Performance management practices within emerging market multinational enterprises: the case of Brazilian multinationals

Abstract

This study advances our understanding of HRM within emerging market multinational enterprises (EM-MNEs) by examining the extent to, and mechanisms by, which Brazilian MNEs standardise or localise their performance management (PM) policies and practices, and the factors that influence their design and implementation. We explored these issues through qualitative case studies of three Brazilian MNEs. The analysis of interview data reveals a strong tendency for Brazilian MNEs to centralise and standardise their PM policies and practices. The key finding of this paper is that PM practices within Brazilian MNEs are not based on indigenous Brazilian practices, but, rather, are heavily influenced by global best practices. The findings are at odds with previous research, which suggests that EM-MNEs apply different HR practices in developed country subsidiaries and developing country subsidiaries. Also, contrary to expectations, our results indicate that institutional distance does not have a significant influence on the adaptation of PM practices at subsidiary level.

Keywords:

• emerging market
• HRM
• multinational enterprise
• performance management
• policy transfer

Introduction

Over the past decade, there has been a rising interest in understanding how emerging market multinational enterprises (EM-MNEs) enter foreign markets and compete internationally (e.g. Luo & Tung, 2007; Ramamurti, 2012). A significant shortcoming of much of the existing literature is a lack of understanding about how EM-MNEs manage their activities and interact with their overseas subsidiaries (Thite, Wilkinson, & Shah, 2012). In this paper, we explore EM-MNEs' performance management (PM) policies and practices. This literature highlights a dilemma that many MNEs face: while standardised PM policies may help the MNE ascertain compliance with its policies and procedures, and ensure consistency in its strategic decisions (Coates, Davis, Emmanuel, Longden, & Stacey, 1992), effective PM policies need to be congruent with national cultural values and local practices (Amba-Rao, 2000; Srinivasa Rao, 2007), and for that they need to vary significantly between and within the MNE depending on host and home country factors (Coates et al., 1992; Rosenzweig, 2006). Accordingly, we view PM policies as a key arena in which the tension between global standardisation and local adaptation of human resource (HR) practices plays out in these firms. Indeed, understanding the tension between standardisation and localisation of management practice in MNEs has been a central question in international HRM literature for many years (see Prahalad & Doz, 1987; Rosenzweig & Nohria, 1994). Recent research has shown that the effective management of the pressures for standardisation and localisation of HR practices results in higher levels of subsidiary performance (Cogin & Williamson, 2014). However, our understanding of how these dynamics unfold in EM-MNEs is poor and there have been calls for further research in this area (Rosenzweig, 2006). Our study is guided by the following research question: to what extent do EM-MNEs standardise or localise their PM policies and practices? For the purpose of this study, PM policies and practices refer to the processes of setting, communicating and monitoring performance targets and rewarding results with the ultimate aim of enhancing organisational effectiveness (Fee, McGrath-Champ, & Yang, 2011, p. 366).

With regard to EM-MNEs, a study of PM policies may shed new light on the diffusion of management policies within EM-MNEs and on how the pressures for global standardisation versus local adaptation are managed in these firms. In particular, the extant research is not clear on the direction of the flow of HR policies between the centre of EM-MNEs and their subsidiaries. While there is evidence that MNEs tend to engage in ‘forward diffusion’ of their home country practices to their overseas subsidiaries (cf. Chang, Mellahi, & Wilkinson, 2009; Gooderham, Nordhaug, & Ringdal, 1998; Mayrhofer & Brewster, 1996), several studies reported that EM-MNEs are often engaged in reverse diffusion of best practices from their subsidiaries in advanced Western countries to the home country (Zhang & Edwards, 2007; Zhang, Tsui, Song, Li, & Jia, 2008). This reflects more recent research in this area, which recognises that standardisation decisions are not solely premised on the export of successful local practices to other units, but rather result from the integration of best practices to achieve economies of scale and scope (Festing & Eidems, 2011). Given the clear distinction between the formal Western instrumental PM system and the indigenous, typically relational, Brazilian PM policies, the results of this study may provide important insights as to what types of PM policies EM-MNEs are adopting.

The study of PM policies and practices is of significant importance because they signal the firm's strategic priorities to subsidiaries, managers and employees, and the types of behaviours that are expected and rewarded by the MNE (Biron, Farndale, & Paauwe, 2011; Fletcher & Williams, 1996). They also have far reaching consequences in assessing and developing employee competence, enhancing performance and distributing rewards (Cascio, 2006; Fletcher, 2001; Schuler, Fulkerson, & Dowling, 1991). In addition, research in the German context found performance and bonus systems to be central in standardisation efforts (Muller, 2001) owing to their strategic significance in the organisational value chain (Festing & Eidems, 2011). Furthermore, reflecting the lack of attention given to PM policies within MNEs in general, very little research is devoted to their study within EM-MNEs (Claus, 2008; Claus & Hand, 2009; Shen, 2004).

We focus on PM policies within Brazilian MNEs. While Brazil is predicted to be one of the leading world economies alongside other BRIC countries (Brainard & Martinez-Diaz, 2009; Hawksworth & Cookson, 2008), and Brazilian firms are internationalising in greater numbers than ever before (Fleury & Fleury, 2011; Lima & de Barros, 2009)3, compared with other BRIC countries, Brazilian MNEs are perhaps the least studied and therefore little is known about their operations overseas (Fleury & Fleury, 2009; Islam, 2012, p. 266). Moreover, the small body of research that explored management within Brazilian MNEs has focused on the headquarter (HQ)–subsidiary relationship and the role of subsidiaries within Brazilian MNEs (Borini, Fleury, Fleury, & Oliveira, 2009; De Miranda Oliveira & Borini, 2012), and broad HRM challenges faced by Brazilian MNEs (Muritiba, Muritiba, Galvao de Albuquerque, Fleury, & French, 2012).

The paper unfolds as follows. We begin by discussing the existing literature with a special focus on the standardisation versus localisation debate, followed by a brief background to Brazilian MNEs. The subsequent methodology section presents the case-study firms and the characteristics of the participating subsidiaries. The core section of the paper is the findings, which analyse the core features of PM policies of Brazilian MNEs and the degree to, and mechanisms through, which PM policies are diffused to the subsidiaries. In the discussion section, we extract general lessons and implications of our findings. In the final section, we briefly discuss limitations and the main conclusion.

Standardisation versus localisation of HR practices

A key tension for any firm operating globally relates to managing the tensions and contradictions emerging from being ‘simultaneously local and global in scope, [and] of being both centralised and decentralised’ in the management of their foreign operations (Evans, Pucik, & Barsoux, 2002, p. 6). This highlights the need for organisations to maintain a ‘dynamic balance’ between globalisation (implementing globally standard practices) and localisation (adapting practices to account for the host environment) if they are to become truly transnational (Bartlett & Ghoshal, 1998). However, as noted above, MNEs do not tend to standardise entire HR systems but rather focus on HR practices that are seen as strategically significant in the value chain (Festing & Eidems, 2011). Equally, recent research suggests that standardisation may occur in a phased way, with standardisation being rolled out in geographically proximate subsidiaries before more distant units (Colakoglu & Caligiuri, 2008).

External factors play a key role in terms of how localised HR practices are in MNE subsidiaries. For example, empirical research suggests that the degree of standardisation is mediated by the level of constraint in the host environment and the economic dominance of the subsidiary's parent country of origin relative to the host environment (Gunnigle, Murphy, Cleveland, Heraty, & Morley, 2002). This is often driven by the adaptation of practices to acquire legitimacy from government, the law, labour unions and other actors in the host environment (Gooderham, Nordhaug, & Ringdal, 1999). Indeed, based on their research, Geppert, Matten, and Williams (2003, p. 833) postulate: ‘the more globalized the strategies and structures of an MNC are, the more it allows for and relies on national specifics to play a key role in its global subsidiaries’. In other words, truly global firms not only acknowledge the need for adaptation of policies in different subsidiary operations, they actually appear to plan for it. However, our understanding is limited by the fact that this empirical work has largely unfolded in the context of MNEs from developed economies operating in similarly developed markets (cf. Chung, Sparrow, & Bozkurt, 2014).

