Human resource policies and company performance: a quantitative approach using longitudinal data

From an outside perspective, human resource management issues are difficult to decipher. Despite a mushrooming of books and significant theoretical developments, it seems that the notion of HRM is still an umbrella term associated with different and contingent practices (Blyton and Turnbull 1992). In essence, rather than a unified theory, HRM appears to be divided by inherent contradictions such as individualism/cooperation, commitment/flexibility, strong culture/adaptability (Legge 1989). This lack of logical consistency raises doubts about the plausibility of HRM models, the efficiency of which, moreover, strongly depends on their organizational context. This means that outside actors have difficulties in collecting and understanding information about them. It is a problem, in the first place, for investors and bankers and is rarely considered in recent research. In fact, if the human resource system can really make a difference - which is the fundamental belief behind the HRM model (Storey 1995: 5) - investors should be given the means to understand and evaluate HRM company decisions. If not, they could be tempted to ignore or distort HRM issues and instead allow themselves to be driven by economic pressures and short-term profits. Should such a situation become reality (Porter 1992), it could prove detrimental to the human aspects of the company's strategy and would probably threaten their future and that of the economy as a whole. To correct this deficiency, it seems important to undertake research which could help in gaining a better understanding of these analytical and informational problems.

The aim of this research is to discover the outside usefulness of the French Company Personnel Report,(1) in other words, its ability to help investors improve their evaluation of future economic performance by gaining a better understanding of company HRM models. It presents an empirical longitudinal study based on 61 large French companies during the period 1982-1989. The results indicate immediate and lagged correlations between profitability or productivity ratios and large HRM indicators.

Research Model and Methodology

There has been an extensive literature on the links between HRM practices and company performance. In all, it may be seen that these specialized studies still leave many questions unanswered (Kleiner et al. 1987; Becker and Gerhart 1996). Useful light can be shed by taking a holistic approach to human resource policy management. In this HR system perspective, the implicit assumption is that single or isolated HR policies may have only limited competitive effects. Some of the most prominent of these global studies can be cited as follows: Denison (1984), Stebbins (1987), Gomez-Mejia (1988), Hansen and Wernerfelt (1989), Albert (1990), Huselid (1995). In France, research by Allouche (1984) and Reyes Garcia (1985) also follows these lines. Their results suggest that it is possible to identify a generally positive relationship between economic performance and employee-centred management in involving and developing the human resources of the company. This relationship is not, however, systematic (Ondrack 1988; Abowd et al. 1990) and the hope that HRM can easily be linked to economic performance still seems illusory (Whipp 1992: 52).

In this paper, by following such holistic approach, we seek to link human resource management to company performance. From a theoretical point of view, this kind of research derives from the seminal Harvard framework (Beer et al. 1984), which first proposes a prescriptive guideline for human resource management. Within constraining situational factors, this framework displays the following process vision of HRM: HRM policy choices [right arrow] HR outcomes [right arrow] long-term consequences. From an outsider's perspective, some HR outcomes (commitment, competence, adaptability...) are particularly difficult to assess. Thus, this paper will overlook these intermediate outcomes and limit the focus to the possible effects and long-term consequences of HRM policy choices. Since the focus is also on investors' points of view, long-term consequences will be reduced to economic performance.

Hypotheses

We can identify at least five major fields for HRM policy choices: wages, training, employment, work systems(2) and industrial relations. However, because available data on the two latter fields are rare in the French Company Personnel Report, we had to exclude them from our research. Focusing on the first three fields, this paper will also test the possible economic effects of what we usually call the 'social climate'.

Considering the first field, that of wages, it is true that we still do not know whether this has a positive and incentive effect on individual or group performance. As far as managers are concerned, a significant correlation is often evident between the level of remuneration (salary + bonus) and either the size of the company (Kostiuk 1990), or the accounting or Stock Exchange performance.(3) The scale of incentives generated by remuneration cannot, however, be explained by these correlations alone. Although high wages and bonuses may indeed improve performance, they can also be the fruit of improved performance or the result of signalling effects (Murphy 1986). Neither does research devoted to employees and general executives reveal any apparent causal relationship between high salaries and firm-level performance. On the contrary, this causal relationship is contradicted by capital market reactions (Abowd 1989). However, it seems that high or increasing wages stimulate intermediate performance and outcomes, such as productivity or employee involvement. Studies converge on this point, despite the diversity of the populations examined (Asch 1990; Ehrenberg and Bognanno 1990; Groshen and Krueger 1990; Holzer 1990). In the light of these latter results, hypothesis H1 can be formulated, as follows: