Institutional Convergence Towards the United States
Through Globalization
University of Groningen Faculty of Economics and Business
Master Thesis International Economics and Business
Name Student: Andrea Jonker Student ID number: 1770586
Student email: andrea.c.jonker@gmail.com Date Thesis: 17th January 2014
Abstract
1. Introduction
the sample and methodology. Section four shows the empirical results. Section five presents a discussion of the main findings and the limitations of this study. Finally, section six concludes.
2. Literature Review
Since North (1990) the institutional dimension of economic development has become one of the most active fields in economics and in the development of literature (Alesina and Perotti, 1994). Many have presented their definitions and views on institutions and institutional change, which are reviewed in this section. The literature on the direct relationship between globalization and institutional convergence is more scarce. Despite this scarcity, several authors present interesting views regarding this topic that contribute to the theoretical model that is built to answer the research question of this study.
2.1 Institutions
Over the last few decades there has been no unanimity in the definition of institutions (Hodgson, 2006). The term institutions is being used in many disciplines (sociology, politics, geography, economic), which makes it hard to present one detailed definition that could apply for all the different spheres. However, there is a consent on a generalization of the definition that is being used by many authors in the different disciplines. This section presents different definitions and concludes with the relevant definition for this research.
of institutions. These definitions all have a similar essence as North’s definition and could be considered general. Despite the fact that Hodgson (2006) disagrees with the consent on this generalization and states that this has led some writers to give up matters of definition, for this research it is deemed sufficient to define institutions according to North.
2.2 Institutional change
Along with the recognized importance of institutions came the prevalent discussion on institutional change. When discussing institutional change, the literature focusses on actors and dynamics along the following two axes: domestic versus international spheres, and macro-level versus micro-level (Wilson, 2009). The international sphere has been discussed in the literature, however, empirical research on this matter is scarce. Providing both empirical and theoretical insights on the matter could offer a useful contribution to the development discussion. Remarkable about the empirical research that is done on institutional change is that most of the empirical studies available take a unilateral perspective. Numerous papers have generated institutions and its development as an explanatory variable that influences different response variables. However, not many have used institutional development as the response variable in an empirical research. Nonetheless, the contribution of this empirical research is deemed important. Understanding the explanatory variables of institutions and its development may provide an additional piece to the big development puzzle. The development puzzle discusses the sources of growth of countries and helps to explain why some countries are poor and other countries are rich.
2009). In addition, the globalized operations can be of other natures, such as military. This study focusses for globalization mostly on economic and military relations with the US. The next section illustrates why the US is used for this study.
2.3 Role of the United States
Relations with the US are considered of higher importance than strong ties with a small economic and political country. The US is widely viewed as an innovative economy, providing great incentives to its entrepreneurs and workers (Acemoglu et al, 2012). The role of world power is also acknowledged by the American government itself. George W. Bush (2002) stated that the US enjoys a position of unparalleled military strength and great economic and political influence. In addition, Gertler (2001) has also pointed that the US has developed an exemplary role in the world economy. He states that since the resurgence of the US economy beginning in the second half of the 1990s, US practices have apparently become the object of global firms’ affections. This caused large corporations in Europe and Asia to adopt the core characteristics of US-style shareholder capitalism. In conclusion, the role of the US is considered most interesting due to its role as global economic and political power.
2.4.1 Path Dependency and culturalism
dependency, and states that social mechanisms lead to institutional inertia. Furthermore, Levi (1997) claims that the entrenchments of certain institutional arrangements obstruct an easy reversal of the initial choice.
An extension of the path dependency theory is the culturalism perspective. This perspective was initiated by Samuel Huntington in the 1990s. According to Huntington (1993) the great divisions among humankind and the dominating source of conflict will be cultural. Huntington referred to this as the ‘clash of civilizations’ and this clash ought to dominate global politics. Civilization identity will be increasingly important in the future, and the world will be shaped in large measure by the interactions among seven or eight major civilizations. These include Western, Confucian, Japanese Islamic, Hindu, Slavic-Orthodox, Latin American, and possibly African civilization. The most important conflict of the future will occur along the cultural lines separating these civilizations from one another. The clash of civilizations is an extension of the path dependency theory as it states that the civilizations are differentiated from each other by history, language, culture, tradition, and most important, religion. These are initial factors with such a fundamental characteristic which indicates that they are not expected to alter rapidly. This view on culture is shared by Glaeser’ et al. (2004) view that culture is a fundamental channel of historical influences. Thus, both the path dependency and culturalism theory stress that countries have initial factors and choices that determine the development of institutions beforehand without the possibility of alterations in the course of time.
2.4.2 Varieties of Capitalism
A theory that builds on the path dependency theory is the theory of varieties of capitalism. Varieties of capitalism is a theory conceived by Hall and Soskice (2001) in Varieties of
Capitalism: The Institutional Foundations of Comparative Advantage. In their book Hall and
institutions in others. In the political economy as a terrain peopled with entrepreneurial actors seeking to advance their interests as they construe them, actors are constrained by existing rules and institutions but also look for ways to make institutions work for them (Hall and Thelen, 2009). Building on path dependent analyses, authors in the varieties of capitalism school contend that early institutional choices shape a country’s development pattern and deter convergence (Wilson, 2009). However, the difference with path dependency and culturalism is that institutions are best seen, not as a rigid matrix of sanctions and incentives to which actors respond in relatively mechanical fashion, but as resources that actors use to attain their ends, in some cases switching from one institution to another when it seems more likely to serve their purposes. It is important to note that this theory does not take into account non-capitalist economies.
In conclusion, path dependency, culturalism, and varieties of capitalism present different outcomes on institutional development of countries determined by an initial choice, culture, and capitalist style that can hardly or not be altered over time. The hypothesis on institutional developmentthat follows from these theories is:
H1: Institutions of countries tend to diverge over time
As H1 already states, institutions tend to be sticky, thus in the case of institutional divergence, the
effect of globalization and economic relations is not present. The literature underpins this non-effect of globalization and economic relations between countries. The path dependency theory, culturalism, and the varieties of capitalism theory claim that national institutions tend to resist reform, even in the face of globalization (Wilson, 2009). This is due to the fact that domestic firms tend to follow a dominant institutional form due to their embeddedness in the national institutional structure. Consequently, national settings strongly condition institutional choices (Thatcher, 2004). Thus, despite the fact that through economic relations foreign companies present distinct institutional models for reformers and firms in the host country, economic actors tend to be tied to existing systems based on their patterns of asset investment. Thus, specific relations with the US would be of no influence according to these theories.
