An empirical investigation of flexibility patterns

3 downloads 0 Views 194KB Size Report
Jan 25, 2010 - 41 Mortimer Street, London W1T 3JH, UK. The International .... the GDP of many countries' (Rubery and Grimshaw 2003, p. 198). In these ...
This article was downloaded by: [Bibliothèque cantonale et universitaire - Fribourg] On: 25 January 2010 Access details: Access Details: [subscription number 788627695] Publisher Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 3741 Mortimer Street, London W1T 3JH, UK

The International Journal of Human Resource Management

Publication details, including instructions for authors and subscription information: http://www.informaworld.com/smpp/title~content=t713702518

Does location matter? An empirical investigation of flexibility patterns in foreign and domestic companies in five European countries Sebastian Schief a a Department of Social Work and Social Policy, University of Fribourg, Fribourg, Switzerland Online publication date: 25 January 2010

To cite this Article Schief, Sebastian(2010) 'Does location matter? An empirical investigation of flexibility patterns in

foreign and domestic companies in five European countries', The International Journal of Human Resource Management, 21: 1, 1 — 16 To link to this Article: DOI: 10.1080/09585190903466830 URL: http://dx.doi.org/10.1080/09585190903466830

PLEASE SCROLL DOWN FOR ARTICLE Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf This article may be used for research, teaching and private study purposes. Any substantial or systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.

The International Journal of Human Resource Management, Vol. 21, No. 1, January 2010, 1–16

Does location matter? An empirical investigation of flexibility patterns in foreign and domestic companies in five European countries Sebastian Schief*

Downloaded By: [Bibliothèque cantonale et universitaire - Fribourg] At: 16:18 25 January 2010

Department of Social Work and Social Policy, University of Fribourg, Fribourg, Switzerland A major field of discussion about globalization is the role multinational enterprises play in an international economy. The essence of the debate lies in the distribution of power between the two centres of influence – multinational corporations, on the one hand, and the institutional setting of nation states, on the other. Is it the local environment (country-specific settings) that exerts the decisive influence on the organization of employment and production or is it the corporate environment (corporate settings) of the companies in question? The aim of this contribution is to analyze flexibility patterns of foreign owned companies compared with domestic owned companies in five European countries with respect to the aforementioned debate. The analysis uses the EUCOWE dataset which is based on an enterprise survey in six countries (France, Germany, the Netherlands, Portugal, Spain and the United Kingdom) in 2003. By comparing foreign owned with domestic owned companies, similarities and differences between companies and between countries were analysed. The results show a hierarchy of effects influencing the flexibility patterns of enterprises. The analysis gives strong support to the argument that the institutional framework of countries is still a key factor for the organization of labour even within multinational corporations. Keywords: comparative research; flexibilization; governance; multinational corporations; work organization

1.

Introduction

As economic globalization advances, the balance between ‘the power of the local environment’ (Flecker 2000, p. 67), that is the influence of the institutions operating at particular production sites and office locations, and the power of companies, particularly of multinational companies, seems to have been lost. The decisive factor in this shift in the balance of power is said to be the threat to move production and hence jobs elsewhere, a threat that can be forestalled only if management’s demands are met. According to this interpretation, this bargaining power places companies in the position to impose lower wages, longer working hours and more flexible systems of work organization. So the basic question is whether the power of companies, and of multinationals in particular, has reached a new level, as economic globalization gathers pace, and, if so, what form this shift in the balance between the power of the local environment and that of companies takes. Companies’ power is determined by the tension between the local environment that is the institutional setting at a given location and the company itself with its particular organizational structure and culture (c.f. Schmidt and Dworschak 2002). Companies will always try to obtain what they regard as the optimal conditions under which to produce

*Email: [email protected] ISSN 0958-5192 print/ISSN 1466-4399 online q 2010 Taylor & Francis DOI: 10.1080/09585190903466830 http://www.informaworld.com

Downloaded By: [Bibliothèque cantonale et universitaire - Fribourg] At: 16:18 25 January 2010

2

S. Schief

goods or deliver services. However, local conditions, and in particular industrial relations and legal arrangements, may stand in the way of companies seeking a free hand to determine working and employment conditions. For some years now, there has been a wide-ranging academic debate on how to assess the strength of companies’ position relative to the local environment (c.f. Chase-Dunn 1990; Krugman 1994; Hirsch 1995; Hirst and Thompson 1996; Schief 2003, 2008). The present investigation seeks to contribute to this debate by examining the influence of corporate organization and institutional setting on companies’ flexibility patterns. Flexibility patterns are defined as various combinations of personnel-related measures that are used to deal with short and long-term fluctuations in production or demand. The investigation focuses in particular on the degree of influence of national institutional settings in comparison to the influence of multinational corporations on the adoption of flexibility patterns. This object of investigation would seem to be particularly well suited to our purpose, since both companies and the institutional setting can be assumed to exert considerable influence on the selection of measures intended to cope with fluctuations in production. The research questions can be expressed as follows: Research question 1: Are certain flexibility patterns used more or less frequently by foreign than by domestic companies and how does the use of these instruments differ in the five countries under investigation? Research question 2: How strong are these influences on flexibility patterns of institutional setting in comparison with corporate organization? Research question 3: Can a hierarchy of influences be established?

The article is divided into four following sections. In section 2, a theoretical framework is outlined that can be used to examine the relationship between companies and their institutional setting under the influence of economic globalization, that is the increasing interdependence of actors in the global economy. This is followed by a description of possible flexibility patterns, while in section 4 the dataset and methodology are explained. The subsequent empirical section of the article begins with an examination of differences in the flexibility patterns adopted by companies in the five countries under investigation. This is followed by a comparison of the patterns used by domestic and foreign owned companies in the five countries and an investigation of German, British and Americanowned companies. Finally, a multivariate analysis of the flexibility instrument ‘flexible working times’ checks for correlations between the various types of influence.