Drivers of standardisation

There are a number of factors that drive the standardisation of practices in MNEs. One key factor is the institutional environment in which an organisation is founded and developed, which impacts on how managerial processes and structures evolve within the organisation and is likely to be reflected in the managerial processes and structures of the firm as it expands internationally (Almond, 2011; Edwards & Ferner, 2002). In line with other ‘Latin’ business systems, Brazil is a society with a high-power distance (Posthuma, Bisseling, & Sobral, 2011). Although there is significant variance between firms located in different regions (Islam, 2012; Lenartowicz & Roth, 2001), Tanure (2004) noted that power concentration is one of the key pillars of the Brazilian management system. Typically Brazilian firms tend to have ‘centralised decision making, with information controlled at top levels and relatively inflexible structures’ (Nicholls-Nixon, Sanchez Garcia, & Rivera Pesquera, 2011). Brazilian MNEs may extend such practices to their subsidiaries located overseas. Typically MNEs from high-power distance cultures tend to favour centralised practices and attempt to exercise control, while those from low-power distance cultures tend to favour consultative management styles with their subsidiaries (Brock, Shenkar, Shoham, & Siscovick, 2008). Brazil also scores high on uncertainty avoidance – the extent to which individuals in a society can tolerate ambiguity (Hofstede, 1994, p. 11; Volkema, 1999). MNEs from such cultures often ‘favour more formalized coordination mechanisms… (and prefer) the appointment of expatriates who are “tried and true” principals or trusted agents’ (Brock et al., 2008, p. 1297).

In addition, Brazilian firms operate in a high collectivist culture (Beekun, Stedham, & Yamamura, 2003) typically adopting a person-centred approach management style, and valuing non-monetary social goals over financial performance (Dant, Perrigot, & Cliquet, 2008; Nicholls-Nixon et al., 2011). Rodrigues and Betania (2002) reported that Brazilian employees feel out of their comfort zone in formal settings and often try to create a climate of personal intimacy and cordiality in business settings (Amado & Brasil, 1991). Thus, this cultural feature may translate into a strong emphasis on social results and relationships over hard performance measures such as financial and productivity measures by Brazilian MNEs. All in all, we expect to see at least some influence of these characteristics on HR practices in Brazilian MNEs.

Institutional drivers within MNE also influence the design, implementation and diffusion of HR practices. In considering PM systems, Decramer, Smolders, Vanderstraeten, and Christiaens (2012, p. 3) argue such systems ‘are shaped and embedded in a specific organizational and institutional context’. Edwards and Ferner (2002) explicitly point to the impact of increased emphasis on global integration of business operations in the MNE in exploring HR policy transfer. There is evidence of the increasing focus on global integration of HR in the contemporary MNE. The desire for global integration is driven by a number of factors including the development of a common corporate culture and the potential to enhance equity and procedural justice within the MNE through the transfer of organisational practices (Rosenzweig & Nohria, 1994; Smale, Björkman, & Sumelius, 2013).

Likewise, the MNE's strategic orientation is an important consideration. MNEs with a strategy to produce or provide globally standardised goods or services will logically desire to monitor subsidiary performance through benchmarking against standard practices that enable the quantification of performance along different dimensions (Morgan & Kristensen, 2006). Conversely, a strategy premised on providing more localised goods and services might emphasise the subsidiary's ability to provide the necessary expertise and skill, and the requirement for global integration may not be as evident (Taylor, Beechler, & Napier, 1996).

Drivers of localisation

There are equally institutional factors in the host country environment that challenge the deployment of standardised PM practices and drive greater localisation of such practices. For example, a significant body of literature challenges the universal applicability of ‘best practice’ PM policies and emphasises the role of national culture and institutions in driving localisation of such practices (Aycan, 2005; Cascio, 2006; Varma, Budhwar, & DeNisi, 2008). Institutional theory research advocates that firms need to conform to the social norms in a given business environment because they cannot survive without a certain level of external social approval (legitimacy) (DiMaggio & Powell, 1983; Meyer & Rowan, 1977; North & Thomas, 1973). Institutional scholars postulate that establishment of legitimacy – the perception that the policies are desirable, proper and appropriate with employees' norms, values and definitions (Suchman, 1995, p. 574) – in the host country is one of the main drivers for adapting practices to host country institutions (Jensen & Szulanski, 2004; Kostova, 1999; Kostova & Zaheer, 1999), and that managers and employees at subsidiary level are more likely to accept and internalise HQ's PM policies if they judge them to be legitimate (Forstenlechner & Mellahi, 2011). Fletcher and Perry (2001), for instance, warned that Western PM policies such as linking rewards to individual performance would be ‘unsafe’ in economies in transition. Similarly, Aycan (2005) advocated a contingency framework that links cultural and institutional factors as well as organisational factors with performance appraisal processes. In particular, in countries high in collectivism and high in power distance such as India, China and Brazil, firms tend to emphasise soft and subjective/indirect PM tools rather than the often used hard objective tools by firms in individualist and lower power distance cultures such as the USA and Northern European countries. Cascio's (2006) review of PM literature reached similar conclusions with regard to reward systems and the communication of PM systems. That is, in contrast to the Western-dominant model of performance-related rewards and direct/explicit communication of performance, in countries high in collectivism and high in power distance we would expect less emphasis on the link between individual performance and individual reward and communication of performance to be carried out in a subtle indirect way (p. 168).

Recent research has begun to delineate when and how local institutions influence PM policy and practice. For example, Cogin and Williamson (2014) displayed that in local environments characterised by higher levels of environmental uncertainty, higher levels of localisation of HR practices were associated with higher level of subsidiary performance. The institutional distance between the host and home countries has also been shown to impact on the standardisation/localisation of management systems (Kostova, 1999; Xu & Shenkar, 2002, pp. 609 and 610).

Institutional distance, defined as the extent of similarity and dissimilarity between the regulatory, cognitive and normative institutions of the two countries (Kostova, 1999; Salomon & Wu, 2012), emerges as a significant moderator of the level of localisation. Xu and Shenkar (2002, p. 610) noted that ‘a large institutional distance triggers the conflicting demands for external legitimacy (or local responsiveness in the host country) and internal consistency (or global integration) within the MNE system’. Generally, the literature indicates that as institutional distance between the home and host countries increases, external legitimacy becomes more important to MNEs than internal consistency (Xu & Shenkar, 2002, p. 614). The underlying premise here is that institutional distance increases local employees' ‘cognitive ability to understand the practice’ and rationale behind it (Jensen & Szulanski, 2004, p. 511). Indeed, research has shown that US MNEs generally introduce standardised practices to subsidaires in geographically and culturally proximate locations before more distant ones (Colakoglu & Caligiuri, 2008). This perhaps explains why Brazilian multinationals entry mode varies according to cultural distance (Ramsey, Barakat, & Monteiro, 2013). A key insight from this literature is that uniform application of PM policies across the MNE tends to break down as the firm ventures into institutionally distant locations, resulting in unique hybrid PM systems displaying both home and host countries characteristics (Lu & Bjorkman, 1997). This is because the larger the institutional distance, the less the compatibilities of the key facets of PM policies, and the harder for MNEs to transfer their HQ practices to host countries (Eden & Miller, 2004; Jensen & Szulanski, 2004, p. 511). Dossi and Patelli's (2008) study of the influence of MNEs' HQ policies and practices on Italian subsidiaries found that HQ influences decrease as institutional distance between the subsidiary and the HQ increases.

Interaction between the drivers of standardisation and localisation

It is evident that home and host country effects do not operate in isolation of each other and that the decision to standardise HR practices is a complex one. Drawing on the work of Smith and Meiksins (1995), Edwards and Ferner (2002) point to the importance of considering the relative performance of the home and host economies in which MNEs are located in understanding how PM practices look in subsidiary operations. Such a perspective suggests that strong economic performance in one country creates pressure for the diffusion to other countries of aspects of the system concerned such as HR practices. Such ‘dominance effects’ are reflective of the fact that at any point in time, countries ‘in dominant positions have frequently evolved methods of organising production or the division of labour which have invited emulation and interest’ (Smith & Meiksins, 1995, pp. 255 and 256; see also Almond, 2011; Pudelko & Harzing, 2007). Specifically, those MNEs from economies, which are higher up the hierarchy, may be perceived to have superior HR policies that may improve managerial practice in the host (Chang et al., 2009). In addition, where the subsidiary is located in a host, which is higher up the hierarchy of nation states, there is a possibility that the HQ will tap into local best practice that offers the potential for reverse diffusion of practices to the HQ (Edwards & Ferner, 2002). However, the interaction between the drivers of standardisation and localisation in EM-MNEs remains under-explored.

