Content uploaded by Michael Lewis-Beck
Author content
All content in this area was uploaded by Michael Lewis-Beck on Apr 23, 2017
Content may be subject to copyright.
Comparative Democracy: The Economic Development
Thesis Revisited
Kenneth W. Moffett
Doctoral Student
Department of Political Science
The University of Iowa
341 Schaeffer Hall
Iowa City, Iowa 52242-1409
Voice: (319) 335-3381
Fax: (319) 335-3400
E-mail: ken-moffett@uiowa.edu
Prepared for submission to the American Political Science Review.
I would like to thank Joel Barkan, Fred Boehmke, Scott Cody, Doug Dion, Vicki Hesli, Ha-
Lyong Jung, Eric Manning, Dan Morey, Tom Rice, Helena Rodrigues, Robert Salmond, Megan
Shannon, Tracy Slagter, Matthew Whittaker, Jeremy Youde and especially Michael Lewis-Beck
for many helpful comments and suggestions on earlier versions of this paper. Any remaining
errors and omissions in this paper are my own responsibility.
Abstract:
Several studies have examined whether economic development causes democracy. This paper
extends Burkhart and Lewis-Beck’s (1994) study in light of theoretical challenges to their
findings and an increasing number of democracies since the late 1980s. I test whether economic
development, world position and whether a nation-state is an oil country affect whether a country
is more likely to be a democratic. I estimate my model on data from 180 nation-states with cross-
sectional time series data from 1972 to 1995. I discover that a country is more likely to be a
democracy as it economically develops. Further, semi-peripheral, peripheral and oil countries are
less likely to be democracies. Finally, my findings hold regardless of how I measure democracy
and after controlling for heteroskedasticity as well as spatial and temporal dependence.
1
Political scientists have long suspected that a relationship exists between economic
development and democracy. One reason offered for this connection is that the populace’s
demand for democracy grows as their country becomes economically better off because such
development can spread authority and democratic aspirations over a wide range of people
(Burkhart and Lewis-Beck 1994; Dahl 1989; Huntington 1991). Also, increased economic
development should foster democracy because nondemocracies that partly develop may change
their political systems because they are able to afford the costs that go along with having a
democratic system.1
Along these lines, many have argued that nation-states that economically develop are more
likely to be democracies (Burkhart and Lewis-Beck 1994; Feng 1997; Jackman 1973; Leblang
1997; Lipset 1959, 1994). Others argue that peripheral and semiperipheral countries are less
likely to be democracies than core countries (Bollen 1983; Burkhart and Lewis-Beck 1994;
Snyder and Kick 1979; Wallerstein 1974). Finally, some suggest that oil countries are less likely
to be democracies (Barro 1999; Smith 2004).
To determine which explanation most accurately explains the conditions under which a
country is more likely to be a democracy, I empirically retest and expand Burkhart and Lewis-
Beck’s (1994) study. A reanalysis is useful for three reasons: 1) more countries have become
democracies since the late 1980s (Freedom House 2003); 2) recent research has theoretically
challenged the basis for their findings; and 3) I would like to see whether economic development
affects a state’s level of democracy in light of the first elections in newly democratic countries,
new methodological techniques that better test this relationship, and using different measures of
1 These costs include, but are not limited to services that the government must provide to its citizens
because those citizens demand them. Examples of such services include infrastructure improvements,
elections and social services. See Boix (2001) and Baum and Lake (2003) for more details.
2
democracy. I will answer my research question in five parts. First, I review the basics of
Burkhart and Lewis-Beck’s (1994) study. Second, I state three reasons why we should reexamine
their study. Third, I discuss the conditions under which we should expect a relationship between
economic development and democracy and derive a series of hypotheses from these conditions.
Fourth, I will test my hypotheses on a dataset consisting of 180 nation-states from 1972 to 1995.
To conclude, I discuss my results.
Background
Burkhart and Lewis-Beck (1994) test whether economic development causes economic
development. Through a pooled, cross-sectional analysis of 131 nation-states from 1972 to 1989,
they discover that countries which are more economically developed are more likely to be
democracies. Also, they find that this relationship partly depends on that state’s position in the
world system. Finally, they learn from Grainger’s (1969, 1988) tests on their data that economic
development “causes” democracy, not the other way around.
Since Lewis-Beck and Burkhart’s (1994) published their article, their thesis has generated
considerable debate amongst scholars who study the relationship between economics and
democracy. For example, Rice and Ling (2002) argue that proponents of the economic
development thesis give little thought to how culture relates to the relationship between
economics and development. Using data from the 1990 World Values and the cumulative
General Social surveys, Rice and Ling (2002) find that whether a country was a democracy in
the 20th century depended more on social capital in the 1800s than 19th century economic
development.
However, two problems plague their study. First, they employ measures of social capital that
“stack the deck” in favor of a statistically significant relationship between social capital and
3
democracy. We see such an effect because they perform an ex post analysis by using
contemporary data as a historical measure of social capital.2 Second, Jackman and Miller (1996a,
1996b) employ measures of social capital that circumvent the ex post problem and discover little
evidence that yields a systematic relationship between political culture and political and
economic performance (Jackman and Miller 1996a, 653).
Other scholars argue that researchers should distinguish between exogenous and endogenous
theories with respect to the relationship between economic development and democracy.
Endogenous explanations say that countries which continue to economically develop are more
likely to be democracies (Przeworski and Limongi 1997; Przeworski, Alvarez, Cheibub and
Limongi 2000). Conversely, exogenous theories argue that countries establish democracies
independent of economic development. However, democracies are more likely to survive in
developed countries (Przeworski and Limongi 1997; Przeworski, Alvarez, Cheibub and Limongi
2000). These researchers argue that once we disentangle these explanations, support for
endogenous explanations falls away in favor of exogenous ones.
While their line of research seems convincing, it comes with two problems. First, they use
proprietary measures to operationalize democracy. While using these measures has some
benefits, the drawbacks of using such measures outweigh the benefits. The main drawbacks are
that the rater may introduce random and/or systematic measurement error (Bollen 1993), the
measure may not be reliable (see Przeworski and Teune 1970) and that such measures are
available for limited time periods. Second, both studies only look at nation-states until 1990.
2 Using such data assumes that a country had a lot of social capital in its history because it currently has a
lot of social capital. One cannot prove such an assumption without accurate social capital measures during
that country’s history.
4
Ending their analysis at 1990 poses significant problems because many countries have become
democracies since that time.
Because more recent research has cast doubt upon exogenous explanations for the
relationship between economic development and democracy — especially Burkhart and Lewis-
Beck’s (1994) — it is useful to extend their dataset and reexamine this relationship for two
additional reasons. Since the late 1980s, we have seen a number of partially developed
authoritarian states become democracies (Freedom House 2003). This rise has four implications.
First, we should observe less conflict in international politics because democracies are less likely
to fight one another (Cedarman 2001; Dixon 1994; Huntington 1991; Maoz and Russett 1993).
Second, economically developing democracies will have larger public sectors that provide
collective goods and transfer payments to their citizens than authoritarian countries (Boix 2001).3
Third, those who inhabit poor democracies are more likely to live longer (Baum and Lake 2003).
Finally, citizens of comparatively wealthier democracies are more educated on average (Baum
and Lake 2003).
Further, major studies that examine the relationship between economic development and
democracy measure democracy differently. Early studies employed proprietary measures to
operationalize democracy (Bollen 1979, 1983; Bollen and Jackman 1985; Jackman 1973; Lipset
1959). However, these measures came with significant drawbacks. Because of these limitations,
most recent research has employed Polity or Freedom House scores to measure democracy
(Baum and Lake 2003; Burkhart 1997; Burkhart and Lewis-Beck 1994; Feng 1997; Leblang
1997). Researchers who favor using Polity scores argue that Freedom House measures are
conservatively biased to favor American allies (Hartman and Hsiao 1988; Nagle 1985; Scoble
3 Examples of collective goods include government regulation and infrastructure. Also, transfer payments
include, health insurance, unemployment benefits and pensions (Boix 2001).
5
and Wiseberg 1988). Others who favor using Polity measures argue that Freedom House scores
contain a liberal bias against American allies (Bozeman 1979; Feen 1985). Proponents of
Freedom House measures argue that these ratings are relatively unbiased (Bollen 1993;
McCamant 1981) and that Polity scores are biased against monarchies (Herb 2004). From this
controversy, one can ask whether any relationship between economic development and
democracy depends on how one measures democracy.
