INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps.
Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.
Bell & Howell Information and Learning 300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA UMI 800-521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with with permission permission of the of copyright the copyright owner. owner.Further reproduction Further reproduction prohibited without prohibited permission. without permission. PETROLEUM PROFITS AND PARTICIPATORY POLITICS:
EVALUATING THE EMPIRICAL VALIDITY OF THE RENTIER STATE THEORY
by
Eric David Hornberger
submined to the
Faculty of the School for International Service
of American University
in Partial Fulfillment of
the Requirements for the Degree of
Master of Arts
in
International Affairs
Committee Approval:
Dr. Allan^t^m,'Chair
TDr. Simona Sharoni
Dean of the School for International Service
lb ______Date
1999 American University Washington, D.C. 20016 TlMERICftH ONIVERSmf UBRftH
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI Number 1398908
___ ® UMI
UMI Microform 1398908 Copyright 2000 by Bell & Howell Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.
Bell & Howell Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. PETROLEUM PROFITS AND PARTICIPATORY POLITICS:
EVALUATING THE EMPIRICAL VALIDITY OF THE RENTIER STATE THEORY
BY
Eric David Homberger
ABSTRACT
This study traces the evolution of the Rentier State Theory, quantifies its primary
components, and rigorously tests the extent to which the negative impact of economic rent
upon the advancement of democracy in the Arab world can be substantiated empirically.
Through a series of multivariate regressions done on the primary components of the Rentier
State Theory and quantitative democratic indicators, little empirical evidence was found to
support the common hypotheses espoused in rentier literature. In fact, there is every
indication that many of the assumptions concerning the relationship between the sources of
state revenue and national political behavior are actually contrary to empirical evidence.
While groundbreaking in bringing a quantitative, more systematic dimension to a largely
qualitative field of political inquiry, further research is needed to further substantiate the
conclusions drawn in this study and to more fully understand the impact of regional oil
wealth in shaping Arab political development.
II
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS
ABSTRACT...... ii
LIST OF TABLES...... iv
Chapter
1. Introduction ...... 1
2. Theoretical Foundations ...... 11
3. Establishing an Empirical Framework ...... 52
4. Analyzing the Results...... 84
5. Conclusion ...... 107
BIBLIOGRAPHY...... 112
iii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES
1. INDICATORS OF THE RENTIER STATE THEORY...... 79
2. MODELING RENTIER INDICATORS OF NATIONAL POLITICAL BEHAVIOR IN TH E ARAB W ORLD...... 86
3. SPECIFYING MODELS OF RENTIER POLITICAL PATTERNS IN THE ARAB WORLD O V E R T IM E ...... 89
4. MODELING FOR RENTIER PATTERNS OF NATIONAL POLITICAL BEHAVIOR WORLDWIDE...... 97
5. SPECIFYING MODELS FOR NATIONAL POLITICAL BEHAVIOR WORLDWIDE OVER TIME...... 99
iv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 1
INTRODUCTION
Democracy in the Arab World
Heralded as “probably the most important political trend in the late twentieth
century” by one of the premiere political scientists of the late 20th century, the ‘third wave’
o f global democratic revolution that occurred between 1974 and 1990 witnessed the
transition of more than thirty countries to democradc systems o f governance.1 From East
Asia to Ladn America, countries that entered the 1980s ruled by autocradc regimes exited
the decade more politically ‘free’ and well on the road to democracy. Yet, despite the global
nature of this trend, the Middle East continues to elude substantive democratization.
According to a recent cross-national, comparative study of democracy’s ‘third wave’, only in
the Middle East and Africa are patterns of governance in the late 20 * century found to be
more autocratic than democratic in only the Middle East and Africa.2 Furthermore, this
study found that while in Africa the prospects for democracy are uncertain, “the prognosis
for democratization in the Middle East is even more tenuous.” 3 With the vast majority of
Arab states entering the new millennium still controlled by either tribal monarchies or
military regimes, these findings are not at all shocking and have come to characterize the
1 Samuel P. Huntington, “How Countries Democratize,” Political Science Ouarterlyvol. 106, p-579. See also an expanded discussion of the ‘Third Wave” in Huntington’s The Third Wave: Democratization in the Late Twentieth Centur/OldahomaCityt University of Oklahoma Press, 1991), p. 579. 2 Keith Jaggers and Ted Robert Gurr, “Tracking Democracy’s Third Wave with the Polity III Data,” Toumal o f Peace Research vol. 32, no. 4 (1995), p. 476.
1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2 politics of the Arab world in the minds of global public opinion, policy experts and scholars
alike.
Scholars o f the Middle East have also noticed this noteworthy political anomaly. In
and investigation of the prospects for democratic change in the aftermath of the Gulf War,
Michael Hudson notes that “. . . authoritarianism has been the dominant feature of modern
Axab politics.”4 Similarly, according to William Quandt, the region categorically suffers from
what he calls a “political deficit” and argues that:
The biggest challenge facing the Middle East in the years to come is the development o f better systems o f governance. This means governments that are accountable, in some acceptable manner, to their people. 5
While Quandt notes that while regional economic challenges remain significant, he finds
that this political challenge is “at the core of the economic deficit.”6 Considering the
enormous resources, relatively high rates of modernization, and integrated nature of trade
relations with the rest of the world, one would assume that this region would have a
heightened sensitivity to global trends, not an aversion to them. W hat is the cause of such
retarded political development? These and many more observations about the seemingly
anachronistic politics of the Axab world have fueled an increasing number of more recent
investigations, making it one of the most studied topics in regional scholarship.
Survey of Broader Scholarly Debate
Within the discussion o f democracy and the Middle East, several areas of analytical
inquiry have polarized in the late 1980s and early 1990s with the intent on isolating and
3 Ibid., p. 478. * Michael C. Hudson, “After the Gulf Wan Prospects for Democratization in the Arab World," Middle East rournal. v. 45, no. 3 (Summer 1991), p. 407. 5 William Quandt, “The Middle East on the Brink: Prospects for Change in the 21st Century," Middle East rournal. v. 50, no. I (Winter 1996), p. II.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3 explaining the reasons for the region’s persistent resistance to global trends of political
liberalization. These broad topic areas can be categorized around four general themes:
political culture, civil society, regional and international context, and political economy.7
Prior to launching into a discussion of che latter of these four major areas, which is the focus
o f this particular inquiry, it may be helpful to survey briefly the other thematic areas in order
to understand better the broader context in which the political economy discussion ensues.
Culture
The first of these general themes involves the distinct Arab and Islamic nature of the
region. The recent manifestation of this rather old debate involving the role of culture in
predisposing particular patterns of national political behavior involves a number o f various
positions ranging from the inherent incompatibility of the Arab/Islamic culture with the
fundamental principals of democracy to the assertion that both authoritarian and democratic
patterns of governance exist within religious and traditional spheres of Arab society.
Probably the most widely debated of the four issue areas, this discussion of culture has
provided a rich and diverse body of literature from such highly regarded regional scholars as
Daniel Pipes, David Pryce-Jones, Michael Hudson, John Esposito, and John Waterbury, just
to name a few.8
6 Quandt, p. 14. See also Allan Richards, “Economic Imperatives and Political Systems,” Middle East rournal . vol. 47, no. 2 (Spring 1993), p. 217. 7 The Erst identification, articulation, and comprehensive analysis of these major themes can be (bund in Rex Brynen, Bahgat Korany and Paul Noble, eds. Political Liberalization and Democratization in the Arab World _, Boulder, Colorado: Lynne Rienner Publishers, Inc., 1995- * For a sampling of noteworthy “cultural” school literature, see Daniel Pipes, In the Path o f God: Islam and Political Power (New York: Basic Books, 1983); David Pryce-Jones . The Closed Circle: An Interpretation of the Arabs (New York: Harper and Row, 1989); Michael Hudson, Arab Politics: The Search (or Legitimacy _ (New Haven: Yale University Press, 1977); John Esposito, The Islamic Threat: Myth or Reality? (Oxford: Oxford University Press, 1992); and John Waterbury, “Democracy without Democrats? The Potential for Political Liberalization in the Arab World,” in Ghassan Salami, ecL, Democracy without Democrats: The Renewal of Potties in the Muslim World (London: LB. Tauris, 1994).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4 While the notion of cultural predeterminalism has intensified in recent decades, the
fundamental presuppositions are not new nor are they specific to the Arab/Islamic region. It
is even doubtful whether this path of inquiry leads to further understanding of the forces of
political change within the region or simply perpetrates a crude fatalism and ridged
stereotypes that reflect a contemporary form of Orientalism.9 Nevertheless, this angle of
inquiry has helped to promote a broader discussion and analysis of current political trends
within academic and policy-making circles.
Civil Society
The second major theme around which another significant body of literature on the
subject of the potential for Arab democracy has emerged is that of civil society. While part
of a much broader discussion that dates back to the observations of the de Toqueville’s
analysis of the democratic experiment in American society, application of the concept of civil
society to the study of the Middle East is relatively nascent and was propelled by the
experience of political development occurring in other regions during the late 1980s.
Scholarly interest in the civil society of the Middle East is certainly growing both inside and
outside the region as evidenced by the number of relatively recent, collaborative research
projects, conferences and symposia dedicated exclusively to the discussion o f the existence,
nature and potential of Arab civil society. Scholars o f this topic tend to emphasize either the
various non-state entities which span both the public and private domain, the quality and
nature of Arab civil society in general, or the relationship and interaction between state and
society. A sample of civil society literature specifically applied to the region can be found in
9 Cultural predeterminalism is not new to the region. See Edward Said’s benchmark work, Orientalism (New York: Pantheon, 1978) for a trenchent analysis o f the tole of culture in Middle Eastern scholarship.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5 the writings of A. Richard Norton, Ernest Gellner and Robert Bianchi.10 As a tool for
understanding of the complex nature of Arab society, the civil society paradigm is most
helpfu.1. However, for analytical purposes, the concept has proven difficult to define, isolate
or quantify as an influence on Arab political development in any particular direction.
External Influences
A third theme that has coalesced in recent years involves the role and extent to which
external influences shape Arab domestic politics. Emerging out of general discussions of the
global diffusion of democratic values, increasing interdependence from economic
globalization, the spread of empowering technological innovation, and the role of
superpower foreign policy, a small body of literature has emerged that seeks to apply these
broad concepts specifically to the region of the Middle East. While the majority of this
literature has been more rhetorical than scholastic, there are a few regional scholars that have
also sought to apply these concepts specifically to internal Arab politics, the most noteworthy
of which are the writings of William Quandt and Lisa Anderson.11 Despite their academic
focus of the external forces that may impact political trends within the region, few scholars
would challenge the ultimate primacy of internal factors.
Political Economy
The fourth and final theme chat has emerged over the last few decades and which is
the subject of this particular inquiry is that of the role of the political economy in shaping
10 See Augustus Richard Norton, “The Future o f Civil Society in the Middle East,” Middle East Journal . vol. 47, no. 2 (Spring 1993); Ernest Gellner, “Civil Society in Historical Context,” International Social Science Journal. vol. 129 (August 1991); and Robert Bianchi, Unruly Corporatism: Associational Life in Twentieth- _ Century Egypt (New York: Oxford University Press, 1989). 11 For an example of the various topics covered under this broad theme, see William Quandt, “American Policy toward Democratic Political Movements in the Middle East,” in Ellis Goldberg, Rasat Kasaba, and Joel Migdal, eds.. Rules and Rights in the Middle East (Seattle: UniversityofWashington Press, 1993); and Lisa Anderson, “Absolutism and the Resilience of Monarchy in the Middle East,” Political Science Quarterly ■ vol. 106, no. 1, 1991, p. 3. Anderson’s article is noteworthy in it she argues that monarchy as it currently exists within the region is largely a colonial legacy and not a particularly indigenous phenomenon as commonly believed.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 6 recent patterns of Arab governance. For sure, the study of relationship between economic
development and political change in general is not, dating back to the time of Aristode, and
has been the subject of intense intellectual debate throughout the last century.12 However,
its systematic application within the Arab context is a rather new development.13 One of the
major theories to develop out of an investigation of the economic influences o f political
development with the Arab world is also loosely connected to external factors. This
particular genre of thought, commonly referred to as the Rentier State Theory, focuses on
the role of the structure of Arab public finance in shaping state/society relations. According
to one of the Rentier State Theory’s chief architects, Giacomo Luciani, “. . . whenever
income or grants from foreign sources are an important source of government revenue, a
tendency towards authoritarian rule is created or reinforced.”14 According to this theory,
external rent or more, domestically passive, foreign revenue is, therefore, considered to have a
powerful effect on strengthening state autonomy and inspiring political passivity among the
broader society. Through this theory, the economic concept of rent is uniquely linked to the
evolution of domestic political structures within the Arab world.
Economic Foundations
While its application in the Middle Eastern context is a relatively modern
development, this concept of economic rent itself is not. The earliest articulation of
economic rent actually emerged with the formation of modern economic theory as set forth
by some of the foremost economic philosophers, namely Adam Smith, David Ricardo and
12 Political economy as a general field o f inquiry has inspired many theories within social scientific research and political discourse, the lease of which include Liberalist, Marxist, Dependancy and Modernization theories. A noteworthy exception to this recent development is the Arab scholar Ibn Khaldun whose writings on the relationship o f economics and politics predated those of the great western political and economic theorists of the Enlightenment which, interestingly enough, were derived entirely from his medieval Arab/Islamic context.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 Karl Marx. These early economists were the first to note and articulate the unique enterprise
of rent as an economic concept with unique economic consequences.15 Stated most
succincdy, economic rent represents the profit earned from the forfeiture of monopolistic use
of real property. In the pre-modern context, this concept was primarily applied to the
agricultural land or mining rights of European aristocracy. These early rentiers, or
proprietors of rent, were able to make sizable profits from their ability to collect a percentage
share of agricultural production or mineral excavation done on their property, even if they
themselves did not participate in the actual laboring process. Wealth was made through
indirect activity, only dependent upon a favorable economic environment and not physical
enterprise.
While key to understanding feudal economies and in developing theories of capital
formation, rent lost much of its theoretical utility and subsequendy lapsed in application
during the industrial age. It was not until the globalization of the petroleum market and the
widespread nationalization of the oil industries within the Arab world in the early 1970s that
the concept re-emerged with renewed interest. Unlike the rent conceptualizations of the
early economists that focused on the economic effects of rent, however, the concept took on
this new, more political dimension when applied to the modern Middle East.
Commonplace Assumptions
With the increasing importance of oil in the late 20th century global marketplace
and the need to understand the reasons for the region’s persistent resistance to global
M Giacomo Luciani, “Economic Foundations of Democracy and Authoritarianism: The Arab World in Comparative Perspective,” Arab Studies Quarterly . vol. 10, no. 4 (1988), p. 15 See Adam Smith, An Inquiry into the Nature and Causes of the Wealth of Nations _ (London: T . Nelson and Sons, 1884); David Ricardo, “On the Principals of Political Economy and Taxation,” in David Ricardo, The Works and Correspondence o f David Ricardo, vol. 1 _ (Cambridge: The University Press, 1993); and Karl Marx, Capital (London: Lawrence and Wisarr, 1972), vol. 3-
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8 democratic trends, the Rentier State Theory became commonly invoked to help explain
domestic politics throughout the Arab world. So commonplace is the assumption of the
influence of economic rent upon Arab politics that in virtually every discussion of political
development within the region, the politics of oil are noted. In analyzing the political crends
of the Arab Gulf states since the Gulf War, Emile Nakhleh exemplifies the tendency to
attribute this democratic development (or at least increased internal dialog of the topic) to
the ending of the oil-inspired, regional economic boom of the 1970s and early 1980s.16
Likewise, in an international conference on civil society in the Middle East held in 1994,
several scholars noted the importance o f rentier economics in regional political
development.17 The rentier assumption is also not uncommon in even newspaper articles.18
The importance of rent in shaping Arab political structures has become so commonplace in
regional scholarship and popular conceptualizations that few have bothered to question its
validity or assess the strength of the empirical correlation between rent and national political
behavior within the Arab world.
Although many scholars of Middle Eastern politics have recognized the important
role that oil wealth, or lack thereof, plays in the development of democracy within the
region, few have devoted attention to explaining or proving this hypothesis. Of those that
have, virtually none have attempted to do so by applying rigorous quantitative methods of
analysis on the region as a whole. In this way, the Rentier State Theory remains a
commonplace assumption that has yet to be fully documented, dissected, or statistically
substantiated.
16 Emile A. Nakhleh, “The Arab World After the Gulf War, Challenges and Prospects,” in Elise Boulding, ed., Building Peace In the Middle East (Boulder. Colorado: Lynne Rienner Publishers, 1994), p. 116. 17 Civil Society and the Prospects for Political Reform in the Middle East Conference Report of final meeting of New York University’s Civil Society in the Middle East held in September/October 1994 in Queenstown, Maryland, p. 4-5-
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 9
Next Steps: Infusing a Quantitative Dimension
While there are a plethora of internal and external forces influencing political
development within the Arab world, the prominence of external revenue sources to Arab
governments is commonly understood as one of the primary factors leading to both the
regional nature of authoritarianism and the lack of democracy within the Arab world. To be
sure, the context o f any discussion into the forces that shape Arab political change can not be
reduced simply to internal or internal economic determinants, to do so would ignore the
historic realities of the tremendous role of the military (Iraq, Syria, Egypt, Algeria), religious
movements (Saudi Arabia, Iran), or international politics (Lebanon, Jordan, Libya, Kuwait,
LLA.E., Palestine), just to name a few, have played in shaping the domestic politics of the
region. Nevertheless, in the interest of evaluating what has become a largely foregone
conclusion among a significant number of regional scholars and in developing a clearer
perspective on the extent to which petroleum has influenced the development of
participatory patterns of governance within the region, it is important to better document,
dissect and substantiate the Rentier State Theory. This particular study undertakes this
significant task in order to discover the extent to which oil revenues play a role in
determining domestic political patterns within the Arab world and the unique political
economy of the region as a whole.
This study is groundbreaking in that it traces the evolution of the Rentier State
Theory, quantifies its primary components, and applies quantitative methodology as a more
rigorous and systematic means to evaluate the extent to which the theory is valid. Through a
series of multivariate regression analyses done on the primary components of the Rentier
“See “Sand Under KadafI is Shifting,” in Los Angeles Times . September 19, 1996.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 10 State Theory against various quantitative indicators for democracy, litde empirical evidence
was found to support the heretofore largely impressionistic, qualitatively derived hypotheses
and assumptions found in rentier literature. To the contrary, there is every quantitative
indication that many of the assumptions about the relationship between various sources of
state revenue and national political behavior are either not statistically verifiable or actually
contrary to empirical evidence. This is particularly evident when analyzed among a broader
sampling of both Arab and non-Arab countries. In short, this study found little evidence to
support the commonly held assumption that petroleum profits are inversely correlated with
participatory politics in the Arab world.
Prior to presenting the statistical results produced through this study, it is important
to first gain a more concrete understanding of the socio-political implications of rent in the
Arab world. By first tracing of the theoretical foundations of the rentier concept itself, the
context in which it developed, and the core hypotheses around which the theory was
constructed, it is possible to construct a quantifiable framework and statistical measures
which facilitate a more empirically-oriented evaluation of the political effects of particular
sources of state revenue that form the basis of the Rentier State Theory.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 2
THEORETICAL FOUNDATIONS
Historical Context
While the study of the interplay of politics and economics dates back to the dawn of
human civilization, its systematic application within the global context only began during
the last half century under the general nomenclature of ‘development’ studies. This broad
field of social scientific inquiry emerged in the mid-20th century as a direct outgrowth of the
political changes and spirit of optimism chat swept the international landscape following the
Second World War. It was during this period that a vast number of the world’s population
was liberated from either foreign subjection or military authoritarianism, resulting in a
plethora of new or renewed nation-states within the international community. Such
newfound freedom within the developing world was widely considered to afford new
opportunities for both economic and political progress, both of which were considered to be
the core components of the modern nation-building process.
At first these two goals of economic and political development were not only
considered parallel but also mutually reinforcing. By the early 1960s, an entire body of
literature emerged which focused on economic and social modernization as the means to
increased democratization. Guided by what became known as che Modernization Theory,
this scholarship quickly dominated the intellectual landscape, providing a logical explanation
for the persistence of authoritarianism within the lesser-developed, non-Western world.
11
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 12 While not without its critics, the Modernization Theory and its proponents remain very
much at the center of any discussion of the democratic evolutionary process.
The Modernization Theory itself is deeply rooted in economic elements of social
scientific thought and commonly substantiated through an empirical method of analysis. At
its most basic level, this theory postulates that the emergence of democracy within the
developing world is the direct result of widespread advances in socio-economic development.