Background on Brazilian MNEs

Firms from Brazil were latecomers in the internationalisation process. The intensified outward FDI from the 1970s was largely due to the international expansion of a small number of large state-owned enterprises (SOEs), most notably Petrobras (oil and gas) and Companhia Vale do Rio Doce, or Vale (mining). From the early 1990s, outward FDI and the transformation of Brazilian firms into MNEs significantly accelerated thanks to a more favourable business environment, particularly Brazil's economic liberalisation and the formation of the South American common market MERCOSUR in 1991 (da Rocha & da Silva, 2009; Fleury & Fleury, 2011). While their origins lay in the early 1990s, Brazilian MNEs – with the rest of Latin American MNEs – came to global prominence after the year 2000 when high economic growth and high commodity prices led to soaring outward FDI, especially in the form of large-scale foreign acquisitions (Casanova, 2009, pp. 10–13).

Given their recent expansion, Brazilian MNEs remain at an earlier stage of the internationalisation process compared with developed country MNEs; for instance, the foreign assets of the top 20 Brazilian MNEs in 2006 ranged between 1% and 46%, with an average of only 20%. Indeed, the share of foreign assets was distorted upwards by Petrobras and Vale, which held more than three-quarters of the total foreign assets of the top 20 MNEs (Fleury & Fleury, 2011, p. 204). Furthermore, most Brazilian MNEs are still largely ‘regional’ rather than ‘global’, with the foreign share of total assets, employment and sales still largely dominated by Latin American markets (Ramsey, Resende, & Almeida, 2009).

Very different typologies of Brazilian MNEs have been proposed (Cuervo-Cazurra, 2008; da Silva, da Rocha, & Carneiro, 2009; Fleury & Fleury, 2009) and we can derive from them that the makeup of Brazilian MNEs is highly heterogeneous. In terms of industrial background, Brazilian MNEs encompass very diverse sectors, ranging from automotive, food and beverage, engineering to cosmetics. In terms of internationalisation motives, they are resource-seeking, market-seeking, efficiency-seeking and strategic-asset seeking (Fleury & Fleury, 2011). In contrast to Chinese or Russian MNEs, Brazilian MNEs are not dominated by SOEs and many leading Brazilian MNEs are privately owned (Fleury & Fleury, 2011). Furthermore, in contrast to MNEs from some other emerging markets, a number of Brazilian MNEs have highly sophisticated world-class technical competences, most notably Petrobras (deepwater oil and gas production) and Embraer (passenger aircraft manufacturing) (Carvalho, Costa, & Duysters, 2010; Fleury & Fleury, 2011).

Methodology

The topic of HR policies in general, and PM policies in particular, in EM-MNEs is an emergent field, which still requires a more careful conceptualisation and theory building, lending itself to a case-study approach as the most appropriate methodological approach (Eisenhardt, 1989).

For the purpose of our investigation, a sample of three Brazilian MNEs was chosen. Their key characteristics are presented in Table 1. All of the firms are headquartered in Brazil and have foreign subsidiaries in both developed and developing countries. The names of the companies have been anonymised for confidentiality, using a pseudonym based on their economic activity.

Table 1 Case study participants.

Company	Economic activity	International experience (2011)	Employees worldwide (2011)	Location of main subsidiaries	Location of interviewed subsidiary	Total interviews
BrazMan	Manufacturing	21 years	10,000	North AmericaLatin AmericaItalyChinaSlovakia	North AmericaChinaSlovakiaMexico	6
BrazCon	Diversified, includes construction and petrochemicals	31 years	130,000	North AmericaLatin AmericaPortugalGermanyAfricaUAE	PortugalGuinea	4
BrazCem	Diversified, includes cement and metals	10 years	40,000	North AmericaLatin AmericaPortugal	North AmericaBolivia	4

We conducted 14 interviews with the relevant managers between October and November 2011: four interviews with BrazCon and BrazCem each, and six interviews with BrazMan (see Table 1). Consistent with the traditions of naturalistic enquiry, the sampling method of selecting participants on the basis of their particular knowledge about the phenomena under study, with the aim of maximising the information that could be obtained, was considered appropriate (see Lincoln & Guba, 1985). Hence the interviewees were all senior managers, who were personally involved in HRM and specifically PM within each company. Interviews lasted on average 45–60 minutes and were recorded digitally and transcribed. However, two interviews lasted longer and some answers were provided in writing in follow-up discussions. All interviews were conducted in English but, in one case, an English-Portuguese interpreter was present during the interview to assist the interviewee. Before the interview, we asked each interviewee to provide basic background personal information on them and – in the case of subsidiaries – background information on each respective subsidiary. During the interview, we followed a semi-structured format that focused on two main aspects of PM – the practice of PM policies, and localisation. In terms of PM practice, the questions covered four areas: the origin of the present PM system; the philosophy underpinning the PM system; how the PM system operates (frequency, techniques, etc.); how the performance data are used (link to rewards, development etc.). In terms of localisation, the questions covered local employees, particularly managers, participation in PM-related policy-making and the extent to which local decision makers in the subsidiary are able to adapt the PM system. In order to avoid a possible bias towards standardisation, we asked each interviewee several differently worded questions about differences between the subsidiary and the HQs, and barriers to diffusion of HQ-level PM policies to the subsidiary. Company documents were also used to supplement the interview data.

We collected data from both HQs and subsidiaries in order to provide a holistic view of how Brazilian MNEs manage their PM policies throughout the firm. For each firm, we spoke with a senior manager at HQ-level (including the firm's director of HR in two out of three cases) and with senior managers in at least two different subsidiaries for each firm.

Given that previous research suggested that emerging market MNEs apply different HR practices in developed country subsidiaries and developing country subsidiaries (Khavul, Benson, & Datta, 2010) and given that there may be differences in terms of forward diffusion and reverse diffusion between developed country subsidiaries and developing country subsidiaries, our research design purposively includes interviews with both a developed country subsidiary and with a developing country subsidiary for each company. The key characteristics of interviewed subsidiaries are presented in Table 2.

Table 2 Characteristics of interviewed subsidiaries.

Company	Location of interviewed subsidiary	Subsidiary employees (2011)	Subsidiary ownership(%)	Subsidiary age (2011)(years)	Nationality of managing director	Percentage of expatriates among top executives^a(%)
BrazMan	North America	60	100	>20	American	0
	China	2300	70	16	Brazilian	30
	Slovakia	2000	100	14	Slovak	13
	Mexico	700	100	7	Brazilian	63
BrazCon	Portugal	250	100	>20	Brazilian	29
	Guinea	1300	100	1	Brazilian	100
BrazCem	North America	2800	100	10	Danish	56
	Bolivia	300	51	2	Brazilian	100

a Includes expatriates as percentage of top executives in the subsidiary; for North America subsidiaries, both US and Canadian executives are counted as ‘local’.

Data analysis was informed by key constructs identified from the literature review. Each of the three authors coded the transcripts independently in an iterative process with refinements of the coding categories agreed after each round of coding. The final analysis reflects the agreed coding of the three authors.

Findings

The practice of PM policies

In order to investigate how the standardisation versus localisation debate unfolded in our case firms, we first set out to investigate the origins of the present PM system of the respective case study firms. Guided by the literature review, we asked participants to outline the development and implementation of PM activities, or bundle of PM activities, deployed in the case firms. For the purpose of investigating the role of standardisation and localisation (see Table 3 for selected quotations), we have coded data according to three main areas: the origin of the present PM system, how the PM system operates (including frequency and techniques used), and how the performance data are used (such as its link to rewards and development) (see Tables 456).

Table 3 Global standardisation versus local adaptation: selected interviewee quotations.

