Vous consultezThe Stock Market and Human Resource Management: Evidence from a Survey of French Establishments

AuteursCorinne Perraudin[*] [*] Centre d’Etudes de l’Emploi (CEE) and CES (CNRS, Université...
suitedu même auteur

1 - Introduction

Corporate governance has been the object of growing attention since the early 1990s. Of primary interest are the determinants and consequences of equity ownership structure, the legal form of the firm, and the composition of the board of directors. Numerous studies have examined the role labour plays, or might play, in shaping business conduct and strategic choice (Blair, 1995; Blair and Roe, 1999; Roe, 2002; Gourevitch and Shinn, 2005). More recently, some studies have explored the influence of corporate governance, and more specifically of stock markets, on human resource management (‘HRM’) practices (see, for example, Gospel and Pendleton, 2005 or Konzelmann, Conway, Trenberth and Wilkinson, 2006). Corporate governance determines the nature of the relations between the main stakeholders in the firm (shareholders, directors, executives and employees). As such, it plays a decisive part in fixing the firm’s objectives and orientating the way it is run. If HRM is considered to be one of the most important strategic variables for companies, then there are good reasons to ask what contribution it makes to achieving the objectives set by the governance.

2 In continental Europe and Japan, changes in stock market law and (to a lesser extent) in corporate law, together with the growing liquidity of financial markets, have increased the sensitivity of managers to the interests of minority shareholders and stock market price (Fanto, 1998; Cioffi and Cohen, 2000; Aoki, 2007). This has led some authors to diagnose a shift of the continental European or Japanese systems of governance towards the American or British models (Hansmann and Kraakman, 2001). This raises the question of the consequences of this trend on HRM (Jackson, 2005).

3 At present, research into the relation between corporate governance and employment follows three main paths.

4 The ‘Varieties of Capitalism’ (VoC) approach, macro-economic and macro-legal, stresses the institutional complementarities that are likely to form on a national level between the stock market and the labour market (Hall and Soskice, 2001; Amable, Ernst and Palombarini, 2005; Deakin and Ahlering, 2005; Barker and Rueda, 2007; Black, Gospel and Pendleton, 2007). The key idea is that a financial system that favours liquidity, as in the United States or Great Britain, may limit the possibilities for employee commitment and cooperation within enterprises, but facilitate the reorganisation of activities. On the basis of cross-sectional and comparative data for OECD countries, Black et al. (2007) observe the negative influence of stock market activity on employment stability; on the other hand, they find no effect on employee training, widely studied in this literature. The VoC approach analyses a first level of interaction between the spheres of finance and employment. But, by definition, it does not allow any sub-national distinctions between firms, according to whether or not they are listed on the stock market or according to the distribution of equity ownership among different types of shareholders.

5 The second line of research examines changes in employment in firms that have become sensitized to the stock market, through qualitative case-studies. Jackson, Höpner and Kurdelbusch (2005) analyse HRM in large listed companies in Germany, which are becoming increasingly concerned with their financial profitability and stock market valuations. Froud, Haslam, Johal and Williams (2000a) and Deakin, Hobbs, Konzelmann and Wilkinson (2006) study British listed companies. These works bring to light a style of HRM associated with the requirements of minority shareholders, a style that cannot be found either in non-listed companies or in listed companies that are insulated, because of their shareholding structure, from the influence of the stock market. Such case-studies approach would be interestingly complemented by statistical or econometric analyses.

6 The third path of research uses data on business enterprises to investigate the consequences of mergers and acquisitions – and therefore of changes affecting equity ownership – on variations in the workforce. This micro-econometric literature has been revived by the use of linked employer-employee data (LEED; see Bryson, Forth and Barber, 2006). The results are mixed: for example, while Lehto and Böckerman (2006) observe that changes in ownership lead overall to reductions in the workforce, Margolis (2006) notes that the workers laid off have profiles that enable them to find new jobs relatively quickly. Goergen, Brewster and Wood (2006) offer an original contribution: on the basis of enterprise-level data from different countries, they analyse the influence of mergers and acquisitions on the way enterprises adjust their workforce (freezing recruitment, early retirement, lay-offs, etc.). The main limitation of this line of research, for our purpose, is that it focuses on one particular aspect of HRM (changes in the workforce) and on one particular moment, when the equity ownership is restructured.

7 Our article differs from and complements the approaches described above by proposing an econometric study based on establishment-level data, through which it is possible to identify a large set of HRM practices, over and above workforce adjustments. The objective is to study the influence of stock market listing and of the identity of the main category of shareholders on different HRM practices[ 1] [ 1] Abe and Hoshi (2007) propose a similar approach for Japan. ...
suite. We use a French linked employer-employee database, that merges together two sources: the 2004-2005 REPONSE survey, carried out by the Research and Statistics Department of the French Ministry of Labour (DARES) and the Annual Declarations of Social Data (DADS) collected by the National Institute for Statistics and Economic Studies (INSEE). The REPONSE survey is based on a sample of 2,930 French establishments with 20 workers or more, representative of the French productive sector, excluding the agricultural and the public sectors. The information supplied on HRM practices, work organisation and industrial relations is very rich. If the data pertain mainly to the establishment level, there is also information available on stock market listing and the distribution of the company’s equity capital. The DADS are an administrative source that provides information about the characteristics of the workforce and the amount of wages paid in the establishments. We end up with a large quantity of information on the competitive environment of these establishments, the distribution of equity ownership and their HRM practices for the year 2004.

8 Our findings confirm the importance of corporate governance as a determinant of HRM practices. More specifically, we observe that stock market listing is associated with a particular profile of labour management. Relatively favourable working conditions (in terms of wages and training) are given to a limited core of employees. Wage bill flexibility is obtained by the extensive use of individual and collective bonuses rather than changes in the number of salaried staff. In parallel, the massive use of commercial contracts (temp agencies and subcontracting) constitutes a complementary form of cost ‘flexibilisation’ that characterises these establishments. We argue that this profile of HRM is consistent with the profitability requirements of the stock market. Another important result is that the identity of shareholders matters yet in a marginal way: the penetration of equity capital by non-resident institutional investors (mainly Anglo-Saxon investment funds) tends to reinforce the specificity of wage policy associated with stock market listing.

9 The article is organised as follows. The second section presents the main characteristics of the French model of corporate governance and examines recent developments in this model, reflecting the growing importance of minority shareholders and stock market. The theoretical links between stock market and HRM are discussed in the third section. The fourth section describes the empirical analysis. The fifth section presents the main findings. The sixth section concludes.

2 - The dynamic of the French model of governance

10 The comparative literature on corporate governance traditionally contrasts continental Europe with the United States and Great Britain (see, for example, Prowse, 1995; Barca and Becht, 2001). The role played by stock markets in corporate control is the main distinctive factor.

11 In the Anglo-Saxon countries, companies are more frequently listed and the stock markets are more active, in terms of volumes transacted. The prime movers in these markets are institutional investors, especially pension funds and mutual funds, which own about half of the shares listed in the United States, the other half being owned by households. If we include insurance companies, the holdings of institutional investors in Great Britain represent more than 70%. Although institutional investors manage large volumes of assets, they diversify their portfolios and only possess a small proportion of shares in any given company: they are, most of the time, minority shareholders. Stock market prices, which in theory capture expectations of dividends and determine the possibilities of making capital gains, are a priori the best indicator of the interests of these investors.

12 The dispersion of equity ownership resulting from portfolio diversification makes direct control of the managers by the stockholders difficult, or even impossible (Berle and Means, 1932). There are, however, a certain number of devices that encourage the managers to behave in line with the interests of the minority shareholders. Some of these devices are of a legal nature. Shareholder activism, involving the submission of proposals at annual shareholder meetings, is common practice among institutional investors in the United States and also, to a lesser extent, in Great Britain (Black, 1998). Likewise, legal actions against managers or directors for the breach of fiduciary duties are frequent in the United States (Hertig and Kanda, 2004). But it is above all those mechanisms that sensitize managers to the stock market price which guarantee the ‘prioritization’ of shareholders’ interests in business conduct: in Great Britain, a flexible regime for hostile takeover bids favours stock market discipline (Deakin, Hobbs, Nash and Slinger, 2003), while in the United States, this role is rather played by share option schemes for corporate executives (Hansmann and Kraakman, 2004, p. 67). Taken together, these mechanisms drive managers to adopt a form of management based on the “creation of shareholder value” (Froud, Haslam, Johal and Williams, 2000a and 2000b; Lazonick and O’Sullivan, 2000; O’Sullivan, 2000; Hossfeld and Klee, 2003): in practical terms, this means achieving maximum financial profitability so as to maintain a favourable stock market evaluation (see below).