Hypotheses
Economic Development and Democracy
We should expect a relationship between economic development and democracy for four
reasons. First, the populace’s demand for democracy grows as their nation-state becomes
economically better off. People within that country tend to demand democracy because rapid
economic growth is more likely to destabilize authoritarian regimes in favor of Democratic ones
(Gasiorowski 1995; Huntington 1991).
Second, the values of a nation-state’s citizens change as that country economically develops
(Huntington 1991; Putnam 1993). More specifically, citizens develop more feelings of
interpersonal trust, life satisfaction, and competence as the economy develops (Huntington 1991;
Putnam 1993). These attributes of social capital strongly correlate with whether democratic
institutions exist.
Third, economic development increases a society’s level of education (Huntington 1991;
Lipset 1959). Education is important to the development of democratic institutions because more
highly educated people, “…tend to develop the characteristics of trust, satisfaction and
competence that go with democracy” (Huntington 1991, 65-66). Also, education favorably
affects the development of democracy because citizens of more educated societies are more
6
likely to tolerate opposing viewpoints and believe in multi-party rather than single-party systems
(Lipset 1959).
Finally, economic development makes more resources available for the government to
distribute to social groups (Huntington 1991). If a government has more resources, then it forces
groups within and outside of the government to compromise with one another to determine who
receives what share of the resources (Huntington 1991). Democracy is more likely when groups
must compromise and accommodate one another in such a situation because such actions teach
members of those groups democratic values. For all of these reasons, I anticipate that:
Hypothesis One: ECONOMIC DEVELOPMENT. A country is more likely to be a democracy
as it economically develops.
World System Position and Democracy
Further, we know that a country’s chances for democracy depend on where it is placed within
the world’s economy (Bollen 1983; Burkhart and Lewis-Beck 1994; Snyder and Kick 1979;
Wallerstein 1974). Dependency theorists argue that core countries have a considerable economic
and political advantage over peripheral and semi-peripheral countries.4 Core countries have an
advantage over peripheral and semi-peripheral nation-states in three respects. First, semi-
peripheral and peripheral countries have production systems within their economies that are
export-oriented and usually specialize in raw materials (Snyder and Kick 1979; Wallerstein
1974). Second, multinational corporations from core countries use these economic structures to
exploit peripheral and semi-peripheral countries’ resources to benefit core nation-states
(Wallerstein 1974). Finally, core countries have substantial control over the activities of political
4 Core nation-states are those countries with high gross domestic products (GDPs), high standards of
living and transnational corporations. Semi-peripheral countries are those states that have medium GDPs,
moderate standards of living and no multinational corporations. Peripheral countries are those nation-
states with low GDPs, low standards of living and are exploited by transnational corporations from core
countries. See Wallerstein (1974) for more details.
7
and economic elites in peripheral and semi-peripheral nation-states that prevent the latter from
breaking off the relationship (Wallerstein 1974). Peripheral and semi-peripheral countries face
serious constraints on their ability to become democracies because of the advantage that core
nation-states have over these countries. Therefore, I anticipate that:
Hypothesis Two: WORLD SYSTEM POSITION. Peripheral and semi-peripheral nation-
States are less likely to be democracies than core states.
Oil Countries and Democracy
Finally, oil countries fundamentally differ from those nation-states that are not major oil
exporters because such countries tend to be one party states or ruled by some kind of monarch
(Herb 2004; Huntington 2004). Oil countries are more likely to have such political structures
because the wealth that oil sales generate goes to the government and/or to multinational
corporations who have operations in those countries (Barro 1999; Przeworski, Alvarez, Cheibub
and Limongi 2000). Further, governments of oil countries rarely redistribute their wealth to the
population that lives in these countries (Barro 1999). Since we know that countries that rarely
redistribute their wealth are less likely to be democracies (see Gasiorowski 1995), I anticipate
that:
Hypothesis Three: OIL COUNTRIES. Oil countries are less likely to be democracies.