Representative quotations supporting standardisation	Representative quotations of the limited examples of adaptation
Quote 1. ‘It's a miracle of it [sic], you are able to keep the same procedure throughout the years at different countries and different sectors’.(BrazCon manager in Guinea subsidiary)Quote 2. ‘All the big practices – promotion, salary increases, talent pool – all that comes from the corporate [headquarter]. The actual initiatives and the systems are from the corporate [headquarter]. We follow suit accordingly’.(BrazMan HR manager in North America subsidiary)Quote 3. ‘We need to keep a company culture. What we say to them, we have the Brazilian culture, we have the Chinese culture, but we have the BrazMan culture. It doesn't matter whether we are in China, we are in Slovakia, we are in Brazil, we need to follow the company culture. (…) If they [subsidiary staff] are not following our values, unfortunately we cannot keep them working for us. The main requirement to keep working for us is to follow our values [sic]’.(BrazMan HR manager in the Brazilian headquarter)Quote 4. ‘When I came onboard the large part of my mandate was to take the North American businesses and to align them from an HR point with the carbon copy in Brazil’.(Vice-President HR for BrazCem in North America)Quote 5. ‘This [PM] system was basically generated in Brazil and then we adopted that system for North America… As a general rule, it's pretty much identical for the both parts … It translates very well between Brazil and North America. I can't think of any differences of the top of my hat’.(Vice-President HR for BrazCem in North America)Quote 6. ‘The company as a whole has global systems regardless of cultural differences because that's the only way to ensure global mobility. It's difficult to move to another country where you find a different way to manage talent, to manage competencies. For this top level, the population we are talking about, the company needs to have global systems’.(Director of HR at BrazCem in the Brazilian headquarter)	Cosmetic adaptationQuote 7. ‘The secret is to be flexible – flexible enough to suit every community, e.g. there may be a different wording or approach for them to understand properly, but don't be flexible about the values and procedure’.(BrazCon HR manager in Guinea subsidiary)Quote 8. ‘Cultural aspects are related to interpretation and understanding (…) We try to stick very close to corporate programmes, following corporate guidelines, corporate timelines, corporate agenda, but we cannot just push it on people. There is a very important role in the organization in communicating and training. For example, the specific cultural example I mentioned [related to the different understanding of professional hierarchies in Mexico]. People focus on the name [of the job title] instead of the concept. We are training them what it actually means, so that they properly understand the concept of the model’.(BrazMan HR manager in Mexico subsidiary)Initial stage adaptationsQuote 9. ‘Initially it wasn't easy for them – subsidiaries – to follow rules and structures developed in Brazil but this changed quickly once they understood why we needed to do it’.(BrazMan Director of HR in the Brazilian headquarter)Quote 10. ‘There was much more of a cultural issue at the outset, all of a sudden there was a mass influx of Brazilians coming to our cement plants, and there were some cultural clashes at the time because there was a little bit of, you know, the Brazilians were ‘smarter’ than the North Americans and they were going to tell the North Americans how to make cement. There was a little of that at the beginning. And the Brazilian management style is still a little bit different from the North American management style, so we had some issues initially. We have both over time adapted to each other, and there is much less of that now. At this point it's a non-issue’.(Vice-President HR for BrazCem in North America)
	Subsidiary size and adaptationsQuote 11. ‘The biggest challenge I found is that the initiatives that are coming out of the corporate [headquarter], because of the magnitude of the initiatives, they have 4000 employees, we have 57, sometimes we have to taper those initiatives to be able to fit the manpower that we have here. So sometimes we need to think outside the box, to make sure that we meet all the criteria that they need, whether it's a volunteer initiative or even when there is something to come with processes, we have to taper that because we don't have the manpower to meet the same kind of outcome or the same kind of number that the corporate have put to us’.(BrazMan HR manager in North America subsidiary)Quote 12. ‘Just because of our size [of the North America subsidiary], we can't completely replicate the scope of the programme [Brazilian corporate university programme] but basically we have worked with Brazil very closely on this, taking the programme that they developed, so we introduced four modules of our corporate university programme to our North American bases and, in proceeding years, we are going to adopt two additional ones, so that we, as far as possible, replicate the whole corporate university model that exists in Brazil’.(Vice-President HR for BrazCem in North America)

Table 4 Characteristics of performance management policies of BrazMan.

	Headquarter	North America subsidiary	China subsidiary	Slovakia subsidiary	Mexico subsidiary
Institutional distance		High	High	High	Low
Subsidiary age		>20 years	16 years	14 years	7 years
Origin of present PM system	Headquarter PM system based on Western best practices adopted throughout the company world-wide, but local HR managers consulted before new HR tools developed
Philosophy underpinning PM system	Globally standardised PM system, PM used as a strategic HR practice, strong emphasis on a common corporate culture that is fully adopted throughout the company world-wide
Use of performance data	Performance data determine rewards and compensation, inform talent management (training and development etc.), and benchmark subsidiaries
Operation of PM system:
Sources of information and feedback	360° system	360° system	360° system	360° system	360° system
Frequency of appraisal	Annual	Annual	Annual	Annual	Annual
Coverage of corporate PM policy in subsidiary	n/a	All administrative staff	All administrative staff	All administrative staff	All administrative staff
Facilitation of corporate PM policy in subsidiary	n/a	Corporate culture, training, reporting lines, use of expatriates	Corporate culture, training, reporting lines, use of expatriates	Corporate culture, training, reporting lines, use of expatriates	Corporate culture, training, reporting lines, use of expatriates
Main barrier to subsidiary diffusion of PM policy	n/a	Small size of subsidiary (about 60 staff versus 2300 in China and 2000 in Slovakia)	Cultural resistance to adoption of 360° appraisals	None identified.	Trade union rules related to blue-collar workers

Table 5 Characteristics of performance management policies of BrazCon.

	Headquarter	Portugal subsidiary	Guinea subsidiary
Institutional distance		Low	High
Subsidiary age		>20 years	1 year
Origin of present PM system	Headquarter PM system adopted throughout the company world-wide
Philosophy underpinning PM system	Globally standardised PM system, strong emphasis on a common corporate culture that is fully adopted throughout the company world-wide, PM used as a strategic HR practice, some flexibility for subsidiary and project directors
Use of performance data	Performance data determine rewards and compensation, and inform talent management (training and development etc.)
Operation of PM system:
Sources of information and feedback	Internally developed evaluation form and personal action plan, influenced by Hays and other international firms	Internally developed evaluation form and personal action plan, influenced by Hays and other international firms	Internally developed evaluation form and personal action plan, influenced by Hays and other international firms
Frequency of appraisal	6–12 months	6–12 months	6–12 months
Coverage of corporate PM policy in subsidiary	n/a	All staff	All staff
Facilitation of corporate PM policy in subsidiary	n/a	Primarily corporate culture, also training, reporting lines, use of expatriates	Primarily corporate culture, also training, reporting lines, use of expatriates
Main barrier to subsidiary diffusion of PM policy	n/a	Autonomy of subsidiary or project director related to project-based nature of industry	Autonomy of subsidiary or project director related to project-based nature of industry

Table 6 Characteristics of performance management policies of BrazCem.

	Headquarter	North America subsidiary	Bolivia subsidiary
Institutional distance		High	Low
Subsidiary age		10 years	2 years
Origin of present PM system	Historically BrazCem allowed subsidiaries considerable autonomy in the design of PM systems (primarily owing to growth through mergers and acquisitions in North America). However, from 2007 a standardised headquarter PM system was introduced and adopted throughout the company world-wide, strongly influenced by international consultancy firms.
Philosophy underpinning PM system	The development of a standardised PM system was premised on modernising governance structures and standardising PM practices as part of this. Now the company has a globally standardised PM system, and strong alignment with corporate strategy and coordination and control of subsidiary operations. Strong emphasis on a common corporate culture that is fully adopted throughout the company world-wide.
Use of performance data	Performance data determine rewards and compensation, inform talent management (training and development etc.)
Operation of PM system:
Sources of information and feedback	Balanced scorecard	Corporate balanced scorecard (objectives, targets, KPIs and initiatives), cascaded down to subsidiary level	Corporate balanced scorecard (objectives, targets, KPIs and initiatives), cascaded down to subsidiary level
Frequency of appraisal	Annual	Annual	Annual
Coverage of corporate PM policy in subsidiary	n/a	All administrative staff	All administrative staff
Facilitation of corporate PM policy in subsidiary	n/a	Corporate culture, training, reporting lines, use of expatriates	Corporate culture, training, reporting lines, use of expatriates
Main barrier to subsidiary diffusion of PM policy	n/a	Some local resistance to performance-based individual rewards in subsidiaries with poor organisational performance	Autonomy of subsidiary linked to joint-venture structure involving a partner with a minority stake (51% BrazCem ownership)

Origins of the present PM system at the headquarter

In terms of the origin of the present PM system, with the exception of compensation systems, PM systems were developed in the HQs in Brazil. In all three cases, interviewees reported that the initial starting point has been the desire for a standard PM framework based on global best practices. These practices were developed in the cases of BrazMan and BrazCem in conjunction with major international Western-based consultancy firms, while BrazCon relied more on internal expertise. All three firms stressed the desire for a professionally operated HQ-designed PM system with universal applicability.