According to the theory, as first articulated through the writings of Daniel Lerner, Seymour
Lipset, Karl Deutsch and Cyril Black, technological innovation and expanded social services
eventually give rise to a more empowered and responsible citizenry.19 Advances in mass
education and communication provide opportunities for even the average citizen to grapple
with complex political issues and engage in the type of meaningful debate and organization
long considered fundamental to an open, democratic society. Thus, while the process of
modernization may begin with widespread changes in technology and economics, it
eventually spills over into the social realm, triggering a popular outcry for more participatory
forms of governance and an increased capacity for democratic participation.
Modernization theorists, however, did not stop at simple observation or rational
constructs drawn from historical antidotes. Rather, their conclusions are typically buttressed
by rich and rigorous quantitative analysis. From the theory’s earliest articulation, abundant
empirical support was amassed to demonstrate the significant statistical correlation between
various indicators of socio-economic modernization and levels of democracy over varied
19 For the earliest articulation o f the modernization theory, see Daniel Lerner, The Passing ofTraditional Society. (Glencoe: Free Press, 1958); Seymour Lipset, “Some Social Requisites o f Democracy: Economic Development and Political Legitimacy,” in The American Political Science Review . vol. 53 (1959), pp. 69-105 and Political Man: The Social Bases of Politics _ (New York: Doubleday, I960); and Karl Deutsch, “Social Mobilization and Political Development," in The American Political Science Review . vol. 55 (1961), pp. 403- 514; and Cyril E. Black. The Dynamics of Modernization ( New York: Harper &C Row, 1966).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 13 spatial and temporal domains.20 This combination of sound theoretical reasoning and ‘hard’
quantitative analysis gready enhanced the theory’s pragmatic, as well as scholastic appeal,
having even served as the principal rationale for the establishment of foreign aid policies and
programs by wealthier, developed countries. Unlike more conflict-oriented or culturally
imperialistic methods to politically opening societies that typified political thought in
previous centuries, the Modernization Theory provided a clear and seemingly apolitical,
incentive-based means to achieve political ends in the mid-ZO* Century.
However, as the democratic experiment in many fledgling nations began to fid ter by
the late 1960s and 1970s or failed to progress as quickly as anticipated, increased attention
was devoted to articulating the apparent limits of the Modernization Theory. From colonial
America to modern India, real world examples of lopsided economic modernization and
political development were not difficult to detect, and many scholars set forth to explain any
apparent deviation from the expected pattern.21 For most, the relationship between
economic and political development was found to be significandy more complex than
originally articulated. More precise definitions, sophisticated statistical techniques, and a few
added qualifications were employed to account for apparent deviations.22 Not surprisingly, it
20 For a greater sampling of the empirical method of inquiry employed by modernization literature, see Phillips Cutright, “National Policical Development: Measurement and Analysis,” American Sociological R eview . vol. 28 (1963), pp. 253-264; James S.Coleman, “Conclusion" in Gabriel A. Almond and James S. Coleman, eds., The Politics o f Developing Areas _ (Princeton, New Jersey: Princecon University Press, I960); and Bruce M. Russet, Trends in World Politics (New York: Macmillan, 1965). More recent applications of the theory can be found in Kenneth Bollen, “Political Democracy and Timing of Development, “ American Sociological Review, vol. 44 (1979), pp. 572-87; Kenneth Bollen and Robert Jackman, “Political Democracy and the Size Distribution of Income,” American Sociological Review vol. 50 (1985), pp. 438-57; Zehra Arac, “Democracy and Economic Development: Modernization Theory Revicited,” Comparative Politics . vol. 21 (1988), pp. 21- 36; and Axel Hadenius, Democracy and Development . (Cambridge: Cambridge University Press, 1992). 21 For a sampling o f literature re-evaluating the Modernization Theory in particular contexts, see Robert A. Dahl. Polvarchy: Participation and Opposition ( _New Haven: Yale University Press, 1971); Roland Pennock, Democratic Political Theory (Princeton: Princeton University Press, 1979); Edward N. Muller, “Dependent Economic Development, Aid Dependency on the United States, and Democratic Breakdown in the Third World,” International Studies Quarterly . 29 (1985), pp. 445-70. 22 For examples of such reconciliatory literature, see Deane E. Neubauer, “Some Conditions o f Democracy,” The American Political Science Review , 61 (1967), pp.l,002-1,009; Kenneth A. Bollen, “Political Democracy
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 14 was in this context that rentier literature first appeared and gained prominence in helping to
explain the seeming incongruencies of the Modernization Theory as it applied to the region
of the Middle East.
At first glance the Rentier State Theory may represent a regionally specific
contradiction to the Modernization Theory or a form of neo-Orientalism. However, upon
closer analysis it can be seen to reaffirm the fundamental correlation between economic and
political patterns, reconciling any apparent regional contradictions by introducing other key
variables into the equation, those associated with the sources and structure of public finance.
Since patterns of public finance are not generally a cultural phenomenon but rather
contextual in nature, rentier literature could not reasonably be considered neo-Orientalist. A
thorough review o f the literature that gave rise to the Rentier State Theory, its underlying
assumptions, and core hypotheses further underscores these actualities and represents a
fundamental first step in evaluating the theory’s empirical validity, as well as regional
applicability.
While a great number of regional scholars have sighted the influence of oil rent on
various facets of Middle Eastern life and global interaction, few have sought to describe in
detail the exact nature of its effects on Arab national political behavior. Those that have,
employ the concept of economic rent to help explain the skewed nature of Arab domestic
politics. For these scholars, what insulates the Arab world from the global trends affecting
other regions is not persistent economic underdevelopment or cultural incongruencies, as
classical modernists or Orientalists would assume, but rather, the oil wealth widely circulated
within it. This emphasis on oil rent as the key factor in shaping the nature of Arab politics is
and Timing o f Development,” American Sociological Review . vol. 44 (1979), pp. 572-587; Samuel Huntington, “Will More Countries Become Democratic?,” Political Science Quarterly . vol. 99(1984), pp. 193-
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 15 what forms the core commonality of all rentier literature.23 Beyond this fundamental
similarity, rentier literature itself is quite varied in scope and level of analysis as can be seen
through its development in three periods over the past two and a half decades.
Early Rentier Literature: Detecting A Deviation
The earliest scholastic applications of the concept of rent in the Middle Eastern
context were clearly a direct outgrowth of the dramatic changes that had occurred within
several countries of the region by the early 1970s. As oil-rich states began to secure
increasing shares of petroleum profits from the global marketplace, indirectly in the form of
oil royalties from foreign commercial interests and ultimately through direct control of
nationalized industries, they were better able to finance the penetration of their own societies
and the consolidation of government authority within their own territories. More
importandy for this study, this unique rentier character enabled these states to do so largely
without a corresponding dependence on extractive capabilities and relative domestic consent
that has long characterized state-society relations in most other nations. Unlike better
known rentier literature appearing by the mid-1980s, the earliest rentier literature focused
almost exclusively on the phenomenon of rent in pre- and post-revolutionary Iran.
Hossein Mahdavy
In his analysis of the patterns and complexities of economic development in Iran,
Hossein Mahdavy was the first to propose that there is a fundamental difference between
218; and. Tacu Vanhanen, “The Level o f Democratization Related to Socioeconomic Variables in 147 States in 1980-85.” Scandinavian Political Studies . vol. 12 (1987), pp. 95-127. 33 Rentier literature differs with other scholarship that may include reference to rent within an exclusivly economic framework. For a sampling of economically oriented literature, see Mahmoud Abdel-Fadil, “The Pure Oil-Rentier States: Problems and Prospects for Development,” Oil and Arab Cooperation . vol. I 5, no. 3 (1979); Mohsen Massarrac, “The Energy Crisis” and Peter Nore, “Oil and the State: A Study of Nationalization in Oil Industry”, both in Petter Nore and Terisa Turner, eds., Oil and Class Struggle (London:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 16 states financed primarily through external sources o f revenue and those based on domestic
extraction. Subsequendy, he was also the first to characterize the oil exporting countries of
the Persian Gulf (namely Iran and its Arab neighbors) as modern ‘rentier states’ and identify
petroleum revenue as a major factor influencing these countries patterns of both economic
and political development. According to Mahdavy, the 1950s represented a landmark period
in which external economic rents, in the form of oil exports as well as transit fees, began to
dominate government revenues within the entire region.24 This form of external rent was
considered to be of particular importance, he thought, for unlike the agricultural rent
articulated by the classical economists, oil rent was based on the exportation of a non
renewable resource, was not dependent upon productive elements within the domestic
economy, and accrued directly to the state from foreign interests. These three factors
combined to allow oil-rich governments, as Mahdavy observed in the case of Iran, to exist
outside of the traditional confines of their own domestic tax base and to grow without
developing corresponding extractive capabilities common in rest of the world. Mahdavy
concluded:
A government that can expand its services without resorting to heavy taxation acquires an independence from the people seldom found in other countries . . . In political terms, the power of the government to bribe pressure groups or to coerce dissidents may be greater than otherwise. By the same token, this power is highly vulnerable since the stoppage of external rents can seriously damage the government finances.25
Zed Press, 1981); and Cyrus Bina, The Economics of Oil Crisis; Theories o f Oil Crisis. Oil Rent, and Internationalization of Capital in che Oil Industry (London: Merlin Press, 1985). 24 Hossein Mahdavy, “The Patterns and Problems of Economic Development in Rentier States: the Case of Iran,” in M. A. Cook, ed., Studies in the Economic History of the Middle East: From the Rise o f Islam to the Present Day (London: Oxford University Press, 1970) p. 428. 25 Ibid., p. 467.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 17 While the political ramifications of rent were not the primary consideration of his work,
these observations stimulated a subsequent body of literature that sought to explain the
depth and breadth of the rentier influence in regional domestic politics.
Theda Scokpol
As an outgrowth of Mahdavy’s analysis of pre-revolutionary Iran, Theda Scokpol was
the next scholar to apply the concept of rent to the socio-political sphere. In her pivotal
work tracing the nature and causes of the Iranian revolution, Scokpol launched the first
analysis of rent with a sociological, rather than economic bent. While Mahdavy observed the
dramatic shift in the nature of Iranian public finance due to the influx of oil rent that was
occurring by the 1970s, Scokpol described in detail the political ramifications of such a
socio-economic shift that culminated in popular revolt leading to the Iranian Revolution at
the end of the decade.
Skocpol chose to begin her analysis by portraying the Iranian regime under the last of
the Pahlavis as a “rentier absolutist state” in which revenues from the exports of oil and
natural gas enabled the state to develop an independent power-base through government
expenditures or ‘buy offs.’26 With the existence of an exogenic source of revenue, the pre
revolutionary Iranian state was not dependent upon the taxes of its own people but rather
mitigated its relationship to the rest of Iranian society through public expenditures on
everything from military build-ups to consumption subsidies. Unlike the absolute
monarchies of earlier centuries, petroleum revenues insulated the Shah from having to rule
through or in alliance with any independent social classes within Iran.
25 Theda Skocpol, “Render State and Shi’a Islam in the Iranian Revoludon,” Theory and Society ■ vol. 11 (1982), pp. 268-269.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 18 By the mid-1970s, the characteristically absolutist, political economy of Iran began
to falter. As global demand for Iranian oil contracted, the Iranian government was forced to
suspend or cancel many of the public projects and programs through which the state had
effectively purchased popular support in the form of public employment and a vast array of
social services. When the Iranian government was no longer capable of bankrolling the
nation through sustained expenditure, Skocpol maintained, the people turned on the Shah
personally as the “autocratic embodiment of state authority.”27 Such social upheaval was to
a large degree the consequence of decades of increased dependence on externally derived oil
rent and the insulated and excessive power that it permitted.
While these early notations of the political dimensions of rent played an important
part in raising the spectacle of the significance of oil revenue in modern public finance, at
least within Iran, they fell short of providing a more comprehensive picture of the true depth
and scope of the political ramifications of the concept. It was not until the effects of nearly
half a decade of declining oil revenues in the early 1980s were being experienced region-wide
that renewed interest in oil rent emerged. As a result, a host of other social scientists and
regional experts set out to more fully analyze the concept as a factor shaping patterns of
political, as well as economic development, within the region as a whole.
Core Rentier Literature: Constructing A Theory
The first and probably foremost comprehensive articulation of the Rentier State
Theory was through a collection of essays that were a part of a four volume series entided,
‘State, Nation, Integration in the Arab World’ undertaken in the mid-1980s. The second
book in the series, The Rentier State, was specifically devoted to an analysis o f the economic
n Ibid., p. 270.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 19 foundations and impact of the state in Arab countries, with particular attention paid to the
importance of regional oil wealth in shaping the character of state/society relations within the
modern Arab world.28 This book represents the most comprehensive investigation of the
topic of rent within the Middle East and the de facto primer on the Rentier State Theory
iself. As with any compilation of separately authored essays, it is by no means homogeneous
but rather contains a variety interpretations along a common theme. While all contributions
examine some element of the modern Arab rentier state and economy, those of Hazem
Beblawi, Giacomo Luciani, Afsaneh Najmabadi, and Dirk Vandewalle focus more precisely
on the political dimensions of rent from either a theoretical or historical framework. Taken
together, these authors were the first to articulate the connection between the unique rentier
economic pattern and authoritarian governance and, in so doing, to provide a theoretically
plausible economic justification for the region’s seeming resistance to global democratic
trends.
In their introduction to The Rentier State. Hazem Beblawi and Giacomo Luciani,
the principal architects and editors of the work, couched the discussion of the rentier state
within development theory and outlined the core political elements associated with the
concept. For these co-editors, the notion of the rentier state emerged out of a broader
investigation into the “nature of the state” which they define as determined by four key
factors: state size relative to the economy, the sources and structure of state income, the
destination of state expenditure, and the laws and regulations affecting economic life.29
While all four are key to developing a comprehensive picture, it was through an analysis of
28 Hazem Beblawi and Giacomo Luciani, eds., The Rentier State (New York: Croom Helm, 1987). See also the single volume reprenc o f the series, Giacomo Luciani, ed., The Arab State (Berkely: University o f California Press, 1990). which contains a sampling of essays previously published in the series. The chapters written by Beblawi, Luciani and Michael Chatelus from The Rentier State were reprinted in The Arab State . 29 Hazem Beblawi and Giacomo Luciani, “Introduction,” in The Rentier State . p. 4.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 20 the first two elements in particular that directly inspired, the notion of the modern rentier
state and the development of its corresponding theory.
According to Beblawi and Luciani, a rentier state is “any state that derives a
substantial part of its revenue from foreign sources and under the form of rent.”30 The
externality to and supremacy of oil rent within Arab public finance is important, because it
allows the state to function independendy of domestic extraction and therefore, forego the
need for the high degree of internal legitimacy that is typically associated with tax-based
forms of state revenue. So long as the state maintains access to constant or increasing levels
of external revenue, it is able to remain insulated from popular demands for political
participation and accountability. However, if it should fail to obtain the necessary revenue
from abroad to satisfy constant or increasing services, it will be forced to either reduce the
levels of those services and/or increase its dependency on domestically derived revenue to
remain financially solvent. In either scenario, the state will inevitably face increased domestic
political pressure. “This issue,” write Beblawi and Luciani, “is closely connected with the
question of legitimation and to the development of democratic institutions."31 This
assumption that there are unique socio-political effects associated with a dependence on
external economic rent is central to the rentier state concept as a whole.
Hazem Beblawi
In his personal contribution to the book, Hazem Beblawi explaines the pervasive
nature of the rentier economic pattern within the Arab world and the underlying causes for
its unique political effects. For Beblawi, the rentier state is more than just a political
construct but rather also a subsystem of a larger economic system. Utilizing the broader
30 Ibid., p. II. 31 Ibid., p. 19.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 21 term rentier state (of being) to refer to the prominence of oil economies in the Arab region,
Beblawi briefly outlines four key characteristics to understanding the nature of a rentier
economy and its socio-political effects.32 Firsdy, there are no pure rentier economies but
rather situations in which rent predominates in varying degrees. Secondly, a rentier
economy relies primarily on externally derived rent. Echoing Mahdavy’s earlier analysis,
Beblawi considers the externality of rent to be crucial to the rentier concept in that the only
real linkage to the rest of the economy is through the expenditure o f profits from overseas
sales. Thirdly, in a rentier economy, only a few are actually engaged in the generation of
wealth. The vast majority of those participating in the wider economy are only involved in
the distribution and utilization of that wealth. And finally, closely related to the third
characteristic, the government within this broader rentier state is the principal recipient and
distributor of rent. For the purposes of understanding the political ramifications of the
concept, it is this fourth element that is of paramount importance and forms the basis of the
concept of the rentier state as a distinct political entity within a broader rentier economy.
The concept of the rentier state is based upon the assumption that a specific rentier
mentality exists among the great majority, if not entirety, of the Arab population. Instead of
seeking productive roles within society through which to obtain financial rewards for work or
calculated risk, Beblawi notes that rentiers seek, as individuals and in groups, only to gain
access to the rent circuit and the wealth that it provides.33 It is in the prevalence of this
mentality among the general citizenry of the Arab world, both oil-rich and oil-poor, and
their perceived (and real) dependence upon the benevolent role of their governments as the
32 Hazem Beblawi, “The Rentier State in the Arab World,” in The Rentier State . pp. 51-52. 33 Ibid., p. 52.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 22 ultimate protectors and distributors of oil wealth that links the economic concept of rent to
the political sphere.
Beblawi does not stop with an economic analysis, but rather points to the unique
political pattern that such a phenomenon encourages. Here too, Beblawi points to the
centrality of the political state in the broader rentier state. Beblawi writes:
The conventional role of the state as provider of public goods through coercion - mainly taxation - is now blurred in the Arab oil states by its role as a provider of private favours through the ruler’s benevolence. Public goods and private favours have thus gone together in defining the role of the state. With virtually no taxes, citizens are far less demanding in terms of political participation. The history of democracy owes its beginnings, it is well known, to some fiscal association (no taxation without representation).34
While it is clear that the Arab states of the Arabian Peninsula, Libya and Algeria have
substantially more access to oil revenue, Beblawi finds the rentier state evident in the non oil-
rich countries as well. Location rents, inter-Arab aid, worker’s remittances, and global geo
politics all combine to heighten the importance of oil throughout the entire region, both in
terms of producing a secondary rentier mentality among the poorer Arabs and in the way
that the rest of the world relates to them. The internal politics of these countries, in many
ways, are not all that unlike those of their wealthier neighbors in that the rentier
phenomenon, albeit less severe, is nonetheless a primary characteristic of the relationship
between state and society.
Giacomo Luciani
Buttressing Beblawi’s findings regarding the political importance of oil rent, Luciani
proposes a new analytical tool through which to understand the nature of the state and
domestic politics within the Arab world. From the premise that “the nature of the sources of
income of the state influences the basic rules of political life in each individual country,”
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 23 Luciani draws distinction between two general types of states, ‘exoteric’ and ‘esoteric’.35
While exoteric states are based primarily on revenue accrued from abroad, ‘esoteric’ states
finance themselves domestically through state-run enterprises and/or taxation. Alternatively,
being the primary recipient of revenues accrued from the international marketplace, exoteric
states play a principally allocative role with regards to society, distributing goods and services
in return for its monopoly of all significant foreign transactions. Esoteric states, on the other
hand, obtain needed revenue through their own productive capacity or that of their societies
from which they extract revenue. Dependence on productivity requires the esoteric state to
earn its regulatory role over the long-term rather than purchase it as is the case with its
exoteric opposite. All Arab states that depend on income from the rest of the world
(typically considered over 40% of government revenue), either through the sale of natural
resources, receipt of foreign aid, or collection of monopolistic transit fees, are inherendy
exoteric or allocative in nature.36
Luciani is careful to disqualify, however, a few common sources of foreign revenue as
either not truly ‘foreign’ in origin or not direcdy accrued to the state. Both taxes on
international trade and income derived from migrant’s remittances, while commonly
misconstrued as foreign sources of revenue, are not considered sources which contribute to
the rentier state. Taxation on international trade should not be considered ‘foreign’ because
ultimately this cost is born not on foreign importer or exporter but rather on the domestic
producer or consumer. Quite common and significant in several Arab countries, for example
in Jordan, Yemen and Egypt, repatriated revenue through expatriate worker’s remittances are
also not considered a source of state income from the rest of the world, because they are
34 Ibid., pp. 53. 35 Giacomo Luciani, “Allocation vs. Production States: A Theoretical Framework,” in The Rentier State . pp. 63-69.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 24 never truly the property of the state but rather of the migrant who circulates it within the
broader domestic economy. This distinction is particularly important in the case of
migrant’s remittances, Luciani concludes, because while they contribute to a broader rentier
character of the economy, they do not necessarily contribute to the creation of a rentier
state.37 Thus, while Beblawi finds a pervasive rentier economic pattern or mentality within
the entire region, Luciani finds noteworthy distinctions between rentier and semi-rentier
states based upon their relative ability to direcdy monopolize rent.
Like Beblawi, Luciani also points to the direct correlation between governance
patterns and sources of income in allocation (exoteric) and production (esoteric) states.