13 This system of control is usually referred to as ‘outsider control’, because it relies first and foremost on players from outside the company (the minority shareholders). The main characteristic of this model is that it takes the stock market price as a central indicator in the management of listed companies.

14 In France (and continental Europe as a whole), the importance of institutional investors is lower – notably due to the slighter presence of pension funds. The possession of shares by non-financial companies is relatively substantial. Unlike the typical shareholdings of institutional investors, these are often ‘blockholdings’. The concentration of ownership is therefore quite high (La Porta, Lopez de Silanes and Shleifer, 1999; Barca and Becht, 2001)[ 2] [ 2] Ownership concentration is favoured by legal rules lying...
suite. It tends to protect or insulate listed companies from the stock market. In addition, employees’ rights to information and consultation, particularly through works councils, constitute an internal counterweight (Rebérioux, 2002). In Germany, these rights are made even more substantial by the system of co-determination, where employee representatives of certain companies are given seats on the supervisory board (with the same rights as the shareholder representatives). This general pattern of corporate control is referred to as ‘insider control’, giving significant voice to agents committed to stable relations with the company (blockholders and employees).

15 This difference in the relative importance attached to stock market valuations within each of the governance systems is not without consequences for the strategic choices and the management of companies. Probably the most direct effect concerns the dividend policies of listed companies: according to La Porta, Lopez de Silanes, Shleifer and Vishny (2000), common law countries (including Great Britain and the United States) have a more generous dividend policy than civil law countries (including France and Germany). This very general result is criticised by Goergen (2007), who nevertheless finds, in a comparison between Great Britain and Germany, that German firms are more likely to cut dividends when their net results fall. These studies suggest that the payment of dividends does represent a more strategic, or more sensitive, variable for listed companies that are under direct pressure from the stock market.

16 However, the above typology is being overturned: the shift of the continental European model of shareholding towards the Anglo-Saxon model is now widely described in the comparative literature (see, for example, Hansmann and Kraakman, 2001; Denis and McConnell, 2003). In the French case, it is possible to identify certain developments that undeniably play a part in shifting the insider model of governance towards a more market-based (outsider) one.

17 The first evidence is the growth in stock market capitalisation, mostly due to an increase in the volume of security transactions. This growth in stock market activity is directly linked to the increasing penetration of these markets by financial investors, not only national but also non-resident. Tirole (2006) estimates that one third of the capital of French listed companies was held by non-residents in 2002, whereas Thomsen (2004) reports that at the beginning of the 1990s, the proportion of ownership by non-residents was less than 15%. In 2005, for the largest companies (included in the CAC40), 46.4% of the equity capital was held by non-residents, with more than 20% for British and US funds looking for international diversification of their portfolios (Poulain, 2006). This increase in the power of minority shareholders in the equity capital of French companies has been accompanied by a decline, but not a collapse, in blockholdings (see Thomsen, 2004, pp. 306-308). Table 1 presents the distribution of ownership of French listed companies in 2004 (date when the REPONSE survey was conducted): considering that the large majority of non-residents are investment funds, we see that institutional investors, French and foreign, represent by far the most important category of shareholders in listed companies today.

Table 1 - Ownership of common stock (as a percentage of outstanding shares) for French listed companies in March 2004

18 In parallel, important changes in financial market law and, to a lesser extent, in corporate law have occurred over the last decade, moving towards greater protection for minority shareholders (Fanto, 1998; Cioffi and Cohen, 2000). The information disclosure obligations of listed companies have been markedly strengthened, notably by three texts: the “New Economic Regulation” Act of May 2001, the “Financial Security” Act of August 2003 and the “Ordonnance” n°2004-604 of June 2004. These texts were directly inspired by the publication, since the mid-1990s, of a series of corporate governance codes, promoting the interests of minority shareholders[3] [3] See the Viénot I (1995) and Viénot II (1999) reports,...
suite.

19 The calculations made by Batsch (2007) on the 30 industrial groups of the CAC40 also reveal that dividends are on the upswing: between 1999 and 2005, the dividend per share practically doubled on average (multiplied by 1.9). Ginglinger and L’Her (2006) point out that share buybacks, by which some of the cash flow can be transferred to shareholders, have become commonplace, encouraged by a change in the regulation in 1998. The standard use of stock options by French listed companies is another marker of the process of ‘financialization’: according to a study by the firm Towers Perrin, published in May 2007, France is the European country in which the proportion of stock options and free shares in executive pay is the highest (50 %, compared with less than 30 % in Great Britain, and more than 65 % in the United States).

20 The use of data from the last two REPONSE surveys (1997-1998 and 2004-2005) confirms these trends. The first evidence is that of an increase in the number of establishments belonging to companies listed on the stock market[4] [4] The figures presented in this article are weighted. They...
suite: in 2004, 23.9% of establishments belonged to a listed company, compared to 21.9% in 1998. 37.6% of the workforce was employed in a listed company in 2004, compared to 34.2% in 1998. The second evidence is that of the increasing importance of institutional investors (resident and non-resident) in the equity capital of listed companies. In 2004, these investors were the largest category of shareholders for 28.4% of the establishments belonging to listed companies (thus outstripping all the other categories), against only 17.7% in 1998. The equity ownership structure of non-listed companies remained much more stable, with a predominant proportion owned by families or individuals.

3 - The stock market and human resource management: some hypotheses

21 The literature on the ‘Varieties of Capitalism’ (Hall and Soskice, 2001) brings to light the way in which the characteristics of the financial markets can be related to certain practices in the field of HRM. A financial system that favours the rapid reallocation of capital (exit strategies for investors) is seen as being complementary to a flexible labour market, with low job protection. Conversely, less liquid financial markets, such as they exist traditionally in the continental European model, promote a certain level of job stability within companies, favourable to the development of specific human capital. If we use the same analytical grid to examine the situation at the corporate level, in a country with a pure ‘insider’ model of governance, we should not find any clear difference in HRM practices between listed and non-listed companies: the weakness of the control mechanisms associated with the stock market should diminish the differences likely to affect listed and non-listed companies. In other words, being listed on the stock market should not, per se, imply a specific style of management. On the other hand, we should expect to observe more pronounced differences in HRM practices between the two groups of companies as the influence of the stock market gains ground. This is the hypothesis underpinning, for example, the study of Jackson, Höpner and Kurdelbusch (2005), who seek to grasp the specificity of HRM practices in the large listed companies in Germany. In the French case, given the developments in corporate governance described in the previous section, we may assume that HRM practices can be differentiated according to whether or not the company is listed and the extent to which its equity capital is held by financial investors.

22 It is possible to specify the ways in which stock market pressure is likely to influence the management, organisation and remuneration of labour. A priori, the most direct effect of the current changes in corporate governance is to strengthen the requirements of financial profitability imposed on listed companies[5] [5] Besides, the process of ‘financialization’ strengthens...
suite. In a capitalist system, every company has to achieve a level of profitability allowing it to cover the cost of capital. Over and above this level, the requirements of profitability will be more or less intense according to the macro-institutional context that underlies the variety of capitalism (Hall and Soskice, 2001; Amable, 2003; Jackson et al., 2005). The process of ‘financialization’ tends to accentuate these requirements. Competing to attract household savings, investment funds seek to offer the highest possible profitability (at a given level of risk) to their beneficiaries. This is not without consequence for the companies in which they own shares. As a result, institutional investors promote new forms of corporate management, based on the maximisation of ‘shareholder value’[6] [6] A conspicuous example of strategic turn by listed companies...
suite.

