«Keskusteluaiheita – Discussion papers No. 1082 Mikko Ketokivi – Jyrki Ali-Yrkkö DETERMINANTS OF MANUFACTURING-R&D CO-LOCATION Financial supports ...»
ETLA THE RESEARCH INSTITUTE OF THE FINNISH ECONOMY
Lönnrotinkatu 4 B 00120 Helsinki Finland Tel. 358-9-609 900
Telefax 358-9-601 753 World Wide Web: http://www.etla.fi/
Keskusteluaiheita – Discussion papers
Mikko Ketokivi – Jyrki Ali-Yrkkö
DETERMINANTS OF MANUFACTURING-R&D
CO-LOCATION Financial supports from Finnish Funding Agency for Technology and Innovation (TEKES) and ETLA-BRIE project are gratefully acknowledged.
ISSN 0781-6847 30.03.2007 Ketokivi, Mikko ja Ali-Yrkkö, Jyrki. DETERMINANTS OF MANUFACTURING-R&D CO-LOCATION. Helsinki, ETLA, The Research Institute of the Finnish Economy, 2007, 28 p. (Keskusteluaiheita, Discussion Papers; ISSN 0781-6847; no. 1082).
ABSTRACT: The phenomenon of physical R&D-manufacturing co-location is interesting, because researchers have made very different observations regarding its prevalence. In some populations co-location of the two functions seems to be the norm; in others, an exception. However, we still do not have an explicit explanatory theory of co-location. In this paper, we look the reasons why manufacturing and R&D may have to be physically co-located. In a sample of 241 Finnish industrial firms, we find that the need for co-location varies drastically from company to company. We further find that product complexity, process complexity and industry clockspeed have an effect on co-location need.
KEY WORDS: Business location decisions, co-location, R&D, manufacturing, contingency theory JEL CODES: D21, D23, F21, F23, L6, O32 Ketokivi, Mikko ja Ali-Yrkkö, Jyrki. MITKÄ TEKIJÄT VAIKUTTAVAT SIIHEN, ETTÄ VALMISTUS JA T&K-TOIMINTA SIJOITTUVAT SAMOILLE ALUEILLE? Helsinki, ETLA, Elinkeinoelämän Tutkimuslaitos, The Research Institute of the Finnish Economy, 2007, 28 s. (Keskusteluaiheita, Discussion Papers; ISSN 0781-6847; no. 1082).
TIIVISTELMÄ: Tuotanto ja t&k-toiminta näyttävät usein sijoittuvan samoille maantieteellisille alueille. Aina näin ei kuitenkaan ole. Tässä tutkimuksessa tarkastellaan sekä teoreettisesti että empiirisesti tekijöitä, jotka vaikuttavat tuotanto- ja t&k-toimintojen yhteissijoittumisen tarpeeseen. Aineisto koostuu 241 Suomessa toimivasta teollisuusyrityksestä. Tilastollisen analyysin tulosten mukaan t&k-ja tuotantotoiminnan yhteissijoittumisen tarve vaihtelee huomattavasti. Yhteissijoittumisen tarve lisääntyy, mitä monimutkaisempi yrityksen tuote tai tuotantoprosessi on.
Myös jatkuva uusien tuotteiden valmistuksen aloittaminen ja vanhojen tuotteiden lopettaminen lisää yhteissijoittumisen tarvetta.
AVAINSANAT: sijaintitekijät, yhteissijainti, t&k, tuotekehitys, valmistus, tuotanto JEL-KOODIT: D21, D23, F21, F23, L6, O32
1.1 APPROACHES TO LOCATION DECISIONS
1.1.1 The External Drivers View
1.1.2 The Intra-Functional Dependencies View
1.1.3 The Cross-Functional Dependencies View
1.2 TENSIONS AND TRADEOFFS BETWEEN THE THREE VIEWS
2. ORGANIZATIONAL DETERMINANTS OF CO-LOCATION
3. EMPIRICAL ANALYSIS
3.3 ROBUSTNESS TESTS
3.3.1 Robustness test 1
3.3.2 Robustness test 2
3.3.3 Robustness test 3
4. DISCUSSION AND CONCLUSION
4.1 MANAGERIAL IMPLICATIONS
4.2 POLICY IMPLICATIONS
4.3 FUTURE RESEARCH
1. Introduction While manufacturing and R&D location decisions have both received a lot of attention in the extant literature, the determinants of co-location of the two—the location of both activities in the same geographic location—has attracted considerably less interest. Empirical studies are scarce, but even more scarce are rigorous theoretical examinations of the topic at the proper level of analysis. The goal of this paper is to start filling this gap by examining the co-location of R&D and manufacturing activities, by looking at both external and internal (to the firm) drivers of colocation. The central question we pose in this paper is: When is co-location necessary for the effective coordination of R&D and manufacturing activities? Examination of this question necessarily implies a detailed micro-organizational analysis of the phenomenon.