2.4.3 Modernization
first, and probably the most influential of all the theoretical paradigms predicting global convergence. The theory assumes that socio-economic and political development is a unilinear process taking on almost identical forms in all societies (Herkenrath et al, 2005). Kreutzmann (1998) specifies this theory by stating that modernization is the process which enables backward countries to escape from tradition, to promote and accelerate transition and finally to overcome underdevelopment. Herkenrath et al (2005) acknowledge that in some societies development processes take longer than others, but eventually all societies will be structurally and culturally similar. Thus, regardless of the initial choices and the past of a country, countries will eventually be able to develop towards an equilibrium of development. Institutions are bound to converge according this theory. Institutional convergence can be understood to have occurred when actors in one place adopt the institutions developed elsewhere, leading to a greater homogeneity in institutional form (Hall, 2003). This leads to the following hypothesis:
H2: Institutions of countries tend to converge over time
2.4.4 Globalization
Path dependency, culturalism, and varieties of capitalism present arguments on a domestic level and state that globalization is irrelevant when it comes to institutional reforms. Modernization theory presents the opposite argument and does not rule out the influence of globalization. According to the modernization theory convergence does not only occur as countries independently develop themselves towards the equilibrium. Convergence occurs through different mechanisms in which globalization plays a key role. Different theories on these mechanisms are developed for a better understanding of the influential factors for institutional development. Herkenrath et al. (2005) present valuable theory on institutional convergence of countries. Herkenrath et al. first distinguish between different types of convergence between countries. One type is the two-sided convergence. This type is most likely to occur among societies that not only occupy a similar structural position in the world-system, they also conceive themselves as culturally similar. A good example of this type is the European Union. In the case of the US, Canada and Great Britain are good examples. However, two-sided convergence can also occur among countries that are not perceived as similar.
requires operations on a worldwide level. Both Gertler (2001) and Herkenrath et al. (2005) support this notion of social learning and state that with social learning countries copy institutions and practices from another country it perceives as more efficient and effective. In the case of the US, American economists and public officials argue that the superior performance of the American economy in the 1990s and the weaknesses of the once-envied Japanese and other Pacific Asian economies have made the American economy and the free market the model for the world (Gilpin, 2011). Thus, this implies that countries can learn from the US and converge towards their institutions. However, it must be noted that it does not necessarily mean that all countries will regard the US institutions as more efficient and effective. Additionally, it cannot be ruled out that the US cannot learn from other countries or other countries cannot learn from each other.
culturalism, and varieties of capitalism, globalization plays a different role in the modernization theory. This leads to the following hypothesis:
H3: Economic relations with the United States have a positive influence on the degree of
institutional convergence between countries and the United States
Not merely economic globalization has a positive influence on the degree of institutional convergence of countries and the US. Globalization in general tends to reinforce the effects of convergence (Drezner, 2001). In addition to economic relations, another factor that contributes to this is the improvement of communication technologies. Advances in telecommunications and computers have made the exchange of ideas across borders considerably easier. Gilpin (1987) already claimed this by stating that the transistor radio and the television set have made people in even the most remote parts of the globe aware of the wealth of others and of the benefits of material progress. Roland (2004) supports this by stating that the institutional changes that took place in Western Europe in the 18th and 19th centuries would be difficult to imagine without the intellectual turmoil created by the Renaissance and the ideas of the Enlightenment, which were spread by communication technology such as the printing press. Thus, communication technology has given countries the knowledge about the institutions in other countries and may lead to the desire of similar institutions. More specific in this study, the hypothesis that follows from this is:
H4: An increase in telecommunications has a positive influence on the degree of institutional
convergence between countries and the United States
2.4.5 Westernization
countries towards their own image. It is important to note that modernization and westernization are similar but no identical theories. The difference between these theories is the final outcome of the institutional development. According to the modernization theory the institutional development of countries will reach an optimum outcome. The westernization theory claims that the institutional development of countries will result in a complete take-over of the Western institutions. However, Western institutions are not necessary the optimum point of development despite the fact that several believe this. Liberals for example, assert that there is a time lag but the gap between rich and poor will eventually disappear as Western economic methods and technology diffuse throughout the world (Gilpin, 1987). Modernization theory states that also Western countries ought to develop their institutions towards the optimum. Thus, H2 which
predicts institutional convergence, also holds for the westernization theory. However, the direction of convergence differs from the modernization theory. Where modernization appears to be led by two-sided convergence, westernization has a one-sided characteristic. Countries ought to converge towards the Western states as they are the ‘best’ and not the other way around.
Herkenrath et al. (2005) state that one-sided convergence is most likely to occur in situations characterized by power imbalances or differences in prestige that is, when one society is less powerful than another in military-economic realm or perceives itself as ‘backward’ in the realm of cultural practices. In accordance with the westernization theory, the Western states are perceived as the superior states. Thus, the one-sided convergence is highly applicable for the US, the biggest and most powerful Western state (section 2.3). Social learning plays a role in one-sided convergence, however this is not deemed as the most important in this theory. Another important mechanism that contributes to the one-sided convergence is coercion. Coercion is defined by the Oxford English dictionary as “the application of force to control the action of a voluntary agent”. Military force is one factor through which coercion can take place. Countries that are dependent on the US on a military level have military alliances with and receive military aid from the US. This dependency and inferiority may lead to institutional convergence of countries towards the US. From this it can be hypothesized that:
H5: Having military alliances and receiving military aid from the United States has a positive
Economic globalization can also contribute to one-sided convergence trough coercion. Drezner (2001) argues that the threat of mobile capital to exit generates pressure to modify regulatory policies. Loosing this mobile capital causes non-converging countries to lose their competiveness in the global economy. Thus, countries that have economic relations with the US feel the pressure of losing the benefits of these relations and therefore adopt their policies towards the policies of the partner country. Additionally, Drezner presents a second possible reason for convergence. The pressure that countries feel could be ideational. States alter institutions and regulations because a set of beliefs has developed sufficient normative power that leaders fear looking like laggards if they do not adopt similar policies. This latter reason is not necessarily reinforced by economic relations, rather by a general image of other countries. However, it does explain why countries converge their institutions. These arguments fit well with H3 which
concludes that economic relations with the US have a positive influence on institutional convergence towards US institutions.
the organization have a higher chance of being influenced by the US and its institutions through these organization. This leads to the following hypothesis:
H6: The amount of credit from the IMF has a positive influence on the degree of institutional
convergence towards the United States.
2.6 Theoretical model
The theoretical models of this study are specified in equation one, two, and three. The dependent variable is the degree of institutional convergence. Formula one presents the two-sided convergence and formulas two and three present one-sided convergence. This one-sided convergence can either occur by the convergence of a country towards the US or by convergence of the US towards a country. Formula two and three distinguish these different types of one-sided convergence.
