How Industry Dynamics Influence the Internationalization–Performance Relationship: Evidence From Technology-Intensive Firms

Introduction

Internationalization of markets has become a phenomenon which firms cannot afford to ignore. From a strategic perspective, this trend requires firms to operate beyond their national boundaries to remain competitive. International operations expand the character and magnitude of strategic options available to firms when compared with those firms that only operate domestically. The need to operate internationally is very critical for firms in technology-intensive (or high-technology) industries. In many instances these industries (e.g., aircraft manufacturing, motor vehicle manufacturing) have been categorized as global industries, and it is understood that international operations is not a matter of choice but a necessity for a firm to be successful (Doz & Prahalad, 1991; Ghoshal, 2003; Leknes & Carr, 2004; Porter, 1986; Prahalad & Doz, 1987; Yip, 2003).

Among the reasons advanced for this notion is the increasing need to share the developmental costs of new technology across many markets, shorter technological life cycles, the need to source technology worldwide to remain competitive, emerging commonality in market needs, and the pressure to be continuously innovative. For example, firms operating in high-technology industries have a large proportion of their resources committed to their products and process. These firms generally invest a larger portion of their sales revenue into research and development and other tangibles than do firms in other industries (see Figure 1). This obviously creates a need for a firm to regain the high investment made in its products and services. One way for the firm to regain its investment is by increasing the sales of its products and services in international markets without being constrained by the size of the home market. For such firms, international operations provide for the exploitation of interrelationships present across geographic regions, allowing them to gain economies of scope, scale, and experience by sharing or coordinating activities across the major global markets (Bartlett & Ghoshal, 1989; Calori, Atamer, & Nunes, 2000; Hamel & Prahalad, 1985; Kogut, 1985a, 1985b; Kumar & Singh, 2008; Porter, 1985).

While previous studies offer us significant insights on the topic of internationalization and performance, our understanding is not definitive and many unexplored avenues still exist. For instance, while there seems to be a significant push toward consensus on the cubic relationship (horizontal S-shaped or three-stage theory) between internationalization and performance (Contractor, 2007; Lu & Beamish, 2004; Thomas & Eden, 2004), at the same time there seems to be strong support for the belief that internationalization performance would be influenced by firm, industry, and country-specific moderators (e.g., Bausch & Krist, 2007; Elango, 2006; Elango & Sethi, 2007; Hitt, Bierman, Uhlenbruck, & Shimizu, 2006). Recent meta-analysis on the topic (Kirca et al., 2011) indicates some degree of support for both assertions.

Theoretical Background

Internationalization refers to the degree to which a firm's sales revenue is gained from operations outside its home country. Reasons advanced for internationalization by Contractor (2007, p. 471) include knowledge derived from abroad, accessing or arbitraging cheaper inputs, exploitation of firm-specific assets in foreign markets, accumulation of global market power, international scale, lowering risk through geographic diversification, and accumulated internationalization experience. To some extent, one would expect these benefits to be relatively greater in technology-intensive industries, given that firms in these industries have higher investments in research and development, significant economies of scale present in the production of these products, a greater need to protect their investment in technology, a higher difficulty involved in the transfer of such complex technology, a greater rapidity of change in technology and its application, and a need to maintain a closer relationship between leading customers and suppliers in many countries to be successful (Bae, Park, & Wang, 2008; Hout, Porter, & Rudden, 1982; Pangarkar, 2008).

Earlier theoretical and empirical work on the multinational firm has also stressed the importance of international operations for such firms with intangible assets. The theory of the MNE firm's core argument is that multinational firms exist because they possess “unique assets” in terms of products, processes, and skills (Hymer, 1976). These unique assets, which are not easily imitable, enable a firm to gain “competitive advantage” or “market power” over its rivals. Firms seek to exploit firm-specific knowledge and intangible wealth to a greater degree by operating in domestic and international markets rather home markets alone (Buckley & Casson, 1985; Caves, 1980; Kindleberger, 1969). Firm-specific assets, like technological competencies, which are tacit in nature, are also hard to transfer across firms or nations (Kogut & Zander, 1993). These circumstances force a firm to internalize transactions by moving overseas. By internalizing foreign activities, a firm can avoid the disadvantages inherent in market transactions and capitalize on its assets. As a result, multinational corporations (MNCs) can be viewed as a case of the multiplant firm operating in different countries by integrating industries through ownership of assets or controlling activities due to structural market imperfections and transaction cost advantages (Dunning, 1988; Horaguchi & Toyne, 1990; Hymer, 1976; Rugman, 1980; Teece, 1981).