We found evidence that companies felt some initial pressure to adapt to local norms around PM in subsidiaries. However, over time, standardisation around the HQ-originated PM system became more evident. Most notably, BrazCem expanded in North America from 2001 through a series of acquisitions of US and Canadian firms, and HR practices, including PM, were left largely unchanged in these firms, as BrazCem's HQ initially focused on a multitude of other financial and operational matters in North America. However, the company began a process of implementing fully standardised HR practices, including PM, in 2007 based on the policies formulated and operated in the Brazilian HQ. When BrazCem made further acquisitions in North America from 2007, interviewees stated that all newly acquired local firms were required to apply the standardised HR practices almost from the start. In two cases (BrazMan China subsidiary and BrazCem Bolivia subsidiary), the subsidiaries were part of a joint venture involving a local partner with different PM systems; nonetheless, in both cases the Brazilian firm has a majority stake and was able to impose most of the HQ PM systems from the start. In all three cases, we found no evidence at all of ‘reverse diffusion’ of PM systems from host country subsidiaries to the HQ.

Interestingly, although subsidiary-level interviewees talked about the fact that the PM system originated from the corporate level and had little say in its design, interviewees constantly referred to PM policies and procedures as best practices and seldom referred to them as Brazilian management practices. Moreover, they often emphasise the fact that they used ‘well known’, ‘international’ or ‘global’ consultancy firms to perhaps legitimise the use, of what they believe were ‘globally accepted’ practices. It seems that the legitimacy is conferred upon, or attributed to, the PM systems by the fact that they were designed by established consultancy firms, are internationally applied by well-known firms and provide a measure of procedural fairness to employees throughout the MNE. Indeed, interviewees frequently talked about their PM systems as though they are pursuing and conforming to what is expected of them as a successful global firm.

Standardisation versus adaptation and variations of PM systems within the case studies

As stated above, with very few exceptions discussed below, the PM policies of all three companies originated in the Brazilian HQs. Consequently, we set out to establish to what extent each respondent firm's HQ in Brazil either promotes standardisation of HQ-originated PM policies or adaptation of HQ-originated PM policies across the firm's subsidiaries. Based on our systematic analysis of interviewee statements related to standardisation and adaptation, a high level of standardisation of practices emerges. In all three cases, the firm's Brazilian HQ expects all subsidiaries to follow HQ-originated PM policies without any major adaptation (a sample of representative quotations are presented in Table 3).

Each of the case study firms naturally has a unique corporate culture and norms that influenced the operation of the PM system. While BrazCon corporate culture emphasises a relatively unique entrepreneurial approach with each project assessed separately, which is related to the project-based nature of the engineering and construction sector; BrazMan emphasises the importance of affiliate productivity and benchmarks productivity between different subsidiaries, which is related to the nature of the manufacturing sector. Such coercive comparisons are commonly deployed as a form of PM in MNEs (see Edwards, 1998).

Although PM systems are developed at, and generally uniformly applied by, the corporate level, there are some slight variations within the three MNEs in how and when they are implemented. As discussed earlier, some variations within MNEs are a result of practical considerations such as the different subsidiary size and legal issues such as compulsory negotiations with trade unions. For example, BrazMan's compensation system is composed of two parts: salary and annual bonus linked to subsidiary and individual performance (see Figure 1). The Italian subsidiary, however, challenged the link between subsidiary performance and compensation, which led to a protracted negotiation with local trade unions. In BrazCem, the compensation model takes into consideration regional differences in cost of leaving between subsidiary locations in the USA and Canada. Furthermore, while BrazCon uses a standard appraisal process for all levels of employees based on key performance indicators and agreed performance targets, in BrazMan and BrazCem, the performance of shop floor employees, and therefore their compensation, is managed by local managers.

Figure 1 Compensation systems.

Nonetheless, there are some commonalities in terms of the operation of the PM system. In line with the above-discussed origin and philosophy of the PM system, evidence of fashionable global PM practices was present in all three cases. For example, BrazMan used 360 appraisals and BrazCem used a balanced scorecard to manage performance throughout the firm. The frequency of appraisal is normally annual (it can be occasionally more frequent in BrazCon when a person's posting to an engineering project is shorter than 12 months), using the same evaluation forms across the entire company. The corporate PM policy applies to all administrative staff globally (ranging from the vice-president to a secretary).

Drivers of standardisation

Alignment with strategy

All interviewees emphasised the importance of aligning PM to corporate strategy. Interviewees revealed that all important aspects of their firm's activities are encapsulated into a standard set of performance targets and objectives against which subsidiary activities are monitored (quotes 1, 2 and 4, Table 3). In line with the literature on global strategic orientation, given that all three firms produce or provide globally standardised goods or services (i.e. manufactured products for BrazMan, engineering projects for BrazCon and cement for BrazCem), all three firms have a strong preference for globally standardised PM policies that not only facilitate the management of individual employee performance but also facilitate the monitoring of subsidiary performance through benchmarking practices, which enable the quantification of performance. Hence, a primary emphasis in the development of the systems was the professionalisation of PM systems and the desire to be perceived as having best practice PM in place. For example, an HR manager at BrazCon in Portugal argued: ‘I started working with BrazCon in 2005 … I think we improved on being less paternalist and more professional… We have created salary tables that are in line with Hays and the other companies’. As a result, the three firms adopted fashionable global best practices.

Consistency and equity across the multinational firms

In the three cases, PM is used as a strategic HR practice to enable the MNE to evaluate and improve corporate and subsidiary performance against pre-set objectives that are aligned with the MNE strategy. Analysis of interview data suggests that PM is used to evaluate, develop and most importantly to inform the compensation of employees. Consistency of PM practices is propagated as central to equitable compensation and a mechanism through which activities throughout the corporations are successfully aligned with corporate goals and objectives. One interviewee emphasised the importance of international comparability and harmonisation and demonstrated (to us) how the ‘job point matrix’ scheme, for example, enables BrazMan to provide similar reward structure for employees doing similar jobs in different parts of the organisation. In BrazCem, managers started using a global platform such as standardised global grade points, which is used throughout the organisation; this alignment took over three years to complete. Also, interviewees talked about possible uncertainty and confusion and potential inefficiency if different subsidiaries adopt standards different from those at HQs, as well as facilitating mobility within the firms (quote 6, Table 3). Generally, a standard economic measure is used to calculate the economic earnings of each subsidiary, which, as explained below, determine employees' annual bonus. With few exceptions, performance measures are subject to strict reporting requirements.

Corporate culture

Interviewees put a strong emphasis on the importance and existence of a common corporate culture in all of our case firms and the PM system appeared to be central to the diffusion of this across the international operations (quote 3, Table 3; also see under section ‘Mechanisms of standardization’).

It is noteworthy that the push for standardisation of corporate PM practices was not always top down (from the centre to the subsidiary), but in some cases, it came from the subsidiaries. For example, the BrazMan subsidiary in Slovakia initially used a different performance distribution curve to that used in the rest of the MNE. In contrast to the Brazilian HQ and other subsidiaries where a standard bell curve measure of performance was used, the managers at the Slovakian subsidiary classified employees using an 80-20 rule – 80% classified as high performers and 20% as low performers and therefore were not entitled to the annual bonus. Over time, as employees became familiar with the practice in the rest of the MNE, they asked the Brazilian HQ to adopt the corporate performance curve, which further underlined the standardisation pressures within the Brazilian firms.