(1) IC = c + β1ECONREL + β2COM + β3MILREL + β5IMF +ε
(2) ICCNTRY = c + β1ECONREL + β2COM + β3MILREL + β5IMF +ε
(3) ICUS = c + β1ECONREL + β2COM + β3MILREL + β5IMF +ε
IC = Degree of institutional convergence ε = error term
ECONREL = Economic relations (include trade, FDI, and aid) c = constant
COM = Level of communication technology in a country MILREL = Military relations with the US
IMF = Level of IMF loans of a country CNTRY = country
US = United States
3. Methods and Data
3.1 Data Sample
to control for effects caused by differences among regions and developed and developing economies. Information on the list of countries selected in the study is available in the Appendix, Table 7 and 8. The remainder of this section describes the measurements of both the dependent and independent variables. Table 9 presents an overview on the sources of the data collected for this study.
3.1.1 Institutional Quality
Institutional quality (IQ) is measured by the Worldwide Governance Indicators (WGI). The WGI constructs aggregate indicators of six broad dimensions of governance. Thus, the concept of institutional quality is defined through six different indicators namely, voice accountability, political stability and absence of violence/terrorism, government effectiveness, regulatory quality, rule of law, and control of corruption. The indicators range from approximately -2.5 (weak) to 2.5 (strong). The years 1996 and 2012 are considered as they are the first and last available years, respectively. As the change of institutions tends to occur over a long period of time it is important to have a long time span. Table 10 shows that the six indicators have high correlations (correlation coefficients are > 0.7). In addition, theoretically the six indicators attempt to indirectly measure one concept. Considering these two points, a factor analysis is conducted. The details of the factor analysis are discussed in section 3.2. The factor analysis generates a factor to describe several variables. In conclusion, institutional quality is measured by the factor of the six indicators for both 1996 and 2012 for the entire sample. This reduces the amount of dependent variables, which facilitates the regression analysis.
3.1.2 Degree of Institutional Convergence
After measuring the factor for IQ of 1996 and 2012, the degree of institutional convergence with the US can be generated. It must be noted that the sample now consist of 181 countries per year, as the US is omitted. Evidently the degree of institutional convergence for the same country cannot be calculated. A distinction is made between the degree of two-sided and one-sided convergence. Institutional convergence is measured as a difference- in- difference indicator. This can be formulated as follows:
(4) IC = |IQUS – IQCNTRY|1996 - |IQUS – IQCNTRY|2012
(6) ICUS = |IQUS – IQCNTRY|1996 - |IQUS,2012 – IQCNRTY,1996|
Formula four describes the two-sided convergence. For two-sided convergence it implies the absolute difference of the IQ factor of the US and a country from the sample in 1996 minus the absolute difference of the IQ of the US and a country from the sample in 2012. The difference of the absolute difference between these two years indicates the degree of two-sided institutional convergence. Formula five presents the absolute difference of the IQ factor of the US and a country from the sample in 1996 minus the absolute difference of the IQ factor of a country in 2012 and the US in 1996. This illustrates whether the IQ of a country in 2012 has changed relatively to the IQ of the US in 1996 compared to the absolute difference of both IQs in 1996. Formula six presents the same formula for the convergence of the US. It must be noted that if the degree of institutional convergence has a negative sign, countries become institutionally more distant to the US.
3.1.3 Economic Relations
Economic relations are empirically defined as trade flows, FDI flows and aid. Trade is determined by the sum of exports and imports. Imports are cost, insurance, and freight (c.i.f.). Exports from the US to a country and imports towards the US are both presented in US dollars.
FDI flows are measured by the US direct investment position abroad on a historical-cost basis per country. The investment position of the US in a country is the sum of the direct investment by American enterprises outside the American boundaries minus the direct investment of foreign country enterprises within the US (Dunning, 1981). The investment position is measured in US dollars and has the advantage of taking into account both directions of FDI flows between the US and other countries.
For all three independent variables the mean of the years 1990 to 1995 is selected as institutional change does not occur instantly. Policies need time in order to have a possible measurable effect. If an affect is present it is more likely that institutional change is affected after several years. Thus, it is deemed more likely that solely flows from the beginning of the analyzed period will have an effect rather than flows from 1996 to 2012. This solves the potential problem of endogeneity as it cannot be stated that institutional quality of 2012 can influence variables means of 1990 to 1995. Additionally, using the mean helps to partially tackle the problem of missing values.
3.1.4 Communication
In order to test the fourth hypothesis an indicator of communication is required. The amount internet users per 100 people in a country is used to measure the communication level of a country. Internet users are defined by the World Bank as people with access to the worldwide network. Internet connections are deemed appropriate as the internet gives direct access to information of all countries. Television and radio are also proper communication technologies, however, the amount of information available is less than on the internet. As internet users are presented cumulative, the data for internet users are for 1996.
3.1.5 Military Relations
Hypothesis five requires proper measurements for military aid and military alliances. The latter presents a dummy variable for having a defense pact with the US. A defense pact implies that the US helps to defend the states included in the alliance. This is considered the indicator with the highest bargaining power as it requires an active involvement of the US. Having a non-aggression or neutrality pact are reckoned to be highly passive as they solely require countries to remain neutral or a promise of non-aggression towards the states in the alliance (Gibler, 2009). The alliances included are since 1990 and still active.
3.1.6 IMF
The possible influence of the IMF is measured through the supply of IMF credit of countries. Use of IMF credit denotes members' drawings on the IMF other than amounts drawn against the country's reserve tranche position. Use of IMF credit includes purchases and drawings under stand-by, extended, structural adjustment, enhanced structural adjustment, and systemic transformation facility arrangements as well as trust fund loans (World Bank). The use of IMF credit is presented in current US dollars. The mean for 1990 to 1995 is also used for IMF credit.
3.1.7 GDP per capita
The model includes GDP per capita to control for differences in GDP per country. GDP per capita is gross domestic product divided by midyear population. GDP is the sum of gross value added by all resident producers in the economy plus any product taxes and minus any subsidies not included in the value of the products. It is calculated without making deductions for depreciation of fabricated assets or for depletion and degradation of natural resources. Data are in current US dollars (World Bank). Furthermore, the data was supplemented with GDP per capita from the United Nations as missing values were present. Here the GDP per capita is also in current US dollars.
3.2 Factor Analysis
First, for this factor analysis it is important to note that the sample size is 364. Both the institutional quality indicators of 1996 and 2012 are added. This ensures that the analysis takes into account the factors for 1996 relative to 2012. In addition, the sample size is considered large enough for the analysis. Comrey and Lee (1992) give the following guide for samples sizes: 50 as very poor, 100 as poor, 200 as fair, 300 as good, 500 as very good, and 1,000 as excellent. Second, the Kaiser rule is used to determine the amount of factors. This rule states that only factors with an eigenvalue equal or larger than one should be retained (Field, 2009). Table 11 presents the eigenvalues of the factor analysis for the six indicators. Following the Kaiser rule it can be concluded that only factor one should be retained. The cumulative proportion adds the proportion of all factors. Thus, factor one explains 100.2% of the total variance. This is the entire variance, which confirms that solely retaining factor one is sufficient for this study.