Data and Methods
My data focuses on 180 countries from 1972 to 1995 and describes the degree to which each
country is a democracy. My dependent variable is each nation-state’s democracy score from
1972 to 1995. Since researchers disagree about how to measure democracy, I use two different
measures.5 First, I measure democracy using Polity III Scores for each country during this time
period. Jaggers and Gurr (1995) coded their democracy rating from zero to ten, with higher
8
rankings assigned to more democratic states.6 Second, I measure democracy using Freedom
House scores. This organization coded each country’s score from two to fourteen based on civil
and political rights within a nation-state, with a lower score indicating a more democratic state. I
rescaled these scores by reducing each by two to create a scale from zero to twelve. The intuitive
meaning of these scores is the same as the original measurement in that a lower score still
indicates a more democratic state.7
My first independent variable is economic development. I use kilowatt hours of electricity
consumption per capita to measure economic development.8 I acquired data for this variable
from the World Development Indicators CD-ROM (1998). I took the natural log of the values for
each country in every year for two reasons. First, "… the relationship between development and
5 One benefit of using different measures is that I can determine whether the relationship between my
dependent and independent variables depends on how I measure my dependent variable.
6 Please see Jaggers and Gurr’s (1995) article for more information on how they operationalize each facet
of a democracy.
7 I recoded these scores to make the Beck, Katz and Tucker (1998) corrections work with my data.
8 Researchers vehemently disagree about how to measure economic development. Some researchers have
measured this concept via gross national or gross domestic product (Barro 1999; Bollen and Jackman
1985; Jackman 1973). One benefit of measuring economic development in this manner is that most
researchers agree that GNP and GDP per capita adequately measures economic development. However,
measuring economic development in this manner has two significant drawbacks. First, values of GNP and
GDP are susceptible to the exchange rate problem (Bollen 1979, 578). This problem occurs when a
researcher converts the GNP/GDP value from one currency to another. The values of GNP and GDP
fluctuate depending on when the researcher converted these values from that country’s currency to
another. Given the floating exchange rate system, these values vary considerably.
Second, we observe a related problem that concerns comparability in that currencies fluctuate in value
as compared with each other (Bollen 1979, 578). Such a problem comes about in two different ways.
First, when researchers convert several nation-states' GDP or GNP into a third currency, values for those
countries partly depend on the exchange rate on the particular day, hour and minute that the conversion
takes place. Second, this problem surfaces when researchers compare the GDP or GNPs of Communist
and noncommunist countries. It is widely acknowledged that Communist GDP and GNP calculations (and
their per capita derivatives) are inaccurate such that the margin of inaccuracy is not consistent across
years and countries.
Because of these problems, I cannot use GNP and GDP per capita to measure economic development.
Instead, I use per capita electricity consumption. Using this measure carried two benefits. First, at least
two major studies have employed this measure as a substitute for GNP and GDP measures (see Bollen
1979; Burkhart and Lewis-Beck 1994). Second, this measure is highly valid at face value because it
carries a .9 correlation with GNP per capita over time (Bollen 1979).
9
democracy is curvilinear and can be best captured by a log transformation of energy
consumption" (Jackman 1973; Bollen 1979, 578). Second, such a transformation, "…reduces the
extreme skewness in the untransformed energy consumption variable" (Bollen 1979, 578).
My second independent variable is world system position. World systems theorists argue that
each country fits into one of three categories: core, periphery or semi-periphery. To measure
world system position, I employ Burkhart and Lewis-Beck’s (1994) ratings for each country.9
Because I want to test whether semi-peripheral and peripheral countries are less likely to be
democracies than core nation-states, I use core countries as my comparison category. I employ
two dummy variables that correspond to peripheral and semi-peripheral nation-states,
respectively. I code the peripheral dummy one of a country is located in the periphery and zero
otherwise. Also, I code the semi-peripheral dummy one if a nation-state is located in the semi-
periphery and zero otherwise. Since dependency theory argues that the effect of world system
position is conditioned upon that state’s economic well-being, I interact each peripheral dummy
with my measure of economic development.
My third independent variable measures whether a country is an oil country. A state is an oil
country if it is a member of the Organization of Petroleum Exporting Countries (OPEC). I code
this variable one if a nation-state is an OPEC member and zero otherwise. My final independent
variable is a lagged, instrumental dependent variable. I include such a variable for two reasons.
First, one must account for other social factors that can have an effect on Democracy (Burkhart
and Lewis-Beck 1994). Second, one must control for any and all unintentionally omitted
independent variables. An instrumental lagged dependent variable performs quite nicely on both
counts.