2. Globalization, multinational companies and national institutional settings In the academic debate on the extent and effects of an increasingly integrated global economy, two matters are the subject of particularly heated debate (for a summary, cf. Schief 2003, p. 33). The first is governance, i.e., the possibility of regulating processes through political action. The second is the role of multinational companies in a globalized economy. One argument heard in the debate takes a largely globalized economy as a starting point and interprets developments since the beginning of the 1980s as a massive push towards globalization. In this increasingly integrated global economy (e.g., Narr and Schubert 1994), multinationals are regarded as the principal actors (Hirsch-Kreinsen 1997). The ‘competitive state’ (Hirsch 1995) exists only as a service provider, regulating in order to optimise markets. However, both the impetus for and the objectives of state action are, according to this line of argument, dependent on capital in the shape of

Downloaded By: [Bibliothèque cantonale et universitaire - Fribourg] At: 16:18 25 January 2010

The International Journal of Human Resource Management

3

multinational companies. States’ attempts to create favourable conditions for multinationals lead to a ‘competing down’ process (Chase-Dunn 1990) between them. The counterargument to this analysis, with its negative view of the influence that can be exerted by nation states, has been advanced by authors such as Krugman (1994) and Hirst and Thompson (1996), who take a fundamentally different view of both the extent and effects of globalization. Drawing on historical analyses, Hirst and Thompson cast doubt on the uniqueness of the present rush towards globalization, pointing to the significantly higher degree of integration in the global economy prior to 1914. Krugman (1994) contests the existence of competition between nation states. In his view, it is national economic factors, in particular productivity, that determine a region’s living standards rather than its position in global competitive markets. Nevertheless, he argues, the competitiveness argument occupies a position of hegemony in political discourse because it is admirably suited to justify unpopular political decisions or to avoid them (Krugman 1994, p. 40). Both Hirst and Thompson (1996) and Krugman (1994) also cast doubt on the notion that the increased mobility of capital is leading to a radical shift in investment and jobs from the developed to the less developed countries. They ground their argument on empirical findings showing that trade, production and financial relations are concentrated within the so-called triad of North America, Western Europe and Japan. In his analysis, Schief (2003) shows that this is also true of direct investments by German companies. Assessment of the extent and effects of an integrated global economy leads logically to attempts to evaluate the opportunities open to multinationals to influence national institutional settings and vice versa. Thus Rubery and Grimshaw (2003), for example, see the rise of multinationals as a potential challenge to the continuation of national employment practices, since ‘some individual multinational corporations can now be considered more powerful than many nation states, and certainly have a value in excess of the GDP of many countries’ (Rubery and Grimshaw 2003, p. 198). In these authors’ opinion, this economic power is combined with political power. Thus the great power of multinational corporations is reflected in the restricted possibilities available to nation states when it comes to regulation. The introduction of new employment management practices within subsidiaries can certainly exert an influence on national institutional settings; however, that influence can often spread much further: More significantly, they can do what indigenous employers may feel unable to do – ignore custom and practice within the country and sector, and thereby introduce novel and different ways of working and new approaches to human resource management . . . it is this influence of MNCs which may prevent or dissuade individual nation states or groups of nation states from pursuing their own system of social and economic organization. (Rubery and Grimshaw 2003, p. 199)

Another aspect of the possible influence exerted by multinational corporations identified by Rubery and Grimshaw is the diffusion of so-called ‘best-practice’ employment models, which in the long term could threaten the survival of alternative employment models. Another proponent of this argument is Streeck (1993), who sees the European integration of the mid-1980s as a basic agreement between the national political elites and large European companies, according to which business would support the western European nation states in regaining control over foreign policy and receive in return an extensively deregulated economy (‘a single market without a single state’) (cf. Streeck 1993, p. 96). The essence of the debate outlined above lies in the distribution of power between the two centres of influence – multinational corporations, on the one hand, and the institutional setting, on the other.1 Is it the local environment (laws, institutions) that exerts

Downloaded By: [Bibliothèque cantonale et universitaire - Fribourg] At: 16:18 25 January 2010

4

S. Schief

the decisive influence on the organization of employment and production or is it the corporate environment, that is characteristics specific to the companies (organizational structure, organizational culture, size) in question? Schmidt and Dworschak (2002) break the question down further by identifying four kinds of influence that can impact on multinationals. The influence of the corporate environment can be divided into the effect exerted by the environment of origin, that is, the country of origin effect, and the effect exerted by the company itself – the transnational effect. For its part, the influence of the local environment can be divided into the host country effect, that is the effect exerted by the host country’s institutional setting, and a discourse effect, that is the influence of assumptions and prejudices about the local environment. Since the analytical distinction between host country and discourse effect cannot be tested empirically, we will summarize these distinctions by referring to the local environment effect in the rest of the article. The core question to be addressed in this paper is which of these effects exerts what influence on companies’ employment policies and production systems? More specifically, our aim is to examine the question by taking the example of companies’ flexibility patterns. We start from the hypothesis that these patterns are to a large extent shaped by the local environment to which companies are adjusting.2 This hypothesis has been tested before by Igalens, Akremi, Demery-Lebrun and Vicens (2002). They demonstrated that flexibility patterns of companies are heavily dependent on national legal and institutional frameworks. By taking the example of a multinational company they were able to show that the subsidiaries of the enterprise reacted to declining demand in different ways depending on their location. 3.

