The Big 4 in Bangladesh: caught between the global and the local

Purpose - The purpose of this paper is to explore the work practices of Big 4 firms in Bangladesh with the aim of exploring the extent to which global professional service firms (GPSFs) can be thought of as being genuinely “global”. Design/methodology/approach - Interviews were undertaken with the vast majority of Big 4 partners in Bangladesh. These interviews explored a number of themes related to the professional service work context in Bangladesh and the relationship between local and global firms. Findings - The central finding of this paper is that although the Big 4 have a long-established presence in Bangladesh, local societal factors heavily influence the realities of work for accountants there. In most cases the Big 4 firms establish correspondent firms (instead of full member firms) in Bangladesh and tend to offer restricted service lines. Additionally, the paper identifies professional, commercial and cultural barriers to greater Big 4 involvement in the local market. Conceptually, the chief contribution of this paper is to explore how the effects of globalizing capitalism and standardised “best practices” in global professional service work are mediated through the societal effects of Bangladeshi society, resulting in the Big 4 having only a tentative presence in the Bangladeshi market. Research limitations/implications - The findings cast doubt on the extent to which self-styled GPSFs are truly “global” in nature. Future work examining the Big 4, or accounting more generally, in the context of globalization, would do well to pay greater attention to the experience of professionals in emerging markets. Originality/value - Whilst there has been much work looking at accounting and accountants in the context of globalization, this work has tended to privilege “core” western empirical settings. Very little is known about professional service firms in “peripheral” emerging markets. Furthermore, this study extends the application of the system, society and dominance framework by mapping the interactions and dynamics of these three sources of influence in the setting of PSFs.

The Big 4 in Bangladesh: caught between the global and the local Introduction Globalization remains one of the most contested concepts in the social sciences, with disagreement over, inter alia, when it started, who benefits/loses from it and what consequences it engenders (Guillén, 2001). There is, however, general agreement that globalization does exist, as evidenced by the influence of institutions and processes that operate on a global scale, such as financial markets, the World Trade Organization, War Crimes Tribunals, etc. (Giddens, 2002;Sassen, 2003). Arguments vis-à-vis the desirability of such institutions and processes can often be placed into one of two, polarized streams of thinking: the globophiles and the globophobes (Held and McGrew, 2007;Kellner, 2009). The former uncritically accept the spread of global institutions and processes as leading to greater democracy, freedom, happiness and wealth (Friedman, 1999(Friedman, , 2005. The latter see these processes as the imposition of western capitalist hegemony, further exacerbating wealth disparities, degrading the biosphere and undermining national sovereignty (Boyer and Drache, 1996). The position adopted in this paper is neither globophilic nor globophobic. Rather, we see globalization, following Kellner (2009), as "a highly complex, contradictory, and thus ambiguous set of institutions and social relations that takes economic, political, social, and cultural forms" (p. 3). Indeed, it is this complex and contradictory nature that renders globalization such an interesting object of analysis.
The globalization of the world economy has been accompanied by the globalization of professional services. According to Morgan (2009), "one of the most important trends of the last decade has been the way in which financial and professional service firms have internationalized to seek out new markets" (p. 588). Similarly, Faulconbridge and Muzio (2012) argue that our understanding of professional work needs to be recontextualized given the now well-established phenomenon of the global professional service firm (GPSF). Faulconbridge and Muzio (2012) lament that research into professional work is still overwhelmingly undertaken within the context of specific nation-states, even though GPSFs are decreasingly constrained by the fetters of national-professional jurisdictions (Suddaby et al., 2007). Local studies therefore need to be understood in a global context (Sikka, 2008).
The accounting literature has seen some important globalization-sensitive work published on the Big 4, although this has tended to privilege either western empirical settings (see e.g. Barrett et al., 2005) or focus on international governance arrangements that are heavily influenced by Big 4 firms based in Western Europe or North America (Suddaby et al., 2007). Very little contemporary work has been undertaken in more "peripheral" or "marginal" (Sassen, 2003) contexts. Although the analysis of the accounting professional in post-colonial contexts is well established, such work tends to be historical in nature (see e.g. Annisette, 2000Annisette, , 2003Dyball et al., 2007). There is still very little known about the way in which the present-day pressures of globalization manifest themselves in emerging markets for professional services. Where such work has been undertaken, more attention has either been paid empirically to the views of "core" participants (see e.g. Cooper et al., 1998) or to what is generalizable globally rather than to what is idiosyncratic locally (see e.g. Ruiz Castro, 2012). This represents a lacuna in our understanding of the GPSF as comparative work has shown that the Big 4 have largely homogenous work processes, career paths and bureaucratic structures across different western contexts (see e.g. Carter and Spence, 2014).
From a globalization perspective, however, one would expect more differences to emerge when comparing global power "cores" with national "peripheries" (Spence et al., 2016). In other words, one would expect analysis of locales where the globalization of financial markets is less embedded to reveal a different set of professional circumstances (Morgan, 2009). As Sassen (2003) astutely observes, globalization is an incomplete project. Nationstates are not necessarily completely undermined by global processes. Rather, what can be seen today is the "partial embeddedness of the global in the national" (Sassen, 2003, p. 14). As such, a more "peripheral" context such as Bangladesh offers a ready test case to assess the extent to which global PSFs are "truly global" (Boussebaa et al., 2012). In this paper we explore the work practices of Big 4 firms in Bangladesh with the aim of assessing the extent to which GPSFs can be thought of as genuinely "global".
The paper is structured as follows. The following section outlines some key arguments from the globalization literature and identifies relevant accounting studies that have sought to situate accountancy practice within the wider context of globalization. The third section 146 AAAJ 30,1 introduces the theoretical framework system, society and dominance (SSD) used in this study. The fourth section provides the socio-economic context of Bangladesh. A subsequent section describes the research method employed to achieve these. The findings of the empirical study are then described with reference to several sub-sections. The key findings and insights offered by the study are then discussed in a separate section before offering some conclusions.

