specific management team, which is posited to be inelastic at least in the short run, is unable to handle effectively the increased demands that are placed on these internally experienced managers due to increased complexity as well as the time and attention that the new managers require from these internally experienced managers. Consequently, inefficiency in the firms current operations will follow if the firm maintains its high rate of growth.
The research proposition that a firm cannot remain operationally effective if it maintains high rates of growth in successive time periods, and that consequently those firms with foresight typically will slow down their growth in the subsequent time period is known as the Penrose effect in the research literature, and this effect of dynamic adjustment costs has been examined and corroborated in a few empirical research studies. However, researchers have not yet examined the Penrose effect in an international business context.
The current paper examines the Penrose effect in an international business context by exploring whether Japanese firms achieve high growth in consecutive time periods in the entered U. S. industries. The empirical results indicate that, consistent with Penroses (1959) resource? based theory prediction, in general, Japanese firms did not maintain high employment growth in two consecutive time periods following their entry into U. S. industries.
We also find empirically that for Japanese multinational firms that entered in U. S. industries where the extent of knowledge tacitness, globalization, and unionization was high, rapid expansion growth in one time period had negative impacts on growth in the subsequent time period.
Thus, dynamic adjustment costs limit the rate of the growth of the firm and the development of dynamic capabilities in this international business context, which suggests that the Penrose effect may be widely applicable to international business and corporate strategy. We are grateful to Richard Levin, Alvin Klevorick, Richard Nelson, and especially Sidney Winter for providing us with the data from their paper in Brookings Papers on Economic Activity (1987).
Published: 2003 URL: http://www. business.uiuc. edu/Working_Papers/papers/03? 0113. pdf EXAMINING THE PENROSE EFFECT IN AN INTERNATIONAL BUSINESS CONTEXT: THE DYNAMICS OF JAPANESE FIRM GROWTH IN U. S. INDUSTRIES BY Danchi Tan Assistant Professor National Chengchi University 64, Chih-nan Rd. , Sec.
2 Wenshan, Taipei 11623 Taiwan Tel: +886 2 2939-3091 ext 81139 Fax: +886 2 2938-7699 Email: [email protected] edu. tw and Joseph T. Mahoney Professor of Strategy Department of Business Administration College of Business University of Illinois at Urbana-Champaign 339 Wohlers Hall 1206 South Sixth Street Champaign, IL 61820 217-244-8257 (office) 217-244-7969 (fax) [email protected] We are grateful to Richard Levin, Alvin Klevorick, Richard Nelson, and especially Sidney Winter for providing us with the data from their paper in Brookings Papers on Economic Activity (1987).
EXAMINING THE PENROSE EFFECT IN AN INTERNATIONAL BUSINESS CONTEXT: THE DYNAMICS OF JAPANESE FIRM GROWTH IN U. S. INDUSTRIES Abstract Penrose (1959) theoretically developed the research proposition that the finite capacities of a firms internally experienced managers limit the rate at which the firm can grow in a given period of time.
One empirical implication that follows logically from this line of reasoning is that a fast-growing firm will eventually slow down its growth in the subsequent time period because its firm-specific management team, which is posited to be inelastic at least in the short run, is unable to handle effectively the increased demands that are placed on these internally experienced managers due to increased complexity as well as the time and attention that the new managers require from these internally experienced managers.
Consequently, inefficiency in the firms current operations will follow if the firm maintains its high rate of growth. The research proposition that a firm cannot remain operationally effective if it maintains high rates of growth in successive time periods, and that consequently those firms with foresight typically will slow down their growth in the subsequent time period is known as the Penrose effect in the research literature, and this effect of dynamic adjustment costs has been examined and corroborated in a few empirical research studies.
However, researchers have not yet examined the Penrose effect in an international business context. The current paper examines the Penrose effect in an international business context by exploring whether Japanese firms achieve high growth in consecutive time periods in the entered U. S. industries. The empirical results indicate that, consistent with Penroses (1959) resourcebased theory prediction, in general, Japanese firms did not maintain high employment growth in two consecutive time periods following their entry into U. S. industries.
