HRM practices abroad: Are SMEs really different? · Italy, Tel. 030.371.5563, Fax 030.370.2448, e-mail: [email protected] . 1 HRM practices abroad: Are SMEs really different?

HRM practices abroad:

Are SMEs really different?

Mariasole Bannòa, Francesca Sgobbi

b

Abstract. Based on a sample of internationalised Italian firms including small, medium, and large companies, the paper explores which variables affect the attitude towards hu-man resource management abroad. The collected data show a variety of differentiated behaviours, which span from the pure control of the foreign investment, to the exploita-tion of the human resources provided by the parent company, to the exploration of the human resources supplied by the host country. While “traditional” size-related effects do not display significant effects, the governance model and the formalisation of HRM practices at the parent company considerably affect the solutions adopted abroad. Key words – HRM practices abroad, internationalisation; SMEs; empirical analysis JEL classification codes – F23, J20, L20, L60 Acknowledgements. We are grateful to Anna Belleri, Alessia Mancini and Federica Pellicciari for valuable help in data collection. We also thank two anonymous referees from the EIBA 2006 committee for their helpful comments on an earlier version of this paper, as well as the FIRB research grant RBNE039XKA_005 for financial support.

a Mariasole Bannò, DIM – Università di Brescia, Via Branze 38, 25123 Brescia, Italy, Tel. 030.371.5563,

Fax 030.370.2448, e-mail: [email protected] b Francesca Sgobbi, corresponding author, DIM – Università di Brescia, Via Branze 38, 25123 Brescia,

Italy, Tel. 030.371.5563, Fax 030.370.2448, e-mail: [email protected]

1

HRM practices abroad:

Are SMEs really different?

1. Introduction Internationalisation processes have been recently involving a growing number of small

and medium enterprises1 (SMEs), whose participation in foreign markets increasingly

entail higher levels of financial and organisational commitment (OECD, 1997; Mariotti and Mutinelli, 2003). For most SMEs, the switch from exports to more binding foreign activities, such as foreign direct investments (FDIs), involves a cultural transition, be-sides a dimensional one, with the most dramatic changes affecting the organisational configuration and the distribution of decision power across the company hierarchy. As a matter of fact, the evolution of the international competitive environment confronts entrepreneurs with the need to reconsider their strategies, and some of their basic values as well. Globalisation processes point out the limits of the entrepreneur-centred model, in favour of more decentralised solutions (Spigarelli, 2003). The variety and the hetero-geneity of international markets make delegation processes a pre-requisite for achieving collective solutions to complex problems, as well as for complementing the entrepre-neurial instinct with rational procedures and processes. More and more often, the entre-preneur of the internationalising SME experiences the need for capable team-mates to increase efficiency, stimulate innovation, manage technological and organisational change, and co-ordinate physical and information flows between the parent company and the foreign subsidiary. However, despite the moves to decentralisation, up to now the empirical evidence suggests that entrepreneurs still play a central role in internation-alised SMEs (McDougall and Oviatt, 2000; Lamb and Liesch, 2002). The entrepre-neur’s competences and capabilities are often critical for company performance in in-ternational markets (Kundu and Katz, 2003; Minguzzi and Passaro, 1993). The opposite pushes between centralisation and decentralisation outline a dilemma of “managing continuity in change”, which affects internationalising SMEs to a larger ex-tent than large-sized enterprises. This dilemma concerns on the one hand the selection of the capabilities which allow replicating abroad the success of the parent company, on 1 According to the European Commission (2003), “the category of micro, small and medium-sized enter-

prises (SMEs) is made up of enterprises which employ fewer than 250 persons and which have an annual

2

the other one the identification and development of those new competences required by

the internationalisation process2. Both processes are critical for SMEs, because they

both involve re-assessment of the entrepreneur’s role. The shift of the company bounda-ries, as well as the introduction of new management techniques for competing abroad, imply a new distribution of decision processes and power inside the organisation. The governance model (which drives the company strategies) and the human resource man-agement system (which selects the employees who assist the entrepreneur abroad) be-come therefore key-variables for understanding how smaller firms organise their foreign activities. The importance of the above issues justify the choice of exploring the human resource management practices at the foreign subsidiaries of internationalised SMEs. The avail-able empirical evidence highlights the extreme variability of HRM models implemented at both small and larger companies (Wright and Brewster, 2003; de Kok and Uhlaner, 2001). An equally large choice appears when considering internationalisation models, even if literature traditionally focused on larger multi-national companies and SMEs-specific research has only recently emerged (Coviello and McAuley, 1999). The gap be-tween the attention devoted to large and small companies further widens when taking into account the variables affecting the approach to HRM in internationalisation proc-esses. While a vast and consolidated literature exists in the case of large firms (see Min-baeva et al., 2003, for a recent survey), SMEs still suffer from the lack of systematic evidence and comprehensive theoretical frameworks (Heneman et al., 2000). Acknowledging the need for specific empirical research, the paper uses original data to investigate the relationship between firm-specific features of internationalised SMEs and their approach to Human Resource Management (HRM) abroad. Among the three major components of international HRM recognised by Schuler et al. (2002) – issues, functions, and policies and practices – our discussion focuses on the last two compo-nents, which involve specific HR initiatives and the development of general guidelines on how individuals will be managed. We claim that the governance model and the characteristics of the HRM system at the parent company, rather than size, drive the role played by human resources in interna-tionalisation processes, hence the success of foreign activities. The data for testing the research hypotheses via statistic and econometric tools were collected between Fall

turnover not exceeding EUR 50 million, and/or an annual balance sheet total not exceeding EUR 43 mil-lion”. 2 In the case of the SMEs located in industrial districts, this dilemma has often been referred to as “the

location paradox” (Storper, 1997; Enright, 1998; Porter, 1998). Also due to information and communica-tion technologies, the globalisation of the whole economic system encourages also firms from less open contexts to exploit their advantages in foreign markets. At the same time, the increased availability of (transferable) standard competences and production models underlines the higher sustainability of com-petitive advantages rooted in local culture and social systems.

3

2003 and Spring 2004 by administering a structured questionnaire to a sample of com-panies from Northern Italy owning FDIs. Contrary to most available empirical evidence, which often focuses on homogeneous samples, the exam of both SMEs and middle-to-large firms allows discriminating be-tween common trends, due to local or contextual factors, and the possible peculiarities of HRM at the foreign subsidiaries of smaller companies. The framework used for con-ceptualising the role of human resources in internationalisation processes resorts to the opposition between two contrasting, yet not mutually exclusive, growth mechanisms (March, 1991). “Resources explorers”, who leverage on newly acquired competences and personnel, oppose to “resource exploiters”, who consolidate their FDIs mainly by deepening and delocalising competences and human resources from the parent com-pany. The empirical analysis confirms the usefulness of the governance model and the HRM system at the parent company to explain HRM practices abroad. On the contrary, size did not prove a significant variable to discriminate among different attitudes. In particu-lar, the data confirm the existence of differentiated approaches also among SMEs and underline the importance of including a wide range of firm- and investment-specific factors when designing the policy tools to support outward internationalisation proc-esses. In order to develop the above outlined reasoning, the paper is organised as follows. Sec-tion 2 synthesises the contributions of literatures to the examined issues. Section 3 dis-cusses the hypotheses which drove the empirical analysis, while Section 4 presents the empirical questionnaire, the examined sample, and the methodology used for testing the research hypotheses. Section 5 discusses the results of the empirical analysis and Sec-tion 6 offers some concluding remarks.

