Timing of Entry in International Market: An Empirical ... · Gerardo R. Ungson*** SAN FRANCISCO STATE UNIVERSITY This study attempts to explain the timing of entry of firms in interna-

Timing of Entry in International Market: An Empirical Study of U.S. Fortune 500 Firms inChinaAuthor(s): Vibha Gaba, Yigang Pan, Gerardo R. UngsonSource: Journal of International Business Studies, Vol. 33, No. 1 (1st Qtr., 2002), pp. 39-55Published by: Palgrave Macmillan JournalsStable URL: http://www.jstor.org/stable/3069573Accessed: 22/11/2009 07:44

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available athttp://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unlessyou have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and youmay use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained athttp://www.jstor.org/action/showPublisher?publisherCode=pal.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printedpage of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

Palgrave Macmillan Journals is collaborating with JSTOR to digitize, preserve and extend access to Journal ofInternational Business Studies.

http://www.jstor.org

http://www.jstor.org/stable/3069573?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp

http://www.jstor.org/action/showPublisher?publisherCode=pal

Timing of Entry in International Market:

An Empirical Study of U.S. Fortune 500

Firms in China

Vibha Gaba* UNIVERSITY OF OREGON

Yigang Pan** YORK UNIVERSITY AND UNIVERSITY OF HONG KONG

Gerardo R. Ungson*** SAN FRANCISCO STATE UNIVERSITY

This study attempts to explain the timing of entry of firms in international markets. Based on the existing literature, we propose a framework that consists of firm-specific factors, industry/market factors, and host countryfactors. Empirical results, based on the entry information of U.S. Fortune 500 firms in China between 1979 and 1996,

INTRODUCTION Firms face three interlocking ques-

tions with regard to international expansion: what market to enter (entry location), how to enter (mode of entry), and when to enter (timing of entry). Earlier

show that larger firms with greater level of internalization and scope economies are likely to enter this foreign market earlier. In addition, non-equity modes, competitors' behavior in the product market, and lower levels of country risk are significantly associated with early entry.

studies have focused on the first two

questions, with particular attention to the prevalence, sequence, trade-offs, and effectiveness of different modes of entry (Kim and Hwang, 1992; Woodcock, Beamish, and Makino, 1994; Madhok,

*Vibha Gaba is a PhD student at the Lundquist College of Business, University of Oregon.

**Yigang Pan is Scotiabank Professor of International Business, Schulich

School of Business, York University, and School of Business, University of

Hong Kong (2000-2002). ***Gerardo Rivera Ungson is the Y.F. Chang Professor of International Business, College of

Business, San Francisco State University, and Professor of Mgmt., LCB, University of

Oregon. We thank Rick Mowday, Alan Meyer, Michael Russo, and three anonymous reviewers for their

suggestions on earlier versions.

JOURNAL OF INTERNATIONAL BUSINESS STUDIES, 33, 1 (FIRST QUARTER 2002): 39-55 39

TIMING OF ENTRY

1997). In contrast, research on the third

question-the timing of entry-has received relatively little attention. More- over, extant studies have focused on ex-

post performance consequences of early or late entry. We argue that understand-

ing the antecedents of entry timing com-

plements the current emphasis on the

consequences of entering in a particular entry order. Finally, we attempt to redress some methodological biases of the

previous studies that have been noted in various reviews of the literature. Collec-

tively, these conceptual and method-

ological issues provide the motivation for this paper.

First, in examining the antecedents of firms' timing of entry in a newly opened international market, our empirical work is based on a sample of U.S. Fortune 500 firms covering both entrants and non- entrants into China during the eighteen- year period of 1979 and 1996. The choice of China offers two advantages. First, prior to 1979, China pursued a socialist economic policy of self-reliance and was closed to foreign business. In 1979, China formally re-opened its market to

foreign investment. Accordingly, the

year 1979 serves as the benchmark for

examining how long a MNC waited before entering China. In the past twenty years, China has become the largest re-

cipient of foreign direct investment

among developing countries (Pan, Li, and Tse, 1999). Moreover, China has been characterized as particularly dis- tinctive in terms of its institutions and form of capitalism (Boisot and Child, 1996).

Second, we distinguish between the actual timing of entry (continuous measure) as opposed to the order of entry. In

many previous studies, the order of entry represents an ordinal ranking that as- sesses market entry in terms of first en-

trants, second entrant (early follower), and late entrants, using a dummy variable measurement. Such self-reported categorical measures have been used ex-

tensively in past, particularly with PIMS data (Lieberman and Montgomery, 1998). Our study employs a continuous measure that operationalized by the time difference between a starting time and the time of entry. Luo (1998) has argued that ordinal rankings are sufficient and

appropriate for analyzing oligopolistic market structures (i.e. few producers and

competitors) in that one can readily identify first movers from followers. In more competitive structures (i.e. many competitors), however, classifications based on ordinal ranking tend to bundle

up, that is, a number of firms can be

designated as first entrants, even when self reports indicate different time of en-

try. Since our analysis spans many industries with various market structures and competitive environments, our use of continuous measure of entry time is, thus, appropriate (Luo, 1998; Lilien and Yoo, 1990). Finally, we employ event

history method that incorporates time

varying explanatory variables and provides more efficient estimates of entry timing through censoring techniques (Tuma and Hannan, 1984).

LITERATURE REVIEW Earlier economic studies on the timing

of entry sought to explain how monopolistic or ownership advantages of the firm determined the choice of entry modes (Hymer, 1976; Knickerbocker, 1973). Buckley and Casson (1981) sought to determine the optimal time to switch between entry modes, in order to mini- mize cost and to capitalize on market

growth. In an attempt to synthesize different theories, empirical findings, and

interpretations, Dunning (1988) pro-

JOURNAL OF INTERNATIONAL BUSINESS STUDIES 40

VEBHA GABA, YIGANG PAN, GERARDO R. UNGSON

posed an eclectic framework that combines the ownership and internalization advantages of firms, and locational advantages of countries. This framework has been applied in explaining international market entry of firms (Agarwal and Ramaswami, 1992).

