Does institutional theory explain foreign location choices in fragmented industries?
There has been a comprehensive review in the International Business (IB) literature on the behavior of firms in concentrated industries and their expansion across national borders. Beginning with the Harvard Multinational Project in the 1960s, IB scholars such as Raymond Vernon and Frank Knickerbocker posited that oligopolistic industries were prone to make aggressive-defensive moves in order to maintain the equilibrium in the subject industry. As such, and because these firms had a valuable position to protect, the decision for foreign direct investment (FDI) was reactionary for many firms. The focus of these studies was not only industries that had oligopolistic tendencies but that were also involved in manufacturing. This combination of highly concentrated, or oligopolistic, industries and manufacturing concerns was logical for the time since most data that was available to academics were of this sort.
One problem with the study of that type of industry makeup is that it represents only a fraction of businesses that operate in the United States. As manufacturing has been displaced overseas, the U.S. economy has become ever more service oriented (Department of State Website 2006). Scholars, however, have not moved with the changing economy as many current studies are interested in diminishing industry structures. Certainly I am not arguing that these industries are irrelevant; instead, the point is that IB needs to incorporate more diverse sets of industry populations even if data collection and theory building is more difficult in this environment.
The central question of this study, as it pertains to international business scholarship, is why do firms that are in fragmented industries compete away from their home country? In other words, it can be completely understood why firms in highly concentrated industries move overseas as they must compete with their few major competitors. However, it is not so evident why firms in fragmented industries would do so. Theoretically, fragmented industries have numerous small sellers who, in many cases, do not compete directly for the same buyer's business. Following on this point, tit for tat moves are not common in highly fragmented space yet many of these firms still operate on the worldwide stage.
This study will focus on the Real Estate industry as a test case for fragmented industries for several reasons.. First, real estate is a highly fragmented industry (U.S. Census Bureau 2008; Ahluwalia and Chapman 2000; Forgey et. al. 1996; Frey 1996; Mahajan 2006; Porter 2003) nearing perfect competition. Even in sub-industries within this broad category (i.e. operative homebuilders, brokerage firms, construction, hotels, etc), one cannot find highly concentrated activity. Secondly, the real estate industry, being in the service sector, can give academics and practitioners a new angle on FDI. With these points in mind, I will focus on real estate firms that are based in the United States and have expanded internationally in order to discern what country characteristics are present for fragmented industry participants to enter. Therefore, in additional to fragmentation as a key topic, location choice will be the outcome studied.
I argue that the current literature neglects industries that do not have consolidated structures and that this neglect may be caused by several factors. First, the study of concentrated industries is facilitated by a greater amount of data. Secondly, by their nature, concentrated industries have less actors to follow which makes this type of study more feasible. Thirdly, many scholars discount industry structure. In other words, in the IB literature, there tends to be an ever increasing focus on entry mode, technology and knowledge transfer, and the like but industry structure is held constant. In other words, if entry mode is the focus of a study, either industry structure is lost because it is not explained or it is lost because fragmented industries are not included. The contribution of this paper is to focus on a fragmented industry in the context of foreign location choice.
Institutional Theory may help to explain many of the activities of Multinational Enterprises (MNE) in both the motivations to move across boundaries and the resultant behaviors once those boundaries are crossed. Following the variation of Institutional Theory proposed by Meyer and Rowan (1977) and developed by DiMaggio and Powell (1983) as well as Scott and Meyer (1983), institutions are considered to be both formal and informal and, as such, can have a profound effect on those individuals or firms that must interact with them. In the context of location choice for U.S. based firms, informal and formal institutions are government structures. This point needs to be expanded. Governments and their agencies are both formal structures in that they operate and create laws and informal in that individuals that have power in the formal structures often create informal entities within government. An important trait of this line of theory is that environments can have multiple institutional setting which help in defining legitimacy in the system (Scott 1987).
An example of the dichotomous (i.e. Formal and Informal) relationship within government structures is that of a regulatory body. Although national or local governments may set forth the legal structure of how industries and firms are regulated, subsystems within the regulatory body can have a persuasive influence on the regulatory outcome (Fineman 1998). If there is an asymmetry between the de jure law and the de facto law, one can argue that this is a lack of regulatory quality. Additionally, some subsystems in weak governmental structures become corrupt. In a corrupt system, meritorious work is considered inferior to both connections and bribes. All of these traits create informal institutions within formal institutions which effect not only firms that are based in those countries but also the location choices made by firms based in other countries.
Two other concepts that are brought to the fields of Strategy and IB through the Sociology literature are that of mimetic isomorphism and legitimacy. Although these concepts are intertwined at some level, they are treated here as separate and stand alone concepts.
Mimetic Isomorphism deals with behaviors of actors in the social structure that act to duplicate or mimic other actors (DiMaggio and Powell 1983; Lai et. al. 2006; Scott 1987). In IB, mimetic behaviors encompass a "follow the leader" mentality and can be seen in many instances in the marketplace. However, mimetic behavior need not be done based on an industry leader; one firms can mimic an industry equal or even a firm considered inferior. Examples of industry-based mimetic isomorphism are seen in cost cutting schemes, corporate structure, and executive pay. As a strategy, mimicry enables firms to reduce risk by following methods and processes that are successful or, at least, appear to be. As such, firms need to be either innovators or imitators in their decision to become international concerns.