Third, after determining the amount of factors that need to be retained, a proper factor extraction method needs to be selected. As the data does not have a multivariate normal distribution, the principal factor extraction method is chosen. The most common extraction method is the maximum likelihood method. However, a multivariate normal distribution is required when using maximum likelihood as the factor extraction method (Treiblmaier and Filzmoser, 2010). For the principal factor extraction a multivariate normal distribution is not a prerequisite, however, it must be noted that the results may be influenced by the occurrence of the non-normality. Table 12 presents the output of the principal factor extraction method of one factor. Column two shows that all six variables have a large loading on factor one, which implies that factor one is defined by all six variables. Furthermore, the sample shows low uniqueness levels for all six variables. Consequently, all variables are considered good items for the factor analysis. Finally, as only one factor is retained and the factor loadings are very clear in this analysis, rotation is deemed unnecessary. The factors can be predicted as the institutional quality of 1996 and 2012. These factors are used for the ordinary least square (OLS) regression analysis which is discussed in the next section.
3.3 Regression Analyses
below, where MILAID stands for military aid received by a country and MILAL is the dummy for the military alliance with the US.
(7) IC = c + β1TRADE + β2FDI + β3ODA + β4INTERNET + β5MILAID + β5MILAL +
β6IMF + ε
(8) ICCNTRY = c + β1TRADE + β2FDI + β3ODA + β4INTERNET + β5MILAID + β5MILAL +
β6IMF + ε
(9) ICUS = c + β1TRADE + β2FDI + β3ODA + β4INTERNET + β5MILAID + β5MILAL +
β6IMF + ε
First, for this study regression analyses are conducted to test whether the degree of institutional convergence has a significant direction. It shows whether countries and the US become institutionally more distant or closer towards each other. Second, the regression models determine the impact of economic relations, communication technology, military alliances and aid, and IMF credit on the degree of institutional convergence. Lastly, robustness checks are generated. Prior to the regression analyses, tests are conducted to inspect the general assumptions of an OLS regression. Table 13 presents the summary statistics of the variables. Leverage and Cook’s D tests indicate several possible outliers. The findings are presented in Figure 1, Table 14 and Table 15. The regression results take into account the possible outliers. Furthermore, a skewness and kurtosis test shows that the data set does not conform to the normality assumption. However, transforming the independent variables into their logarithm is not a plausible solution, as the results remain the same with logTrade and over 100 observations are missing when using logFDI and logODA. Therefore, the analysis proceeds with the non-normality and it must be taken into account when discussing the results. The data set does conform the homoscedasticity and multicolinearity assumptions as can be seen from the summary statistics and Table 16.
4. Regression Results
4.1 Two-sided Convergence Results
The first part of the regression analysis tests whether the degree of institutional convergence has a significant direction. For the degree of institutional convergence to increase, a negative relationship between the initial IQ difference in 1996 and the degree of institutional convergence of 1996-2012 must exist. The following regression model is used:
(10) IC = c + (IQCNTRY – IQUS)1996 + ε
Table 1 regression 1-1 presents the results for this regression. The coefficient is positive and significant at a 1% level, which means that it presents strong evidence to conclude that countries become institutionally distant from the US. Regressions 1-2 to 1-11 are the results for regression model (7) and test which factors influence the degree of institutional convergence. Regression 1-1 presents all three variables as non-significant. There is insufficient evidence to conclude influence or non-influence of the three independent variables on the degree of institutional convergence. Regression 1-2 shows the coefficients for a sample that exclude the outliers mentioned in section 3.3. Evidently, the outliers appear to be influential observations since omitting them results in the significance of ODA at a 5% level. From this it can be concluded that there is enough evidence to state that ODA has a negative influence on the degree of institutional convergence. This implies that, all other things being equal, when ODA increases with one million dollars, institutions become more distant by 3.31e-09 points.
ODA and a military alliance does not explain whether countries become distant from the US or the US becomes more distant from the countries. The next section attempts to solve this question.
4.2 One-sided Convergence Results
As for the two-sided convergence regression, the first part of the regression analysis tests whether the degree of both one-sided institutional convergence types have a significant direction. The formulas for these regressions are as follows:
(11) ICCNTRY = c + (IQCNTRY – IQUS)1996 + ε
(12) ICUS = c + (IQCNTRY – IQUS)1996 + ε
Table 2 and regression 2-1 show the results for regression (11). The coefficient is significant and negative as in regression 1-1 which implies that countries become institutionally more distant from the US with 0.091 points. Remarkable is that the influence of ODA found in the two-sided convergence regression cannot be found in the one-sided convergence. Furthermore, as for two-sided convergence, internet has a positive relationship on the degree of one-two-sided convergence. However, this relationship becomes insignificant when adding the economic relations variables. Regressions 2-5 to 2-9 provide insufficient evidence to conclude anything about their influence on the dependent variable. Regression 2-10 shows that by adding the military alliance dummy, both the dummy and internet become significant. From this it can be concluded that, all other things being equal, an increase of one internet user results in an increase of the one-sided convergence of countries by 0.038 points. Having a military alliance causes countries to become institutionally more distant from the US by 0.253 points.
Table 1 Cross-Sectional OLS Regressions of Two-sided Institutional Convergence on Economic and Military Relations
Numbers in parentheses are absolute values of t-statistic.
Asterisks indicate statistical significance at the ***1, **5, and *10 percent level.
a: Sample excludes four outliers: Germany, Mexico, UK, and Egypt. This holds for regression 1-3 through 1-11
Independent variable
Regression (dependent variable is Institutional Convergence)
1-1 1-2 1-3a 1-4 1-5 1-6 1-7 1-8 1-9 1-10 1-11 (CNTRY96 – US96) -0.196 (3.49)*** Trade -2.02e -12 (0.53) -1.47e-12 (0.30) -6.25e-13 (0.13) -6.58e-13 (0.11) -7.23e-13 (0.12) -1.76e-12 (0.30) FDI 6.37e-12 (0.78) 7.13e-12 (0.52) 2.47e-12 (0.17) 1.72e-12 (0.09) 1.81e-12 (0.10) 9.14e-12 (0.49) ODA -6.39e-10 (1.38) -3.31e-09 (2.52)** -3.09e-09 (2.31)** -3.18e-09 (2.33)** -3.33e-09 (2.42)** -2.73e-09 (1.94)* Internet 0.038 (1.90)* 0.020 (0.87) 0.021 (0.87) 0.024 (0.95) 0.030 (1.20)
Military Aid -5.72e-11
(0.21) -6.03e-11 (0.22) -5.44e-11 (0.20) -7.18e-11 (0.27)
IMF credit 3.75e-11
Table 2 Cross-Sectional OLS Regressions of One-sided Convergence of a Country on Economic and Military Relations
Numbers in parentheses are absolute values of t-statistic.