Recent business strategists and marketing researchers extended earlier clas- sical works with multivariate frame- works that included strategic variables (firm and industry) and competitive/environmental variables (country and industry) (Hill, Hwang, and Kim, 1990). Several studies have investigated the order of entry at the level of product sectors (Lieberman and Montgomery, 1998; Kerin, Varadarajan and Peterson, 1992; Robinson, Fornell and Sullivan, 1992; DeCastro and Chrisman, 1995). These studies suggest that order of entry, competitive strategy and performance are in- terrelated. Despite the flurry of studies, the issue of whether first movers have an enduring advantage over followers in terms of long-term overall performance is still debatable (Isobe, Makino and Montgomery, 2000; Kerin et al., 1992; Lieberman and Montgomery, 1998). Nonetheless, there are supporting evidence that surviving first movers achieve higher market shares and profitability, largely as result of economic, preemptive, technological, and behavioral reasons (Kerin et al., 1992; Lilien and Yoon, 1990). Even so, follower advantages can stem from lower innovation costs, rapid reverse engineering, market knowledge from consumers already being familiar with their products, cost savings from imitation, and even scope economies that arise from experience curves (Kerin et al., 1992). Lambkin (1988) reported that pioneers, early followers, and late entrants have significantly different strategic profiles and performance, and that

pioneers tended, on the average, to out- perform later entrants.

Majority of this research has focused on the entry into particular product sectors in the U.S. domestic market (Lamb- kin, 1988; Lieberman and Montgomery, 1998; Mitchell, 1991). Findings suggest that the timing of entry affects competitive position, specifically the ability and competencies of a firm to realize its ob- jectives in attaining or sustaining a competitive advantage (Green, Barclay and Ryans, 1995). In international business, it is important to know when foreign firms enter a newly opened country market (Rivoli and Salorio, 1996; Mas- carenhas, 1997). With this premise, recent studies have examined the role of learning (Chang, 1995), uncertainty (Rivoli and Salorio, 1996), and investment as future options (Kogut and Kula- tilaka, 1994).

Recent studies have examined the order of entry into international markets (Mascarenhas, 1992, 1997; Madhok, 1997; Shaver, Mitchell, and Yeung, 1997; Luo, 1998; Luo and Peng, 1998; Pan and Chi, 1999; Pan, Li, and Tse, 1999; Isobe, Makino and Montgomery, 2000). With few exceptions, such studies have utilized the PIMS data base for which entry timing is ex-post, exogenous classification of order of entry. These studies rely on firms' self-reports where firms categorize themselves as first- mover, early follower, or late follower. Because there are biases with abnor- mally large numbers of firms claiming the first-mover status (Lieberman and Montgomery, 1998), researchers have recommended the use of actual times of entry in order to provide more accurate measures (Lilien and Yoon, 1990; Luo, 1998). Taken collectively, by measuring time in months we are able to provide a fine partitioning of entry timing, as op-

VOL. 33, No. 1, FIRST QUARTER, 2002 41

TIMING OF ENTRY

posed to a more coarser partitioning of entrants into first movers and early/late followers.

A related problem, is that, in explaining the performance consequences of entry timing, previous studies that em- ployed conventional regression analysis tended to include only surviving firms, that produce a "survivor" bias toward higher performance (Mitchell, 1991). One remedy, espoused by Mascarenhas (1997), who also examined the effect of pioneering (first movers) on the market share of firms in an international market, is to include both surviving and failed firms in the analysis. Our use of event history methods includes both entrants and non-entrants in estimating coefficients and testing hypotheses, thereby avoiding some of these limitations noted by Mitchell (1991). Finally, while the order of entry can be meaningfully interpreted in the context of a single product market sector, our study covers many different product market sectors, therefore a continuous measure is preferred.

THEORETICAL BACKGROUND AND CONCEPTUALIZATION

Based on our review, we identify three theoretical thrusts that guided our hypotheses. First, the strategic goals, capabilities, resources, and intent of a firm define its position in the international market, that it, the timing of entry is invariably related to the resources and competencies of the firm, also called the "resource-based theory of the firm" (Lieberman and Montgomery, 1998). Second, the ability of firms to assess market signals and opportunities (Porter, 1991), i.e. "information processing and market signals perspective," is another key factor in explaining the entry timing decision. The decision to enter will de- pend on whether market growth is antic-

ipated (Green et al., 1995), as well as the relative presence of competitors in the market (Knickerbocker, 1973). Third, the attractiveness of the host country and the level of risk are important determinants of the timing of entry, since a country with significant growth prospects and a low risk environment is more likely to attract foreign firms to enter early (Lieberman and Montgomery, 1998). Taken altogether, we propose that the timing of entry is influenced by firm's capabilities and/or firm-specific factors (resource-based), by perceptions of the market/environmental opportunities (information processing), and by host country variables (locational factors). We also examine the relationship between entry timing and selected mode of entry.

Firm-specific Factors Level of Internationalization. Level of

internationalization refers to the degree of involvement of a firm in the international market (Li, 1995; Sullivan, 1994). Firms that have a significant overseas engagement can benefit from the learning and experience associated with their operations in foreign markets, as well as their extensive market networks (Chang, 1995; Shaver et al., 1997). Because they can leverage their accumulated knowledge and experience more readily, firms with a higher degree of international experience are more likely to enter a newly opened international market earlier. They can cross-support and cross-subsi- dize their entry in the new market with their existing operations in other foreign locations. While there is no guarantee that a firm can successfully leverage its experience from one market to another, we argue that the experienced firms are in a better position to overcome the risks and uncertainties in the initial phase of the market entry. Thus, we hypothesize:


VIBHA GABA, YIGANG PAN, GERARDO R. UNGSON

Hypothesis 1: Firms with a higher level of international experience enter a new international market earlier than firms with a lower level of international experience.