In manufacturing, internationalization can be accomplished through export, licensing, joint venture or foreign direct investment. This is not the case with service industries where no concrete product is in question. First, export is not applicable. Secondly, while feasible, joint venture may be less valuable for a service firm because there are limited advantages when production is absent from the equation. Thirdly, licensing and franchising is appropriate for some service-oriented firms such as hotels yet not for others such as homebuilders. FDI can give service firms the quickest access to foreign markets in that there are lower set-up costs relative to manufacturing firms.
Reverting back to mimetic isomorphism and legitimacy, theory indirectly states that firms in fragmented industries should be less prone to imitation. This could be true for two major reasons. First, firms who participate in perfect competition environments have less at stake to protect. In an oligopolistic structure, if a firm does not protect its share, then its competition will move to capture it. However, industries that are fragmented create an environment that is less about protection because both the input markets and output markets tend to be fragmented (Briesmeister and Fisher 1998; Caves and Porter 1978; Dess 1987; Dollinger 1990). Secondly, fragmented industry actors may not know what their competition is doing because information tends to also be fragmented and asymmetrical. Taking this a step further, firms may not completely understand who their competition is or which firm is a market leader.
If this is the case, then how does a firm gain legitimacy? Legitimacy (Von Bertalannfy 1975; Terreberry 1971; Scott 1987) can be earned by a firm by numerous societal stakeholders namely customers, suppliers, competitors, trade groups, and the like. In an international setting, host governments and local social structures are also a source of legitimacy and, as such, can have a significant influence on entry choice by firms. Conversely, governments can be legitimatized by entrants. Although the literature covers firm legitimacy, I argue that legitimacy goes both ways. In countries where firms aggregate, these firms act to legitimatize doing business in that country. Additionally, fragmented industry firms legitimize to a greater extent than do concentrated industry firms because, theoretically, fragmented industries should be less prone to follow the leader or competitors.
Firms in concentrated space follow the leader because they have to maintain legitimacy. The only way to do this is through mimetic isomorphism and, along the way, to gain a competitive advantage through small asymmetries in their own structure and resources (Barney 1991). Simply put, there is just not enough difference in rigid industries that have only a few major players to set firms apart from each other. However, if one looks at the opposite structure, one would expect to find the opposite. That is, fragmented industries should have a wider number of influential societal stakeholders that give their firms legitimacy. This may mean that each influential stakeholder's power is diluted in its effect on the subject firm.
The literature concerning international location choice is extensive yet multi-directional. To make a case for the theory, variables, industry and industry structure chosen for this study, I break down the literature review into three sections.
Institutional theory has had a limited exposure in the context of International Business studies. However, a number of studies have utilized it including Brouthers (2002) which combines Institutional Theory and Transaction Cost Economics; Oliver (1997) uses broad influences such as government and society in the context of firm competitive advantage; Aguila et. al. (2004) in explaining the international technology management of information systems; and Lai et. al. (2006) who describe the theory in the context of global supply chains.
The variable types included in this study can be classified as related to cultural distance, corruption, economic freedom, and institutional quality. Several empirical works in the IB literature have focused on cultural factors including Benito and Gripsrud (1991); Brouthers (2002); Brouthers and Brouthers (2001); Kogut and Singh (1988); Rivoli and Salorio (1996); Loree and Guisinger (1994); and Shenkar (2001). Additionally, Tihyani, Griffith and Russell (2005) created a meta-analysis of cultural distance in entry mode and diversification.
Other studies which have focused on different variables of interest are Habib and Zurawicki (2002) who studied the impact of corruption on FDI; Lorree and Guisinger (1994) who examined policy variables including political stability and infrastructure; and Globerman and Shapiro (2002) who examined the statistical importance of governance attributes such as legislation, regulation and legal systems.
Therefore, there are many works that give the variables in this work precedent to be used in regard to FDI and location choice. The extant literature which incorporates cultural distance includes measures both from Hofstede (2004) and Kogut and Singh (1988). Additionally, the work by Rioli and Salorio (1996) deals with uncertainty and its relationship with FDI.
Industry and Industry Structure
Although the inclusion of the real estate industry in the IB literature is infrequent, there are several studies that incorporate it. Ball (2007) writes on the dichotomy between localization and globalization in the real estate services industry while Gotham (2006) stresses the regulatory structure of the state in the context of the U.S. real estate industry restructuring.
In terms of industry structure, there are relatively few works that handle industry structure that is highly fragmented. Early works as well as some recent ones deal with the exact opposite--oligopolistic industries. These include Vernon (1974, 1979), Knickerbocker (1973), Yu and Ito (1988), Ito and Rose (2002) amongst many others. As described earlier, much of the data compiled from the 1960s onward was that of manufacturing industries that had very high concentration. Therefore, the academic literature was skewed toward the study of such industries because researchers could only empirically test the data that was available.
Finally, as discussed in the opening section, service industries have received more attention in the past decade of research but still not enough in proportionality to the real world. Nig, Cho and Krishnan (1986) as well as Miller and Parkhe (1998) study the international expansion and entry mode of banking firms.