23 Value-based management models, which first appeared in the 1990s[7] [7] On the diffusion of these models, see for example Cooper,...
suite, are all founded on the same principle: there is ‘creation of shareholder value’ when the (financial) profitability achieved by the company is higher than the profitability expected by the market (the cost of equity capital for the firm). The most widely-used tool of shareholder value creation, the Economic Value Added (EVA), expresses this approach most clearly. Presented by its advocates as being closely correlated to the stock market price, this indicator makes it possible to operationalize the requirements imposed by investors. R denotes the net result (net operating profit after tax), k the cost of capital (or the equilibrium return on equity as calculated by the Capital Asset Pricing Model), EC the book value of equity capital and ROE (Return On Equity) the financial profitability (the ratio of the net result to equity capital, R / EC). We can then write two equivalent expressions of EVA for a firm at time t:

24

25 According to this approach, therefore, the wealth actually created for shareholders is the value added over and above the profitability expected by the market (ROE – k). The market equilibrium return (k) is considered as the minimum return, a benchmark, on the basis of which the real creation of value can be appreciated (Batsch, 1999). Managers are invited to maximise shareholder value in each financial period, and this should guarantee them a favourable evaluation in the stock market. From an operational point of view, as managers have no direct influence over the cost of capital (k), the requirement of shareholder value creation ultimately comes down to a requirement for the maximisation of financial profitability (R / EC).

26 Accounting analysis can be used to appreciate the different paths open to obtain maximum financial profitability (Froud et al., 2000a and 2000b). As with any ratio, there are two possibilities. The first considers the denominator, the capital invested (EC): share buybacks, which reduce total equity capital, are used ever more frequently by listed companies (see Ginglinger and L’Her, 2006). The second considers the numerator, the net result (R, the difference between turnover or total income and total costs). It is at this level that HRM practices can be used as strategic leverage. More precisely, two strategies can be distinguished:

27 • The first strategy could be described as ‘defensive’ or ‘low road’. It consists in minimising labour costs, the primary component in operating costs. In its most direct form, this may be achieved by reducing the workforce, through ‘downsizing’. Company restructuring over the last two decades has often appeared to be guided by the desire to reduce the proportion of value-added devoted to labour, in order to satisfy the profitability constraints imposed by the rise to power of market finance[8] [8] We should note, however, that no systematic empirical link...
suite (see, for example, Froud et al., 2000a for the British case and O’Sullivan, 2000 for the case of the United States during the 1980s and 1990s). For a given size of the workforce, a low road strategy can also take the form of a restrictive pay policy or the limitation of training expenditures. This hypothesis is envisaged by Black et al. (2007), who use macroeconomic data to test the impact of stock market activity on the training effort of firms – without obtaining any significant results.

28 • The second strategy has more of an ‘offensive’, ‘high road’ nature. It acknowledges the fact that certain short-term costs can, over the medium to long term, increase total income and so both net result and profitability. The motivation of employees (through pay policy for example) and high expenditures on training can increase productivity and the company’s innovation capacity, so maintaining a competitive advantage. Consequently, a strategy of short-term cost minimisation can conflict with a strategy of maximising profitability over a longer-term horizon.

29 In sum, the question of the impact of profitability requirements on the level of pay and on training expenditures remains a priori open.

30 Yet a prominent effect of stock market pressure is to strengthen the damaging consequences in case of poor (short term) financial results: as the EVA definition makes clear, any “destruction” of shareholder value (financial profitability that is positive but below the cost of capital) runs the risk of provoking a fall in the price of the company’s shares. Accordingly, listed companies might seek greater control over the variations in profit than non-listed companies. Here, the flexibility of operating costs might be exploited (Colasse 2001; Froud et al. 2000b). Labour costs, as the main component of operating costs, are a natural candidate to that end. They might be adjusted through the use of flexible forms of employment and flexible pay practices. By using temporary labour arrangements, through employment contracts (fixed-term contracts) or commercial contracts (temporary agency workers and subcontracting), the company can resort to a workforce without any long-term commitment. It can adjust the quantity of labour used over the very short term. Pay policy is another tool of labour cost ‘flexibilisation’. Beyond the wages, employers have the possibility of implementing, on a selective and reversible basis, individual bonus schemes (linked to attendance or individual performance appraisal for example) or collective bonus schemes (linked to financial performance or sales notably). An indication of this effect is provided by Jackson (2005) and Jackson et al. (2005) who observe the massive use of performance-related pay schemes by German firms pursuing shareholder value-based management strategies.

31 To sum up, the growing importance of stock market valuation in the strategies of French listed companies is pushing up the requirements of financial profitability. HRM practices are likely to be influenced, to the extent that they can contribute to meeting these requirements: in particular, we expect to find greater use of flexible contracts and pay practices in listed companies whose capital is primarily owned by financial funds, whether resident or foreign. The effects on labour remuneration and training expenditure are difficult to interpret a priori: companies might be prone to adopt a defensive strategy, with low investment in human resources, or a more ‘offensive’ management style. The following section therefore sets out to evaluate the existence of a style of HRM consistent with the pressure of the stock market, drawing on the data of the 2004-2005 REPONSE survey.

4 - Empirical analysis

32 The REPONSE (Workplace Industrial Relations) survey constitutes a unique source of information in France from which one can combine relatively complete information about both HRM practices and the ownership structure of companies, for a representative sample. To obtain information on the pay level, we have drawn information from the DADS (Annual Declarations of Social Data), merged with the REPONSE database. The DADS give us exhaustive, administrative information on wages, as well as on worker characteristics.

33 Our objective is to estimate the influence of corporate governance on different HRM practices, controlling for an extensive list of firm and worker characteristics. We now present in more detail the variables describing corporate governance and HRM practices, the control variables and our empirical strategy.

4.1 - Corporate governance variables

34 The 2004-2005 REPONSE survey contains two questions related to corporate governance: one concerning the listing of the equity capital and the other concerning the nature of the principal category of shareholders (French or foreign institutional investors, French or foreign non-financial companies, families or individuals, State or employees). We first exclude companies primarily owned by the State or employees, given their specificities. We end up with a sample of 2,521 establishments (out of the 2,930 establishments in REPONSE) among which 993 belong to listed companies (see table 2). More than 24% of French establishments with 20 or more employees (excluding the agricultural sector and public sector) belong to listed companies, employing 37.6% of the labour force (i.e. about 3,610,000 employees). Beyond quotation, we differentiate listed companies according to the nature and nationality of the largest category of shareholder (see table 2). The increased penetration of financial investors, particularly Anglo-Saxon ones, has been identified as a driving force in the evolution of the French model of corporate governance. We therefore expect the identity of shareholders to influence the degree of ‘financialization’ of the companies. More precisely, we expect listed companies in which institutional investors are the main category of shareholders to be more directly involved in value-based management strategies, and specifically when these investors are non-resident.

Table 2 - Listing and ownership of equity capital

4.2 - HRM variables

35 In the previous section, we set out the hypothesis that stock market valuation can have significant consequences on HRM practices. We investigate these effects at four different levels: (i) on the use of temporary work arrangements, (ii) on the use of flexible pay systems, (iii) on the level of pay and (iv) on the training expenditures. Each of these practices is considered at the establishment level. This enables us to examine in the most precise way the effective employment practices put in place.

36 (i) Among the flexible work arrangements, we take into account temporary agency workers, fixed-term contracts and subcontracting. In the survey, the use of both temporary workers and fixed-term contracts is given as a percentage of total workforce. These types of contract are quite common in France (nearly 40% of establishments use temporary workers and nearly 60% used fixed-term contracts, according to the 2004-2005 REPONSE survey). We do not intend to explain the complete distribution of those variables, nor the difference between non users and occasional users. Rather, we are interested in the difference between intensive and non-intensive use, so as to focus on clear-cut cost flexibilisation strategies. Further, the utilisation rate of these types of contract varies greatly from one sector to another (as Table A.1 in Appendix illustrates): the value of the third quartile ranges from 0 to 21% for temporary workers or from 0 to 15% for fixed-term contracts. We therefore examine the behaviours of intensive use on a sector-by-sector basis: we create a dummy variable that takes the value of ‘1’ if the establishment has a rate of use equal to or greater than the third quartile in its sector, ‘0’ otherwise. In the survey, the question concerning the use of subcontracting distinguishes uniquely between establishments who do or do not use it. We define a dummy which takes the value of ‘1’ if the establishment uses subcontracting, ‘0’ otherwise.

37 To complete our analysis of the way companies manage their labour requirements, we exploit information on variations in the salaried workforce (be they open-ended or fixed-term contracts) over the three years preceding the survey. The REPONSE survey does not provide precise figures for workforce variations over previous years, but indicates the direction of these variations – stable, increasing or decreasing – according to the management representatives. Using this information, we can analyse situations where the workforce increased or decreased rather than stayed stable.