9 See Appendix B for more details on how these ratings were constructed and expanded.
10
Because both measures of my dependent variable are discrete and ordered in a cross-
sectional dataset, I test my hypotheses via ordered probit with Beck, Katz and Tucker (1998)
corrections. I cannot employ Ordinary Least Squares (OLS) to test my hypotheses for three
reasons: 1) the observations are not independent of each other; 2) the errors are heteroskedastic;
and 3) there is a considerable amount of missing data.10 To control for temporal dependence, I
employ Beck, Katz and Tucker’s (1998) corrections which create a series of splines. To correct
for heteroskedastic errors and spatial dependence, I employ White (1980) standard errors.
Results
I present my results in Table One. Model One shows the results for Polity III Scores whereas
Model Two displays the results for Freedom House scores.
[TABLE ONE ABOUT HERE]
Model One shows strong support for the economic development hypothesis because the sign
for that coefficient is positive and statistically significant. Also, Model One confirms both facets
of the world system position hypothesis. Model One shows that peripheral and semi-peripheral
nation-states are less likely to be democracies than core countries because the signs on the
coefficients for these variables are negative and statistically significant. Further, Model One
confirms that oil countries are less likely to be democracies because the sign for that coefficient
is negative and statistically significant. Finally, Model One shows that knowing the values of all
of the independent variables is quite useful in predicting the value of my dependent variable.
Model Two demonstrates support that favors the economic development hypothesis because
the sign which corresponds to that coefficient is negative and statistically significant.11
10 I do not rectify my missing data problem because Amelia does not allow an analyst to impute missing
data in cases where there is an ordered dependent variable contained within a time series, cross-sectional
dataset.
11 Since the Freedom House scores are lower more democratic states, my finding holds.
11
Moreover, Model Two confirms both parts of the world system position hypothesis in that the
signs for both coefficients are positive and statistically significant. Additionally, Model Two
shows that oil countries are less likely to be democracies because the sign for that coefficient is
positive and statistically significant. Finally, Model Two shows that knowing the values of all of
the independent variables is quite useful in predicting the value of my dependent variable.
Discussion and Conclusion
I measured my dependent variable using Polity and Freedom House scores and found that
countries are more likely to be democracies as they economically develop. I have three possible
explanations for my finding. First, people within economically developing societies are more
likely to demand democracy. Second, economic development creates more resources for the
government and other actors to distribute to social groups. These groups must work with one
another to determine how they plan to allocate the resources. Finally, economic development
affects other factors that push a nation-state into becoming a democracy — like education and
interpersonal trust (Huntington 1991; Lipset 1959; Putnam 1993).
In addition, I discover that peripheral and semi-peripheral countries are less likely to be
democracies than core ones. More specifically, peripheral countries are even less likely to de
democracies than semi-peripheral ones. I have two explanations for my discovery. First, core
countries economically and politically exploit peripheral and semi-peripheral ones such that
these countries are less likely to be democracies. Second, semi-peripheral countries are more
likely to be democratic than peripheral ones because core nation-states repress the semi-
periphery much less than the periphery. Such a relationship holds because semi-peripheral
countries have more ability to resist the core’s influence than peripheral states (see Wallerstein
1974). Semi-peripheral states can resist core nation-states in a limited sense because the many
12
states within the former category are trying to industrialize their economies. Past exemplars of
this trend include South Korea, Malaysia and Thailand.
Finally, I learn that oil countries are less likely to be democratic than non-oil exporting ones.
I have three explanations for my finding. First, oil countries are less likely to redistribute their
wealth to their citizens than other nation-states (Barro 1999). Second, such states tend to have
oppressive one party or monarchical political systems which enforce the non-redistribution of
wealth (Herb 2004; Huntington 1991). Finally, such states have considerable economic
inequalities between rich and poor. We know that countries with great income inequality are less
likely to be democracies (Gasiorowski 1995).
To conclude, I discover strong evidence that supports the economic development, world
system position and oil country hypotheses. Also, I find that my results hold regardless of how I
measure democracy and after correcting for observational interdependence, spatial dependence
and heteroskedasticity. Further, my results hold in light of recent theoretical challenges to
Burkhart and Lewis-Beck’s (1994) piece. Finally, my results hold after many new democracies
have held one election.
