constructions of numerical models used in regulation. By addressing these issues that are usually overlooked and by conducting studies that have a different theoretical point of departure, findings may be uncovered that allow us to raise questions that might not otherwise have been apparent. By adopting a critical stance and questioning some of the fundamental ideas that form the foundations of regulation, possibilities emerge from which new insights can be gained and new forms of knowledge created. For purposes of this study, the problem that must be addressed is that the critical literature has cast doubt on the validity of the models used to measure risk (Chua, 1996; McGoun, 1992, 1995; Young, 2001). This is problematic since Basel II is intended to reward banks that create and implement numerical models for measuring risk in their operational decision making. Given the severity of the criticisms of quantitative models in the literature, it is of interest to learn how users of such models view them in practice. To investigate the opinions of banking staff involved directly or indirectly with such models in daily routines, the researcher, of course, must go out into practice.

There are several perspectives to be aware of when interviewing staff in organisations (Townley et al., 2003). From an ethical perspective, it is important not to dismiss an interviewee’s reasons for, and explanations of, particular decisions. Whether the interviewee’s statements are true or not, we should regard them nonetheless as sincere. From a methodological perspective, the assumption that the statements are sincere allows the researcher to consider why the interviewee presents them as such. Finally, from a political perspective, recognition must be given to the practical considerations of workplaces that affect an interviewee’s responses. In this study, it is argued that a common frame of reference is established that makes staff at banks believe in numbers and, in particular, their predictive ability. It is important to give voice to the bank officers since they are at risk of being held responsible for “weak” performances, due to a high capital requirement or, in a worst case scenario, bank failure, due to insufficient capital cover. 4.1. Selection of interviewees This study is based on empirical data from 25 interviews with staff from four publicly listed banks in Sweden. With regard to operational structure, three of the banks used a centralized management structure and the fourth used a more decentralized structure. The author selected interview candidates from annual reports and enquiries made at the banks. When a potential interviewee was unavailable, often that individual recommended another person. The author conducted the interviews at the banks’ offices in the autumn of 2005, using a tape recorder. Each participant was interviewed once for a period of 30–40 min, and the interviews continued until the answers were repeated and additional comments would no longer be of value (Glaser and Strauss, 1967). Interviewees held various organisational positions including chief financial officer, business area manager, head of auditing, head of internal control, project leader for Basel II, head of risk management, and head of credit risk. This wide distribution of employee positions allowed for the collection of a broad range of empirical data about opinions of Basel II at the banks. 4.2. The interview questions In Basel II, incentives are provided for risk measurement by allowing for lower capital requirements. The critical literature of risk measurement is based on a logical reasoning (cf. Chua, 1996; McGoun, 1992, 1995; Young, 2001). As a complementary approach to such investigation, this study uses data from face-to-face interviews. A study of Basel II in practice, as perceived by users, provides a new avenue of investigation. To carry out this research, broad questions designed to elicit users’ opinions of Basel II implementation were formulated for use in the interviews. These semi-structured interviews allowed the interviewees to give extensive responses to general questions in a conversational format (Hartley, 2004; King, 2004; Kvale, 1983). Furthermore, this semi-structured interview approach allowed the author to pose additional, spontaneous questions as information was disclosed.

G. Wahlström / Management Accounting Research 20 (2009) 53–68

As the purpose of the study was to analyze bank officers’ perceptions of Basel II, the prepared interview questions required the interviewees to address their personal knowledge of and experience with the implementation of the regulation. The prepared interview questions, posed to all interviewees, were as follows: How does the banking community perceive the new capital requirements? Does the general perception differ from your own view? In your experience, what are the strengths of the new capital requirements? What are its weaknesses? 4.3. Analysis of the interviews The use of semi-structured interviews produces extensive empirical data, both in terms of breadth and depth. In order to demonstrate how the findings have been reached in this study, a specific method for interview data analysis was used (Bryman and Bell, 2003; Hartley, 2004; King, 2004). Immediately after each interview, using five or fewer keywords, the author summarized the interview in a field book in order to highlight and link the principal ideas gained from all interviews. Within a day or two after each interview, the author created an interview transcript4 from the tape recording. In reading the transcripts, the author assigned a key word to each paragraph that described its content. This procedure provided an overview of the interviews in total, leading to the next step in the analytical process in which the following two post hoc analytical categories were created: The positive opinions expressed of Basel II, and The negative opinions expressed of Basel II. Using these two categories of opinion, the author reread the transcripts and prepared a final summary analysis. In brief, the interviewees expressed strong overall support for Basel II and regarded the regulation, without exception, as a step forward. However, there were strong negative opinions of various aspects of Basel II as well. The following two sections present and analyze the positive and negative opinions in both table and text format. 5. The positive opinions of Basel II The general impression of Basel II by the interviewees was favourable. The specific reasons for approval are described in Table 1. 5.1. Confirmation of current banking practice The interviewees generally believed that Basel II improved the banking investing environment. One principal reason was their belief that Basel II released capital, resulting in more investments and greater profitability.

4 ‘Interviewees’ responses quoted in this study have been translated from Swedish to English by the author.

59

Table 1 The positive opinions of Basel II. 1. Confirmation of current banking practice. 2. Measuring risks with the banks’ own models. 3. More efficient internal systems and control of risk. Bank staff at the centralised banks and staff working primarily with risk measurement expressed these positive opinions most frequently.

They believed Basel II allows capital reserves that better match the actual capital requirements of banks. A project leader for Basel II explained this positive opinion of Basel II: “I think that one fundamental aspect of the banking industry is that the old rules went out of fashion and did not disclose the risk that was actually there. It is a common viewpoint in the banking industry that there is a need to refine the risks and to set a capital requirement that better corresponds to the risks in the bank’s operations . . . and there has been a desire amongst the banks to work internally to get a better understanding of what capital the banks actually need in relation to the risks and the desired level of risk in the banks. It is not a good thing if your internal measurements come to the conclusion that you need a certain amount of capital to be able to cope with particularly serious events and that the supervisory requirement is at a totally different level. It is obviously a great advantage if the methods you use internally can also be used to measure the legal capital requirement.” (Project leader for Basel II) This comment reveals strong support for Basel II. Regulation can to some extent be seen as restricting behaviour (Bromwich and Hong, 2000; Danielsson et al., 2005; Jönsson, 1991; Puxty et al., 1987), and if new regulation conforms to already established practice, banking staff have few reasons to object to it, as no significant changes are required. Thus, if a new regulation confirms the already established banking practice, the regulation is acceptable. 5.2. Measuring risk with the banks’ own models Additionally, the interviewees were very positive towards risk modelling and the ability of such models to quantify levels of risk. There was a strong consensus that such models were overwhelmingly beneficial for the banks. The emphasis on risk measurement in Basel II was regarded as being more in line with the actual situation for the banks, and thus the interviewees were positive about the opportunities for banks to use their own models for risk measurement. Risk measurement is a fundamental activity for banks. Therefore, for some banking staff – those comfortable with and used to models for risk measurement – such models are natural and unproblematic (Chua, 1996; Knights and Vurdubakis, 1993; McGoun, 1992, 1995; Porter, 1995). The head of the risk management department at one bank stated: “With Basel II, we have a more refined way of measuring risk. This would have continued in spite of Basel II since the industry has always had an interest in being more skilled in judging the risks for different customers