38 (ii) Concerning flexible pay practices, the REPONSE survey gives information about the existence of individual and collective bonus schemes for managerial employees on the one hand, and for non-managerial employees on the other. We create four dummies (equal 1 if bonus are used, 0 otherwise) so as to account for a possible differentiation of strategies of wage ‘variabilisation’ between managerial and non-managerial workers.

39 (iii) The DADS data provide precise information about average and median net hourly wage levels in the establishment. We do not intend to explain individual wages through workers characteristics, as it is usual in the human capital theory. Rather, we consider here average (and median) wage as a strategic variable for the firm, a variable that might be analysed through firm-level attributes including ownership[9] [9] For a similar approach, see for example Burgess, Lane and...
suite. However, we control as precisely as possible for the characteristics of the workforce in the establishment. A high level for the average (or median) wage, controlling for workers characteristics, tends to indicate the existence of a rent for employees. Note that our data exclude top executives remunerations. Accordingly, our empirical study cannot be related to the topic of (excessive) managerial compensation in listed companies (see e.g. Bebchuk and Fried, 2005).

40 (iv) Finally, the REPONSE survey provides data about the level of training expenditures as a percentage of the total wage bill of the establishment. Although the initial variable is subdivided into five classes, we use the 3% level of expenditures as a benchmark for high employer commitment to training. This threshold corresponds to the average rate of expenditures in France in 2004 (Idmachiche, 2007). We therefore construct a dummy that takes the value of ‘1’ if training expenditures are equal to or higher than 3% of the wage bill, ‘0’ otherwise.

41 Table 3 presents the distribution of the different HRM variables, as well as a first exploratory analysis of the relations between these variables and those relating to corporate governance. We observe, for example, that the use of individual bonuses for managerial employees is much more frequent in establishments belonging to listed companies than to non-listed ones, and this difference is even more pronounced when the primary category of shareholders are institutional investors (French or foreign). Econometric analysis enables to further investigate these links.

Table 3 - Human resource management practices by equity ownership structure

4.3 - Control variables

42 The same set of control variables is used in the different regressions, even if some modulations have been made where necessary. These variables concern the structural characteristics of the establishment, the commercial context in which it operates and the socio-demographic characteristics of the workforce. The details and distribution of these variables are presented in Table A.2 in the Appendix.

• For the establishment, we include its sector of activity, its size in number of employees and its age, but also the turnover level of the company it belongs to (as proxy for the size of the firm). Beyond the economic justification for including such variables, we also use them as representatives of the design feature of the REPONSE survey (based on size and sector) in the econometric analysis[10] [10] See Reiter, Zanutto and Hunter (2005) for a methodological...
  suite.
• Following previous research in HRM (see e.g. Schuler and Jackson, 2005), we take into account the role the context plays in shaping human resources policies and practices. A set of variables then describes the market environment in which the establishment evolves: its market share, the predictability of demand for its products, the fact of being a subcontractor (for at least 10% of its turnover), having recourse to subcontractors, the state of the market over the three years preceding the survey (growth, stable or decline) and the existence or not of an unexpected shock in demand during the last year. We have also chosen to introduce a variable describing the competitive strategy adopted. Deakin et al. (2006) have shown, on the basis of monographs on British listed companies, that the way in which stock market pressure impacts on HRM is not uniform, but depends on different mediations, specific to each company, and notably the strategy they adopt on the products market (low price versus high quality).
• The characteristics of the workforce are taken into account through the structure of occupational groups (proportion of managers and supervisors; technicians and professionals; clerical workers; frontline workers), the proportion of employees aged under 40 and the proportion of women[11] [11] The socio-demographic characteristics of the workforce are...
  suite. We use categorical variables, except for regressions on the wage levels. In this latter case, the linear specification of the regressions (see infra) and the need to take into account more precise information about the characteristics of the workforce lead us to use continuous variables.

4.4 - Empirical specification

43 According to the nature of the explained variable, the estimations are computed either on the basis of linear regressions using Ordinary Least Square method (for the wage level variables), dichotomous logit estimations (for the different forms of work arrangements and pay schemes as well as for training expenditures) or multinomial logit estimations (for workforce variations). We then estimate three types of equations of the following form:

44

45 where Pij is the probability of occurrence of practice i={1,…, 8} for establishment j, Lj a dummy variable taking the value ‘1’ when the company to which the establishment belongs is listed (‘0’ otherwise), Xjk the set of K controls (for a list see Appendix, table A.2), ai, bi and (ci1 ;… ; ciK) the estimated coefficients and εj the independent and identically distributed (i.i.d.) random noise. Practices are the following:

• P1j : the probability for establishment j to have a rate of use of temporary workers equal to or greater than the third quartile of the sector
• P2j : the probability for establishment j to have a rate of use of fixed-term contracts equal to or greater than the third quartile of the sector
• P3j : the probability for establishment j to resort to sub-contracting
• P4j : the probability for establishment j to use individual bonus schemes for non-managerial employees
• P5j : the probability for establishment j to use individual bonus schemes for managerial employees
• P6j : the probability for establishment j to use collective bonus schemes for non-managerial employees
• P7j : the probability for establishment j to use collective bonus schemes for managerial employees
• P8j : the probability for establishment j to spend more than 3% of the wage bill on training

46 where Pij is the probability of occurrence of case i={1,2} for establishment j, P3j the probability of occurrence of case 3 (reference category), Lj the listing variable, Xjk the set of K controls, ai, bi and (ci1 ;… ; ciK) the estimated coefficients for case i and εj the i.i.d. random noise. Alternative situations are the following:

• P1j : the probability for establishment j to have an increase in workforce size (over the last three years)
• P2j : the probability for establishment j to have a decrease in workforce size
• P3j : the probability for establishment j to have stability in workforce size (reference category)

47 where j refers to the establishment, Lj is the listing variable, Xjk the set of K controls, ai, bi and (ci1 ;… ; ciK) the estimated coefficients and εj the i.i.d. random noise. Vij, i={1,2}, refers to the following:

• V1j : the average net hourly wage in establishment j
• V2j : the median net hourly wage in establishment j

We also address the possible role of institutional investors as a determinant of HRM practices. To do so, we use two more dummies: RII, which takes the value ‘1’ if resident institutional investors are the prime category of shareholders (‘0’ otherwise) and NRII, which takes the value ‘1’ if non-resident institutional investors are the main shareholders (‘0’ otherwise). We then run all the previous regressions including, next to the listing dummy (Lj), crossed effects of listing according to the nationality of institutional investors (Lj*RIIj and Lj*NRIIj). In the next section, we report the results of this alternative model only when the coefficients associated with crossed effects are significant[12] [12] The complete results are available from the authors upon...
suite.

Table 4 - Forms of temporary work arrangements (results of logit estimation)

5 - Estimation results

48 We successively present results of econometric regressions (i) on the use of temporary work arrangements, (ii) on the use of flexible pay systems, (iii) on the level of wages and (iv) on the level of training expenditures. As a first, general comment, we observe that stock market listing has a significant influence on all the practices. Furthermore, once listed, the fact that institutional investors are the prime category of shareholders, be they French or foreign, does not influence the use of temporary work arrangement nor the level of training expenditures. However, it does shape establishment’s pay policy.

49 (i) The use of each of the three forms of temporary labour mobilisation (agency workers, fixed-term contracts and subcontracting) is estimated separately. Table 4 presents the results of the estimations. We observe a particularly strong tendency to resort to agency workers or subcontractors among establishments belonging to listed companies. On the other hand, those establishments make significantly less use of fixed-term contracts. Establishments subject to stock market pressure therefore tend to use commercial contracts (temp agencies and subcontractors) rather than employment contracts (fixed-term contracts) to meet their temporary manpower needs.

To complement the analysis of labour mobilisation strategy, we also estimate the influence of listing on changes in the size of the establishment’s workforce. We observe a negative significant effect of stock listing on increases in the size of the workforce (see Table 5): growth in staff numbers is less likely than stable workforce in establishments belonging to listed companies. But it has no significant effect on reductions.
Taken together, these results suggest that listed companies favour the use of commercial contracts, to the detriment of new hiring (whether in open-ended or fixed-term contracts). While providing some flexibility in cost management, such a strategy allows to stabilize the workforce around a relatively limited core group of employees and to limit the costs of labour turnover. Neither the forms of temporary work arrangements nor changes in the workforce size are significantly related to the presence of institutional investors, whatever their nationality.
(ii) Alongside the recourse to employment flexibility, we also investigate the use of variable pay practices through the use of individual and collective bonus schemes. Four models are estimated so as to differentiate between the determinants of the implementation of individual or collective bonus schemes, granted to managerial employees or non-managerial employees (Table 6).