Real Estate and Real Estate-Related
The real estate industry including hotels1 includes all firms that work primarily with real property. Major sub-groups are operative builders, general contractors, real estate brokerages, title insurance companies, real estate lessors, and hotels. This industry was chosen because it characterizes one that has been neglected in the Strategy and IB literature except for a few instances. Additionally, real estate is an interaction of two industry types which have not been prevalent in location studies. This interaction includes service industries which are also highly fragmented. Why is this interaction important to the literature?
The short answer is that fragmented service industries have not received the attention that they deserve. Concentrated industries consumed the early IB literature for good reason. First, data sets compiled in the 1960s and 1970s were of concentrated, and often oligopolistic, industries because it was much easier to collect data in this space. Secondly, many early studies were interested in manufacturing concerns which tended to be oligopolistic. However, the composition of the U.S. economy has changed in the past fifty years. The majority of U.S. business revenues are now derived from the service sector yet the academic literature has not kept up proportionately (Department of State Website 2006).
Another reason that this industry type is important is that, by its nature, it is filled with anomalies. A fragmented industry has low barriers to entry which leads to numerous small entrants. Some of these entrants remain alive for a long period of time (i.e. Holiday Inn, Centex Homes) while some enter and exit with great frequency. Therefore, the composition of the industry and, derivatively, the composition of international entry are dynamic in nature.
The real estate industry has some attributes worth discussing which may be considered anomalies. First, since much of the industry is land-based, location choice is important. Although the macro choice of which country to locate is important, some countries have complex local legal and regulatory structures that either are in addition to or replace the national ones. Secondly, property rights are inherently important. In an extreme environment, property rights are either de facto or de jure absent. However, even in more liberalized regions, property rights' law can be ambiguous and difficult to fully protect. This protection, therefore, is a transaction cost borne by firms which is meant to counteract value dissipation (Foss and Foss 2005) by local forces. Thirdly, many of the sub-groups in certain real estate industry expand their influence through different mechanisms. While export is irrelevant, licensing is sparse and used mainly for branding. An important governance mechanism is that of franchising. Real estate brokerages and hotels, for the most part, expand and govern through this structure. However, it tends to be the parent company (Franchisor) which chooses the location of expansion.
Characteristics of Fragmented Industries
Fragmented industries have several common characteristics that are present albeit to different extents. Porter (1980) describes fragmented industries as those that contain all or some of the following:
* Low Entry Barriers
* Lack of Power Advantages with Buyers
* Lack of Power Advantages with Suppliers
* Lack of Economies of Scale
* Lack of Economies of Scope
* Regional Issues such as High Transport Costs
All points have merit but three are key. First, low entry barriers are a given when discussing fragmentation because ease of entry disallows large firms to set the tone. Simply put, the lack of high entry barriers is an open invitation for more small companies to join the industry. One finds that in fragmented industries a plethora of small firms, many privately held operating alongside large, publicly traded corporations (Wright et al. 2004). The second common trait is a lack of power advantages with buyers in fragmented industries as buyers often are equally fragmented. Because buyer consolidation, usually through multiple distribution levels, is absent in these industries, power is absorbed in demand as opposed to supply (Briesemeister and Fisher 1998). This tends to lead to a greater variance in pricing than if buyers were a more solid cohort. The third aspect is low economies of scale which contribute to industry fragmentation as sparse resource origins supply different industry actors but in a non-unified way. As will be seen, low economies of scale in fragmented industries are not in all inputs but often in a few important ones.
Porter's definition concentrates on industries that tend to produce something. However, in the new economy, many firms that are service oriented tend to be in fragmented space. With the advance of technology, especially the internet, firm entry in the service sphere has become increasingly more tenable. This study combines the interaction between the service sector and fragmented structure.
Measurement of Fragmentation
Fragmentation measures typically have two variants. First, there are Concentration Ratios (CR) which measure a certain number of industry leaders. [CR.sub.4] and [CR.sub.8] levels are most common in the academic literature and they measure the top four or eight industry leaders, respectively, by market share. Mathematically, these levels are represented by:
[n.summation over (i=1)]MSi
In this case, the summation of market shares (MS) of N market participants are simply calculated to derive at a number. In a highly concentrated industry, a [CR.sub.4] level is greater than 40 percent and often above 60 percent (Caves and Porter 1978). In a fragmented industry, these levels may fall to below 20 percent. The closer an industry approaches perfect competition, the closer the [CR.sub.4] level approaches zero.
Another measure of industry concentration is the Herfindahl-Hirshman Index (HHI). This differs from the previous equation in that the market shares are squared as shown by the equation:
[n.summation over (i=1)][(MSi).sup.2]
The major difference is that the HHI (Federal Reserve Bank of Atlanta 1993) accounts for large industry players by squaring the market shares. The index has values that range from perfect competition (0) to pure monopoly (10,000). As a mental exercise, think of two industries. One has market leaders with respective market shares of 50 percent, 10 percent, 10 percent, and 10 percent while the other has a split of four leaders with 20 percent per firm. In each case, the CR4 level equates to 80 and this industry would be considered oligopolistic (CR4>60). However, if the market shares are squared to derive at an HHI value, Industry 1 has HHI=2,800 while Industry 2 has HHI=1,600. (2) These two values are significantly different because the HHI accounts for firms with very large market shares exponentially while the CR levels are a simple summation.