60

G. Wahlström / Management Accounting Research 20 (2009) 53–68

by measurement techniques. With Basel II, you have an additional incentive to develop your own models, which is something the regulator also wants. The bank should have advanced models because the banking supervisor may then set a lower capital requirement.” (Head of risk management) This comment indicates both approval of Basel II’s own risk measurement techniques and of the incentives Basel II offers for developing such techniques internally. The interviewee recognizes the positive goals of a regulation that provides for or encourages developing risk models that permit lower capital requirements that lead to increased yields, tighter controls and better decision making As Solomons (1983) observes, if a regulation is to be accepted broadly in practice, the regulation must be useful for control and decision making. 5.3. More efficient internal systems and the control of risk A risk measurement model is a powerful tool since it enables abstractions of a complex reality (Chua, 1996; Hirst, 1981; Porter, 1995; Thompson, 1967). The listed banks in this study have a wide diversity in operations in terms of both market segment and geographic area. Therefore, the language of mathematics is a useful communication tool when a variety of banking activities and staff are involved. A Basel II project manager said: “With Basel II, we will improve our judgement of our portfolio. As we have branches in different countries, such as Finland, Norway and Sweden, Basel II will help us create a similar foundation for judgement of our profitability. Our calculations on the profitability and the business operations will improve. As well, we can use the information with customers to explain why we need to charge a certain rate. Our credit policy will become more visible since we can explain to the customers that the loans will require a certain amount of capital reserves for which there is a cost to the bank.” (Project manager for Basel II)

by several experts. What I want to say is that the risk measurement is based on information developed ourselves and is presented by people from headquarters.” (Division manager) Those bankers who support these numbers have speciality training, deep experience and high-ranking positions at headquarters. These qualifications and job descriptions ensure that these individuals possess essential knowledge and power (Chua, 1996; Shapin, 1984; Verrecchia, 1982). The selection of such people to supervise the Basel II implementation can sustain a belief in a bank that risk measurement is the right way forward. The interviewees believed better risk measurement would better control banking risk. The assumption is that “properly” measured risk will automatically result in the generation of useful information. The usefulness of such information will be the natural by-product of a more “correct” measurement of risk (Young, 2006). One division head explained: “The positive element, as I see it, is that the new rules create much better and much more transparent control over risks. And I can say that, in our division, we have worked with these questions long before Basel II was introduced for larger corporations. So now we are communicating our ideas to other divisions in the bank. If I look upon what the bank has achieved to implement those ideas in the whole bank, I see much better control of the risk situation. You have processes that are linked to risk handling, more risk-based pricing, and the possibility for analyzing changes in risk in a totally different way. From our bank’s perspective, Basel II has created far better controls, lower risks of credit losses and improved possibilities for setting a better price for the not-so-good customers, Thus, it is possible to differentiate between customers and have risk-based pricing.” (Division head)

“I rely on the information 100%. This [Basel II implementation] is an enormous project in the banks and we have spent a lot of resources over several years and have placed our best person as head of the project supported

Basel II’s statistical approach appealed to many of the bank staff interviewed, especially those working closely with Basel II, such as risk managers and project leaders for Basel II and those with similar functions. The interviewees’ support for Basel II was more frequently expressed by officers working in the centralised banks.6 Thus, positive opinions were more common amongst staff at three of the four banks included in the study. To conclude, the consensus was that Basel II’s emphasis on the measurement of risk using numbers imposed a degree of uniformity on the banks. Uniformity in this area was viewed positively by the interviewees who believed it improved control of risk across different operations and on equal terms. For people convinced of the advantages that accrue to centralized organisations, uniformity and improved systems of control are appealing, especially so in banks where operations span a vast variety of different activities and geographic locations. Thus, the value of centralised management practice is enhanced by Basel II since

5 Jorion (2002, p. 911) describes the VAR measure in the following way: “A company disclosing, for instance, that its daily VAR is $30 million at the 95% level, means that there is only a 5% chance the firm will incur more than a $30 million loss over the next day.”

6 From a historical perspective, the most decentralised bank in Sweden, Svenska Handelsbanken (SHB), has also been the most successful in terms of yield. SHB was the only bank that did not have to apply for funding during the financial crisis in the beginning of the 1990s (Wallander, 1999).

The interviewees revealed an innate confidence in specific numbers, such as those calculated using risk measurement models, one of which commonly used in practice is Value-at-Risk5 (VaR) (Basel Committee on Banking Supervision, 1996; Danielsson, 2002; Danielsson et al., 2002; Szegö, 2002). Thus calculating VaR is similar in both Stockholm and London. This provides a sense of security since the method of calculating VaR is uniform across borders (Chua, 1996; Danielsson et al., 2004). A division manager said:

G. Wahlström / Management Accounting Research 20 (2009) 53–68 Table 2 The negative opinions of Basel II and distribution of responses. 1. A knowledge gap between banking staff members. (Present in three of four banks and expressed primarily by bank officers with operational functions.) 2. The lack of applicability of the models in practice. (Present in two banks and expressed primarily by bank officers with operational functions.) 3. Different interpretations of a vague regulation. (Present in all four banks and expressed primarily by bank officers with operational functions.) 4. Overly resource intensive. (Present in all banks and expressed primarily by bank officers with operational functions.) 5. Disadvantage for decentralised management structures. (Present in two banks and expressed primarily by bank staff with operational functions.) Bank staff at the decentralised banks and staff working primarily operations expressed these negative opinions most frequently.

the regulation leads to increased uniformity and control and, as a result, a further concentration of power at the banks’ headquarters. 6. The negative opinions of Basel II In spite of the strong support for Basel II, the interviewees also expressed important misgivings. From the interview data, the following negative opinions of Basel II were identified (Table 2). The negative opinions, most common amongst bank officers with primarily operational functions, were expressed in various degrees by the different interviewees. In some cases, these different opinions were articulated by staff from only two banks, whilst in others, they were expressed by staff at all four banks. As far as the negative opinions common to bank staff from all banks, there was no criticism or questioning of the fundamental approach adopted in Basel II. Instead, the criticisms concerned aspects of Basel II that were considered essentially uncontroversial, such as the fact that the implementation of Basel II was very resource intensive. Such criticism is natural when extensive sets of new regulations have to be implemented. Additionally, whilst the bank staff with the most decentralised management structure held all the negative opinions that are detailed below, staff at the more management centralised banks were less emphatic in their criticism of Basel II. As researchers have noted, change is generally more acceptable when the change conforms to pre-existing beliefs and is perceived as legitimate and worthwhile (Bateson, 1972; Hedberg and Jönsson, 1978; Jönsson and Lundin, 1977; Young, 2003). Thus, the finding that the most negative opinions were held by bank staff operating under the most decentralised management structure is by no means unexpected. These bank officers were more unreceptive to a regulation that could threaten their current frame of reference as far as management structure. In decentralised organisations, employees are typically trusted to make the “right” decisions, and thus centrally located control systems are less important. If regulations, such as Basel II, are imposed on staff, extensive new systems need to be implemented that require employees to report