We also find empirically that for Japanese multinational firms that entered in U. S. industries where the extent of knowledge tacitness, globalization, and unionization was high, rapid expansion growth in one time period had negative impacts on growth in the subsequent time period. Thus, dynamic adjustment costs limit the rate of the growth of the firm and the development of dynamic capabilities in this international business context, which suggests that the Penrose effect may be widely applicable to international business and corporate strategy. Keywords:
The Penrose effect, dynamic adjustment costs, dynamic capabilities 2 Section 1: Introduction How a firm evolves over time has been an important issue in the fields of strategic management and industrial organization economics (Kor and Mahoney, 2000; Nelson and Winter, 1982). Looking at the historical business record from an organizational capabilities and technology trajectories perspective, Chandler (1990) suggests that modern business enterprises arise from the economies of scale and scope that are made possible by the development of new technologies.
Furthermore, a number of researchers who approach these business issues more deductively in economic science come to a similar conclusion to Chandlers (1990) more inductive business history methodology by maintaining that a firms behavior is best understood as a path-dependent process, and that organizational capabilities develop dynamically (see e. g. , Nelson and Winter, 1982; and Teece, Pisano, and Shuen, 1997). All agree that history matters.
In the current paper, we contribute to the dynamic capabilities research literature (Teece, Pisano and Shuen, 1997) and maintain that the history of a firms strategic moves will matter a great deal in the operational effectiveness of their subsequent moves. In particular, we argue from a resource-based perspective (Penrose, 1959) that a firm that attempts to alter its resource base and to develop its organizational capabilities to meet market change adaptively is likely to incur dynamic adjustment costs, and we examine empirically to what extent a firm that expands internationally incurs such adjustment costs.
Dynamic adjustment costs occur when adjustments of productive resources (such as hiring new employees and new managers) disrupt current operations (Hamermesh and Pfann, 3 1994; Lucas, 1967; Mortensen, 1973; Treadway, 1970). Due to dynamic adjustment costs, there are limits to the growth rate at which a firm can increase its resource base at any point in time.
Our primary goal in the current paper is to make clear that dynamic adjustment costs need to be in the foreground for both empirical testing and theorizing about the rate of the growth of the firm and the development of firm-specific resources and dynamic capabilities. We focus on a major source of dynamic adjustment costs: the inability of a firm to adjust its managerial resources to the desired level in a timely way to match adaptively to a change in the market (Hay and Morris, 1991; Ingham, 1992; Penrose, 1959; Rubin, 1973; Slater, 1980). Penrose (1959) argues that a firms expansion requires the services from managers who have experience internal to the firm.
Since such managers must be developed within the firm over time and could not be hired from the outside, a firm that needs to maintain effectiveness in its current operations could only increase its managerial resources in a controlled and incremental fashion. A fast-growing firm is likely to incur managerial problems (Slater, 1980), and consequently can achieve only little growth in the subsequent time period. Otherwise, inefficiencies in current operations would result because the new managers will require too much time and attention from the experienced managers.
Thus, in the current paper we emphasize that the research on dynamic adjustment costs (Hay and Morris, 1991) should be joined with research on the firm building dynamic capabilities (Teece, Pisano and Shuen, 1997). To be even more precise, we argue here that Penrose (1959) is the seminal work that connects the dynamic adjustment cost research literature and the 4 dynamic capabilities research literature. Indeed, both research literatures cite Penrose (1959) as a seminal contribution (see Hay and Morris, 1991 for dynamic adjustment costs; and Teece Pisano and Shuen, 1997 for dynamic capabilities).
When considering some of the empirical research literature on dynamic adjustment costs, the impact of the managerial constraint on firm growth has been cited as the Penrose effect in the research literature (Hay and Morris, 1991), and has been examined in a number of empirical research studies (e. g. , Gander, 1991; Orser, Hogarth-Scott and Riding, 2000; Shane, 1996; Shen, 1970; and Thompson, 1994). However, there has been little empirical work in examining the Penrose effect in an international business context, which is the focus of the current paper.