This article is largely motivated by interesting yet almost opposing empirical observations regarding the co-location of R&D and manufacturing in extant research. In some contexts, co-location seems to be the norm rather than the exception: Ambos (2005), for instance, found that among his sample of large German industrial firms, 79% of research laboratories were colocated with production. Similar observations have been made with Japanese data (e.g., Kenney
and Florida, 1994). In stark contrast, other research has led to completely opposite conclusions:
R&D laboratories in Swedish and British industrial firms tend to be separated from production facilities (Håkanson and Nobel, 1993; Pearce, 1989). However, evidence regarding British firms appears mixed as Hood and Young (1979) in turn observed that foreign companies in the UK in the main co-located their R&D laboratories with production. Anecdotal case study evidence also suggests that the need for co-location can vary drastically from context to context (e.g., Clark and Wheelwright, 1993; Govindarajan and Gupta, 2001). Of course, it is difficult—and somewhat meaningless—ex post to construct theoretical interpretations about the need to co-locate based on past observations made by someone else. One of the primary reasons is that these empirical observations by and large only reflect what is, not what should be. The need for colocation is about what should be, not what de facto is. As Ambos (2005: 404) notes, “[colocation] may simply be a coincidence.” Clearly, more primary theoretical and empirical research is needed.
To be sure, location decisions—both R&D and manufacturing—have received volumes of theoretical and empirical attention from economists (e.g., Mueller and Morgan, 1962; Pearce and Singh, 1992; Smith,  1991), researchers of industrial organization (e.g., Porter, 1986, 1990), as well as organizational and operations management researchers (e.g., Brush et al., 1999;
Ferdows, 1989; Kuemmerle, 1999; Schmenner, 1982). However, what is missing from the literature is a theoretical account of the potential factors that may make co-location necessary. Extant theoretical explanations are not only cursory but also equivocal on the matter: while some researchers have argued that physical co-location of activities is central to coordination both within and across business functions, others have proposed that co-location is overrated in the sense that there are alternative and less expensive mechanisms for coordination (e.g., Rafii, 1995), particularly after the advent of advanced information technologies, starting in effect already with the invention of the telegraph in the 1800s (Appold, 2005). As a result, geographic dispersion of R&D activities within the firm is commonplace (e.g., Howells, 1995; Rugman and Verbeke, 2001). The geographic dispersion of manufacturing activities within the firm, in turn, has been widespread for decades; for instance, year 2008 will mark the 100th anniversary of geographically dispersed production at General Motors.
With the rising concern in Western economies that manufacturing activities and jobs are being relocated to Eastern Europe, Asia, as well as Latin America this article has economic and political relevance as well: What are the implications of this new geographic division of manufacturing tasks on R&D activities? Are decisions about the location of R&D activities subject to the same laws and regularities as manufacturing? Under what conditions must R&D and manufacturing be co-located? How should policy-makers take into account the proposition that colocation may be a necessity for some companies but not others?
1.1 Approaches to Location Decisions While the research interests of economists and organization scholars have coincided in studying (co-)location decisions, economists have in the main been interested in the location-specific drivers, the “locational pulls”, such as access to local technologies and know-how (e.g., Audretsch and Feldman, 1996), whereas organizational researchers have focused not as much on locational as they have on industry- and especially firm-specific considerations (Brush et al., 1999) and even at a more micro-level, cross-functional interdependencies within the firm (Adler, 1995; Clark and Wheelwright, 1993). Our theory builds on the latter, organizational and micro-organizational analysis. However, we will contrast this approach with the macro-level approaches.
Toward this end, it is useful to distinguish between two types of determinants for location decisions: environmental and organizational. Environmental factors may further be divided into location-specific, industry-specific and market-specific; what they have in common, however, is that they are exogenous to the organization and often beyond the sphere of influence of organizational decision-making. Organizations adapt to these contingencies. These are also factors whose effects manifest themselves as tendencies in larger populations, even across national economies, although their effects on individual firms may vary (Kuemmerle, 1999).
Organizational factors, in turn, are organization-specific and subject to both strategic and technological considerations, both of which are, at least in part, under the sphere of management influence. These are factors whose effects are more subtle and more difficult to ascertain empirically without conducting detailed micro-level analyses of organizational activities. At the same time, investigating these organizational factors is important, because Howells (1990), among others, has indeed argued that internal corporate considerations are more important than external, environmental factors in influencing the pattern of corporate R&D location. Helpman (2006) makes a similar argument regarding strategic firm behavior: firm-level examinations are required to understand the prevalent within-industry heterogeneity in location decisions. Finally, whether it is the organizational or environmental factors that matter, understanding managerial decisionmaking is the key to understanding co-location. After all, the decisions to co-locate or not colocate R&D and manufacturing are ultimately made in corporate and business unit management teams; these micro-level activities lay at the foundation of all macro-level phenomena observed in micro- and macroeconomic studies. Failure to understand organizational decision-making means failure to understand co-location.
We can distinguish between three different approaches to the study of location decisions (Figure 1). All the approaches are indispensable, but they take a strikingly different approach to the phenomenon.
1. The External Drivers View
2. The Intra-Functional Dependencies View
3. The Cross-Functional Dependencies View Figure 1. Different views on location decisions
1.1.1 The External Drivers View The proponents of this approach look at the various locational drivers or “locational pulls” for activities and functions. Locational drivers include factors such as low-cost labor, advanced local technology base or developed infrastructure. This is the view most often used by economists, and arguments often invoke the concept of comparative advantage. The external drivers are not, however, limited to economic factors, they also include political, and more recently, also sociological and even psychological factors. After all, in a classic study on manufacturing location decisions in the U.S., Mueller and Morgan (1962) discovered that by far the most common specific reason for location decisions was “personal reasons or chance,” not proximity to customers and suppliers or potential labor advantages.