|
Article Excerpt
Introduction
The critical role of knowledge as a source of competitive advantage has heightened interest in understanding how firms identify, acquire, and use externally-generated knowledge. In addition to formal arrangements, like acquisitions and alliances, firms might also seek external knowledge through indirect means of knowledge spillovers: localized knowledge spilling across organizational boundaries by informal means such as common buyers and suppliers, chance meetings of different firms' scientists and engineers, and employees switching jobs. Because knowledge is partially tacit and localized, its transfer requires frequent interaction that proximity facilitates. Therefore, location is a key parameter that firms can use to increases their exposure to potential knowledge spillovers.
Although knowledge spillovers are clearly a channel for acquiring external knowledge, their relationship to firms' location choices is largely unexplored. The localized knowledge literature in economics, spurred by Jaffe et al. (1993), focuses on the existence, measurement, and diffusion of localized knowledge and assumes, due to its focus on knowledge flows, that locations are exogenous elements to knowledge spillovers. Typically using patent citations as the unit of analysis, most papers neglect firm heterogeneity and competition, and thus the strategic use of knowledge spillovers. As a result, firms are cast as passive actors that source and receive spillovers rather than active entities that can affect the occurrence of knowledge spillovers through their location decisions.
Location decisions are central to the literature in international business on motives for foreign direct investment, which suggests that firms may expand abroad to access new technologies, skills, or knowledge that are not available in their home countries (Cantwell 1989, Chung and Alcacer 2002). While focused on firms' international expansion to access new knowledge, this literature still does not provide a strategic view of location decisions to benefit from knowledge spillovers because of its focus on country, or at best country-industry conditions. As a result, firm capabilities, which determine whether firms can absorb desired knowledge, are mostly ignored and knowledge spillovers are not isolated from other mechanisms used to seek knowledge.
Economics provides a careful treatment of spillovers from localized knowledge, but assumes location decisions as exogenous and neglects firm heterogeneity. International business examines firms' international location decisions, but its focus on country conditions overlooks firm capabilities and does not single out knowledge spillovers. Taking these two literatures as starting points, we propose a more comprehensive and general framework where location choices are no longer exogenous elements to knowledge spillovers but are the result of active decisions made by firms to maximize net knowledge spillovers and to preserve or enhance their competitive position. Under this framework, knowledge spillovers are not exogenous events resulting from the prevailing geographic configuration of economic actors, but are the result of firms viewing the economic landscape and locating strategically.
Specifically, we synthesize arguments from absorptive capacity (Cohen and Levinthal 1990) and knowledge leakage (Shaver and Flyer 2000) to develop a framework where firms are neither equally equipped to receive knowledge nor homogeneously willing to serve as sources of spillovers. Given these differences, firms would strategically choose locations to gain exposure to others' localized knowledge while reducing leakage of their own knowledge to competitors. This paper complements and advances both the localized knowledge literature in economics and the international business literature by not only endogenizing location choices with respect to knowledge flows but also by exploring firm-specific strategic considerations related to knowledge spillovers that may emerge once a firm settles in a new location.
We expect that firms' technological capabilities will affect their baseline preferences among knowledge sources. We consider three knowledge sources--industry, academia, and government. They produce technical knowledge and associated innovations that vary in two dimensions--basicness and appropri-ability. Public sources (academic and government) produce knowledge that is more basic and less appropriable, while private sources (industry) produce less basic, more appropriable knowledge. For firms as receivers, more basic, less appropriable innovations are farther from commercialization, making academic and government technical activity less desirable and industry activity more so--although, more technological advanced firms, those with greater knowledge stocks, are better able to successfully receive spillovers from all sources. The more technologically advanced the firm, the more likely it can find profit opportunities for even public sources' less commercial innovations.
While receiving knowledge spillovers, firms also serve as potential knowledge sources. The same knowledge stock that gives a technologically advanced firm the ability to receive spillovers may make it attractive as a knowledge source, especially for less advanced competitors. Thus, when locating to maximize potential spillovers, firms are subject to these two competing constraints: maximizing benefits from locations' knowledge activity and minimizing spillovers to competitors. These constraints provide opposite incentives for firms' location strategy. Advanced firms, although better suited to experience knowledge spillovers, may nevertheless distance themselves from external innovation sources; less advanced firms, although most interested in obtaining knowledge spillovers, may not have the ability to absorb them. We look to empirical tests to see how firms resolve this tension.
For our examination, we focus on first-time entrants because incumbent firms have prior investments that may affect subsequent location choices and create dependence among observations by the same firm. Our sample consists of first-time foreign entrants to the United States from 1985 to 1994. We examine where these first-time entrants choose to locate as a function of location-industry-year specific technological activity by source and as a function of their firm-specific R & D intensity.