The current paper attempts to fill this research gap by examining the Penrose effect in international expansion. The Penrose effect arises from the lack of suitable management for coordinating increased complexity of an organization during its expansion. However, the use of new organizational innovations, such as a multidivisional organizational structure (Chandler, 1962), may help reduce this complexity and hence the need for managerial resources, which in turn may mitigate the Penrose effect. In addition, the need for closely coordinating subsidiaries might be reduced if the subsidiaries are located overseas (Penrose, 1959).
As a result, the time and efforts that managers must spend in managing international operations may be reduced. However, the Penrose effect may still exist as the head office must plan international expansion and may have to maintain a certain level of coordination among its overseas units. The current paper, thus, not only tests empirically for the existence of the Penrose effect that has been 5 corroborated in previous empirical studies, but also suggests a new path for future business research that explores theoretically the conditions under which the Penrose effect is more likely to prevail, and then tests these conditions empirically.
In the next section, we briefly summarize Penroses (1959) argument on the limitations to the growth of firms and empirical studies that have examined this proposition. We then explore the applicability of Penroses (1959) argument in an international business context and we propose several conditions that may moderate the Penrose effect in international expansion. Section 3 provides the theory and hypotheses. Section 4 outlines our methodology and section 5 presents our empirical results. We then provide our discussion and conclusions in Section 6.
Section 2: Background Penrose suggests that a firm can be viewed as a collection of productive resources (1959: 24). No matter what industries a firm is in, the firm relies on its managers to direct and coordinate its productive resources, and to capitalize on production opportunities for the firm. However, to provide proper service to the firm, managers must have experience internal to the firm and experience working with other people within the firm as a team (Becker, 1964; Castanias and Helfat, 1991; Penrose, 1959).
Consequently, the capacities of these managers (henceforth, internal managerial capacities) shape the scope and complexity of activities that a firm can undertake. Since internally experienced managers cannot be hired from outside and must be developed within the firm over time, there are limits to the rate at which a firm can 6 expand its activities at any time. A fast-growing firm is thus likely to encounter managerial problems because it cannot adjust its managerial resources to the desired level in a timely fashion.
To express these ideas compactly: managerial time and attention are the scarce resources that are the binding constraint on the rate of the growth of the firm (Penrose, 1959; Slater, 1980). In other words, dynamic adjustment costs place a limit on the rate of developing and deploying dynamic capabilities The impact of this managerial constraint on the growth of the firm has been cited as the Penrose effect in the research literature and has been empirically examined in a number of studies (Hay and Morris, 1991).
One strategic management implication of the Penrose effect is that a fast-growing organization tends to stagnate in the subsequent time period due to managerial limitations. Shen (1970) finds empirically that the Penrose effect was responsible for the negative correlation coefficients between growth rates of 4000 Massachusetts manufacturing plants in 1948-53 and in 1953-57. Shens (1970) study shows empirical evidence of the Penrose effect at the plant level. Nevertheless, a firm can also grow via establishing new plants.
The use of organizational innovations, such as a multidivisional organizational structure (Chandler, 1962; Williamson, 1975), may help reduce organizational complexity that potentially arises from increased organizational size. Orser, Hogarth-Scott and Riding (2000) report that among the 1,004 small and medium-sized Canadian firms studied, fewer than one quarter of these firms had two consecutive years of revenue increases. However, this empirical study does not provide reasons 7 why there were firms that were not subject to the Penrose effect and were able to achieve growth in consecutive time periods.
Gander (1991) examines empirically the managerial limitations on firm growth by investigating whether there are decreasing (growth) returns to managerial resources (i. e. , managerial diseconomies). Gander (1991) suggests that as the firm doubles its size, the firm has to utilize more than double its managerial resources to maintain effective coordination. Hence, Gander (1991) expects that managerial intensity in an industry (proxied by the ratio of managerial employment to industry asset size) should increase with the size of firms in the industry.
Gander (1991) tests this hypothesis using aggregate two-digit SIC U. S. industry data, and finds this hypothesis supported empirically for the 1977-1980 period, but not for the 1983-1986 period. Two studies explored empirically the Penrose effect of firm expansion by franchising. Thompson (1994) suggests that different forms of firm expansion require different levels of managerial resources. Thompson (1994) expects that a firm is likely to expand via franchising initially in order to economize on its managerial resources.