Companies’ flexibility patterns

Martinez-Lucio and Blyton (1995) define flexibility as the ‘freedom to vary or adjust the quantity and quality of the labour input in response to changes in demand’. This labour input can be regulated in various ways. Atkinson (1984) suggests in his ‘flexible firm’ that companies should, on the one hand, respond to changes in products and production by providing a core workforce with rather rare skills that cannot be bought readily. On the other hand, changes in volume or demand are more easily mastered by peripheral groups of workers with low-level qualifications. Atkinson’s suggestions turn out to be advice for functional and numerical flexibility. In the literature on companies’ flexibility patterns, a distinction is usually made between the flexibility that is dependent on the internal labour market and the flexibility that can be achieved by recourse to the external labour market (cf., for example, Atkinson 1984; Delsen 2002). The second dimension of the various flexibility patterns is the distinction between numerical and functional or quantitative and qualitative flexibility (Atkinson 1984; Delsen 2002). On the basis of these different dimensions, four types of workforce flexibility can be identified, as shown in Figure 1. Internal numerical flexibility can be achieved by use of the core workforce alone. Management can, for example, cut working time, arrange short-time working, give permission for overtime or increase working time flexibility through the organization of working time. A distinction should be made between a rigid increase or reduction in the volume of the labour input and a variable working time system that can absorb fluctuations in output. However, management can also enhance and/or change the quality or function of its core workforce (internal functional flexibility). Some of the options here include job rotation, group work and multitasking. The basic idea is to extend the area within which

The International Journal of Human Resource Management

5

Downloaded By: [Bibliothèque cantonale et universitaire - Fribourg] At: 16:18 25 January 2010

Figure 1. Types of flexibility and their characteristics. Source: Atkinson (1984), Delsen (2002), Lehndorff and Voss-Dahm (2005).

employees can be deployed through further training, which tends of course to be a longerterm strategy for increasing flexibility. External numerical flexibility is produced when companies take recourse to the external labour market in order to deal with fluctuations in production. This usually means the use of additional labour, for example through new hires, fixed-term appointments or the use of temporary agency staff. External functional flexibility, on the other hand, is achieved when parts of the production process or certain services are outsourced or parts of companies are hived off. Freelancers should also be included here, since they may also take on certain tasks previously carried out within the company. In this way, employment contracts can be replaced by contracts for services (cf. Delsen 2002, p. 255). It should be noted that these four types of flexibility are not, in reality, mutually exclusive. Companies can, for example, make simultaneous use of elements of internal numerical flexibility and of external functional flexibility. Lehndorff and Voss-Dahm are thus quite right to point out that the internalization of markets within companies and changes in labour market regulations are increasingly blurring the differences between the various flexibility patterns (cf. Lehndorff and Voss-Dahm 2005, p. 14; Flecker 1999, p. 20). Because of this increasing lack of differentiation between the four types of flexibility, our analysis will be located one level lower. Our focus will be on the actual instruments used to create flexibility, since they can be more clearly distinguished from each other and are easier to operationalize. Among the many flexibility instruments that can be associated with one or other of the individual types of flexibility, we have selected those for analysis we regard as particularly influential. Thus the flexibilization of production will be investigated by means of the following instruments: . overtime (paid and unpaid); . flexible working times (flexi-time, working-time accounts); . use of external workers (contract workers, agency and temporary workers, employees on call); and . recruitment and/or dismissal. The first two flexibility instruments can be assigned to the internal numerical flexibility pattern, while the other two are external flexibility instruments. Whereas new hires and dismissals can be assigned unambiguously to the numerical type of flexibility, the use of external workers cannot be unequivocally associated with the numerical or functional type. We are not concerned here with the exclusivity of flexibility instruments but rather with the frequency with which these instruments are used by various types of company.

Downloaded By: [Bibliothèque cantonale et universitaire - Fribourg] At: 16:18 25 January 2010

6

S. Schief

4. Data and methods The empirical investigation draws on the data set from the EUCOWE project on working times and operating hours in Europe.3 As part of this project, a company survey on working time and operating hours was carried out in six European countries – France, the UK, the Netherlands, Portugal, Spain and Germany.4 The observational survey unit is the establishment. Since there are no Spanish data for certain parts of the investigation relevant to our present purpose, this country is not included in our analysis. Part of this survey deals with fluctuations in production or service provision and the possible means of dealing with these fluctuations. This set of questions can be used to operationalize the four flexibilization instruments selected for our analysis. The companies were first asked whether they actually had to deal with variations in production. Each company that replied in the affirmative was then asked about the methods they used to do so. It is the answers provided that constitute the empirical base for our investigation of flexibility. The companies were asked about their use of overtime (paid and unpaid), flexible working time (working time accounts, flexitime, etc.), recruitment and/or dismissals and their use of external labour, that is the allocation of certain tasks to people outside the company (contract workers, agency and temporary workers, employees on call). Companies of all sizes and in all sectors were included in the EUCOWE survey. Data were gathered from a total of 17,442 companies in the five countries. However, the present investigation includes only companies with 250 and more employees, since multinational corporations are the main focus of our analysis. Bodies incorporated under public law and non-profit-making organizations were also excluded. In total, 1683 companies fulfil these criteria. Of these, 884 (52.6%) stated that they have to deal with variations in production or service provision, and it is these companies that are analysed here. Weighted data (establishment proportional weight) were used for the following bivariate analyses, while the multivariate analysis was carried out with unweighted data. 5. The empirical investigation Our analysis of companies’ flexibility patterns is divided into three stages. In the first, the flexibility patterns within the five countries for which data are available are examined (local environment effect). This is followed by a comparison of domestic and foreign owned companies in these countries (transnational effect, country of origin effect). In the third stage of the analysis, the influence of specific countries of origin (Germany, the UK and USA) on flexibility patterns (country of origin effect) is investigated. The final stage comprises a multivariate analysis of flexible working times. By structuring the investigation in this way, we will be able to examine the three effects outlined above and the correlation between them. 5.1

Flexibility patterns within the countries under investigation

In this first part of the analysis, the influence of the local environment on companies’ flexibility patterns is investigated. The aim, therefore, is to ascertain whether there are statistically significant differences between different local environments in the use of particular flexibility patterns. In general, significant differences are found for all types of flexibility, as can be seen from Table 1. Great Britain is the country that makes by far the greatest use of overtime to absorb fluctuations. More than nine out of ten companies with at least 250 employees in

The International Journal of Human Resource Management

7

Table 1. Share of companies with 250 þ employees using particular flexibility instruments, by country and sector.