Extant research in the field of strategic management has been examined to see whether internationalization of a firm's operations leads to increased firm performance. Despite the strong theoretical rationale and numerous advantages advanced by practitioners and scholars, studies investigating the relationship between internationalization and performance have not always reported encouraging results. Currently, one may claim with many caveats that extant research on this topic offers support for three types of relationships between internationalization and performance, namely, linear (positive/negative), quadratic (inverted U- and U-shaped), or cubic (varied horizontal S-shaped). However, one must also acknowledge that a few studies report a lack of any relationship (for a review, see Annavarjula & Beldona, 2000; Li, 2007). The varied finding has led some scholars (Hennart, 2007; Verbeke, Li, & Goerzen, 2009) to argue that there is no singular theoretical rationale supporting this relationship. This is a troubling implication for international operations by firms, considering that, on average, large high-technology firms have about 45% of their revenue generated in foreign markets.

Several scholars working on this topic caution that internationalization has significant costs that could potentially reduce the benefits, leading to a poor or even negative relationship with performance. One reason is that firms face increased setup, coordination, monitoring, product modification, and operational costs due to operations being spread across large distances and diverse markets. Second, a firm's operational risk is also increased in three significant ways with international operations. These are risks due to changing political situations in the host countries, fluctuations in currency exchange values, and managerial mistakes due to liability of foreignness and newness (Hitt, Hoskisson, & Ireland, 1994; Hoskisson & Hitt, 1990; Zaheer, 1995).

Undoubtedly, competing in multiple countries is more complex than competing within a firm's home country. In international markets, a firm faces a customer mix with different preference functions as well as a competitor mix with different factor endowments. International markets force a firm to respond to different sets of environmental controls and host/home governmental restrictions/inducements, in addition to societal values originating in different countries internalized in firms' cultures. However, as this study's focus is on high-technology firms, one would assume that, relative to other industries, the pressures to adapt to local environments, as well as the extent of the host government pressure to force firms to adapt, would be much less. Also, these firms' intangible and tangible assets are harder to imitate and usually protected by patents and other forms of entry barriers. In industries characterized by such global pressures, the benefits of international operations will far exceed their costs (Doz, 1985; Elango, 2011; Porter, 1980; Sethi & Elango, 1999). Therefore, such assets, when utilized in different markets, are likely to be very profitable for firms in these industries, leading me to expect a strong relationship between internationalization and performance, despite the potential costs and risks of internationalization.

The second dimension this study incorporates is the influence of industry dynamics on the internationalization–performance relationship. The underlying logic for such an inclusion is that industry characteristics will influence the viability of the firm's internationalization strategies. Earlier work done on this topic clearly supports the fact that industry structural characteristics and competitive dynamics limit the ability of a firm to exploit the benefits of globalization (Birkinshaw, Morrison, & Hulland, 1995; Johansson & Yip, 1994). This influence of industry on firm profitability and conduct has been well established under the structure-conduct-performance (S-C-P) paradigm in industrial organization literature. This paradigm subscribes to the view that the underlying economics of an industry is believed to control and affect competition, and thus the profitability of all firms in that industry. Available research suggests that a firm's capacity to earn profits is highly correlated to the attractiveness and underlying dynamics of the industry in which the firm operates (Caves, 1980; Caves & Porter, 1978; Scherer, 1980). Therefore, I look at two industry variables in particular, namely, the extent of intrafirm trade in the industry and the differential industry growth rate between home country and global markets.

Intrafirm trade is the international exchange of goods and services within a multinational enterprise and refers to the extent to which a firm's various value-added activities are globally integrated. It allows a firm to produce components or products by building global scale facilities at cost-efficient locations and sharing the production with various units of the firm throughout the world. This allows the firm with international operations to gain several advantages: protect its technology, avoid high transaction costs/risks involved in dealing with the market, and diminish currency risks, while gaining global scale and scope economies. According to the OECD, intrafirm trade of MNCs represent a significant portion of foreign trade between countries (Kobrin, 1991; Makhija, Kim, & Williamson, 1997; OECD, 1993).

