HRM and Performance

Achievements and Challenges

David E. Guest, Jaap Paauwe, Patrick M. Wright, David E Guest, Jaap Paauwe, Patrick Wright, David E. Guest, Patrick M. Wright, David E Guest, Jaap Paauwe, Patrick Wright

• Publisher: Wiley
• Year: 2012
• Print ISBN: 9781405168335
• eBook ISBN: 9781118482636
• Edition: 1
• Topic: Business
• Subtopic: Human Resource Management
• Index: Business

The link between HRM and performance has become an important policy issue at both a national and a corporate level.

HRM and Performance draws on the knowledge and expertise of a number of leading international scholars in the field of HRM to provide a comprehensive overview of the current state of HRM and identify fruitful directions for theory, research and practice. A central question throughout is - what's next for HRM and what are the keys to the future of managing people and performance?

CHAPTER 1

HRM AND PERFORMANCE: WHAT DO WE KNOW AND WHERE SHOULD WE GO?

JAAP PAAUWE, PATRICK, WRIGHT AND DAVID GUEST

‘Based on four national surveys and observations on more than 2000 firms, our judgment is that the effect of a one standard deviation change in the HR system is 10–20% of a firm’s market value.’ (Huselid & Becker, 2000, p. 851; emphasis added)

‘The existing evidence for a relationship between HRM and performance should be treated with caution.’ (Wall & Wood, 2005, p. 454)

‘After hundreds of research studies we are still in no position to assert with any confidence that good HRM has an impact on organization performance.’ (Guest, 2011, p. xx)

Practitioners interested in human resources management (HRM) have long sought to convince others of its value. Drucker (1954) referred to ‘personnel’ managers as constantly worrying about ‘their inability to prove that they are making a contribution to the enterprise’ (p. 275). More recently Tom Stewart described HR leaders as being ‘unable to describe their contribution to value added except in trendy, unquantifiable and wannabe terms’ (Stewart, 1996, p. 105).

In response to this longstanding and often repeated criticism that HR does not add value to organizations, academic research has exploded over the past 20 years, seeking to show that HRM practices are related to firm performance. Huselid’s (1995) groundbreaking study showed that a set of HR practices he called ‘high-performance work systems’ (HPWSs) were related to turnover, accounting profits and firm market value. This study served as the springboard for a significant body of research confirming empirical relationships between HR and performance.

However, in spite of the vast body of research that has emerged over the past two decades, as these quotes from some distinguished academic researchers suggest, divergence exists regarding what we can conclude about the relationship between HRM practices and firm performance. Is the HRM–performance relationship one that is strong, universal and causal, or is it potentially weak, contingent and even spurious? More importantly, what is the underlying research base from which we can answer that question, and how can we improve that base in order to answer it in a way that is valid, reliable and practically important? The purpose of this book is to attempt to bring together some of the leading researchers in this area to provide insights into what we know, what we need to know, and how we can begin on a journey to improve our knowledge of the relationship between HR and firm performance.

In this chapter, we will first present an overview of what we seem to know about this relationship. We will trace some of the streams of this research in an effort to provide a foundation for how we have arrived at this point in our knowledge base. We will then lay out some of the unanswered questions that have emerged from the research to date on the relationship between HRM and performance. Within this context we will show how the authors have attempted to provide some answers and set out future directions as an overview to the rest of the book.

What do we know about the HRM–performance relationship?

The development of theory and research on the relationship between HRM and performance began in the 1980s. A series of articles and books by authors such as Fombrun et al. (1984) and Miles and Snow (1984) began to link business strategy to human resource management. The Harvard group (Beer et al., 1984) and Schuler and Jackson (1987) began to argue for a clear and systematic integration between the strategy of the firm and the HRM practices used to manage the workforce of that firm. In the UK, writers such as Guest (1987) and Storey (1992) took a normative perspective, suggesting the need for external and internal fit of HRM, contrasting it with the pluralism of an industrial relations perspective.

During this time, others such as Walton (1985), while less concerned with fitting HRM to strategy, highlighted the need for a shift from control to commitment as the basis for management of people at work. In parallel with this, authors such as Foulkes (1980) and Peters and Waterman (1982) provided glimpses of evidence about successful organizations that seemed to apply the ‘high commitment’ HRM principles. Thus, this early phase presented the foundational arguments that (a) HRM practices should be integrated with the strategy of the firm in order to be maximally effective, and (b) that certain ‘high commitment’ HRM practices were more effective for generating higher firm performance relative to control-oriented practices.