Downloaded By: [Bibliothèque cantonale et universitaire - Fribourg] At: 16:18 25 January 2010

N ¼ 884

Germany

France

UK

Netherlands

Portugal

Overtime Manufacturing Services Total

67.6 66.4 67.1

64.7 56.5 60.5

88.1 94.1 91.3

84.7 75.3 78.8

62.5 40.0 54.3

Flexible working times Manufacturing Services Total

83.1 74.5 79.6

42.8 41.6 42.2

18.2 20.8 19.5

15.6 29.2 24.1

20.2 42.0 27.9

External labour supply Manufacturing Services Total

45.9 31.2 39.9

88.4 73.4 80.7

71.0 49.9 59.8

85.0 68.1 74.5

36.4 9.8 26.6

Recruitment and dismissal Manufacturing Services Total

49.3 42.8 46.7

6.0 15.7 11.0

63.8 41.9 52.2

38.1 44.0 41.7

55.1 68.6 60.4

Note: Figures in bold have significant differences (Cramer’s V). Source: EUCOWE, own calculations.

the UK use this instrument to deal with variations in production or service provision. Thus overtime can be seen as the British flexibility instrument. An investigation by Pannenberg and Wagner (2001) confirms the high level of overtime use in the UK. Whereas just about 80% of companies of this size in the Netherlands use this form of flexibility, the figure for Germany is only two-thirds and for France about 60%. The lowest figure – about 55% – is observed for Portugal. The result for the second flexibility instrument, flexible working times, is quite different. Flexible working time means the use of working-time accounts or flexi-time in order to deal with fluctuations in demand. Companies in Germany make by far the greatest use of this form of flexibility. Eight out of ten companies in Germany use this instrument, which can be seen as the German flexibility instrument. The widespread use of flexible working times in Germany is also confirmed by the ISO study on working and operating hours (Bauer, Groß, Munz and Sayin 2002). German companies are followed, at a considerable distance, by their counterparts in France, where only slightly more than four out of ten companies report using this form of flexibility. The shares in Portugal (27.9%), the Netherlands (24.1%) and the UK (19.5%) are even lower. The use of external labour, such as temporary agency workers, is encountered most frequently in France and can be regarded as the predominant flexibility instrument in France. About eight out of ten companies in France state that they use this form of flexibility. The share of companies in the Netherlands that do so is only insignificantly smaller, at 74.5%. Six out of ten companies in the UK have recourse to this instrument, while only about 40% of their counterparts in Germany use it. In Portugal, on the other hand, the share is even lower, at only 25%. The recruitment or dismissal of workers as a means of coping with variations in production or service provision is used most frequently by companies in Portugal (60.4%). In France, on the other hand, only slightly more than 10% of companies with 250 or more

Downloaded By: [Bibliothèque cantonale et universitaire - Fribourg] At: 16:18 25 January 2010

8

S. Schief

employees report using this instrument. Lying between the extremes are the UK, Germany and the Netherlands with figures of between 40 and 50%. In some countries, there are large differences in the use of flexibility instruments between manufacturing and service companies. Across all the countries in the investigation, considerably fewer service companies make use of external labour than companies in the manufacturing sector. This difference is most evident in Portugal (26.6%) and the UK (21.1%). In Portugal, moreover, considerably fewer service-sector companies use overtime (about 20%) in order to deal with variations in demand; on the other hand, the share of companies in the manufacturing sector that make use of flexible working times is only about half that in the service sector. It is also interesting that recruitment and dismissal are used as flexibility instruments in about 63% of British manufacturing companies but only in about 42% of service-sector companies. As can be seen in summary form in Table 2, there are large differences between the flexibility patterns in use in the countries investigated. To exaggerate somewhat for the purposes of the argument, the UK can be characterized as the overtime country and Germany as the flexible working time country. In the Netherlands just as many companies use overtime as use external sources of labour. In France, companies prefer to deal with variations in demand by calling on external sources of labour, while in Portugal overtime and recruitment or dismissal are the types of flexibility used by the greatest share of companies. Table 2.

The main flexibility instruments in the countries investigated.

Country

Main flexibility instrument(s)

Germany France UK Netherlands Portugal

Working time flexibility External labour Overtime Overtime external labour Overtime recruitment and dismissal

Source: EUCOWE, author’s representation.

5.2 Foreign vs. domestic companies: flexibility patterns compared In this section, we test for the existence of influence exerted by a company’s environment, that is the transnational and country of origin effects. In concrete terms, this will involve testing whether there are significant differences in flexibility patterns between foreignowned and domestic companies.5 The relatively small number of foreign-owned companies in our sample does not enable us to make a further distinction between manufacturing and service companies. In Portugal, no significant differences between foreign- (FO) and domestically owned (DO) companies were found for any of the indicators. For the other four countries, on the other hand, such differences can be observed for at least three of the four indicators. In Germany, fewer foreign than domestically owned companies use overtime to deal with variations in demand (DO: 69.2%; FO: 62.7%). On the other hand, considerably more foreign-owned companies than domestic companies call on external labour sources (DO: 35.9%; FO: 49.5%). The difference between the two groups of companies with regard to flexible working times is small, but significant: a greater share of foreign-owned companies uses this form of flexibility than companies in German ownership (DO: 79.1%; FO: 81.9%). Recruitment and dismissal are reported significantly more frequently as a source of flexibility by foreign-owned companies than by domestic companies

The International Journal of Human Resource Management

9

Table 3. Share of companies with 250þ employees that use particular flexibility instruments, by country and domestic (DO) or foreign ownership (FO).