Asterisks indicate statistical significance at the ***1, **5, and *10 percent level. All regression exclude the outliers Germany, Mexico, UK, and Egypt.
Independent variable Regression (dependent variable is one-sided convergence of countries towards the US)
2-1 2-2 2-3 2-4 2-5 2-6 2-7 2-8 2-9 2-10 (CNTRY1996 – US1996) -0.091 (1.77)* Trade -3.12e-12 (0.71) -1.96e-12 (0.44) -3.58e-12 (0.66) -3.63e-12 (0.67) -4.60e-12 (0.85) FDI 1.91e-11 (1.53) 1.26e-11 (0.95) 1.81e-11 (1.08) 1.81e-11 (1.08) 2.50e-11 (1.47) ODA -4.92e-10 (0.41) -1.91e-10 (0.16) -2.33e-10 (0.19) -3.57e-10 (0.28) 2.04e-10 (0.16) Internet 0.033 (1.85)* 0.027 (1.34) 0.030 (1.34) 0.032 (1.42) 0.038 (1.67)*
Military Aid -6.28e-11
(0.25) -3.84e-11 (0.16) -3.33e-11 (0.13) -4.96e-11 (0.20)
IMF credit 3.24e-11
Table 3 Cross-Sectional OLS Regressions of One-sided Convergence of the US on Economic and Military Relations
Numbers in parentheses are absolute values of t-statistic.
Asterisks indicate statistical significance at the ***1, **5, and *10 percent level. All regression exclude the outliers Germany, Mexico, UK, and Egypt.
Independent variable Regression (dependent variable is one-sided convergence from the United States towards countries)
3-1 3-2 3-3 3-4 3-5 3-6 3-7 3-8 3-9 3-10 (CNTRY1996 – US1996) -0.474 (28.55)*** Trade -4.34e-12 (1.27) -3.13e-12 (0.91) -8.24e-13 (0.20) -9.03e-13 (0.22) -1.63e-12 (0.40) FDI 6.26e-12 (0.65) -4.03e-13 (0.04) -8.52e-12 (0.68) -8.41e-12 (0.67) -3.30e-12 (0.26) ODA -8.05e-10 (0.88) -4.93e-10 (0.53) -5.76e-10 (0.61) -7.49e-10 (0.80) -3.31e-10 (0.34) Internet 0.026 (1.89)* 0.028 (1.80)* 0.024 (1.44) 0.027 (1.60) 0.031 (1.85)*
Military Aid 5.99e-11
(0.32) 4.67e-11 (0.25) 5.37e-11 (0.29) 4.16e-11 (0.23)
IMF credit 8.46e-11
The one-sided convergence of the US shows a stronger degree of institutional distant of 0.474 points compared to 0.091 points of one-sided distance of countries. Similarly, the presence of a military alliance has a stronger negative effect on countries than on the US. The amount of internet users causes a stronger convergent effect on countries towards the US than on the convergence of the US towards the countries.
4.3 Robustness Checks Two-sided Convergence
Table 4 Robustness Checks Two-sided Convergence
Numbers in parentheses are absolute values of t-statistic.
Asterisks indicate statistical significance at the ***1, **5, and *10 percent level All regression exclude the outliers Germany, Mexico, UK, and Egypt.
Independent variable Regression (dependent variable is Institutional Convergence) 4-1 4-2 4-3 Trade -1.98e-12 (0.33) -4.93e-12 (0.71) -2.83e-12 (0.47) FDI 8.07e-12 (0.43) 1.98e-11 (0.90) 1.18e-11 (0.62) ODA -2.55e-09 (1.74)* -2.70e-09 (1.81)* -2.73e-09 (1.86)* Internet 0.019 (0.54) 0.037 (0.98) 0.027 (0.76)
Military Aid -9.19e-11
(0.34)
-2.15e-10 (0.77)
-9.98e-11 (0.37)
IMF credit 9.87e-11
(0.94) 1.02e-10 (0.79) 1.07e-10 (1.02) Military Alliance -0.287 (1.83)* -0.255 (1.35) -0.307 (1.96)*
GDP per capita 7.09e-06
(0.43) 2.04e-07 (0.01) 0.000 (0.71) Africa -0.519 (2.06)** Europe -0.519 (2.09)**
Latin America and Caribbean -0.521
Table 5 Robustness Checks One-sided Convergence of Countries Towards the US
Numbers in parentheses are absolute values of t-statistic.
Asterisks indicate statistical significance at the ***1, **5, and *10 percent level All regression exclude the outliers Germany, Mexico, UK, and Egypt.
Independent variable Regression (dependent variable is Institutional Convergence) 5-1 5-2 5-3 Trade -4.63e-12 (0.85) -6.16e-12 (0.96) -4.81e-12 (0.87) FDI 2.49e-11 (1.44) 3.05e-11 (1.49) 2.57e-11 (1.47) ODA 2.23e-10 (0.17) 2.04e-10 (0.15) 1.83e-10 (0.14) Internet 0.037 (1.14) 0.044 (1.28) 0.038 (1.17)
Military Aid -5.18e-11
(0.21)
-1.28e-10 (0.49)
-5.35e-11 (0.21)
IMF credit 8.27e-11
(0.86)
8.25e-11 (0.84)
8.45e-11 (0.88)
Military Alliance 8.27e-11
(1.78)*
-0.207 (1.18)
-0.259 (1.80)*
GDP per capita 7.74e-07
(0.05) -5.23e-06 (0.30) 1.83e-06 (0.12) Africa -0.336 (1.43) Europe -0.267 (1.08)
Latin America and Caribbean -0.363
Table 6 Robustness Checks One-sided Convergence from the US Towards Countries
Numbers in parentheses are absolute values of t-statistic.
Asterisks indicate statistical significance at the ***1, **5, and *10 percent level All regression exclude the outliers Germany, Mexico, UK, and Egypt.
Independent variable Regression (dependent variable is Institutional Convergence) 6-1 6-2 6-3 Trade -1.46e-12 (0.36) -2.10e-12 (0.45) -2.01e-12 (0.49) FDI -1.46e-12 (0.19) 1.28e-12 (0.09) -6.31e-14 (0.00) ODA -4.71e-10 (0.47) -5.08e-10 (0.50) -5.89e-10 (0.59) Internet 0.040 (1.66)* 0.051 (1.99)** 0.045 (1.85)*
Military Aid 5.74e-11
(0.31)
5.13e-12 (0.03)
5.24e-11 (0.28)
IMF credit 1.14e-10
(1.60) 1.20e-10 (1.67)* 1.20e-10 (1.67)* Military Alliance -0.175 (1.64) -0.188 (1.47) -0.189 (1.76)*
GDP per capita -5.60e-06
(0.50) -9.13e-06 (0.71) -2.46e-06 (0.21) Africa -0.323 (1.89)* Europe -0.353* (1.95)
Latin America and Caribbean -0.193
4.4 Robustness Checks One-sided Convergence
Table 5 presents the robustness checks results for the first type of one-sided convergence, where countries would institution wise move towards or away from the US. Internet and military alliance showed a significant influence on one-sided convergence. However, controlling for GDP per capita generates a non-significant internet coefficient. Thus, internet appears to be less robust and the effect of internet is affected by the difference in GDP per capita. Military alliance is deemed robust as it remains significant when controlled for GDP per capita. Military alliance appears to be less robust when adding region dummies, however, this is not the case when adding the developed countries dummy. This result was also found in Table 4. Furthermore, both region and developed countries dummies are insignificant, thus no conclusions can be drawn regarding their influences.