61

data to headquarters. Thus administration, as well as control, will increase. Management policies such as centralised control and centralised administration contradict organisational traditions in decentralised management practice. For this reason, international regulations imposed on employees in decentralised organisations can easily be regarded as unnecessary and undesirable. Such an attitude may have exacerbated the various negative opinions of Basel II identified in this study, particularly amongst staff at the decentralised bank. 6.1. A knowledge gap between banking staff members Normally, senior level managers at bank headquarters have deep experience in different areas of operations gained from many years, for example, at the local branch offices. Recently, banks have hired much younger and less experienced specialists to develop risk models and to collect data to measure risk. These specialists in risk measurement may be twenty or even thirty years younger than senior managers, and their background and experience may be primarily theoretical rather than practical. Suspicion stemming from generational and competence differences, that is, different types of knowledge between the two groups, seems almost inevitable. As one head of internal control said: “For me, the new capital requirements reveal contrasts between our risk organisation, with its PhDs and statisticians on the one side, and the managers who run the bank on the other. Those in the risk organisation are, so to speak, convinced of the reliability and importance of risk measurement. They are the ones who are believed to know all about risk in the bank. Because they, and it is really only they, think that risk measurement can solve all kinds of problems. How can this be possible? If, at the heart of the bank, the senior bankers want to develop the bank’s business in a certain direction, then how can the people in the risk group think that they should develop operations in another? This symbolizes an extreme for me. They [the risk organisation people] are the ones who should be best at risk measurements. It is really important, but sometimes you can be so devoted to your discipline that you cannot become sufficiently critical of the numbers. You come to believe that you can solve all kinds of problems by measurements of risk.” (Head of internal control) Some interviewees thought Basel II was too complex, which suggests that people in different parts of an organisation may find it difficult to understand the highly technical and abstract language used by risk measurement specialists. It is therefore plausible that the degree of technical knowledge and understanding is also a problem in the implementation of Basel II. A retail division manager commented: “The weakness of Basel II is its complexity. It [risk measurement] forms its own tradition in the theoretical world and at the universities. And soon there will be just three people in each bank who really understand the rules and can explain them. This is a great problem in the running of the business. How can you explain, not just

62

G. Wahlström / Management Accounting Research 20 (2009) 53–68

to one person but also to thousands of people, what the rules actually mean? They will have to call us and ask. Only giving support to people in the business operations is really a weakness.” (Retail division manager) If people use numbers without knowing how they are computed, the numbers can become enslaving since information remains invisible to the users (ter Bogt, 2003; Chua, 1996; Young, 1996). 6.2. The unreality of the models in practice Bank officers did not question the models from a technical point of view, but rather in terms of whether the models corresponded with reality. A chief financial officer at one bank explained this criticism, as follows: “I am convinced that the mathematicians have made models that will work technically. The question is whether they meet reality? How will they be received? Will they be used? What effect is achieved when performance is evaluated? And of course the concept of Basel II is built on integration; if we are allowed to have our own models then the organisation must also use them. It is not sufficient to submit a report with a computed number of capital requirements and then not actually incorporate them in the organisation. It must be integrated.” (Chief financial officer) In the business world, a fundamental belief is that numbers are an important means of evaluating performance (Chua, 1996; Halkos and Salamouris, 2004; Hirst, 1981; Porter, 1995; Thompson, 1967). Many academics argue that the numerical calculation of risk is a valid and scientific exercise and therefore worth performing when evaluating creditworthiness (Ball and Forster, 1982; Chua, 1986, 1996; Verrecchia, 1982). Clearly, some practitioners agree since bankers are eager to develop models, collect data for measuring risks, and, indeed, implement these procedures in everyday decision making. According to the head of auditing at one bank: “There is a tendency that all banks follow in the same direction. If you listen to the major banks, they all say that they are working for greater quality in their creditors, and that the new rules will lead to lower capital requirements. We already have a balance sheet that meets high standards of quality and there has been a rush towards getting a balance sheet similar to ours. This leads to greater competition in areas that previously did not exist. Previously, much capital has drained into areas regardless of whether they are high or a low risk business. Now our competitors are trying to improve their balance sheets, whilst we have had a low risk profile all the time.” (Head of auditing) If practitioners are convinced of the value of a certain practice or regulation, it will come into effect. In the case of Basel II, the banking community has approved the regulation’s incentive for decision-makers to use statistically computed numbers. However, as banks develop these similar systems to value their businesses, there is a risk, some interviewees indicated, that the banks will act en masse. Some interviewees singled out mortgage loans in particu-

lar as a business area where all the banks wanted to increase their presence since credit losses in this area are extremely low. The market conditions for mortgages were described as tough, with margins constantly decreasing. The opposite, however, was said to be the case for extending credit to businesses in real estate where loans were no longer as attractive as they had once been. This situation had led to increases in interest rates charged on loans to real estate companies. Such a shift in the banks’ operations provides a hint of the extent to which Basel II has been embraced by the industry and has resulted in a similar strategic action. A staff member at one bank’s headquarters said: “My understanding is that Basel II will lead to a change in where the capital goes. Mainly, with a lower capital reserve requirement set for mortgage loans, this activity will increase. This shift is already taking place. A few years ago, one of our chief competitors got out of the mortgage area almost entirely, but now this bank is re-entering this segment at full speed, the same as everyone else in our business. We have all seen that because of Basel II it is possible to carry mortgage loans with very low or without any capital requirements. The area that I think will lose due to Basel II is the real estate industry. Interest rates will rise, but certainly Basel II has driven the marginal rate down for us on mortgages since everyone is running into this segment. I think it is a bit worrying that we have this herd behaviour in the banking industry due to Basel II since the models are built upon limited statistical data. For instance, we now have huge commitments in the Baltic, and for this region the economic development in recent years has been an enormous success story. Yet the problem is that our data for the Baltic area are from only those few successful years.” (Staff at headquarters) The negative effect when banks act en masse is believed to contribute to the creation of economic cycles. At the time of this study, each bank had submitted its minimum capital requirements model to the banking supervisory authority and all expected to receive approval. However, they accepted that their capital requirements were subject to revision from the supervisory authority, depending on future economic conditions. Generally, the interviewees took a rather pessimistic view of possible supervisory action. According to the interviewees, the supervisory authority would increase the level of capital requirements during times of financial distress, leading to a deeper downturn in the economy. Additionally, the interviewees believed the opposite action was also possible so that if good times resulted in lower capital requirements, a financial bubble would be created or sustained. It seems that Basel II created a different role for the supervisory authority other than just issuing regulations and supervising practice. The interviewees stated they believed Basel II caused supervisory authorities to take a much more active role in influencing money supply. Consequently, the banking supervisory authority may take on a function akin to the role of central banks. The interviewees were concerned about procedural and theoretical problems that might arise when a supervisory authority takes on central bank responsibilities.