My piece suggests at least two avenues for future research. First, one could extend my
dataset to more effectively account for Huntington’s (1991) two turnover test. Finally, one could
extend my model to more explicitly account for social (e.g. education and social capital) and
natural resource-based factors that may help explain why some countries are democratic and
others are not.
13
Appendix A: Summary Statistics of Dependent and Independent Variables:
Variable Number of
Observations Mean Standard
Deviation Minimum Maximum
Polity Scores 3331 3.562 4.305 0.000 10.000
Freedom House
Scores 3941 6.400 4.115 0.000 12.000
Logged Per
Capita Energy
Consumption
4017 6.177 1.760 1.800 9.950
Oil Country 4098 .073 .261 0.000 1.000
Peripheral
Country*Logged
Per Capita
Energy
Consumption
3657 3.280 2.783 0.000 9.41
Semi-Peripheral
Country*Logged
Per Capita
Energy
Consumption
3657 1.867 3.294 0.000 9.95
14
Appendix B: World System Position Ratings
I used Burkhart and Lewis-Beck's (1994) position ratings for each country to construct my
second independent variable, world position. Since their data only goes to 1990, these rankings
are somewhat problematic as many countries have come into (or out of) existence since then. For
countries that began since 1990, I constructed ratings based on how well they meet the specified
conditions that determine whether a nation-state is a peripheral, semi-peripheral or a core
country. If they met the conditions for a peripheral country, then they were operationalized as
such. The same goes for semi-peripheral and core nation-states. These rankings will be listed
starting with the country's name, and a p, s or c by it. These letters signify periphery,
semiperiphery, and core as defined by Wallerstein (1974) and others. These ratings are as
follows:
Angola, p; Armenia, p; Azerbaijan, p; Belarus, s; Botswana, p; Croatia, s; Czech Republic, s;
Estonia, s; Georgia, p; Guinea-Bissau, p; Kazakhstan, p; Kyrgyz Republic, p; Latvia, s;
Lithuania, s; Macedonia, s; Moldova, p; Mozambique, p; Namibia, p; Oman, s; Palau, p; Papua
New Guinea, p; Russia, s; Slovakia, s; Slovenia, s; Spain, c; Taiwan, s; Tajikistan, p;
Turkmenistan, p; Ukraine, s; Vietnam, p; Vietnam p; Yemen (South, when it existed), p; Yemen,
p.
I changed one nation's rating from the Burkhart and Lewis-Beck (1994) ratings. I changed South
Korea's rating from semiperiphery to core because this nation has enlarged its economy such that
it is the eleventh largest economy in the world according to the last World Bank GDP figures
(World Bank 1998). Moreover, this nation also has several transnational corporations, where
previously none existed. Given these facts, as well as this nation's tremendous increases in per
capita energy consumption and GDP since 1961 and the inception of the Developmental State
under General Park Chung-Hee, I could not call South Korea a semiperipheral nation. It has
become a core nation.
15
References
Barro, Robert 1999. "Determinants of Democracy." The Journal of Political Economy 107:158-
183.
Baum, Matthew A. and David A. Lake 2003. “The Political Economy of Growth: Democracy
and Human Capital.” American Journal of Political Science 47: 333-347.
Beck, Nathaniel; Jonathan N. Katz and Richard Tucker 1998. “Taking Time Seriously: Time-
Series Cross-Section Analysis with a Binary Dependent Variable.” American Journal of
Political Science 42: 1260-1288.
Boix, Charles 2001. “Democracy, Development and the Private Sector.” American Journal of
Political Science 45: 1-17.
Bollen, Kenneth A. 1979. "Political Democracy and the Timing of Development." American
Sociological Review 44: 572-587.
Bollen, Kenneth A. 1983. "World System Position, Dependency, and Democracy: The Cross-
National Evidence." American Sociological Review 48: 468-479.
Bollen, Kenneth A. 1993. "Liberal Democracy: Validity and Method Factors in Cross-National
Measures." American Journal of Political Science 37: 1207-1230.
Bollen, Kenneth A. and Robert W. Jackman. 1985. "Political Democracy and the Timing of
Development." American Sociological Review 50: 438-457.
Bozemann, Adda 1979. “Review of Freedom in the World: Political Rights and Civil Liberties,
1978.” Orbis 23: 213-220.