Since production states depend upon domestic mobilization for the generation of public
revenue taxation, either direcdy or indirecdy through publicly owned industries, they must
maintain a reladvely high degree of credibility and acceptance on the part of their own
populations. According to Luciani, “[t]his establishes a link between the ability to raise taxes
and legitimacy, which is captured in the saying ‘no taxation without representation.”38
Juxtaposed, however, since allocation states generate wealth from abroad and do not need to
extract revenue from their own citizens, the reverse is true. Luciani notes:
Democracy is not a problem for allocation states. Although they may find it expedient to set up some kind of representative body to vent and control some of the resentment that even court politics generates, these bodies inevitably have a very tenuous link to their apparent constituency: their debates are followed with indifference by the public and the ruler can disband them and meet practically no resistance whatsoever. More commonly, representative bodies do not exist at all and to the need of establishing them little more than lip service paid. . . The fact is there is ‘no representation without taxation’ and there are no exceptions to this version of the rule.39
36 Ibid.. p. 70. 37 Ibid. HLuciani, “Allocation vs. Production Sates,” in The Rentier S ate . pp. 73. 39 Ibid., pp. 74-75.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 25 Without significant economic motivation for opposition, allocation or rentier states remain
in effect unchallenged, except when they are widely perceived to have foiled to extract
revenue from abroad. Regardless, Luciani concludes, “[t]he foct remains that even limited
revenue from abroad dramatically improves the state’s ability to buy legitimacy through
allocation and increases regime stability.”40
Oddly enough, the only case in which Luciani finds reason to doubt the collinear
relationship between abundant rent and political absolutism is in the case of pre
revolutionary Iran, about which most prior work on the subject of the rentier state was
concerned. However, he offers little explanation beyond his observation of the Shah’s
preoccupation with “promoting aggressive industrialization, even at the cost of exacerbating
class conflict, [more] than with buying political support.”41 It was left to another politically
oriented contribution to The Rentier State to further explore and explain this seeming
contradiction.
Afsaneh Najmabadi
In the only single-state case study within the book, Afsaneh Najmabadi devotes his
attendon to articulating evidence of the rent-induced causes of the Iranian revolution of
1979. His analysis represents a natural extension of the earlier writings of Mahdavy and
Skocpol, only with an opportunity for greater historical hindsight and in light of more recent
theoretical developments with regards to the rentier state concept itself. Najmabadi finds
that, though it does deviate somewhat from the anticipated rentier pattern as articulated by
Beblawi and Luciani, the Iranian revolution was still very much influenced by the
fundamental change in state/society relations as a result of growing state dependence on oil
"Ib id ., p. 76. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 26 revenues after the mid-1950s.42 Despite typical indicators to the contrary, including huge available oil revenues and a strong, regionally unrivaled military, the Pahlavi dynasty was toppled from within in only a matter of months. Even more perplexing, such rapid upheaval did not produce the split within the Iranian ruling class that would have been expected. Najmabadi attributes this peculiarity not on the commonly-credited repressive nature of the Shah’s regime but rather on the almost total depoliticization that had occurred in post-1953 Iranian society. As the Pahlavi regime gained a greater share of growing profits from a nationalized oil industry, it became less dependent and detached from its own population. By the mid- 1970s, revenues from the sale of oil and gas on the international market almost exclusively insulated the state from reliance on domestic revenues and the tacit social contract that it is assumed to require. According to Najmabadi: This increasing independence from reliance on a tax base made it possible for various economic and social policies to be followed without much regard for social consent. A curious inversion o f the classical formula ‘no taxation without representation’ occurred; the Iranian state felt no compulsion to be representative since it was effectively not taxing the population. The population itself gave up political claims on the state, since it was not being taxed; the Pahlavi state was not ‘its’ state in more ways than one. Rather than the state representing and insuring the interests of certain privileged strata, social classes were adapting themselves to state policies that were beyond their power to influence. When they did come into conflict with state [sic], opposition took the form of rejecting the state in its totality, rather than first pressing it for meaningful reforms.43 When an economic slump occurred in the mid to late 1970s, the state was curtailed in its ability to continue to purchase popular trust and loyalty through increasing expenditure on public projects and social programs. As the Iranian welfare state seemed to falter in meeting its end of the bargain of absolute political control in return for Iranian society’s a share in the 42 Afsaneh Najmabadi, “Depoliticisation of a Rentier State: The Case of Pahlavi Iran,” in The Rentier State . Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 27 oil profits, a sharp re-politicization of the Iranian lower and lower middle classes and mobilization of the historically disempowered traditional elites occurred. The resulting popular uprising was therefore comprised of an unlikely alliance of the lower and lower middle classes who were most immediately negatively affected by the economic downturn, and the disempowered elites (merchants, landlords, and clergy). W ith reconciliation impossible after years of power consolidation under the person of the Shah who was unaware of the extent of societal discontent, these unlikely allies joined forces to oust the Shah as the single embodiment of state authority. The Iranian case represents a unique and seemingly contrary twist to the previously argued relationship between rent and politics. While at first glance the Shah’s regime seemed to failed to exhibit the ability to buy-off political opposition despite access to a generally increasing source of external revenue, Najambadi’s analysis depicts a regime that was so effective in buying political quietess that over a period of over two and a half decades it effectively depoliticized the whole o f Iranian society. However, this depoliticization ultimately raised the stakes of the domestic political game so much so that when there was even the slightest perception of decreasing economic benefits, formerly politically marginalized social groups were dramatically re-politicized and aimed at overthrowing the state in order to protect their narrowly prescribed interests and insure continued access to the domestic Iranian rent circuit. The revolutionary case of Iran actually represents, therefore, the socio-political ramification of extreme rentier state behavior not an alternative pattern as one might presume at first glance. pp. 211-213. * Ibid., pp. 215-216. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 28 Even post-revolutionary Iran, Najambadi argues, represents only a seeming wrinkle in the Rentier State Theory. While economic policy has not been a high priority of the new Islamic regime, there were almost immediate noteworthy steps taken to reduce government dependence on oil revenues and reorient the source of public revenue towards increased taxation. However, instead of signaling a corresponding increase in participatory politics, as would be expecced, legitimacy was gained through religious ideology that cushioned the state from the participatory links of increased taxation. “Since taxes are projected as one’s mandatory religious duty,” Najambadi writes, “their payment no longer entides one to to [sic] a say in the affairs of the state.”44 Najambadi concludes that post-revolutionary Iran, as with the period preceeding the Ianian revolution, represents not only a new twist but powerful confirmation of the rentier state paradigm. Dirk Vandewalle Dirk Vandewalle finds similar confirmation of the rentier state paradigm in the state- building experiences of Algeria and Libya. While the modern Iranian state benefited from a long history of reinforced centralized authority spanning thousands of years, the identity of nation, let alone state, in Algeria and Libya is a relatively modern phenomena. As was the case for a great number of Arab countries, the consolidation of absolute state authority in both these two North African countries was not financed through years of incremental domestic extraction but rather with revenues gained through windfall petroleum profits in the 1970s. Such financial autonomy allowed these states to develop independently of and eventually dominate their own societies. This lack of “convergence between state and society,” as Vandewalle refers to it, allowed the governments of both Algeria and Libya to evolve is somewhat of a political vacuum, being subject primarily to the biases of a very small 44 Ibid., pp. 225-226. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 29 circle of elites.45 While the evolution of the state in both of these countries was greatly affected by the relative absence of capital scarcity due to the relative abundance of rentier income, they nevertheless developed with differing characters due to the particularistic differences in their leadership and historical experiences. The socialist tendencies and industrialization policies of the post-colonial Algerian leadership of Houari Boumedienne provided the ideal backdrop for the eventual take-over of the Algerian state by increasingly powerful administrative elites in the 1970s. As the state apparatus expanded to regulate and administer the growing revenues collected from the international oil market and the public industries that these revenues helped to finance, the Algerian state increasingly relied on the expertise of technocratic and bureaucratic professionals. When Boumedienne died in office in 1978, a power struggle emerged between these new administrative elites and the traditional party regulars, unions, and students that opposed them. “The conflict was resolved,” Vandewalle notes, “in favour of the bureaucratic/technocratic class. . . [and] the traditional sources of power - personal, class or religious - [were] replaced by organizational ones. . which were no doubt butressed by vast sums of hydrocarbon revenues realized in the 1970s and early 1980s.46 By the mid- 1980s, however, these new elites found it increasingly difficult to sustain the large state apparatus on decreasing oil revenues and were forced to undertake progressive liberalization o f both the economic and political sectors simultaneously. In sharp contrast, however, the Libyan state entered the 1980s without the entrenchment or even cultivation o f technocratic and bureaucratic elites. The Libyan Free Officer’s revolution of 1969 and its ideological aftermath effectively destroyed the power of 45 Dirk Vandewalle, “Political Aspects of State Building in Rentier Economies: Algeria and Libya Compared,” in The Rentier State . p. ISO. 46 Ibid., p. 163-164. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 30 the administrative elites that had emerged in the early oil era as rivals to the ideological movement. Instead of embracing these elites as a means toward entrenching political power and fostering coherent economic strategies, as did Boumedienne, Qadhafi remained deeply distrustful of bureaucratic procedures and the potential political rival represented by the growing technocratic class.47 He pursued a populist political ideology that reinforced and heightened the rentier character of the Libyan state and economy. Personalization of power and the creation of a consumption society enabled the Libyan state to retain its autonomy and authority without a large bureaucracy. Despite decreasing revenues from abroad, Libya’s small population, abundant oil reserves, and lack of an organized opposition allowed the Libyan state to escape the political pressures faced by its Algerian neighbor in the late 1980s. Vandewalle’s analysis demonstrates two important elements of domestic politics in Arab rentier states: the tendency toward disconnected state/society relations due to the availability of external public finance resources and the diversity of state structures that exist within Arab rentier states due to other factors. The latter is import in understanding the variety of governance structures found throughout the region, even among similar financially-situated countries. Other Contributors While not specifically addressing the topic of the domestic political impact of rent in Arab countries, the remainder of essays found in The Rentier State provide a broader, as well as more in-depth understanding of the characteristics of rentier economies and states which prove important background for any further analysis. These contributions can be divided into two broad categories, those analyzing the influence o f rent in the structure o f Arab "Ib id ., p. 167. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 31 public finance and those covering the broad economic implications of rent for state policies within the region. Leading the discussion of the structure of Arab public finance, Thomas Stauffer provides a detailed analysis of the various and peculiar nature of Arab state revenue sources. According to Stauffer, “many - if not most - of the economies of the Middle East are characterized by an unusual degree of dependence upon non-renewable assets.”48 These non-renewable assets, including extractive rents from oil and phosphates, unrequited transfers o f foreign aid, quasi-rents in the form of worker’s remittances, locational rents from pipelines and canals, and portfolio income from reinvested rents overseas, all combine to bring a unique character to both Arab economies and the governments that regulate them, both oil-rich and oil-poor. Stauffer insists that some qualifications to these external sources of revenue, particularly in accounting for a ‘depletion charge’ on non-renewable resources, reorienting the calculation of remittances, and making use of a rentier multiplier, are necessary to gain a true perspective on the national accounts of Arab countries when drawing comparisons with other non Arab countries.45 While the net affect of such adjustments would vary from councry to country, they all work to significandy increase income disparity typically noted within the region and, in the case o f the powerful rentier multiplier, reflect the overwhelming dependence of all Arab states on rent from abroad, particularly regional petroleum revenues. Joining Stauffer in an analysis of public finance patterns within the Arab world, Fathallah Oulalou and Larbi Jaidi focus on the nature and character of tax structures within North Africa. The fiscal system of these countries, Oulalou and Jaidi maintain, have a Thomas Stauffer, “Income Measurement in Arab States,” in The Rentier State . p. 22. 49 Ibid., p. 23. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 32 common origin in the colonial period and remain largely unaltered to this day. Their cross national analysis finds that the structure of tax revenue within this sub-region in general is heavily weighted toward indirect taxes that they conclude to be the result o f a “lack of domestic integration and . . . extroverted orientation of Maghrebi countries.”50. Oualalou and Jaidi find that the majority of public revenue is financed, not through the extractive means employed by most, but rather by tapping exogenous sources, such as the international oil and lending markets, which only serves to further foreign dependence and lessen the states’ dependence on their own citizens. Future economic growth and well being within the Maghrab will only be achieved and sustained through greater reliance on the domestic economy. While Oualalou and Jaidi examine the fiscal state of affairs in the western-most portion of the region, Hesham Garaibeh rounds out the survey by analyzing the public revenues of Jordan, Egypt and Kuwait. In general, Garaibeh finds a noteworthy distinction in the structure of government finance between oil-rich and oil-poor Arab countries. According to Garaibeh, . . . oil revenues induced oil exporting countries not to impose any significant taxes and remain with an underdeveloped domestic taxation system, while the rest of the Arab countries were obliged to impose different types of taxes to partly finance their development programmes . . . [which] in turn, led to the establishment of a relatively modern taxation system.51 However, as the global market for oil began to falter, even the oil exporting countries began to examine domestic taxadon as a means of off-setdng declining external revenues. Despite this obvious difference in the need for reliance on internal sources of revenue, Jordan, Egypt and Kuwait’s domestic revenue structures, like those of their North African neighbors, are 50 Fathallah Oualalou and Larbi Jaidi, “Fiscal Resources and Budget Financing in the Countries of the Maghreb (Algeria, Morocco, Tunisia, Libya and Mauritania),” in The Rentier State . p. 172. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 33 primarily comprised, of indirect taxes, such as import/export duties and stamp taxes. Garaibeh finds, however, that significant variation does exist in these state’s relative preference of nontax revenues over direct taxes as supplemental sources of revenue.52 Nevertheless, domestic sources of revenue for many, if not most, countries within the region remain secondary to external sources in public finance, namely oil, remittances, aid or deficit financing through borrowing. While the writings of Stauffer, Oulalou, Jaidi, and Garaibeh focus on the nature and structure of public finance in the Arab world, those of Mahmoud Abdel-Fadil, Michel Chatelus, and Hamid Ait Amara emphasize the unique role that Arab governments play in their own societies. According to Abdel-Fadil, in the Arab countries with significant petroleum exports, “the state becomes the main intermediary between the oil sector and the rest of the economy” through the allocation of petroleum profits.53 Lavish, oil-financed public projects and programs, including the construction of highways, massive irrigation schemes, and widespread social services and subsidies, not only form the basis of the domestic economies of these countries but also stimulate the dependence of their own citizens upon the political elite who have control over these external revenues. The unique nature of rent-financed, government income also has a number of secondary effects. For the oil-exporting countries themselves, ‘external recycling’ o f oil rent in the form of overseas portfolio investment, Abdel-Fadil adds, further reinforces the rentier 51 Hesham Garaibeh, “Government Income Sources and the Development of the Taxation System - the Case of Jordan, Egypt and Kuwait," in The Rentier State ■ p. 194. 52 Ibid., pp. 196-209. Garaibeh notes that after indirect taxes, non-tax revenues are second-most important to the Jordanian state, while in Egypt, it is direct taxes. Domestic revenues play a little role in the finance o f the Kuwaiti government and direct taxes are non-existent, at least for the period under investigation. 53 Mahmoud Abdel-Fadil, “The Macro-behaviour of Oil-rentier States in the Arab Region,” in The Rentier State, p. 83. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 34 pattern and paramount role of the state within society.54 In non-oil rich countries too, oil rent induces rentier patterns through expatriate worker’s remittances and regional foreign aid transfers, though only the latter has the same government role-enhancing effect. In his analysis o f the state attitudes and resulting policies that shape the industrial and service sectors of Arab societies, Michel Chatelus finds surprisingly litde variation between states and citizens within the region with regards to rentier behavior. According to Chatelus: Observing the political economy of state industrialisation in the Arab Middle East we suggested a generalisation of the rentier state hypothesis to the great majority of Arab states. Induced allocation effects in non-oil states are just as determinant as direct allocation in oil countries. . . For non-oil countries - even when a genuine productive sector exists - a growing part of economic activity is linked to oil money produced through labour and capital movements.55 While early rentier scholars found immediate evidence to support the Rentier State Theory among oil exporting countries, Chatelus joins several of his contemporaries, notably Beblawi and Luciani, in noting parallel behavior even among the non principal, oil exporting countries or semi-rentier states within the region. Having access to petroleum profits, even if only secondarily through the flow of expatriate remittances and inter-Arab aid, finances a more or less dominant role for Arab states within their own societies. “In this [rent-rich regional] context,” Chatelus writes, “the role of the state is paramount: it is the unavoidable instrument of resource allocation whether in liberal or socialist regimes.”56 This role further buttresses these states ability to more comprehensively involve themselves and govern the affairs of their citizens. For Arab citizens, Chatelus writes, “getting access to the rent circuit 54 Ibid., p. 86. 55 Michel Chatelus, “Policies for Development: Attitudes Toward Industry and Services,” in The Rentier State . p. 110. * Ibid., p. 111. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 35 is a greater preoccupation than reaching productive efficiency.”57 Not only are regional industrial efforts, both private and public, stifled by the very mindset of society in general which is often more interested in quicker means of wealth accumulation, but so to are any tendencies towards meaningful political opposition. The final contribution to The Rentier State to also emphasize the trans-regional applicability of the Rentier State Theory and the centrality of the state in Arab society is that of Hamid Ait Amara. His analysis of the agricultural sector within the Arab world compliments Chatelus’ industrial and service sector orientation and serves to underscore many of the same themes. According to Ait Amara, the 1970s witnessed a restructuring of economic activity in favor of hydrocarbon, manufacturing, and service sectors over agriculture.58 As a result, a great number of Arab countries in which cultivation has been the traditionally dominant economic activity, experienced a decline in agricultural productivity and an increase in reliance on foodstuff imports which necessitated the introduction of wide spread subsidies and price stabilization programs. For others, where import dependence and subsidization were historically commonplace, capital surpluses from increasing oil revenues enabled the undertaking of elaborate schemes to expand their limited agricultural production. Nevertheless, in both instances, the advent of rentier economic patterns had dramatic trans-regional effects that further enhanced the role of the state in Arab societies. By the mid-1980s, however, many of the marginal rentier states within the region, including Morocco, Tunisia, Egypt and Jordan, could no longer afford to continue to counter the effects of a declining agricultural sector. The resulting civil unrest with regards to rising prices that manifest itself particularly in the poorer Arab countries has significandy restricted 57 Ibid. 58 Hamid Ait Amara, “The State, Social Classes and Agricultural Policies in the Arab world," in The Rentier State, p. 140. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 36 the economic policy options available to the state, both worsening the fiscal viability of these governments and contributing to the growing instability in state/society relations throughout the region. Together these three sections of The Rentier State paint a comprehensive picture of the far-reaching impact that external rent, primarily hydrocarbon profits, have within the Arab world during the last few decades. From oil-rich to oil-poor, though in varying degrees, the entire region has been affected by the ready availability of vast capital surplus revenues throughout the 1970s and early 1980s. Such exogenous revenue created a trans- regional economic structure in which symbiotic relationships redefined roles and responsibilities of both state and society. Petroleum exporting nations received expatriate labor, access to transportation routes, and general regional stability from their Arab neighbors in return for a share in the profits. Arab states, as the necessary facilitators o f these abundant revenues, traded a vast network o f social services, subsidies, and public projects in return for the complete and unwavering support of their own citizens. Without the fiscal necessity of directly taxing its own citizens to raise public revenues, Arab states managed to escape the “no taxation without representation” axiom and developed into largely autocratic regimes that remained insulated from popular pressures of direct political accountability or representation. Likewise, Arab citizens traded their political influence over this expanded state for access to public welfare at the minimum and, perhaps even, a share in the vast economic profits themselves. Democratic progress in the Arab world was not stifled by a lack of economic progress, as the modernization theory hypothesizes, but rather a unique type of dramatic economic progress known as rentierism. The Rentier State represented the first comprehensive attempt to theoretically identify and articulate rent as a powerful economic force capable of influencing patterns of political development. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 37 Giacomo Luciani (1988) Besides The Rentier State, another piece o f scholarship produced by Giacomo Luciani was introduced during this period that reiterates many of the central themes to Rentier State Theory. In a 1988 article appearing in the Arab Studies Quarterly. Luciani provides further discussion of the underlying assumptions of the theory itself. As previously indicated, this theory is based upon the assumption that public finance structures both reflect and shape domestic political patterns. Luciani writes: Control of public finance is essential to the legitimacy of government. Practically all policies are implemented through taxation, the allocation of fiscal resources and expenditures, or both. While public finance may not be the only factor legitimizing a specific government, it is an essential component of the combination. To tax and to allocate fiscal revenue is the very essence of government.59 In this sense, Luciani contends that the way in which a government finances itself is just as important as the allocation of revenues in shaping its relationship with society. The typical pattern of the expansion of government (penetration) through increased domestic taxation (extraction) necessitates concessions of power, or at least accountability and voice, in the allocation of this public revenue. “In general,” Luciani notes, democratic governments are in a better position to raise taxes because of the principal of “no taxation without representation. [Conversely] . . . no representation is offered whenever the state is substantially independent of domestic taxation.60 With most if not all of Arab states being primarily financed through external sources, be them hydrocarbon profits, foreign aid, transit fees or indirectly, through expatriate remittances, Luciani concludes that it is not surprising to find an abundance of authoritarian regimes, even among the wealthiest countries within the region. 59 Giacomo Luciani, “Economic Foundations of Democracy and Authoritarianism: The Arab World in Comparative Perspective,” Arab Studies Quarterly . vol. 10, no. 4 (1988), pp. 460. “ Ibid., p .463- Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 38 Having underscored the importance of taxation in general in fostering democratic patterns of governance, Luciani is careful to differentiate between direct and indirect taxes. “From a political point of view,” Luciani contends, the distinction must be drawn between expenditure that is financed by revenue obtained from taxes that require compliance, and expenditure that is financed by non-tax revenue, taxes on international trade and excises, or deficit spending. The former supports a demand for democracy, the latter commonly does not.61 Direct taxes, or those taxes that require overt popular compliance, involve the surrender of private funds to the state in order to advance the perceived public good. This form of revenue, which represents the majority of public revenues in most democratic societies, is thought to require a high degree of regime legitimacy to be successfully imposed. Indirect and non-tax revenues, however, are levied as the government’s share for enabling a fair commercial exchange in which immediate benefits are received by willing parties who take part in the transaction. Unlike direct taxation, these means of obtaining revenue are relatively easy to collect and therefore, do not require the same degree of sustained popular support. Not only taxation in general but the type of tax itself plays a key role in determining the degree of state autonomy from popular pressure for a greater voice in governance. Evaluating a period immediately following nearly a decade of generally high, yet fluctuating petroleum profits, when oil prices had only just begun to plateau and then decline, core rentier literature was dedicated almost exclusively to laying the theoretical framework of the Rentier State Theory. This new theory, though the concept of rent itself not new, provided justification for the asymmetrical relationship between economic and political progress in the Arab world that represented a qualification to the dominant Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 39 Modernization Theory that dominated the discussion of political development. It remained for subsequent literature, however, to evaluate the validity of this concept by analyzing the development patterns of particular countries and sub-regions over a period of increasingly scarce petroleum profits. Post-Rentier Literature: Analyzing the Pattern Unlike the rentier literature that appeared in the late 1980s which articulated the general patterns, theoretical peculiarities, and unique political consequences of rent in the Arab world, that of the early 1990s was largely dedicated to qualitatively evaluating or further analyzing the affect of rent fluctuations on select countries and sub-regions. Having been made fully aware of the unique political economy of the Arab world and presented with a reasonable explanation for regional exceptionalism to the dominant assumptions of the relationship between wealth and democracy, post-rentier scholarship in the early 1990s sought to reinterpret the Rentier State Theory in light of sustained periods of decreasing oil revenues and what appeared to be widespread, though often subde, moves toward more participatory forms of political structures on the part of many, if not most, states in the Arab world. Jill Crystal In her analysis of political patterns in the oil-rich countries of Kuwait and Qatar, Jill Crystal examines the impact of oil on the formation and destruction of political coalitions and state institutions. This pioneering study was the first to trace how traditional rulers in the oil era utilized oil revenues for the penetration, consolidation, patronage, and co-option o f their own societies to foster greater dominance and stability. Crystal traces the 61 Ibid., p. 466. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 40 relationship between the tribal sheiks and merchant classes, which she argues underscores the validity of the rentier state paradigm in regional political inquiry. The political economy of Kuwait and Qatar were historically balanced through a symbiotic relationship between ruling and trading families in which neither were able to totally dominate. In the pre-oil era, the merchant class checked the traditional political power of the tribal sheiks through their control of the country’s economic resources. As the state’s primary means to extract revenue, the ruling families depended on the economic successes and voluntary cooperation of the leading merchant families. W ith the blossoming of the petroleum industry, however, ruling elites were freed from their dependence on domestic sources of revenue almost entirely. “Where once these revenues had to be squeezed from the population, through the merchants, who in turn extracted a political price,” Crystal notes, “the rulers now received revenues independently.”62 The result of the state’s growing dependence on external revenues was a fundamental restructuring of political life in which a tacit trade of wealth for power consolidated absolute political power under the leadership of this new rentier state. “The merchants’ withdrawal from public politics,” Crystal concludes, suggests that participation demands are tied to extraction o f taxes and to the ability of those who mediate that extraction to influence the distribution of extracted wealth. Since in the rentier oil economies extraction of wealth from the population, by the state, does not occur, neither does the demand for political participation.63 With this observation, strong evidence for an inverse relationship between rentier-derived wealth and democracy as postulated by the Rentier State Theory was at least qualitatively affirmed. Increased authoritarianism was the price that Kuwaiti and Qatari people, as largely 62 Jill Crystal, Oil and Politics in the Gulf: Rulers and Merchants in Kuwaic and Qatar _ (Cambridge, Great Britain: Cambridge University Press, 1990), p. 6. 63 Ibid., p. 10 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 41 mitigated through the merchant families, were forced to pay for a share in the profits that their governments secured from abroad. It was not until nearly a decade of declining oil profits began to financially cripple both governments that the merchant class again began to assert itself in the political arena. According to Crystal, Oil . . . initially had a depoliticizing effect, removing previously active groups, in particular the merchants, from politics. But as merchant dissatisfaction grew with the fall in state spending in the 1980s, the merchants began to return to political life, in alliance with other opposition groups. 64 Crystal’s observation of the re-politicization of the merchants in Kuwait and Qatar following periods of scarce external rents is remarkably similar to the phenomenon detected by Najmabadi among the traditional elites (which included merchants) in pre-revolutionary Iran. However, it was left to subsequent scholars to expand the scope of the Rentier State Theory’ applicability throughout the remainder of the region and evaluate in greater detail the political effects o f declining oil revenues. Rex Brynen Since the writing and compilation of The Rentier State, it had been widely assumed that if the Rentier State Theory was correct, than a decline in petroleum profits over the long-run should stimulate societal pressures for participatory politics. While Najmabadi and Crystal’s analysis of the oil-exporting countries of Iran, Kuwait and Qatar seemed to evidence promising signs in this direction, little direct evidence had been presented to substantiate this logical corollary. In what could be considered the second single-most pivotal piece of rentier scholarship, Rex Brynen not only provides qualitative support for this reversed or post- Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 42 rentier assumption, but also strong evidence of the trans-regional applicability and qualitative viability of the Rentier State Theory. Unlike previous rentier scholarship, particularly those involving detailed case studies, Brynen uses the unlikely subject of the oil-poor, semi-rentier state of Jordan to highlight his observations about the political effects of rent within the Arab world. According to Brynen: The relatively litde examination of the political economy of rentierism to date has largely been confined to the hyper-rich tribal monarchical autocracies of the Gulf. Little attention has been paid to the political dynamics of semi-rentier states, and virtually none to the prospects for transition from neo-patrimonial authoritarianism associated with rentier politics to more liberal aryl participatory patterns of government. Yet in many ways, the semi-rentiers provide a critical test of the model given their generally weak financial reserves and consequent vulnerability to fluctuations in regional economic conditions.65 Since it had already been argued that even the oil-poor, or semi-rentier, states within the region manifest many of the same rentier political patterns as their oil-rich neighbors, Brynen considered Jordan an ideal subject for analysis. While it could take decades for prolonged periods of suppressed petroleum profits to alter the state/society relations in the oil-rich countries, the oil-poor do not have the financial reserves, overseas investments, or favorable borrowing status to weather even the slightest of economic storms. The Jordanian state and domestic politics, about which Brynen writes, were significantly altered during the 1970s and 1980s. Like many of its rentier neighbors, the Jordanian government did not feel the need to develop substantial extractive mechanisms for raising public revenue from its own people. Rather, it depended on its good relational standing and geo-political situation to secure large sums of revenue from wealthier Arab states, either direcdy through foreign aid or indirecdy through the re-circulated expatriate 64 Ibid., p. 185. 65 Rex Brynen, “Economic Crisis and Post-Rentier Democratization in the Arab World: The Case of Jordan,” Canadian Toumal o f Political Science . vol. 15, no. 1 (March 1992), p. 75. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 43 income earned From neighboring countries’ booming petroleum industries and swelling downstream service sectors. As predicted by the Rentier State Theory, politics during this period likewise witnessed a similar political disconnect between state and society. The Jordanian state remained insulated from both the need to rely on the domestic taxation as the primary source of public funds and the political accountably long associated with it. While not a rentier state in the classical sense, the Jordanian state manifest many of the same core economic and political characteristics of its wealthier Arab neighbors. However, by the mid 1980s Jordan’s economic, as well as political, situation had begun to change. As the global market for petroleum waned and petrodollars became harder to secure, the Jordanian state was forced to drastically reduce expenditures and increase levels of domestic taxation. Utilizing annual financial statistics from sources such as the IMF, World Bank, and the Jordanian government itself, Brynen highlights a steady decline in grants and remittances and corresponding increase in domestic taxation throughout the 1980s.66 Conversely, relying on more qualitative documentation, he notes a corresponding increase in both popular pressure for and government acquiescence in the form of democratic reforms during the same period.67 Brynen argues that faced with prolonged economic hardship and inability to continue to finance budget shortfalls by taking on more international debt, the Jordanian state had litde alternative but to forge “a new social contract, an implicit - and sometimes explicit - quid pro quo whereby a real democratic opening [was] offered for acceptance of continued economic austerity.”68 The Jordanian political system, while still rarely classified as a democracy, evidences some o f the most widespread political freedoms within the Arab world, including an active parliament, a 66 Ibid., pp. 84-92. 67 Ibid., pp. 87-94. 68 Brynen, “Economic Crisis and Post-Rentier Democratization in the Arab World,” p. 92. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 44 plethora of political parties, a relatively free press, among others. W ith this observation and more systematic quantitative documentation, Brynen further substantiated the Rentier State Theory as a viable explanation for both the historic lack of democracy within even a semi- rentier state and the seeming political reversal of the late 1980s and early 1990s. Rayed Khalid Krimly While rentier case studies by 1992 had grown to included a broad range of Arab countries with varying access to external rent, conspicuously absent from this number was an analysis of the leader of the petroleum club, Saudi Arabia. As the largest and probably most influential Gulf state with the largest know oil reserves, it is not surprising that Saudi Arabia is also considered to maintain one of the most authoritarian political systems in the Arab world. However, it too not only manifests the powerful political effects o f rent, but also, as Rayed Krimly highlights, the reverse effects of declining rent. As the culmination of his doctoral studies in political economics, Krimly chose to analyze the economic potential of his home country of Saudi Arabia in light of the increasingly popular rentier state paradigm. Besides providing overwhelming evidence confirming the strong render character of the modern Saudi state, Krimly’s analysis is particularly noteworthy in that its underlying theory offers a challenge to the previous assumption that access to abundant external rent allows for growing state autonomy. Up to this point, rentier literature tended to depict rentier states, especially those of the Gulf, as largely insulated from their own societies and capable of a great deal of political autonomy. From a perspective of maximizing economic growth, Krimly finds that the rentier state structure is actually quite limited by its own internal political constraints. “[S]ince oil is a depletable nonrenewable resource,” Krimly maintains, Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 45 . . . the goal of achieving real and sustained economic development in rentier states takes an added significance. Instead of utilizing current oil revenues to develop alternative sources of income for the future, the oil state spends most of its revenues on imports and consumption to both buy political passivity from its subjects and enrich the few members that dominate its social structure. 69 During the 1970s and early 1980s, when petroleum profits were at a historic high, the Saudi Kingdom spent enormous sums o f money on lavish public works projects, consumption subsidies, and social welfare programs aimed at legitimizing the state and its autocratic governance in the eyes of its own people. When profits slumped throughout the latter half of the 1980s, the Saudi state was forced to sell off foreign assets and deplete any assured capital for future investment in order to maintain government spending and continue to purchase its legitimacy. Krimly argues that this actual interpenetration of state and society is reflected in the Saudi states’ inability to make and undertake wise long-term economic decisions due to the short term political costs inherent in doing so.70 This complex dilemma faced by all rentier states, which inevitably threatens the long-term viability of the society as a whole, is a sharp reminder of the fragility, yet importance of balancing state-society relations within the Arab world. F. Gregory Gause, III A similar challenge to some prior assumptions about rentier politics is offered by Middle Eastern scholar G. Gregory Gause, III in his analysis of che ‘oil monarchies’ of the Gulf. His observations of the domestic and security challenges of the Arab Gulf States echo many earlier themes about the preeminence of rent in Arab Gulf economies and unique patterns of governance that they promote, particularly in che 1970s and 1980s. However, 65 Rayed Khalid Krimly, The Political Economy of Rentier States: A Case Study o f Saudi Arabia in the Oil Era. 19*50-1990. Unpublished dissertation submitted to the Columbian College and Graduate School of Arts and Sciences o f The George Washington University, February 14, 1993, p. 4. 70 Ibid., pp. 398-399. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 46 breaking somewhat from the expected post-rentier analyses of Crystal, Brynen and, to a lesser extend, Krimly, Gause provides an alternative explanation for the increased demands for political participation felt throughout the Arab world in the 1990s. While Gause finds ample evidence to support the Rentier State Theory with regards to the unique economic causes of the strengthening of authoritarianism within the Axab world during the 1970s and 1980s, he attributes raising expectations not declining rents to be the primary cause o f signs of reversal in the 1990s. According to Gause: As the role of the state in these countries has grown, it has begun to call forth new demands for representative institutions and responsible government from society. Those demands spring from the very processes of state growth and expansion occasioned by the oil boom. The recent upsurge in political activity in the G ulf monarchies is not only consistent with the realities of the rentier state and its relationship to society, but is in fact generated by those realities.71 When rentier states took on a greater responsibility in not only promoting, but actually maintaining, the general welfare of their citizens, they permanently raised Arab societies’ expectations of the state. “Much as taxpayers want responsible government to spend their money,” Gause adds, “the beneficiaries of rentier states want responsible governments to sign their checks.”72 The sustained slump in oil prices and continued increases in population throughout the 1990s only served to narrow the political options of rentier states and expand the number of increasingly sensitized stakeholders. The political prospects of the entire region in the future, therefore, are of either increasing crises and conflict, gradual economic and political liberalization, or most likely for the majority of Axab states, some combination of the two. 71 F. Gregory Gause, III, Oil Monarchies: Domestic and Security Challenges in the Arab G ulf States New York, Council on Foreign Relations Press, 1994, p. 81. 72 Ibid. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 47 Contemporary Discussion: Reevaluating Theoretical Validity The latest scholarship concerning the rentier state paradigm has emerged as part of a larger discussion evaluating the complexities of emerging democratization in che Arab world. In a very real sense, the dramatic and sustained decrease in petroleum profits throughout the early 1990s did not result in the widespread and concrete political change as many rentier scholars predicted. Instead the political development of the region was largely mixed. While democratic pressures and rumblings made themselves manifest in many countries throughout the region, as previously discussed, noteworthy reversals also took place in at least one Arab country and several others experienced little if any tangible democratic growth More expectant scholars, like Giacomo Luciani, felt compelled to explain the potential reasons for the limited and slow political progress despite theoretical predictions to the contrary. Giacomo Luciani (1995) In his latest published work on the importance of rent within the Arab world, Luciani does not abandon the rentier state paradigm but rather chooses to qualify and temper its overly deterministic undertones. While the rentier state paradigm has become “widely accepted,” Luciani notes, . . . the dynamic relationship between rent and political order is far from being automatic. In other words, the idea that changes in rent availability are necessarily linked to inverse changes in accountability, representation, and participation (“democratization”) cannot be proven in general, and is in fact not supported by experience. A relationship does exist, but it is one that necessitates more complex historical and circumstantial analysis to be properly understood. 73 (emphasis is that of the author) 73 Giacomo Luciani, “Resources, Revenues, and Authoritarianism in the Arab World: Beyond the Rentier State,” in Rex Btynen et ai., Political Liberalization and Democratization in the Arab World pp. 211-212. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 48 This is a significant admission of limited explanatory value for one of the chief architects of the Rentier State Theory and somewhat of a strong indictment on subsequent rentier scholarship that has attempted to “prove” the concept’s theoretical validity. The reasons for a variety o f post-rentier experiences within the region, Luciani maintains, is due to both the differing degrees and nature of rentierism among Arab states and the range of economic resources available to them.74 While classic rentier states may still possess the economic reserves to stave-off the need to cultivate more accountable sources of income which would necessitate a renegotiation of the tacit political pacts between ruler and ruled, the options available to semi-rentier states are significantly narrower. Even among these states, however, varied political responses are inevitable. “As existing regimes cannot credibly undertake economic reforms,” Luciani writes, “a way out will have to be provided either by a turn toward democratization or by the advent of a different kind of authoritarian regime.”75 Perhaps this would explain the relatively opposite experiences of Algeria and Jordan to declining availability to external rent. Despite this apparent failure of the theory to categorically predict the political results of declining external rent, Luciani finds democratic transition the only means of maintaining a fiscally viable Arab state over the long-run, regardless of short-term political circumstance and evidence to the contrary. Conclusion What then can be said about the development of rentier literature over the last two and a half decades? As can be seen through this “brief’ yet comprehensive survey, the Rentier State Theory emerged gradually as the result of a handful of regional and political economy-oriented scholars attempt to explain the Arab world’s persistent authoritarianism 7< Ibid., pp. 212-214. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 49 and seeming imperviousness to global democratic trends. Like no other region, democracy in the Arab world over the last century has been historically noteworthy only in its absence, and rentier scholars sought to develop an economically-oriented paradigm to understand the nature of Axab political practices and institutions. While rentier scholarship is relatively diverse in both scope and depth, this cross- disciplinary body of literature unilaterally posits a fundamental relationship between the nature o f public finance and the domestic politics necessary to maintain that nature. In the case of a number of Arab states, the exportation of petroleum (and to a lesser degree other hydrocarbons) allows for an external means of subsidizing the revenue that must be raised domestically by most other governments. The availability of this external revenue, or rent, frees the state from the need to extract taxes from its own citizens and, in most cases, enables some level of the reverse flow of funds from state to citizen. Therefore, for the rentier state, the political principal of “no taxation without representation” is reversed to being one of “no representation without taxation”; and domestic politics become notably authoritarian. This is the essence of the Rentier State Theory or paradigm. However dear, this theory is not without a considerable number of complexities. The simplicity of the Rentier State Theory is further complicated by the nature of its fundamental component, the availability of external rent. While the sale of hydrocarbons is the primary means to collecting external rent from the rest of the world, a number of downstream “rents” also contribute to the making of rentierism and a rentier state. Locational rents, Arab aid, and expatriate remittances are also forms of external rents, though the latter differs considerably in that it does not accrue directly to the state. These 75 Ibid., p.225. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 50 downstream or secondary rents join with the petroleum industry to create the regional nature of the rentier state phenomenon which includes both rentier and semi-rentier states. Over the course of two and a half decades, both the economics and politics of the Arab world have been transformed. Prior to the oil boom of the early 1970s, most Ajrab states were either weak from lack of revenue or from the failure of over-aggressive socialist economic programs. With the exponential influx of external revenue that accompanied the nationalization of the oil industry within the Arab world, the growth of production capabilities, and the dramatic increase in the price of oil, the Arab state was able to consolidate its control and influence over its own territory and society. Those states with the most access to external rents and the smallest populations became the wealthiest and established elaborate welfare schemes that effectively purchased political power from their own populations. Citizenship became the means to securing elaborate social services, and for many, enormous personal fortunes. States throughout the Arab world became largely autonomous from their own populations, bound only by their inability to control the revenue they obtained from abroad and desire to perpetuate their relatively absolute power. When the global market for petroleum subsided in the early 1980s and subsequently protracted into the 1990s, Arab states were faced with real economic and potential political crisis. W ith decreasing external rents, most Arab states sought a variety of means to maintain their fiscal solvency and the programs that perpetrated their authoritarian character, depending upon the economic resources available to them. The few with very little finance options, have resorted to reducing levels of public expenditure, raising levels of domestic extraction, or a combination of the two which have produced a variety of political results, ranging from increased participatory politics (democracy) to internal chaos. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 51 However groundbreaking and noteworthy, observations of the political effects of the rentier state in the Arab world can be characterized largely by sweeping generalizations and statistically unsubstantiated correlation. Rentier literature has struggled to continue to describe and offer predictable political outcomes to the evolution of modern Arab public finance. While its theoretical explanations appear logically sound and qualitative documentation rather rich, rigorous empirical support for the theory has not been forthcoming and most recent experience proved rather mixed. There is ample opportunity for improvement in understanding the extent of the validity to the underlying assumptions within the theory. This study seeks to significandy enhance this body of literature on the impact and influence of renderism on the domestic political systems of the Arab world by employing a different angle of approach. Instead of relying exclusively on one or two case examples through which to analyze the Rentier State Theory and its corollary, as has been the case in previous inquiries, this study seeks to broaden the investigation by evaluating a cross-section of oil-rich and oil-poor countries within the region using a more empirically-oriented method of inquiry. It is hoped that through this deviation from the established pattern of topic literature new light will be shed on the determinants of the regional anomaly of persistent autocracy in the late 20th century Arab world. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 3 ESTABLISHING AN EMPIRICAL FRAMEWORK Towards a New Methodology In his critique of the study of Middle East politics since the Second World War, James Bill notes that “political scientists in the United States have made little progress in the past 50 years in understanding and explaining Middle East political systems.”76 This is a rather sobering incitement of modern regional inquiry in general and, by extension, that of the Rentier State Theory. The primary reasons for such a continued lack of indepth understanding of Middle Eastern political inquiry, Bill maintains, is due largely to research shortcomings and flawed methodological approaches. According to Bill, the region’s complex political nature, the lack of interdisciplinary perspective, inadequacy of the research skills and tools of analysis, the lack of coordination between regional experts and social scientific theorists, and the propensity for single country orientation are a few of the major reasons for limited progress in understanding the political processes of the Middle East.77 The body of rentier literature evidences mixed results with regards to these five areas. To their credit, rentier scholars have done quite well in avoiding many of the pitfalls that plague other regional scholarship. They are painfully aware of the complex nature of Middle Eastern politics, the need for interdisciplinary orientation, the importance of a 76 James A. Bill, “The Study o f Middle East Politics, 1946-1996: A Stocktaking,” Middle Ease Journal . v. 50. no. 4 (Autumn 1996), p. 501. 77 Ibid., p. 502. 52 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 53 healthy balance of fieldwork and theory construction, and the necessity of a trans-regional level of analysis in developing a comprehensive picture of the forces which shape political trends within the region. The very nature of their topic of study and line of approach demonstrates their commitment to accounting for these shortcomings common in other areas of Middle Eastern political inquiry. As previously noted, few if any scholars ascribing to the rentier state paradigm would support the use of any single theory to explain in totality the complex nature of Middle Eastern politics. However, the frequent absence of reference to or discussion of the profound influence of external rent in shaping the internal politics of the region in the great majority of regional political inquiry necessitates the focused nature of their investigation. The fact that the political influence of oil within the region is so commonly.held among regional scholars, but rarely substantiated, makes the significance of rentier literature particularly poignant. Furthermore, frequent qualifications are commonly made within rentier literature in distinguishing the degree of external rent’s influence shaping the region’s political orientations, processes and institutions, most notably the distinction between rentier and semi-rentier states. While it represents a common critique, oversimplification of what is a complex issue is not a fair evaluation of the Rentier State Theory or the majority of scholarship produced by its intellectual champions. Similarly, rentier literature demonstrates a fundamental understanding of the importance of interdisciplinary orientation when grappling with the complexities of Arab domestic politics. In feet, the rentier state paradigm is rather unique in this regard in that it provides a political economy approach, representing a hybrid of the fields of political science and economics. It employs both political and economic components in order to understand one of the fundamental causes of the region’s complex political nature. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 54 Rentier literature also demonstrates a relatively healthy tension between field research and theoretical analysis rarely evidenced in other areas of regional scholarship. While the primary architects o f the Rentier State Theory itself are largely theoretical in orientation, their work is frequently buttressed by more focused research into the areas of Arab public finance and economic structures. Support for the Rentier State Theory is frequendy cited from a combination of primary and secondary sources, multi-national and national financial publications from inside and outside the region. Furthermore, the demographic character of rentier scholars themselves is rather diverse, including both men and women, Arabs and non- Arabs, political and social scientists, and regional and theoretical experts. Such diversity of both researcher and research material is rarely found elsewhere within Middle Eastern political inquiry. Finally, rentier scholarship has included a broad range of levels of analysis which include single-country, sub-regional and regional studies. This variation has been particularly important in the formulation of sound generalizations about the relationship between external rent and politics within the region and, ultimately, to the generation of the Rentier State Theory itself. It has also been beneficial for more close evaluation and analysis of the workings of rent within the Axab political arena. Initially articulated solely in the Iranian context, over the course of the past two and a half decades, others broadened the applicability of the theory to both Arab rentier and semi-rentier states alike utilizing case studies of individual countries and sub-groups of Arab states. Despite these achievements, render literature has not totally avoided mistakes Bill notes as common among other Middle East polidcal scholarship. One of the primary flaws of rentier scholarship is its lack of statistical sophistication and quantitative rigor. While the theory draws heavily from observed economic phenomena that easily lend themselves to Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 55 quantification, surprisingly limited analytical and quantitative methods have been employed to substantiate the empirical generalizations o f the theory itself. M ost rentier scholarship, as evidenced in the previous section, is content with qualitative description, logical inference, implied relationships, and simple diagrams through which to support its hypotheses. While the theory has attracted significant qualitative description and general popular support among regional scholars, at least to some extent, it lacks an empirical analytical dissection. It is the absence of this quantitative method of analysis among rentier literature that provides the impetus of this study. Having combed through the existing works on the subject, this particular study will enhance the analysis of Rentier State Theory by attempting to correct for this apparent methodological shortcoming. New insights and concrete understanding are hoped to be gained by offering more rigorous quantitative analysis of the common assumption of the inverse correlation between petroleum profits and participatory politics within the late twentieth century, Arab world. Particular attention will be paid to identifying those variables associated with an Arab rentier state, identifying the particular challenges and obsticles to an introduction o f more complex quantitative analysis in this field, constructing a model through which to evaluate the Rentier State Theory itself, and predicting the expected outcome and anticipated variance within that model. Establishing a Rentier Equation As has been previously noted, the Rentier State Theory postulates two major economic dimensions of the state which affect its political nature, namely its size relative to the economy and the sources and structure of its income.78 It is argued that these two 78 Beblawi and Ludani, p. 5. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 56 components, uniquely modified through the availability of vast amounts of external rent, combine to provide the modern Arab state with a characteristically authoritarian nature. Stated otherwise, the rentier state thesis can be represented in its most basic form through the simple equation: Political Nature of the State = Relative State Size + Sources and Structure of State Income It is for this reason, rentier state theorists argue that the Arab world remains largely resistant to global democratic trends of the late 20th century, not the cultural determinants or arrested economic development that others frequendy contend. Since the theory seeks to offer an explanation of a particular pattern of political development found in the Arab world, one must begin with a brief discussion of the political nature of the state and establish a means of quantitatively describing what is an undoubtedly a qualitative concepc. The Political Nature of the State The Rentier State Theory is a political economy explanation of patterns of Arab national governance. Prior to launching into the relatively uncharted territory of quantifying the core economic components of the theory itself, one must first identify and operationalize the principal phenomenon against which these components can be measured. Despite its featured presence within social scientific literature throughout much of history, democracy as a form o f governance remains a largely allusive concept that eschews precise definition. Nevertheless, this lack of precision has not generally infringed on the formation of a general understanding of the principal elements associated with a democratic form of political system or the process of transitioning towards democracy. Nor has the inherent qualitative nature of the subject effectively stifled quantitative measurement of the concept. Language aside, there is notable commonalties among varying expressions of democracy. Most definitions focus on two distinct features or components of a democratic Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 57 system, one involving political rights, and the other civil rights or liberties.79 In a very structural sense, political power in a democracy is selected, accountable to, and removable by its contituency. Its will has a direct impact on the policies and actions of those who rule individually and collectively over a democratic state. These political rights are made both meaningful and effective through certain guaranteed civil rights or liberties which insure a broad degree of political expression and mobilization among the electorate. Democracy, therefore, can be defined as a political system in which ultimate political power resides with the majority of those who are governed and in an environment in which the governed are free to express and impress their individual and collective will upon both society and state. To the extent that both elements are present or becoming more common within the political system, democracy can be said to exist or in the process o f developing. These two elements not only form the basis for a qualitative definition of democracy but also form its quantitative measurement. Over the last two decades, sophisticated measures of democracy have been developed and compiled into comprehensive data sets by a number of scholars, including Kenneth Bollen, Mark Gasiorowski, Zehra Arat, Tatu Vanhanen, Raymond Gastil, Michael Coppedge and Wolfgang Reinicke, and Keith Jaggers and Ted Gurr.80 Each tend to emphasize particular nuances of political systems, cover 79 Alex Inkeles, “Introduction,” in On Measuring Democracy; Its Consequences and Concomitants _ (New Brunswick, New Jersey: Transaction Publishers, 1991), p. ix; see also Michael Saward, “Democratic Theory and indicies of Democratization,” in Defining and Measuring Democracy (Thousand Oaks. California: Sage Publications Inc., 1994), pp. 6-24. 10 See Kenneth Bollen, “Issues in the Comparative Measurement of Political Democracy,” American Sociological Review ■ vol. 45, no. 2 (June 1980), pp. 370-390 and “Liberal Democracy: Validity and Method Factors in Cross-national Measures," American Tournal of Political Science vol. 37, no. 4 (November 1993), pp. 1207-1230; Mark Gasiorowski, “The Political Regimes Project,” in Studies in Comparative International Development. vol 25, no. 1 (Spring 1990), pp. 109-125; Zehra Arat, Democracy and Human Rights in Developing Countries (Boulder. Colorado: Lynne Rienner, 1991); Tatu Vanhanen, The Process of Democratization (New York: Crane Russak, 1990); Freedom House, annual, Freedom in the World (New York: Freedom House Press, 1974-1998); Michael Coppedge and Wolfgang Reinicke, “Measuring Polyarchy," Studies in Comparative International Development _, vol. 25, no. 1 (Spring 1990), pp. 51-72; and Keith Jaggers and Ted Robert Gurr, “Tracking Democracy’s Third Wave with the Polity III Data,” Tournal of Peace Research . vol. 32, no. 4 (1995), pp. 469-482. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 58 various spatial and temporal domains, and employ differing scaling methods. Only two, however, provide measures for a relatively comprehensive collection of nation-states on an annual basis. These form the sources for measures against which economic indicators of the Rentier State Theory can be compared and an empirical model for the theory developed and evaluated. Political Nature of the State (AUTOC, DEMOC, POL, RIGHTS, LIB and FREE) In their construction of a set of quantitative indices of national political behavior, Keith Jaggers and Ted Gurr write, In its simplist conceptualization democracy is defined by what it is not; democracy is the opposite of autocracy. . . From this perspective, democracy and autocratic systems are assumed to occupy two ends of a single political continuum.81 This continuum perspective forms the basis of their Polity III data set which includes two indicators of regime type and eight indicators of political authority for 161 countries from 1946 to 1994. These regime type indicators (democracy and autocracy) each form an 11- point measure (ranging from 0 to 10) derived from five of eight political authority indicators included in the data set, namely the degree of constraints on chief executive, competitiveness of political participation, competitiveness of executive recruitment, openness of executive recruitment, and degree of regulation of political participation.82 The unique continuum framework and great number of cases under consideration has made the Polity data series (Polity I, II and HI) the most widely employed measure of democracy within the empirical social sciences.83 11 Jaggers and Gurr, p. 469. 82 Keith Jaggers and Ted Robert Gurr, Polity III Code Book . reference number ICPSR 6695 (Ann Arbor Michigan: Inter-university Consortium for Political and Social Research, 1995). 83 Sara McLaughlin, ec al., “Timing the Changes of Political Structures: A N e w Polity Database,” loumal of Conflict Resolution . vol. 42, no. 2 (April 1998), p. 232. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 59 Since the level of democratization in the Arab world is commonly held to be quite low, the Polity III indicators of democracy (DEMOC) and autocracy (AUTOC) are both necessary to observe the whole gambit of regime types within the region. By combining the two indicators of regime type by subtracting the 11-point measure of autocracy from the 1l- point measure for democracy and adding 10, a single regime type index ranging from 0 to 20 can be constructed into a single variable (APOL).84 Therefore, the greater the aggregate index number, the more democratic the regime or state. Since Polity III indicators are designed to measure polity persistence over the long-term, they are somewhat limited in their ability to capture short-term political fluctuations in the level of autocracy and democracy within a particular country. Nevertheless, they represent one of the most reliable and popular empirical indicators of democracy available for use in cross-national, social scientific inquiry. The Comparative Survey of Freedom, developed by Robert Gastil and maintained by a team of researchers at the Freedom House, also represents an equally valid annual indicator of democracy. While the Polity III indicators tend to focus more on institutional democracy, the Freedom Survey, as it is commonly referred, is comprised of interpretations of the level of democratic rights (individual freedoms) and influence that exist within 227 sovereign states and related territories between 1973 to the present. The Freedom Survey is comprised of two primary dimensions and a combined index drawn from the indicators political rights (RIGHTS) and civil liberties (LIB), each measured on a seven-point scale (ranging from 1 to 7). Political rights refer to those rights associated with meaningful participation of the individual in the political process, including the right to vote, compete M This combined index to gauge democracy is suggested by Jaggers and Gurr, p. 473-474; and Michael Mousseau, “Militarized Interstate Conflicts,” in Tournal o f Conflict Resolution . vol 42, no. 2 (April 1998), p. 215. It is Mousseau’s suggestion to make the index a positive number by adding 10 to the combined index. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 60 for public office, and elect representatives to have a decisive vote on public policies. Civil liberties, on the other hand, refer to rights of expression, organization, and freedom of activity, including religious practice and affiliation, mobility, and education. These two indicators of political rights (RIGHTS) and civil liberties (LIB) can be combined to form a single index, known as the Freedom Rating (FREE), on a 13-point scale (ranging from 2 to I4).85 The greater the numeric rating o f the indicators of political rights, civil liberties and freedom within the society, the generally less democratic the system of governance. This emphasis on the individual within a democracy captured by the Freedom Survey represents a fine compliment to the more structural interpretation of democracy tracked in the Polity III data set. While the Polity III and Comparative Survey of Freedom represent the only sources of annual measures of democracy within the national political systems and measure relatively distinct nuances of national political behavior from the individual to the structural, they remain highly correlated. In their introduction of the Polity III version of their Polity data set, Jaggers and Gurr compared the relationship between all seven of the other available time- series empirical measures of democracy drawn from a range of subjective to objective data. They found a relatively high degree of correlation between all sources, with correlation coefficients ranging from .72 to .93.86 Most noteworthy for this inquiry, the indicators of Polity III (combined index) and the Comparative Survey of Freedom (political rights and civil liberties) were found to be highly correlated at to each other at .92 and .87 and to the other sources, ranging from the .81 to the .93 levels. Despite the conceptual and methodological diversity among them, this relatively high degree of correlation further K Raymond D. Gastil, Freedom in the World. 1988-1989 (New York: Freedom House, 1989), p. 25. 86 Jaggers and Gurr, “Tracking Democracy’s Third Wave,” p. 475. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 61 validates the integrity of both sources relative to each other and all other existing sources measuring similar political phenomena. Together APOL (DEMOC minus AUTOC) and FREE (RIGHTS plus LIB) and their component indicators represent the measures against which the empirical elements of the Rentier State Theory can be compared and evaluated. After having established a means of operationalizing the political nature of the state, attention must be given to dissecting the two causal elements that comprise this quantitative model of the Rentier State Theory. The Relative Size of the State The size of the state relative to the national economy is a critical component of the Rentier State Theory, because it indicates the general magnitude of the state’s role within society. If the state’s relative size is small compared to the general economy, it remains largely peripheral to its own society or at least rivaled in its influence and control by other domestic entities (i.e. business interests). However, if it is large, it is more important in the daily lives of its citizens and a prize worth pursuing. According to one o f foremost scholars of the politics of public finance, Margaret Levi, The greater che revenue of the state, the more possible it is to extend rule. Revenue enhances the ability of rulers to elaborate the institutions of the state, to bring more people within the domain of those institutions, and to increase the number and variety of the collective goods provided through the state.87 There is, then, a direct correlation between the role of the state and its revenue and expenditure. Over both the long and short-run, state revenue and state expenditure are likely co be highly correlated, unless of course the state chooses to borrow against expected future revenue and postpone a corresponding reduction in a given year’s expenditure. In 87 Margaret Levi, O f Rule and Revenue (Los Angeles: University o f California Press, 1988), p. 2. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 62 either case, the greater the state’s budget in terms of revenue or expenditure, the more likely it is to be influential within society. However, as it relates to relative state size, there is a fundamental difference between state revenue and expenditure. This stems from the fact that societal penetration requires expenditure by the state. It is these expenditures that necessitate the state to finance itself through the extraction of revenue. In his discussion of state size, Luciani notes, “indeed state income cannot in any meaningful sense be considered as being part of GDP until it is actually spent. . .'m Furthermore, state expenditure is more commonly associated with the Rentier State Theory in that it is considered to be able to allow the state to purchase political power within society.89 State expenditure is, therefore, a more relevant indicator of state size when considering the state’s influence within society than is state revenue. Relative State Size (SIZE) Operationalizing relative state size is actually quite simple, at least in so far as determining broad patterns. By dividing state expenditure by the gross domestic product (state expenditure/GDP), the relative importance of the state within the economy can be determined.90 For this analysis, the function of state expenditure divided by the gross domestic product is captured in the variable SIZE and represented as a percentage.91 The greater the percentage, the larger the size o f the state and generally, its significance within society. M Luciani, “Allocation vs. Production States,” in The Rentier State . p. 65. 89 See Beblawi and Luciani, “Introduction,” in The Rentier State . p. 11. 90 State expenditure divided by GDP is the commonly utilized method o f determining relative state size within rentier literature. See Gause, pp. 46-51; Krimley, p. 190; Brynen, p.73. 91 Figures for state expenditure and GDP are derived from the International Monetary Fund, International Financial Statistics Yearbook and Government Financial Statistics Yearbook _ (Washington: International Monetary Fund, various). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 63 The Nature and Structure of State Income If the size of the state affects its relative importance, it is hypothesized that the nature and structure o f its income influence the direction of its political orientation. In general, state income is derived from one of two sources, either tax or nontax revenue. According to the Rentier State Theory, it is the degree to which the state relies on each of these fiscal components and the unique nature of their composition that determines its likelihood of being autocratic or democratic in nature. While this theory is typically considered germane in describing the determinants of political patterns in the Arab world, its central premise should remain universally applicable, since it is not based upon cultural or regionally specific determinants. However, in order to empirically evaluate the extent of the validity of this political economic hypothesis, one must first understand the fundamental nature and components of tax and nontax revenue and their assumed political implications. Tax Revenue and Its Political Implications: According to the 17th century finance minister under Louis XIV of France, Jean- Baptiste Colbert, “The Art of Taxation consists of so plucking the goose as to obtain the largest amount of feathers with the least amount of hissing.”92 As this colorful analogy implies, taxation has never been considered simply a fiscal concept. Rather, it represents a complex administrative tool of the state that involves political trade-offs between state and society. The ruler’s desire for more revenue is weighed against both the ability and will of his or her constituents to pay. A number of relatively recent studies have echoed this significant political nature of taxation and sought to describe the trade-offs necessary for successful taxation to be 92 Quoted from Sven Sceinmo, Taxation and Democracy (New Haven, Connecticut: Yale University Press, 1993), p. 19. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 64 implemented. At the conclusion of her study of the historical relationship between rule and revenue, Margaret Levi writes: Rulers maximize revenue to the state, but not as they please. Nearly all are predatory, whether they want to be or not, and all operate within significant constraints. . . Always and everywhere, the fundamental constraints are political-economic. They rest on political management of economic and political resources.93 Levi’s study highlights the dual nature of constraints of taxation, both economic and political. Similarly, in his analysis of the relationship between tax and politics in industrialized societies, Guy Peters finds that national political factors are not only of importance but of greater influence in the formation of tax policies than are economic, legal or redistributional considerations. According to Peters, Taxes are simply too important as political actions - in both real and symbolic terms - to permit other types o f factors (important as they certainly are) to have anything other than a secondary influence over the final selection of tax instruments. Taxation, therefore, is just as much if not more political as it is fiscal. The political nature of taxation is assumed to be strongly directional and require a commensurate level of legitimacy of the state. Political theorists have long noted the correlation between taxation and representative forms of governance.95 Robert Bates and Da-Hsiang Donald Lein go so far as to propose a particular formula for increased revenue extraction. They find through their analysis of the historical relationship between taxation and democracy that “revenueseeking [sic] governments may well find it to their advantage to 93 Levi, p. 184. M Peters, p. 21. 95 See Albert O. Hirschman, Exit. Voice and Loyalty: Response to Decline in Firms. Organizations and States _ (Cambridge, Massachusetts: Harvard University Press, 1970) and The Passions and the Interests: Political Arguments for Capitalism Before Its Triumph (Princeton. New Jersey: Princeton University Press, 1970) fora more robust discussion of this particular topic o f political economic inquiry. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 65 strike bargains with citizens whose assets they seek to tax.”96 This observation notes the strong positive relationship between taxation and democracy, in that revenue extraction promotes democracy. Margaret Levi also notes this correlation when she concludes that “[representative institutions ‘legitimize’ tax power.”97 Therefore, there is a logical and historical correlation between increased political liberalization, or democracy, and increased revenue extraction. This should not be surprising, considering the well-known democratic cry, “no taxation without representation” that so captures the demands of citizens for a greater political role within modern society and frequendy noted in rentier literature. Taxadon, therefore, has long been considered both highly political and directly correlated with the rise of participatory politics. However, not all forms of taxation should be considered equally influential in the advancement of democracy. Most scholars of taxation, whether from an economic or political perspective, differentiate between those taxes that are direct and those that are indirect. Direct taxes, such as those on income, property, payroll and social welfare, are charged direcdy to the taxpayer without immediate benefit. Indirect taxes, such as those on expenditure and transactions, however, are associated with the more involuntary taxpayer action in receiving state services or engaging in commercial exchanges. Generally, direct taxes are considered to be more “visible” to those being taxed, since they provide no immediate benefit and are not “hidden” by another transaction. They, therefore, are considered to require a greater political price on the part of the state to collect.98 By choosing to rely more or less on a particular type of tax base, the state can alter the inherent 96 Robert Bates and Da-Hsiang Donald Lien, “A Note on Taxadon, Development, and Representauve Government,” in Politics and Society . 15:1 (1985), p. 53. 97 Levi, p. 180. 91 For a broader discussion o f direct and indirect taxes and their differing political implicadons, see Peters, pp. 22-58. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 66 accountability and legitimization necessary to successfully extract revenue from its own citizens. The highly contested debates over tax structure common among many esoteric states further evidences this reality. Tax Revenue (INCTAX, SOCTAX, PAYTAX, PROPTAX, EXPTAX, INTLTAX and TAXIMP) Tax revenue can be operationalized in seven variables derived from statistics of central government revenue. It has already been established that direct taxes consist of taxes on income, property, payroll and welfare, while indirect taxes consist of taxes on expenditure and transactions. The direct taxes on income, property, payroll and social welfare are indicated separately as a percentage of government revenue in this inquiry (INCTAX, PROPTAX, PAYTAX, and SOCTAX, respectively) and derived from IMF statistics for consolidated central government, or budgetary central government when consolidated figures are not available.99 Indirect taxes, including taxes on expenditure and international trade, are also indicated separately as a percentage of government revenue (EXPTAX and INTLTAX, respectively) and derived from the same IMF statistical data. The final indicator of taxation is an aggregate of all six of che above mentioned direct and indirect tax components divided by the total amount of government expenditure to provide a relative indicator of tax impact (TAXIMP). These seven independent variables together represenc a strong indicator of the importance of taxation within a country during any given year. They are frequently utilized in individual, aggregate or slightly modified aggregated forms in rentier literature to highlight the limited and skewed nature of tax policy within the Arab world.100 99 Scads tics for che different components of government revenue sources, total government revenue and expenditure ate derived from the Internadonal Monetary Fund, Government Financial Statistics Yearbook _ (Washington: International Monetary Fund, various). 100 See Beblawi and Luciani, The Rentier State : and Rex Brynen, pp. 73-80. Brynen uses the term “domestic tax” to differentiate taxes on international trade and. transactions from among ocher forms o f taxadon chat are levied on the domesdc economy. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 67 Nontax Revenue and the Rentier State If increased taxation is theoretically associated with increased participatory politics, particularly that which involves a relatively high degree o f popular consent and legitimization, than it stands to reason that a lack of taxation would encourage state autonomy and political authoritarianism. The existence of abundant nontax revenue remains the only means to finance state penetration of society without being forced to make the political concessions or bargains long associated with taxation. It is this assumption, uniquely modified by the abundance of rent-based nontax revenue within the Arab world that forms the core hypotheses of the Rentier State Theory. Since the dependence of nontax revenue in state finance is certainly not exclusive to che Arab world, it is important to understand the unique nature of nontax revenue within this particular region which gave rise to the notion of a rentier state and its corresponding theory. According to the chief architects of the Rentier State Theory, Hazem Beblawi and Giacomo Luciani, a rentier state is defined as “any state that derives a substantial part of its revenue from foreign sources and under the form of rent.”101 Having access to significant external revenue sources that are easily monopolized and that have an unusually large ‘value added’ relative to actual cost of production liberates the state from having to rise revenue from its own population. This distinguishes it from the dominant, largely extractive model of state finance found in Western industrial and emerging democracies and referenced by Luciani as “esoteric.” Likewise, the monopolization of vast amounts of external rent frees the rentier state and the semi-render state, albeit to a lesser degree, from the widespread mobilization associated with the socialist state paradigm found elsewhere in the post industrial world. The rentier state’s unique position as the principal facilitator of external Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 68 rent and the domestic economy radically strengthens its economic and political position vis- a-vis its own society. It is able to generate an effective welfare state and the dependence of society it inspires without sophisticated tax instruments or social mobilization that requires more accountable, power-sharing arrangements with its citizenry. As previously defined, the rentier state itself is created and sustained through a unique pattern of public finance based upon economic rent, a particular form of nontax revenue. Instead of raising funds through domestic taxation or state-sponsored economic enterprises, the rentier state in the Arab world depends to a large degree on financial resources accrued from rent earned abroad. External rent is derived from foreign sources and maintains litde linkage to the rest of the economy.102 This externally of revenue further insulates the state from its own citizens by detaching it almost completely from the productive elements of the domestic economy. The principal source of rent in the Arab world, the sale of hydrocarbon natural resources, requires very litde involvement from the domestic economic sector to be extracted and sold on the international market. Yet, the value-added to this economic resource, even in its raw state, is among the highest in the world, even among manufactured goods. Most rentier scholars consider “rentier” those states whose income is constituted from 40-50% or more of this external rent.103 In the Arab world, the lion’s share of external rent is obtained through the sale of petroleum products on the international market. However, other forms of external rent are also common within the region, including location rents (transit fees), foreign aid, overseas investment income, and worker’s remittances. Whereas all external rent contributes to the making of a rentier economy, it is that which flows directly to the state in the form of nontax 101 Beblawi and Luciani, p. 11. 102 See Madhavy, p. 58; Beblawi, “The Rentier State in the Arab World,” in Beblawi and Luciani, p. 5.1; and Krimley, p. 183. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 69 revenue that contributes to the establishment of a rentier state.104 While great disparity does exist within the region with regards to rentier sources of nontax revenue, all Arab states are financed, at least to some extent, by external rent derived direcdy or i n d i r e c t l y from petroleum profits circulating within the region as frequently cited throughout rentier literature. Nontax Revenue (STINCOME, GRANTS, CAPREV, VIABLE and NONIM P) The relative role of nontax revenue in state finance can be operationalized through four variables: state income, grants, capital revenue, and deficit financing. The external rent associated with the rentier and semi-rentier state phenomenon is primarily captured in the form of state income or foreign grants, at least that which is accrued directly to the state in the form of revenue.105 State income includes revenue from entrepreneurial and property profits and fees and fines collected by the state. Most external rent collected directly or indirecdy from petroleum revenues, in the form of the international sale of petroleum products, transit fees related to the movement of petroleum products to market, and/or “petrodollar” investment income from global financial markets, is considered a part of state income. Grants or foreign aid also represent a form of external rent. While not all grants are a by-product of petroleum revenues, within the Arab world, recalculated petroleum profits represent a significant means through which the rentier state phenomenon of a few oil-rich states spread throughout the region. According to one study, inter-Arab foreign aid 103 See Luciani, “Allocation vs. Production States,” in Beblawi and Luciani, p. 70; and Gause, p. 43. 104 See F. Gregory Gause III, “Regional Influences on Experiments in Political Liberalization in the Arab World,” in Rex Brynen, et. al eds., Political Liberalization and Democratization in the Arab World pp. 291- 293. 105 While worker’s remittances or repatriated income is also noted as a significant quasi-external rent within rentier literature, it is not directly accrued to the state in the form o f revenue. Rather worker’s remittances Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 70 represented nearly 75% of all aid offered by the petroleum-rich Arab countries in the late 1970s, when petrodollar aid was at its highest relative levels.106 Regardless of their origination, foreign grants serve to enable the state to augment its coffers without extracting revenue domestically. Capital revenue and deficit financing, while less fundamental in the creation of a rentier or semi-rentier state, are means through which the state can raise revenue also without resorting to domestic extraction. Capital revenue in monetary terms is typically relatively small compared to other forms of nontax revenue. It is derived primarily through the sale of capital assets of the state (i.e. land and state interest in public corporations) and therefore extremely limited in its revenue generating potential over the long term. Deficit financing through borrowing, however, has become a significant source of nontax revenue in most countries in the short-term, but can not be sustained indefinitely. Furthermore, since borrowing serves as not only a means of meeting budget deficits but also a means of making a return on investment by the lender, this form of nontax revenue has the long-term disadvantage of having to be repaid with interest and, in the case of the developing world, inviting external economic and political pressure that can actually erode state autonomy. In this sense, borrowing and grants flowing into the region from abroad are often offered with economic and political “strings” attached, as is often the criticism of debtor countries against the international lending institutions of the wealthier countries. In the Arab world, borrowing has not been as prevalent a phenomenon on the whole thac it has been in the rest of the world, but one that, since the oil bust years of the 1980s and 1990s, is only increasing. serve to contribute to a broader rentier economy, not a rentier state. For a more detailed discussion of worker’s remittances, see Beblawi and Luciani, pp. 13-15. 106 Andre Simmons, Arab Foreign Aid (East Brunswick. New Tersev: Associated University Presses, Inc., 1981), p. 27. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 71 All four of these components of nontax revenue are easily operationalized, both independendy and collectively. Since state income, grants and capital revenue are all financial in nature and contribute directly as nontax revenue in national accounts, they are best indicated as a percentage of state revenue (STINCOME, GRANTS, and CAP REV, respectively).107 Clear financial data on borrowing is a bit more difficult to obtain and is therefore derived from the surrogate statistic of the state’s budget deficit (state revenue divided by state expenditure) under the assumption that any deficit must be made up through borrowing.108 Since borrowing impacts the fiscal viability of the state over the long- run, it is represented by the variable “VIABLE” and indicated as a percentage, with a score of 100 representing parity between state expenditure and state revenue during that year. Budget surpluses and deficits are reflected as scores greater or less than 100, respectively. Together these four variables (STINCOME, GRANTS, CAPREV, and VIABLE) present a means of quantifying the relative role of nontax revenues to any given state and therefore, a rough means of capturing both the relative dependence of that state on external rent and independence from the need for imposing taxes to raise state revenues. Nontax revenues, therefore, can be generally considered to be inversely related to tax revenues in the composition of state revenue and therefore, directly associated with increased state fiscal autonomy. When modified by the relative size of the state within the national economic system as a whole, the sources of state revenue can be considered to also influence the political nature of the state. This relationship between state size, state revenue and the national political system lies at the heart of the Rentier State Theory and the means through which it can be quantitatively evaluated. 107 Figures for state income, grants and capital revenue are all derived from IMF, Government Financial Statistics Yearbook . various years. State income is indicated as “nontax revenue” by the IMF. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 72 Adding Dimension to the Model In order to construct a more dynamic model that reflects the theoretical context, the importance of change among the principal components, a potential time lag or lead associated with any socio-political behavior, and corrects for nonadditivity, a few other variables must be also included. While few of them are based upon concepts found direcdy in rentier literature, they are included as logical extensions of the theory and frequent components of any robust and multi-dimensional quantitative analysis. Controlling for Modernization (GDPCAP) Firsdy, as has been previously discussed, the Rentier State Theory does not as much represent a deviation from as an important qualifier on the Modernization Theory. Therefore, the variable most associated with the Modernization Theory, GDP per capita, will also be included in the analysis. However, to assist in comparing this indicator of economic progress across multiple countries, this variable is standardized through the use of the IMF’s monetary measure, the SDR, which represents a monetary value independent of any particular country’s currency and facilitates cross-national comparison. Therefore, standardized GDP/capita (GDPCAP) is factored in SDRs. Nevertheless, GDP per capita is only a rough indicator for development (or modernization) and therefore, limited in its explanatory ability. Adjusting for Change (CSIZE, CTAXIMP, CNONIMP, CGDPCAP) A second dimension is added to adjust for annual change in the aggregate components of state size, tax and nontax revenues, and modernization. While not previously noted within rentier literature as a factor influencing national political behavior in the Arab IM Figures for state expenditure and state income are derived from IMF . Government Financial Statistics Yearbook. various years. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 73 world, change in the principal components of the Rentier State Theory may play as much a role in shaping democratization or a lack thereof as absolute financial figures. Therefore, the change in value of the principal aggregate variables, SIZE, TAXIMP, NONIM P and GDPCAP, from one year to the next is captured in the change variables CGDPCAP, CSIZE, CTAXIMP and CNONIMP and reflected as a percentage of change from the previous year. The dramatic increase o f any of these elements in one year could logically contribute to a noticeable change in the political sphere. Factoring Time Lag and Direction of Influence or Time Lead Logically, it is also safe to assume that political change may not occur in the same year as any given adjustment in state size and finance structures. Therefore, a time lag or delayed influence variables must also be included within the equation. By adding a year to the dependent variables utilized to quantify the political nature of the state, the relative time- lag associated within the model can also be evaluated. The indicator “P” followed by the number of years of delay is added to each of the six dependent variables in order to assist in the assessment of time-lag within the model (i.e. AUTOCPIO would imply a ten year delay on the dependent variable quantifying autocracy). Similarly, the actual direction of influence on the relationship between the independent variables and the dependent variables or time lead can be factored by subtracting one year. This is coded in the data with a “M ” followed by the number o f years o f reverse affect or time lead (i.e. AUTOCM1 indicates a one year lead time in influence or a change in dependent variable affecting the independent variables). There is at least a theoretical possibility that increased democracy, autocracy, civil liberties or political rights could assist in the modification of state size, revenue sources, GDP/cap, instead of the reverse which is implied in the equation. While a time lead in any given model could facilitate a Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 74 better understanding between the dependent and independent variables, it is of limited utility in evaluation of the statistical significance of the Rentier State Theory since causality is only specified as unidirectional within rentier literature. Correcting for Nonadditivity (STAXIMP and SNONIMP) The final dimension added to the model is that of multiplicity or nonadditivity between the two principal components of the Rentier State Theory, state size and state revenue. The assumption of the rentier equation is that the relationship between state size and state revenue is additive, that is: Political Nature o f the State = Relative State Size plus Sources and Structure of State Income However, there is a real possibility that state size actually magnifies or multiplies the influence of state tax and nontax revenue sources in shaping the political nature of the state, that is: Political Nature of the State = Relative State Size times Sources and Structure of Sate Income To factor this potential magnifying relationship, two additional control variables are added to the model (STAXIMP and SNONIMP). STAXIMP represents the sum of SIZE (state expenditure/GDP) times TAXIMP (tax revenues/state revenues) and reflects the increased importance of tax revenues when the state is large or more influential in society. Similarly, SNONIMP represents the sum of SIZE times NONIMP (nontax revenues/state revenues). If the importance of the source of revenue increases exponentially with the relative size of the state, than these variables would add a significant dimension to the analysis. Scope o f Analysis: Contextualizing the Model When establishing an empirical model for the Rentier State Theory and evaluating its statistical significance, it is important to consider the spatial and temporal context in which Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 75 the theory itself developed and to which it is thought to apply. Since the phenomenon of the external rent-financed, modern state is considered exclusive to the Arab world, it should be of litde surprise that the theory was established as a means of describing the significance of external rent in shaping national political behavior within the Arab world alone. Spatial Domain From Maghrab to Makrek, the core of the Arab world can be said to include a total of 17 countries, including: Morocco, Algeria, Tunisia, Libya, Egypt, Palestine, Lebanon, Syria, Jordan, Saudi Arabia, Yemen, Oman, the United Arab Emirites (UAE), Qatar, Bahrain, Kuwait and Iraq.109 While only half of these countries can be considered “rentier” by definition due to the relative size o f their petroleum resources within their countries, all of these states have been frequendy cited within rentier literature as being significandy affected by the rentier phenomenon. Together, they comprise the principal subjects of any comprehensive evaluation of the theory. Yet, since the theory itself evolved as a means of further qualifying or explaining the limits of the global applicability of the more dominant Modernization Theory in political economic inquiry, the political evolutionary experiences of all countries of the world could be considered important in evaluating the basic assumptions of the theory with regards to state size and revenue sources. As with most quantitative inquiries, however, theoretical delineation of spatial domain is gready restricted to the uncomfortable confines of data availability. This inquiry is no different and therefore, 109 Mauritania, Sudan, Djibuti, Cyprus, Turkey, Iran, and Israel are also sometimes considered part of a broader definition o f the region itself, but are peripheral to the core of the Arab world, because they contain political, religious, linguistic, cultural, and/or significant ethnic charateristics that make them considerably different than those defined here as part of the strictly Arab world. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 76 ultimately includes only a cross-section of total of ten of these Arab countries and 29 non- Arab countries in the global community of nation-states.110 Temporal Domain Besides the specification of the spatial domain or geographic context in which the Rentier State Theory is applicable or which facilitates its evaluation, the temporal domain of the theory and its evaluation must be also established. Again, rentier literature itself proves particularly instructive in isolating the period under analysis. Ironically the earliest articulation of economic rent in the Arab context actually began at the dawn of the global oil boom era with Hossein Mahdavi’s documentation in 1970 of his observations concerning the skewed development patterns of Iran. Beginning in 1973, with the four-fold increase in the price of petroleum in the global market and its subsequent affect on Arab state revenues, the state rapidly became the dominant player in Arab society, in a time-frame unparalleled in history. From the early 1970s, rentier literature evolved in its ability to interpret Arab national political events in an era of the past two and a half decades which included both dramatically elevated and suppressed petroleum profits. The Rentier State Theory, therefore, is best analyzed over these past two and a half decades (1970-1995) when both the influence of external rent and advent of the scholarship it inspired were made manifest. As with spatial domain, this temporal domain is significandy restricted by the availability of data to include varying ranges of years during this timeframe depending on the country. While informadon on most o f the 25 year span is available for most countries, a few such as Bulgaria, China and 110 These 39 countries under analysts include: Bahrain, Egypt, Jordan, Kuwait, Morocco, Oman, Syria, Tunisia, UAE and Yemen from the Arab World and Australia, Austria, Bulgaria, Canada, China, Costa Rica, Denmark, Ethiopia, France, India, Indonesia, Iran, Israel, Kenya, Malaysia, Mexico, Myanmar, Nepal, Norway, Panama, Sierra Leone, Sri Lanka, Thailand, Turkey, UK, USA, Uruguay and Venezuela. While all Arab countries with available data were included, a few countries outside the Arab world were deliberately excluded in order to construct a more balanced, representative sample of non-Arab countries against which to Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 77 Yemen only allow for 5 or 6 years of analysis. They were nevertheless included due to their representational importance within the general data. Establishing an Empirical Model: Minimizing Specification Error As was articulated earlier, the Rentier State Theory can be quantitatively evaluated by testing the essential relationships captured in the simple equation: Political Nature of the State = Relative State Size + Sources and Structure of State Income where the political nature of the state represents the observed phenomenon or dependent variable and operationalized indicators from the Polity III and Freedom Survey data sets, and the relative state size and sources and structure of state income represent the predictors or independent variables and operationalized using financial statistical data from IMF international and government statistics yearbooks. The sources and structure of state income can be further disected into smaller blocks of variables representing tax and nontax, direct and domestic forms of revenue. Table 1 presents the various indicators of the Rentier State Theory to test the principal hypotheses associated with it. The more simplistic equation of the Rentier State Theory can be further refined using the following, more empirically-oriented, equation: APOL or FREE = a + b(SIZE) + [b(INCTAX) +b( PAYTAX) + b(PROPTAX) + b(SOCTAX) + b(EXPTAX) + b(INTLTAX)] T+[ b(STINCOMEO + b(GRANTS) + b(CAPREV) + b(VIABLE) W + b(GDPCAP) + [b(CGDPCAP) + b(CSIZE) + b(CTAXIMP) + b(CNONIMP)Jc + [b(STAXIMP) + bCSNONIMPJW + e where “a” represents the intercept or constant, “b” represents the partial slope coefficient of the independent variables, and “e” represents the random error associated with the equation. Together, INCTAX, PAYTAX, PROPTAX SOCTAX, EXPTAX and INTLTAX represent compare the regional phenomenon o f the Rentier State Theory. Limited data availability for former Soviet Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 78 indicators of tax revenue, while STINCOME, GRANTS, CAPREV and VIABLE represent indicators of nontax revenue. CGDPCAP, CSIZE, CTAXIMP, CNONIMP represent the indicators of change in principal independent variables and STAXIMP and SNONIMP represent those variables added to the model to correct for the additivity assumption inherent in the equation but not necessarily in the relationship between the principal components of the state that affect its political nature. The tax revenue component of state revenues can be further divided between direct taxes (INCTAX + PAYTAX + SOCTAX) and indirect taxes (EXPTAX and INTLTAX) in order to compare the importance of tax “visability” in affecting national political patterns. Domestic taxes would include all tax revenues except those on international trade and transactions. Dummy variables are also inserted into the equation in order to correct for the panel effects associated with multiple year, cross-national analysis. The specification error associated within this general empirical model of the Rentier State Theory is minimized by realizing its theoretical limitations in explaining the all the nuances of the theory itself and by eliminating insignificant indicators associated with this more general equation. It is important to note that the indicators proposed to capture the core elements of the theory in this equation are only approximate indicators of the rentier phenomenon itself. As was previously discussed, external rent is not isolated as such within the nontax revenue indicators of state income (STINCOME) and foreign grants (GRANTS), rather included within them. Therefore, these indicators may also include state income derived from domestic sources (i.e. state-owned industries) or foreign grants which are derived from non-petroleum resources (i.e. U.S. aid to Egypt). Nevertheless, a general pattern should be observable from these indicators that either supports or contradicts the theory or can be explained to do so more qualitatively. Those indicators that are less and the majority of Sub-Saharan countries reduces the truly representative nature of the overall sample. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 79 Table 1: Indicators of the Rentier State Theory Dependent Variables Independent Variables Politv III Index (APOL) Relative State Size: Autocracy (AUTOC) State Expenditure/GDP (SIZE) Democracy (DEMOC) State Revenue Sources: Freedom Survev Index /FREE) Tax Revenues (TAXIMP) Political Rights (RIGHTS) Direct Taxes Civil Liberties (LIB) Income Taxes (INCTAX) Pay Taxes (PAYTAX) Property Taxes (PROPTAX) Social Welfare Taxes (SOCTAX) Indirect Taxes Taxes on Goods and Services (EXPTAX) International Trade Taxes (INTLTAX) Nontax Revenues (NONIMP) State Income Revenues (STINCOME) Foreign Grants (GRANTS) Dimensional Indicators: Capital Revenues (CAPREV) Deficit Financing (VIABLE) Time Lead (dependent variables minus 1) Dimensional Indicators: Tim e Lag (dependent variables plus 1-12) Modernization (GDPCAP) Change (CGDPCAP, CSIZE, CTAXIMP, CNONIMP) Nonadditivity (STAXIMP, SNONIMP) significant or insignificant are anticipated to be identified, qualified and/or eliminated entirely from the model when they are regressed more comprehensively. Uncovering the Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 80 significance and direction of these indicators in affecting the national political orientation of Arab states, or perhaps any state in the world, is the primary focus of the statistical analysis undertaken in this study. Evaluating the Data: Minimizing Measurement Error As with all quantitative inquiry, the validity of the results are only as good as the ingredients, or data, behind the them. This particular empirical inquiry benefits from relatively reliable data, at least as much as can be reasonably expected. The previously proposed indicators associated with operationalizing the Rentier State Theory into an empirically testable equation are derived from three distinct sources. The dependent variables, of which there are two, are derived from the Polity III data and the Freedom Survey. While both o f these sources are fundamentally limited due to their very nature, reducing the highly qualifiable subject of national political patterns into quantifiable data, they are nevertheless highly regarded by and frequendy utilized in social scientific empirical research. The reliance of this inquiry on these two sources is justified largely due to the fact that they represent the only sources for time-series, cross-national quantification of regime data currendy available. Since they are highly correlated to each other and other more limited sources quantifying the same phenomenon, despite the use of differing research strategies and qualitative base sources, they can also be reasonably considered of high integrity. The indicators provided by each source are kept distinct within the equation in order to avoid scaling and interpretation problems associated with the combination of loosely related variables. The independent, causal variables in the model are also considered to be of reladvely high integrity due to the nature of their source. By relying on one, highly regarded source Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 81 for all financial statistics in the data, the independent variables are relatively Free From both random and nonrandom bias and facilitates cross-national comparison. To be sure, additional cases or deeper specification oF variables could be undertaken in operationalizing and evaluating the Rentier State Theory but not without a significant increase in the measurement error associated with more multi-source derived data. Financial statistics are available For nearly every country oF the world, if one is willing to rely on each government’s own measurements and methods oF documentation. Rather than subject this inquiry to such high levels oF potential error and measurement bias and in the interest oF Facilitating cross- national comparability over time, only IMF sources are utilized.111 The IMF is well-known to be one oF the most comprehensive and reliable sources oF global financial statistics currendy available. By restricting the sources oF the various independent variables to only those oF the IMF, the maximum reliability oF data and minimal measurement error is assured at the cost oF a more limited spadal and temporal domain. Framing a Quantitative Analysis: Oudining the Hypotheses to be Evaluated Discovering the statistical significance oF core hypothesis oF the Rentier State Theory is at the center oF this inquiry. According to this hypothesis, the unique render nature oF large portions oF Arab state revenue gready insulates the state From political accountability to its citizenry and contributes to its autocratic character. IF the theory is correct, then, one would a strong correlation between the degree oF dependence on nontax revenue andthe level oF autocratic and democratic patterns within the region. The greater the relative size oF the state and the level oF its fiscal insulation through nontax Forms oF revenue, generally, the greater the likelihood oF it being more autocratic and less democratic in nature. 111 Namely, the IMF’s International Financial Statistics and Government Financial’Statistics Yearbooks. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 82 As a natural correlation to this hypothesis, also frequently cited in rentier literature, tax revenues should be equally influential when regressed against autocracy and democracy. The greater the dependence on nontax revenues, the more likely a state to be more autocractic and less democratic. The greater the dependence on tax revenue, however, the more likely it is to be democratic and less autocratic. By evaluating this hypothesis throughout the entire modern oil era (the 1970s through early 1990s), the significance of the relationship between state size, revenue sources and national political orientation for any given state should become more apparent. Furthermore, by extending this analysis across the whole of the Arab world, to the extent possible given the availability of data, the regional nature of the rentier phenomenon also should be detected. This empirical analysis should indicate to what extent the thus far exclusively qualitative articulation of the Rentier State Theory can be quantitatively verified or supported. Beyond the evaluation of this core hypothesis, a quantitative regression of the empirical variables associated with the Rentier State Theory should indicate those variables that have the most significant effect on shaping the state’s political orientation. It is assumed that the more “rentier” the source of state revenue (i.e. the nontax revenues of state income and foreign grants), the more insulated it is from direct accountability to its own citizens and less likely it is to enter into the power-sharing relationship with its own people that is the basis of democratic politics. The more “visible” and direct the fiscal relationship between state and citizen (i.e. the tax revenues collected on income, property, and payroll), the greater the likelihood of fostering national democratic political behavior. Therefore, not only the extent to which the theory itself remains statistically significant, but also the individual components associated with it, can be evaluated through this inquiry. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 83 Thirdly, since the fiscal determinants of the political orientation of the state (i.e. state size and revenue sources) are not unique to Arab countries, only the relative importance of external rent in the composition of Arab state revenue, then the corresponding rentier national political pattern should be substantively different in other countries than in those of the Arab world. In order to evaluate the regional exceptionalism implied in the theory, this quantitative analysis will also apply rentier indicators to the global community of states to facilitate cross-regional comparison. If the Rentier State Theory does truly provide an explanation for a unique pattern of political behavior within the Arab states compared to the whole world, then this should be also evident through this inquiry. The same rentier indicators are utilized to discover the extent to which they prove significant in explaining national political behavior worldwide and which of those indicators are relatively more important in influencing those measures most associated with increasing democratization. Finally, as is the case in most empirically oriented, social scientific inquiries, it is anticipated that some variance of pattern between countries and within any explanatory model developed will exist. These variances could be related to the primacy of external rent in state revenue between countries, such as semi-rentier and rentier states, or related to a deviance from the assumptions implicit in the proposed rentier equation. Analysis of the results of the multivariate regression done on the most significant models should provide valuable insights into the limitations in determining the causal orientation of rentierism, if indeed it exists, within the Arab and non-Arab world. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 4 ANALYZING THE RESULTS Bringing an Empirical Dimension to The Render State Theory As previously asserted, the Rentier State Theory evolved over the period of the last two and a half decades as a means to explain the anomaly of persistent authoritarianism in the Arab world despite relative modernization within Arab societies and global trends toward increasing democratization. The core hypotheses associated with the theory were established primarily through the use of qualitative methods of analysis involving a series of country and sub-regional case studies. While the theory itself has been largely embraced by the majority of regional scholars as a powerful means of understanding the unique pattern of Arab political development, or lack thereof, it has yet to be subjected to rigorous statistical analysis, despite the predominately quantitative nature o f the subject. In order to gain a deeper understanding of the Rentier State Theory and assess its statistical significance, this particular study seeks to complements the existing body of rentier literature by infusing a more rigorous and systematic method of inquiry into an otherwise exceedingly impressionistic and exclusively qualitative genre of literature. Having traced the evolution of the Rentier State Theory and establishing a quantitative means of describing and analyzing it, a series of multivariate regressions were done on the various rentier influences of national 84 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 85 political behavior in the Arab and non-Arab worlds alike.112 The results of these series of multivariate regressions not only serve to better substantiate the validity of the Rentier State Theory itself but also identify the components associated with quantifying the rentier state that are most statistically significant in explaining various types of national political patterns. The first series of regressions were undertaken to discover those components associated with measuring democracy (or autocracy) that are most affected by rentier indicators, assess the significance of individual variables in shaping Arab national political patterns, and establish a basis for developing more explanatory statistical models through which to evaluate the Rentier State Theory. The core components of state size, state revenue sources (tax and nontax revenues), and modernization, as well as the she “dimensional” indicators associated with factoring for change and nonadditivity that were identified in the previous section, were regressed on the six dependent variables associated with measuring various aspects of democracy in national political behavior. Table 2 presents the results of these comprehensive, unspecified regressions. The first noteworthy observation of the results of these regressions is the relative impact of the rentier indicators in shaping autocracy (AUTOC) and civil liberties (LIB). Given the relatively low levels of political development within the Arab world, which was a principal impetus in the development of Rentier State Theory, it is not surprising that those national political indicators least direcdy associated with democratic political systems among those of the Polity III and Freedom Survey indexes are better described by rentier indicators than chose of democracy (DEMOC) and political rights (RIGHTS). The combined indexes of APOL and FREE, since they contain both measures of democracy and autocracy and 112 Ail regressions in this study use the ordinary least squares (OLS) method o f regression and are run at the standard 95% confidence level. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 86 political rights and civil liberties respectively, are o f only marginal value in explaining the nature of political change associated with the various indicators of the rentier state. This is most likely due to their less specific nature. With clearly the lowest inherent error (root Table 2: Modeling Rentier Indicators of National Political Behavior in the Arab World Independent Dependent Variables Variables Polity III Index Freedom Survey Index TAXIMP 356 (3.34)* -.125 (-2.52)* 331 (3.87)* 331 0.87)* -.038 (-1.39)*** -.041 (-1.77)** INCTAX -.139 (-2.03)* .010 (31) -.129 (-337)* -.044 (-1.49)“ * -.024 (-1.46)*** -.020 (-138)— PAYTAX -.285 (-2.08)* .066(1.04) -319 (-2.85)* .149 (2.38)* .090 (2.57)* .059 (1.93)** PROPTAX -.543 (-4.05)* .152(2:44)* -390 (-531)* .089 (1.46)— .038(1.11) .059(1.72)'* SOCTAX -.132 (-1.82)** .0 09(37) -.123 (-3.02)* -.081 (-2.57)* -.039 (-231)* -.04Z{-2.74)» EXPTAX -.182 (-2.72)* .021 (3 8 ) -.161 (-438)* -.013 (-.46) -.007 (-.42) -.0006(-.45) INTLTAX -.083 (-1.06) -.025 (-.69) -.109 (-2-46)* -.058 (-1.69)** -.032 (-1.65)** -.026 (-1.57)— NONIMP -.005 (-.18) .000 (.01) -.005 (-31) .021 (1.49)— .013 (1.69)** .008(1.13) STINCOME .109 (138)— -067 (-1.81)** .043 (.96) -.056 (-1.50)*** -.027 (-131)"* -.028 (-1.58)— CAPREV .098 (1.00) -.100(-230)* -.003 (-.05) -.048 (-1.06) -.035 (-1.38)*“ -.0I3(-39) GRANTS .064 (.77) -.040 (-1.03) .024 (.53) -.054 (-139)*** -.026 (-130)*** i -328 (-143)— VIABLE -.146 (-1.85)** 087(238)* -.059 (-1.33)*** -.059 (-1.33)*** .039 (1.90)** I .037(2,07)* GDPCAP .000 (8.92)* .000 (9.91)* -.000 (-11.92)* -.000 (-10.61)* -.oooGuao)? CSIZE -.849 (-.60) , ; ;472072) ' -377 (-.47) 3 35 (35) .155 (.41) ; .080(35) CTAXIMP -2352 (-2.00)* ■ .969 (1.78)** -138 (-2.10)* 1.167(2.10)* 391 (1.90)** 376(2;I4)* CNONIMP -.001 (-.07) ^^^ooa^oiQ!-;• -.001 (-.17) .003 (.89) .002 (1.02) .O O l'W . CGDPCAP -1.776 (-234)* ' .832(236)* ' -.944 (-2.13)* .526 (1.42)*** 324 (1.55)“ * 30Z(r.I2) STAXIMP .003 (2.82)* > -.001 (-2.83)* .001 (2.67)* -.001 (-2.39)* -.001 (-232)* -,000(-235)*: SNONIMP .000 (30) : -.000 (-.03) .000 (.33) -.001 (-1.51)*** -.000 (-1.70)** -.000(-1.15) Adjusted Rz .6310 3405 .6126 .6421 .6089 .6386 Root MSE 4.9677 23107 2.784 23821 13368 1.1547 Constant 7373 (230)* 5.023(3.41)* 2396 (1.30)*** 10.076 (6.89)* 4.864 (5.92)* 531(735)* Observations 801 ‘ 801 801 845 845 845 NOTE: Coefficients given for indicators of the most explanatory model. T-statistics given in parentheses. F-statistic equals zero for all models. *P<.05, **P<.i0,*** P<.25 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 87 MSE) and higher levels of specificity (adj. R2), these two dependent variables are those upon which more powerful explanatory models were developed and analyzed for the Arab world. However, even these unspecified models appear to indicate at least some connection between rentier indicators and national political behavior. When regressed against each of the six indicators for national political development, the rentier indicators appear to account for between 61% to 64% o f the level of relative democracy in the Arab world over roughly the last two and a half decades. This relatively high correlation, by social scientific standards, between the models on the sue dependent variables regressed by the comprehensive set of time-constant rentier indicators further underscores both their qualitative integrity in describing elements associated with democratization and that of the independent variables. With between 1.15 to 4.97 error associated with these more comprehensive models and a wide disparity in the relative explanatory power among the independent variables, it is evident that these models can be further refined to assist in constructing more statistically significant models to explain the importance of rentierism in the Arab world. Evaluating the Statistical Significance of the Rentier Pattern in the Arab World By isolating those dependent variables measuring democratization and constructing more specified models which factor the most significant rentier indicators over time, the Rentier State Theory as a viable explanation in determining national political behavior in the Arab world proves statistically insignificant. Table 3 presents the results of more focused statistical models generated on the Polity III indicator of autocracy (AUTOC) and the Freedom Survey indictor of civil liberties (LIB) which appeared the most affected by rentier Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 88 indicators over time.113 The results generated by these models include only those independent variables that maximize the relative explanatory power of rentier indicators on autocracy and civil liberties and signal the significance of tax and nontax revenue. This is particularly true when coupled with the other components such as state size, modernization, and change but not as would be assumed in the rentier state paradigm. Analyzing Autocratic Political Patterns While at first glance, the core elements of the Rentier State Theory when regressed against autocracy appear to contradict prior qualitative evidence, the actual relationship between autocracy and rentier indicators prove inversely causal when factored with a time lead in the most statistically significant model as depicted in Table 3. When regressed against autocracy (AUTOC) in the same year, the majority of identified rentier indicators account for 64.1% of autocratic patterns in the Arab world with a relatively high degree of error (Root MSE 2.312). All tax indicators, except social welfare taxes (SOCTAX) and taxes on international trade, and all of the nontax indicators, except deficit spending (VIABLE), contribute to the overall significance of the model. However, the nontax indicators, particularly those associated with external rent which are of most importance to the Rentier State Theory, appear negatively correlated to autocracy with varying levels of significance ranging between 95% to 76% certainty (P values between .053 and .242). Similarly, the anticipated relationship between those tax revenues included in the model and autocracy implied in the 113 In determining the most statistically significant or “best fit” models using both autocracy (AUTOC) and civil liberties (LIB) as the dependent variables, those rentier indicators or independent variables that failed to minimize the error associated with the model (Root MSE) and maximize the model’s statistical significance (Adj. R2) were systematically discarded. The remaining independent variables are only those which, regardless o f their own statistical significance, contributed to establishing the most significant base models. Time lead or lag was than introduced on these “best fit” base models for autocracy and civil liberties in order to test the directionality of the dependent and independent variables. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 89 Rentier State Theory appears contrary to the results of this regression, except in the case of the summary statistic of relative tax impact on state expenditures (TAXIMP) which is of lesser direct theoretical importance. These results would imply that there does not appear to Table 3: Specifying Models of Rentier Political Patterns in the Arab World Over Time Independent Dependent Variables Variables Polity III Index Freedom Survey Index LIBPl TAXIMP -.128 (-3.54)' -.123 (-3.44)’ C ^ -.035 (-3.15)' -.039 (-3.58)' INCTAX .010 (1.17)"' .009 (1.06) -.013 (-2.87)' -.014 (-3.07)' PAYTAX .113 (2.03)' .128 (2-23)* .072 (2.57)' .059 (2.12)' PROPTAX .154 (329)' .155 0-34)* .065 (2.83)' .064 (2.70)' SOCTAX .022(1.92)" -.035 (-539)' -032 (-5.03)' EXPTAX .022 (1.91)” .023 (2.10)' INTLTAX -.014 (-1.60)'” -.022 (-2.57)' NONIMP -.000 (-1.16)*' -.000 (-1 .1 8 )'" .009 (2.01)' .006 (126)'" STINCOME -.074 (-2.07)* -.070 (-1.96)" -.015 (-1.44)'” -019 (-1.88)” CAPREV -.110 (-2.59)* -.100 (-2.43)' GRANTS -.056 (-1.47)**” -.048 (-1.27)" -.013 (-1.04) -020 (-1.67)” VIABLE .096 (2.66)* .091 (2.56)' .023 (229)' .027(2.81)' GDPCAP -.000 (-7.17) -.000 (-7.14)' -.000 (-1130)* -.000 (-12.99)' CSIZE .022 (.036) P'1 -.100 (-.16) CTAXIMP 397(1.65)" .747(3.09)' CNONIMP CGDPCAP .710 (1.99)* ■ T.if-VPr'-y-": 1 .584 (1.63)" STAXIMP -.001 (-9.41)' -.001 (-9.67)' -.000 (-6.13)* -000 (-5.61)* SNONIMP -.000 (-2.05)* -.000 (-130)"* Adjusted R2 .6408 .6385 .6405 .6356 Root MSE 2312 23327 1.1512 1.1607 Constant 5.756(6.79)* 6.032(7.11)* 5.68 (18.30)* 5.471 (17.66)' Observations 802 fi| 835 811 824 NOTE: Coefficients given for indicators of the most explanatory model. T-statistics given in parentheses. F-statistic equals zero for all models. *P<.05, **P<.10,*** P<.25 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 90 be empirical evidence to substantiate the Rentier State Theory, though the relatively sizable amount of error associated with the model (Root MSE of 2.312) does weaken its explanatory power. However, when factored with a time lead, the directionality of the relationship between these rentier indicators and autocracy assumed within the model and the Rentier State Theory itself is fundamentally compromised.114 Analyzing these rentier determinantsof autocracy, therefore, is of little conclusive value when evaluating the statistical validity of the Rentier State Theory. The second regression against autocracy, the results for which are also provided in Table 3, subtracts one year from the dependent variable (AUTOC) to provide a reverse causality model (AUTOCM1). This modified model of autocracy contains less error (Root MSE of 2.3052 instead of 2.312) and is only slightly less statistically significant (Adj. R2of .6385 instead of .6408) than the same year model and is therefore, considered of more explanatory value. In essence, this model evidences that there is more of an influence of autocracy on the various rentier indicators than the inverse that is implied in the original model. While the tax and nontax indicators included in this modified model also appear to contradict the relationship hypothesized by the Rentier State Theory, the causality assumption challenged by this model make autocracy not a substantive indicator of theoretical validity in this instance. As would be expected, when further modified for a one- year time lag through the off-setting of the autocracy indicator by one year (AUTOCP1), the model does not statistically improve in its explanatory, causal or directional significance. 