2. SMEs, human resources, and internationalisation processes Before specifying the hypotheses which drove the empirical research, this Section will outline the most relevant contributions of literature to the issues examined in the present paper. In particular, the following paragraphs will survey the peculiarities of HRM prac-tices at SMEs, the role played by human resources in internationalisation processes, and the specific features of organisational learning triggered by foreign investments. 2.1 HRM and SMEs

By exploring the peculiarities of HRM practices at SMEs3, many contributions have

identified the organisational features and behaviours that explain the structural weak-nesses of smaller companies (low specialisation, lack of critical competencies, succes-

3 For an extended survey, see Heneman et al. (2000).

4

sion problems due to entrepreneur-centric vision), as well as their specific strengths (polyvalence of human resources, flexibility, simpler co-ordination tasks). In recent years, HRM has acquired a growing importance also among SMEs (de Kok and Uhlaner, 2001). In a knowledge-based economy, small and large enterprises share the need for more and more qualified human resources. However, SMEs encounter greater difficulties in attracting and retaining them (Audretsch and Thurik, 2000 and 2001). Literature has outlined the correlation between the informal nature of both organisa-tional structures and HRM systems at SMEs (Aldrich and Langton, 1997; Arthur and Hendry, 1990; Marlow and Patton, 1993). However, the debate concerning the relation-ship between the degree of formalisation of the HRM system and the company perform-ance is still open. Some authors emphasise the advantages of best practices (MacDuffie, 1995), which somehow prefigure the diffusion of relatively uniform solutions among firms, while others claim the importance of diversified HRM practices also among smaller companies. According to the supporters of the latter approach, differentiation follows from the close relationship between firm specificities, environmental features, and HRM (see, e.g., de Kok and Uhlaner, 2001; de Kok et al., 2003). Among influential firm- and environment-specific variables, the most frequently reported factors include firm size, strategy, culture, and the peculiarities of the other companies and institutions operating in the same market (Daft, 1998). While theoretical contributions have outnumbered quantitative studies (Heneman et al., 2000), the available evidence on the diffusion of HRM practices suggests that, in gen-eral terms, SMEs tend to adopt less formal solutions than larger companies. However, the variety of the adopted solutions also among SMEs suggests that size cannot be the only significant variable to explain the company choices in the area of HRM (de Kok et

al., 2003). 2.2. Human resources and the internationalisation of SMEs

For a long time, the study of internationalisation patterns has focused on the problems and specificities of large multi-national companies, regarding SMEs as a group of rather homogeneous and undifferentiated actors. The constraints imposed by the reduced availability of financial and managerial resources (Buckley, 1989) justify the traditional choice of modelling the pattern of international growth followed by SMEs as an incre-mental process. The stage theory (Johanson and Vahlne, 1977) shapes the internation-alisation process of the SME as a path along subsequent steps, from exports via indirect sales, to direct sales in foreign markets, to the delocalisation of productive activities to geographically, but also “psychically”, distant countries. According to this stream of lit-erature, in each phase the SME accumulates the competencies, the know-how, and the financial resources to face the following step in a successful way.

5

However, the dynamics actually followed by SMEs in their internationalisation paths question the effectiveness of the stage theory. First, also due to the introduction of inno-vative management and organisational techniques associated with information and communication technologies (Gillespie et al., 2001), the participation of SMEs in inter-national markets has recently displayed acceleration rates never witnessed before (Mu-tinelli, 2001; Prasad, 1999; Mariotti and Mutinelli, 2003; OECD, 1997). Second, the empirical evidence suggests the diversification, rather than the standardisation, of the solutions adopted also by SMEs in pursuing their commercial and productive expansion in international markets (Oviatt and McDougall, 1994). On the one hand, the develop-ment patterns often depart from those forecasted by the stage theory, including leap-frogging and returns (Kundu and Katz, 2003; Lamb and Liesch, 2002). On the other one, the delocalisation of productive activities is by no means the necessary conclusion of an internationalisation process (Gankema et al. 1997; Minguzzi and Passaro, 1993). Some empirical studies openly suggest that the trajectories followed in international

markets do not depend on size only4. In particular, growing consent recently accrued to

the so-called network perspective (Axelsson and Easton, 1992), which suggests that the internationalisation pattern also depends on the network of inter-personal and inter-organisational relationships. According to this approach, the launch of foreign activities follows not only from the rational choices of the top management, but also from the dy-namic formal and informal relationships between the organisation – hence, its members – and the environment. However, a satisfactory model for describing the peculiarities of the internationalisation patterns followed by SMEs is still missing. Coviello and McAuley (1999) underline the limits deriving from reading this issue through the lens of a single theoretical frame-work, suggesting the opportunity of a more holistic approach. Despite often qualitative and not systematic, the available evidence nonetheless emphasises the central role played by human resources as a key success factor of international growth (Welch and

Luostarinen, 1988)5. In Italy, 60.7% of exporters, and 71.8% of consolidated exporters,

point out the lack of specialised personnel as the main constraint experienced in recent years (Esposito, 2001). Based on case studies from German manufacturing SMEs, Brüs-

4 For example, by examining a sample of 53 manufacturing companies, Minguzzi and Passaro (1993) ver-

ify the correlation between the entrepreneur’s openness to innovation and the growth of export activities. Based on a representative sample of 297 small Italian multi-national companies with FDIs, Mutinelli (2001) shows the relationship between firm size and motivations to the foreign investment. In particular, the propensity to market-oriented investments increases with size. On the contrary, smaller companies more frequently motivate their FDIs with the attempt to access material inputs and technologies. Besides firm size, however, the internationalisation strategy strongly correlates also with the industrial sector and the peculiarities of the destination country. 5 Once more, this topic has been more frequently studied in the case of large multi-national firms. Ex-

plored issues space from HRM in the international context (the so-called Strategic International Human Resource Management, Taylor et al., 1996), to the problem of knowledge transfer to and from foreign branches (Minbaeva et al., 2003), to the management of expatriates (Tan and Mahoney, 2003).

6

sig et al. (2001) recognise diversified patterns of internationalisation and suggest a posi-tive correlation between the approach to foreign activities and the competencies of the employees operating at the interface between the parent company and the foreign branch. Once more based on case studies, this time from the Netherlands and the USA, Prasad (1999) outlines the importance of foreign acquisitions as a means for compress-ing the costs and time undertaken by SMEs to learn new critical competencies. 2.3. Internationalisation processes and learning

According to the definition provided by Argyris and Schon (1996), organisational learn-ing takes place when the members of an organisation experience a problem and investi-gate it in the organisation interest. Competing in international markets involves contacts with new actors, exposes the company to higher variability, imposes new quality and innovation targets, and often also requires to diversify the company offer to meet spe-cific local requirements. The need to develop ad hoc tactics and strategies provides a “natural” opportunity of organisational learning, which originates from a process of cir-cular causation between the stock of tangible and intangible resources and the experi-ences accumulated in international markets (Caroli and Lipparini, 2002). On the one hand, available resources affect learning paths, because they constrain the accessible strategies to internationalisation. On the other one, the stock of available resources and competences constitutes not only the start-up of international trajectories, but also their result. By enacting learning mechanisms, the foreign experience increases the organisa-tional base of the firm, also affecting subsequent foreign investments. The learning cy-cles between the choice of going abroad and the increase of knowledge from the new environment turn the internationalisation process into an intrinsically uncertain and dy-namic path of learning (Lamb and Liesch, 2002). The development of new knowledge by the members of the organisation represents a mandatory, however not sufficient, condition to increase the resources of the firm by en-riching its knowledge base. As underlined by Argyris and Schön (1996), learning from organisational investigation turns into organisational learning when it becomes “embed-ded in the images of organisation held in its members’ minds and/or in the epistemo-logical artefacts embedded in the organisational environment. The expectation is that new knowledge, attitudes and skills will be acquired and applied in existing or new con-texts”. Consequently, also learning paths along international trajectories emphasise the importance of increasing the firm human capital (i.e. a resource for managing intra-firm organisational processes), as well as its relational capital, which concerns the relation-ships between the company and the external environment. The above outlined processes of internationalisation-related organisational learning ac-quire a special meaning in the case of SMEs. First, the progressive nature of the interna-tionalisation patterns at SMEs suggested by the stage theory directly relates to the spe-cific nature of learning processes. If the progressive enrichment of knowledge and com-