Burkhart, Ross E. and Michael S. Lewis-Beck 1994. "Comparative Democracy: The Economic
Development Thesis." American Political Science Review 88: 903-910.
16
Burkhart, Ross E. 1997. “Comparative Democracy and Income Distribution: Shape and
Direction of the Causal Arrow.” The Journal of Politics 59: 148-164.
Cedarman, Lars-Erik 2001. “Back to Kant: Reinterpreting the Democratic Peace as a
Macrohistorical Learning Process.” The American Political Science Review 95: 15-31.
Dahl, Robert A. 1989. Democracy and its Critics. New Haven, Connecticut: Yale University
Press.
Dixon, William J. 1994. “Democracy and the Peaceful Settlement of International Conflict.” The
American Political Science Review 88: 14-32.
Feen, Richard H. 1985. “Review of Freedom in the World: Political Rights and Civil Liberties,
1984-1985 by Raymond Gastil and Country Reports on Human Rights Practices for 1984
by U.S. Department of State.” Washington Quarterly 8: 237-241.
Feng, Yi 1997. “Democracy, Political Stability and Economic Growth.” British Journal of
Political Science 27: 391-418.
Freedom House 2003. Freedom in the World 2003: The Annual Survey of Political Rights and
Civil Liberties. Landham, Maryland: Rowman and Littlefield Publishers.
Gasiorowski, Mark J. 1995. “Economic Crisis and Political Regime Change: An Event History
Analysis.” The American Political Science Review 89: 882-897.
Granger, C.W.J. 1969. "Investigating Causal Relations by Econometric Models and Cross-
Spectral Methods." Econometrica 37: 424-438.
Granger, C.W.J. 1988. “Some Recent Developments in a Concept of Causality.” Journal of
Econometrics 39: 7-21.
Hartman, John and Wey Hsiao 1988. “Inequality and Violence: Issues of Theory and
Measurement in Muller.” American Sociological Review 53: 794-800.
17
Herb, Michael 2004. “No Representation Without Taxation? Rents, Development and
Democracy.” Comparative Politics, Forthcoming.
Huntington, Samuel P. 1991. The Third Wave: Democratization in the Late Twentieth Century.
Norman, Oklahoma: Oklahoma University Press.
Jackman, Robert W. 1973. “On the Relation of Economic Development to Democratic
Performance.” American Journal of Political Science 17: 611-621.
Jackman, Robert W. and Ross A. Miller 1996a. “A Renaissance of Political Culture?” American
Journal of Political Science 40: 632-659.
Jackman, Robert W. and Ross A. Miller 1996b. “The Poverty of Political Culture.” American
Journal of Political Science 40: 697-716.
Jaggers, Keith and Ted Robert Gurr 1995. “Tracking Democracy’s Third Wave with the Polity
III Data.” Journal of Peace Research 32: 469-482.
Jaggers, Keith, and Ted Robert Gurr. 1995. Polity III: Regime Change and Political Authority,
1800-1994 [Computer file]. 2nd ICPSR version. Boulder, CO: Keith Jaggers/College
Park, MD: Ted Robert Gurr [producers], 1995. Ann Arbor, MI: Inter-university
Consortium for Political and Social Research [distributor], 1996.
Leblang, David A. 1997. “Political Democracy and Economic Growth: Pooled Cross-Sectional
and Time-Series Evidence.” British Journal of Political Science 27: 453-466.
Lipset, Seymour Martin 1959. "Some Social Requisites of Democracy: Economic Development
and Political Legitimacy." American Political Science Review 53: 69-105.
Lipset, Seymour Martin 1994. “The Social Requisites of Democracy Revisited.” American
Sociological Review 59: 1-22.
18
Maoz, Zeev and Bruce Russett 1993. “Normative and Structural Causes of Democratic Peace,
1946-1986.” The American Political Science Review 87: 624-638.
McCamant, John F. 1981. “A Critique of Present Measures of ‘Human Rights Development’ and
an Alternative.” In Global Human Rights: Public Policies, Comparative Measures, and
NGO Strategies, ed. Ved P. Nanda, James R. Scarritt and George W. Shepherd, Jr.
Boulder, Colorado: Westview Press.