Global Professional Service Firms
There is a growing body of literature that studies accounting's relationship to globalization (Barrett et al., 2005;Cooper et al., 1998;Gallhofer and Haslam, 2006;Gallhofer et al., 2011;Graham and Neu, 2003;Lehman, 2009;Samsonova, 2009;Suddaby et al., 2007). One set of arguments in this body of research has been that local traditions and rules are dismantled with the aim of achieving global uniformity. Suddaby et al. (2007), for example, examine the role of large accounting firms such as the Big 4 "in the emergence of a transnational regulatory field in professional services" (p. 333). In this sense, the Big 4 and other GPSFs are presented, not as mere products of globalization, but key architects of it as well (Morgan, 2009).
A number of studies have explored the influence of globalization on the role of audit firms and the emergence of audit regulations in different contexts. Barrett et al. (2005) look at the coordination of multinational audits in different sites of a large multinational audit firm and argue that such coordination "links the global and the local in a dialectical manner" (p. 1), without the one dominating the other. Local appropriations of global prescriptions are always possible (Barrett et al., 2005, p. 21). Samsonova (2009) examines the emergence of audit regulation in the context of globalization, specifically from the perspective of Russia. Applying the concept of "transnational communication" she focusses on relevant actors to show how global policy templates "penetrate the local audit environment". Samsonova (2009) concludes that "representation of globalization as an impersonalised power ruling the world fails to convey the increasingly transnational nature of the present-day world order where, apart from the nation-states, cross-border activities of various private and public actors and organisations with transnational jurisdiction increasingly shape national practice localities" (p. 528).
There has been considerable work on PSFs in other professional jurisdictions as well. Muzio and Faulconbridge (2013) recount the failed attempt of English law firms to dominate the Italian legal market, highlighting various regulatory, cultural and institutional factors in the local context that effectively thwarted the imposition of global prescriptions. The implication of this study is that attempts to globalize a "one firm" model almost inevitably come undone in contexts where formal institutional coverage is strong. Conversely, Faulconbridge and Muzio (2015) tell a slightly different story in Germany, where English law firms were more successful in actively shaping institutional arrangements to their own advantage, leading to greater uniformity in the activities of PSFs across borders.
The work of Boussebaa on management consultancies is also very relevant to the present study. In analyzing consulting PSFs across different national contexts, Boussebaa (2009Boussebaa ( , 2015a has shown that "competing rationalities emerge […] from different national institutional contexts" (p. 829). Even when pursuing uniform strategies in different national units, this leads to inter-unit conflict over economic rewards. As a result, putatively "global" firms end up having de facto "polycentric" (Boussebaa, 2015b) control regimes, leading to tensions between "core" and "peripheral" offices. In a related study, Boussebaa et al. (2012) note that these tensions undermine the global pretensions of the firms in question as they "were in reality far from having transcended national boundaries" (p. 479). In other words, there is a "fundamental clash between a shared aspiration to be truly "global" and the reality of local, institutionally conditioned material interests" (Boussebaa et al., 2012, p. 482).

Big 4 in Bangladesh
One theme related to this concerns the form of international expansion that PSFs take. Boussebaa and Morgan (2015) recently identified four different forms of PSF internationalization: network, project, federal and transnational. They explained such differentiation based on different modes of governance, client requirements and forms of knowledge. However, little is known about the internationalization of PSFs into emerging countries.
Even where works do seek to shed light on the expansion of professional services into emerging markets, they tend to offer little detail about the extent to which there are differences between PSFs in "core" vs "peripheral" contexts. For example, Cooper et al. (1998) examine the case of opening a new office in Russia by one of these firms (Big 6 at that time). Using theories of identity, nationalism and imperialism the study notes that the competitive position of the firm in Russia was compromised as a consequence of the views taken by the American members of the firm as an opportunity to support existing multinational clients at the cost of local opportunities. However, Cooper et al.'s (1998) study privileges the views of Big 4 participants in Canada, the UK, the USA and Germany (these being the investing firms in the new Russian venture) rather than Russian participants themselves. The opportunity to highlight local idiosyncracies is therefore very limited from this perspective. Even where studies do engage local actors in more peripheral environments, they often downplay the local context. Ruiz Castro (2012), for example, looks at gender imbalances in Big 4 firms in Mexico but it is not clear to what extent her findings are specific to the Mexican context or generalizable to Big 4 firms globally. The sole published study to date that has been undertaken on the Big 4 in Bangladesh, equally struggles to speak to the globalor otherwisenature of GPSFs. Siddiqui et al. (2013) examined whether Big 4 affiliated firms attract audit fee premiums in Bangladesha market "characterized by low levels of audit fees, concentrated ownership structure, and lack of demand for quality audit services"(p. 332). They interestingly conclude that Big 4 affiliated firms do not generally earn fee premiums in Bangladesh, although this does not permit much wider reflection on the nature of global firms, which is the central concern of this study.
There has therefore been only very limited work looking at the market for professional services in emerging economies. What we do know about "global" PSFs tends to be based on data from OECD countries. The empirical coverage of "global" PSFs has curiously not been "global" itself. More work is therefore needed looking at more peripheral contexts . Indeed, Sassen (2003) argues that peripheral contexts, rather than global financial centres, actually hold the key to understanding how globalization actually operates: Studying the global, then, entails not only a focus on that which is explicitly global in scale, but also a focus on locally scaled practices and conditions articulated with global dynamics, and a focus on the multiplication of cross-border connections among various localities fed by the recurrence of certain conditions across localities (p. 3).
Beyond offering empirical novelty, it is important also to provide a conceptual framework capable of offering wider insights into, not only professional service work, but the nature of PSFs in processes of globalization per se.