Downloaded By: [Bibliothèque cantonale et universitaire - Fribourg] At: 16:18 25 January 2010

Germany

France

Netherlands

UK

Portugal

N ¼ 866

DO

FO

DO

FO

DO

FO

DO

FO

DO

FO

Overtime Flexible working times External labour Recruitment/dismissal

69.2 79.1 35.9 44.6

62.7 81.9 49.5 50.6

56.7 44.0 80.5 13.0

70.4 37.9 81.7 6.1

87.3 11.8 46.8 55.5

94.6 26.0 70.8 47.7

80.8 22.1 82.2 48.9

78.9 24.3 67.1 38.1

53.2 30.0 21.5 62.6

55.9 25.0 33.3 54.1

Note: figures in bold have significant differences (Cramer’s V). Source: EUCOWE, author’s calculations.

(DO: 44.6%; FO: 50.6%). Thus three of the four flexibility instruments are used more by foreign-owned companies than by those in German ownership. In France, on the other hand, more foreign-owned companies use overtime in order to deal with variations in demand than companies in French ownership (DO: 56.7%; FO: 70.4%). However, fewer foreign-owned companies rely on dismissals or recruitment (DO: 13.0%; FO: 6.1%) and flexible working times (DO: 44.0%; FO: 37.9%) than is the case with French-owned companies. On the other hand, there are no differences between the two groups of companies with regard to the most frequently used flexibility instrument in France, namely recourse to external labour. In the UK as well, three of the four flexibility instruments are used more by foreignowned companies than by companies in British ownership. The only one used more frequently by British-owned companies is the recruitment or dismissal of workers. The differences in the use of external labour and of flexible working times are considerable. More than twice as many foreign-owned as British-owned companies state that they use flexible working times in order to deal with fluctuations in production or the provision of services (DO: 11.8%; FO: 26.0%). The share of foreign-owned companies that use external labour for this purpose is about 25 percentage points higher than that of British companies with 250 or more employees (DO: 46.8%; FO: 70.8%). The use of overtime, the most frequently used form of flexibility in the UK, is even more widespread among foreignowned companies than among those in British ownership (DO: 87.3%; FO: 94.6%). In the Netherlands, it is striking that there are no significant differences between foreign- and Dutch-owned companies in the use of overtime, the most common flexibility instrument. Furthermore, considerably fewer foreign companies than Dutch-owned ones make use of external labour (DO: 82.2%; FO: 67.1%) and dismissals and/or recruitment (DO: 48.9%; FO: 38.1%) in order to deal with variations in demand. On the other hand, flexible working times are used by a higher share of foreign-owned companies (DO: 22.1%; FO: 24.3%). In Germany, France and the UK, the flexibility instruments most frequently reported by domestic companies are also those most frequently mentioned by foreign-owned companies. In the case of the Netherlands, two main flexibility instruments were identified, one of which – overtime – is just as widely used in foreign-owned as in domestic companies. However, there are considerable differences in the use of external labour, which can be observed much more frequently in Dutch companies. No significant differences between foreign-owned and domestic companies could be observed in Portugal, meaning that no effects of the above-mentioned type could be detected. The key

10

S. Schief

Downloaded By: [Bibliothèque cantonale et universitaire - Fribourg] At: 16:18 25 January 2010

lesson to be drawn from this part of the analysis is that effects attributable to the companies’ environment were certainly detected in four of the five countries investigated, but that there were no differences between foreign and domestically owned companies with regard to the main flexibility instruments used.

5.3 Flexibility patterns in British, German and American companies abroad As is evident from the previous section, foreign ownership does in some cases exert a clear influence on companies’ flexibility patterns, albeit without fundamentally challenging the dominant national patterns. This influence has two components: the transnational effect, i.e., the effect of the company itself, and the country of origin effect. In order to be able to investigate these effects in greater depth, we will examine the flexibility patterns adopted by German, British and American companies with 250 or more employees in the countries under investigation because these are major countries of origin of companies in the dataset operating abroad. German companies in Germany and British companies in the UK are of course excluded. Table 4 presents the results for the three groups of companies. Table 4. Share of companies with 250 employees and more that use certain flexibility instruments, by selected countries of origin.

Overtime Flexible working times External labour Recruitment/dismissal

British companies

German companies

US companies

63.0 47.0 54.5 26.5

86.2 46.8 64.7 17.7

81.3 39.0 67.0 32.0

Note: figures in bold have significant differences (Cramer’s V). Source: EUCOWE, author’s calculations.

For all four indicators, there are statistically significant differences between the groups of companies. For example, while about two-thirds of British companies abroad state that they use overtime as a means of dealing with variations in production or service provision, more than 80% of American and more than 86% of German companies use this instrument. Interestingly, however, neither for the German nor the British companies do these findings correspond even remotely to the values recorded for the use of overtime in Germany and Great Britain (cf. Table 1). Significant differences can also be observed in the use of flexible working times. Whereas an almost equal share of British and German companies (47%) use this instrument, only about 39% of American do so. As with overtime, there is a very considerable difference between the share of German companies abroad and that of companies in Germany that use this instrument. The same is true of British companies abroad and companies in the UK. British companies abroad make least use of external labour, with only 54.5% reporting that they do so, compared with about two-thirds of German and American companies abroad. The use of recruitment and dismissals as a means of dealing with variations in demand is most frequent among American companies abroad (32.0%). About one-quarter of British companies and just 18% of German companies have recourse to this flexibility instrument. For this type of flexibility as well, there is an enormous difference between British companies abroad and companies in the UK.