Table 6 presents the robustness checks results for the second type of one-sided convergence, where the US would move towards or away from countries. Only internet remains robust when controlling for GDP per capita. Military alliance is less robust and becomes non-significant. A striking change that has occurred is the significant coefficient for IMF credit when controlling for regions and developed countries. The amount of IMF credit has a positive influence on the degree of one-sided convergence. Furthermore, it can be concluded that, for all other things being equal, countries from Africa and Europe have a lower degree of institutional convergence than countries from the Middle East. The same conclusion was made for the two-sided convergence. Finally, no conclusion can be drawn from the distinction between developed and developing countries as the dummy coefficient is not significant.
5. Discussion and Limitations
The aim of this study was to assess whether economic and military relations influence the degree of institutional convergence between countries and the US. In addition, telecommunications, and IMF credit are considered. The first two hypotheses state whether convergence or divergence occurs. The results present strong evidence to conclude that countries and the US become institutionally distant from each other. This effect is significant for two-sided convergence and for both types of one-sided convergence. Thus, H1 is accepted. This is what was predicted by path
that globalization has no effect on institutional change. The results present contradictory findings. The amount of ODA has a significant negative effect on the degree of two-sided institutional convergence. H3 predicts that economic globalization causes institutional convergence. H3 is
rejected as a positive relationship was predicted, however, the divergence predicting theories are incorrect by stating that globalization has no effect at all on institutional change. Furthermore, it cannot be concluded whether ODA causes countries to diverge from the US or the other way around. Both one-sided divergence results are insignificant. H3 is also rejected for FDI and trade
as both coefficients are non-significant.
Whereas the first three theories did not predict a relationship at all, modernization and westernization were also incorrect on predicting the direction of the relationship between ODA and the degree of institutional convergence. A reason for generating these unexpected results on ODA may be due to a possible selection bias. The US does not donate ODA to all countries. Approximately 42% of the countries receive no ODA at all. Thus, the difference in effect on institutional convergence may not necessarily lie in the volume of ODA but in the difference between countries that do receive ODA and countries that do not. Adapting ODA into a dummy variable may provide better insight on the matter. Countries that do not receive ODA are the reference group against countries that do receive ODA. Regressing the degree of institutional convergence on the ODA dummy presents insufficient evidence to state that receiving countries have a significant lower degree of institutional convergence than non-receiving countries. However, this does not necessarily solve the possible selection bias. When only considering countries that receive ODA in a regression analysis, the coefficient of ODA remains significant except when military alliance is added. Nonetheless, it can be concluded that the significant negative effect of ODA cannot be assigned to a selection bias.
in their best interest to retain their power. However, Boone (1996) finds no empirical evidence that alternative political regimes use aid differently. Thus, results regarding this relationship are mixed. Another possible reason for the negative influence of ODA is mentioned by Tungodden et al (2004). They claim that the influence of aid on policy changes is conditioned by macroeconomic support. Thus, it could be that the negative influence of ODA is due to bad macroeconomic support.
All types of convergence indicate that the amount of internet connections has a positive influence on the degree of institutional convergence. However, when controlling for GDP per capita, only the convergence of the US remains robust. This implies that H4 is accepted and that
the US converges towards countries when the amount of internet users increases. Another significant coefficient is the military alliance dummy. This coefficient’s effect also contradicts the expectations. H5 predicted a positive relationship between having a military alliance and the
degree of institutional convergence. In addition, it is important to note that the robustness checks show that military alliance becomes insignificant when controlling for the different regions. Thus, the influence of military alliance stems from the difference in regions. It can be concluded that the power position of the US is not effective to coerce countries into altering their institutions to those of the US and H5 is rejected.
For the two-sided convergence it can be concluded that regions differ significantly in the degree of institutional convergence. Countries from the Middle East have a significantly higher degree of institutional convergence than countries from Africa, Europe, Latin America and Caribbean, and Oceania. The two-sided convergence only states that the institutional quality of countries from the Middle East and the US converge. It does not show whether this is due to the convergence of the Middle East or convergence of the US. One-sided convergence specifies this and shows that the US significantly converges towards the Middle East compared to Africa and Europe.
As for internet, IMF also has a positive effect on the degree of institutional convergence. When controlled for regions and development, IMF has a significant positive relationship on the degree of institutional convergence of the US towards countries. This positive relationship is not in line with the expectations as it was predicted that countries would converge towards the US when receiving more IMF credit. The opposite is the case which implies that H6 is rejected.
implies that westernization does not hold for this study. This finding is quite puzzling as the literature claims that the US uses IMF to pursue American financial and foreign policy objectives (Oatley and Yackee, 2004). As for ODA, there must be controlled for a selection bias as not all countries receive IMF credit. A regression that includes solely countries that receive IMF credit, generates a non-significant coefficient. This implies that a selection bias is present. The significance of IMF found in Table 6 is not due to the level of IMF. The influence stems from the difference between countries that receive IMF credit and countries that do not receive IMF credit. Where it was expected that trade, FDI, military aid and alliances would have a significant influence on the degree of institutional convergence, this is not the case. Several limitations may have caused this insignificance. In this research the model specifies solely direct linear effects of the variables. The relationships between the variables may also be exponential or u-shaped. Additionally, it could be the case that indirect effects exist. For example, trade might affect a third variable, which is not included in this analysis, that in turn might influence the degree of institutional convergence. Another possible limitations of this research is the time frame of the data set. Institutions tend to change over a long period of time. Significant changes over 17 years may not occur which leads to insignificance. The World Governance indicators for institutional quality do not present a larger time frame. A different data set might provide a larger time frame, however, it must be noted that most institutional data solely includes political institutions. Thus, it only captures the political aspects of institutions, which might require a different theoretical framework. In addition, aside from a factor analysis, there exists multiple methodology to deal with the six institutional quality indicators. Handling the six indicators in a different manner may generate different results.