System, Society and Dominance (SSD) Effects
In the broader literature on cross-national analysis of multinational firms, two distinctive positions can be distinguished. On the one hand, there are studies which emphasize the importance of globalizing trends, considering multinational companies (MNCs) as "globalizing agents" that are "denationalising from their domestic origins in the course of developing genuinely global strategies of operation" (Sklair, 2001, p. 48). These studies argue for the gradual removal of national diversity and increased homogenization of 148 AAAJ 30,1 organizational practices. On the other hand, there are those which show a penchant for highlighting areas of diversity and heterogeneity when applied to the internationalization of firms (e.g. Whitley, 1992), leading to the celebration of diversity and, according to some commentators, the obfuscation of processes of power and domination (Willmott, 2011). While acknowledging the merits of these studies, scholars have argued that neither of the above mentioned approaches fully recognizes the complexity of real economies because the nature of work within modern multinational organizations is more complex than global/ local and convergence/divergence dichotomies (Smith, 2008). In an attempt to transcend such polarization in the analysis of multinational firms, Smith and Meiksins (1995) elaborate an alternative theoretical perspective, namely, the SSD framework.
The SSD framework suggests that work organization is shaped by a three-way interaction, in which the nature of the political economic system ("system effects"), the societal features within which the firm is located ("societal effects") and the characteristics of the dominant economy, leading sectors and leading firms ("dominance effects") all influence organization and work in any country. Smith and Meiksins (1995) argue that each of these three sources of influence need to be considered when looking at how multinationals (of which Big 4 firms can be considered-specific variants thereof) behave and co-ordinate work in different countries. Moreover, the interrelations between these different sources of influence also need to be drawn out by empirical work. A schema of the SSD framework is presented in Figure 1.
"System effects" in this framework refer to the common elements of a certain type of political economy, be it capitalism or state socialism. System effects also refer to the forces that transcend national borders and have fundamental implications for industrial society, such as the development of global science and technology. Invariably, the mode of production surrounding Big 4 professional service firms is a capitalist one, underpinned by competitive market relations, the accumulation of capital and the antagonistic Big 4 in Bangladesh capital-worker relationship. Such a system has a general tendency to engender contradictions, conflicts and uncertainties. In addition, international markets for professional services, as a product of globalization, also impact upon the organization of PSFs in different countries. However, system effects are institutionally mediated through the specific institutional complexes in different countries, these being designated by Smith and Meiksins (1995) as societal effects.
"Societal effects" highlight the enduring institutional differences across countries. They suggest that organizations are embedded in societies in which formal institutional rules as well as informal social norms and customs all play important roles in shaping work and organizational practices. Indeed, capitalism cannot be thought of as uniform or homogenous in every context, but perhaps takes several varieties (Hall and Soskice, 2001). An alternative interpretation from regulatory theory would characterise global capitalism by both unity and heterogeneityor argue that it is variegated ( Jessop, 2012). As we will detail in the next section, societal effects in Bangladesh include strong family networks, existence of accounting dynasties and nepotism, informality, the emphasis on social capital and so on, which are different to western economies.
"Dominance effects" refer to the "best practice" generated and diffused by the "society-in dominance" within the global economy at a particular period of time. Such effects reflect "the uneven nature of economic development and economic power", and explains the "tendency for one society to take the lead in evolving work organisation and business practices considered more efficient than those operating within other countries" (Smith and Elger 2005, p. 66). As firms with Anglo-American origins, and as PSFs more generally can be thought ofat least to some extentas architects of globalization, the Big 4 might be characterized as being implicated in the globalization of Anglo-American models of corporate governance. At a field level, they might also be thought of as proselytizers of Anglo-American modes of professional organization.
The SSD framework allows us to integrate the international and national levels of business context, thus offering a holistic approach to investigating the extent to which the Big 4 can be considered truly "global" in nature. First, the notion of "system effects" recognizes that certain organizational models or practices could be adopted in Bangladesh by virtue of their general and systemic nature, for instance, the pursuit of efficiency and commercial benefits. Moreover, the identification of "dominance effects" is particularly pertinent in analyzing the Big 4. For example, dominance effects might be evident from the standards, procedures and practices that have been developed in a more western context. Finally, while both system and dominance effects highlight the influence at the global level, the attention paid to "societal effects" recognizes that firms are context-dependent and not subject wholly to a standardising or universal rationality. They allow us to look into the specific settings in which Bangladeshi firms are embedded and investigate the extent to which these settings influence the adoption and implementation of a certain standard or "best practice".
Apart from the identification of these three sets of effects, it is crucial to understand the interactions between them and how they play out in Big 4 firms in Bangladesh. We recognize that the SSD effects are all present in Bangladeshi firms. They work in consistent or inconsistent directions, and give rise to the specific configurations of organizational practices in Bangladesh. As Smith (2008) has emphasized, "it is not the aim of the SSD model to elevate one effect above another, but, rather, to locate analytically the sources of change and continuity within the international firms" (p. 41). In other words, the specific constellation and interrelation of these three sources of influence on Big 4 work practices will vary from country to country and can only be uncovered through careful empirical analysis. It cannot be known a priori, for example, the extent to which dominance effects override or combine with local culture to reshape institutions. Neither is it clear the 150 AAAJ 30,1 extent to which Bangladesh, as a developing economy, is more likely to passively accept what global headquarters consider "best practice" or whether local idiosyncrasies might prevail instead. Mapping the interrelations between these three sources of influence on Big 4 Bangladeshi work practices permits an elucidation of how the global interacts with the local (Sassen, 2003). In turn, consideration of the dynamics of these different sources of influence permit consideration of the extent to which the Big 4 can be considered truly "global" in nature (Boussebaa et al., 2012).
The socio-economic context of Bangladesh Bangladesh separated from Pakistan in 1971 following a nine month long liberation war (Sisson and Rose, 1991). Immediately after its independence the Bangladeshi government pursued socialist policies for a brief period ( Jahan, 1973) whereby it nationalized all its mills and factories, only to then reverse these decisions a few years later in 1975 following a military coup (Rahman, 1993). Since then, induced by international agencies such the World Bank and the International Monetary Fund (IMF), a private sector-led denationalization policy has been pursued leading to the privatization of major state owned enterprises (Boughton, 2001;Rodrik, 2006). In 1991 an ostensible parliamentary democracy was introduced following a political struggle (Rashiduzzaman, 1997), leading to a development strategy based on market economy principles. A series of reforms, including trade liberalization programmes, were introduced to improve the corporate governance and capital market structure in Bangladesh. During this period foreign direct investment (FDI) increased significantly from a meagre $3.2 million in 1990 to $1,501.6 million in 2013 (http:// databank.worldbank.org/data/views/reports/tableview.aspx, 22 June 2015). Although the Big 4 and other GPSFs have had a presence in the country since the 1970s, their visibility has increased in recent times with the prospect of steady economic growth (on average 6 per cent per annum) and Bangladesh aiming to become a middle income nation by 2021. Most of the economic forecasts by market analysts, such as Goldman Sachs, PwC, J.P. Morgan, Moody and Standard and Poor, point in this direction.
Mainly inspired by market economy principles the current economic mode of production is organized via a capitalist system. However, the development of an ideal-type form of modern capitalism is thwarted by many traditional and cultural features in Bangladesh such as colonialism, exploitation, inequalities, families and kinship (Uddin and Choudhury, 2008;Uddin and Hopper, 2001). These phenomena of family connections and kinship are deeply rooted in the socio-political and economic history of Bangladesh (Uddin and Choudhury, 2008) and have led to the development of what some have referred to as crony capitalism (Uddin and Hopper, 2001).
Bangladesh, like many other Asian business systems (Witt and Redding, 2013) such as India, Japan and China might be characterized by strong family networks, informality and forms of social capital that are different to western economies. Most of the business conglomerates in Bangladesh are owned by families and are privately held companies. Spence et al. (2016) talk about the "existence of accounting dynasties and nepotism" in the Bangladeshi accounting profession. Family loyalties and connection to the closer circuits of political power determine the likelihood of better outcomes in business transactions rather than professional skills and judgement (Uddin and Choudhury, 2008). Mair et al. (2012) observed informality in dispute resolution systems which tended to privilege more powerful groups in society, usually operating via favours and bribes. Accumulation of social capital in Bangladesh depends on family networks and political connections. Social status is determined by where you come from rather than what you can do. According to McKague et al. (2015) on Hofstede's (1984) cultural framework, Bangladeshi society would be characterized by high-power distance, uncertainty avoidance, masculinity and group orientations.