Size of Firm. Historically, firm size has been related to market power in both domestic and international contexts

(Chandler, 1962; Hymer, 1976; Knicker-

bocker, 1973). The conventional argu- ment for firm size is premised on how

larger-sized firms exercise their power. Thus, it is argued that larger firms com-

pete in a broader spectrum of products and markets using scale and scope economies (Chandler, 1962). They are able to make preemptive moves that limit or

prevent later entrants from gaining access to suppliers, markets, customers, and other scarce assets (Kobrin, 1991; Lieberman and Montgomery, 1998).

Larger firms have more resources to invest in innovations, to pursue aggressive expansions, to be able to incur the costs and bear the risk, and to achieve a better

performance (Cohen, 1996). Other ad-

vantages include access to privileged learning channels (Tan and Vertinsky, 1996), risk reduction through wider port- folios (Chang, 1995), and from a stronger bargaining power to gain concessions and incentives from host country government (Brewer, 1993; Pan, 1996). Ad- ditional empirical support comes from Lambkin (1988) who reported that pioneering entrants into new markets are able to enter on a large scale and invest

heavily in building a strong market position, while late entrants are predomi- nantly firms with smaller size. Collec-

tively, within this perspective, we hypothesize that larger-sized firms are more likely to enter a foreign market earlier than smaller-sized firms.

Despite its intuitive appeal, there are a number of studies that do not empiri- cally support some of these hypothesized effects of firm size (see review by Kimberly, 1976), prompting researchers to conclude that firm size, in of itself, is not a precursor to market entry, but that the effects of size are what are conse-

quential (Evans, 1976; Kimberly, 1976). In related formulations, researchers have

expanded the effects of size in terms of overall ownership advantage (Dunning, 1988; Rivoli and Salorio, 1996). Owner-

ship advantage is defined generally as

"any income generating assets which makes it possible or firms to engage in

foreign production" (Rivoli and Salorio, 1996, p. 344). Within such formulations, the timing of market entry becomes con- ditional. Rivoli and Salorio (1996) point out, for example, that firms can't afford to delay the entry if the ownership ad-

vantages possess are less unique and

non-monopolistic (Casson, 1987). Be- cause firm size is not a unique form of

ownership advantage (Casson, 1987), there are conditions when a firm might actually delay market entry (Rivoli and Salorio, 1996).

The possible reconciliation of conventional arguments about firm size and

contemporary treatments of ownership advantage might come with the size of the targeted entry market.1 Specifically, we argue that firms will capitalize on their size without too much delay particularly when facing particularly large markets, such as China. This is because

larger firms will be able to leverage their resources when demand is high-a condition that a smaller-sized firms might not be able to capitalize as effectively. While uncertainty conditions posited by Rivoli and Salario (1996) prevail, larger- sized firms are in a better position to

adopt standardized marketing strategies


TIMING OF ENTRY

in particularly large markets, as in the cases of Coca Cola, Caterpillar, Nike, Sony, Marlboro, Phillips, Toyota, Volvo and Kodak. Thus, we hypothesize:

Hypothesis 2: For a potentially large host country market, larger firm size will encourage earlier entry.

Scope Economies. The impact of economies of scope on entry timing comes from two sources. First, a broader scope of products and services means a wide portfolio of offerings to choose from, and thus, means a better chance of providing the right product and service to the newly opened market. In this case, firms with a broader strategic scope are better prepared to handle the uncertainty about the types of product that are needed in the new market (Kerin et al., 1992). Sec- ond, a broader scope of products enables the firm to develop a synergy across different product sectors (Shaver et al., 1997). This synergy gives rise to both efficiency and quality in product devel- opment, product line extension, production, distribution, and market support (Lambkin, 1988; Green et al., 1995; Reddy, Holak, and Bhat, 1994). Taken together, firms with a broader product scope have a higher likelihood of entering the market early. Thus, we hypothesize:

Hypothesis 3: Firms with a broader scope of products and services enter a new international market earlier than firms with a narrower scope.

Industry/Market Factors

Competitor behavior. In markets of high uncertainty and risk, firms are influenced by what other firms in their industry do (Casson, 1987; Porter, 1991). Competitive behavior, as reflected in the entry decision of competitors (the pres-

ence of prior entries), is a significant signal for market entry decisions for two reasons. First, in the absence of their own calculations, firms use the behavior of other firms as a justification of their own behavior. When many firms are observed as investing in an overseas market, hitherto non-entrants infer from their competitors' behavior that such entry is profitable, and that the earlier they enter the faster they too will be able to earn such profits. Second, firms often have a sense of paranoia and fear that either the opportunity will be com- pletely gone, or those that have entered will put up entry barriers high enough to deter subsequent entries (Porter, 1991). Mitchell (1991) suggests that firms are likely to enter earlier when many firms exist and the threat of competition is high.

While, there may be some informa- tional benefits to a delay, this benefit must be weighed against the costs of delay-the risk of entry by other firms and foregone cash flows. In fact, a counter- vailing position is that some firms might avoid entering markets for which they anticipate intense competition (Knicker- bocker, 1973). Even so, while firms might choose to delay their entry and await new information, but they do not always have the opportunity to do so (Rivoli and Salorio, 1996). Specifically, strategic conditions might make it im- perative for the firm to invest quickly and preempt existing or potential competitors (Lieberman and Montgomery, 1998). In such a scenario we would expect firms to take competitive behavior in the form of the number of prior entries into the host market (in its product sector) as a signal of profitable opportunities that might exist in the market. Thus, we hypothesize:



Hypothesis 4: The entry timing decision of a firm will be affected by its competitors' behavior. In particular, entry by competitors will accelerate the entry of a firm into an international market.

Product Market Growth. It is intuitive that product markets with high growth potential will attract firms to enter the market early, because such markets offer ample room for them to grow and pros- per (Green et al., 1995). Firms will move into markets that are characterized by a strong and rising consumer demand. Hennart and Park (1994) found that the entry of Japanese firms in the U.S. market was motivated principally by the growth prospect of the market. There- fore, a potential high growth rate in a product market serves as a strong signal of attractive market. The growth potential is a strong incentive for firms not only to enter the market but also to enter early (Lilien and Yoon, 1990). Thus, we hypothesize:

Hypothesis 5: The greater the market growth in a specific product sector in a new international market, the earlier a firm will enter that market.