7

petencies is the solution adopted by SMEs to fill up their structural gap of resources in comparison with larger companies, the cumulative nature of learning processes con-strain the international growth along incremental paths. Second, as the network perspec-tive points out, learning processes at SMEs are often the collective result of interaction among different actors (Lamb and Liesch, 2002). This remark also suggests that an un-derstanding of learning processes due to international growth should take into account the relational network surrounding the firm. Third, some authors suggest the existence of a “threshold cultural level”, below which the firm lacks the capability to benefit from the increased relationships offered by the international context and fails to establish the virtuous circle between competencies and internationalisation (Minguzzi and Passaro,

1993)6. The risk to fall short of the minimal threshold – with the result of blocking, or

aborting, the internationalisation process – looks higher for SMEs, where the modes and rhythms of organisational learning are often entrepreneur-centred. Moreover, the typical channel chosen by small companies to start their internationalisation path, i.e. indirect export sales, structurally misses most of those relational stimuli required to activate fruitful learning processes.

3. Research hypotheses The main contribution of literature to the analysis of the internationalisation patterns followed by SMEs consists in recognising, beyond the common trends which distin-guish smaller from larger companies, the existence of differentiated paths also among the former. Despite the inconclusive nature of the available evidence, this hypothesis also extends to the area of human resource management. In order to qualify the attitude to HRM at the foreign branch of internationalised companies, the present paper resorts to two opposed growth mechanisms: resource exploiting and resource exploring. The first mechanism leverages on the human and relational capital accumulated and devel-oped at the parent company, while the second one bases on the internalisation of the human resources offered by the host country. The above mechanisms are well known and documented in literature. Resource exploit-ers heavily resort to the expatriation of managers and technicians in order to control their foreign activities by sharing explicit and tacit knowledge about procedures, tech-nologies, and company values with the management of the foreign subsidiary (Tan and Mahoney, 2003). In this case, the management of the foreign subsidiary mainly consists of expatriates with a relatively reduced discretional power. On the contrary, resource explorers grant their foreign managers – mostly indigenous employees – large auton-

6 The concept of “threshold cultural level” as a pre-requisite for organisational learning in international

markets echoes the concept of absorptive capacity introduced by Cohen and Levinthal (1989) to justify the different capabilities displayed by firms in imitating and developing their competitors’ innovations. Mahnke et al. (2003) explicitly refer to this concept for appraising the relationship between knowledge management tools and the performance achieved by the foreign branches of a large German multi-national company.

8

omy and encourage their contributions to enrich the organisational knowledge base (Minbaeva et al., 2003). In this second scenario, the importance recognised to the direct experience of markets and institutions in the host country motivates the recruitment process (Tan and Mahoney, 2003). Even if conceptually opposed, resource exploiting and exploring can in principle co-exist in the same company. The above recalled processes of circular causation at the base of the organisational learning can be interpreted as the continuous swing between the exploitation of existing resources and the internalisation of new ones. If resource exploitation is the initial solution for selecting the available strategies to grow abroad, resource exploration represents the mechanism to enrich the base of organisational knowledge and open up new opportunities. The paper maintains the general hypothesis that the propensity towards the exploitation or the exploration of human resources, hence the specific HRM solution adopted at the foreign subsidiary, reflects the management perception about the role played by human resources, but also the strategies which translate the company vision into actual organ-isational solutions. A proxy for characterising the firm vision has been recognised in its government model, i.e. the governance structures, norms, and values driving the com-

pany7. The analysis of the relationship between ownership and control allows to classify

governance models along a continuum spanning from the family-centred model, where ownership overlaps with control, to the managerial model, where owners delegate dis-cretional power to independent managers (Albertini, 2002). The family-centred model, characterised by the direct involvement of the entrepreneurs and their relatives in the company management, the extensive resort to direct supervision, and the strong de-pendence between operative decisions and entrepreneurial insight, looks intuitively cor-related with the exploitation of existing human resources. The resort to expatriate man-agers and technicians favours the direct control of the international activities and the sharing of the organisational knowledge. However, it also risks to constraint the explo-ration of new strategies and opportunities. On the contrary, the separation between ownership and management at the parent company is expected to favour the delegation of discretional power also at the foreign subsidiary and the emphasis on the internalisa-tion of externally available human resources. Consequently, the first hypothesis tested by the empirical analysis can be detailed as follows.

Hp1: Managerial governance models favour the exploration of the human resources

available in the foreign markets. Vice versa, family-centred governance models

are more frequently associated with the exploitation of existing human re-

sources.

7 For a survey of the meanings and interpretations given to the concept of governance, see Learmount

(2002).

9

The second hypothesis concerns the relationship between the degree of formalisation of the HRM system at the parent company and the HRM style at the foreign branch. De Kok and Ulhaner (2001) outline that the attempts of literature to define the formalisa-tion of the HRM system are still far from converging. However, growing degrees of formalisation are usually associated with more intense use of written procedures, regu-lar application of standard routines within the organisation, and control over employees. Our empirical analysis resorts to similar variables in order to qualify the nature of the HRM system at the sampled companies. Especially in the case of SMEs, the nature of the HRM system at the parent company may provide a useful tool for understanding the attitude towards HRM abroad. More of-ten than at larger companies, the competencies and know-how of SMEs display a tacit nature, directly embedded in the company products and routines. As a matter of fact, the exchange of knowledge in tacit forms reduces the need for formal HRM techniques. However, the switch to international competition, where physical distances impose the sharing of explicit and codified knowledge, risks to hamper the learning processes of those smaller companies still anchored to tacit models of knowledge creation and so-cialisation. The lack of formal tools to turn new experience into organisational learning could ban less endowed companies from a significant share of the potential benefits from their international activities. In particular, we expect the tools used for co-ordinating human resources at the parent company to display significant correlation with HRM practices abroad. The resort to sophisticated co-ordination tools is expected to accompany resource-exploring strate-gies for penetrating mature and demanding foreign markets. On the contrary, the exploi-tation of existing human resources looks more appropriate when the FDI targets lower labour costs or, more generally, lower input costs.

Hp2: The higher the formalisation of the HRM system at the parent company, the

more frequent the adoption of resource-exploring strategies abroad.

The governance model and the style of HRM look as promising concepts for under-standing the role played by human resources in the internationalisation process. How-ever, also the role of more traditional variables, such as firm size or age, should be taken into account. Their significant impact on the attitude to HRM abroad would support the thesis of a deterministic relationship between firm age, dimensional growth, and the de-velopment of a linear trajectory to internationalisation. However, the literature exam-ined in Section 2 suggests that “ideal” linear paths are often interrupted by leapfrogging, back steps, or sudden abandonment. Consequently, the empirical research explores a third hypothesis to appraise the explicative power of different proxies for the size of the parent company in comparison with its governance model and the formalisation of the HRM system.

10

Hp3: Size-related variables do not discriminate between the HRM practices of SMEs

and larger companies abroad.