Drivers of localisation

In the cases where adaptation of PM policies occurred, it was driven, primarily, by regulatory and logistical requirements. Several interviewees referred to ‘cosmetic’ (rather than genuine) adaptation whereby the wording of policies was adapted to account for local contexts (quotes 7 and 8, Table 3) or where the bonuses may be paid at different times of the year. Our interview data pointed to host country legal requirements and subsidiary size as the main drivers of adaptation.

Legal adaptations

Legal adaptations are naturally mandatory when operating in a given jurisdiction. Interviewees highlighted the fact that age-related anti-discrimination legislation in the USA prohibits the consideration of age in performance evaluation, while age may be considered in performance evaluations in Brazil. As another reported example, legal rules related to trade unions are different in North America, where a company may negotiate many different individual agreements with trade unions, whereas trade unions in Brazil are more centralised operating on a sectoral basis, leading to different levels of complexity in labour negotiations. Similarly, the legal rules related to trade unions are different in Mexico where companies bound by a collective agreement with a trade union are compelled to assign specific categories to the job positions of blue-collar workers which may be different to those used in the HQ PM system although there are no legal restrictions for white-collar staff.

Contingency factors

Small-size subsidiaries often lack the logistical ability to replicate all aspects of the PM-related procedures prescribed by the HQs (quotes 11 and 12, Table 3). According to interviewees, being in a smaller subsidiary may, for instance, make it difficult to replicate all training activities of the HQ (e.g. implementing all modules of the same training course at BrazCem) or make it difficult to meet the same initiatives set by the HQ (e.g. a global volunteering scheme at BrazMan). For instance, the North America subsidiary of BrazMan employs just around 60 people, while the Slovakia subsidiary employs 2000 people; hence, the North America subsidiary finds it difficult to replicate all HQ initiatives in the same way as much larger subsidiaries.

National culture

Adaptations of PM systems due to national culture differences were less prominent than legal and logistical/contingency factors. The two main examples of cultural adaptations in subsidiaries were specifically related to enhancing subsidiary performance rather than simply conforming to the local culture. In one instance, the Chinese subsidiary of BrazMan introduced a salary bonus for workers who come to work on time because the lack of punctuality in China was a persistent problem (see Figure 1). In another instance, the North American subsidiary of BrazCem decided to only pay performance bonuses to individual employees if the subsidiary has reached its HQ-set targets, while in Brazil an employee may still receive an individual performance-related bonus even if the Brazilian plant has not achieved its targets – which reflects the more performance-driven culture in the USA – and this is believed to further helped to motivate employees towards better performance (see Figure 1).

Standardisation, adaptation of PM systems and institutional distance

Given the posited importance of institutional distance in the extant literature, we set out to establish to what extent home country and host country institutional environments influence our sample firms' practices.

There is keen understanding by subsidiary employees that their parent company comes from Brazil and recognition of institutional distance (sometimes labelled differently as ‘ways of doing things’ or similar) between the HQ and the subsidiary, which manifests itself, inter alia, in the Brazilian management style and legal differences. However, as indicated earlier, contrary to the ‘dominance’ literature that might suggest that subsidiaries from developed countries would take the lead and engage in forward diffusion of practices to subsidiaries and HQs located in relatively less developed ones, our case studies reveal that subsidiaries located in the USA and Europe were largely passive adopters of HQ practices.

This willingness to adopt HQ practices is related to the fact that Brazilian MNEs are flexible and willing to learn from outsiders and to diffuse PM practices that are more likely to be found in a Western firm than a typical Brazilian one. As an HR manager at BrazCon in Portugal argued:

The company philosophy does not have to change much. But it is important for everybody to be open to things outside the company as well … What we are trying to do in BrazCon is bringing good practices in HR, good practices in engineering, finance or whatever area, from outside, from the market and from the other companies or even from universities.

Indeed, interviewees constantly referred to PM policies and procedures as best practices and did not label the practice as a ‘Brazilian’ PM system. As an HR manager at BrazMan in Mexico noted:

It is easier for us to follow corporate guidelines also because people down here in Mexico use similar tools as in Brazil (…). When I went to Brazil I saw everybody use the same tools, I know them just under a different name because of my background working in other companies (…). The tools and methodologies that would have been incorporated in our headquarter in Brazil, they are the best practices that are used in international companies.

Even at the early stage when one might expect Brazilian MNEs to look towards their subsidiaries in developed countries to provide best practices, this was not the case. However, this may also point to the contradiction that although the PM practices might not have diffused from the subsidiaries to the HQ, they diffused more indirectly from the host to home economy through major international consultancy firms. To put this differently, employees believed that legitimacy was conferred upon the PM system not on the basis of national institutional norms of either the home country or the host country, but rather global norms related to universally accepted corporate practices.

Our interview data strongly suggest that the influence of institutional distance was significantly greater at the early stage of the firms' internationalisation (quotes 9 and 10, Table 3). The most prominent example was the BrazCem subsidiary in North America during 2001–2007, which was already discussed above. Similarly, in the early stage of internationalisation, BrazMan and BrazCon tried to transplant corporate PM practices throughout the firm but faced some initial resistance. For instance, BrazMan faced initial resistance to the adoption of the 360 appraisal approach in its Chinese subsidiary. As a BrazMan interviewee outlined: ‘initially it wasn't easy for them – subsidiaries – to follow rules and structures developed in Brazil but this changed quickly once they understood why we needed to do it’.

We specifically set out to understand the extent that the company was under more pressure to adapt practices in subsidiaries located in high-distance countries compared with low-distance countries, but our interview data did not point to any notable differences that affect the operation of PM systems. Indeed, it is noteworthy that managers sometimes perceive institutional distance between subsidiaries in different countries as an equal, if not a more significant, challenge than institutional distance between Brazil and the subsidiary. Two interviewees at BrazCem noted significant institutional differences between Canada and the USA (countries that can be regarded as having low institutional distance between each other). The director of HR at BrazCem in Brazil said:

I perceive several differences between Brazil and North America. But I also perceive differences between Canada and the United States. For instance, the way they deal with the labour relationship. The US is much more competitive, whereas in Canada there is much more protectionism.

Nonetheless, BrazCem interviewees maintained that these institutional differences do not have any significant influence on the operation of the PM system beyond taking legal differences into account with regard to contract design and taking into account cost of living differences with regard to setting a specific salary. The interviewees in all three companies maintained that institutional differences have ultimately been overcome in the pursuit of a standardised PM system.

Mechanisms of standardisation

Given the central role of standardisation of HQ PM systems, we analysed the mechanisms of standardisation in our sample firms in order to be able to understand how firms are able to align PM systems between the HQ and the subsidiaries.

As mentioned above, a common corporate culture played a key role in disseminating values and policies in all of our case firms. The PM system appeared to be central to the diffusion of this across the international operations, with clearly formulated written elements of the respective firm's values, emphasised by words such as integrity, winning spirit and teamwork for BrazMan and words such as trust, self-development and reinvestment for BrazCon. In all three cases, the Brazilian HQ takes the dissemination of corporate values within the entire organisation very seriously.

In the case of all three MNEs, there was almost no adaptation of the corporate culture and values in the subsidiaries. Only one interviewee in North America mentioned legal adaptations, by noting that the Brazilian HQ was unable to implement the same wording of the code of conduct in North America because of legal restrictions.

In all three MNEs, there are regular communications between HQ and subsidiaries at the level of HR professionals and senior-level executives. For instance, BrazMan conducts joint teleconferences or physical meetings monthly and holds an annual week-long meeting for principal managers from all subsidiaries across the globe. In addition, there are many informal communications between HQ and subsidiaries, largely by email. Similarly, BrazCon and BrazCem have reported high levels of interactions involving HR professionals and senior-level executives. Indeed, there was a higher level of interactions between HR professionals, in contrast to other types of professionals, which points to the strategic importance of HR practices, including PM systems, for the control of subsidiaries by the HQ.

The use of expatriates was highly uneven. The percentage of expatriates among top executives in a subsidiary ranged from 0% to 100%. Of the eight different subsidiaries we interviewed, six have an expatriate as managing director, five of which are Brazilians (see Table 3).

Brazilian MNEs use expatriates strategically when they deem it necessary and they occasionally assign HQ staff to subsidiaries to facilitate the diffusion of values and procedures. For example, BrazMan's current director of HR was previously assigned to Europe for about a year and another senior HR staff was previously assigned to China for about two years, each of whom had the mission to set up the firm's standardised HR practices, including PM procedures, in the respective subsidiaries.