Host Country Factor Risk Condition. A potential obstacle

for firms wishing to enter into a foreign market is the inadequate legal, political and regulatory framework-the 'institutional environment' (Brewer, 1993; Keefer and Knack, 1997). Deficiencies in the institutional environment of the host country may deter the entry of foreign firms into the market or at the least post- pone their entries (Boisot and Child, 1996). Given the high degree of central- ized control in China and closed nature of the Chinese market prior to 1979, it is

likely that the firms would evaluate the

level of political and business risk very carefully before making an entry decision (Pan and Chi, 1999). Thus, we hypothesize:

Hypothesis 6: The lower the level of risk in a new international market, the earlier a firm will enter that market.

Mode of Entry The timing of entry and mode of entry

are inextricably related (Mascarenhas, 1997; Madhok, 1997). However, even

though the literature on entry modes is extensive (Buckley and Casson, 1998; Kim and Hwang, 1992; Woodcock, Beamish and Makino, 1994), how timing of entry is affected by the mode of entry has received relatively little attention.

Modes of entry can be viewed as equity-based (i.e., equity joint ventures and wholly owned subsidiaries) and non-equity-based (i.e., exporting; licensing; non-equity alliances) (Pan and Tse, 2000). By adopting an equity-based mode, firms inevitably set up a new entity in the host country, whether an EJV or a wholly owned subsidiary. This re- quires an on-going decision-making and management of the entity. Further, equity-based modes involve investment in assets that are often irreversible and non-redeployable. In contrast, non-equity modes are based on specific contracts on a fixed duration. Therefore, the level of risk tends to be higher with equity-based modes.

From the timing of entry perspective, it appears that an early entry is associated with a higher level of risk and uncertainty, especially at the early phase of market opening in the host country, holding other factors constant (Rivoli and Salorio, 1996). Therefore, we expect that early entries are associated more likely with non-equity modes of entry.


TIMING OF ENTRY

Given that non-equity modes are faster to

plan and implement, and they require smaller resource commitment (Madhok, 1997). Thus, earlier entry might be hence

prompted by less resource commitments

(Isobe, Makino and Montgomery, 2000). In other words, managers may be more

willing to invest earlier if they have less resource commitments at stake. Taken

together, we hypothesize that non-equity modes are easier to adopt for early entry:

Hypothesis 7: Firms adopting non-equity entry modes are likely to enter a new international market earlier.

METHODOLOGY

Sample In this study, we analyze the timing of

entry of U.S. Fortune 500 firms into China during the period of 1979-96. China re-opened its door to foreign firms in 1979 after three decades of isolation. This provides a starting point from which the delay of entry by foreign firms can be examined. In order to provide consistency, we only included those For- tune 500 firms that appeared on the list in both 1979 and 1996. The sample we used is made up of 271 Fortune 500 firms. We used the China Business Re- view (a bimonthly trade journal published by the U.S.-China Business Coun- cil that reports foreign business activities in China) to gather the information on whether and when each of these 271 firms had entered China in this 18-year period. Of these 271 firms, only 126 firms entered China during the study period. In analyzing hypothesis 4, we included firms outside the original Fortune 500 sample to construct an adequate measure of our independent variable

(competitive intensity).

Dependent Variable

Timing of entry. The dependent mea-

sure, is the duration (in months) that an U.S. Fortune 500 firm waited before en-

tering China. For the entry timing, the duration is the difference between the actual time of entry (month and year of

entry) and the starting point when it was

possible to enter China, i.e., January of 1979. For those firms that had not entered at the end of our cutoff point (end of 1996), the dependent measure is the full duration between 1979 and 1996.

Independent Variables Level of internationalization. Two

variables were considered as measures of internationalization-the number of geo- graphic segments in which the firm is

present and the percentage of foreign sales of the total sales of a firm (Sullivan, 1994). Sales figures were deflated by the U.S. producer price index (Paasches Price Index, 1990=100). In addition, we

lagged sales of firms by one year. Both measures were collected from COMPU- STAT (Standard and Poor). We found that the inclusion of both measures re- sulted in high multicollinearity (0.74) between the two. To redress this problem, we used principal component anal-

ysis to create a single variable as a linear function and accounting for 99% of the variation of these two variables.

Size of firm. We measured the size of firms using the firm's assets with data for the period of 1979-96 from COMPU- STAT. While we hypothesize a mono- tonic relationship between firm size and

entry timing, we used the log of firm assets to control for possible non-linear-

ity in the relationship. Scope economies. The number of busi-

ness sectors that a firm participated in is used to measure the breadth of business

scope. The business sectors are catego-



rized according to the 4-digit SIC codes. These data also came from COMPU- STAT.

Competitors' behavior. We operationalized this variable by the number of firms that entered China in the same product sector in the previous year before that particular firm entered. The ra- tionale is that the greater the number of prior entries, the more competitive pressure the firm is under, which may prompt it to enter in the following year (Wernerfelt and Karnani, 1987). The number of prior entries includes all firms, U.S. Fortune 500 firms, Japanese and other non-U.S. firms.

Product market growth. The sampled firms were broadly classified according to the three digit international SIC (ISIC) codes of United Nations Industrial De- velopment Organization (UNIDO). Using the UNIDO database, growth rates for gross output and the number of enterprises for each industry in China were obtained for the period 1979-93. The figures for 1994-96 were extrapolated using the average growth rates of the previous five years. A geometric average of growth rates of gross output and the number of enterprises was used as a proxy for product market growth.

Level of risk in host country. This variable is operationalized by the risk assessments from two independent international investor risk services, Inter- national Country Risk Guide (ICRG) and Business Environmental Risk Intelli- gence (BERI). Risk of expropriation measures the risk of confiscation and forced nationalization of foreign enterprises. Risk of repudiation of contracts measures the risk that the government will repudiate or otherwise unilaterally change the terms of contract with foreign businesses. Given the high correlation between the two and in order to cover

the period of 1979-96, we collapsed these two measures into one index. For ease of interpretation, we re-scaled the composite index so that higher numbers imply higher risk.