4. Methodology

4.1. The questionnaire and the sample

The data to test the suggested hypotheses were collected by administering a structured questionnaire to a sample of internationalised firms from the Northern Italian provinces of Brescia, Bergamo and Verona. In order to point out the possible peculiarities of smaller companies, besides SMEs we also sampled large and medium-to-small firms. The investigation involved companies holding production facilities abroad, excluding purely commercial and logistic investments. This choice dramatically reduced the size

of the observed universe8. However, since production facilities abroad usually require

more articulated organisational structures, the focus on FDIs promised to offer clearer indications on the driving forces behind specific models of HRM. The questionnaire included four sections. The first one collected general information about the firm and data to identify the prevailing governance model. The second section examined the HRM at the Italian headquarters, while the third section explored the mo-tivations and the strategies driving the internationalisation process. Eventually, after asking the interviewed firms to focus on their most strategic foreign subsidiary, the last section concerned the management of human resources abroad.

The the database Reprint9 allowed to identify the observed universe. At the end of 2003,

Reprint reported 177 firms holding FDIs in the province of Brescia, 104 firms in the province of Bergamo, and 73 firms in the province of Verona for which an address was identifiable. After a preliminary telephone contact, the questionnaire was sent to candi-date firms by e-mail or fax, addressing either the entrepreneur, the personnel manager, or the employee responsible for HRM. A first round of interviews, limited to a sample of SMEs from the province of Brescia, was run in Fall 2003. Data collection was subse-quently extended to the remaining sampled firms between May and June 2004. In both rounds, firms were repeatedly solicited to fill up the questionnaire. We collected 57 questionnaires, with a response rate of 16.6%, 48 of which could be used in the following analyses (27 firms from Brescia, 12 from Bergamo and 9 from Verona). A follow-up survey of no respondents showed that most frequently reported motivations for missing answers were the economic difficulties currently faced by the

8 In the case of European SMEs, the data provided by the Observatory of European SMEs (Jakobsen and

de Voss, 2004) show that only 3% of firms hold at least one FDI.

9Reprint, developed and maintained at Politecnico di Milano and University of Brescia, represents the of-

ficial source of the reports on outwards and inwards Italian FDIs periodically released by ICE (the Italian Institute for Foreign Trade).

11

firm, the failure of the FDI initiative, and the vacation of the personnel manager due to mission abroad. The examined firms operate in different productive sectors, which nevertheless reflect the typical mode of foreign investment displayed by Italian companies, focused on tra-ditional and mechanical sectors (Mutinelli, 2001). Indeed, most of interviewed firms operate in the metal-mechanical industry (48%) and in the electric and electronic one

(15%)10

. Table 1 reports the descriptive statistics for the whole sample and the sub-samples com-prising the 31 SMEs and the 17 larger enterprises. On average, both at large companies and at SMEs the FDIs employ significantly less people than the parent company. Not surprisingly, most sampled firms accompany an intense export activity to their foreign operations. The most striking difference between large and small companies is the latter’s higher propensity to greenfield investments abroad. This behaviour probably derives from the lower initial resources required by this strategy, which meets the financial constraints often affecting SMEs. On the contrary, larger companies prefer acquisitions, a quicker solution to achieve the desired configuration abroad.

Table 1 – Descriptive statistics for the empirical sample

Whole sample

(N=48) SMEs (N=31)

Large firms (N=17)

Numerical variables µ σ µ σ µ σ

Firm size [employees] 273.23 356.7 115.39 64.76 561.06 479.75 Firm age [years] 44.30 27.68 38.00 22.68 55.41 32.64 FDI sizea) [employees] 87.34 10.51 65.53 72.72 125.82 132.58 FDI age a) [years] 7.93 6.57 7.38 5.57 8.88 8.09 Number of FDIs 2.8 2.05 2.60 2.34 2.25 1.39

Dummy variables [%] [%] [%]

Exports over 25% of turnover 83.33 80.77 81.25 Belongs to a group 68.09 54.84 93.75 Market seeking FDIsb) 97.92 96.88 100.00 Resource and labour seeking FDIs b) 96.88 78.13 58.82 Greenfield FDI a) 52.08 61.29 35.30 FDI towards industrialised countries a) 52.08 41.93 70.59

a) Refers to the most strategic FDI owned by interviewed firms b) Percentage of interviewed firms recognising this target as “important” or “very important” Despite a widespread focus on market-seeking strategies, SMEs see in FDIs also an im-portant source to access additional resources, namely lower costs for labour and mate-

10

Taking into account the comparatively low number of interviewed firms, the industry diversification could bias the empirical evidences. However, industry has not always proven a discriminating variable. For example, Knight (2001) studies the relationship between entrepreneurial orientation and performance in international markets in a sample of 268 SMEs from the USA, and finds no significant industry-related differences.

12

rial inputs. Not surprisingly, only 42% of the sampled SMEs (against 71% of large companies) located their most strategic FDIs in industrialised countries, where the high comparative cost of the production inputs motivates market-seeking rather than re-source-seeking strategies. 4.2. The empirical methodology

In order to investigate which factors affect the approach to HRM abroad, the empirical analysis first concentrated on developing a satisfactory proxy to discriminate between human resource exploitation and exploration. Given the comparatively reduced number of available observations, cluster analysis appears as the best statistic classification tools

for characterising the style displayed by the examined firms in HRM abroad11

. Variables from the fourth section of the questionnaire – the one investigating the HRM policies of the interviewed companies at their most strategic foreign subsidiary – offer the chance to group the sampled firms according to their attitude to HRM abroad. The classification resulting from the cluster analysis, from pure “controllers” of the for-eign investments to proactive “explorers” of the opportunities offered by the host coun-try, is assumed as a proxy for the attitude towards HRM abroad. This measure provides a useful dependent variable for testing the impact of the governance model (hypothesis 1), the HRM style at the parent company (hypothesis 2), and size (hypothesis 3) via econometric regressions. Given the ordinal nature of the dependent variable and the impossibility to quantify the “distance” between the subsequent values it takes, ordered models look as the best func-tional form to run the econometric estimates (Green, 1993, p.672). Ordered regressions model the dependence of a polytomous ordinal response on a set of either numerical or categorical predictors. In particular, the ordered logit model estimates the effects of in-dependent variables on the log odds of having lower rather than higher scores on the dependent variable. If the ordered dependent variable Y can assume J distinct values, the relationship between the log odds ratio and the K independent regressors xk can be expressed in the following way:

( )( ) �

=

−=��

�

�

��

�

�

>≤ K

kkkj

k

kX

XjYp

XjYp

1

ln βα , for j = 1 to J-1,

where αj are the intercepts indicating the log odds of lower rather than high scores when

all independent variables are equal to zero and βk represents the change in the log odds

corresponding to a unit increase in xk12

.

11

Cluster analysis is a multivariate procedure for detecting natural groupings in data when neither the number nor the members of the subgroups are known (Heir et al., 1996). 12

Consequently, keβ represents the odds ratio corresponding to a unit increase in xk, and is called the odds ratio of xk. Since odds ratios of independent variables do not depend on j, the ordered logit model is also called proportional odds model (Rudolfer et al., 1999).

13

Hypothesis 3, which concerns the impact of size-related factors on the attitude towards HRM abroad, has been tested by running analyses of variance (ANOVA) with the or-dered clusters as the discriminating factor and firm age and firm size as the explored variables.

4.3 The attitude to HRM abroad The cluster analysis allowed to identify the different attitudes to HRM abroad displayed by the 48 sampled firms. The primitive variables used for running the cluster analysis (Table 2) primarily concern the policies followed to fill up the empty positions at the foreign subsidi-ary. QUAL_ABROAD identifies the companies which hired local qualified personnel (technicians or executives), while EXPAT concerns the expatriation of Italian employees. In order to capture the firm capability to use the FDI as a source of organisational knowledge, the variable LEARN apprises the transferability of accumulated experience to subsequent FDIs. Eventually, by evaluating who managed the FDI stat-up, the variable START ap-

prises the role of HRM abroad at the beginning of the foreign venture.