Intrafirm trade is critical in high-technology industries, as firms have high investments in research and development (R&D) that need to be protected; rely on product- and process-oriented skills, which do not allow for easy transfer; and produce goods that have high economies of scale (Kogut & Zander, 1992). To a certain extent, empirical evidence is also supportive of the notion that high-technology firms need to be vertically integrated to be successful (Hamilton, 1990; Itami & Numagami, 1992; Roberts, 1991; Schoonhoven, 1984). However, the benefits of intrafirm trade cannot accrue to firms that do not have significant international operations. Extensive international operations are needed to allow a firm to perform value activities around the world to provide for exploitation of intrafirm trade. Hence, in industries with large intrafirm trade, one could see a reduced payoff at low levels of internationalization, as these firms will be unable to benefit from intrafirm trade. However, when a firm's international operations exceed a certain threshold, it would be able to benefit significantly from intrafirm trade. Apart from the cost advantages, such firms would also be more competitive, as they would be able to use their leverage across various markets through transfer pricing mechanisms, in addition to being able to withstand competitive actions such as dumping or cross-subsidization by rivals. Therefore, in industries with intrafirm trade, the benefits a firm gets from internationalization are likely to be significant at high levels of internationalization.

The second variable this study examines is the influence of the differential growth rate of the global markets relative to the domestic market. The rationale for inclusion of this variable is straightforward and may be elucidated as follows. The growth of markets in which the firm operates usually sets a “cap” on firm growth and profitability. For example, a firm operating in a growing market will be able to grow and make profits much more easily than a firm in a mature market. Lack of growth and tapering off of market demand usually sets firms into a desperate battle for market share, which results in a lower profit margin for all players. Empirical findings also support the notion that the market growth rate impacts the profitability of firms (Oster, 1994; Schmalensee, 1989). In such circumstances, internationalization of a firm's operations allows the firm to overcome the lack of growth in local markets by diversifying into growing markets. Firms with international operations that are located in stagnant or low growth markets at home may not have to fight as hard for the market share of domestic customers who yield only marginal revenues. In such instances, these firms will be able to exploit their operations more effectively and economically in the international marketplace. Therefore, in the context of this study, I believe that when global markets are growing at a faster rate than domestic markets, the performance benefits derived from internationalization would be greater for a firm (and vice versa). In the next section, I will elaborate the methodology and data collection procedure of this study.

Sample and Methodology

In this study, I focus on five high-technology industries (Aircraft and Spacecraft; Pharmaceuticals; Office, Accounting, & Computer Machinery; Radio, Television, and Communication Equipment; Medical, Precision, and Optical Instruments) and five medium-high-technology industries (Electrical Machinery and Apparatus; Motor Vehicles, Trailers, and Semi-Trailers; Chemicals; Railroad Equipment and Transport Equipment; Machinery and Equipment) based on the classification system of the OECD. These industries have a high R&D intensity relative to production value, varying from 12.7% for Aircraft and Spacecraft manufacturing to 1.9% for Machinery and Equipment manufacturing (36.5 and 5% in research intensity when expressed in terms of value added). These industries conduct the majority of their R&D activity in OECD nations and are believed to represent a significant portion of the collective wealth in knowledge capital in these nations. To study these industries, I decided to focus on the five countries that have the largest numbers of such firms: the United States, Japan, the United Kingdom, Germany, and France.

This resulted in a final sample of 795 firms after deleting the outliers. The United States has the largest representation (299 firms), followed by Japan (222), Germany (130), United Kingdom (86), and France (58).

While a significant debate exists on the best way to measure internationalization, this study follows a widely used measure (Pattnaik & Elango, 2009). I operationalize internationalization as the foreign sales of the firm divided by the total sales. This measure allows us to capture the extent and importance of foreign market exposure to a firm, while also allowing for comparisons with previous studies. Another advantage of this measure is that data is available for many firms, thereby facilitating testing of the internationalization effects using a large cross-country sample. This is not the case with other measures. Another potential measure of Internationalization is the foreign asset ratio. This measure, however, is unduly affected by asset age, asset valuation and depreciation procedures, and other accounting treatments, which vary by country. However, I acknowledge that a limitation of the foreign sales ratio is that it captures one form (internationalization through sales), but not necessarily all forms of internationalization.

Firm performance is measured as operating profit margin, a widely used measure to compare firms in a relative context. I chose operating profit margin over other measures such as return on assets (ROA) or return on investment (ROI), as I felt that it allows for the capture of profit margins undiluted by other firm-related expenses, taxation and accounting treatments and artifacts, which vary significantly across the five countries studied. Second, a ratio such as operating profit margin tells us how much surplus the firm has from each unit of sales revenue, thus serving as a proxy for management efficiency and a firm's ability to withstand price competition and currency exchange rate fluctuations.