G. Wahlström / Management Accounting Research 20 (2009) 53–68

6.3. Different interpretations of a vague regulation In Sweden, the national supervisory authority interpreted Basel II as creating a more uniform regulatory framework, although the regulation was still regarded as somewhat vague by many of the interviewees. A retail division head at one bank commented: “In Basel II there are very few explicit guidelines about what Basel II actually demands. There were unclear rules, but they have become clearer and clearer over time since the SFSA [Swedish Financial Supervisory Financial Authority] has issued its interpretation. We have had a dialogue with the SFSA and they have done a good job. But in the beginning, the rules were extremely unclear, so much so that they created a lot more work for us. OK. So we can, for instance, use internal rating models. That’s fine. But how? And to what extent? Is it up to us to decide? But to be frank with you, we want to know the type of framework we are operating in, and this framework has been unclear.” (Retail division head) One of the consequences of the vaguely stated regulation of Basel II is that the opportunities for national supervisory authorities to make their own interpretation of Basel II have increased. As the retail division head pointed out, their interpretation was reached jointly with representatives of the banks. In this instance, a dialogue between the supervisory authority and the banks about how the regulation should be interpreted increased the likelihood of acceptance by the users since their points of view were ventilated and could be incorporated into practice. However, a vague regulation can also create problems if the interpretation of Basel II differs too much amongst countries since eventual approval of the banks’ internal models are believed to affect conditions for competition. In particular, interviewed managers were worried that national supervisors might interpret Basel II differently. A chief financial officer said: “We are afraid that supervisory authorities in countries will interpret Basel II differently, in spite of Basel II being made into a directive. There are so many possibilities for the national supervisory authorities to choose from. We are afraid that the Swedish supervisory authority will interpret the regulation more strictly than the supervisory authorities in other countries. So we are afraid that there may be competitive disadvantages for Swedish banks with non-Swedish banks, including those with branches here in Sweden. Such banks may have an advantage since their branches are under supervision from their banks’ home countries.” (Chief financial officer) With the possibility of so many interpretations of vague rules by the supervisory authorities and bank representatives, it is easy to imagine a problematic situation for all parties involved. As will be analysed in more depth below, the interviewees emphasised that they wished to gain approval from the supervisory authority for their own models and data in measuring risks, but they also indicated that they did not want to be exposed as unable to meet the requirements. To this end, especially if there is significant

63

room for manoeuvring on the interpretation of the regulation, there was the likelihood of doubt and delay. Many managers seemed to ask: “Are we doing the right thing?” As a consequence, this uncertainty contributed to making the implementation of Basel II overly resource intensive, a criticism discussed further in the next section. 6.4. Overly resource intensive Interviewees expressed negative opinions about Basel II because of the costs associated with its implementation. A division head said: “I think many of us can agree on the need for new rules. Now, when it finally becomes a bit clearer what Basel II is actually about, it is much more extensive than any of us could have imagined. It is a complex and complicated regulation that is very expensive to implement as well. In the beginning, when we talked with the Banking Association, the banks, and the supervisory authority, there were many people who were well aware of the tradition of measuring risk and they advocated the need for rules based on risk measurements. But I do not believe anyone really understood how expensive it was going to be. For instance, IT solutions needed to be developed, and so forth.” (Division head) There was a commonly held opinion amongst the banks that the implementation of Basel II created such large costs that smaller banks, for example, niche banks, would not have sufficient resources to develop models and collect data. A staff member from one bank headquarters said: “Developing the models and collecting the data demand significant resources and smaller banks have no possibilities to do this, so they will hang on to conventional standards. But then they will not enjoy the benefit of the new system. They will be unable to operate with a smaller amount of capital and competitively will be unable to really judge the risk in a similar way. Thus small banks will accept creditors at a certain price. And maybe it will turn out that these banks will accept large numbers of these creditors and then their portfolios will suddenly consist of creditors that no one else wants.” (Headquarters staff member) There seemed to be a widely held belief, at least amongst the centralised banks, that the ability to develop models, collect data and gain approval by the SFSA was indicative of the competency of the banks’ human resources departments and an indicator of the overall quality of the organizations. The risk measurement information appeared to give these interviewees a sense of safety because they believed it reduced the risk of being stuck with “bad” customers. On the other hand, interviewees in the more decentralised bank emphasised the importance of establishing personal relationships with customers over long periods of time and of creating a spirit of mutual trust between business parties. Bank officers from the decentralised bank considered it a significant advantage to know their customers personally. In such cases, it was emphasised that the customer and the bank could both, over time, fulfil their obligations and duties, leading to a situation

64

G. Wahlström / Management Accounting Research 20 (2009) 53–68

where the customers’ reliability was established. For these bankers, the important element in creating a creditworthy customer base was in the direct, personal contact with customers, and less in the indirect, impersonal quantification of risk models. 6.5. Disadvantage for decentralised management structures The bank officers at the decentralised bank thought that Basel II favoured centralisation as an approach to managing operations. Thus, in many ways, the new regulatory regime put banks with a decentralised management structure at a disadvantage. The task of developing models and collecting data was, amongst the banks in the study, concentrated at the headquarters level, as it has been on previous occasions when new measures for evaluation had been introduced (Chandler, 1977, 1980; Porter, 1995). As such, Basel II was said to contribute to increasing centralisation at the banks. A division head said: “I think that knowledge about Basel II varies a lot within the bank. Although we are one of the four largest banks in Sweden, it is true even for us. As usual, regulations for capital requirements are something that originate at headquarters and then are spread out in the organisation. The organisation is starting now to be aware of the new rules. And we have talked about capital allocation and new capital requirement regulations for several years now, but the competence is here at the headquarter.” (Division head) In summary, as the interviewees stressed, relatively few people understand Basel II. Those few people who do, work at the banks’ headquarters. Overall, whilst Basel II was generally perceived as important and beneficial, for the reasons outlined in the previous section of this paper, there was a concern that implementation of Basel II would further increase the power of banks’ headquarters, thus increasing centralisation of operations. In Swedish banks, where many decisions are made at the branch level, decentralised management structure is traditional (Jönsson, 1995; Wallander, 1999), especially compared to other countries’ management models used, for instance, in American companies (Chandler, 1977). International regulations based on centralisation are a challenge for Swedish banks, particularly so for the most decentralised banks that lack the culture of centralised management (Jönsson, 1995; Wallander, 1999). If Basel II is to be fully implemented and its benefits realised, Swedish banks will have to make adjustments to their overall management structure. In particular, the Basel II requirement that the information on risk measurement must be used in day-to-day activities (Basel Committee, 2004a) is likely to be troublesome for banks accustomed to operating under decentralised management. 7. Discussion and suggestions for future research In this study, the chief purpose was to investigate the perceptions of Basel II amongst Swedish bankers through the use of semi-structured interviews conducted at Sweden’s four largest banks as they worked to implement the