Thanks to the possibility of choosing ex-post the final number of clusters, we preferred hierarchical analysis over the K-means procedure. The performed cluster analysis re-sorted to average linkage between groups as the clustering algorithm and on squared Euclidean distances as distance measures. We checked for the presence of outliers, the representativeness of the examined sample, and the correlation among the variables en-tered in the clustering algorithm. In the latter case, the low and/or poorly significant cor-relation indexes allow to exclude the existence of multicollinearity. Moreover, in order to avoid the risk of overestimating the impact of more widely dispersed measures, all the primitive variables entered the clustering algorithm after standardisation in the range 0 to 1.

Table 2 – Primitive variables entering the cluster analysis Variable Description Min Max µ σ QUAL_ABROAD Recruitment of qualified foreign employees (technicians

or executives) abroad 0 1 0.458 0.504

EXPAT Expatriates from the parent company 0 1 0.438 0.501 LEARN Capability to transfer the acquired experience to subse-

quent FDIs 0: Experience hardly transferable; 1: Experience transferable only to similar countries; 2: Experience transferable

0 2 0.688 0.879

START Delegation of decisional power at the FDI start-up 0: entrepreneur in charge; 1: Family manager in charge; 2: Italian employee in charge; 3: Foreign employee in charge

0 3 1.292 0.617

48 Observations The cluster analysis identified four mutually exclusive groups, whose members share a common attitude towards HRM abroad. With the purpose of characterising this attitude, the 4 clusters were ordered according to the growing propensity to explore local human resource, labelling the subsequent groups as “Controllers”, “Exploiters”, “Mixers”, and

14

“Explorers”. The average value and the standard deviation assumed by the primitive variables for each cluster are reported in Table 3. Table 4 further characterises the 4 clusters by exploring the average values assumed by significant variables from the ques-tionnaire. Explorers and Exploiters show the largest differences in hiring policies: while the first cluster only hires employees from the foreign country, the second one only resorts to the expatriation of Italian personnel. Between Explorers and Exploiters, the cluster of Mix-ers adopts both policies. Controllers display a “passive” attitude towards human re-sources abroad, delegating all recruitment issues to the foreign subsidiary. Not surpris-ingly, Controllers mostly started their FDI via acquisition, internalising an already exist-ing organisation. Moreover, Controllers also characterise for the comparatively low interest to invest in industrialised countries, which usually require more qualified human capital. On the contrary, 3 out of 4 Explorers located their most strategic FDI in an industrialised coun-try.

Table 3 – The 4 Clusters: descriptive variables

Variable Cluster Controllers

N = 11 Exploiters

N = 15 Mixers N = 10

Explorers N = 12

µ σ µ σ µ σ µ σ QUAL_ABROAD 0.000 0.000 0.000 0.000 1.000 0.000 1.000 0.000

EXPAT 0.000 0.000 1.000 0.000 1.000 0.000 0.000 0.000

LEARN 1.333 0.617 1.000 0.000 1.400 0.699 1.417 0.793 START 0.667 0.900 0.455 0.820 0.400 0.699 1.167 0.937

Table 4 – Descriptive statistics for the 4 Clusters

Variable Cluster Controllers

N = 15 Exploiters

N = 11 Mixers N = 10

Explorers N = 12

µ σ µ σ µ σ µ σ Firm size [employees] 184,87 112.45 304,00 412.33 218,50 247.11 401,08 540.09

Firm age [years] 40,20 23.76 48,09 22.49 38,30 28.10 51,55 37.19

FDI sizea) [employees] 93,14 82.24 122,36 160.17 80,90 83.93 53,30 59.03

FDI age a) [years] 10,38 9.25 8,82 6.23 8,20 6.80 8,08 2.50 Number of FDIs 2,85 2.94 2,45 1.63 2,50 2.12 2,08 1.16

[%] [%] [%] [%]

Exports over 25% of turnover 92 80 78 73

Market seeking FDIsb) 93 100 100 100

Resource and labour seeking FDIs b) 73 91 70 50 Greenfield FDI 40 55 70 55 FDI towards industrialised countries 40 55 40 75 Belongs to a group 90 82 89 71

a) Refers to the most strategic FDI owned by interviewed firms b) Percentage of interviewed firms recognising this target as “important” or “very important”

15

As expected, the Exploiters’ resort to expatriation results into the lowest capability of transferring the acquired experience to subsequent FDIs. On the contrary, both Mixers and Explorers are more confident about the transferability of their experience abroad. The positive correlation detected by the cluster analysis between the attitude to hire abroad and the capability of benefiting from previous FDI experience suggests that when firms feel ready to internalise external human resource, they can also turn their experience into explicit, hence transferable, routines and procedures. Also the delega-tion of discretional power during the start up of the FDI is higher for Explorers and lower for Exploiters, confirming the propensity of the second cluster to replicate abroad the control exerted at the parent company. Even if Explorers show a higher average size, both SMEs and larger enterprises are pre-sent in all the four clusters in similar proportions. This result provides preliminary evi-dence in support of our third research hypothesis, which concerns the low significance of firm size as a discriminating factor among HRM practices abroad. Taken together, these characteristics suggest that Explorers resort to hiring qualified foreign personnel at the foreign subsidiary, to the delegation of discretional power dur-ing the FDI start-up and to the capability of replicating the accumulated experience. On the contrary, Exploiters priviledge expatriation policies from the Italian parent company and strict control at the start up of foreign subsidiary. The two approaches coexist among Mixers, while controllers are the extreme case where the firm belongs to a group and the parent company only marginally participates in HRM abroad. 4.4 The explanatory variables

Table 5 reports the description and the main statistics for the independent variables used to test the research hypotheses. The choice of the explanatory variables reflects the at-tempt to capture, respectively, the governance model of the sampled firms, the HRM style at the parent company, and some critical FDI- and firm-specific issues. In principle, the involvement of younger members of the entrepreneurial family among the employees (YOUNG), as well as the participation of family members in the com-pany management (MNGR_REL) reflect the prevalence of a family-centred governance model, rather than a managerial one. An additional variable to characterise the govern-ance model is represented by the degree of delegation in HRM issues (HRM_RESP). Literature has often underlined the symbolic value attached by entrepreneurs to the HRM function (see e.g. de Kok et al., 2003) and the importance of its delegation to complete the switch to a fully managerial model. The HRM style at the parent company is characterised through the attitude towards training (TRNG_NEW, TRNG_RESOURCE, TRNG_PROPOSAL), the existence of a

16

firm-level labour agreement13

(FIRM_AGR), the recruitment policy (RECR_AG), and the importance attached to formal tools for performance appraisal (IDX). The underly-ing idea is that the resort to explicit rules and routines to govern the employees’ train-ing, the bargaining over employment conditions at the company level, the capability to share the company recruitment needs with an external agency, as well as the structured evaluation of the employees’ performance all increase the degree of formalisation in HRM. Moreover, in order to check the coherence between the attitude to HRM abroad and the actual discretional power enjoyed by foreign managers, we introduced MGT_AUT, which measures the decisional autonomy of management at the foreign subsidiary. Control variables primarily check the impact of firm size (LG_SIZE). In order smooth the weight of the largest companies, as well as to facilitate the comparison between es-timated coefficients, firm size enters the empirical estimates as a logarithmic variable. An alternative measure of the firm size is supplied by firm age (LG_AGE), which prox-ies for the organisational knowledge accumulated by the company since it started. If the company followed a linear growth path, firm age should also represent a reasonable proxy for the company size. Also firm age enters the econometric regression in loga-rithmic form. Moreover, control variables also take into account the FDI origins (the acquisition of an already existing organisational structure is expected to favour lower control by the par-ent company) and the complementary activities performed at the foreign subsidiary. In line with the average behaviour displayed by Italian outwards FDIs, only 6 firms de-clared the delocalisation of design, project, research or development activities. How-ever, 34 out of the 48 examined FDIs accompany their operations with sales and mar-keting activities. We assume that the need to deal with local clients drives the parent company towards the internalisation of human resource abroad, thanks to their better knowledge of local markets and institutions. The last independent variable (CONSULT) verifies the relationship between the attitude to HRM abroad and the resort to external consultancy services at the launch of the FDI. The need for external support could sig-nal the lack of proper organisational and cultural tools to face the foreign investment, hence a more conservative approach.