However, in a fragmented industry, the effect is reversed. Imagine an industry where there are 10 players with an average of four percent of market share and then a number of tiny firms each with less than one percent of the market share. The HHI would equate to 10*([4.sup.2]) or 160 plus the aggregated amount of the small firms. However, since the square of a market share under one produces a smaller value, even if there were hundreds of small firms, the aggregated figure would not have a significant influence on the HHI. In this case, squaring small values keeps the HHI low.
This study will attempt to find empirical evidence to support or refute Institutional Theory in general and both mimetic isomorphism and legitimacy in particular. Additionally, through a simple cluster analysis, I attempt to find out if characteristics of certain countries can be proxies for legitimatization between firms in the real estate industry and those countries.
The study is designed as follows. First, samples of leading firms in the real estate industry were chosen. Leading firms were chosen because there is much more detailed information about their previous international entry choice than smaller, private firms. Secondly, a count was tabulated as to which countries these firms chose to locate. This second point also entails that countries were noted where no companies chose to locate. Thirdly, the country sample was divided into two. The first sample is comprised of 68 countries that were entered and the second is comprised of 22 countries which were entered from zero to one percent relative to the total amount of entries. Fourth, a probit regression was utilized to model the propensity of firms to choose the first set of countries over the second set. The binary dependent variable (1=High Entry, 0=No Entry) was regressed on country data which will be described in the Data Section.
The specific hypotheses to be tested include:
H1: Firms that are located in fragmented industries will tend not to mimic competitors in their choice on international entry.
H2: Nations that are perceived to have high corruption are less likely to attract firms in fragmented industries.
H3: Nations that have higher economic freedom are more likely to attract firms in fragmented industries.
H4: Nations that have a high power distance are more likely to attract firms in fragmented industries.
H5: Nations that have low individualism/high collectivism are more likely to attract firms in fragmented industries.
H6: Nations that have high uncertainty avoidance are more likely to attract firms in fragmented industries.
H7: Nations that are politically stable, have an effective government, have high regulative quality and a strong rule of law will tend to attract fragmented industry firms
Data for this analysis was retrieved using multiple sources. First, a comprehensive list of real estate firms was found in Hoover's Directory. Of these firms, a sample was selected of leading firms in four industry subgroups--Commercial Brokerage, Residential Brokerage, Property Managers, and Hotels. Some subgroups, such as operative homebuilders, were intentionally not selected because they almost never directly compete internationally.
Secondly, country data were derived from several sources:
1. Data concerning cultural issues were taken from the Geert Hofstede Cultural Dimensions website.
2. Corruption Perceptions Index values come from Transparency International.
3. An Index of Economic Freedoms was compiled from the Heritage Foundation.
4. Data concerning Political Stability, Government Effectiveness, Regulatory Quality, Rule of Law, and the Control of Corruption were taken from the Kaufman et al. survey (2006)
5. Finally, economic variables of interest were found at the World Development Indicator Database.
It is important to reiterate the unit of analysis. This study is based on a country or country by industry unit of analysis evidenced by the documentation of entry into countries by firms in the real estate industry. The dependent variable is based on a count of entries into the 90 countries in question. The model, which will explained below, is based on the number of times a country was entered by firms in the real estate industry. Additionally, all independent variables are based on country data which supports the unit of analysis for the study.
Probit Regression is the technique used in this analysis. Probit analysis is relevant when the response variable has a binary value (Johnston, 2007; Tabacknick and Fidell 2007). In this case, the Dependent Variable is Entry or No Entry and, therefore, a link function of the Generalized Linear Model (GLM) is appropriate. Typical linear regressions require that the dependent variable is continuous and not either binary (Bounded between 0 and 1) or discrete (Bounded between 0 and [degrees][degrees]) More specifically, the probit regression to be modeled:
P(Y= 1 given that X = x) = 0(X'[beta]) (1)
Manipulating both sides yields:
[0.sup.-1]P(Y = 1 given that X = x) = (X'[beta]) (2)
In this case, I am modeling the propensity, or probability, that the dependent variable (Y) is equal to 1 (1=Entry), given that we have certain independent variables (X). Additionally, probit analysis is based on the central density function (0) which is shown in Equation (2). (X'[beta]) represents a vector of both independent variables and Beta coefficients which will be described in detail in another section of this analysis. The cumulative central density function (CDF) is represented by:
[[integral].sup.Z.sub.-[infinity]] 1/[square root of 2[pi]] exp (1/-2 [Z.sup.2])d z
Therefore, I am modeling the propensity that a firm will enter those countries that have had the highest density of previous entry (Y=Entry=1) based on the explanatory variables listed in Section 6.2 below.
As discussed previously, the response variable is a binary dependent variable with takes the value of 1 for Entry and a value of 0 for No Entry. In this study, it is important to note the mechanism for calculating this. 90 countries were included in the study of which 68 were included in the category of Entry and 22 were included in the category of No Entry. No entry, in the case of this study, was defined as countries that were entered from zero to three times in total. In the 90 country sample, there were a total of 509 entries with the range, per country, of zero to 19. A country was labeled as No Entry if it had between zero and one percent of total entries.