The International Journal of Human Resource Management

11

Downloaded By: [Bibliothèque cantonale et universitaire - Fribourg] At: 16:18 25 January 2010

Essentially, these results can be interpreted as supporting the existence of a country of origin effect, since different flexibility patterns are associated with the various countries of origin. Finally, for all four types of flexibility investigated, significant differences were observed between the companies from three different countries of origin. However, certain reservations arise when it comes to the comparison of these results with those for companies in Great Britain and Germany. Considerable differences were observed between companies in Great Britain and British companies abroad, and the same applies to companies in Germany and German owned companies abroad.6 5.4

Multivariate analysis of flexible working times

The results obtained to date make the presumption of a hierarchy of effects appear plausible. The local environment effect seems to exert a considerably stronger influence on companies’ flexibility patterns than the transnational and country of origin effects, i.e., the effects produced by the companies’ characteristics. In order to verify this hierarchy of effects, logistic regressions for one of the types of flexibility investigated, namely flexible working times, will be calculated by way of example (Table 5). This type of flexibility was selected because it was here that the greatest differences between the countries were observed in the descriptive analysis, as well as significant differences between the various types of company. Dummies for the countries are generated as independent variables for the local environment effects. The independent variable ‘existence of a foreign owner’ is used to represent the transnational effect. In order to be able to investigate the country of origin effect as well, three common nationalities of foreign owners (the USA, UK and Germany) are introduced into the regression as dummies. In addition to these effect variables, dummies for age, sector and size of the company are included in the logistic regression as control variables. Additionally, whether membership of a corporate group exerts any influence was tested. A step-wise procedure was adopted in order to enable us to check the influence of the three effects on each other. In model 1 (Table 5), the only variable included in addition to the control variables is ‘foreign owner’. A negative effect on the probability of the use of flexible working times is observed for companies in foreign ownership. Domestically owned companies have a higher probability of using flexible working times in order to deal with variations in demand. This provides empirical confirmation of the existence of the transnational effect. In model 2, in addition to the control variables, the variables for the three most frequent countries of origin for foreign-owned companies (Germany, the UK and the USA) are introduced in order to test for the country of origin effect. Whereas no significant effect is found for British owners, companies in both German (i.e. German-owned companies operating abroad) and American ownership have a significantly lower probability of using this type of flexibility than companies not in German or American ownership. However, the value for German-owned companies is only weakly significant. Overall, the existence of a country of origin effect can be established by means of this model, even if it applies only to two out of three countries of origin and to varying extents. In model 3, both the transnational and country of origin effects are included, in addition to the control variables. The results show that the most stable effect is found in Americanowned companies. The already weak effect for German-owned companies is no longer significant in model 3, which is also true of the foreign company effect overall. This combination of results can be interpreted by means of the hierarchy of effects posited in the descriptive analysis. The transnational effect (foreign companies) and country of origin

12

S. Schief

Table 5. Logistic regressions for companies with 250þ employees for the dependent variable ‘flexible working times’.

Downloaded By: [Bibliothèque cantonale et universitaire - Fribourg] At: 16:18 25 January 2010

Model 1

Model 2

Model 3

Sector Services Manufacturing

20.007 2 0.010 Reference category

Size 500þemployees 250– 499 emp.

0.655*** 0.636*** Reference category

Age Betw. 5 and 10 years old Older than 10 years Less than 5 years

2 0.464 2 0.370 2 0.651 2 0.188 Reference category

20.373 20.362

Individual company or group Individ. comp. Group

0.129 0.151 Reference category

0.116

Transnational effect Foreign owner Domestic owner

2 0.385** Reference category

Country of origin effect German owner No German o. American owner No American o. British owner No British o.

– 2 0.757þ Reference category – 2 0.740** Reference category – 2 0.474 Reference category

Local environment effect France Netherlands UK Portugal Germany Constants Pseudo-R2 Chi2 N

– – – – – – – – Reference category 2 0.032 2 0.113 0.0257 0.0300 30.66*** 36.07*** 862 869



2 0.030

0.654***

Model 4 2 0.104

0.111

20.336 20.303

20.415þ

20.132

20.062

20.665

0.159

20.657*

20.687*

20.388

20.289

– – – –

21.920*** 22.538*** 22.709*** 22.448***

20.048 0.0316 37.68*** 862

1.882*** 0.1984 236.32*** 862

Note: þ p , 0.1 *p , 0.05 **p , 0.01 ***p , 0.001. Source: EUCOWE, author’s calculations.

effects (German and American companies) can be detected independently of each other, but as soon as both effects are tested for in the same model, only the country of origin effect for the US remains. This latter effect is, therefore, stronger than the transnational effect. Thus the fact of being American in origin is of greater importance for the lower probability of using flexible working times than the mere fact of being a foreign company. In the final stage of the multivariate analysis, the dummies for the countries under investigation, i.e. the companies’ locations, are added in order to test for the local environment effect. The addition of the country variables increases the pseudo-R2, which is a measure of the explanatory power of logistic regressions,7 from 0.03 to about 0.2. In addition, all the country effects are highly significant. Thus the greatest contribution to explaining the use of flexible working times is made in this model by the company’s local

Downloaded By: [Bibliothèque cantonale et universitaire - Fribourg] At: 16:18 25 January 2010

The International Journal of Human Resource Management

13

environment, which can be interpreted as further clear evidence of the hierarchy of effects. In all the other countries, the probability of using this flexibility instrument is considerably lower than in Germany. The greatest difference is between Germany and the UK, while the smallest is between France and Germany. However, the country of origin effect remains for American-owned companies. Such companies obviously have a lower probability of using flexible working times, regardless of location. The significant effects of the size of a company in models 1 to 3 and of individual company status in model 4 constitute a further interesting aspect. It is known from various studies that larger companies also make more frequent use of flexible working times (Bauer, Groß, Lehmann and Munz 2004). This finding is confirmed in the first three models. However, this size effect disappears in favour of the influence of membership of a corporate group when the country variables are incorporated into the model. Obviously the decisive effect that increases the probability of using flexible working times is this membership of a group, rather than sheer size. We assume that this finding results from the greater learning capabilities of organizations within groups as compared with single companies. 6.