13

The Italian law sets two levels for bargaining over labour contracts. At the national level, trade unions and employers unions define the normative contents of national labour contracts by industry and set the minimum wage. At the firm level, an additional bargaining process between local trade unions and man-agement may result in a firm-level contract which integrates the national industry agreement and sets the rules for the employees’ variable wage.

17

Table 5 – The independent variables

Variable Description Min Max µ σ

Government model MNGR_REL At least one relative of the entrepreneur(s) among managers 0 1 0.500 0.505 YOUNG At least one younger relative of the entrepreneur(s) among

employees 0 1 0.596 0.496

HRM_RESP Company position in charge of HRM 0: Entrepreneur; 1: Employee; 2: Family manager; 3: Non-family manager

0 3 1.521 1.238

HRM at the parent company TRN_NEW Training programmes for new employees

0: No training programmes; 1: For managers only; 2: For all employees

0 2 1.468 0.830

TRN_RESOURCES Training also based on internal resources 0 1 0.313 0.468 TRN_PROPOSAL Employees can propose training initiatives 0 1 0.646 0.483 FIRM_AGR Firm-level labour agreement 0 1 0.187 0.394 RECR_AG Hiring also based on recruitment agencies 0 1 0.375 0.489 IDX Employees’ performance monitored via formal indicators

0: No indicators; 1: For managers only; 2: For all employees 0 2 0.822 0.888

MGT_AUT Management autonomy at the subsidiary 0: Most important operative decisions need authorisation; 1: All strategic decisions need authorisation; 2: Most strategic decisions need authorisation; 3: Large autonomy

0 3 1.478 1.005

Control variables LG_SIZE Logarithm of firm size (number of employees) 2.303 7.587 5.096 1.031 LG_AGE Logarithm of firm age (years) 1.386 4.754 3.559 0.736 ACQUISITION FDI started as an acquisition 0 1 0.532 0.504 MKT&SALES Foreign branch in charge for marketing & sales activities 0 1 0.708 0.459 CONSULT External consultants accompanied the launch of the FDI 0 1 0.560 0.501

48 Observations

5. The empirical analysis The regressions to test the three research hypotheses were run by using SPSS 11.0, which models the ordinal regression through the PLUM methodology, derived from McCullagh (1980). The correlation matrix (Table 6) reveals comparatively high correla-tion indexes between LG_SIZE and FIRM_AGR, which consequently entered separate estimate models (respectively, Model 1 and Model 2 in Table 7). The results of the empirical estimates are synthesised in Table 7. All estimated models present a good fit with data and a highly satisfactory explicative power (Nagelkerke R2 ranges from 76% of Model 1 to 74.3% of Model 2). The ordinal regressions confirm the significant relation between the governance model and the attitude towards HRM abroad (Hypothesis 1). Managerial models tend to couple with an explorative attitude, while family-centred models prevail among Controllers and Explorers. In particular the presence of family managers discourages the adoption of an

18

exploratory attitude (coefficient of MNGR_REL negative and significant), while the delegation of HRM direction to a non-family manager is more probably adopted by Ex-plorers. Note that only the non-family HRM director determines a positive increase of the probability to be an Explorer in comparison with the baseline (i.e., when the entre-preneur is also the HR manager). On the contrary, intermediate forms of delegation , to an employee controlled by the entrepreneur, or to a family manager, are more probably associated with a passive or conservative attitude towards HRM abroad. Contrary to expectation, the coefficient of YOUNG is positive and significant: the in-volvement of the younger generation in the company business, despite perpetuating the family-centred model of governance, increases the log odds of higher scores on the de-pendent variable. However, besides a tendency to the family-centred model, YOUNG could also signal the presence of new business ideas coming from the younger genera-tion, as well as at least partial delegation of leadership and control by the entrepreneur. Both elements would explain the positive impact of the probability to adopt an explora-tory attitude towards HR abroad. Hypothesis 2 stated that a more formalised HRM style at home should couple with less control abroad. The regressions show that a certain degree of formalisation helps the adoption of a more exploratory attitude. For example, the capability to detail the com-pany HR needs to an external recruitment agency (RECR_AG) displays a positive and significant impact on the probability of being an Explorer. In a similar manner, the at-tention to formal training for new employees (TRNG_NEW) and continuous training (TRN_RESOURCE and TRN_PROPOSAL) are positively associated with the log odds of higher scores for the dependent variables. Also the explicitation of the rules and in-centive schemes stated by the firm-level labour agreement (FIRM_AGR) has a signifi-cant and positive effect over the exploration of human resources abroad. However, the HRM system at the parent company should also be flexible enough to grab the benefits of more open attitude. The acceptance of training suggestions also from employees (TRNG_PROPOSAL) positively affects the probability of being an Explorer. At the same time, the use of formal indicators to monitor the management performance (IDX) has a negative coefficient: too tightly controlled managers at home correspond to less autonomous foreign branches. Not surprisingly, only when foreign manager are granted high degrees of autonomy the foreign subsidiary displays a more explorative attitude towards human resources (MGT_AUT). The estimated regressions fully support Hypothesis 3: firm’s size in terms of both em-ployees number (LG_SIZE) and accumulated experience (LG_AGE) positively effect the probability of adopting an exploration-oriented attitude to HRM abroad (Model 1). the ordinal regressions confirm that SMEs adopt all the outlined behaviours to manage their human resources abroad: if significant differences exist among Explorers, Mixers, Exploiters, and Controllers, they are not size-related.

19

Table 6 – The correlation matrix

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.

1. 4_CLUSTERS 1

2. MNGR_REL 0.054 1

3. YOUNG -0.106 0.338** 1

4. HRM_RESP -0.060 -0.085 -0.227 1

5. TRN_NEW -0.291** -0.124 -0.214 0.163 1

6. TRN_RESOURCES 0.079 0.135 -0.023 -0.007 0.244* 1

7. TRN_PROPOSAL -0.176 -0.044 -0.184 0.279* 0.230 -0.123 1

8. FIRM_AGR -0.071 0.160 -0.298** 0.248* 0.250* 0.252* 0.202 1

9. RECR_AG -0.180 -0.086 -0.131 0.163 0.314** -0.035 0.034 0.041 1

10. IDX 0.030 0.192 -0.030 0.077 0.104 0.292** -0.118 0.075 0.220 1

11. MGT_AUT -0.237 0.000 0.176 -0.144 -0.065 -0.051 -0.053 0.060 -0.226 -0.099 1

12. LG_SIZE 0.082 0.009 0.252* -0.258* -0.208 -0.101 -0.061 -0.397*** -0.116 -0.088 -0.097 1

13. LG_AGE 0.023 0.061 0.232 0.026 -0.055 0.031 -0.119 -0.354** 0.196 0.029 -0.183 0.370*** 1

14. ACQUISITION 0.146 0.065 -0.228 -0.181 0.184 -0.042 0.134 0.240 -0.037 -0.027 -0.078 -0.276** -0.212 1

15. MKT&SALES 0.021 0.367*** 0.077 0.064 -0.007 0.235 -0.004 -0.044 -0.213 -0.042 -0.111 0.107 0.126 -0.238 1

16. CONSULT 0.119 -0.042 -0.192 0.139 0.091 0.051 0.214 -0.007 0.098 -0.201 -0.152 -0.131 0.078 -0.771 0.081 1

* significant at the 10% level; ** significant at the 5% level; *** significant at the 1% level