Gross Domestic Product (Ln_GDP).
Gross Domestic Product is utilized as an independent variable by taking its natural log. GDP is important as it can explain how a nation's economic size influences firms to enter. Using the natuaral log of GDP is useful in that it narrows the large range of data values that occurs when using the actual GDP.
Hofstede Power Distance (Hof_PDI).
The Power Distance of a nation is the degree to which individuals in that culture accept the difference between those that have power and those that do not. For this study, Power Distance is used to measure a nation's tolerance for the powerful to run the institutions, both formal and informal, in the country. For this study, PDI is expected to have a negative association with the propensity of a firm to enter.
Hofstede Individualism (Hof_Ind).
Individualism measures the "extent that individuals are integrated into groups (Hofstede 2004)." In other words, does a culture nurture the individual or does it value being part of some bigger unit of analysis. The higher the Individualism score, the more the culture in question values individual efforts.
Hofstede Masculinity (Hof_Mas).
Masculinity measures the gender roles in a culture. Higher masculinity scores means that the culture is more assertive while lower masculinity scores can be interpreted as being more caring.
Hofstede Uncertainty Avoidance (Hof_UAI).
Uncertainty avoidance measures the collective degree to which cultures or nations avoid uncertain situations both in business settings as well as other types of settings. Hofstede states that it is the extent to which "... a culture programs its members to feel either uncomfortable or comfortable in unstructured situations."(Hofstede 1980). UAI is important as a variable because as a nation's uncertainty scores rise, it can be stated that this nation has more rules and attempts to control outcomes more frequently.
Corruption Perceptions Index (CPI).
The Corruptions Perception Index is produced by Transparency International and is a perceived score of corruption by experts surveyed by the organization. A higher score (Close to 10) is indicative of a nation that is free of corruption whereas a score closer to zero represents a highly corrupt society. The CPI is used in the model because corruption of institutions of government can affect the decision of firms to enter these countries. In other words, the more corrupt a country is perceived to be, the more its institutions, both formal and informal, are not aligned to merit.
Index of Economic Freedoms (IEF).
The Index of Economic Freedoms measures the extent to which a nation has a free atmosphere to conduct economic activity. It can be argued to be, in the case of a low score, a measure of a nation's propensity to have non-market institutions control a greater degree of the economy.
Kauffman Survey Variables.
The Kaufman Survey rates nations on six important variables on a standardized scale with a zero mean. Four of the six variables are important for this study and include:
1. Political Stability (Pol_St)
2. Government Effectiveness (Gov_Eff)
3. Regulatory Quality (Reg_Qlt)
4. Rule of Law (Rule_Law)
These variables are directly relevant for the study because institutional theory, as asserted by Scott (1995), deals with institutions which are "... social structures that have attained a high degree of resilience. They are composed of cultural-cognitive, normative and regulative elements ..." As government effectiveness, political stability, regulatory quality and rule of low dissipate at the national level, firms should reject these nations. These variables are important for the institutional legitimacy of the nation in question because institutional stability should be paramount to FDI.
RESULTS AND DISCUSSION
Table 1 includes the results of several models of the Probit Regression. The full model (Model 1) tested all variables simultaneously. Model 1 corresponds with Hypothesis 1 which is a test of Institutional Theory. H1 is rejected and, therefore, mimetic isomorphism and legitimacy factors do contribute to fragmented industry. Support for the alternative hypothesis, in this case, is support for Institutional Theory within industry structures that should not portray such behaviors. Predictors that were significant in Model 1 include the Ln of GDP, Hofstede's Uncertainty Avoidance and Regulatory Quality.
Model 2 tests all variables except for Hofstede's Individualism and Masculinity variables which were both insignificant in all iterations of the probit analysis. Although the same three predictors were found to be significant in this model as in the previous one, the AIC value improved (Smaller is better) from 62.25 to 59.24.
Model 3 discriminated against all non-significant predictors and included just those independent variables that held highly significant in Models 1 and 2. This model was both the most significant and parsimonious of all iterations with just an intercept and three predictors. The AIC dropped to 53.21 and the model p-value was <0.0001. Furthermore, the intercept, which was insignificant previously, became highly significant along with the three predictors. Model 3 can be written as:
Model 3 states that the propensity of a country to be entered by a highly fragmented firm is positively related to the country's GDP, Uncertainty Avoidance, and Regulatory Quality. In other words, if a country had LN_GDP of $30.00, an Uncertainty Avoidance value of 46 and a Regulatory Quality score of 1.5, then the probability that this country would be entered calculates to standardized z-score of +4.39. These figures just discussed are those of the United States and these results make sense. Although the probability value of z=4.39 is not valid because the Probit results are of firms that are based in the United States, it makes sense that a country such as the United States would have a probability approaching one of being entered.
Likewise, and conversely, if a country had LN_GDP of $20.00, an Uncertainty Avoidance value of 54 and a Regulatory Quality score of -1.2, then the probability of this country to be entered is -3.62. These values come from the African nation of Liberia which is at the other end of the spectrum from the U.S. on many of the economic and regulatory issues in question. With a z-score of -3.62, only 1 in approximately 6,800 firms in fragmented industries would enter Liberia. These predictive results using Probit Analysis in Model 3 seem to offer a model that has statistical significance as well as real-world intuitiveness.