6. Conclusions
Appendix
Table 7 Country Sample
Africa Mauritius Kyrgyz republic Hungary
Algeria Morocco Lebanon Iceland
Angola Mozambique Oman Ireland
Benin Namibia Pakistan Italy
Botswana Niger Qatar Latvia
Burkina Faso Nigeria Saudi Arabia Liechtenstein
Burundi Rwanda Syrian Arab rep Lithuania
Cameroon São Tomé and Principe Tajikistan Luxembourg
Central African republic Senegal Turkmenistan Macedonia, fyr
Chad Seychelles United Arab emirates Moldova
Comoros Sierra Leone Uzbekistan Netherlands
Congo, dem. Rep. Somalia West bank and Gaza Norway
Congo, rep. South Africa Yemen, rep. Poland
Côte D'ivoire Sudan Portugal
Djibouti Swaziland Europe Romania
Egypt, Arab rep. Tanzania Andorra Russian federation
Equatorial guinea Togo Austria Serbia
Eritrea Tunisia Azerbaijan Slovak republic
Ethiopia Uganda Belarus Slovenia
Gabon Zambia Belgium Spain
Gambia, the Zimbabwe Bosnia and Herzegovina Sweden
Ghana Bulgaria Switzerland
Guinea Middle East Croatia Turkey
Guinea-Bissau Afghanistan Cyprus Ukraine
Kenya Albania Czech republic United kingdom
Lesotho Armenia Denmark
Liberia Iran, islamic rep. Estonia
Libya Iraq Finland
Madagascar Israel France
Malawi Jordan Georgia
Mali Kazakhstan Germany
Latin America and Caribbean Oceania North America
Antigua and barbuda Fiji Bermuda
Argentina New zealand Canada
Bahamas, the Papua new guinea United states
Barbados Samoa
Belize
Bolivia Asia
Brazil Bahrain
Cayman islands Bangladesh
Chile Bhutan
Colombia Brunei darussalam
Costa rica Cambodia
Cuba China
Dominica Hong kong sar, china
Dominican republic India
Ecuador Indonesia
El salvador Japan
French guiana Korea, dem. Rep.
Grenada Korea, rep.
Guatemala Lao pdr
Guyana Macao sar, china
Haiti Malaysia Honduras Maldives Jamaica Mongolia Martinique Myanmar Mexico Nepal Nicaragua Philippines Panama Singapore
Paraguay Sri lanka
Peru Taiwan, china
Puerto rico Thailand
Suriname Vietnam
Trinidad and tobago Uruguay
Table 8 Development Categories
Developing regions Developed regions
Africa Northern America
Americas excluding Northern America Europe
Caribbean Japan
Central America Australia and New Zealand
South America Asia excluding Japan
Oceania excluding Australia and New Zealand
Table 9 Variable Sources
Table 10 Correlations (n=364) VA PS GE RQ RL CC VA 1.000 PS 0.729 1.000 GE 0.854 0.767 1.000 RQ 0.854 0.707 0.923 1.000 RL 0.855 0.805 0.947 0.901 1.000 CC 0.816 0.768 0.927 0.857 0.930 1.000 VA = Voice accountability
PS = Political stability and absence of violence/terrorism GE = Government effectiveness
RQ = Regulatory quality CC = Control of corruption
Variable Source Year
Institutional quality Worldwide Governance Indicators, World Bank 1996, 2012 Exports, Imports IMF Direction of Trade and Census Bureau U.S. Mean 1990-1995
FDI flows Bureau of Economic Analysis Mean 1990-1995
AID US U.S. Official Development Assistance Database Mean 1990-1995
Internet users World Bank 1996
Military alliance Correlates of war Since 1990 – current
Military aid US Agency for International Development Mean 1990-1995
IMF credit World Bank Mean 1990-1995
Table 11 Factor Eigenvalues and Proportion (n=364) Factor Eigenvalue Cumulative Proportion
Factor 1 5.078 1.002 Factor 2 0.073 1.017 Factor 3 0.028 1.022 Factor 4 -0.024 1.018 Factor 5 -0.030 1.012 Factor 6 -0.059 1.000
Table 12 Loadings and Uniqueness (n=364)
Variable Factor 1 Uniqueness
VA 0.886 0.202 PS 0.805 0.321 GE 0.972 0.047 RQ 0.930 0.105 RL 0.974 0.048 CC 0.942 0.010
Table 13 Summary Statistics
Mean Standard Deviation 1 2 3 4 5 6 1. Institutional Convergence -0.094 0.796 2. Trade (million) 5810 22000 0.024 3. FDI (million) 3500 11100 0.065 0.723 4. ODA (million) 32.84 118.91 -0.032 -0.054 -0.062 5. IMF credit (million) 234 7.82 -0.115 0.157 0.018 0.021 6. Military Aid (million) 40.6 2.68 0.007 -0.009 -0.028 0.515 -0.026 7. Internet 1.18 2.92 0.107 0.227 0.276 -0.112 -0.118 -0.0055
Figure 1 Scatterplot y and x’s
The table shows that the data presents some potential problems as several data points are far away from the rest of the data points. The data set may contain problems through outliers. Outliers cause great problems as they may have potential great influence on the regression coefficient estimates.
Table 14 Leverage Scores
Country Leverage China 0.037 Netherlands 0.042 Germany 0.062 Switzerland 0.065 Bermuda 0.071 Mexico 0.096 Japan 0.349 Canada 0.518 United Kingdom 0.627
Egypt, Arab Republic 0.863
The leverage of an observation is based on how much the observation's value on the predictor variable differs from the mean of the predictor variable. The greater an observation's leverage, the more potential it has to be an influential observation. In general, a point with a leverage greater than (2k + 2)/n. Here k is the number of predictors and n is the number of observations (n=149 due to missing values). Thus, observations that have a leverage higher than (2 x 3 +2)/149 = 0.05 require the most attention as they could be influential. The table presents the ten highest leverage scores. The countries with a leverage greater than 0.05 are Germany, Switzerland, Bermuda, Mexico, Japan, Canada, United Kingdom, and Egypt.
IC TradeMean FDIMean ODAMean -2 0 2 -2 0 2 0 1.000e+11 2.000e+11 0 1.000e+112.000e+11 0 5.000e+10 1.000e+11 0 5.000e+101.000e+11 0 5.000e+08 1.000e+09 1.500e+09
Table 15 Cook’s D scores Country Cook’s D Philippines 0.038 Germany 0.036 Mexico 0.119 United Kingdom 0.720
Egypt, Arab Republic 7.222
An additional test for influential observations is through Cook’s D. Cook's D is a good measure of the influence of an observation and is proportional to the sum of the squared differences between predictions made with all observations in the analysis and predictions made leaving out the observation in question. The cut-off point for Cook’s D is 4/n = 4/149 = 0.027. The table presents the observations that have a higher estimate and thus could be observations of influence. Remarkably, Egypt and the UK again present the highest scores and also Mexico and Germany coincidence with the leverage scores table. Thus, these countries require extra attention.