In our earlier work (Chung and Alcacer 2002), we focused primarily on how cross-country industry-level heterogeneity in technological environments impacted the relative attractiveness of different types of locations for different types of foreign direct investment, and offered preliminary evidence about the impact of firm-specific heterogeneity (e.g., see Table 4 in our earlier paper). The research reported here significantly extends that more preliminary analysis by specifically focusing on the theoretical drivers and empirical consequences of firm-level heterogeneity. To focus on the role of firm differences, our analysis incorporates and then synthesizes a range of theoretical arguments from economics, international business, and strategy, and then focuses on testing those arguments on a more narrow range of transactions (namely, greenfield investments) for smaller geographic units (economic areas instead of states) that better capture physical proximity.
We find that economic areas with any level of academic technical activity attract first-time foreign entrants. First-time entrants are disinterested, on average, in areas rich with government and industry technical activity. When exploring the influence of firm heterogeneity, we find that less technologically advanced firms favor economic areas with any level of academic activity and high levels of industrial innovative activity. In contrast, technologically advanced firms favor only locations with high levels of academic activity and avoid economic areas with any level of industrial activity. Our results reveal location behavior that is consistent with firms positioning to maximize their potential for knowledge spillovers.
Limitations of our empirical approach call for conservative interpretation of our results. We focus on location, a pre-condition for knowledge spillovers, instead of measuring knowledge spillovers directly through increases in productivity or information flows. As such, we are neither able to determine whether knowledge actually spilled from source to receiver nor able to exclude market exchanges of knowledge that require physical proximity to transfer. (1)
We proceed as follows. In the next section, we discuss agglomeration economies emphasizing one of its main mechanisms, knowledge spillovers. We then look at how heterogeneity in knowledge spillovers originates from differences among potential sources and receivers. We discuss how these differences are reflected in firms' location choice strategies. Firms trying to maximize potential spillovers and minimize spilling their own knowledge stocks to competitors give rise to differences in firms' propensity to agglomerate. Next, we explain our data, methods, and results. Finally, we highlight our findings, suggest ideas for future research, and discuss implications for firm strategy.
Location Choices and Knowledge Spillovers
Agglomeration economies argue that proximate economic activity benefits firms. Marshall (1920) suggests three mechanisms that induce firms to agglomerate: access to skilled labor (labor market pooling), access to specialized suppliers (shared inputs), and inter-firm knowledge spillovers. Within economics, substantial attention has been paid to the existence, measurement, and diffusion of knowledge spillovers. For example, Jaffe et al. (1993) offer evidence of the link between physical proximity and knowledge spillovers by showing that inventors are more likely to cite other inventors who are geographically proximate. Measuring knowledge spillovers as increases in total factor productivity, Adams and Jaffe (1996) find that a firm's plants in the same state where R & D activity is performed are more productive than its out-of-state plants, demonstrating that knowledge transfer decreases with distance even within the same firm. In an international context, Branstetter (2001) shows that home and foreign country R & D spending spills over to increase firms' patent output. Rosenthal and Strange (2003, p. 387) show that agglomeration economies attenuate rapidly with distance and speculate that the cause is rapid decreases in knowledge spillovers: "information spillovers that require frequent contact between workers may dissipate over a short distance as walking to a meeting place becomes difficult or as random encounters become rare."
Although this literature establishes that knowledge spillovers exist, that they are geographically constrained, and that they have a significant economic effect, it fails to recognize that these very same findings make knowledge spillovers an important basis for firms' competitive advantage. By assuming that the geographic distribution of firms is exogenous, this literature is silent on how firms can alter, via location choices, the intensity and directionality of knowledge flows that provide competitive advantages. (2) For example, in Adams and Jaffe (1996), firms may have considered intra-firm knowledge flows when locating its plants, which then partially determines the observed distance-total factor productivity relationship; in Branstetter (2001), multinational firms may have chosen specific locations to maximize intra-firm knowledge flows. These examples highlight the need for a framework linking firms' strategic location choices and knowledge spillovers.
We argue that location choices may be endogenous to knowledge spillovers: locating proximate to other firms, the pre-condition for knowledge spillovers to happen, is an active decision that firms make depending on many factors, among them the potential to benefit from localized knowledge.
In contrast to the localized knowledge literature in economics, location choice is central to the international business literature. Recent research examines knowledge seeking as a motivation for international expansion. Cantwell (1989) argues that firms may supplement their existing technical capabilities by expanding geographically to access new technology, skills, or knowledge. Several empirical studies offer evidence that firms expand abroad to seek knowledge by setting up R & D facilities or manufacturing sites (Kogut and Chang 1991, Chung and Alcacer 2002).
Although suggesting that firms actively expand abroad to seek knowledge, the arguments in the international business literature are built around country environmental conditions: differences in technological environments across countries and countries' industries are the main drivers of firm behavior. As a result, this literature does not address within-industry firm heterogeneity and its theoretical predictions apply to all firms in an industry or to all firms in a given country. For example, although in our previous paper (Chung and Alcacer 2002) we found that most firms in the pharmaceutical industry follow a location strategy consistent with knowledge seeking, we were unable to explore within-industry differences driven by firm heterogeneity. Even when there is a clear industry effect that encourages knowledge seeking, a location's strategic benefits will vary by firm.
By not addressing within-industry firm heterogeneity, the literature in international business does not provide a clear...
|