Big 4 in Bangladesh
Principally instigated by the World Bank and IMF, the Bangladeshi corporate governance reforms were shaped by a rational/legalistic Western Anglo-American model of corporate governance. However, this has been mediated by the traditionalist socio-cultural factors noted above (Uddin and Choudhury, 2008). For example, with family dominated boards whereby it would not be unusual to see a father acting as the chairman, while his son or daughter serves as CEO. This means that corporate decision making is more likely to be driven by family interests rather than the interests of other shareholders and wider stakeholder groups in society. In this scenario managers would be rewarded for family loyalty rather than the exercise of professional judgements in decision making. This clearly undermines the expectations of a rational legalistic model of corporate governance.
In Bangladesh, foreign accounting firms essentially partner with local firms. As such, the Big 4 operate in the country through local avatars, each of which is legally obliged to sign off on statutory audits in their local names. There are historical reasons that explain this phenomenon. Most of the Big 4 firms entered the country before 1971 during the period of Pakistani dominance. Hoda Vasi was established in 1972 by taking over from its predecessor J.F. Ferguson and Company which operated in Pakistan for a long period of time. One of the founding partners of Rahman Rahman Huq was an ex-partner of what was previously Price Waterhouse Peat. The current senior partner of A Qasem & Company was trained in the UK with PwC. One of the founding partners of S.F. Ahmed worked in one of the local predecessors of a Big 4 firm. Thus, most of the founders of the Big 4 affiliates were either trained or worked in the UK in a Big 4 firm or in one of the Big 4's predecessors. These historic links perhaps nurtured a desire among the partners to have some affiliations with the Big 4 firms when they returned to their native Bangladesh to start audit practices.