While these foundational arguments sparked thinking, they alone could not provide convincing evidence of the potential value of HRM. However, the 1990s served as the springboard for what would end up being a vast and growing empirical literature. Huselid (1995) provided the seminal work in his study of more than 800 corporations, revealing an empirical relationship between the HPWSs (similar to the ‘high commitment’ practices discussed above) and important corporate performance variables such as the gross rate of return on assets (a measure of accounting profits) and Tobin’s Q (a measure of the value of the firm). This study has become the central node in research on the HRM–performance relationship, but was by no means alone. Arthur (1994) found an empirical link between strategy of steel mini-mills and HRM practices. Ichniowski et al. (1995) found a relationship between HRM practices and operating performance of steel mini-mill manufacturing lines. MacDuffie (1995) presented evidence of ‘bundles’ of HRM practices and measures of manufacturing performance within the automobile industry. Finally, Delery and Doty (1996) explored universalistic, contingency and configurational models of HRM as predictors of firm performance among a sample of banks. They found support for the universalistic (i.e. a similar set of practices consistently related to performance) but little support for either contingency or configurational approaches. In Europe, and particularly the UK, a group of authors reacted strongly against this emerging stream of research, arguing that it represented a new and more subtle form of exploitation of workers (Keenoy, 1990; Keenoy & Schwan, 1990; Blyton & Turnbull, 1992; Legge, 1995). Their analysis was conceptual rather than empirical and partly as a result, received only limited attention among researchers (Keegan & Boselie, 2006). Nevertheless, they raised the important question, largely neglected in the early research on HRM and performance, about the impact of HRM on employees.

Since these early studies, the empirical research has continued unabated and expanded globally. For instance, Guthrie (2001) replicated Huselid’s methodology in a sample of New Zealand firms and found a similar relationship between HPWSs and firm performance. Guest et al. (2003) related HR practices to both past and subsequent objective productivity and profitability data, as well as current subjective productivity and financial performance estimates, among a sample of 366 companies in the UK. Boselie et al. (2003) explored the role of sectoral/institutional factors in The Netherlands and showed that the effect of HRM on performance is lower in highly institutionalized sectors (like hospitals and local government) than in a less institutionalized sector like hotels. More recently Takeuchi et al. (2007) found a relationship between HPWSs and firm performance among a sample of Japanese firms. There is now a growing body of research on Chinese organizations that shows a similar pattern of results (e.g. Liao et al., 2009).

There is little doubt that in the past 20 years some progress has been made in the analysis of the relationship between HRM and performance. On balance, however, progress has been modest. This is reflected in the rather optimistic conclusions from some of the main overview articles that have appeared. Becker and Gerhart (1996), indicated that the conceptual and empirical work had progressed far enough to suggest that the role of human resources can be crucial. Similarly, Paauwe and Richardson (1997), based on an early review of 22 studies, concluded that HRM activities give rise to HRM outcomes that will influence the performance of the firm. More specifically and more positively, Huselid and Becker (2000) indicated that the effect of one standard deviation change in the HR system leads to a 10–20 per cent increase in a firm’s market value.

However, on a more cautious note, Wright and Gardner (2003a), reflecting on the available evidence, concluded that HR practices are at least weakly related to firm performance. By 2005, Boselie et al. (2005), drawing on a comprehensive sample of 104 studies, concluded that much (though by no means all) of the empirical research shows that HRM is associated with organizational performance. Based on a selection of 25 mainly American so-called high-quality studies, Wall and Wood (2005) concluded – even more cautiously – that the evidence for an effect of HRM on performance is promising but only circumstantial due, for the most part, to inadequate research design. Thus, 19 of the 25 studies they examined reported statistically significant positive relationships between HR practices and performance, but the effect sizes are typically small and the majority of studies failed to consider whether it is the HRM system (the ‘gestalt’) generating the effects or just specific component/individual practices. Overall, therefore, they conclude that ‘The existing evidence for a relationship between HRM and performance should be treated with caution’ (Wall & Wood, 2005, p. 454).