Conclusions

We may now answer the questions raised at the beginning of the article. First, we can state that there are different flexibility patterns between the countries of investigation, a socalled local environment effect. For example, whereas the instrument most frequently used by companies in Germany faced with variations in production or service provision is flexible working times, companies in the UK report that their preferred flexibility instrument is overtime. On the basis of these differences in the use of the various types of flexibility, specific flexibility patterns for the five countries under investigation can be constructed. The differences between these national flexibility patterns are very pronounced, and the local environment effect is accordingly very powerful. The reasons for the country differences are to be found in cultural and regulatory effects. On the one hand, enterprises have to handle the national labour law and collective agreements which reduce the options to cope with variations in production. On the other hand, institutional isomorphism leads to similarities in the behaviour of organizations in sectors and other entities without regulation (DiMaggio and Powell 1991). The findings are a strong indicator that the powerlessness of nation-states is clearly overemphasized by authors such as Rubery and Grimshaw (2003). Second, we detected that there are differences between flexibility patterns of foreign and domestic companies. The influence of a company’s country of origin and that of the host country were both demonstrated unequivocally. The second stage of our analysis provided further confirmation of the existence of the transnational and country of origin effects, although their influence appears to be considerably reduced. Certainly there are statistically verified differences between the flexibility patterns adopted by domestic and foreign-owned companies; however, the differences between the two groups are considerably smaller than those between the various countries. Thus, for example, foreign-owned companies in France were found to make considerably greater use of overtime to deal with variations in demand than domestically owned companies; in the Netherlands, on the other hand, considerably more domestic than foreign-owned companies regard the use of external labour as a suitable means of absorbing such variations. Another significant finding from this part of the analysis is that no differences were found within the various countries between domestic and foreignowned companies with regard to the main types of flexibility.

Downloaded By: [Bibliothèque cantonale et universitaire - Fribourg] At: 16:18 25 January 2010

14

S. Schief

The specific investigation of the country of origin effect, carried out by comparing the flexibility patterns adopted by American, German and British companies operating abroad, also revealed significant differences between these groups of companies. German and American companies operating abroad report considerably more frequent use of overtime than British companies operating abroad. On the other hand, considerably more German and British companies abroad than their American counterparts use flexible working times in order to deal with variations in demand. However, the strength of the country of origin effect has to be put into context, since the differences between German companies abroad and companies in Germany as well as between British companies abroad and companies in the UK are very large, which is further evidence that companies tend to adapt to their local environments to a considerable degree. On the one hand, the detected differences between foreign and domestic companies as well as between companies of different origin show that there is restricted influence of the companies in deciding how to cope with variations in production. On the other hand, the results show that this particular influence is weaker than the influence of the cultural and institutional surrounding conditions. Third, we analysed the research question concerning a hierarchy of influences on flexibility patterns by testing all effects namely the country of origin, transnational and local environment effects through a logistic regression for the flexibility instrument ‘flexible working times’. The results obtained confirm with some certainty the existence of all the effects. The local environment effect clearly emerges as having the strongest influence. The country of origin effect has significantly less influence, since it could be detected in all four patterns only for American companies. No effect could be observed for British companies, while for German-owned companies the effect loses its significance when it is tested with the transnational effect. The weakest influence is that exerted by the transnational effect. This effect, generated by the simple fact of operating internationally, can be observed only when neither of the other two effects is included in the model. To that extent, all the effects have a significant influence, but the basic shape of the flexibility pattern is determined by the local environment in which a company is operating. Only then do the country of origin and transnational effects make themselves felt in the hierarchy of effects. Some clear conclusions can be drawn with regard to the debate, outlined at the beginning of this article, on the power of the local environment, about the extent to which national policies can influence companies in general and multinational companies in particular. We have been unable to find any empirical evidence to support the argument that the influence of multinational companies in a globalized economy is considerably greater than that of national institutional environments. On the other hand, the massive influence of the local environment effect shows what opportunities for shaping companies’ behaviour actually exist here. At this level, the basic conditions for companies are laid down; these are then modified in different ways by individual companies but are not fundamentally called into question. The results clearly indicate that there is no need to capitulate when it comes to regulating and influencing the organizational behaviour of companies at national level. It can indeed be argued that ownership matters (Mason and Encarnation 1994) and that a company’s country of origin is also influential. However, the key factors determining work organization still seem to be located in the local environment in which companies operate.

The International Journal of Human Resource Management

15

Acknowledgements The author is grateful to the support of this research by the Hans– Boeckler – Stiftung and the European Commission. Moreover, he would like to thank Steffen Lehndorff and two anonymous reviewers of the initial article for helpful comments and suggestions.

Notes 1.

Downloaded By: [Bibliothèque cantonale et universitaire - Fribourg] At: 16:18 25 January 2010

2.

3. 4. 5.

6. 7.

It has to be mentioned that there are international regulations that gain more importance (ILO, EU regulation, etc.), but up to this point the regulations are rather weak. Given the wide-ranging nature of the local environment vs. company question, simply to investigate the ‘flexibility model’ aspect may appear to be narrowing the focus too much. However, if general statements are to be examined empirically, there is no option but to concentrate on specific issues in order to permit operationalization. This does not of course preclude the investigation of other questions. This project, entitled ‘A Comparative Study of Operating Hours, Working Time and Employment in France, Germany, the Netherlands, Portugal, Spain and the United Kingdom’ (EUCOWE), was funded by the European Commission. For a detailed description of the study see Delsen, Bosworth, Groß and Munoz de Bustillo y Llorente (2007). A distinction will be made between wholly domestic-owned companies and companies that have at least one foreign part owner. Of course domestically owned companies can also be transnational, but we regard this lack of clarity as tolerable, since we are dealing with companies operating in their own national environment. On the other hand, the transnational effect makes itself felt primarily outside of the country of origin. Regrettably, because of the relatively small number of companies in our sample, we are unable to analyse companies from individual countries operating in individual host countries, in order to control for the effect of different host countries. In contrast to the R2 of linear regressions, pseudo-R2 is only an approximate measure of the explained variance.