20

Table 7 – The ordinal regressions Model 1 Model 2

β Std error β Std error Threshold

Controllers 8.355** 4.142 10.727*** 3.596 Exploiters 11.330*** 4.460 13.706*** 3.998 Mixers 13.254*** 4.564 15.453*** 4.138 Explorers (baseline)

Governance model

MNGR_REL -1.965 1.531 - 3.131** 1.430

YOUNG 3.584** 1.531 4.802*** 1.543

HRM_RESP: External manager 4.166** 1.787 4.638*** 1.773 HRM_RESP: Entrepreneurs' relative -4.122** 1.872 - 3.198* 1.727 HRM_RESP: Employee -3.931** 1.946 -4.192** 1.802 HRM_RESP: Entrepreneur (baseline)

HRM at the parent company

TRN_NEW: Training for all 4.891** 1.968 4.571*** 1.742 TRN_NEW: Training for new managers only 16.891*** 4.969 17.368*** 4.758 TRN_NEW: No training for new employees (baseline)

TRN_RESOURCES 3.026** 1.319 3.615*** 1.324

TRN_PROPOSAL 4.504*** 1.417 4.638*** 1.390

FIRM_AGR - - 3.768*** 1.412

RECR_AG 3.543*** 1.379 - -

IDX: performance indicators for all employees 0.306 1.223 1.265 1.117 IDX: performance indicators for managers only -3.801*** 1.686 -2.929** 1.348 IDX: no performance indicators (baseline)

MGT_AUT: large autonomy for foreign management 6.940*** 2.634 7.295*** 2.409 MGT_AUT: authorisation for most strategic decisions - 0.325 1.500 0.214 1.219 MGT_AUT: authorisation for all strategic decisions 2.053 1.837 2.416 1.663 MGT_AUT: authorisation for main operative decisions (baseline)

Control variables

LG_SIZE - 1.092 0.890 - -

LG_AGE 0.828 0.594 - -

ACQUISITION -3.405*** 1.312 -2.187** 0.986

MKT&SALES 2.627** 1.279 2.903** 1.275

CONSULT -7.113*** 2.104 -6.435*** 1.997

N 48 48

-2 Log Likeliwood 68.897*** 54.576***

Nagelkerke R2 0.760 0.743

Link function: Logit * significant at the 10% level; ** significant at the 5% level; *** significant at the 1% level

Also the investment-specific control variables carry interesting results. The positive co-efficient of MKT&SALES confirms that the direct contact whit local clients encourages the internalisation of locally available human resources. Moreover, the negative coeffi-cient of ACQUISITION supports the intuition that when the FDI results from an acqui-

21

sition, the parent company will privilege the formal control of the foreign subsidiary. However, the low involvement into HRM issues abroad also limits the incentives to ex-plore the knowledge transfer opportunities provided by the FDI experience. Accor4ding to expectations, CONSULT shows a negative coefficient: the resort to external counsel-ling signals the lack of the organisational capability to properly exploit the human re-sources available in the foreign context.

6. Concluding remarks Wright and Brewster (2003) suggest an effective metaphor to stress the difficulty of generalising the application of HRM techniques: “HRM is not lycra”. Also in the case of international growth, rather than looking for the “optimal model”, it is interesting to show how different firms, displaying a sort of “equifinality” (Katz and Kahn, 1966), can attain similar results also through extremely diversified paths. The reduced size of the examined sample imposes caution in generalising the obtained evidence. Some interesting results however emerge. First, cluster analysis has outlined important differences among the HRM models adopted by internationalised firms. Con-trollers, Exploiters, Mixers, and Explorers show significantly different behaviours in term of recruitment policies abroad, capability to transfer the acquired experience to subsequent FDIs, and power delegation at the FDI start-up. Second, similarity in firm size is not enough to equal SMEs to a homogeneous universe, also when dealing with internationalised firms and the role of human resources abroad. The ordinal regressions report no significant influence of firm size over the probability of adopting an exploratory rather than exploiting attitude towards HRM at foreign sub-sidiaries. Moreover, also the knowledge base accumulated at the parent company, proxied for by the age of the parent company, does not apparently influence the choice of HRM policies abroad. These are the most innovative result of the paper, which de-serve further analysis and extension to wider and possibly international empirical sam-ples. Third, also after controlling for size effects, the governance model and the HRM style at home matter. In general terms, managerial modes of governance and formalised HRM systems favour exploration-oriented behaviours abroad. However, a careful interpreta-tion of these results is needed. As for the governance model, the degree of power dele-gation outside the entrepreneurial family plays an important role for explaining the adoption of an exploring behaviour towards human resources abroad. Nonetheless, the econometric tests show that the involvement the entrepreneurial family’s younger gen-eration in the business positively affects the probability of being an Explorer. This evi-dence suggests that the company attention towards a diversified range of growth oppor-tunities, which is compatible with various governance models, could be a more effective variable for discriminating among different attitudes to HRM abroad. In the case of the HRM style at the parent company, the empirical analysis suggest that if more structured

22

practices favour indirect control, hence power delegation, also at the foreign subsidiary, too tight rules shift the emphasis to direct control and privilege the controller or the ex-ploiter model over the explorer one. The above empirical findings hold significant consequences. Firms are increasingly conscious that HRM practices abroad are becoming a critical source of competitive ad-vantage in global as well as multi-domestic markets for both larger and smaller multi-national enterprises and this applies to small as well as large multi-national enterprises (Schuler et al., 2002). However, our results stress the importance of developing a coher-ent approach including not only HRM practices and policies at the foreign subsidiary, but also the governance model and the HRM style at the parent company. The empirical results suggest some interesting implications also for the ever expanding private and public offer of services to support the internationalisation of firms. Differentiated pat-terns of growth in foreign markets imply diversified training needs, also in the case of SMEs. Consequently, training programmes in the field of internationalization should avoid addressing companies mainly on size terms. Initiatives should be tailored on more detailed features than size only, including the governance model and the HRM system at the parent company.

Bibliographical References Albertini, S. (2002). Modelli di corporate governance e gestione delle risorse umane

nelle PMI distrettuali, XVII Convegno AIEL, Salerno, 26-27 September. Aldrich, H.E. and Langton N. (1997) “Human Resource Management Practices and Or-

ganizational Life Cycles”, in Paul D.R. et al. (eds.), Frontiers of Entrepreneurship

Research, Babson Park, Mass.: Babson College, 349-357. Argyris, C. and Schön, D. (1996). Organizational learning II: Theory, method and

practice, Reading, Mass.: Addison Wesley. Arthur, M. and Hendry C. (1990). “HRM and the emergent strategy of small to medium

sized business units”, International Journal of Human Resource Management, 1/3, 233-250.

Audretsch, D.B. and Thurik, A.R. (2000). “Capitalism and Democracy in the 21st cen-tury: From the Managed to the Entrepreneurial Economy”, Journal of Evolutionary

Economics, 10/1, 17-34. Audretsch D.B., Thurik A.R. (2001). “What is New about the New Economy: Sources

of Growth in the Managed and Entrepreneurial Economies”, Industrial and Corpo-

rate Change, 10/1, 267-315 Axelsson, B. and Easton, G. (eds.) (1992). Industrial networks: a new view of reality,

Routledge: London. Brüssig, M., Gerlach, L. and Wilkens, U. (2001). The development of globalisation

strategies in SME and the role of human resorce management, Human Resources

23

GLOBAL Management Conference. Comparative HRM - Learning from Diversity, Barcelona, 19-22 June.