Non-Equity Mode of Entry. Mode of entry of firms was coded using a dummy variable. All non-equity modes were coded as 1, and 0 otherwise.2

Control Variable With cyclical fluctuations in the Chi-

nese economy that affected foreign investment in China, we introduced control dummies for four different periods: Phase I (1979-83)-Rehabilitation and Market Incentives; Phase II (1984-87)- Market Reforms; Phase III (1988-91)-Re- trenchment, and Phase IV (1992-96)-Ex- pansion of Foreign Investment (For more details, please see Bell, Khor, and Koch- har, 1993). A negative and significant coefficient, for any of the period dummies will signify that the waiting time until entry is significantly shortened during this period.

Given that Fortune 500 firms are large diversified firms, the entry into a host country market carries a strategic impli- cation for the whole firm, as opposed to its competitive standing on a particular product sector. Therefore, apart from product market growth and competitive pressure that are measured at the level of each industry, we control for industry specific effects by inserting a separate intercept term for each industry in the equations yielding a fixed-effects model (Hsiao, 1986).

Method of Analysis Given that over half of the firms in the

sample had not entered China in this 18-year period, we encountered the issue of right censoring in the data. Event history analysis is well suited to longitudi-


TIMING OF ENTRY

nal data sets because it can handle the

right censored cases, i.e., cases where no event has occurred at the cut off date. The information of censored cases is also

incorporated into the estimates using their survival function values at the time of censoring, that is, the probability that an event will occur at some time beyond the observed duration.3

We used an accelerated event-time method that assumes that the event times (the waiting time until entry into China in this study) are distributed ac-

cording to a parametric baseline model that would hold if all independent variables were set to zero (Kalbfleisch and Prentice, 1980).4 Covariates are then es- timated as exponentially multiplicative accelerations or decelerations of the baseline distribution. The natural loga- rithm of the survival time is expressed as a linear function of the covariates.

log Ti = xl3 + ras,

where Ti is the (possibly unobserved) time until entry by firm i, xi is the vector of covariates associated with the ith firm, and 3 is the vector of coefficients associated with each independent variable and

ei is the error term. Once the distribution of the error term is specified, estimation

proceeds by maximizing the log-likelihood for the censored data.

The accelerated event-time model re-

quires the specification of an underlying distribution for the survival time (i.e., the waiting time until entry into China). The generalized gamma distribution with a density function for T is particularly flexible and used in this study (Mitchell, 1989):

f(K) = IKi(K-')-2 exp[K-(KE - exp(KE))]

K# O: '(K-')

(2r)-1/2 exp(-w2/2) K=0

where a is the scale parameter and K is the shape parameter. To ensure that the model assumptions are met, we did pre- tests and used STATA 6.0 version in our

analyses.

RESULTS AND DISCUSSION

Table 1 shows that means, standard deviation and correlation between the

independent variables. Of the 271 firms in the sample, a total of 126 firms had made their initial (first) entry into China

by 1996.

TABLE 1

MEANs, STANDARD DEVIATION, AND CORRF.ATION

Variables Means (S.D.) (1) (2) (3) (4) (5) (6)

1) Internationalization 11.929 (17.384) 1 2) Economies of scope 2.433 (1.512) 0.048* 1 3) Size of firms 8.094 (1.474) 0.046* 0.107* 1 4) Product market

growth 0.127(0.190) -0.050* -0.008 0.012 1 5) Competitors'

behavior 25.449 (28.55) -0.025 -0.02 0.125* 0.385* 1 6) Environment risk 2.830 (1.301) 0.109* 0.039* -0.228* 0.034* -0.034* 1

NOTE. *p < 0.05.



TABLE 2

TIMING OF ENTRY: EVENT HISTORY ANALYSIS

Model 5 (Controlling for Indnitly

Variables Model 1 Model 2 Model 3 Model 4 Effects)

Internationalization -0.032 -0.045 -0.031 -0.029 (0.004)*** (0.007)*** (0.005)*** (0.006)***

Size of firms -0.397 -0.423 -0.306 -0.322 (0.08)*** (0.106)*** (0.075)*** (0.08)***

Economies of scope -0.146 -0.226 -0.151 -0.152 (0.049)*** (0.070)*** (0.051)*** (0.05)***

Product market -0.182 0.143 -0.232 -0.316

growth (0.337) (0.522) (0.369) (0.353) Competitors' -0.003 -0.011 -0.008 -0.005

behavior (0.002)** (0.004)*** (0.003)*** (0.003)* Environment risk 0.111 0.393 0.321 0.329

(0.052)** (0.144)*** (0.191)* (0.196)* Control variables

Phase II -0.938 -0.875 (0.271)*** (0.272)*** 1

Phase mI 0.085 0.132 (0.266) (0.548)

Phase IV 0.052 -0.068 (0.565) (0.263)

Intercept 9.661 5.92 9.465 8.498 8.938 (0.711)*** (0.119)*** (0.918)*** (0.977)*** (1.122)***

Chi-square statistic 71.12*** 10.81*** 119.54*** 129.65*** 159.43***

Log likelihood -262.95 -325.72 -236.89 -231.831 -216.94 No. of observations 316 311 271 271 271 No. qf events 133 139 126 126 126

NOTE. Negative accelerated event-time coefficients indicate early entry. Standard errors in parentheses. ***p < 0.01; **p < 0.05; *p < 0.1 (two-tailed tests)

Table 2 and 3 shows the results of using firm specific variables only; model accelerated event-time model estimates 2 uses industry and country specific for the timing of entry in testing H1 to variables; model 3 combines model 1 H7. A positive (negative) coefficient im- and model 2; model 4 controls for period plies that the covariate exercises a posi- effects by including period dummies for tive (negative) influence on waiting time. the four phases of market reforms in Thus, a unit increase in the covariate is China; model 5 adds industry specific interpreted as a firm delaying (hastening) dummies to control for unobserved in-

entry into China. In effect, if we hypoth- dustry heterogeneity. In Table 3, we test esize an earlier entry, a negative coeffi- if non-equity mode of entry are corre- cient would support our expectation. lated with early entry into China with