new capital reserve requirements specified by the new banking regulation. Basel II provides powerful incentives to promote the successful measurement of risk by banks and actively encourages them to develop and implement their own analytical models based on their own data. However, the critical literature has seriously questioned this numerically based approach to the measurement of risk. The question that thus demands our attention is: How do perceptions define contextual outcomes within a Basel II implementation setting? It is argued here that this question can most usefully be addressed by investigating and analysing opinions of bank officers whose day-to-day work is affected by the Basel II regulations. The study revealed strong, favourable support for the new regulations but also sharp, unfavourable criticism. The interview results, categorised as positive and negative opinions, are illustrated through representative interviewees’ statements accompanied by analysis and discussion. A suggestion for a way to resolve these conflicting opinions is proposed, followed by suggestions for future research, presented in the last section. 7.1. Summary of results and explanation The positive opinions of Basel II revealed in these interviews were in no sense unexpected since Basel II is well established and supported in practice. Bank interviewees in this study recognized the benefits associated with using their own risk measurement models that Basel II encourages and permits. For example, the interviewees appreciated Basel II’s more sensitive and context applicable regulation for minimum capital reserves and its encouragement of the development of internal systems that are more effective and efficient in risk control. These positive opinions were articulated by banking staff working closely with risk measurement, such as risk managers and project leaders for Basel II. However, it was clear that such opinions were much more strongly expressed by staff at the three banks in the study with centralised management structures and much less so by staff at the fourth bank that uses a more decentralised branch structure. In the study, the negative opinions expressed by the interviewees primarily questioned the functionality of Basel II. Except for rather “minor” objections – Basel II is too resource intensive and too vaguely written, for instance – the interviewees raised some fundamental criticisms, ranging from specific concerns questioning whether the risk measurement models would actually work in practice to broader concerns questioning the ability of Basel II to protect the banking sector from financial distress. These criticisms, particularly the doubts about the usefulness of measurement risk models, hinted at a problem of knowledge clash between different areas of operational practice within the banks, specifically between operating managers and risk managers. Additionally, the interviewees claimed that Basel II had the potential to create increased centralisation of management decision making at the expense of bank systems that tend to delegate more authority to the branches. Unsurprisingly, staff in operations at the decentralised bank in this study voiced this negative opinion. Of most interest for this study are the negative opinions that the interviews brought forth. These findings

G. Wahlström / Management Accounting Research 20 (2009) 53–68

suggest there are serious, potentially negative outcomes in the implementation of Basel II in the banking industry. For instance, if, as a result of Basel II, the concept of centralised management diminishes, even replaces, more decentralised management structures, there is a risk of disillusionment and demotivation, leading to productivity loss amongst employees working in banks where the branches have enjoyed considerable discretion in decision making. Equally possible and equally discouraging is the potential outcome that banking staff at decentralised banks will tend to ignore Basel II as mere window dressing for the industry. It is useful to ask why there is such a diversity of opinion on Basel II. One answer may be related to the various banking responsibilities of the interviewees. Based on their responses in the study, it appears that the positive opinions were more strongly emphasised by Basel II project leaders, risk managers and headquarters personnel directly involved with the implementation of the new regulation. In contrast, the negative opinions came from banking staff primarily involved with operational assignments, particularly those at the bank with the most decentralised management structure. Having identified the bank employees who were most negative toward Basel II, we may ask why the nature of their work distinguishes them from the employees generally positive toward Basel II. The answer appears to be a matter of employees’ pre-existing frames of reference. As noted, those individuals identified in the study who most strongly supported Basel II were the risk managers, project leaders and other staff directly involved with the measurement of risk. The commonality of this group is that all individuals were quite comfortable in the tradition of risk measurement. In working within a framework such as Basel II, they were positively disposed to the concept of and benefits associated with risk measurement. Thus, it was natural for this category of staff to appreciate the regulation since it accords with their pre-existing frames of reference. From their perspective, if Basel II improved internal systems and facilitated risk management, the good practice of centralised banks would be advanced. Additionally, the universal support for Basel II by banking authorities such as central banks and prominent public figures in the industry confirmed the positive opinions held by these bank officers. On the other hand, the study’s interviewees who expressed negative opinions of Basel II held positions such as head of auditing, head of retail division and credit officer. In their work, this group was involved in the day-to-day, operational tasks of foreseeing and avoiding problems. They were most interested in practical, problem-solving information and in the various, complex social relationships their work required. For them, their previous and personal experience was considered vital to the success of their work and, as a result, they had little interest in the numerical quantification of a specified risk and little confidence in the ability of numbers to accurately represent complex and dynamic realities. 7.2. The contribution to the literature The findings of this study contribute to the literature on several dimensions. Whilst earlier critical literature of

65

risk measurement by Chua (1996), McGoun (1992, 1995), and Young (2001), takes a logical reasoning perspective, this study moves the analysis one step forward by providing evidence of users’ perceptions from everyday practice. Moreover, this study contributes to the literature of management research and accounting regulation in practice as called for by Bromwich and Hong (2000) in their detailed study of the British Telecoms accounting system where they encouraged more studies of application from highly regulated industries. This study expands on the work of Bromwich and Hong (2000) in at least two aspects. In contrast to their investigation of the accounting system in the telecom industry, evidence is presented here on how interviewed staff in a different regulated industry actually perceived regulation promulgated by an international organisation. The question that then arises is how the potential problems arising from the negative opinions of Basel II can be addressed. One suggestion is to engage representatives from the academic community, as well as banking staff, in an industry-wide debate. In open forums, through public debate, the objections to Basel II may be examined, discussed and resolved, even avoided. This suggestion fits well in the literature that is concerned with the role of researchers in society (Briloff, 1993; Lee, 2006; Sikka et al., 1995; Tinker, 2002; Tinker and Carter, 2002; West, 2003; Willmott et al., 1993). In particular, Briloff (1993), Chua (1996) and Lee (2006) have all argued that university education has little relevance for everyday practice. If researchers in accounting, finance and economics engage more with the topic of bank regulations in a discussion of the possible consequences of Basel II, the issue would be opened up to public debate with the participation of practitioners and academics in several areas. Such a discussion from many angles could lead to an understanding of regulation that is closer to everyday practice. In addition, university education might begin to address more directly the questions and problems typical of the so-called real world of work. In return, in an open exchange of ideas with the world of academics, bank managers might also benefit from approaching their work from a more theoretical angle. 7.3. Suggestions for future research In conclusion, it may be said that this study raises a number of issues about the perceptions of Basel II implementation. Although the focus of the study is a Swedish setting, these issues may be relevant to bankers in other settings that fall under the Basel II dominion, as well as to other regulated industries affected by international regulation. Possibly the most important of these issues concerns how an organisation’s management structure affects the implementation of an international regulation imposed upon it. The study reveals that implementation of outside regulation is best supported by a management structure built on centralisation. The concern is that resistance to a new regulation, carried to an extreme, has the potential to harm the organisation if it regards the regulation as mere window dressing and fails to take advantage of the benefits implicit in implementing it. Yet some form of a centralised management structure seems essential for