Table 5 - Changes in the workforce size (results of multinomial estimation)

Table 6 - Individual and collective bonus schemes of establishments (results of logit estimation)

50 Listing has a very significant influence on the distribution of nearly all forms of bonus. Being listed distinctly increases the probability of using individual bonus schemes for managerial employees and collective bonus schemes for all the employees. We therefore observe a practice, linked to the stock market, of intensive pay ‘variabilisation’ – a similar result to that obtained by Jackson et al. (2005) for Germany.

51 Introducing crossed effects of listing and the presence of institutional investors as main shareholders has a significant effect on pay schemes, but only in the case of foreign holdings (Appendix, table A3)[ 13] [ 13] We might also stress the positive link between the presence...
suite: non-resident financial investors are associated with a greater use of individual bonus schemes for non-managerial employees. Their presence in the equity capital of listed companies then seems to strengthen the logic of pay ‘variabilisation’, by extending it to non-managerial workers.

52 In sum, establishments belonging to listed companies make particularly frequent use of temporary workers, subcontractors and flexible pay schemes (this last effect is amplified by foreign holdings). These three tools of HRM enable companies to ‘variabilise’ their costs by modulating both the volume of labour used (via temporary workers and subcontractors) and the cost of the labour (via bonus schemes).

53 (iii) Next, we turn to the influence of listing and the identity of shareholders on wage levels. Two complementary variables are examined: the average wage in the establishment and the median wage. The results of these estimations are presented in Table 7.

54 Corporate governance has a strong significant effect on wage levels, both average and median: they are higher in establishments belonging to listed companies. Wages do not, therefore, follow a rationale of cost minimisation within establishments belonging to listed companies. Once again, this result echoes that of Jackson et al. (2005) for German firms: although they observe that the pressure exerted by financial markets leads companies to reduce the workforce, they also show that the wages paid to the remaining employees tend to rise.

55 Testing the influence of the identity of shareholders we find that, once again, the presence of institutional investors has a different effect according to their nationality (see appendix, table A4). Whereas French financial investors do not influence the level of pay, foreign investors are significantly associated with higher median and average wages. In doing so, they tend to reinforce the effect of stock market listing.

Table 7 - Hourly wages levels within establishment (results of OLS estimation)

Table 8 - Training expenditures (results of logit estimation)

(iv) We finally turn to training expenditures. Table 8 presents the results of the estimation: we find that the coefficient on listing is positive and significant. Stock market listing is positively linked to high levels of training expenditures. Once again, the empirical evidence does not support the hypothesis of a ‘defensive’, ‘low road’ strategy in HRM for listed companies, induced by stock market pressure. Testing for the identity of main shareholders does not yield any significant result here.

56 The results can be summarized as follows. Our estimations shape the contours of a particular profile of HRM associated with listing on the stock market. This profile is characterised by three main features:

• Firstly, empirical evidence confirms the hypothesis of a strong search for cost flexibility associated with stock market valuation. This flexibility is achieved through the implementation of individual and collective bonus schemes (wage flexibility) and the use of commercial contracts (temp agencies and subcontracting).
• Secondly, our results suggest that listed companies tend to maintain the workforce around a relatively limited core group of employees: the negative effect of listing on the use of fixed-term (employment) contracts as well as the reluctance to hire over the last three years are evidence of this strategy.
• Last but not least, our estimations show that listed companies adopt an offensive (high road) rather than a defensive strategy for HRM. Controlling for firm and worker characteristics, we find that wage levels and training expenditures are higher in establishments belonging to listed companies. However, this commitment to human resources is concentrated on a relatively limited number of long-term core employees, isolated through the intensive use of commercial contracts.

Another important result is that this HRM profile is only marginally influenced by the identity of the main shareholders: the presence of institutional investors as the main shareholders in listed firms has a significant influence in only 3 estimations out of 11. Moreover, these cases only concern non-resident investors. More precisely, foreign financial investors appear to influence the pay policy in a way that reinforces the effect of listing (higher wages and more flexible pay schemes).

57 These findings, based on cross-sectional data, are in line with the conclusions reached by Jackson et al. (2005). Their analysis synthesises a collection of studies, mainly monographs, recording the changes in HRM practices in German listed companies subject to increasing stock market pressure[14] [14] Höpner (2001) constructed an index of sensitivity to shareholder...
suite. They observe a tendency to reduce the number of salaried staff, to fall back on a stable core of employees, most often without mass lay-offs (using voluntary departure, early retirement, etc.). Within the company, average wage levels tend to rise, while the use of individual and collective bonus schemes is becoming widespread. These converging findings are of a particular interest. Indeed, France and Germany are, to a large extent, undergoing similar transformations in corporate governance. Accordingly, these findings might convey a broader understanding of the way evolutions in corporate governance impact on business conduct and labour management in so-called ‘insider’ model countries.

58 While the growing sensitization of the executives of German listed companies to the interests of minority shareholders can hardly be disputed, the German system still exhibits specific fundamental characteristics (Schmidt, 2003; Lütz and Eberle, 2007). In particular, labour law and corporate law provide for important worker involvement in corporate governance, which has remained quite stable over the last 15 years. Some authors argue that this strong institutionalization of worker voice has influenced the way the shareholders’ growing requirements for financial return have been met. Vitols (2004), for example, observes the emergence of a hybrid model, which he calls ‘negotiated shareholder value’, where the implementation of value-based management principles are negotiated with labour representatives.

59 In France, German-style board-level representation for workers does not formally exist. However, the growth of employee stock ownership tends to increase worker voice: the board of directors (or the supervisory board) should include two worker-shareholder representatives, with voting power, whenever employees own more than 3% of the equity capital of their company (Code du commerce, art. L.225-23 and 225-71). Concerning the works council, the French Comité d’entreprise is a very different institution from the German Betriebsrat: it contains representatives of both workers and managers, is chaired by a representative of the employer, and has no codetermination (veto) rights. Nevertheless, the powers of the French works council should not be understated. Article L.431 of the Labour Code, for example, requires the employer to provide the works council with the information it may wish to obtain on the general functioning of the company. On some aspects, the information rights held by the work council are even greater than those of shareholders: managers must inform worker representatives (and not the shareholders) whenever the trade of a block holding is expected (Labour Code, art. L.432-1)[15] [15] Another important right is the possibility for the works...
suite. Over the last decade, legal developments (and in particular the 2001 “New Economic Regulation” Act) have reinforced those powers, with a tendency to move worker representative rights closer to the shareholders’ traditional privileges. For example, just like a shareholder or a group of shareholders holding more than 5% of equity capital, the works council has the right to submit proposals to be included with a company’s proxy statement at the general meeting (Labour Code, art. L.432-6-1). The New Economic Regulation Act also established the obligation for the company making a takeover bid to inform and consult the works council of the targeted firm about its industrial projects; if it does not, the voting rights acquired during the operation are cancelled (Labour Code, art. L.432-1).

60 Besides the field of information / consultation / codetermination, French law offers particularly strong protections for industrial actions (strikes), as the ‘leximetric’ (quantification of law) and comparative analysis conducted by Deakin, Lele and Siems (2007) demonstrates. It might be argued that those rights, just like co-determination provisions in Germany, strengthen the bargaining position of worker representatives, with some influence on the way financial requirements are delivered.

61 To sum up, in France, just like in Germany, the diagnosis of a one-dimensional movement of convergence towards Anglo-Saxon standards needs to be qualified. Accordingly, it is interesting to note that HRM practices associated with shareholder value in Germany appear somewhat similar to the practices we observe in France. Taken together, those findings suggest that, because of labour law involving workers in corporate governance matters, continental European countries might find a specific way to cope with stock market requirements. In a nutshell, this way appears to be based on intensive investment in human capital, but for a limited core of stable employees.