Table 2 presents five models: model 1 and without industry fixed effects (re: attempts to explain the entry timing by models 6 and 7). In terms of or hypoth-


TIMING OF ENTRY

TABLE 3 TIMING OF ENTRY: EVENT HISTORY ANALYSIS

Model 7 Independent (Controlling for Variables Model 6 Industry Effects)

Internationalization -0.023 -0.02 (0.004)*** (0.004)***

Size of firms -0.151 -0.155 (0.054)* * (0.055)***

Economies of scope -0.055t -0.074 (0.038) (0.034)**

Product market -0.387 -0.203 growth (0.302) (0.362)

Competitors' behavior -0.004 -0.005 (0.002)** (0.003)*

Environment risk 0.274 0.252 (0.148)* (0.122)**

Non-Equity Mode -1.295 -1.283 (0.196)*** (0.176)***

Control variable Phase I -0.724 -.707

(0.237)*** (0.214)*** Phase III 0.042 0.028

(0.470) (0.410) Phase IV -0.132 -0.113

(0.233) (0.200) Intercept 7.104 7.714

(0.680)*** (0.654)*** Model Log Likelihood -192.18 -177.06 No. of observations 271 271 No. of events 126 126

NOTE. Negative accelerated event-time coefficients indicate early entry. Standard errors in parentheses. ***p < 0.01; **p < 0.05; *p < 0.1 (two-tailed tests) tp < 0.15.

eses, the results across the models

(Model 3-7) are broadly consistent in terms of the sign, the magnitude and the

significance of coefficients. To control for industry effects, i.e. to test whether

industry differences affect entry timing, we performed a likelihood ratio test to

compare the two models (Model 6 & 7). The likelihood ratio statistic = 0.13278, and has a x2 (25) distribution. The critical value at 5% level is 37.652. As such, the

inclusion of industry fixed effects does not

improve the analysis significantly. The same is true for models 4 and 5.

Our first three hypotheses relate early entry with a higher level of internationalization (H1), larger firm size in a particularly large host country (H2), and broader scope of products and services (H3). Results are supportive of our three

hypotheses. Specifically, we found that firms with higher level of international-



ization enter an international market earlier (p<0.01), which indicates that knowledge-based capabilities are important for foreign market entry decisions. The results indicate that the size of firms has an expected impact on the early entry into China (p<0.01). As hypothesized, the advantages of size enable firms to muster resources, extend support among related product sectors, and ex- ploit economies of scale, scope, and learning in particularly large markets (Kobrin, 1991; Lambkin, 1988). Further- more, the broader product scope enables firms to offer the right product, which facilitates early entry into an international market (p<0.05) (Lieberman and Montgomery, 1998). These results are consistent with resource-based arguments that early entries are different from late entries in terms of their resources and capabilities (Robinson et al., 1992; Lieberman and Montgomery, 1998).

Support for the information processing and market signaling explanation (hypotheses H4-H5) is mixed. In examining Hypothesis 4 that relates entry timing with competitors' behavior, we found that prior entry by competitors forces firms to enter a new foreign market earlier (p<0.1). As hypothesized, firms are less likely to delay their entry into China when they see a lot of their competitors have moved in (Rivoli and Solario, 1996). However, hypothesis 5 that relates early entry with higher market growth is not supported. It is plausi- ble that entry-timing decisions are based on expectations of product market growth, not actual growth. Thus, a firm is likely to enter early if it forecasts a high market growth potential, rendering actual rates to be less consequential. There are indications that such a case might have occurred in China during the

First Phase where expectations of market growth opportunities were overly opti- mistic (Oksenberg et al., 1998). Taken altogether, there is mixed support for the information processing or market signaling perspective.

In relation to hypothesis 6 that relates early entry with level of risk, we found that firms evaluate political and business risk in making their entry timing decision (p<0.05). Favorable risk conditions accelerate the entry timing. Thus, locational features are important. In examining hypothesis 7 that relates early entry with non-equity modes, we found that choosing a non-equity mode of entry ac- celerates the entry of a firm into a foreign market (p<0.01). Compared to equity modes, non-equity modes are of short duration and easier to plan and implement. Hence, we found that firms that adopt non-equity modes tend to enter earlier into a foreign market. Due to longer horizon of planning and imple- mentation, equity-based modes are less susceptible to pressure from competition and short-term risk fluctuations.

Taken collectively, there are several implications for practicing managers and strategies. First, firm capabilities-internationalization, size, and scope-matter. Such provide the antecedents of early market entry. Strong capabilities enable firms to make an early entry. As shown in other studies, the motivation for an early entry comes from in part the supe- rior performance (Luo, 1998; Luo and Peng, 1998; Pan and Chi, 1999; Isobe, Makino and Montgomery, 2000). Sec- ond, the timing of entry is tempered by entry mode. Our study suggests that managers are more willing to enter markets earlier when the resource commitment is lower as in the case of non-equity entry modes.


TIMING OF ENTRY

Methodologically, we attempted to avoid the limitations of previous studies. We made substantial efforts to construct the archival data that included a continuous measure of how early (or how de-

layed) each firm's entry into China. Our use of actual timing of entry is more pre- cise than self-reported measures and nominal classifications, such as those used from the PIMS database. However, continuous measures may not be as un-

ambiguously interpreted when studying the order of entry issues. Finally, by us-

ing the event history method, we were able to incorporate time varying explanatory variables in the analysis as well as control for firms that had not entered China by the end of 1996. By using China as the host country, which re-opened its door in 1979, we were able to test the

hypotheses in a real setting.