66

G. Wahlström / Management Accounting Research 20 (2009) 53–68

uniform implementation of a new company-wide regulation. Further research into the issue of the most effective management structure for the adoption of imposed regulation is of interest, whether in the banking or some other regulated industry. Paradoxically, the most decentralised bank of the four banks investigated in this study was the only bank that did not have to ask the Swedish government for funding at the time of the national financial crisis in the beginning of the 1990s. Furthermore, the profitability of this bank during the 1990s was much better than that of the three centralised banks in the study (Wallander, 1999). Given this record of stability and profitability, a question arises: How would such successful, decentralised banks react to a regulation imposed upon them that has more obvious benefits and advantages for banks with centralised structures? Such a question leads to further questions: How would existing management styles be challenged? How would working conditions be influenced? How would employee motivation and morale be affected? Such questions are important since the advantages associated with decentralised organisations are, primarily, related to employee motivation and engagement, which, as has been shown, lead to greater productivity than in centralised organisations (Jönsson, 1995; Wallander, 1999). Especially in the banking industry, a marginal business, such questions require further investigation. Acknowledgements The author would like to thank the anonymous reviewers of this journal and Al Bhimani for their comments. I am grateful for research support from Jan Wallanders Stiftelse, Tom Hedelius Stiftelse and Tore Browaldhs Stiftelse. Appendix A. Description of Basel II Basel II consists of three interrelated pillar: minimum capital requirements, a supervisory review process and market discipline (Basel Committee, 2004a). The first of these pillars provides for the calculation of the total minimum capital requirements for credit, market and operational risk. There are different methodologies to compute the minimum capital requirements, and the banks are encouraged to use one of the more advanced approaches. In Basel II, measurement of risk is equated with sophistication. Instead of “only” assessing risks, Basel II clearly states that banks should move towards the measurement of risk and should use the information for internal decision making. Interestingly, a bank cannot revert to a “simpler” approach once it has been authorized to use a more advanced one except under conditions where the bank no longer meets the qualifying criteria. These are criteria that also involve the generation of qualitative information. It is up to the supervisory authority to approve each bank’s approach/methodology, and thus Basel II is interwoven with other areas in accounting, such as the supervisory review processes (the second pillar) and information to the market, the so-called market discipline (the third pillar). To conclude, it is perhaps easy to think that Basel II is a essentially a conglomerate of statistical demands, but this

impression would be misleading since all three pillars have qualitative, as well as quantitative, characteristics. With regard to the first pillar, when calculating their capital requirements for credit risk, banks have to choose between two broad methodologies. They can either measure credit risk in a standardised manner, supported by an external credit assessments, such as, for instance, Standard and Poor’s rating system, or they can use an internal rating system for measuring credit risk. For operational risk, the framework outlines three methods of calculation: the basic indicator approach, the standardised approach and, finally, the advanced measurement approach. Market risk, can be measured by one of two broad methodologies. One alternative is to measure market risk in a standardised way, and here the framework identifies the following risks: interest rate, equity position, foreign exchange and commodities risk. A second method allows banks to use risk measures derived from their own internal risk management models. It is the latter approach that is emphasised and regarded as the most desirable way of measuring risk. The second pillar has its focus on supervisory review. Here, Basel II begins by outlining the importance of supervisory review, followed by key principles for the supervisory reviewers. As an example of one of its principles, the fourth principle states that “supervisors should seek to intervene at an early stage to prevent capital from falling below the minimum levels required to support the risk characteristics of a particular bank and should require rapid remedial action if capital is not maintained or restored” (Basel Committee, 2004a, p. 165). In the second pillar, the specific issues for supervisors are explicitly stipulated in the form of a framework for the supervisory process. Although numbers and measurement of risk are central in this second pillar, it contains a qualitative element as well. The third pillar, market discipline, contains information about disclosure requirements for banks. These requirements are intended to complement the other two pillars. It must be stressed, however, that disclosure to the market has an extremely important function since the reliance on internal methodologies gives banks much more discretion in assessing capital requirements. Therefore, it is important that a bank’s disclosure should be consistent with ways in which senior management and the board of directors assess the various levels of risk in the bank. Furthermore, good disclosures are an effective way of informing the market about a bank’s exposure to the risks regulated by the requirements of the first pillar, thus providing a consistent and systematic framework that enhances comparability. In summary, Basel II describes an extensive framework for regulatory and operational practice. For instance, Basel II covers and encourages the measurement of risk for individual banks, emphasising, and indeed rewarding, internal methodologies for decision making in the banks. Basel II also sets out guiding principles for the supervisory review and regulatory functions, as well as setting out requirements concerning disclosures to the market. Thus Basel II is both multifaceted and complex, combining quantitative rigour with qualitative insights. It encourages banks to develop their own tailored methods of actually measuring risk, rather than simply relying on broader external assessments.