Table 16 Multicolinearity Test Scores
Variable VIF 1/VIF
FDI 1.97 0.508
Trade 1.97 0.509
ODA 1.00 0.996
This table shows the results of a multicolinearity test that presents colinearity diagnostics of all independent variables. As all VIF scores are < 10 and 1/VIF < 1, it can be stated that there is no multicolinearity among the variables.
References
Acemoglu, Daron, James A. Robinson, and Thierry Verdier. Can't We All Be More Like
Scandinavians? Asymmetric growth and institutions in an interdependent world. No. w18441.
National Bureau of Economic Research, 2012.
Alesina, Alberro, and Roberto Perotti. "The political economy of growth: a critical survey of the recent literature." The World Bank Economic Review 8.3 (1994): 351-371.
Alesina, Alberto, and Beatrice Weder. Do corrupt governments receive less foreign aid?. No. w7108. National Bureau of Economic Research, 1999.
Alonso, José Antonio, and Carlos Garcimartín. "The determinants of institutional quality. More on the debate." Journal of International Development 25.2 (2013): 206-226.
Aoki, Masahiko. Toward a comparative institutional analysis. MIT press, 2001.
Barro, Robert J., and Jong-Wha Lee. "IMF programs: Who is chosen and what are the effects?." Journal of Monetary Economics 52.7 (2005): 1245-1269.
Boone, Peter. "Politics and the effectiveness of foreign aid." European economic review 40.2 (1996): 289-329.
Bush, George W. The national security strategy of the United States of America. EXECUTIVE OFFICE OF THE PRESIDENT WASHINGTON DC, 2002.
Comrey, Andrew Laurence, and Howard Bing Lee. A first course in factor analysis. Routledge, 1992.
Crane, David. “Reforms Needed to Restore IMF Credibility.” Toronto Star, 2000. DeCoster, Jamie. "Overview of factor analysis." Retrieved May 24 (1998): 2006.
Drezner, Daniel W. "Globalization and policy convergence." International studies review 3.1 (2001): 53-78.
Dunning, John H. "Explaining the international direct investment position of countries: towards a dynamic or developmental approach." Weltwirtschaftliches Archiv 117.1 (1981): 30-64.
Ebbinghaus, Bernhard. "Can path dependence explain institutional change?: Two approaches applied to welfare state reform." (2005).
Field, Andy. Discovering statistics using SPSS. Sage publications, 2009. Gardner, Conrad. "QFINANCE-The Ultimate Resource." (2009).
Gertler, Meric S. "Best practice? Geography, learning and the institutional limits to strong convergence." Journal of Economic Geography 1.1 (2001): 5-26.
Gibler, D. International Military Alliances, 1648-2008. Congressional Quarterly Press. 2009. ISBN: 978-1-56802-824-8
Gilpin, Robert. Global political economy: Understanding the international economic order. Princeton University Press, 2011.
Glaeser, Edward L., et al. "Do institutions cause growth?." Journal of economic Growth 9.3 (2004): 271-303.
Greif, Avner. Institutions and the path to the modern economy: Lessons from medieval trade. Cambridge University Press, 2006.
Hall, Peter V. "Regional institutional convergence? Reflections from the Baltimore Waterfront." Economic Geography 79.4 (2003): 347-363.
Hall, Peter A., and David W. Soskice, eds. Varieties of capitalism: The institutional foundations of comparative advantage. Vol. 8. Oxford: Oxford University Press, 2001.
Hall, Peter A., and Kathleen Thelen. "Institutional change in varieties of capitalism." Socio-Economic Review 7.1 (2009): 7-34.
Herkenrath, Mark, et al. "Convergence and Divergence in the Contemporary World System An Introduction." International Journal of Comparative Sociology 46.5-6 (2005): 363-382.
Hodgson, Geoffrey M. “What are institutions.” Journal of Economic Issues 40.1 (2006): 1-25. Huntington, Samuel P. "The clash of civilizations?." Foreign affairs (1993): 22-49.
International Monetary Fund (2013): Direction of Trade Statistics (Edition: June 2013). Mimas, University of Manchester. DOI: http://dx.doi.org/10.5257/imf/dots/2013-06
Kreutzmann, Hermann. "From modernization theory towards the ‘clash of civilizations’: directions and paradigm shifts in Samuel Huntington's analysis and prognosis of global development1." GeoJournal 46.4 (1998): 255-265.
Lai, Brian, and Daniel S. Morey. "Impact of regime type on the influence of US foreign aid." Foreign Policy Analysis 2.4 (2006): 385-404.
Levi, Margaret. "A model, a method, and a map: Rational choice in comparative and historical analysis." Comparative politics: Rationality, culture, and structure 28 (1997).
North, Douglass C. Institutions, institutional change and economic performance. Cambridge university press, 1990.
Oatley, Thomas, and Jason Yackee. "American interests and IMF lending." International Politics 41.3 (2004): 415-429.
OED Online. December 2013. Oxford University Press. 16 January 2014
<http://www.oed.com/view/Entry/35725?redirectedFrom=coercion>.
Pierson, Paul. "Increasing returns, path dependence, and the study of politics." American political science review (2000): 251-267.
Rietveld, Toni, and Roeland Van Hout. Statistical techniques for the study of language behaviour. Berlijn: Mouton de Gruyter, 1993.
Roland, Gerard. "Understanding institutional change: fast-moving and slow-moving institutions." Studies in Comparative International Development 38.4 (2004): 109-131.
Singer, J. David, and Melvin Small. 1966. "Formal Alliances, 1815-1939." Journal of Peace Research 3:1-31.
Small, Melvin, and J. David Singer. 1969. "Formal Alliances, 1815-1965: An Extension of the Basic Data." Journal of Peace Research 6:257-282.
Streeck, Wolfgang, and Kathleen Thelen. "Introduction: Institutional change in advanced political economies." Beyond continuity: Institutional change in advanced political economies (2005): 1-39.
Tungodden, Bertil, et al., eds. Toward Pro-poor Policies: Aid, Institutions, and Globalization. World Bank-free PDF, 2004.
Walton, John, and Charles Ragin. "Global and national sources of political protest: Third world responses to the debt crisis." American Sociological Review (1990): 876-890.
Williamson, John. "Democracy and the “Washington consensus”." World development 21.8 (1993): 1329-1336.
Wilson, Scott. "Remade in China: foreign investors and institutional change in China." OUP
Catalogue (2009).
Woods, Ngaire. "The United States and the international financial institutions: Power and influence within the World Bank and the IMF." Foot, McFarlane and Mastanduno (Eds.), US Hegemony and International Organizations, Oxford (2003): 92-114.
World Bank. 2013. World Development Indicators 2013. Washington, DC: World Bank. DOI: 10.1596/978-0-8213-9824-1.
Yifu Lin, Justin, and Jeffrey B. Nugent. "Institutions and economic development." Handbook of