On a more positive note, Combs et al. (2006) carried out a meta-analysis of 92 studies on the HR–firm performance relationship and found that an increase of one standard deviation in the use of high-performance work practices (HPWP) is associated with a 4.6 per cent increase in return on assets, and with a 4.4 percentage point decrease in turnover. Hence their conclusion that ‘HPWPs’ impact on organizational performance is not only statistically significant, but managerially relevant’ (p. 518).

Whether examining individual studies, the systematic reviews or the meta-analytic summaries of this literature, what we do know is that HRM practices seem to be consistently related to performance. However, the reviews reveal different levels of confidence about the strength of the association, about the quality of the research on which it is based and about the practical conclusions we can draw from it about the impact of HRM. Therefore, before concluding that we have found the truth and there is no more need for research in this area, we may first want to look at it with a more critical eye. Such a critical view reveals that while this empirical body of research has consistently demonstrated a relationship between HRM practices and performance, it has revealed a number of problems. These centre around two basic themes: theoretical ambiguity and empirical invalidity.

Theoretical ambiguity

Regarding the theoretical ambiguity, Guest (1997) neatly summarized the challenge when he stated this literature needed ‘a theory of HRM, a theory of performance, and a theory of how the two are linked.’

First, in attempts to articulate a theory of HRM, Guest (1997), along with others such as Becker et al. (1997) utilized expectancy theory (Vroom, 1964) to theorize about the core HRM practices. This basic framework was adopted and adapted to what has emerged as the AMO (ability, motivation, and opportunity) framework (Appelbaum et al., 2000; Purcell & Hutchinson, 2007). This framework suggests that HR practices can influence the skills, competencies and abilities of the workforce to provide a strong human capital base. Second, practices can affect the motivation and commitment of employees, engaging them to want to behave in ways that benefit the firm. Finally, HR practices such as job design and participative processes can provide the opportunities for the skilled and motivated workforce to positively affect organizational outcomes. This has become the most well-accepted framework for understanding HRM practices. However, while the framework has been widely accepted, there is no consensus about the specific practices that should be considered. This is a major problem. After all, how can we ever make progress in this field if we do not agree on what constitutes one of the main independent variables, namely HR practices?

Regarding a theory of performance, since Huselid’s (1995) study, most of the research has focused on performance from the standpoint of the firm. For instance, in their meta-analysis on HPWSs and performance, Combs et al. (2006) noted that of the 92 studies they included, ‘[a]ccounting returns were most frequently studied (35 effects), followed by productivity (32), retention (23), multidimensional (22), growth (16), and market returns (8)’ (p. 510).

However, a somewhat separate line of research on HRM has explored employee-centred outcomes, assuming a need to look beyond firm performance. For instance, Ramsay et al. (2000) suggested that HRM might not be in the workers’ best interests, and some studies critical of HRM based on workers responses supported this (e.g. Cappelli & Neumark, 2001; Godard, 2004).

Recently, a more nuanced view is emerging, suggesting that worker outcomes can be considered a central issue, and that it is possible that HRM can lead to both enhanced workers’ well-being and higher performance (Peccei, 2004; Guest, 2011). For instance, Kehoe and Wright (2010) found that HRM practices were related to employees’ affective commitment to the firm, their willingness to exhibit discretionary behaviour, and their intent to stay with the firm. In a more complex analysis, Gardner et al. (2011) found that motivation and opportunity enhancing practices were positively related to affective commitment and negatively related to turnover. However, ability-enhancing practices, consistent with labour market theory, were unrelated to affective commitment and positively related to turnover. Thus, theory and research regarding what constitutes the various aspects of performance affected by HRM has expanded beyond the pure financial measures. This view is further reinforced by research in public sector organizations where conventional private sector financial indicators are not relevant (see, for example, Messersmith et al., 2011). This supports the view that the outcomes of HRM are multi-faceted and that we need to look beyond financial performance (Boxall & Purcell, 2003; Paauwe, 2004).

Finally, regarding the theory linking HRM and performance, Wright and McMahan (1992) presented six theories that had been used in exploring the relationship between the two. However, by far the most popular theory in the 1990s was the resource-based view of the firm, which argues that when a firm’s resources are valuable, rare, inimitable, and non-substitutable, they can be a source of sustainable competitive advantage. Wright et al. (1994) explored how a firm’s human resources could meet these criteria and thus constitute a potential source of sustainable competitive advantage. While popular, this theory provides only a very generic argument for how HR practices might be related to performance, and does not help to understand the specific nature of this relationship. One of the challenges, therefore, is how far it can be adapted to offer more specific proposals.