6 - Conclusion

62 This work has enabled us to explore the way the profitability requirements imposed by the stock market impact on HRM practices. The empirical evidence exhibits a relatively homogeneous effect of stock market listing: the predominance of institutional investors in the equity capital is only occasionally influential and does not qualitatively change the results. More precisely, the presence of French institutional investors has no significant influence on HRM policies, whereas foreign investors tend to reinforce the characteristics of pay policy associated with stock market listing. This result suggests that if the shareholder-value approach may have been primarily carried by institutional investors during the 1990s, it has now penetrated all companies listed on the stock market.

63 Financial return requirements result in HRM practices focused on a limited core of employees who enjoy relatively favourable working conditions, in terms of wages and training. However, these employees are not entirely shielded from the business cycle, because of the very pronounced use of variable and reversible forms of pay. These characteristics suggest that listed companies have taken a “high road” to profitability, based not so much on the minimisation of costs as on the enhancement of human capital coupled with a strong incentive policy.

64 The very marked use by these establishments of commercial contracts (temp agencies and subcontractors) to adapt to manpower needs is another characteristic feature, calling for further reflection. On the basis of the 2004-2005 REPONSE survey, Perraudin, Petit, Rebérioux, Thèvenot and Valentin (2008) report the unfavourable working conditions among subcontractors. In parallel, Perraudin, Thèvenot, Tinel and Valentin (2006) show that the profitability of these firms, engaged in a strategy of cost minimisation, is relatively low. Taken together, these results suggest that part of the profitability of listed companies is obtained precisely by means of outsourcing. This interpretation, which requires empirical confirmation, identifies listed companies with profit centres, satisfying the requirements of market finance through a dual strategy, combining enhancement of human capital on the one hand with cost minimisation, by means of outsourcing, on the other. In this context, a more profound understanding of the process of ‘financialization’ can only be achieved by widening the field of analysis beyond the sole category of listed companies. The transformation of financial systems, marked by the rise to power of stock markets, might then be seen to participate in a global reconfiguration of productive systems.

Annexe

Appendix

Intensity of use of temporary workers and fixed-term contracts as a percentage of salaried employees (third quartile by sector)

Table A1 - 

Temporary workers Fixed-term contracts Agri-food industry 9 8 Consumer goods industry 1 4 Automotive industry 7 1 Capital goods industry 10 4 Intermediate goods industry 7 4 Energy 3 2 Construction 21 4 Commerce 1 6 Transports 5 5 Financial activities 0 6 Real-estate activities 3 11 Business services 2 5 Personal and domestic services 0 9 Education, health, social services 0 15 Administration in private sector 0 4 Note: quartiles calculated on weighted distributions Field: establishments of 20 and more employees in the private sector (excluding agricultural sector) Source: 2004-2005 REPONSE survey, management representatives questionnaire, Dares

Distribution of control variables

Table A2 - 

Frequency Sector Agri-food industry 3.43 Consumer goods industry 4.18 Capital goods and automotive industries 5.92 Intermediate goods industry / energy 11.92 Construction 8.54 Commerce 20.58 Transports 7.25 Financial and real-estate activities 4.72 Business services 16.47 Personal and domestic services 7.14 Education, health, social services 9.85 Establishment size Less than 50 employees 64.37 From 50 to 100 20.28 From 100 to 200 9.27 From 200 to 500 4.82 500 and more 1.27 Age of establishment Less than 10 years 13.01 10 to 50 years 67.87 More than 50 years 19.13 Company turnover in € million Less than 5 18.94 5 to 10 32.81 10 to 100 30.38 More than 100 17.87 Competitive strategy Price 20.83 Innovation 6.89 Quality 56.4 Originality, reputation, diversity 12.23 Not applicable 3.64

Frequency Market share Less than 50% 61.29 More than 50% 12.85 Not applicable 25.87 State of the market Growing 56.43 Stable 28.79 Declining 14.79 Difficulty in predicting demand 71.21 Unusual variation in demand 41.56 Subcontractor 15.9 Prime contractor 53.71 Proportion of women Less than 15% 24.81 From 15% to 60% 48.77 More than 60% 26.43 Proportion of employees aged under 40 Less than 40% 19.51 From 40 to 70% 52.27 More than 70% 28.23 Proportion of managers & supervisors; technicians & professionals Less than 15% 29.3 From 15% to 30% 25.92 From 30 to 50% 18.24 More than 50% 26.54 Number of observations in the sample 2521 Weighted number of observations 112653 Note: in weighted % of establishments Field: establishments of 20 and more employees in the private sector (excluding agricultural sector) Source: 2004-2005 REPONSE survey, management representatives questionnaire, Dares

Individual and collective bonus schemes when specifying the ownership structure

Table A3 - 

Individual bonuses Collective bonuses For non-managerial employees For non-managerial employees For managerial employees Ownership of equity capital Listed -0.106 (0.109) 0.556 (0.125) *** 0.593 (0.132) *** Listed and french institutional investors 0.106 (0.201) 0.465 (0.274) * 0.489 (0.289) * Listed and foreign institutional investors 0.447 (0.202) ** -0.370 (0.231) -0.400 (0.243) * Note: Standard error in parenthesis. *, **, *** indicate that the coefficient is significant to the 10%, 5%, 1% level respectively. Same control variables as in table 6. Full results are available upon request. Field: establishments of 20 and more employees in the private sector (excluding agricultural sector). Source: 2004-2005 REPONSE survey, management representatives questionnaire, Dares

Hourly wages levels within establishment when specifying the ownership structure of the firm (results of OLS estimation)

Table A4 - 

Average net hourly wages (in log) Median net hourly wages (in log) Ownership of equity capital Listed 0.022 (0.012) * 0.019 (0.009) ** Listed and french institutional investors -0.009 (0.018) 0.005 (0.016) Listed and foreign institutional investors 0.041 (0.021) ** 0.043 (0.017) ** Note: White robust estimates of the standard error are in parenthesis. *, **, *** indicate that the coefficient is significant to the 10%, 5%, 1% level respectively. Same control variables as in table 7. Full results are available upon request. Field: establishments of 20 and more employees in the private sector (excluding agricultural sector) Source: 2004-2005 REPONSE survey, management representatives questionnaire, Dares

Bibliographie

References

Abe, M. and T. Hoshi (2007), “Corporate Finance and Human Resource Management”, in M. Aoki, G. Jackson and H. Miyajima (eds.), Corporate Governance in Japan: Institutional Change and Organizational Diversity, Oxford, Oxford University Press.

Amable, B. (2003). The Diversity of Modern Capitalism, Oxford, Oxford University Press.

Amable, B., Ernst, E. and S. Palombarini, (2005). “How do financial markets affect industrial relations: an institutional complementarity approach”, Socio-Economic Review, vol. 3, pp. 311-330.

Aoki, M. (2007). “Whither Japan’s Corporate Governance?”, in M. Aoki, G. Jackson and H. Miyajima (eds.), Corporate Governance in Japan: Institutional Change and Organizational Diversity, Oxford, Oxford University Press.

Barca, F. and M. Becht, (eds.) (2001). The Control of Corporate Europe, Oxford, Oxford University Press.

Barker, R. and D. Rueda, (2007). “The Labor Market Determinants of Corporate Governance Reform”, CLPE Research Paper, n° 5, Osgoode Hall Law School, York University, Toronto.

Batsch, L. (1999). Finance et stratégie, Paris, Economica.

Batsch, L. (2007). “Variations empiriques autour des groupes du CAC40, 2000-2005”, paper presented at the CEPN seminar (University Paris Nord), 9 February 2007.

Bebchuk, L. and J. Fried, (2005). “Pay without Performance: Overview of the Issues”, The Journal of Corporation Law, vol.30, pp.647-673.

Berle, A. and G. Means, (1932). The Modern Corporation and Private Property, new ed. 1967, New York, Harcourt, Brace and World.

Black, B. (1998). “Shareholder Activism and Corporate Governance in the United States”, in The New Palgrave Dictionary of Economics and the Law, vol. 3, pp. 459-465.

Black, B., Gospel, H. and A. Pendleton, (2007). “Finance, Corporate Governance, and the Employment Relationship”, Industrial Relations, vol. 46(3), pp. 643-650.

Blair, M. (1995). Ownership and Control. Rethinking Corporate Governance for the Twenty-first Century, Washington DC, Brookings Institution.

Blair, M. and M. Roe, (eds.) (1999). Employees and Corporate Governance, Washington DC, Brookings Institution.