LIMITATIONS AND FUTURE RESEARCH

While our study incorporated the im-

pact of broad government regulatory influences on MNEs' entry timing strate-

gies during the period of 1979-1996, we

recognize that there might be regulations unique to each industry. To statistically control for the unobserved effect, we in- serted a separate intercept term for each

industry in the estimation. We note that the results remain robust even after con-

trolling for industry effects. Future research should explicitly address the im-

pact of industry-specific influences on MNE's entry timing. Second, while China as a host country of investment

provides diverse institutional and com-

petitive conditions across provinces, we were not able to test for the impact of state and provincial differences on MNE's entry timing strategy due to the lack of provincial level data for the 1979- 1996 period. It is reasonable to assume that some state/provincial regulations

preclude entry by foreign firms (Nike ac-

knowledged that the company was forced to go to Tianjin and Shanghai in 1980, and not their initial choices), or

specify the nature of entry mode (cur- rently, in many provinces, entry can

only occur through joint ventures, for

example). In such cases, the order of en-

try can be biased, if not changed as a

consequence of these regulations. Be- cause of this limitation, a more penetrat- ing account of the timing entry decision can be gleaned more effectively with a

disaggregated analysis at the provincial and local level.5 Finally, while we only examine the first entry of firms in China, the fact is most of these Fortune 500 firms have multiple entries over time. Thus, in order to obtain a more compre- hensive account of the timing of entry decision, specifically how previous investments and resources influence sub-

sequent decisions, future research should examine the sequence of entries by each firm over time. Moreover, it might be instructive to study the pattern of se-

quential entry as well as what point in time firms decide to switch from one mode of entry to another. For example, when do firms decide to move from non-

equity modes to equity modes of entry in an international market? What deter- mines the timing decisions of sequential entries in a foreign market? Such studies will allow us to understand when and how multinational firms enter new markets, conduct operations, and learn from their experiences as an integral part of their global strategies.

NOTES

1. We acknowledge the suggestions and guidance of two reviewers in our

developing an enlarged discussion of firm size.



2. We thank the reviewer for making this suggestion.

3. Within event history analysis, it is

acknowledged that left censoring (a case of firm entry before 1979) can be a problem unless cases are negligibly small (Tuma and Hannan, 1979). In our study, this situation is not problematic in that, prior to 1979, China had severely re- stricted, if not allowed foreign entry. Thus, left censoring is not a problem due to historical context and circumstances.

4. Since our data set did not meet the

proportional hazard assumptions, we chose accelerated failure time model to

analyze the entry timing of firms into China.

5. We are grateful to the reviewer for

pointing this out.

REFERENCES

Agarwal, Sanjeev & Sridhar N. Ra- maswami. 1992. Choice of Foreign Market Entry Mode: Impact of Owner-

ship, Location, and Internalization Factors. Journal of International Busi- ness Studies, 23 Spring: 1-27.

Beamish, Paul W. 1993. The Character- istics of Joint Ventures in the People's Republic of China. Journal of Interna- tional Marketing, 1(2): 29-48.

Bell, Michael, Hoe Ee Khor & Kalpana Kochhar. 1993. China at the Threshold of a Market Economy. Paper No.107, International Monetary Fund Paper, Washington, D.C.

Boisot, Max & John Child. 1996. From Fiefs to Clans and Network Capitalism: Explaining China's Emerging Eco- nomic Order. Administrative Science

Quarterly, 41: 600-628. Brewer, Thomas L. 1993. Government

Policies, Market Imperfections, and

Foreign Direct Investment. Journal of International Business Studies, 24(1): 101-20.

Buckley, Peter J. & Mark C. Casson. 1981. The Optimal Timing of a Foreign Di- rect Investment. The Economic Jour- nal, 91:75-87.

& . 1998. Analyzing For-

eign Market Entry Strategies: Extend-

ing the Internalization Approach. Jour- nal of International Business Studies, 29(3):539-562.

Casson, Mark. 1987. The Firm and the Market: Studies on Multinational En-

terprise and the Scope of the Firm.

Cambridge, MA: MIT Press.

Chandler, Alfred D. 1962. Strategy and Structure: Chapters in the History of Industrial Enterprise. Cambridge: M.I.T Press.

Chang, Sea Jin. 1995. International Ex-

pansion Strategy of Japanese Firms:

Capability Building Through Sequen- tial Entry. Academy of Management Journal, 38(2):383-407.

Cohen, Wesley M. 1996. Firm Size and the Nature of Innovation Within Industries: The Case of Process and Product R&D. Review of Economics & Statistics, 78(2): 232-43.

De Castro, Julio 0. & James J. Chrisman. 1995. Order of Market Entry, Compet- itive Strategy, and Financial Perfor- mance. Journal of Business Research, 33:165-77.

Dunning, John H. 1988. The Eclectic Par-

adigm of International Production: A Restatement and Some Possible Exten- sions. Journal of International Busi- ness Studies, 19(1):1-31.

Evans, William M. 1976. Organization Theory: Structures, Systems, and En- vironments. New York: Wiley.

Green, Donna H., Donald W. Barclay & Adrian B. Ryans. 1995. Entry Strategy and Long Term Performance: Concep- tualization and Empirical Evidence. Journal of Marketing, 59:1-16.


TIMING OF ENTRY

Hennart, Jean-Francois & Young R. Park. 1994. Location, Governance, and Stra-

tegic Determinants of Japanese Man-

ufacturing Investments in United States. Strategic Management Journal, 15:419-36.

Hill, Charles W. L., Peter Hwang & W. Chan Kim. 1990. An Eclectic Theory of the Choice of International Entry Mode. Strategic Management Journal, 11:117-128.

Hsiao, C. 1986. Analysis of Panel Data, NY: Cambridge University Press.

Hymer, Stephen. 1976. The Interna- tional Operations of National Firms,

Cambridge, MA: MIT Press. Isobe, Takehiko, Shige Makino & David

B. Montgomery. 2000. Resource Com- mitment, Entry Timing, and Market Performance of Foreign Direct Invest- ments in Emerging Economies: The Case of Japanese International Joint Ventures in China. Academy of Man-

agement Journal, 43(3): 468-484. Kalbfleisch, John D. & Ross L. Prentice.

1980. The Statistical Analysis of Fail- ure Time Data. New York: Wiley.

Kerin, Roger A., P. Rajan Varadarajan & Robert A. Peterson. 1992. First-mover

Advantage: A Synthesis, Conceptual Framework, and Research Proposi- tions. Journal of Marketing, 56:33-52.