Research method
In order to explore the principal research objectives, a qualitative field study was undertaken between April 2012 and March 2013. We interviewed senior members of Big 4 firms in Bangladesh. Using personal contacts, the partners and managing partners of each firm were contacted and interviews arranged. The sample is skewed towards partners as it was perceived that they would have the most comprehensive overview of their firms' relationships with global and regional centres. The average duration of each interview was 50 minutes approximately.
As at 1 July 2012 there were 23 partners in Big 4 firms in Bangladesh. We successfully managed to interview 18 of these individuals as well as a partner in a non-Big 4 yet internationally recognized firm. Four of the above interviewees are managing partners in each of the Big 4 firms in Bangladesh. Eight of them qualified from the Institute of Chartered Accountants of England and Wales (ICAEW). Seven of them were, at one point, presidents of the local institute -The Institute of Chartered Accountants of Bangladesh (ICAB). Out of the 19 interviewees three were female. Table I shows the breakdown of the interviewees by firms and their profiles.
It is worth pointing out at this point that Big 4 firms operate in Bangladesh under the status of member firms or correspondent firms. Member firms share the same name as the global firms and become an integrated part of the network. They are shown as "member firms" within the firms' global website and are expected to follow similar quality standards. They are expected to pay a royalty fee to the centre (global parent). On the other hand, correspondent firms are not an integral part of the global firms and do not appear within the firms' global website. They operate via a loose cooperation agreement with the global firms. The terms and conditions of these agreements could vary from country to country. More discussion on this convoluted relationship will be offered in the next section.
The majority of the interviews were conducted in the interviewees' place of business in two major cities in Bangladesh -Dhaka and Chittagong. This is subject to one exception.

152
AAAJ 30,1 In the case of a retired partner the interview took place at the interviewees' residence located in one of the exclusive residential areas of Dhaka city. Subject to one exception all interviews were carried out in a face-to-face format. The research team was involved in various stages of planning, conducting, coding and analyzing the interviews.
A common interview guide was used during the interviews, which themselves followed a semi-structured format. Interviewees were asked their views on a number of themes from which globalization-related themes might be explored, such as: broad career histories of each interviewee, legal status of the firm in Bangladesh, continuing professional development both in Bangladesh and internationally, involvement with professional institutes, promotion processes, views about the firm's relationship with the Big 4 "parent", history of this relationship and the general context of professional services in Bangladesh.
All interviews were recorded (with the promise of anonymity) and subsequently transcribed. The interviews took place in English, which is the working language of these firms. The transcripts were subsequently analyzed and coded using an inductively generated coding schema (Miles et al., 2013). The guiding heuristic during the generating of this coding schema was the identification of sources of tension or difference between local Bangladeshi firms and their global "parents".

Findings and analysis
In this section, we report our findings on how the Big 4 have internationalized in Bangladesh; our analysis is informed by the SSD framework. We find that while the systemic and dominance effects have driven the Big 4 to enter the Bangladeshi market, these effects are mitigated by societal effects, which results in any global aspirations that the Big 4 might harbour being thwarted somewhat.

System and dominance effects: entering into the Bangladeshi market
In the first instance, all the Big 4 firms have entered into the Bangladeshi market, although through variegated means. One firm has full member firm status providing the Bangladeshi firm with full access to their parent's intranet, entitling them to send their practitioners to global training courses and allowing them to use the global firm's logo on stationery and letterhead on its communications. This organizational arrangement also has resource and

153
Big 4 in Bangladesh governance implications, meaning that the subsidiary has to pay annual dues to the global centre, as well as having its files inspected on a regular basis by the global centre.
The remaining Big 4 firms in Bangladesh, however, do not have a full member status (at the time of data collection). In fact, the internationalization of these firms is, to a large extent, consistent with what Boussebaa and Morgan (2015) call a "network" form, which involves independent firms in different national contexts linking together to provide international services to their clients. The local avatars of these firms each have less familial relationships with their Big 4 parents. The nomenclature used to designate these relationships could be "technical assistance" firm, "correspondent firm" (we use this phrase for each non-member firm for the sake of brevity and clarity), "independent corresponding" firm, "associate" or "cooperative" firm, depending on the organizations in question. Each of these terms is indicative of a definite, although distanced, connection. For example, the managing partner of one local firm described the effective meaning of these terms: Associates, cooperating, correspondent […] you can have a number of expressions, basically it doesn't mean anything other than that, we're a preferred choice, you know, for these international firms (Interviewee 11).
The wide presence of the Big 4 in Bangladesh reflects the influence of system and dominance effects. In terms of system effects, capitalist competition and the pursuit of profit in the global market make the affiliation an ideal choice for both the Big 4 and the Bangladeshi firms. Indeed, the commercial benefits of affiliation appear to be significant as essentially all of the major audit contracts of multinational firms come the way of these local member/correspondent firms. One interviewee described how the commercial benefits worked both ways, for the global and the local firm: Curiously, the commercial benefits that result from these guaranteed referrals possibly offer one explanation for why, unlike their western counterparts (Spence et al., 2016), Bangladeshi partners do not see themselves as entrepreneurial deal-makers but, rather, as accountants whose primary role is to deliver good quality work: Because we get so much business from [one of the Big 4 firms], to a great extent we do not have to aggressively look for business. If we did not get this big chunk of business coming in from [one of the Big 4 firms], chances are we would have to go and hunt for business, which we now don't have to (Interviewee 4). Dominance effects come into play when the Bangladeshi firms are pressurized to conform to the technical, quality and procedural standards of their Big 4 "parents". One interviewee described such diffusion of best practice guidelines: […] say, for example, [name of a Big 4], they had a client in Bangladesh […] [and] the global auditor is [in need of some input from us] […] they will send us their methodology, their systems and their reprints. We will review locally and then […] pour the information into the system and send to the global auditor, they will review it, and […] and make some queries […] we will then respond to the queries, and […] when they are happy, our job is finished (Interviewee 17).
Given this, being affiliated with the Big 4 offers opportunities for local firms to learn "best practices" from the leading firms. According to our interviewees, these learning activities often take place in global training courses and joint projects with international member firms: Overall, the systemic imperatives of efficiency and profit making, along with the pressure for the local Bangladeshi firms to conform to "best practices" underpin the Big 4s' internationalization into Bangladeshi market through establishing local affiliations.