Intrafirm trade was measured as an index to allow for capture of the extent of intra-industry trade as a proportion of the total trade in this industry. In the context of this study, this measure is used to understand the extent of globalization potential offered by the industry and its effects on the relationship between internationalization and performance. A high score (close to 1) would indicate that the industry has a high degree of intrafirm trade, and a low score (close to 0) would indicate the low degree of intrafirm trade (Makhija, Kim, & Williamson, 1997). The difference in market growth rate was operationalized as the difference in industry growth rate between the global and home market of the firm. A score greater than zero would indicate that global markets are growing faster than domestic markets, while a score less than zero would indicate domestic markets are growing faster than global markets. Industry data were obtained from OECD databases. Additional details on the operationalization procedure for the variables used are provided in Table 1.

Table 1. Variable Definitions

Consistent with previous research practice on this topic, the regression models incorporated five firm-level control variables (firm size, firm growth rate, debt ratio, research intensity, and product diversification) as well as six dummies in the testing of the relationships. These five variables need to be controlled for, as it has been established that they are related to performance. Firm size was measured by the number of employees, as it is well known that large firms have scale or scope advantages (or disadvantages) over smaller firms. Debt ratio is included in the models to control for the effect of financial capital structure on performance. Similarly, firm growth rate is controlled, as firms with higher growth are likely to have a lower performance in the short run. The linkage between firm resources and performance is one of the themes in the resource-based view of the firm in internationalization performance literature (Kotabe, Srinivasan, & Aulakh, 2002). Therefore, industry-adjusted firm research intensity is included as a control variable in the models tested. This measure of R&D expenditure controls for both within-industry and cross-industry differences in R&D expenditure. Finally, product diversification is also incorporated into the models, as product diversification and its influence on internationalization performance has been report earlier (Hitt, Hoskisson, & Kim, 1997; Sambharya, 1995).

The sample of firms came from five countries over a three-year period (1998–2000). This time period was chosen because this was latest time period for which corresponding industry data were available at a disaggregated level needed for this study. Five dummies were created for each of the countries and three dummies were created to control for country effects and time-related events that may have happened across the years. The six [i.e., (5 – 1) + (3 – 1)] dummies were incorporated in all the models tested. The operationalization of the variables and the sources of information for the various elements in the study are described in Table 1. My review of the correlation tables (not reported here) indicates that the likelihood of multicollinearity issues influencing or biasing the findings of the study is minimal. Consistent with statistical conventional recommendations, independent variables were mean-centered in the regression models to remove nonessential multicollinearity. In each of the regression models, variance inflation factor (VIF) scores were checked to ensure that assumptions of the models tested are not violated.

During the process of data collection, an expert panel was used to ensure that the data was matched appropriately, as two sources of data were used. The expert panel of four members is used because this study uses firm level data from the WorldScope database which reports data using Standard Industrial Classification (SIC), whereas OECD uses the International Standard Industrial Classification (ISIC) industry classification. The four-member expert panel (which did not include me) consisted of two professors working in the areas of strategy and international business, one business librarian, and one senior executive from industry. Each of these individuals had work experience that ranged from 15 to 20 years and a combined experience of 70 years. Each respondent was given a list of SIC codes and the identified ISIC-based high-tech industries. They were asked to match corresponding SIC codes for each of the high-tech industries studied. Only when full agreement was reached among all the four experts did I proceed to the next stage of identifying companies operating in these industries.

Empirical Analysis

The empirical analysis used in this study is based on regression models. Seven models are used to answer the research questions of this study, and the results are reported in Table 2. The first question sought the answer to the relationship between internationalization and performance among technology-intensive firms. Previous research has indicated support for three types of relationships between internationalization and performance: linear, quadratic, and cubic. Therefore, to answer this question, four separate regression models were used: one for each of the relationships and one control model. Model 1 is the simplest model, wherein a linear relationship is tested. Model 2 tests for the quadratic relationship, and Model 3 tests for the cubic relationship between internationalization and performance. Therefore, in Model 1 a linear term is added over the control model, and in Model 2 the square term is added over Model 2, and finally the cubic internationalization term is added to Model 3.