References Atkinson, J. (1984), ‘Manpower Strategies for Flexible Organizations’, Personnel Management, August, 28 – 31. Bauer, F., Groß, H., Munz, E., and Sayin, S. (2002), Arbeits – und Betriebszeiten 2001, Ko¨ln: ISO. Bauer, F., Groß, H., Lehmann, K., and Munz, E. (2004), Arbeitszeit 2003. Arbeitszeitgestaltung, Arbeitsorganisation und Ta¨tigkeitsprofile, Ko¨ln: ISO. Chase-Dunn, C. (1990), Global Formation: Structures of the World Economy, Cambridge, MA: Basil Blackwell. Delsen, L. (2002), ‘Flexibility Conflict?’ in Operating Hours in Europe: State-of-the-art-report on Operating Hours Research in France, Germany, the Netherlands, Portugal, Spain, the United Kingdom, Sweden and the USA, eds. F. Bauer, H. Groß and G. Sieglen, Ko¨ln: ISO, pp. 217– 236. Flecker, J. (1999), ‘Sachzwang Flexibilisierung’? Unternehmensreorganisation und flexible Bescha¨ftigungsformen, Wien: Forba-Schriftenreihe 2/99. Flecker, J. (2000), ‘Transnationale Unternehmen und die Macht des Ortes’, in Grenzenlose Kontrolle? Organisatorischer Wandel und politische Macht multinationaler Unternehmen, eds. C. Do¨rrenba¨cher and D. Plehwe, Berlin: Edition sigma, pp. 45 – 70. Delsen, L., Bosworth, D., Groß, H., and Munoz de Bustill y Llorente, R. (eds.) (2007), Operating Hours and Working Times. A Survey of Capacity Utilisation and Employment in the European Union, Heidelberg: Physica-Verlag. DiMaggio, P.J., and Powell, W.W. (1991), ‘The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields’, in The New Institutionalism in Organizational Analysis, eds. W.W. Powell and P.J. DiMaggio, London: University of Chicago Press, pp. 63 – 82. Hirsch, J. (1995), Der nationale Wettbewerbsstaat, Berlin: Edition ID-Archiv. Hirsch-Kreinsen, H. (1997), ‘Globalisierung der Industrie: ihre Grenzen und Folgen’, WSI Mitteilungen, 7, 487– 494.

Downloaded By: [Bibliothèque cantonale et universitaire - Fribourg] At: 16:18 25 January 2010

16

S. Schief

Hirst, P., and Thompson, G. (1996), Globalization in Question: The International Economy and the Possibilities of Governance, Cambridge, UK: Polity Press. Igalens, J., El Akremi, A., Demery-Lebrun, M., and Vicens, C. (2002), ‘La flexibilite´ dans le secteur ae´rospatial’, Notes du LIRHE: Universite´ de Toulouse. Krugman, P. (1994), ‘Competitiveness: A Dangerous Obsession’, Foreign Affairs, 73, 2, 28 – 46. Lehndorff, S., and Voss-Dahm, D. (2005), ‘The Delegation of Uncertainty. Flexibility of Service Work and the Role of the Market’, in Working in the Service Sector – A Tale from Different Worlds, eds. G. Bosch and S. Lehndorff, New York: Routledge, pp. 89 – 315. Martinez-Lucio, M., and Blyton, P. (1995), ‘Industrial Relations and the Management of Flexibility: Factors Shaping Developments in Spain and the United Kingdom’, International Journal of Human Resource Management, 6, 2, 271– 291. Mason, M., and Encarnation, D. (1994), Does Ownership Matter? Japanese Multinationals in Europe, Oxford: Clarendon Press. Narr, W.-D., and Schubert, A. (1994), Welto¨konomie: die Misere der Politik, Frankfurt/Main: Suhrkamp. Pannenberg, M., and Wagner, G.G. (2001), ‘Overtime Work, Overtime Compensation and the Distribution of Economic Well-being: Evidence for West Germany and Great Britain’, discussion paper series Forschungsinstitut zur Zukunft der Arbeit 318, Bonn: IZA. Rubery, J., and Grimshaw, D. (2003), The Organization of Employment – An International Perspective, Basingstoke: Palgrave Macmillan. Schief, S. (2003), Globalisierung. Entwicklungspfade der Internationalisierung deutscher Konzerne, Frankfurt/Main: Peter Lang Verlag. Schief, S. (2008), ‘Patterns of Flexibility in Domestic and Foreign Enterprises’, in Comparative Analyses of Operating Hours and Working Times in the European Union, eds. L. Delsen, F. Bauer, G. Cette and M. Smith, Heidelberg: Physica-Verlag, pp. 143– 164. Schmidt, W., and Dworschak, B. (2002), ‘Entgelt – und Arbeitszeitregime unter dem Einfluss von Internationalisierung und multinationalen Unternehmen: ein deutsch-britischer Vergleich’, Schlussbericht des Schwerpunktprogrammes Regulierung und Restrukturierung der Arbeit in den Spannungsfeldern von Globalisierung und Dezentralisierung der Deutschen Forschungsgemeinschaft, Erlangen, 69 – 98. Streeck, W. (1993), ‘The Rise and Decline of Neocorporatism’, in Labor and an Integrated Europe, eds. L. Ulman, B. Eichengreen and W. Dickens, Washington, DC: The Brookings Institution, pp. 80 – 101.