Societal effects: tentative presence of the Big 4 in Bangladesh
Despite the well-established presence of the Big 4 firms in Bangladesh, our findings reveal that such a presence is tentative, fragile and incomplete, instead of being fully integrated or "global". As we mentioned, the majority of correspondent firms in Bangladesh do not have a full member status with their Big 4 parents. In other words, these local firms do not share in profits/losses with their global partners. They do not have full access to the global firms' resources and they attend training courses only if they are invited. Nevertheless, correspondent firms do feel as if they are part of a wider group, as the following exchange indicates:

Interviewer: As an independent member of [xxx] do you feel part of the [xxx] family?
Interviewee: I think I feel part of the extended family! As we can use the name on the [business] card and sometimes we can refer to them (Interviewee 18, emphasis original).
In some cases, local firms are not permitted to use the logo of their Big 4 partners. Indeed, the bye-laws of the ICAB state that firms need to trade under the name of their founding partners. In spite of such limitations, interviewees often felt that they were part of the Big 4: We are not [name of Big 4], we are not a member firm, but we feel that we are [name of Big 4]. We are proud to be the Big 4 (Interviewee 19).
These firms operate in a liminal space, being local correspondents instead of full members of the Big 4 firms. Such non-member status is something that one might think is indicative of a transitional stage, which full member status would eventually supersede the current arrangement. Our study suggests, however, that this non-member status is something that some of the firms have lived with for a long time. For example, in the case of one firm, links were first formed 37 years ago as a result of a Big 4 firm's need to have a local partner for one of their global clients. The date of the beginning of this relationship suggests that the globalization of professional services goes back at least 37 years in Bangladesh: […] in 1978 we started with [name of parent]. So, that was when we first audited a multinational client in Bangladesh on behalf of [name of parent]. So that was how the link first started (Interviewee 11). Therefore, this liminal space that the firms occupy is one that is well established and has become routine for them. In this sense, societal effects play a significant role in keeping system and dominance effects at bay.
The "correspondent" relationship also leads to contradictions and tensions in the field which might be hard to sustain in the long term. The ICAB can only place restrictions on the provision of statutory audits. Other professional services, such as consulting or tax advice can be offered by foreign firms, although only when there is the opportunity of a long-term contract. Otherwise, non-audit services will be outsourced to regional centres: […] if they get a job such as internal audit, or some other job other than statutory job, they send those jobs to India. […] From India they get it done and maybe, in some cases they come from 155

Big 4 in Bangladesh
Pakistan but recently if they get a good, big project, may be, for 5, 4 or 3 years, then they set up an office here. I know [name of a Big 4], they have an office here (Interviewee 18).
This gives rise to the situation whereby only one Big 4 firm has a consultancy practice in Dhaka, but they need to refer statutory audit work to a local firm, their correspondent firm. Such scenarios do not arise in the case of full member firms; full member firms provide a full panoply of services to clients: audit, tax and advisory. We were told by one interviewee that advisory work was the most lucrative service line in terms of billing rates per hour. Those who are correspondent firms are therefore not fully fledged PSFs, but are essentially audit and tax firms. This situation, we were told, was unlikely to change in the foreseeable future for a number of reasons, which we will elaborate upon below.
Why do the Big 4 not institute a more comprehensive delivery of professional services in Bangladesh? While the system and dominance effects offer explanations for why and how the Big 4 and the Bangladeshi firms could both benefit from internationalization, they do not sufficiently explain why the Big 4 have not instituted a full takeover of the market for professional services in Bangladesh. Here, the societal effects of Bangladesh and how these societal effects mediate the system and dominance effects warrant attention.
In the field of professional services, three societal effects of Bangladesh are evident, namely, the low level of professionalism, the low-audit fees associated with professional service work, and the strong presence of accounting dynasties and nepotistic appointments. Each of these impacts on the system and dominance effects evident in the Big 4 in Bangladesh.
First, there is a general perception among local firms that the level of professionalism is not sufficiently high at the moment to keep up with "best practices" and therefore that greater international integration with the Big 4 could not be justified: If we become member firm of [one of the Big 4 firms], then we have to follow all the rules, regulations and standards. In our country, the professionals we have, I think, our expertise is not that level, which is the main limitation (Interviewee 19).
Similar views were expressed in other firms as well: In terms of quality and [in respect of] all other things this profession is not that much developed in this country (Interviewee 1).
It is curious that Bangladeshi accountants cite the comparative lack of professional quality as a potential barrier to entry here. The vast majority of interviewees were trained with the ICAEW (a historical dominance effect) and were all actively involved in the ICAB, some of whom were the past-presidents. Their involvement in the profession was significant and their identity was very much built around the rigorous application of auditing procedures, something that is not commonly articulated by senior Big 4 accountants in other contexts (see e.g. Cooper and Robson, 2006). Our interviewees thus simultaneously place the traditional accounting professional on a pedestal in terms of valuing independence, but then feel somehow unable to embody such a role themselves because of their limited ability to keep up with "best practice" guidelinesa dominance effect.
Another societal effect that affects the dominance and system effects is the relatively low fees for professional service work. This concern was articulated by interviewees from more than one firm, although in different ways. For example, the following interviewee laments the Bangladeshi view of audit as a legal requirement rather than as a value-added service: I think one question is generally asked by different quarters why the audit profession in Bangladesh is not developed that much. I think one of the reasons is that audit fee is very low […]. Unfortunately in our country people see audit as legal requirement means audit for advocacy, audit has no value [added dimension] here. That is the main problem (Interviewee 18).