G. Wahlström / Management Accounting Research 20 (2009) 53–68

References The American Accounting Association (AAA), 1977. Statement on Accounting Theory and Theory Acceptance. AAA, Sarasota, FL. Ball, R., Forster, G., 1982. Corporate financial reporting: a methodological review of empirical research. J. Acc. Res. 20, 161–234. Basak, S., Shapiro, A., 2001. Value-at-risk based risk management: optimal policies and asset prices. Rev. Finan. Stud. 14, 371–405. Basel Committee on Banking Supervision, 1996. Overview of the Amendment to the Capital Accord to Incorporate Market Risk. Bank for International Settlements, Basel. Basel Committee on Banking Supervision, 1999. A New Capital Adequacy Framework. Bank for International Settlements, Basel. Basel Committee on Banking Supervision, 2001a. Consultative Document, Pillar 3 (Market Discipline), Supporting Document to the New Basel Capital Accord, Issued for Comment by 31 May 2001. Bank for International Settlements, Basel. Basel Committee on Banking Supervision, 2001b. The New Basel Capital Accord: an explanatory note. Bank for International Settlements, Basel. Basel Committee on Banking Supervision, 2004a. International Convergence of Capital Measurement and Capital Standards. Bank for International Settlements, Basel. Basel Committee on Banking Supervision, 2004b. History of the Basel Committee and its membership. Bank for International Settlements, Basel. Basel Committee on Banking Supervision, 2006a. Home-Host Information Sharing for Effective Basel II Implementation. Bank for International Settlements, Basel. Basel Committee on Banking Supervision, 2006b. Report in International Developments in Banking Supervision. Bank for International Settlements, Basel. Bateson, G., 1972. A theory of play and fantasy. In: Bateson, G. (Ed.), Steps to an Ecology of Mind. The University of Chicago Press, Chicago, pp. 177–193. Berger, P., Luckmann, T., 1966. The Social Construction of Reality. Doubleday, New York. Bhimani, A., 1993. Indeterminacy and the specificity of accounting change: Renault 1898–1938. Acc. Organ. Soc. 18, 1–39. Blejer, M.I., 2006. Economic growth and the stability and efficiency of the financial sector. J. Bank. Finan. 30, 3429–3432. Blum, J.M., 1999. Do capital adequacy requirement reduce risks in banking. J. Bank. Finan. 23, 755–771. Boyle, D., 2001. The Tyranny of Numbers: Why Counting Can’t Make Us Happy. Haper-Collins, London. Bricker, R., Borokhovich, K., Simkins, B., 2003. The impact of accounting research on finance. Crit. Perspect. Acc. 14, 417–438. Briloff, A., 1993. Unaccountable accounting revisited. Crit. Perspect. Acc. 4, 301–335. Broadbent, J., Laughlin, R., 1994. Moving towards an accounting that will be enabling: accounting, Habermas and issues of gender. Seminar paper. Department of Accounting and Finance, University of Wollongong. Bromwich, M., Hong, C., 2000. Costs and regulation in the U.K. telecommunications industry. Manage. Acc. Res. 11, 137–165. Bryman, A., Bell, E., 2003. Business Research Methods. Oxford University Press, Oxford. Calem, P.S., LaCour-Little, M., 2004. Risk-based capital requirements for mortgage loans. J. Bank. Finan. 28, 647–672. Canning, J., 1929. The Economics of Accountancy: A Critical Analysis of Accounting Theory. Ronald Press Company, New York. Carnegie, G., Napier, C., 1996. Critical and interpretative histories: insights into accounting’s present and future through its past. J. Acc. Aud. Account. 9, 7–39. Cerny, P.G., 1994. The infrastructure of the infrastructure? Toward embedded financial orthodoxy in the international political economy. In: Palan, R.P., Gilles, V. (Eds.), Transcending the State—Global Divide. Lynne Rienner, Boulder, CO, pp. 208–211. Chandler, A.D., 1977. The Visible Hand: The Managerial Revolution in American Business. Harvard University Press, Cambridge. Chandler, A.D., 1980. The United States. In: Chandler, A.D., Daems, H. (Eds.), Managerial Hierarchies: Comparative Perspectives on the Rise of the Modern Industrial Enterprise. Harvard University Press, Cambridge, pp. 9–40. Chua, W.F., 1986. Radical developments in accounting thought. Acc. Rev. 61, 601–633. Chua, W.F., 1996. Teaching and learning only the language of the numbers—monolingualism in a multilingual world. Crit. Perspect. Acc. 7, 129–156. Collins, H., 1985. Changing Order. Sage, London.

67

Cooper, R., 1990. Explicating the logic of ABC. Manage. Acc. (November), 58–60. Crockett, A., 2000. Marrying the micro- and macro-prudential dimensions of financial stability. The General Manager of the Bank of International Settlements. http://www.bis.org/review/rr000921b.pdf. Danielsson, J., 2002. The emperor has no clothes: limits to risk modelling. J. Bank. Finan. 26, 1273–1296. Danielsson, J., Jorgensen, B.N., De Vries, C.G., 2002. Incentives for effective risk management. J. Bank. Finan. 26, 1407–1425. Danielsson, J., Shin, H.S., Zigrand, J.-P., 2004. The impact of risk regulation on price dynamics. J. Bank. Finan. 28, 1069–1087. Danielsson, J., Taylor, A., Zigrand, J.-P., 2005. Highwaymen or heroes: should hedge funds be regulated? A survey. J. Finan. Stab. 1, 522–543. Danielsson, J., Zigrand, J.-P., 2008. Equilibrium asset pricing with systematic risk. Econ. Theory 35, 293–319. Dent, J.F., 1991. Accounting and organizational cultures: a field study of the emergence of a new organizational reality. Acc. Organ. Soc. 16, 705–732. Dunbar, N., 1999. Inventing Money: The Story of Long-Term Capital Management and the Legends Behind it. John Wiley, New York. Fleming, R.J., Graci, S.P., Thompson, J.E., 1990. Characteristics of the work of leading authors in The Accounting Review, 1926–1945. J. Acc. Hist. 17, 5–25. Fleming, R.J., Graci, S.P., Thompson, J.E., 1991. Tracing the evolution of research in The Accounting Review through its leading authors: the 1946–1965 period. J. Acc. Hist. 18, 26–53. Fleming, R.J., Graci, S.P., Thompson, J.E., 2000. The dawning of the age of quantitative/empirical methods in accounting research: evidence from the leading authors of The Accounting Review. J. Acc. Hist. 27, 43–72. Flesher, D., Flesher, T., 1986. Ivar Kreuger’s contribution to U.S. financial reporting. Acc. Rev. 61, 421–434. Funnell, W., 1998. Accounting in the service of the holocaust. Crit. Perspect. Acc. 8, 435–464. Furfine, C., 2006. The costs and benefits of moral suasion: evidence from the rescue of Long-Term Capital Management. J. Business 79, 593–622. Glaser, B.G., Strauss, A.L., 1967. The Discovery of Grounded Theory. Aldine de Gruyter, New York. Greenspan, A., 1998. Statement before the Committee on Banking and Financial Services. Fed. Res. Bull. (December), 1046–1050. Halkos, G.E., Salamouris, D.S., 2004. Efficiency measurement of Greek commercial banks with the use of financial ratios: a data envelopment analysis approach. Manage. Acc. Res. 15, 311–332. Hartley, J., 2004. Case study research. In: Cassell, C., Symon, G. (Eds.), Essential Guide to Qualitative Methods in Organizational Research. Sage, London, pp. 323–333. Hedberg, B., Jönsson, S., 1978. Designing semi-confusing information systems for organizations in changing environments. Acc. Organ. Soc. 3, 47–64. Heid, F., 2007. The cyclical effects of the Basel II capital requirements. J. Bank. Finan. 31, 3885–3900. Hirst, M., 1981. Accounting information and the evaluation of subordinate performance. An institutional approach. Acc. Rev. 56, 771–784. Hopwood, A.G., 1983. On trying to study accounting in the contexts in which it operates. Acc. Organ. Soc. 8, 287–305. Jacobs, B., 1999. Capital Ideas and Market Realities. Blackwell, Oxford. Jarrow, R.A., 2007. A critique of revised Basel II. J. Finan. Serv. Res. 32, 1–16. Johnson, H., Kaplan, R.S., 1987. Relevance Lost; the Rise and Fall of Management Accounting. Harvard Business School Press, Boston. Jorion, P., 2000. Risk management lessons from Long-Term Capital Management. Eur. Finan. Manage. 12, 6–12. Jorion, P., 2002. How informative are value-at-risk disclosures. Acc. Rev. 77, 911–932. Jönsson, S., 1991. Role making for accounting while the state is watching. Acc. Organ. Soc. 16, 521–546. Jönsson, S., 1995. Goda Utsikter (Good Prospects). Nerenius & Santerus Förlag, Stockholm. Jönsson, S., 1996. Accounting for Improvement. Pergamon, Oxford. Jönsson, S., 1998. Relate management accounting research to managerial work. Acc. Organ. Soc. 23, 411–434. Jönsson, S., Lundin, R., 1977. Myths and wishful thinking as management tools. Stud. Manage. Sci. 5, 157–170. King, N., 2004. Using interviews in qualitative research. In: Cassell, C., Symon, G. (Eds.), Essential Guide to Qualitative Methods in Organizational Research. Sage, London, pp. 11–23. Knights, D., Vurdubakis, T., 1993. Calculations of risk: towards an understanding of insurance as a moral and political technology. Acc. Organ. Soc. 7-8, 729–764.