One more model shall be discussed where I aggregated the four Kaufmann variables into one variable: Aggregat. These four variables measure governance structures both formal and informal. In order to capture the four together, I tested the Probit analysis using all predictors. While the model is highly significant, the AIC score was no better than Model 1. However, the Intercept value is significant in Model 4 as well as the new aggregated value (p=.021).
What do these results have to say about the stated hypotheses? Hypothesis 1, which was rejected, confirms Institutional Theory in the context of fragmentation. In my opinion, since fragmented industry firms should be the least likely to group together and be effected by institutions in their location choice, this is very strong support for the theory and the two features of mimetic isomorphism and legitimacy. Simply put, leading firms in fragmented industries follow each other when the country's attributes work toward their favor. In addition, they legitimatize each other in that they mimic the other's location choices. More importantly, country's are legitimatized or delegitimatized through the entries or lack thereof.
Hypothesis 2 is not supported although the predictor CPI is somewhat significant. In Model 4, the p value for CPI was 0.06 which is very close to the significance level. Additionally, the CPI consistently had the most significant value of the insignificant predictors. However, the sign is the opposite of that expected. I hypothesized that as nations become more corrupt (Lower CPI score), the propensity for that nation to be entered would diminish but the opposite is true. As the CPI score rises (This means less corrupt), the probability that the country is entered diminishes.
Hypothesis 3 was not supported in any of the models as the Index of Economic Freedom (IEF) score had no significance. The same was true for Hypothesis 4 and 5 which dealt with Hofstede's Power Distance (HofPDI) and Individualism (HofIND) respectively. Hypothesis 6 received high support and the Uncertainty Avoidance measure was significant in all iterations. The coefficient was positive in all models denoting that as uncertainty avoidance increases so does the propensity for the subject country to be entered. I argue that this is also support for Institutional Theory because countries that are less comfortable in uncertain situations tend to fall back on the institutions, both formal and informal, that make up society. Therefore, since countries that have higher uncertainty avoidance also have a higher entry propensity, this is in line with Institutional Theory.
Finally, Hypothesis 7 was supported. When all four variables--Political Stability, Government Effectiveness, Regulatory Quality, and Rule of Law--were tested, only Regulatory Quality was significant. However, this predictor was the most significant of any original independent variable in all models in terms of consistency. This is an extremely important point in testing Institutional Theory because, as explained earlier in this paper, the regulatory environment of a country is an important attribute when evaluating to enter it or not. Also, regulatory structures encompass both formal (legal) and informal (societal, corruption, etc) aspects of a country's institutional makeup. Furthermore, when the four variables were aggregated, this combined score was significant adding that the combination of good government, a strong rule of law and political stability were important in the legitimization of those countries.
To summarize, I argued that according to Institutional Theory, firms in fragmented space would not mimic their competitors. If this statement stood, it would have undermined the theory. However, this statement was rejected. I also argued that legitimacy which is written in the literature was not always firm to firm but can be from firm to country. The results of this study confirm that a nation, through its various institutional factors, is legitimatized or not by firms in fragmented industries. As would be expected, the relative size of the country mattered. All models had a significant and positive coefficient for LN_GDP which ranged from 0.549 to 0.779 meaning that as the natural logarithm of GDP rose so did the propensity of firms to enter. This makes common sense since there are more opportunities in countries with higher GDP.
However, one point should be made which is not seen in the statistical results. Although it seems like the firms in this highly fragmented industry mimicked and legitimatized each other, there were dispersions. I expected that in fragmented industries which do not have clear and large leaders that firms may choose to enter countries that have less overall entry because this would be opportunistic. Although this does not show up in the mathematical figures, counties such as Lebanon had about as much entry as did Denmark. Anomalies such as this are difficult to bring out in testing but, qualitatively, they are important in studying differing industry structures.
CONTRIBUTIONS, LIMITATIONS AND FUTURE RESEARCH
The main contribution of this paper was to test the validity of an existing theory in the context of industry structure. As stated in the Results, institutional aspects of nations and cultures such as uncertainty avoidance and regulatory quality are important to attract firms. Firms, on the other hand, legitimatize these countries while also mimicking and legitimatize each other. Another contribution of this paper is adding work on fragmented industries to the literature which is much needed. Finally, scholars have slowly begun to work on building research on the service sector in recent years. This paper adds to that growing literature.
This study had several limitations. First, the study was done using one industry and, therefore, the results can be generalized narrowly. I believe that the results can be generalized beyond the real estate industry to other highly fragmented space. However, the results cannot be generalized to industries as a whole because I was interested in testing a certain type of structure. A second limitation is that that there is no differencing between industry structures. The propensities of fragmented firms to enter countries are measured here but what is not measured is how much more or less probable it is for those countries to be entered by a highly concentrated industry. The third limitation is that the sample is comprised of leading real estate firms and, therefore, may not be representative of all real estate firms. One problem in performing research on fragmented industries is that, since they are a compilation of hundreds or even thousands of firms, information is nearly impossible to obtain on tiny firms.