156
AAAJ 30,1 From a conventional public interest perspective this "legal requirement" view of audit could possibly be interpreted positively, but from a commercial standpoint it has the effect of dampening client enthusiasm for the whole audit enterprise. Ideological views of what audits can achieve for companies aside, the economic situation in Bangladesh simply means that the client base is very weak, as the following interviewee outlines: [one] limitation is, our client base [is very small] […] Bangladesh is a very small country, our professional fee is very low. So it will be not possible for us to comply with [Big 4] standards […] as the fee level is very low (Interviewee 19).
The relatively low-fees paid for professional service work in Bangladesh represent a societal barrier for Bangladeshi firms to developing full member status with the Big 4. This is not only due to the fact that the annual royalties they have to pay to the global firms would be insurmountable, but also because of the expensive cost related to the full conformity with the "dominant" work processes and information systems. For example, many businesses in Bangladesh currently use pirated software on their computers. This is something that would have to change as part of the risk management and governance requirements of being part of a global organization. The following, seemingly trivial examples were cited as serious indicators of the extra burdens that would be placed upon local firms should they ever adopt full member status vis-à-vis their global "parents": By way of elaboration, although these software packages are readily available and used in pirated versions, certain software systems cannot be plugged into global intranets unless they are purchased with institutional licenses. Again, this thwarts attempts at implementing "best practice" guidelines.
The same concern also reduces local firms' enthusiasm to participate in the expensive training courses and continuing professional development within the Big 4 networks: No, but we are always encouraged to participate in workshops. Every day I check emails of [name of Big 4] Singapore. They are very expensive. For two day's workshop they charge one thousand dollars. Plus hotel, plus ticket (Interviewee 12).
Last but not least, the existence of accounting dynasties and nepotistic appointments in Bangladesh thwart attempts at implementing "best practice" guidelines, particularly global human resource management (HRM) practices such as appraisals, promotions and retirements. Some of the local firms are essentially family firms, run by different generations of the same family. This means that one can make partner within six years of being in the firm if they come from the right family compared to the UK or Canada where making partner typically takes a minimum of 15 years. So partners are younger in Bangladesh. At the other end of the spectrum, partners can also be much older as, unlike in most western countries, there are no forced retirement ages written into partnership agreements. A number of interviewees noted that most of the senior partners in their firms were in their 70s although Big 4s tend to have a set retirement age of 65.
This was mentioned as a potential barrier to greater Big 4 involvement by a number of interviewees. Indeed, one interviewee described how they had formally adopted their global parent's policy of forced retirement at 65, but they managed to retain a clause whereby a partner could continue working beyond this age provided that he received unanimous approval from the other partners. This gave rise to the situation of one individual working until the age of 75, at which point he decided himself that he would stop working.
Even retirement in itself would not guarantee a break with the old accounting elite. In one firm we were told that partners technically do not retire, but merely become "inactive". They still turn up to partner meetings, firm's social events and apparently have a significant impact upon the governance of the firm. This firm's global parent has, we were told, a real problem with this scenario, but the local firm is in something of a double-bind as its own partnership agreement allows it. It is the subject of continual tension between the two entities. Essentially, dominance effects are mediated by societal effects here. Asian business systems tend to privilege seniority (Witt and Redding, 2013); yet "global"or westernapproaches to promotion, at least in the Big 4, tend to privilege performance and socially construct age in a way that frames those of a certain age as unable to work at the levels of intensity required of a contemporary Big 4 partner. More comprehensive internationalization on the part of the Big 4 would necessarily entail a denationalization of particular components of the national such as the privilege accorded to older partners in the Bangladeshi context. However, these individuals are deeply embedded into their local contexts and thus retirement ages present a clear example of the local and the global being in tension with each other.
In summary, the incomplete presence of the Big 4 in the Bangladesh is manifest in the interplay between SSD effects. At the global level, system and dominance effects drive the Big 4 to enter into the Bangladeshi market and to establish affiliations with local firms. At the local level, societal effects prevent the Big 4 from more fully integrating themselves into the market for professional services in Bangladesh. Therefore, globalizing forces are mediated through local institutional and cultural factors, explaining the merely tentative presence of the Big 4 in Bangladesh, as illustrated in Figure 2.