68

G. Wahlström / Management Accounting Research 20 (2009) 53–68

Kvale, S., 1983. The qualitative research interview: a phenomenological and a hermeneutical mode of understanding. J. Phenomenol. Psychol. 14, 171–196. Lee, T.A., 1995. Shaping the US academic accounting research profession: the American Accounting Association and social construction. Crit. Perspect. Acc. 6, 241–261. Lee, T.A., 2006. The war of the sidewardly mobile corporate financial report. Crit. Perspect. Acc. 17, 419–455. McGoun, E.G., 1992. The CAPM: a Nobel failure. Crit. Perspect. Acc. 4, 155–177. McGoun, E.G., 1995. The history of risk measurement. Crit. Perspect. Acc. 6, 511–532. Merino, B.D., Neimark, M.D., 1982. Disclosure regulation and public policy—a reappraisal. J. Acc. Pub. Pol. 1, 33–57. Morris, S., Shin, H.S., 1999. Risk management with interdependent choice. Oxf. Rev. Econ. Pol. 15, 52–62. Munro, R., 1995. Managing by ambiguity: an archaeology of the social in the absence of management accounting. Crit. Perspect. Acc. 6, 433–482. Pederzoli, C., Torricelli, C., 2005. Capital requirements and business cycle regimes: forward-looking modelling of default probabilities. J. Bank. Finan. 29, 3121–3140. Persuad, A., 2000. Sending the Herd of the Cliff Edge: The Disturbing Interaction Between Herding and Market-sensitive Risk Management Practices. Larosiére Prize Winning Essay, Institute for International Finance. http://www.erisk.com/news/features/persuad.pdf. Peuro, S., Jokivuollo, S., 2004. Simulation based stress tests of banks’ regulatory capital adequacy. J. Bank. Finan. 28, 1801–1824. Porter, T.M., 1995. Trust in Numbers. Princeton University Press, Princeton. Power, M., 2004. Counting, control and calculation: reflections on measuring and management. Hum. Rel. 57, 765–783. Puxty, A.G., Willmott, H., Cooper, D.J., Lowe, T., 1987. Modes of regulation in advanced capitalism: locating accountancy in four countries. Acc. Organ. Soc. 12, 273–291. Scapens, R., 1990. Researching management accounting practice: the role of case study methods. Br. Acc. Rev. 22, 259–281. Shapin, S., 1984. Pomp and circumstance: Robert Boyle’s literary technology. Soc. Stud. Sci. 14, 481–520. Sikka, P., Willmott, H., Puxty, T., 1995. The mountains are there: accounting academics and bearings of intellectuals. J. Acc. Audit. Account. 8, 113–140.

Solomons, D., 1983. The political implications of accounting and accounting standard setting. Acc. Bus. Res. 13, 107–118. Sterling, R.R., 1990. Positive accounting: an assessment. Abacus 26, 97–135. Sterling, R.R., 1993. The subject matters of accounting. In: Mumford, M.J., Peasnell, K.V. (Eds.), Philosophical Perspectives on Accounting: Essays in Honour of Edward Stamp. Routledge, London, pp. 123–155. Szegö, G., 2002. Measures of risk. J. Bank. Finan. 26, 1253–1272. ter Bogt, H.J., 2003. Performance evaluation styles in governmental organizations: how do professional managers facilitate politicians’ work. Manage. Acc. Res. 14, 311–332. Thompson, J.D., 1967. Organizations in Action. McGraw Hill, Montreal. Tinker, T., 1997. Dialectic of the value form: the social evolution of capital markets in the US. In: Knights, D., Tinker, T. (Eds.), Financial Institutions and Social Transformation. St. Martin’s Press, New York, pp. 28–46. Tinker, T., 2002. Beyond the Brilovian critique: traditional vs. organic intellectuals in critical accounting research. Acc. Publ. Int. 2, 68–87. Tinker, T., Carter, C., 2002. Beyond the Brilovian critique: Abraham Briloff’s lonely hearts club tribute band. Acc. Publ. Int. 2, 97–117. Tomkins, C., Groves, R., 1983. The everyday accountant and researching his reality. Acc. Organ. Soc. 8, 361–374. Townley, B., Cooper, D.J., Oakes, L., 2003. Performance measures and the rationalization of organizations. Organ. Stud. 25, 1045–1071. VanHoose, D., 2007. Theories of bank behaviour under capital regulation. J. Bank. Finan. 31, 3680–3697. Verrecchia, R.E., 1982. The use of mathematical models in financial accounting. J. Acc. Res. 20, 1–42. Wallander, J., 1999. Budgeting an unnecessary evil. Scand. J. Manage. 15, 405–421. Watts, R., Zimmerman, J., 1986. Positive Accounting Theory. Prentice Hall, Englewood Cliffs, NJ. West, B.P., 2003. Professionalism and Accounting Rules. Routledge, London. Willmott, H., Puxty, T., Sikka, P., 1993. Losing one’s reason: on the integrity of accounting academics. J. Acc. Audit. Account. 6, 98–110. Young, J., 1996. Institutional thinking: the case of financial instruments. Acc. Organ. Soc. 21, 579–600. Young, J., 2001. Risk(ing) metaphors. Crit. Perspect. Acc. 12, 607–625. Young, J., 2003. Constructing, persuading and silencing: the rhetoric of accounting standards. Acc. Organ. Soc. 28, 621–638. Young, J., 2006. Making up users. Acc. Organ. Soc. 31, 579–600.