Becker et al.’s (1997) box and arrows model presented a more specific theoretical framework for explaining the mediating mechanisms between HRM and performance. They suggested that HRM practices have their most direct impact on the employee skills and motivation, which, in turn, results in creativity, productivity, and discretionary behaviour. Employees’ behaviour influences the firm’s operating performance, which leads to profitability, growth, and market value.

The review by Boselie et al. (2005), concluded that up until then, the three most popular theories being used in HRM and performance research were contingency theory, the resource-based view (RBV) and the AMO framework. If we include papers published since 2000, more than half use AMO theory (Paauwe, 2009). Both RBV and contingency theory focus on the organizational level, whereas AMO theory focuses on the importance of taking into account employee-level factors. Overall we can discern a lack of attention being paid to the institutional context (Paauwe & Boselie, 2003).

Thus a number of attempts have been made to develop a theory of HRM, a theory of performance, and a theory of how the two are linked. However, all are ‘works in progress’ and none have achieved consensus support among researchers. There is therefore still a need for more theoretical and conceptual development.

Empirical invalidity

Even if there was consensus about theory, as the opening quotes illustrate, a number of researchers have taken issue with the validity of the research base exploring the HRM–performance relationship. Gerhart et al. (2000b) first suggested that the single respondent measures of HR practices lacked reliability. They noted that while low reliability could diminish the observed relationship between HR and performance, it could also inflate it, if much of the error variance was systematically biased. They suggested that the traditional point estimates of the dollar value gains associated with increasing HR practices by a standard deviation, if corrected, would be far beyond what seems reasonable, and thus suggested that some systematic bias might exist.

Later researchers investigated some of the potential forms of systematic bias. Guest et al. (2003) were the first researchers to note that while the assumed direction of causality was from HR practices to performance, this should not be taken for granted. Their own data suggested that performance causing HR practices was an equally plausible explanation of the association. Wright et al. (2005) followed up on this research and showed that their HR practice measures were as strongly related to past performance as to future performance. In addition, they found that when controlling for past performance, the relationship between HR practices and future performance was substantially reduced. Neither set of researchers suggested that HR had NO impact on later performance, but raised the question about how much of the observed relationship was directly causal. Wall and Wood (2005) argue that the lack of rigorous longitudinal studies on the HRM–performance relationship precludes any firm conclusions regarding the causal nature of this relationship. Some more recent studies have taken careful account of causality, using longitudinal research (van Veldhoven, 2005; van de Voorde et al., 2010b) and they also indicate reverse causation, that is, that performance influences subsequent HR practices and well-being.

Another issue that may call into question the strength of the HRM–performance relationship deals with response patterns of the people filling out surveys. Implicit performance theory, rooted in the leadership literature, has shown that respondents observing the same leader will report leadership behaviours consistent with what they are led to believe was the leader’s group’s performance (Rush et al., 1977). Gardner and Wright (2010) presented executives and students with descriptions of high and low-performing organizations and found that respondents reported greater use of HR practices in the high-performing firms. This opens up the possibility that in addition to actual performance causing HR practices, there might be an effect of performance on respondents’ reports of HR practices.

The result of all this led Paauwe (2009) in his overview to conclude that in spit...

Table of contents

• Cover
• Contents
• Title
• Copyright
• Preface
• Editor biographies
• Author biographies
• Chapter 1: HRM and Performance: What Do We Know and Where Should We Go?
• Chapter 2: HRM, Well-Being and Performance: A Theoretical and Empirical Review
• Chapter 3: Building Highly-Performing Work Systems: Analysing HR Systems and Their Contribution to Performance
• Chapter 4: Reconceptualizing Fit in Strategic Human Resource Management: ‘Lost in Translation?’
• Chapter 5: HRM and Performance: The Role of Effective Implementation
• Chapter 6: Strategic HRM and Organizational Behaviour: Integrating Multiple Levels of Analysis
• Chapter 7: Measuring Human Capital: A Strategic Human Resource Management Perspective
• Chapter 8: Measurement of Human Resource Practices: Issues Regarding Scale, Scope, Source and Substantive Content
• Chapter 9: Research on Human Resources and Effectiveness: Some Methodological Challenges
• Chapter 10: Progress and Prospects for HRM–Performance Research in Small and Medium-Sized Businesses
• Chapter 11: Progress and Prospects
• References
• Index