Discussion and conclusion
Overall, we can conclude, along with other studies that PSFs fail to be truly "global" (see Boussebaa et al., 2012;Muzio and Faulconbridge, 2013). What we offer beyond existing work is a novel explanation as to why this is the case. Both global and local factors explain the current status and work practices of Big 4 avatars in Bangladesh. Or, in terms of the theoretical framework employed here, the Big 4 in Bangladesh are caught between system, societal and domination effects. The legal requirement to sign off on audits in the name of

158
AAAJ 30,1 local firms is clearly a societal effect. This does not constitute a protectionist mechanism so much as a cultural norm. After all, India, Japan and China also require this of Big 4 firms, yet for all intents and purposes the Big 4 are the Big 4 in those countries. Other barriers to greater integration offer a greater explanation of the somewhat tentative presence of the Big 4 in Bangladesh. Best practice guidelines, which are associated with domination effects, in terms of audit methodologies and HRM practices are only partially implemented in the local context. This is due to commercial barriers (e.g. the cost of software and training programmes) as well as deference given to local, cultural factors (societal effects) such as the seniority principle and the way in which networks of nepotism and patronage operate. In many ways, these scenarios are the product of a lack of systemic effect in the form of an already well-developed market for professional services in the country. Should there be more money to make, one would anticipate that the local firms would all move quickly to full membership status and that promotion processes, software packages and the training of personnel would similarly be brought more into line with global standards. Without the economic incentives to do so, it is unlikely that these things will change in the near future.
One curious finding relates to the articulation of relatively rigorous discourse of professionalism that revolves around ethics, integrity and independence in contradiction to commercialism and the proliferation of non-audit services. In this sense, Bangladeshi accountants come across as more archetypal professionals than their western counterparts. Yet in another sense they lament their inability to implement "best practice" as set out by global headquarters. It may well be that traditional professional virtues of independence and public service are not best served by what the Big 4 set out as "best practice". Indeed, the Big 4's increasing detachment from such virtues has been well documented (see e.g. Suddaby et al., 2007). It would be interesting to observe in the future the effects that any increased adoption of "best practices" has on Bangladeshi understandings of what it means to be a professional accountant.
Overall, societal, domination and system effects combine to produce a situation that gives the Big 4 only a tentative presence in Bangladesh. In spite of such barriers, a number of interviews were nevertheless optimistic about the possibilities of greater Big 4 involvement in the Bangladeshi market for professional services. For example: Many local big [business] groups are now approaching us. They want to get their financial statements signed off by Big 4 [affiliated] firms like us but five years, ten years back it was not like that. The scope is enormous (Interviewee 14). This statement recognizes that the Bangladeshi economy has grown in recent years and that FDI has increased substantially. Therefore, it is not inconceivable that the client base changes further in recent years, bringing with it concomitant effects on fee income, views towards the value of audit and professional quality standards. Indeed, some interviewees are of this view, as the following quote indicates: We are the member firm that means we are under the strict quality control review of parent monitoring and controlling system. So, that is really important. But the other three firms, they are saying they are representative of those Big 4s, they are not under the control and supervision, in terms of methodology, tools and audit process of the Big 4s. So, [as part of] globalization process in the developing countries the Big 4s, they are moving to the developing countries. For example, KPMG in Bangladesh [has member status]. So, may be in the next five to ten years time we are expecting that there would be PwC in Bangladesh, the Deloitte in Bangladesh. I think if they are coming as soon as possible that would be actually beneficial to us. Then you can say the playing field would be level, because now the playing field is not level [with only one full member firm in Bangladesh] (Interviewee 1).
As Sassen (2003) astutely observes, globalization is an incomplete project. Local institutional factors are not necessarily completely undermined by global processes, even 159 Big 4 in Bangladesh though MNCs often possess the symbolic and material resources to transcend local institutional factors (Smith and Meiksins, 1995). What this study has shown is the way in which domination and system effects can be subject to societal mediation and in some sense thwarted. By looking at the Big 4 in Bangladesh, we can learn much about professional service work in peripheral contexts, but we can also learn about the wider aspirations of international firms and their hesitant attempts to extend their global reach. Essentially, the Big 4 are present in Bangladesh, but in a rather tentative fashion. Their restricted service lines and lack of legal status represent accommodations with the local economic and regulatory contexts, such as the ICAB's insistence on local firms having to sign off statutory audits, the entrenchment of an existing accounting elite and the fact that the economic conditions are simply not tantalizing enough to justify full membership status for most of the firms. Part of this explains why the Big 4 are unwilling to invest themselves more fully in Bangladesh, but equally the legal barriers are more demand-side than supply-side. Morgan (2009) argues that developing country governments often show resistance to granting MNCs the same rights as local players in an attempt to build local industry. In the case of the Big 4, although they arguably pave the way for their multinational clients to operate in different contexts, they themselves do not bring capital or FDI.
In spite of these barriers, the brand of the Big 4 is present and their influence upon professional practice and the structures of local firms is evident from the empirical narrative above. That one of the local firms has full member status, for example, is indicative of the potential that KPMG's core sees in this peripheral context. One would anticipate that, should Bangladesh's economy continue to grow and present greater opportunities for professional services, the grip that these local institutional factors have on professional service work in the country have, will loosen.
Future research might usefully look at the market for professional services in other emerging economies, particularly in Asia. Work on "global" PSFs will be hampered if its empirical remit is limited only to western economies. The SSD framework employed here might usefully be taken to more "peripheral" contexts. Each country will have its own different systemic, societal and domination effects as well as different interrelations of these three. The results presented above may not be representative of emerging economies in general, or even of Asian business systems. Overall, more work is needed that explores the way in which global prescriptions overcome or are thwarted by societal and systemic factors at the local level.