Table 2. Regression Results Between Internationalization and Performance for the Year 1998–2000 for 795 Firms

Following conventional practice, to determine the best fitting model, Model 1 is tested over the control model (also referred to as baseline model), Model 2 is tested over Model 1, and Model 3 over Model 2, using the statistical significance of the incremental F values. The idea is to test the research model (which is based on a prior research hypothesis) over a more parsimonious model to determine if the additional variable in the research model adds value in explaining the incremental variance. From results reported in Table 2, based on the incremental F values, Model 2 is the best-fitting model among Models 1, 2, and 3. Model 2 is preferred over Model 1, as the incremental R-square is significant, indicating that Model 2 explains additional variance even after factoring the extra variable in Model 2 relative to Model 1. Model 3 is rejected, as its incremental R-square has a negative increment. Model 2 supports a quadratic relationship between internationalization and performance. In this model, both linear and quadratic terms have positive loadings, and both are statistically significant.

Models 4 through 7 seek to answer the second question posed by this study. To do this, Models 4 and 6 add one of the two industry variables to the best fitting model from the previous analysis (i.e., Model 2), to be used as a baseline for testing the usefulness of Models 5 and 7. Model 5 tests for the interactive effects of intra-firm trade on the internationalization–performance relationship. Two interactive terms are added for intrafirm trade, as the earlier arguments suggest that intrafirm trade will have differing effects, requiring a linear interactive and quadratic interactive term. Stated differently, it was argued that the benefits of internationalization are reduced with lower levels of intrafirm trade, while the benefits of internationalization would increase at higher levels of intrafirm trade. Model 7 tests for the interactive effects of differential growth rate on the internationalization–performance relationship. In the case of differential growth rate, only one interaction term is added, in contrast to Model 5. This is because arguments made earlier call for only a linear effect, wherein differential growth rate is desirable at all levels of internationalization. Results indicate that both Model 6 and Model 7 offer increased explanations of the relationship studied. The statistically significant results for the interactive terms in Model 5 and Model 7 indicate an interactive effect of these variables in the relationship between internationalization and performance.

Previous research on this topic has also indicated that product diversification moderates the relationship between internationalization and performance (Hitt, Hoskisson, & Kim, 1997; Sambharya, 1995). In order to verify this hypothesis, I ran supplementary models to test for this particular interactive effect (results not reported in Table 2). No support of this hypothesis was found, leading me to infer that interaction with internationalization and product diversification is an assertion that cannot be supported in the case of technology-intensive firms.

Additionally, as a robustness check, I repeated the above tests with three other sets of models. First, using lagged-structure design, I regressed performance in year t + 1 on company attributes in year t, as it is reasonable to expect a time lag between a firm's effort to operate abroad and its outcome in the international market in the subsequent year. Second, to improve reliability with respect to autocorrelation and heteroskedasticity issues, I conducted two separate tests. In the first, I also repeated the models using TSREG procedure in SAS despite having only a three-year panel. This resulted in the sample being reduced significantly (close to 50%), as TSREG requires a balanced sample. In the second test, I repeated the analysis year by year (i.e., without pooling of data). However, the pattern of results remained consistent in all of the instances, giving me greater confidence in the results reported.

Discussion of Findings

The results of this study show that a positive quadratic relationship exists between internationalization and performance for high-technology firms. In other words, the payoffs from internationalization increases at a quadratic rate as firms increase their exposure to foreign markets. The findings strongly validate the long-held contention of scholars and practitioners that significant benefits exist for a firm undertaking internationalization of operations (Contractor, 2007). The quadratic relationship between internationalization and performance indicates that, at the early stages of internationalization (below 31.1%), benefits increase at a lower pace, possibly due to the initial additional costs of internationalization incurred by the firm. At the later stage, when a firm has established its international operations (internationalization above 31.1%), the benefits accrue at a quadratic rate with internationalization.