Future research should start with the second limitation. A study comparing the same countries and variables but a different industry structure would result in the difference for Country X to be entered due to Variable Y between fragmented space and concentrated space. Secondly, a qualitative study of the aspects of Institutional Theory--mimetic isomorphism and legitimacy--would be important to find out how fragmented firms copy and legitimate competitors. There are many works concerning these aspects in business research that focus on either all industries or concentrated industries.
Aguila, A., Bruque, S. and Padilla, A. (2002). Global Information Technology Management and Organizational Analysis: Research Issues, Journal of Global Information Technology, 5(4).
Ahluwalia, G. and Chapman, J. (2000). Structure of the Residential Construction Industry, Housing Economics (48) 10: pp. 7-11.
Ball, M. (2007). Localization Versus Globalization: Some Evidence from Real Estate Services Organizations, Journal of Housing Built Environment, Vol 22: pp. 91-106.
Barney, J. (1991). Firm Resources and Sustained Competitive Advantage, Journal of Management (17) 1: pp. 99-120.
Benito, G and Gripsrud, G. (1992). The Expansion of Foreign Direct Investments: Discrete Rational Location Choices or a Cultural Learning Process?, Journal of International Business Studies, 23(3): 461-476.
Briesemeister, D. and Fisher, N (1998). The 5-Cs: A Modern Framework for Consolidation of Fragmented Industries, Trustees of Dartmouth College.
Brouthers, K. (2002). Institutional, Cultural and Transaction Cost Influences on Entry Mode Choice and Performance, Journal of International Business Studies, 33(2): 203-221.
Brouthers, K. and Brouthers, L. (2001). Explaining the National Cultural Distance Paradox, Journal of International Business Studies, 32(1): 177-189.
Caves, R. and Porter, M. (1977). Market Structure, Oligopoly, and Stability of Market Shares, The Journal of Industrial Economics (26) 4: pp. 289-313.
Chen, H. and Chen, T. (1998). Network Linkages and Location Choice in Foreign Direct Investment, Journal of International Business Studies, 29(3): 445-468.
Contractor, F. and Kundu, S. (1998). Modal Choice in a World of Alliances: Analyzing Organizational Forms in the International Hotel Sector, Journal of International Business Studies, 29(2): 325-358.
Dess, G. (1987). Consensus on Strategy Formulation and Organizational Performance: Competitors in a Fragmented Industry, Strategic Management Journal (8): pp. 259-277.
DiMaggio, P and Powell, W. (1983). The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields, American Sociological Review, 48: 147-160.
Dollinger, M. (1990). The Evolution of Collective Strategies in Fragmented Industries, Academy of Management Review (15) 2: pp. 266-285.
Economic Review--Federal Reserve Bank of Atlanta (1993). Calculation of the Herfindahl-Hirschman Index, (78) 1: pp. 28-30.
Fineman, S. (1998). Street-Level Bureaucrats and the Social Construction of Environmental Control, Organization Studies, 19(6): 953-974.
Forgey, F., Mullendore, W. and Rutherford, R. (1996). Market Structure in the Residential Real Estate Brokerage Market, Journal of Real Estate Research (14) 1/2, pp. 107-115.
Foss, K. and Foss, N. (2005). Resources and Transaction Costs: How Property Rights Economics Furthers the Resource- Based View, Strategic Management Journal (26) pp. 541-553.
Frey, E. (2003). Building Industry Consolidation, Builder Magazine and NAHB Economics, August pp. 7-12.
Globerman, S. and Shapiro, D. (2002). Governance Infrastructure and U.S. Foreign Direct Investment, Journal of International Business Studies, 34: 19-39.
Gotham, K. (2006). The Secondary Circuit of Capital Reconsidered: Globalization and the U.S. Real Estate Sector, American Journal of Sociology, 112(1): 231-75.
Habib, M. and Zurawicki, L. (2002). Corruption and Foreign Direct Investment, Journal of International Business Studies, 33(2): 291-307.
Hofstede, G.(2004). Cultures and Organization: Software of the Mind. New York: McGraw-Hill.
Iizuka, M. (2006). Low-Tech Industry: A New Path for Development? The Case of the Salmon Farming Industry in Chile, 4th Globelics in India.
Ito, K. and Rose, E. (2002). Foreign Direct Investment Location Strategies in the Tire Industry, Journal of International Business Studies, 33(3): 593-602.
Johnston, G. (2007). SAS Software to Fit the Generalized Linear Model, The SAS Institute Inc., Cary, NC.
Kim, W. and Hwang, P. (1991). Global Strategy and Multinationals' Entry Mode Choice, Journal of International Business Studies, 23(1): 29-53.
Kogut, B. and Singh, H. (1988). The Effect of National Culture on the Choice of Entry Mode, Journal of International Business Studies, 19(3): 411-432.3-98.
Lai, K., Wong, C. and Cheng, T. (2006). Institutional Isomorphism and the Adoption of Information Technology for Supply Chain Management, Computers in Industry 57(1):
Loree, D. and Guisinger, S. (1995). Policy and Non-Policy Determinants of U.S. Equity Foreign Direct Investment, Journal of International Business Studies, 26(2): 281-299.