Bouton, D. (2002). Pour un meilleur gouvernement des entreprises cotées, Report of joint committee MEDEF and AFEP / Agref.

Bryson, A., Forth, J. and C. Barber, (eds.) (2006). “Making Linked Employer-Employee Data Relevant to Policy”, DTI Economics Occasional Paper No. 4, Department of Trade and Industry, London.

Burgess, S., Lane, J. and D. Stevens, (2000). “Job Flows, Worker Flows, and Churning”, Journal of Labor Economics, vol. 18, N° 3, pp 473-502.

Capelle-Blancard, G. and N. Couderc, (2006). “How do shareholders respond to downsizing? A meta-analysis”, Working Paper CES.

Cioffi, J. and S. Cohen, (2000). “The State, Law and Corporate Governance: the Advantage of Forwardness”, in S. Cohen and B. Gavin (eds.), Corporate Governance and Globalization. Long Range Planning Issues, Cheltenham and Northampton, Edward Elgar, pp. 307-349.

Colasse, B. (2001). L’analyse financière de l’entreprise, Paris, La découverte, coll. Repères.

Cooper, S., Crowther, D., Davies, M. and E. Davis, (2000). “The Adoption of Value-Based Management in Large UK Companies: A Case for Diffusion Theory”, 23rd annual Congress of the European Accounting Association, Munich, March 29-31.

Cozian, M., Viandier, A. and F. Deboissy, (2005). Droit des sociétés, 18e ed., Paris, Litec.

Deakin, S. and B. Ahlering, (2005). “Labour Regulation, Corporate Governance and Legal Origin: A Case of Institutional Complementarity?”, University of Cambridge Centre for Business Research Working Paper n° 312 and ECGI - Law Working Paper 72/2006.

Deakin, S., Hobbs, R., Konzelmann, S. and F. Wilkinson, (2006). “Anglo-American corporate governance and the employment relationship: a case to answer?”, Socio-Economic Review, vol 4, number 1, pp. 155-174.

Deakin, S., Hobbs, R., Nash, D. and D. Slinger, (2003). “Implicit contracts, takeovers and Corporate Governance: In the Shadow of the City Code”, in D. Campbell, H. Collins, and J. Wightman, (eds.), Implicit Dimensions of Contract, Oxford, Hart, pp. 289-331.

Deakin, S, Lele, P. and M. Siems, (2007). ‘The evolution of labour law: calibrating and comparing regulatory regimes’, International Labour Review, vol. 146, n°3-4.

Denis, D. and J. McConnell, (2003). “International Corporate Governance”, ECGI - Finance Working Paper, n° 05/2003.

Earle, J. and A. Telegdy, (2007). ‘Ownership and Wages: Estimating Public-Private and Foreign-Domestic Differentials using LEED from Hungary, 1986-2003’, NBER working paper, 12997.

Fanto, J. (1998). “The role of Corporate Law in French Corporate Governance”, Cornell International Law Journal, vol. 31, pp. 31-91.

Froud, J., Haslam, C., Johal, S. and K. Williams, (2000a). “Restructuring for shareholder value and its implications for labour”, Cambridge Journal of Economics, 24(6), pp. 771-98.

Froud, J., Haslam, C., Johal, S. and K. Williams, (2000b). “Shareholder value and financialization: consultancy promises, management moves”, Economy and Society, 29(1),. pp. 80-110.

Ginglinger, E. and J.-F. L’Her, (2006). “Ownership Structure and Open Market Stock Repurchases in France”, European Journal of Finance, Vol. 12, No. 1, pp. 77-94.

Goergen, M. (2007). “What Do We Know About Different Systems of Corporate Governance?”, ECGI - Finance Working Paper n° 163/2007.

Goergen, M., Brewster, C. and G. Wood, (2006). “The Boundaries of Governance: The Effects of the National Setting and Ownership Changes on Employment Practice”, ECGI - Finance Working Paper n° 136/2006.

Gospel, H. and A. Pendleton (eds.) (2005). Corporate Governance and Labour Management: An International Comparison, Oxford, Oxford University Press.

Gourevitch, P. and J. Shinn, (2005). Political Power and Corporate Control. The New Global Politics of Corporate Governance, Princeton, Princeton University Press.

Hall, P. and D. Soskice, (2001). “An Introduction to Varieties of Capitalism”, in Hall, P. and Soskice, D. (eds.). Varieties of Capitalism: The Institutional Foundations of Comparative Advantage, Oxford, Oxford University Press, pp. 1-70.

Hansmann, H. and R. Kraakman, (2001). “The End of History for Corporate Law”, Georgetown Law Journal, vol. 89, pp. 439-68.

Hansmann, H. and R. Kraakman, (2004). “The Basic Governance Structure”, in R. Kraakman, P. Davies, H. Hansmann, G. Hertig, K. Hopt, H. Kanda and E. Rock (eds.). The Anatomy of Corporate Law. A Comparative and Functional Approach, Oxford, Oxford University Press, pp. 33-70.

Hertig, G. and H. Kanda, (2004). “Related Party Transactions”, in R. Kraakman, P. Davies, H. Hansmann, G. Hertig, K. Hopt, H. Kanda and E. Rock (eds.), The Anatomy of Corporate Law. A Comparative and Functional Approach, Oxford, Oxford University Press, pp. 101-130.

Höpner, M. (2001). “Corporate Governance in Transition: Ten Empirical Findings on Shareholder Value and Industrial Relations in Germany”, Max Planck Institut Discussion Paper 01/5, Cologne.

Hossfeld, C. and L. Klee, (2003). “Performance Reporting on EVA by French and German Companies”, 2nd workshop on Performance and Management Control, Nice, September.

Idmachiche, S. (2007). “La dépense nationale pour la formation professionnelle continue et la formation en 2004”, Premières Informations Premières Synthèses, DARES, n° 02.1, January.

Jackson, G. (2005). “Stakeholders under pressure: corporate governance and labour management in Germany and Japan”, Corporate Governance, vol. 13(3), pp. 419-428.

Jackson, G., Höpner, M. and A. Kurdelbusch, (2005). “Corporate governance and employees in Germany: Changing linkages, complementarities, and tensions”, in H. Gospel and A. Pendleton (eds.), Corporate Governance and labour management: an international comparison, Oxford, Oxford University Press, pp. 84-121.

Konzelmann, S., Conway, N., Trenberth, L. and F. Wilkinson, (2006). “Corporate governance and human resource management”, British Journal of Industrial Relations, vol. 43, pp. 541-567.

La Porta, R., Lopez-de-Silanes, F. and A. Shleifer, (1999). “Corporate Ownership Around the World”, The Journal of Finance, vol. 54, pp. 471-517.

La Porta, R., Lopez-de-Silanes, F., Shleifer, A. and R. Vishny, (2000). “Agency Problems and Dividend Policies around the World”, The Journal of Finance, vol. 55, pp. 1-33.

Lazonick, W. and M. O’Sullivan, (2000). “Maximising Shareholder Value: A New Ideology of Corporate Governance”, Economy and Society, vol. 29(1), pp. 13-35.

Lehto, E. and P. Böckerman, (2006). “Enemy of Labour? Analysing the Employment Effects of Mergers and Acquisitions”, Munich Personal RePEc Archive Paper.

Lütz, S. and D. Eberle, (2007). "On the Road to Anglo-Saxon Capitalism? German Corporate Governance Regulation between Market and Multilevel Governance", CLPE Research Paper, n° 4/2007, Osgoode Hall Law School, York University, Toronto.

Margolis, D. (2006). “Should Employment Authorities Worry about Mergers and Acquisitions?”, Portuguese Economic Journal, vol. 5(2).

O’Sullivan, M. (2000). Contests for Corporate Control. Corporate Governance and Economic Performance in the United States and Germany, Oxford, Oxford University Press.

Perraudin, C., Petit, H., Rebérioux, A., Thèvenot, N. and J. Valentin, (2008). “Une gestion de l’emploi qui dépasse le cadre de l’entreprise”, forthcoming in T. Amossé, C. Bloch-Lordon and L. Wolff (eds), Relations sociales en entreprise, Paris, coll. Recherche, La Découverte.

Perraudin, C., Thèvenot, N., Tinel, B. and J. Valentin, (2006). “Sous-traitance dans l’industrie et ineffectivité du droit du travail: une analyse économique”, Economie et Institutions, N° 9, pp. 35-55.