These findings are not entirely consistent with many past reported findings, encouraging the need to explain why this may be. As noted by one of the reviewers of this article for this journal, some of the findings are intuitive while others are counterintuitive and therefore offer interesting implications. First, this study is unique in that it focuses on high-technology firms, unlike other studies. For reasons argued earlier, because many technology-oriented industries are global in nature, one would assume the benefits of internationalization are also more significant in such cases. For instance, Elango (2011), in his study of US firms, argues that the benefits of internationalization would be higher in the case of global industries. Additionally, considering that these firms possess significant R&D resources, one would expect a strong relationship between internationalization and performance (Kirca et al., 2011). Second, if one takes a close look at the quadratic (inverted U- and U-shaped) or cubic (horizontal S-shaped) relationships reported earlier, this study's findings are consistent at least with one portion (positive slope) of the earlier findings. Support is not found for the declining portion (negative slope) of the internationalization–performance relationship, leading one to speculate why. Two untested explanations can be offered that need further empirical validation. It could be that, over time, these high-technology firms have evolved and developed organization capabilities in systems and procedures to overcome liabilities and minimize excessive costs in foreign operations. This assumption can be made, as the sample firms are relatively large and these firms could have overcome the negative phases of this relationship. Extant research has also shown/argued firm size and firm age as significant factors in influencing this relationship (Bausch & Krist, 2007; Contractor, 2007). Additionally, as mentioned earlier, it is possible that firms in these industries have relatively less pressure to adapt to local conditions, gain significant scale economies, and potentially do not face the costs related to liability of foreignness (Hymer, 1976; Zaheer, 1995) as assumed, serving as an enabler in this regard. However, one should exercise due caution, as some of the interactive terms in the models loaded negatively, indicating that under some circumstances a negative relationship exists between international operations and performance. Therefore, it could quite well be that earlier studies findings can be replicated contingent on specific industry conditions. Finally, in a limited way, the findings also support the assertion that the internationalization–performance relationship would be driven by differing strategic motivations as well as environmental factors (Verbeke, Li, & Goerzen, 2009).

An important contribution of this study is that it allows firms to recognize that the internationalization–performance relationship is influenced by the context of the industry in which it operates. While the emblematic caveats apply in reading Figures 3 and 4, the statistical support received by the interactive terms in the models tested indicates that the payoffs to internationalization would vary by the extent of intrafirm trade and differential market growth in the industry. In Model 5, the linear interactive term loaded negatively, while the squared interactive term loaded positively, indicating that the benefits of internationalization vary according to the degree of intrafirm trade.

Figure 3 provides an illustration of this moderating effect of intrafirm trade on the relationship between internationalization and performance. It illustrates that, in industries with low levels of intrafirm trade, the benefits of internationalization increase at a much lower rate. However, when the extent of intrafirm trade increases in an industry, firms with international operations obtain greater benefits in international operations. Therefore, in industries with large intrafirm trade, internationalization brings in increases in performance at a faster rate. This calls for firms in technology-intensive industries characterized by extensive intrafirm trade to expand aggressively and operate extensively in international markets. Conversely, in industries with reduced intrafirm trade, firms may find the benefits of internationalization to be reduced, though this reduction is not severe enough to result in a negative relationship.

Figure 4, which is based on Model 7, also offers important implications when internationalization may not lead to increases in performance. For example, when domestic markets are growing at a faster pace than international markets (characterized by a negative differential growth rate), low levels of internationalization are negatively related to performance. However, when a firm has significant operations in international markets, the impact of slow-growing domestic markets is much less. This is not the case for firms located in markets where domestic markets are growing at a relatively slower pace than foreign markets. In such instances, internationalization offers increasing returns. Using the notion of international product life cycles, a concept that argues that markets grow in different cycles across global markets, one can see how international operations bring the potential for risk reduction. Assuming this to be the case and knowing that returns to internationalization vary according to differential market growth rates, firms may want to time their overseas expansion strategies by factoring in both global and domestic market growth rates. Finally, in the model tested, the variable differential market growth rate loads negatively on performance. Therefore, firms located in domestic markets that are growing at a slow rate compared to international markets should recognize that international operations is one way to mitigate this negative impact on performance.

Concluding Comments

This study adds important findings to the internationalization performance literature in several ways: First, it offers clear justification of the benefits of internationalization for firms in technology-intensive industries and supports the benefits of internationalization (e.g., new opportunities for growth; scale and scope economies; factor advantages; greater returns to distinctive capabilities; learning from host markets; risk reduction; as well as many other competitive benefits such as foreclosing markets to rivals and the ability to engage in intense competition in home markets) argued in the literature. In particular, it should be noted that this study uses a focused sample of ten technology-intensive industries, and its sample size is large enough to draw implications with confidence. Therefore, I emphasize that technology-intensive firms are better off internationalizing their operations. If initial returns are disappointing in international operations, this should not be a reason for managers to curtail or minimize such operations, but they should aggressively expand to cross the threshold (> 31%) to ensure that increasing benefits of internationalization (which are quadratic in nature) accrue to the firm. Second, this is one of the few studies to incorporate industry context and highlight the importance of factoring in industry growth rates and intrafirm trade while planning international operations. As illustrated by the slopes in Figures 3 and 4, these two factors attenuated or enhanced the extent of benefits flowing to a firm during internationalization. Managers may want to recognize how the two factors (differential market growth rates and intrafirm trade) constrain or enhance performance in international operations and use them opportunistically, as they offer important options for risk reduction.