What is common among all three MNEs is that there were a significantly higher number of Brazilian expatriates in subsidiaries when they were newly established, and there are fewer expatriates today. The youngest subsidiary, BrazCon's subsidiary in Guinea (one-year old), has the highest number of expatriates (280 expatriates out of 1300 employees), which is attributed to skills shortages in that country. The two subsidiaries that have 100% share of expatriates among top executives are both newly established (1–2 years old). Older subsidiaries have a much lower expatriate share of top executives, as all three MNEs try to lessen their reliance on expatriates over time. The main reasons cited by our interviewees for their localisation efforts was high cost of expatriate postings and no necessity to use expatriates any longer.

In all three MNEs, there is emphasis on common HR-related training, often using external third-party vendors to deliver training activities to employees. The third parties, including universities and consultancy firms, are often headquartered in the USA and Europe, but there is nonetheless an emphasis on common systems and training content. As the vice-president HR for BrazCem in North America summarised:

What it [Brazilian corporate university programme] allows us to do is to share a common sense of principles, beliefs and values, practices among all our businesses, so that at the end of the day it doesn't matter whether you are working in Brazil, or working in Canada or the US, so the core fundamentals of the business are going to be the same.

All three Brazilian MNEs also had formal training on corporate values, with the purpose of instilling the same Brazil-originated corporate culture and values across all subsidiaries. However, BrazMan actually discontinued such formal training; until the late 1990s, the BrazMan HQ organised workshops for instilling corporate culture in subsidiaries, but this practice was discontinued over 10 years ago, as it was felt that the firm's ethos and values were by then well understood in subsidiaries and formal training was no longer necessary – instead, informal communications, socialisation and occasional expatriate postings continue to be used to instil a common corporate culture. BrazCon and BrazCem continue to regularly use training on corporate values, in addition to the use of regular communications and the use of expatriates. As an alternative to formal training sessions and expensive expatriate assignments, a firm may send subsidiary staff to the Brazilian HQ for a period of time as a way of ensuring diffusion of corporate values and practices to the subsidiary. As an HR manager of BrazMan in Mexico subsidiary reported:

Here there is [sic] a lot of people who travel a lot to Brazil in order to train about the process, in order to understand how BrazMan works in a specific area. A lot of people from production, R&D, IT, people from every area have been in Brazil in order to meet the people, to meet the team and also to learn all the practices, systems and everything.

In summary, the three case-study firms rely on common corporate values, regular interactions, expatriate assignments and common training to varying degrees for aligning PM systems between the HQ and the subsidiaries.

Discussion

This study set out to explore PM policies in EM-MNEs' with a particular focus on Brazilian MNEs. It aimed to shed light on the extent to which EM-MNEs standardise or localise their PM policies, and examine the factors that influence the design and implementation of their PM policies. At a macro level, our study builds on, and extends, research that explores the extent to which there is convergence or divergence of HR practices at a global level. At an organisational or meso level, our study sheds light on debates around how MNEs balance the dual pressures for global standardisation versus local adaptation of management practices. Our particular focus on PM systems is premised on the centrality of PM systems to the coordination and control of foreign subsidiaries of MNEs and the key role it plays in developing employee competence, enhancing performance and distributing rewards. Given the limited research on these debates in the context of EM-MNEs, these questions are particularly apposite.

The first key implication of our findings is that while we do see a strong desire for centralised and standardised PM systems in Brazilian MNEs, with the exception of compensation policies, there is relatively little evidence of a strong home country impact on the PM systems. This is because the practices themselves are not reflective of Brazilian traditions, but rather are premised on Western best practices (we return to this issue below). In all three cases, the firm's Brazilian HQ formulates PM policies centrally and expects all subsidiaries to implement these policies without any major adaptation, while there has been virtually no adaptation of the corporate culture and values in the subsidiaries. Frequent HQ–subsidiary interactions, common training programmes and the occasional strategic use of expatriate assignments aided this standardisation. In all three cases, PM is used as a strategic HR practice to enable the MNE to evaluate and improve corporate and subsidiaries performance against preset objectives that are aligned with the MNE global strategy. This desire for standardisation is reflective of broader trends towards greater global integration in MNEs (Farndale, Scullion, & Sparrow, 2010).

In relation to the origin of the HQ-designed practices, however, interviewees consistently refereed to these systems as best practices and placed a strong emphasis on the role of ‘well known’ and ‘global’ consultancy firms in informing the design of the policies. In addition, it appeared that the corporate interviewees perceived the development of these systems as legitimising their status as successful global firms. Our findings give only partial support to the concept of ‘dominance effects’ in that it is not simply a question of adopting a dominant nation's practices by the HQ but rather a question of the existence of standardised professional practices in a given global issue arena, which directs our attention to institutional change agents such as global consultancy firms and professional associations. This finding draws our attention to the supply side of corporate-level practices within EM-MNEs (Pudelko & Harzing, 2007). It points to an important question around how professional practices are conceptualised and measured in studies on policy diffusion. For example, a quantitative measure that explored where a policy originated rather than what the specific policy was could interpret our finding as a home country effect (it was diffused from the HQ in a standardised way) when in fact it represented the re-exporting of Western practice. This finding fits well with, and extends, recent literature on the role MNEs are playing in diffusing ‘best practices’ globally (Brewster, Wood, & Brookes, 2008; Pudelko & Harzing, 2007). Such global best practices emerge as a reference point for standardised HR practice regardless of their origin (Chung et al., 2014; Pudelko & Harzing, 2007).

We found that Brazilian MNEs tend to ‘re-export’ Western practices rather than diffuse conventional local ones. We believe that the quest for legitimacy and the strong yearning to appear as global MNEs, is what is driving their adoption of legitimised global best practices. We trace this to the fact that for over a decade Western PM practices (e.g. Balance Scorecards) have become fashionable in Brazil (Wood & Caldas, 2002), while indigenous Brazilian management studies have struggled to gain legitimacy (Rodrigues, Gonzalez Duarte, & de Padua Carrieri, 2012). Central to the adoption of these Western practices has been the predominance of American and European text books in Brazilian business schools and the adoption of predominantly American and European texts and theories. This combined with the role of the business media and management gurus (such as outlets like Harvard Business Review, which is published in Portuguese) advocating latest management fads and fashions, have further pressed the adoption of Western practices (Cooke, Macau, & Wood, 2013). Indeed, EM-MNEs have weak firm specific management advantages and often use their internationalisation strategy as a platform to emulate Western management practices (Luo & Tung, 2007). Given the relatively strong arguments for the adoption of the Asian model based on an efficiency logic, the shunning of this model for the Western alternative suggests that adoption is driven by legitimacy rather than efficiency per se. Indeed, given that Brazilian MNEs were relatively later adopters of PM systems, this may not be surprising. The institutional literature argues that later adopters will often be those seeking to obtain legitimacy, regardless of the extent to which the practice is perceived to impact on organisational performance (Tolbert & Zucker, 1983). These pressures for external legitimacy are more likely to emerge as significant for EM-MNEs as they may not be particularly well known or received in host economies (Kostova & Zaheer, 1999). Thus, the search for external legitimacy may be a particulate acute and expected driver of practice adoption. Indeed, further unpacking this question in EM-MNEs would be a very useful research effort.

Our analysis indicates that, in addition to utilising Western consultancy firms to develop management systems, management at the corporate level, albeit implicitly, relied on the ‘Western consultancy label’ to strengthen the credibility, thereby legitimising, of the management practices themselves. This points to the significant role consultancies play in the diffusion of best practice and the importance of normative isomorphism whereby actors such as consultancies reinforce and perpetuate the diffusion of models of best practice (Di Maggio & Powell, 1983; Suddaby & Greenwood, 2001). Corporate-level interviewees' exuberance about ‘global best practices’ was palpable. Interviewees also spoke of how the standard corporate practices helped reinforce the firm's strategy and harmonise activities in geographically dispersed subsidiaries. Appositely, we found little evidence of reverse diffusion of HR practice, as conceptualised by Edwards (1998). Rather, the diffusion from some of the Western hosts to the HQ was indirect through the consultancy firms as opposed to directly through the MNE.