As the managerial limit decreases with experience, the firm will replace its franchised outlets by hierarchical ones. Based on a sample of 200 franchise chains in 15 U. S. industries, Thompson (1994) finds empirically that the proportion of hierarchical outlets of these firms had a convex relationship with franchise experience. Similarly, Shane (1996) suggests that contractual organizational forms such as franchising economize on the costs of monitoring employees.
Since such monitoring costs 8 increase with firm size in the process of growth, the use of contractual forms help relieve managerial limits to firm growth. Based on a sample of 138 firms in the United States, Shane (1996) finds empirically that firms following a franchise strategy did enjoy a higher growth rate of establishing outlets. In general, empirical research studies exploring the Penrose effect provide supportive, but not robust, empirical evidence that dynamic adjustment costs are important for understanding the rate of growth and the development of dynamic capabilities.
In addition, the extent to which a firms expansion is subject to managerial limits seems to vary with the types of expansion considered in these empirical studies (e. g. , plant expansion, franchise, etc. ). Given the current state of empirical research concerning the Penrose effect, it is unclear to what extent a firm that expands into a foreign market incurs managerial constraints on its rate of growth. The current paper is, to our knowledge, the first empirical paper to look at the Penrose effect in an international business context.
In the next section, we develop the theory and hypotheses. Section 3: Theory and Hypotheses According to Penrose (1959), planning and executing expansion projects require the services of internally experienced managers. The reason being that the process of decision-making and coordination is too complex to be codified as a management blueprint that newly-hired managers could implement, and consequently the firm must, to some extent, rely on managers experience internal to the firm and on their experience working with other 9 people within the firm as a team (Penrose, 1959).
Since internally experienced managers could not be hired from managerial labor markets and could only be developed within the firm over time, there are limits to the rate at which a firm can grow at any time. A firm that expands faster than it can increase its internal managerial capacities is likely to incur managerial problems and reduced effectiveness in its current operations (Ingham, 1992; Slater, 1980). These managerial problems then may hamper the firms growth and the development and deployment of dynamic capabilities in the subsequent time period.
International expansion via direct investments is a corporate-level strategy that allows a firm to deploy and develop its organizational capabilities (Chang, 1995), but it also requires the services of a firms experienced managers. Specifically, planning an overseas investment involves a series of complex decisions. For example, managers must decide the location and the scale of (sunk cost) investments, the mode of entry, the methods of financing, and so forth. To evaluate different alternatives properly, the managers must understand the firms resources profile and the strategic goals of the firms international expansion.
Such knowledge is firm specific and could only be embodied in managers who have experience within the firm. While planning international expansion requires the services of internally experienced parent managers, whether and to what extent managing international operations requires such managerial services does not have a clear theory-based answer. Because countries differ in cultures, languages, infrastructures, and so forth, managing foreign operations may require skills and knowledge different from what parent managers have learned from their domestic operations 10 (Luo and Peng, 1999).
Consequently, the experience of parent managers may not be as economically valuable in international operations, and the multinational firm could solicit help from local personnel to manage effectively its foreign operations. Given that a multinational firm could expand in a foreign market through acquiring existing operations and hiring local managers, to understand how managing international operations is subject to finite internal (parent) managerial capacities, we next explore what experienced parent managers could bring to the management of international operations that is unique and cannot be replaced by the services of new-hires.
Let us first examine what managers of a multinational parent firm may do in managing overseas operations. Managers of a multinational parent firm facilitate, as well as manage, the interdependencies among the firms subunits (including the headquarters) so that the firm can achieve operational and strategic synergies. Managers provide proper coordination and control to the foreign affiliates by collecting and evaluating the feedback from these foreign affiliates. In addition, these managers may sometimes directly involve (as expatriates) the daily operations of the foreign affiliates.
If fulfilling these managerial tasks does not require managers to be equipped with parent-specific knowledge and skills (that can only be accumulated over time through working within the multinational headquarters), the multinational firm can utilize effectively newly-hired managers from local or home managerial labor markets and the multinational firm should encounter less serious managerial constraints on the rate of growth as it expands in the foreign market. 11
By this logic, the key theoretical question then is under what conditions will a multinational firms international expansion require substantial parent-specific managerial skill? We argue that the level of interdependencies and the type of control systems employed within the multinational firms dictate how much parent-specific skill is required for managing international operations. Specifically, a multinational firm can implement three types of control over its foreign operations output control, behavior control, and social control (Eisenhardt, 1985; Hennart, 1991; Ouchi, 1979).