Nagle, John D. 1985. Introduction to Comparative Politics: Political Systems and Performance
in Three Worlds. Chicago, Illinois: Nelson-Hall Publishers.
Przeworski, Adam and Fernando Limongi 1997. “Modernization: Theories and Facts.” World
Politics 49: 155-183.
Przeworski, Adam and Henry Teune 1970. The Logic of Comparative Social Inquiry. Melbourne,
Florida: Krieger Publishing Company.
Przeworski, Adam; Michael E. Alvarez; Jose Antonio Cheibub and Fernando Limongi 2000.
Democracy and Development: Political Institutions and Well-Being in the World, 1950-
1990. Cambridge University Press: New York City, New York.
Putnam, Robert D. 1993. Making Democracy Work: Civic Traditions in Modern Italy. Princeton,
New Jersey: Princeton University Press.
Rice, Tom W. and Jeffrey Ling 2002. “Democracy, Economic Wealth and Social Capital:
Sorting Out the Causal Connections.” Space and Polity 6: 307-325.
Scoble, Harry M. and Laurie S. Wiseberg 1981. “Problems of Comparative Research on Human
Rights.” In Global Human Rights: Public Policies, Comparative Measures, and NGO
Strategies, ed. Ved P. Nanda, James R. Scarritt and George W. Shepherd, Jr. Boulder,
Colorado: Westview Press.
19
Smith, Benjamin 2004. “Oil Wealth and Regime Survival in the Developing World, 1960-1999.”
American Journal of Political Science 48: 232-246.
Snyder, David and Edward L. Kick 1979. "Structural Position in the World System and
Economic Growth, 1955-1970: A Multiple Network Analysis of Transnational
Interactions." American Journal of Sociology 84: 1096-1126.
Wallerstein, Immanuel 1974. "The Rise and Future Demise of the World Capitalist System:
Concepts for Comparative Analysis." Comparative Studies in Society and History
16: 400-415.
White, Halbert 1980. “A Heteroskedasticity-Consistent Covariance Matrix Estimator and a
Direct Test for Heteroskedasticity.” Econometrica 48: 817-838.
World Bank 1998. World Development Index CD-ROM.
20
Table One: Ordered Probit Analysis on the Effects of Economic Development, Peripheral, Semi-
peripheral and Oil Country Status on Democracy, N=4098.
Independent Variable Polity Scores
(Model One) Freedom House Scores
(Model Two)
Logged Per Capita Energy
Consumption
.240***
(.051) -.182***
(.025)
Oil Country
-.594**
(.204) .246**
(.082)
Peripheral Status*Logged Per
Capita Energy Consumption
-.121***
(.048) .118***
(.021)
Semi-Peripheral
Status*Logged Per Capita
Energy Consumption
-.102*
(.045) .106***
(.020)
Lagged Democracy Score
.608***
(.039) .868***
(.038)
Spline One
.0005
(.003) .0003***
(.0001)
Spline Two
.012
(.003) -
Spline Three
-.004
(.001) -
Cut One
1.868
(.302) -.666
(.238)
Cut Two
2.233
(.302) .625
(.198)
Cut Three
2.684
(.296) 1.952
(.197)
Cut Four
3.023
(.302) 3.102
(.208)
Table One Continued on Next Page
21
Table One (Continued): Effects of Economic Development, World Position and Oil Country on
Democracy Using Splines.
Independent Variable Polity Scores
(Model One) Freedom House Scores
(Model Two)
Cut Five
3.229
(.313) 3.703
(.222)
Cut Six
3.609
(.346) 4.503
(.258)
Cut Seven
4.047
(.376) 5.215
(.297)
Cut Eight
4.576
(.429) 5.963
(.328)
Cut Nine
5.800
(.472) 6.896
(.370)
Cut Ten
6.554
(.471) 7.850
(.399)
Cut Eleven
- 8.811
(.436)
Cut Twelve
- 9.810
(.453)
N 3051 3368
Pseudo R2 .565 .500
Log Likelihood -2067.13 -4193.066
Wald Chi-Squared (7) 389.58 789.17
Prob>Chi-Squared .0001 .0001
Notes: * denotes p<.05, ** denotes p<.01, ***denotes p<.001; all two-tailed tests. Also, STATA
dropped splines two and three in Model Two due to collinearity.