Buckley, P.J. (1989). “Foreign Direct Investment by Small and Medium Sized Enter-prises: The Theoretical Background”, Small Business Economics, 1, 89-100.

Caroli, M. and Lipparini, A. (2002). Piccole imprese oltre il confine, Roma: Carocci E-ditore.

Cohen, W. and Levinthal, D. (1989). “Innovation and learning: the two faces of R&D”, Economic Journal, 99, 569-596.

Coviello, N. and McAuley, A. (1999). “Internationalisation Processes and the Smaller Firm: A Review of Contemporary Research”, Management International Review,

39/3, 223-256. Daft, R.L. (1998). Organization Theory and Design, 6th ed., New York: West Publish-

ing Co. de Kok, J., Uhlaner, L. and Thurik, R. (2003). Human Resource Management with

Small and Mediumsized Enterprises: Facts and Explanations, Centre for Advanced Small Business Economics (CASBEC), Rotterdam: Erasmus University.

de Kok, J. and Uhlaner, L.M. (2001). “Organization Context and Human Resource Management in the Small Firm”, Small Business Economics, 17, 273-291.

Enright, M.J. (1998). “Regional Clusters and Firm Strategy”, in Chandler A.D. (ed.), The Dynamic Firm: The Role of Technology, Strategy, Organization and Regions, Oxford: Oxford University Press.

Esposito, G. F. (2001). Internazionalizzazione e territorio, Roma: Assocameraestero. European Commission (2003). Commission Recommendation of 6 May 2003 concern-

ing the definition of micro, small and medium-sized enterprises, 2003/361/EC, Brussels, 6 May 2003.

Gankema, H.G., Snuit, H.R. and Van Dijken, K.A. (1997). “The internationalisation process of small and medium sized enterprises: An evaluation of the stage theory”, in Donckels, R. and Miettinen, A. (eds.), Entrepreneurship and SME Research: On

its Way to the Next Millenium, Aldershot: Ashgate Publishing, 185-197. Gillespie, A, Richardson, R. and Cornford, J. (2001). Regional Development and the

New Economy, European Investment Bank Papers, 6/1, 109-131. Hair, J.F., Anderson, R.E. and Tatham, R.L., Black W.C. (1996). Multivariate Data

Analysis with Readings, Englewood Cliffs: Prentice Hall.

Heneman, R.L., Tansky, J.W. and Camp, S.M. (2000). “Human Resource Management practices in small and medium-sized enterprises: unanswered questions and future research perspectives”, Entrepreneurship theory and practice, 25/1, 11-26.

Jakobsen, L. and de Voss, V. (2004). “Internationalisation of SMEs”, Observatory of European SMEs 2003 No. 4, European Commission, Luxenbourg.

Johanson, J. and Vahlne, J. (1977). “The internationalisation process of commitments”, Journal of International Business Studies, Spring/Summer, 23-32.

24

Knight, G.A. (2001). “Entrepreneurship and Strategy in International SMEs”, Journal of

International Management, 7, 155-171. Kundu, S.K. and Katz, J.A. (2001). “Born international SMEs: Bi-Level Impacts of Re-

sources and Intention”, Small Business Economics, 20, 25-47. Lamb, P.W. and Liesch, P.W. (2002). “The Internationalization Process of the Smaller

Firm: Re-framing the Relationships between Market Commitment, Knowledge and Involvement”, Management International Review, 42/1, 7-26.

Learmount, S. (2002). Theorizing Corporate Governance: New Organizational Alterna-

tives, Working Paper No. 237, ESRC Centre for Business Research, Judge Institute for Management, Cambridge: University of Cambridge.

MacDuffie, J.P. (1995). “Human Resource Bundles and Manufacturing Performance: Organizational Logic and Flexible Production Systems in the World Auto Indus-try”, Industrial and Labor Relations Review, 197-221.

Mahnke, V., Pedersen, T. and Verzin, M. (2003). The impact of knowledge management

on MNC subsidiary performance: the role of absorptive capacity, CKG Working Paper: 2003-10.

Mariotti, S. and Mutinelli, M. (2003). Le vie di internazionalizzazione delle PMI, Mila-no: Il Sole 24 Ore.

Marlow, S. and Patton, D. (1993). “Research Note-Managing the Employment Rela-tionship in the Smaller Firm: Possibilities for Human Resource Management”, In-

ternational Small Business Journal, 11/4, 57-64. March, J.G. (1991). “Exploration and Exploitation in Organizational Learning”, Or-

ganization Science, 2/1, 71-87. McCullagh, P. (1980). “Regression models for ordinal data”, Journal of the Royal Sta-

tistical Society - Series B, 42, 109-142. McDougall, P.P and Oviatt, B.M. (2000). “International Entrepreneurship: the Intersec-

tion of Two Research Paths”, Academy of Management Journal, 43/5, 902-906. Minbaeva, D., Pedersen, T., Bjorkman, I., Fey, C. and Park, H. (2003). “MNC Knowl-

edge Transfer, Subsidiary Absorptive Capacity and Knowledge Transfer”, Journal

of International Business Studies, 34, 530-549. Minguzzi, A. and Passaro, R. (1993). “Cultura imprenditoriale e processi di internazio-

nalizzazione”, Sviluppo & Organizzazione, 139. Mutinelli, M. (2001). “L’internazionalizzazione produttiva delle imprese italiane: carat-

teristiche strutturali, tendenze evolutive e motivazioni”, in Garavello O., Barba Na-varetti G. (eds.). Investimenti esteri delle imprese italiane nei paesi emergenti, Mi-lano: Giuffrè Editore.

OECD (1997). Globalization and Small and Medium Enterprises (SMEs). Paris: Or-ganization for Economic Cooperation and Development.

Oviatt, B.M. and McDougall, P.P. (1994).”Toward a Theory of International New Ven-tures”, Journal of International Business Studies, 25/1, 45-64.

25

Porter, M.E. (1998). “Cluster and the New Economy of Competition”, Harvard Busi-

ness Review, November-December. Prasad, S.B. (1999). “Globalization of Smaller Firms: Field Notes on Processes”, Small

Business Economics, 13. 1-7. Schuler, R.S., Budhwar, P. S. and Florkowski, G. W. (2002). “International human re-

source management: review and critique”, International Journal of Management

Reviews, 14, 41-70. Rudolfer, S.M., Paliouras, G. and Peers, I.S. (1999). “A comparison of logistic regres-

sion to decision tree induction in the diagnosis of Carpal Tunnel Syndrome”, Com-

puters and Biomedical Research, 32, 391-414.

Spigarelli, F. (2003). “I processi di internazionalizzazione delle PMI. Un’analisi em-pirica sul contesto produttivo marchigiano”, Economia & Management, 3.

Storper, M. (1997). The Regional World, New York: Guildford Press. Tan, D., Mahoney, J. (2003). “Explaining the utilization of managerial expatriates from

the perspectives of resource-based, agency, and transaction costs theories”, Ad-vances in International Management, 15, 179-205.

Taylor, S., Beechler, S. and Napier, N. (1996). “Towards an integrative model of strate-gic international human resource management”, Academy of Management Review, 21, 959-85.

Welch, L. and Luostarinen, R. (1988). “Internationalization: Evolution of concept”, Journal of General Marketing, Winter, 34-57.

Wright, P. and Brewster, C. (2003). “Learning from diversity: HRM is not Lycra”, In-

ternational Journal of Human Resource Management, 14, 1299-1307.