Keefer, Philip & Stephen Knack. 1997.

Why Don't Poor Countries Catch Up? A Cross-National Test of An Institu- tional Explanation. Economic Inquiry, Vol. XXXV:590-602.

Knickerbocker, Fredrick T. 1973. Oli-

gopolistic Reaction and Multinational

Enterprises. Boston, MA: Harvard Uni-

versity Press. Kim, W. Chan & Peter Hwang. 1992.

Global Strategy and Multinationals'

Entry Mode Choice. Journal of Interna- tional Business Studies, 29-53.

Kimberly, John, R. 1976. Organizational Size and the Structuralist Persepective: A Review, Crtique, and Proposal. Ad- ministrative Science Quarterly, 21(4): 571-597.

Kobrin, Stephen J. 1991. An Empirical Analysis of the Determinants of Global

Integration. Strategic Management Journal, 12: 17-31.

Kogut, Bruce & Kulatilaka N. 1994. Op- erating Flexibility, Global Manufactur-

ing, and the Option Value of a Multi- national Network. Management Sci- ence 40(1), 123-139.

Lambkin, Mary. 1988. Order of Entry and Performance in New Market. Strategic Management Journal, 9:127-40.

Li, Jiatao. 1995. Foreign Entry and Survival: Effects of Strategic Choices on Performance in International Mar- kets. Strategic Management Journal, 16:333-51.

Lieberman, Marvin B. & David B. Mont-

gomery. 1998. First Mover (Dis)advantages: Retrospective and Link with the Resource-based View, Strategic Man-

agement Journal, 19:1111-1125. Lilien, Gary L. & Eunsang Yoon. 1990.

The Timing of New Entry: An Explor- atory Study of New Industrial Prod- ucts. Management Science, 36 (5):568- 85.

Luo, Yadong. 1998. Timing of Invest- ment and International Expansion Performance in China. Journal of Inter- national Business Studies, 29 (2):391- 408.

& Mike W. Peng. 1998. First Mover Advantages in Investing in Transitional Economies. Thunderbird International Business Review; 40(2): 141-163.

Madhok, Anoop. 1997. Cost, Value and

Foreign Market Entry Mode: The Transaction and the Firm. Strategic Management Journal, 18 (1):39-61.



Mascarenhas, Briance. 1992. First Mover Effects in Multiple Dynamic Markets.

Strategic Management Journal, 13 (3): 237-243.

. 1997. The Order and Size of En-

try Into International Markets. Journal

of Business Venturing, 12:287-299. Mitchell, Will. 1989. Whether & When?

Probability and Timing of Incumbent's

Entry Into Emerging Subfields. Ad- ministrative Science Quarterly, June: 208-229.

. 1991. Dual Clocks: Entry Order Influences on Incumbent and New- comer Market Share and Survival: When Specialized Assets Retain Their Value. Strategic Management Journal, 12 (2):85-100.

Oksenberg, Michael, L. Swaine & L.

Lynch. 1998. The Chinese Future. Santa Monica, CA: Pacific Council for Asia-Pacific Policy.

Pan, Yigang & Peter S K Chi. 1999. Fi- nancial Performance and Survival of Multinational Corporations in China.

Strategic Management Journal, 20: 359-374.

, Shaomin Li, & David K. Tse. 1999. The Impact of Order and Mode of Market Entry on Profitability and Market Share. Journal of International Business Studies, 30(1):81-104.

. 1996. Influences on Foreign Eq- uity Ownership Level in the Joint Ven- tures in China. Journal of International Business Studies, 27 (1): 1-26.

& David K. Tse. 2000. The Hier- archical Model of Market Entry Modes. Journal of International Busi- ness Studies, 31(4): 535-554.

Porter, Michael E. 1991. Towards a Dy- namic Theory of Strategy. Strategic Management Journal, Winter Special Issue, 12:95-117.

Reddy, Srinivas K., Susan L. Holak, & Subhod Bhat. 1994. To Extend or Not

to Extend: Success Determinants of Line Extensions. Journal of Marketing Research, 31:243-262.

Robinson, William T., Claes Fornell, &

Mary Sullivan. 1992. Are Market Pio- neers Intrinsically Stronger Than Later Entrants? Strategic Management Jour- nal, 13 (8):609-24.

Rivoli, Pietra & Eugene Salorio. 1996. For-

eign Direct Investment and Investment Under Uncertainty. Journal of Interna- tional Business Studies, 26:335-358.

Shaver, J Myles, Will Mitchell, & Ber- nard Yeung. 1997. The Effect of Own- Firm and Other Firm Experience on

Foreign Direct Investment Survival in the United States. Strategic Manage- ment Journal, 18(10):811-824.

Sullivan, Daniel. 1994. Measuring the

Degree of Internationalization of a Firm. Journal of International Busi- ness Studies, 25(2):325-342.

Tan, Benjamin & Ian Ventinsky. 1996.

Foreign Direct Investment by Japanese Electronics Firms in the United States and Canada: Modeling the Timing of

Entry. Journal of International Busi- ness Studies, 27(4): 655-681.

Tuma, Nancy B. & Michael T. Hannan. 1979. Approaches to the Censoring Problem in Analysis of Event Histo- ries. In K.F. Schuessler (ed.), Sociolog- ical Methodology. San Francisco: Jos- sey Bass: 209-240.

& . 1984. Social Dynamics: Models and Methods. Orlando, FL: Ac- ademic Press Inc.

Woodcock, C. Patrick, Paul W. Beamish &

Shige Makino. 1994. Ownership-Based Entry Mode Strategies and International Performance. Journal of International Business Studies, 25 (2):253-73.

Wernerfelt, Birger & Aneel, Karnani. 1987. Competitive Strategy Under Un-

certainty. Strategic Management Jour- nal, 8:187-194.


Documents

Timing of Entry in International Market: An Empirical ... · Gerardo R. Ungson*** SAN FRANCISCO STATE UNIVERSITY This study attempts to explain the timing of entry of firms in interna-