Poulain, J.-G. (2006). “La détention du capital des société françaises du CAC40 par les non-résidents fin 2005”, Bulletin de la Banque de France, mai, n°149.

Prowse, S. (1995). “Corporate Governance in an International Perspective: a Survey of Corporate Control Mechanisms among Large Firms in the U.S., U.K., Japan and Germany”, Financial Markets, Institutions and Instruments, vol. 4, pp. 1-61.

Rebérioux, A. (2002). “European Style of Corporate Governance at the Crossroads: the Role of Worker Involvement”, Journal of Common Market Studies, vol. 40 (1), pp. 111-34. Reprinted in Clarke T. (ed.) 2005. Corporate Governance, Volume III: European corporate governance, London, Routledge, pp. 64-87.

Reiter, J., Zanutto, E. and L. Hunter, (2005). “Analytical Modeling in Complex Surveys of Work Practices”, Industrial and Labor Relations Review, vol. 59, N° 1, pp 82-143.

Roe, M. (2002). “Corporate Law’s Limits”, Journal of Legal Studies, vol. 31(2), pp. 233-271.

Schmidt, R. (2003). "Corporate Governance in Germany: An Economic Perspective", CFS Working Paper No. 2003/36.

Schuler, R. and S. Jackson, (2005). “A Quarter-Century Review of Human Resource Management in the U.S.: The Growth in Importance of the International Perspective”, Management Revue, vol. 16 (1), pp. 1-25.

Thomsen, S. (2004). “Convergence of Corporate Governance during the Stock Market Bubble: Towards Anglo-American of European Standards?”, in A. Grandori (ed.) Corporate Governance and Firm Organization. Microfoundations and Structural Forms, Oxford, Oxford University Press, pp. 297-317.

Tirole, J. (2006). The Theory of Corporate Finance, Princeton, Princeton University Press.

Towers Perrin, (2007). “Intéressement à long terme: pratiques et tendances en France aujourd’hui”, press release, 29 May.

Viénot, M. (1995). Le conseil d’administration des sociétés cotées, Report of the joint CNPF-AFEP committee, Paris.

Viénot, M. (1999). Rapport du comité sur le gouvernement d’entreprise, Joint CNPF-AFEP committee, Paris.

Vitols, S. (2002). “Shareholder value, management culture and production regimes in the transformation of the Germany chemical-pharmaceutical industry”, Competition and Change, vol. 6, pp. 309-325.

Vitols, S. (2004). ‘Negotiated Shareholder Value: The German Variant of an Anglo-American Practice’, Competition and Change 8 (4): 357-374.

Notes

[ *] Centre d’Etudes de l’Emploi (CEE) and CES (CNRS, Université Paris I); corinne.perraudin@mail.enpc.fr

[ **] Centre d’Etudes de l’Emploi (CEE) and CES (CNRS, Université Paris I); heloise.petit@univ-paris1.fr

[ ***] EconomiX (CNRS, Université Paris X); antoine.reberioux@u-paris10.fr
This article originates in a research founded by the Dares (the Research and Statistics Department of the French Ministry of Labour). We would like to thank Thomas Amossé, Eve Caroli, Richard Crabtree, Simon Deakin, Howard Gospel, Loup Wolf, Serge Zilberman and two anonymous reviewers for helpful comments. We also benefited from discussions with the participants at the 19th Annual Meeting on Socio-Economics (Copenhagen, June 2007), at the 27th Annual meeting of the Association d’Economie Sociale (Nanterre, September 2007) and at the PSE seminar ‘Marché du travail et inégalités’ (March 2008). The usual disclaimers apply.

[ 1] Abe and Hoshi (2007) propose a similar approach for Japan. However, there are two important differences. Firstly, their sample only contains 58 companies. Secondly, all these companies are listed, only differing in the distribution of their equity ownership. This work cannot, therefore, be used to compare the direct, pure effects of listing on the stock market.

[ 2] Ownership concentration is favoured by legal rules lying in the domain of corporate law. In France, this essentially involves double voting rights (Hansmann and Kraakman, 2004, pp. 55-56), voting caps to protect against hostile takeovers (Cozian, Viander and Deboissy, 2005, p. 290) and shareholder pacts (Cozian et al., 2005, p.305).

[ 3] See the Viénot I (1995) and Viénot II (1999) reports, and the Bouton report (2002).

[ 4] The figures presented in this article are weighted. They relate either to all establishments in the French private sector (excluding agricultural sector) with 20 or more employees (125,200 establishments) or to all the employees in these establishments (9.6 million employees).

[ 5] Besides, the process of ‘financialization’ strengthens the requirements of informational transparency for listed companies. We leave this question aside in this article.

[ 6] A conspicuous example of strategic turn by listed companies in reaction to the increasing pressure of stock market is reported by Vitols (2002), with the ‘Big Three’ German integrated chemical/pharmaceutical companies (Hoechst, Bayer and BASF). Whereas those companies tended in the post-war period to prioritize growth (in particular through diversification) and investment over profitability, the transformation in corporate governance in the mid-1990s led them to focus on stock price and financial profitability. In the case of Hoechst, this involved quite a radical change of business model, with the abandonment of the integrated strategy.

[ 7] On the diffusion of these models, see for example Cooper, Crowther, Davies and Davis (2000) and Hossfeld and Klee (2003).

[ 8] We should note, however, that no systematic empirical link has been evinced between a rise in stock market price and a cut in the workforce (see Capelle-Blancard and Couderc, 2006). While recognising that restructuring can be motivated by financial profitability, it is clear, however, that shareholders do not always value downsizing.

[ 9] For a similar approach, see for example Burgess, Lane and Stevens (2000) or Earle and Telegdy (2007).

[ 10] See Reiter, Zanutto and Hunter (2005) for a methodological discussion of the role of weights in analytical modelling in complex surveys of work practices.

[ 11] The socio-demographic characteristics of the workforce are drawn from the DADS.

[ 12] The complete results are available from the authors upon request.

[ 13] We might also stress the positive link between the presence of French institutional investors and the use of collective bonuses (for managers and non-managers). Yet these results did not appear robust enough to interpret them: the effects are significant only at a 9% level. Similarly, foreign holding is negatively linked to the use of collective bonuses for managerial workers, but only to the p-value of 9.9%. Some further studies would be needed to confirm this result.

[ 14] Höpner (2001) constructed an index of sensitivity to shareholder value, including four items: the quality of financial information communicated to shareholders, the importance of communication efforts with minority shareholders, the explicit introduction of EVA-type value-based management tools and the use of complementary payments linked to stock market valuations of the company (e.g. stock options) as part of executive compensation packages.

[ 15] Another important right is the possibility for the works council to call in an expert accountant, in order to obtain a counter-valuation of the information communicated by the employer (Labour Code, art. L.434-6).

Résumé

Cet article étudie l’influence de la structure de propriété des entreprises (cotation et identité des actionnaires) sur la gestion de l’emploi. Nous utilisons l’enquête REPONSE 2004-2005, fondée sur un échantillon de 2930 établissements de 20 salariés et plus, représentatif du secteur marchand français. L’analyse économétrique confirme l’importance de la propriété comme déterminant de la gestion de l’emploi, que ce soit en termes de formes temporaires de mobilisation du travail (recours à l’intérim, aux CDD et à la sous-traitance), de politique salariale (niveau des rémunérations et usage de primes) ou de formation.
Classification JEL – G34, J3, M5.

Mots-clés

gouvernance d’entreprise, propriété sociale, gestion de l’emploi

PLAN DE L'ARTICLE

• 1 - Introduction
• 2 - The dynamic of the French model of governance
• 3 - The stock market and human resource management: some hypotheses
• 4 - Empirical analysis
  • 4.1 - Corporate governance variables
  • 4.2 - HRM variables
  • 4.3 - Control variables
  • 4.4 - Empirical specification
• 5 - Estimation results
• 6 - Conclusion
• Annexe

POUR CITER CET ARTICLE

Corinne Perraudin et al. « The Stock Market and Human Resource Management: Evidence from a Survey of French Establishments », Recherches économiques de Louvain 4/2008 (Vol. 74), p. 541-581.
URL : www.cairn.info/revue-recherches-economiques-de-louvain-2008-4-page-541.htm.
DOI : 10.3917/rel.744.0541.