Lurz, B. (2002). Consolidation is Temporary, Professional Builder (67) 4: pp. 87-89.
Mahajan, S. (2006). Concentration Ratios for Businesses by Industry in 2004, Economic Trends, Office for National Statistics: pp. 25-47.
Makadok, R. (1998). Can First-Mover and Early-Mover Advantages Be Sustained in an Industry with Low Barriers to Entry/Imitation?, Strategic Management Journal (19) 7: pp. 683-696.
Meyer, J and Rowan, B. (1977). Institutional Organizations: Formal Structure as Myth and Ceremony, American Journal of Sociology, 83: 340-363.
NAHB (2007). www.nahb.org.
Nigh, D., Cho, K. and Krishnan, S. (1986). The Role of Location-Related Factors in U.S. Banking Involvement Abroad: An Empirical Examination, Journal of International Business Studies, 17(3): 59-72.
Oliver, C. (1996). Sustainable Competitive Advantage: Combining Institutional and Resource-Based Views, Strategic Management Journal, 18(9): 697-713.
O'Toole, P. (2002). Consolidation: Swift and Massive, Professional Builder (67) 4: pg. 67.
Porter, M. (1980). Industry Structure and Competitive Strategy: Keys to Profitability, Financial Analysts Journal (July/August): pp. 30-40.
Porter, M. (2003). The U.S. Homebuilding Industry and the Competitive Position of Large Builders, Centex Investor Conference, New York.
Rivoli, P. and Salorio, E. (1996). Foreign Direct Investment and Investment Under Uncertainty, Journal of International Business Studies, 27(2): 335-357.
Scott, W. and Meyer, J. (1983). Organizational Environments: Ritual and Rationality. Sage Publishing: Beverly Hills, CA.
Scott, W. (1987). The Adolescence of Institutional Theory, Administrative Science Quarterly, 32(4): 493-511.
Scott, W. (1995). Institutions and Organizations. Sage Publishing: Thousand Oaks, CA.
Shenkar, O. (2001). Cultural Distance Revisited: Towards a More Rigorous Conceptualization and Measurement of Cultural Differences, Journal of International Business Studies, 32(3): 519-535.
Tabachnik, B. and Fidell, L. (2007). Using Multivariate Statistics. Pearson: Boston, MA.
Terreberry, S. 1971. The evolution of organization environments, Administrative Science Quarterly, 12: 590-613.
Tihanyi, L., Griffith, D. and Russell, C. (2005). The Effect of Cultural Distance on Entry Mode Choice, International Diversification, and MNE Performance: A Meta-Analysis, Journal of International Business Studies, 36: 270-283.
Vernon, Raymond (1974), "The Location of Economic Activity," in John H. Dunning, ed., Economic Analysis and the Multinational Enterprise, London: George Allen & Unwin, Ltd., 89-114.
Vernon, Raymond (1979), "The Product Cycle Hypothesis in a New International Environment," Oxford Bulletin of Economics and Statistics, 41 (November), 255-67.
Von Bertalanffy, L. (1975) Perspectives on General System Theory, George Braziller, New York.
Wright, P. et al. (2004). Are Competitors Advantageous or Disadvantageous in Consolidated Versus Fragmented Industries?, Academy of Strategic Management Journal (3): pp. 47-64.
Yu, C. and Ito, K. (1988). Oligopolistic Reaction and Foreign Direct Investment: The Case of the U.S. Tire and Textiles Industries, Journal of International Business Studies, 19(3): 449-460.
Richard S. Brown, Temple University
Table 1--Probit Regression Results Model 1 Model 2 Model 3 Model 4 Model Significance 0.001 * 0.001 * 0.001 * 0.001 * AIC Value 62.25 59.24 53.21 62.11 Intercept -8.308 -9.389 -14.97 -20.406 p=(.328) p=(.230) p=(.000) * p=(.000) * LN_GDP 0.677 0.674 0.549 0.779 p=(.012) * p=(.007) * p=(.000) * p=(.000) * Hof_PDI -0.027 -0.017 -0.01 p=(.303) p=(.334) p=(.601) Hof_IDV -0.007 0.007 p=(.770) p=(.713) Hof_MAS 0.026 0.01 p=(.332) p=(.623) Hof_UAI 0.035 0.033 0.029 0.036 p=(.017) * p=(.024) * p=(.017) * p=(.005) * CPI -0.769 -0.801 -0.831 p=(.130) p=(.113) p=(.061) IEF -0.101 -0.073 0.047 p=(.191) p=(.291) p=(.265) Pol_St 0.148 0.171 p=(.766) p=(.705) Gov_Eff -0.168 0.283 p=(.919) p=(.855) Reg_Qlt 3.653 3.109 1.038 p=(.026) * p=(.031) * p=(.000) * Rule_Law 0.425 0.123 p=(.701) p=(.900) Aggregat 0.569 p=(.021) * * Significant at .05
|Printer friendly Cite/link Email Feedback|
|Author:||Brown, Richard S.|
|Publication:||Journal of International Business Research|
|Date:||Jan 1, 2011|
|Previous Article:||Top 250 global retailers: on-line features of retailer websites.|
|Next Article:||EPS as a measure of intercompany performance: Philippine evidence.|