Second, our findings seem at odds with previous research that suggests that EM-MNEs apply different HR practices in developed country subsidiaries and developing country subsidiaries (Khavul et al., 2010). We have specifically put this question to interviewees, but not a single interviewee hinted at such a distinction. Even more significantly, our findings also seem to contradict previous research that pointed towards significant cultural adaptation among Brazilian MNEs. For example, Muritiba, Muritiba, de Albuquerque, Bertoia, and French (2010) analysed Brazilian MNEs at generally an earlier stage of internationalisation, whereas all three Brazilian MNEs in our sample are relatively experienced in international markets. The three Brazilian MNEs in our sample have internationalisation experience of 10–31 years (with an average of 20.5 years), compared with 2–17 years (average 8.5 years) in the study of six Brazilian MNEs by Muritiba et al. (2010). Given that our research findings suggest that cultural differences were most challenging at the initial stages of internationalisation, this points to the conclusion that the more experienced a Brazilian MNE becomes, the less consequential local cultural adaptation of HR practices is.

Third, given that in each case we conducted interviews at subsidiary locations that were institutionally distant and institutionally proximate, we can also conclude that institutional distance did not have a significant influence on the adaptation or otherwise of PM practices at subsidiary level. In part, this finding can be explained by the fact that the practices reflected Western best practice rather than home country practices per se. This finding suggests that institutional distance while a valuable construct in the international business literature, perhaps misses some of the nuances of the reality of policy transfer within MNEs. Specifically, it may not be the institutional distance between the home and host subsidiaries that determines the challenges of standardisation but rather where the practices themselves were developed and how legitimate the practices are perceived to be in the host. Broadly, these findings resonate with a recent study by Brewster et al. (2008, p. 333) who note that

what firms do represents a product of the relative strength of competing forces regulating their behaviour – formal laws, informal norms and practices, ownership structures, and relations with stakeholders …what firms do represents not just a product of context, but rather trade-offs and compromises between competing pressures and influences.

It is plausible that as these EM-MNEs increase their experience in managing global operations, they learn how to establish relationships with local stakeholders and adopt global best practices that the legitimacy of HQ approaches to the management of subsidiary issues (as evidenced through PM in our study) become more accepted as legitimate and the relative influence ease with which HQ can diffuse corporate initiatives increases.

Another possible explanation for the lack of engagement with local norms could be that the Brazilian MNEs that we examined had in effect delegitimised the local Brazilian model in the home country by implementing Western practices. Therefore, it is not surprising that the MNEs might not facilitate the adaptation to local indigenous norms, given they had delegitimised such practices in the home operation by divorcing their PM systems from Brazilian norms.

Where variations of PM did exist across different subsidiaries, these adaptations of HR practices can be largely divided into two main types: legal and subsidiary size factors. However, they are relatively minor in all three cases, while the entire PM system remains highly standardised across all subsidiaries. Indeed, it is notable that small subsidiary size was considered a greater impediment to the implementation of central PM practices than cultural factors.

In line with the above argument, what our study does suggest is that the PM system, alongside other HR practices, has evolved over time in Brazilian MNEs. In the early stage of internationalisation, there was a considerably higher degree of local adaptation in all three cases, as a result of cultural factors, initial technical challenges and the fresh acquisition of smaller local foreign firms with different HR practices. However, within a space of no more than 6–8 years, all three Brazilian MNEs were able to firmly establish standardised PM practices in their respective subsidiaries based on the policies formulated and operated in the Brazilian HQ.

From an institutional theory perspective, it seemed clear that the practices examined in this study were relatively insulated from cultural and institutional baggage in the host and home country institutional environments. This finding draws our attention to early management and organisation behaviour scholarship that already recognised that corporate practices in MNEs can be insensitive to institutional-cultural environments and would lead to a convergence of professional practices (Kerr, Dunlop, Harbison, & Myers, 1960; Hickson, Hinings, McMillan, & Schwitter, 1974). In this context, our findings do not suggest that institutions are unimportant but they rather emphasise the key role of normative isomorphism in adopting standardised global HR practices among EM-MNEs. Normative isomorphism can explain why the HQ of BrazMan and BrazCem began moving towards what they considered more ‘modern’ and ‘legitimate’ HR systems by the early 2000s and why there is little evidence of resistance to standardised HR practices at subsidiary level. We suspect that the enthusiastic HQ and subsidiary-level support, or at least absence of explicit resistance, stems from the adopted practices compelling normative (global best practices) and rational (global comparability and equity) normative isomorphism logics.

These findings are in line with emerging institutional theory applications from very disparate fields such as accounting (Brandau, Endenich, Trapp, & Hoffjan, 2013; Rodrigues & Craig, 2007) and environmental management (Levy & Kolk, 2002; Zelli & van Asselt, 2013), which have recently provided evidence that managerial practices are increasingly converging globally towards international – particularly Anglo-American – professional standards in very specific ‘global issue domains’, particularly as a result of normative isomorphic pressures. This institutional scholarship directs our attention to the importance of increasingly highly specialist and complex professional standards within a given global issue domain in the context of the increasing fragmentation of specialist professional fields of knowledge globally. By extension, HRM scholars would be advised to move beyond the current focus on home country or host country institutions – and the related mimetic isomorphism logics – in explaining the adoption of standardised HR practices by MNEs, towards a focus on the development of global issue domains and the cognitive aspects of professional standards – and the related normative isomorphism logics.

Limitations, future directions and conclusion

Although this exploratory study adds to our understanding of PM in EM-MNEs in general and Brazilian MNEs in particular, it has several limitations. First, as with all case study research, the results of this study have to be interpreted with caution. An important limitation lies in the sample size and type of MNEs studied. Our primary data are from three firms that raise questions of generalisability. Although our sample covers firms with different levels of international experience (see above), they are all relatively mature MNEs, and therefore our results may not be generalisable to newly internationalised firms. Moreover, the three firms in our sample are all large firms; hence, our results may not be generalised to small- and medium-sized MNEs. Furthermore, we purposely focused on PM systems. While this exclusive focus helped us gain a deeper understanding of PM policies within EM-MNEs, it restricts the interpretation of our results. Caution is warranted in the generalisation of our results to other HR policies. These limitations point to opportunities for future research, which could fruitfully examine HRM policies within young and small EM-MNEs.

Second, our results show that Brazilian MNEs exhibit a high degree of standardisation of centralised PM practices across all their global subsidiaries and exhibit a considerable desire for global integration. However, a closer look at these practices revealed that at the corporate level Brazilian MNEs do not use, and therefore do not diffuse, traditional Brazilian management practices; rather they use global Western best practices and re-export them back to their subsidiaries in both developed and emerging economies. These results underscore the importance of examining the supply side of HQ policies rather than just, as has been the case in previous studies, looking at the magnitude of adaptation by subsidiary level. More generally, we hope that our findings help researchers on HRM in EM-MNEs refine the notion of what is meant by home country practices.

Third, all our interviewees were executives or HRM managers involved in, and most of them were responsible for, the implementation of corporate-level HR practices. Future studies involving both top management and lower level managers and employees at the receiving end of PM practices may provide a fuller picture and deeper understanding of the dynamics involved in the diffusion of corporate practices to subsidiaries located overseas. Also, while our study pointed to the evolution of PM practices over time as a result of internationalisation stages, it was carried out at a single point in time. Given the evolutionary nature of PM practices in EM-MNEs, future longitudinal studies would provide useful insights into how HR practices evolve over time. The current study also did not analyse the effectiveness of PM policies in terms of internalisation of the policies and or their implication on organisational performance. Future research could examine the link between EM-MNE PM policies and organisational performance.

Acknowledgement

The authors are very grateful to Camila Yamahaki and Suzana Rodrigues for their advice and their kind assistance in obtaining access to the HRM departments of the Brazilian case study firms. The authors are also very grateful to the editor and two anonymous reviewers for their helpful comments, which have significantly helped to improve this article.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

^1. Email: [email protected]

^2. Email: [email protected]

^3. For instance, in 2009, the 30 largest Brazilian MNEs accounted for over USD 61 billion in foreign sales and employed about 179,000 employees abroad (Lima, 2011). In 2010, FDI by Brazilian MNEs reached over USD 15.6 billion (Lima, 2011).