Output control involves parent managers evaluating foreign affiliates performance based on financial reports, and does not require substantial time and efforts from internally experienced parent managers (Lord and Ranft, 2000). However, output control is not effective when an affiliates individual performance is costly to measure. One such condition is a team production situation, in which an individual subunits contribution to the whole organization is difficult to measure accurately (Eisenhardt, 1985). In this case, individual subunits, if evaluated based on measurable outcome, may under/over-produce unmeasured goods/bads.
To alleviate this positive/negative externality problem, a firm may implement behavior and/or social control (Ouchi, 1979). Doing so often requires the parent firm to transplant and to diffuse corporate policies, practices, and culture in the foreign operations (Hennart, 1982, 1991). Such tasks call for internal managerial capacities because only people who have experience within the parent firm for a sufficient period of time can understand the relationships between the headquarters and foreign subsidiaries and can appreciate and adhere to such corporate practices and culture.
12 The nature of coordination among subunits within the multinational firm therefore influences the extent to which a firm is subject to the Penrose effect during the process of international expansion. Compared with domestic expansion, international expansion might be less subject to dynamic adjustment costs that are due to finite internal (parent) managerial capacities.
First, while a firm in general applies coherent and consistent business policies over its domestic operations and maintains close coordination between them, it does not necessarily do so for its overseas operations (Penrose, 1959) because: (1) the interdependencies between domestic and overseas markets may be low as the firm may deal with different competitors and customers in each market; and (2) national differences in cultures, languages, and institutions may make such close coordination very costly.
In addition, geographical distance between headquarters and foreign subsidiaries makes it difficult for the multinational parent firm to monitor closely its foreign operations, increasing the contractual hazards that are due to agency problems (Shane, 1996). To resolve such business problems the multinational parent firm can implement a multidivisional structure and organize its overseas operations into autonomous profit centers.
Holding each operation accountable for its own economic performance reduces the need of the multinational parent firm for monitoring the overseas operations, allowing the parent managers to concentrate on longer-term strategic planning for further expansion (Chandler, 1962; Williamson, 1975). In this case, firm growth in international business markets may be less vulnerable to the Penrose effect. 13.
Of course, an important premise underlying the use of an autonomous, multidivisional organizational structure within a multinational firm is that the activities between autonomous units can be effectively separable and that there is little interaction required among these units (Williamson, 1975). However, this premise of low economies of scope between subsidiaries does not always hold true. For example, individual subsidiaries may share common technologies or may involve joint production (Chandler, 1990).
To maintain autonomous status of individual subsidiaries is likely to stifle exchanges among the subsidiaries and to produce suboptimal results (Lord and Ranft, 2000). To deal with this high interdependency among subunits, the multinational parent firm may attempt to coordinate subunits more closely and to impose behavior and/or social control over its foreign operations (ODonnell, 2000). Doing so requires parent-specific knowledge and capabilities.
Since managers with such knowledge and skills could only be developed within the multinational firm over time through training and learning-by-doing on the job (Kuemmerle, 1997), if growing fast in a foreign market, such a firm is likely to encounter the Penrose effect as the coordination complexity within the multinational firm is likely to increase at a rate faster than its internal managerial resources can handle effectively.
Therefore, in industries where close coordination within a multinational firm is required and where foreign subsidiaries cannot be effectively managed as autonomous, independent sub-units, the Penrose effect is expected to occur. Industries that are characterized by a high extent of knowledge tacitness are likely to be international business contexts where foreign 14 subsidiaries cannot be effectively managed as autonomous, independent sub-units because well-codified blueprints for management cannot be readily provided.
Therefore, dynamic adjustment costs are more likely to limit the rate of the growth of the multinational firm and its development of dynamic capabilities in industries that are characterized by a high extent of knowledge tacitness. The tacitness of knowledge is a matter of degree (Nelson and Winter, 1982). Much knowledge remains tacit because it cannot be articulated in a complete and synchronous fashion (Nelson and Winter, 1982; Polanyi, 1962).
Since tacit knowledge cannot be made explicit completely and thus must be partially learned through the process of trial and error (Winter, 1987), its transfer relies on the initiative of knowledge receivers to engage actively in the learning process, and on the willingness of knowledge transferors to demonstrate their knowledge on the job and to mentor knowledge receivers (Hitt, Bierman, Shimizu, Kochhar, 2001; Szulanski, 1996).
Organizing overseas subsidiaries as autonomous, independent sub-units is likely to suppress the interaction between knowledge transferor and receiver that is typically crucial in the transfer of tacit knowledge (Winter, 1987), and hence may stifle the flow of tacit knowledge within a multinational firm (Hedlund, 1994; Lord and Ranft, 2000).
To facilitate the sharing and the transfer of tacit knowledge among the subunits within the multinational firm, the managers often need to build up lateral integrating mechanisms (such as inter-subsidiary meetings and committees) to increase contacts between managers from different subsidiaries (ODonnell, 2000; 15 Rugman and Verbeke, 2001; Subramaniam and Venkatraman, 2001). These managers may also need to institute in the overseas subsidiaries a corporate culture that values learning and sharing (Liker, Fruin, and Adler, 1999).
Such managerial tasks require the services of internally experienced parent managers (Kuemmerle, 1997). New hires do not have the intra-firm social relationships to facilitate the interaction among subsidiaries (Bartlett and Ghoshal, 1992), and these new hires have not yet been acculturated to appreciate and adhere to the corporate culture (Edstrom and Galbraith, 1977). A firm that grows fast in industries that are characterized by a high extent of knowledge tacitness is likely to lack sufficient managerial capacities to facilitate the flow of tacit knowledge within the firm.
Thus, the Penrose effect is expected to occur. This theoretical reasoning leads then to our first hypothesis: H1: The Penrose effect is more likely to occur in industries characterized by a high extent of knowledge tacitness. Another industry where a multinational headquarters is likely to coordinate its overseas subsidiaries closely is global industries. In a global industry, a multinational firms competitive position in one country is significantly influenced by its position in other countries (Porter, 1986:113).
This competitive influence may be due to global configuration of production activities (i. e. , a joint production situation), or the sharing of the same brand names among subsidiaries in different locations. In such an industry, the productivity of a subsidiary is likely to be significantly influenced by that of other subsidiaries, and high non-priceable inter16 dependencies among subsidiaries are expected to occur (Kobrin, 1991).
Since in global industries the individual performance of subsidiaries does not accurately reflect their contribution to the multinational firm as a whole, managing the highly interdependent overseas subsidiaries as autonomous, independent subunits will be problematic (Hennart, 1982, 1991; ODonnell, 2000). Instead, the multinational firm must coordinate the behaviors (such as service standards, purchases of common inputs, and marketing) of the subsidiaries worldwide (Porter, 1986; Roth, Schweiger, and Morrison, 1991).
The close coordination of subsidiary activities is likely to require the service of managers who have a comprehensive understanding of the nature of interdependence among subsidiaries and the role of individual subsidiaries in the firm. Such knowledge is firm specific and could be embodied only in internally experienced managers. Since internally experienced managers must be developed within the multinational firms over time and new hires cannot replace their services in the short-term, we therefore expect that the Penrose effect is more likely to occur in global industries.
H2: The Penrose effect is more likely to occur in (highly interdependent) global industries. We have argued that in industries where close coordination among subsidiaries is desired, managing foreign subsidiaries requires the services of managers who have experience within the multinational firm. A multination firm that grows fast in such industries is likely to incur the Penrose effect because the firm cannot adjust its managerial resources fast enough to manage increased coordination and complexity in international operations.
Here the source of dynamic 17 adjustment costs arises from the market frictions in managerial labor markets (Becker, 1964; Williamson, 1975) internally experienced managers cannot be recruited from the outside. Sometimes a firm cannot adjust its managerial resources effectively due to institutional barriers (a structural market friction). One such condition is when a multinational firm enters into a highly unionized foreign industry.