114 The same compromise of the directionality assumption was detected when similarly regressed against political rights (RIGHTS). The “best fit” model on political rights, holding time constant (RIGHTS), produced an adjusted R 2 value of .6101 and a Root MSE value of 1.3348, when factored with a time lead of one year (RIGHTSMI), the model improved significandy with an adjusted R 2 value of .6156 and a Root MSE value o f 1.3193. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 91 Analyzing Civil Liberties Patterns Unlike the previous series of regressions that established models of the most significant rentier indicators against autocracy over time, the second series of models in Table 3 evaluate the statistical significance of the same independent variables against the Freedom Survey’s indicator for civil liberties (LIB). This series of models prove of more explanatory value and statistical significance in evaluating the validity o f the Rentier State Theory, albeit utilizing a less structural indicator of democracy, that of individual civil liberties. While che most significant model constructed around autocracy appears to violate the causality assumption implied in the Rentier State Theory and be of marginal statistical significance, the most significant model utilizing civil liberties as the benchmark for measuring democratization eliminates causality and time lag as influential factors. When regressed against civil liberties, the indicators of the rentier state prove more enlightening with regards to statistically evaluating the Rentier State Theory. The first and most statistically significant model utilizing civil liberties as the dependent variable provides mixed results with regards to che hypothesized relationship between state revenue sources and national political behavior. In evaluating rentier patterns in the Arab world, civil liberties clearly provides the best descriptive value with less than half the degree of error found in the best autocracy model (Root MSE of 1.1488) and the highest explanatory significance (adj. R2 of .6412), accounting for 64.1% of the variance in civil liberties in chose Arab countries analyzed. This model includes all indicators of sources of state revenue (tax and noncax), except taxes on expenditure (EXPTAX) and capital asset revenue (CAPREV), modernization (GDPCAP), and a number of the “dimensional” indicators. The aggregate indicator measuring the relative impact of nontax revenue on state expenditure (NONIMP), while Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 92 directionally consistent with the Rentier State Theory, is only accurate 77.8% of the time (P value of .222), one of the least significant levels within the model. State income and grants, while of more substantially greater statistical significance (P values of .033 and .085 respectively), appear to have an inverse affect on civil liberties that would be expected. With every 2% decline in state income and grants as a portion of state revenue, there is a corresponding point increase o f repression of civil liberties on a seven point scale.115 These results would appear to contradict the direction of the relationship assumed by the Rentier State Theory, since these indicators represent the principal measures of hydrocarbon external rent (either direct or indirect) as a portion of state revenue. Given the relatively high confidence levels of these results, 97% and 91% respectively, litde statistical evidence exists in this model to substantiate the hypothesized importance of external rent as a principal determinant of democratization. The final nontax revenue source, while not direcdy associated with the Rentier State Theory, is also worth noting. Increased deficit spending (VIABLE), however, does appear direcdy correlated to decreasing civil liberties (coefficient of .030) and of reladvely high stadsdcal significance (P value of .002). While not a source of state revenue derived from hydrocarbon profits per se, borrowing does represent the common means of deficit financing and allows the state to avoid potentially politically cosdy domesdc extraction. This could, quite logically, enhance the state’s repressive nature, at least in short-term. Nevertheless, since it is not a fundamental component of nontax revenue cited in rentier literature, it is not particularly helpful in substantiating or not substantiating the Rentier State Theory in and o f itself. 115 As indicated by a coefficient or slope of -.021 for state income and -.020 for grants. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 93 While also not directly the focus of rentier literature, tax revenue represents the theoretical antithesis of the rentier state phenomenon. The tax revenue component of the model, like nontax revenues, provides similarly mixed results. The aggregate indicator of tax revenue as a component of state expenditure (TAXIMP) appears highly statistically significant (P value of .000) and inversely related to increased repression of civil liberties. With every 4% proportional decrease of tax revenues in state expenditure (coefficient of - .039), there is a point increase in the civil liberties indicator which increased repression. Conversely, it can be said that for every 4% increase in relative tax impact, civil liberties are enhanced. This bodes well for the assumed correlation between revenue extraction and individual freedoms that is captured in the adage, “no taxation with out representation” at least in the Arab world. Similarly and as anticipated, income taxes, social welfare taxes, and taxes on international trade, are also considerably statistically significant (P values of <.02) and negatively correlated with civil liberty suppression. Payroll and property taxes, however, also are similarly statistically significant but inversely correlated to civil liberties. This implies that the “domesticness” and “visability” of the tax has less to do with increasing civil liberties than does increased taxation in general. Surprisingly, GDP per capita, as the principal indicator of the contending Modernization Theory, also proves both statistically significant and negatively correlated with civil liberties suppression in the Arab world. According to this model, a slight decrease in GDP per capita (coefficient o f -.0001) increases civil liberties suppression. This is counter to what would be expected if the Arab world truly represented an anomaly in applicability of the Modernization Theory. The significance of this finding must be interpreted carefully, however. While GDP per capita is presented in a standardized measure to facilitate comparability, it is still measured nominally and therefore does not fully account for Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 94 inflation. Nevertheless, the correlation of “modernization” and “democracy” found in this model does represent another noteworthy statistical contradiction to the rentier assumption. The final three indicators included in the model serve to add dimension to the aggregate indicators for tax and nontax revenue impact on state expenditures by introducing “change” and “nonadditivity” elements. The significance of “change” in relative tax impact (CTAXIMP) and the magnifying influence of state size on both relative tax (STAXIMP) and nontax impact (SNONIMP), while they serve to enhance the descriptive power of the model, are of less direct significance on evaluating the empirical validity of the Rentier State Theory. Nevertheless, since they contribute to the overall strength of the model, they are worthy of notation. Both the change in relative tax impact (CTAXIMP) and the indicator of relative tax impact multiplied by state size (STAXIMP), which corrects for the additive assumption inherent in the regression model, prove highly statistically significant with confidence levels of 96% and 99.9% respectively. This would imply that state size exponentially effects the importance of both tax and nontax revenue, respectively. However, while the change indicator for relative tax impact was positively correlated with civil liberties suppression, the nonadditivity indicator was not. Every 48% increase in the percentage of tax revenues in state expenditure was associated with a point increase in civil liberties suppression. This could very well describe the reality that a dramatic increase in revenue extraction may necessitate increased coercion o f the citizenry on the part o f the state in order to collect. Conversely, when magnified by state size, tax impact maintains a negative correlation to civil liberties suppression. This is what would have been expected given the similar negative correlation between relative tax impact itself (TAXIMP) and the dependent variable. The nonadditivity indicator for relative nontax impact (SNONIMP), while inversely correlated to Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 95 the civil liberties indicator, is of lesser statistical significance given its relatively low confidence rating o f only 79% (P value of .209). Like relative nontax impact (NONIMP), interestingly enough, nonadditive relative nontax impact is of dubious explanatory value in this particular model. While the regression of rentier indicators against the Freedom Survey’s indicator of civil liberties produces little evidence to support the core hypothesis o f the Rentier State Theory, it is important to note the significant spatial limitations of the model’s applicability on the ten countries sampling the Arab world. This particular model describing the rentier influences on civil liberties could be considered highly statistically significant for only half of the Arab countries included in the model. The confidence level of the direction of the relationship between the independent variables and the dependent variable is particularly strong, above the 99.9% level, in the case of Oman, Syria and Tunisia . It is also highly correlated (coefficients ranging between 1.07 and 1.71) for these three countries. For Egypt and Morocco, the explanatory confidence of the model is also significant, since it is above the 95% level. Both rentier (Oman) and semi-rentier (Egypt, Morocco, Syria and Tunisia) states are included among these countries, which lends credibility to the model in describing its applicability to both oil-rich and oil-poor countries. For two other countries, namely Jordan and Kuwait, the model is of significantly less statistical value, maintaining confidence levels of only 61% (P values of .288 and .286).116 It is clear that the model is of little explanatory power in the case of Bahrain, UAE and Yemen. For these countries, there is little statistical confidence that the results depicted in the model are not just the result of chance (P values o f .775, .946 and .725 respectively). Each of these countries has significant 1,6 In the case o f Kuwait, the IS90 year was eliminated from the regression due to the invasion and occupation o f that country by Iraq. Stare finance statistics during this year were significantly disrupted during 1990 and the exiled Kuwaiti state’s finance patterns significandy skewed. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 96 anomalies in the pattern of state finance during the period under investigation, which quite possibly are the causes for the model’s limited explanatory value for them.117 With this model only accurately statistically describing half of the countries included with relative certainty, it is of only limited or qualified explanatory value for the region as a whole. Nevertheless, the observations enabled through a quantitative analysis of the link between external rent and autocracy within at least a sizable cross-section of the Arab world present a significant challenge to the core assumptions of the Rentier State Theory. In order to determine the utility of the rentier paradigm among the broader community of nation states, these same indicators were regressed including the whole cross-regional sample of Arab and non-Arab countries worldwide. Evaluating the Statistical Significance of Rentier Indicators Worldwide Since the independent variables utilized to evaluate the extent of the rentier state phenomenon in the Arab world can be equally applied to Arab and non-Arab states alike, the next series o f regressions were undertaken on the entire 39 country sample. The results depicted in Table 4 for all rentier indicators regressed against the six dependent variables measuring different, yet highly correlated, elements of democracy in order to form the basis of developing more statistically significant models from which to draw further observations. As was the case in analyzing rentier patterns in the Arab world, only those dependent 117 Over the course of the last two and a half decades, Bahrain’s hydrocarbon reserves have all but depleted. Lost hydrocarbon revenue has been somewhat countered through increased financial market activity and downstream hydrocarbon industries, including refinement of a sizable share o f Saudi oil. The unique nature of the UAE, which is comprised of a loose federation of autonomous emirates financed primarily through the redistribution of petroleum proceeds fiom the emirate of Abu Dhabi, significantly skews the pattern of state finance found in the rest of the Arab world. The UAE state is almost exclusively financed through a transfer of petroleum profits in the form of a grant fiom Abu Dhabi to the UAE state. While not a foreign grant, per se, this transfer is an entirely rentier phenomenon and therefore, treated like a foreign grant within the data set. This statistical exception coupled with the complexity inherent within the UAE state itself that necessitated that exception undoubtedly limits the applicability o f the model for UAE. The limited availability o f statistical data Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 97 Table 4: Modeling For Rentier Patterns of National Political Behavior Worldwide Independent Dependent Variables Variables Polity III Index Freedom Survey Index APOL AUTOC T DEMOC I FREE RTftHTS T.IR •f'-y-:‘Jr7;7uM7TJ:r . ' } TAXIMP 308 (3.44)* -MS (2*24)* : .159(335)" -.107 (-2.57)* -.050 (-2.04)’ -.058 (-269)" INCTAX --291 (-4.33)* -U9X3.45)* y^.l72t4& )v' .151 (5.07)" .075 (430)" .076 (4.98)" PAYTAX -.682 (-6.59)* % i42*i?i|^5®^ .501 (9.87)' 276 (9.30)" 225 (8.64)" PROPTAX -.462 (-3.36)’ -267(3.79)* -.195 (-2.68)* .387 (5.75)" 205 (5.20)" .182 (5.27)" -.030 (-.39) .096 (2.75)" .067 (327)" .029 (1.62)""" SOCTAX v v l . KiTjA'V'* 2 K •" EXPTAX -.197 (-3.00)* -x; .066;(Jt .132 (4.67)" .071 (430)" .061 (420)" INTLTAX -.240 (-3.51)" .113(3.83)" .070 (4.06)" .043 (2.84)* NONIMP .007 (.39) .001 (.09) .004 (.85) -.004 (-.79) STINCOME 083 (1.27) .019 (.58) .019 (.96) .001 (.03) CAPREV .070 (.94) 053 (1.43)— 029 (133)— 024 (1.27)" GRANTS .051 (.75) .022 (.64) .019 (.95) .003 (.17) VIABLE 101 (-1.56) .012 (-.37) -.012 (-.65) .000 (.02) GDPCAP -4.86 (-.01) 000 (-1.26) .000 (-1.78) -4.800 (-.43 CSIZE -.554 (-.69) .397 (.97) .221 (.92) .177 (.84) CTAXIMP -.436 (-.65) 069 (.20) -.014 (-.07) 084 (.47) CNONIMP .001 (.28) -.000 (-.05) .001 (-.41) -.001 (-.58) CGDPCAP .110 (-.28) .230 (-1.01) .068 (-.51) .162 (-139) STAXIMP .003 (5.37)" -.001 (-4.93) -.001 (-4.18)" -.001 (-4.85)* SNONIMP -.000 (-36) -.000 (-.12) -.000 (-.88) .000 (.76) Adjusted Rz .8887 .8710 .8469 .8543 Root MSE 27279 1.4301 .83637 .73323 Consume 9.069 (438)" 7.762(7.44)" 3.47(5.70)" 429 (8.02)* Observations 801 ■ IB 845 845 845 NOTE: Coefficients given for indicators of the most explanatory model. T-statistics are given in parentheses. F-statistic equals zero for all models. *P<.05, **P<.10,*** Pc.25 variables that produced the greatest relative explanatory power with the least error were further specified. and the disruptive merger of North and South Yemen in the early 1990s into one country and state that resulted in a brief civil war contribute the limited explanatory significance of the model. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 98 As was the case in applying the rentier indicators within the Arab world, the aggregate indicators for the Polity III index (APOL) and Freedom Survey index (FREE) appear substantially less explanatory than their component elements measuring autocracy (AUTOC), democracy (DEMOC), political rights (RIGHTS), and civil liberties (LIB). Therefore, further specified models were run only on those four elements. All four of these indicators measuring different nuances of democracy produced relatively significant models with limited degrees of error, even prior to being further specified. They also produce models with a relatively high degree of correlation, ranging from 85% to 90% in explanatory power and only 1.45 to .73 degrees of error. However, regressions run against political rights and civil liberties on the worldwide sample presented the same causality or directionality violations as did autocracy and political rights when considering only the Arab world.118 Table 5 presents the results of the first set of regressions aimed at identifying the most significant determinants of autocracy and democracy and includes the results of those models when factoring various degrees o f time lag. The development o f further specified models further illuminates the importance of tax over nontax revenue in determining national political behavior. Evaluating Autocratic Determinants Worldwide The first set of results presented in Table 5 highlight the most significant rentier determinants of autocracy around the world. This most specified and yet, statistically 118 The “best fit” model regressed on political rights, holding time constant (RIGHTS), produced an adjusted R2 value of .8483 and a Root MSE value of .83307. When factored with a time lead of one year (RIGHTSM1), the model improved significandy with an adjusted R 2 value of .8572 and a Root MSE value of .80415. Similarly, the most explanatory civil liberties model produced an adjusted R 2 value o f .8554 and a Root MSE value of .73021 and with a lead time of one year (LIBMl), improved significandy with an adjusted R2 value of .8606 and a Root MSE value of .71785. Since this evidence suggests that changes in polidcal rights and civil liberties produces an affect on the independent variables, instead o f the assumed causal reladonship of render indicators on the dependent variables, they are of limited value in understanding the forces that shape national polidcal behavior. Regressions run on autocracy (AUTOC) and democracy (DEMOC), however, do Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 99 Table 5: Specifying Models for National Political Behavior Worldwide Over Time Independent Dependent Variables Variables AUTOC I DEMOC I DEMOCPSlDEMOCPllDEMOCPlDEMOCPS I DEMOCP 1 I D E M O C P 1 I DEMOCI1 SIZE TAXIMP -.136 C-637)*r; St08;(523) .072 (3.75)’ -.015 (-.72) -.030 (-1.31)“ * INCTAX -.098 (-5.09)* -.004 (-.17) ' .018 (.74) -.022 (-2,07)*. PAYTAX -.434 (-9.53)' -222 (-3.93)’ $ -.164 (-2.59)’ -.160 C-2.r4)* -.151 (-1.92)” .176 (-2.47)* PROPTAX tv, r — 4WW.*-* vt** SOCTAX EXPTAX .092 (-5.01)* IW7 (2.03)* ,■-•.020 INTLTAX -.112 (-6.71)’ -.055 (-2.72) NONIMP STINCOME CAPREV GRANTS VIABLE GDPCAP CSIZE CTAXIMP .116 (-.37) .089 (.25) CNONIMP CGDPCAP 348 (2.36) .097 (-.39) STAXIMP .002 (5.41)* .001 (1.87)” SNONIMP Adjusted R2 .9160 .9226 Root MSE 1299 12461 Constant 2.335 (2.99)* 2.43 (2.60)* Observations i i i i 761 S jjS S 512 ISSSS NOTE: Coefficients given for indicators of the most explanatory model. T-statistics are given in parentheses. F-statistic equals zero for all models. *P<.05, **P<.10,*** P<.25 significant model retains relative state size, all but one of the indicators of taxation, a few nontax revenue indicators, relative change in tax impact, and the two nonadditive control variables on tax and nontax revenues. With an adjusted R2 value of .8693, che model can be not appear to evidence the same reverse directionality with regards to the independent variables and are Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 100 said to describe 87% of the variance in autocracy around the world with a modest amount of error (Root MSE of 1.3956). When considering autocracy worldwide, the rentier indicators overall produce a much more significant model than for the Arab world alone and one in which the results are largely contrary to what might have been anticipated, at least with regards to the direction of influence of various state revenue sources. Firsdy, the relative size of the state appears to be an important determinant within the model. Three of the four indicators containing, at least to some degree, the indicator of relative state size remained within the most specified model describing the influences of autocracy. While state size-related indicators o f relative state size (SIZE) and the nonadditive indicator of relative tax impact times relative state size (STAXIMP) appear highly significant (P values of .032 and .000 respectively), the nonadditive indicator for relative nontax impact times relative state size (SNONIMP) is notably less significant (P value of .234). Relative state size is found to be correlated with autocracy, meaning the greater the size of the state, the generally more likely the national political system within the country is to be autocratic. For every 1.7% increase in the relative size of the state, autocracy increases one degree (coefficient of .017). When multiplied by relative tax and nontax impact on state expenditures, however, size appears to be inversely related with autocracy. This reversal would appear to indicate the relative primacy of tax and nontax impact over relative state size in determining autocratic patterns. The sources of state revenue, both tax and nontax, are particularly noteworthy within the “best fit” model describing the influences of autocracy around the world. While tax revenues generally appear positively correlated to autocracy, nontax revenues evidence a therefore, of more explanatory value. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 101 negative correlation. This observation significantly challenges the common assumption concerning the importance of state legitimacy in enhancing revenue extractive capabilities. On the whole, taxation is found to be highly significant in influencing patterns of autocracy. All six of the seven possible indicators of various tax sources, as well as the aggregate measure of the relative impact of taxation in state expenditure are found to explain 99.9% of all variance in autocracy. Only social welfare taxes (SOCTAX) were found to be statistically insignificant and to not enhance the explanatory power of the model. The annual rate of change in tax impact (CTAXIMP), while also statistically insignificant, does improve the overall model’s explanatory power and was therefore included. 119 Relative tax impact, oddly enough, was the only tax indicator that did not appear to be positively correlated with autocracy. Since social welfare taxes were found to be insignificant, this would seem to suggest that state expenditures are considerably negatively correlation with autocracy, though it is hard to be certain since state expenditure was not analyzed as a separate indicator. Of those determinant taxes considered negatively correlated with autocracy, taxes on payroll and property appear the most powerful, with coefficients of .265 and .286 respectively. By and large, it can be concluded with a relatively high degree of certainty that there is little variation in the positive influence of taxation on fostering autocracy within the Arab and non-Arab world. It can be therefore hypothesized and statistically substantiated that the process of revenue extraction actually encourages autocratic governance patterns, perhaps due to the necessity of enforcing taxation compliance upon a reluctant citizenry. 1,9 W ith a P value o f only .493, little conclusive observations can be drawn from the “change” variable for relative tax impact (CTAXIMP). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 102 The influence of nontax forms of state revenue on autocracy also appears to be highly consistent throughout the world, albeit with less relative certainty than does taxation. Factoring state income (STINCOME), capital revenue (CAPREV), and deficit financing (VIABLE) into the model improved its overall significance, while relative nontax impact (NONIMP) and grants (GRANTS) did not. The effects of state income and deficit financing were found to be of greater certainty (P values at .022 and .000 respectively) than capital revenue (P value of .153). State income and capital revenues were found to exert a negative influence over autocracy, meaning that only when the relative proportion of state income and capital revenues to state revenues decreased does autocracy increase. This observation, again, would appear opposite than the state political autonomy hypothesis assumed in rentier literature. According to this model, state political autonomy does not appear to increase with increased access to non-extractive revenue sources, at least those obtained direcdy by the state through its own productive efforts and that of the sale of its capital assets. As state revenues increase relative to state expenditures, also known as increased state fiscal viability, however, so does autocracy. Here again, commonplace assumptions of the Rentier State Theory concerning the influence of nontax revenues appear converse to what would be expected. The applicability of this particular model specifying the most significant determinants of autocratic governance patterns worldwide is enhanced when considering the 39 countries under analysis. The model proves highly statistically significant at above the 99.9% confidence level (P values of <.000) for 26 o f those countries , including seven o f the ten countries sampling the Arab world. Four other countries were found to have significance above the 97% level. The model could be only considered statistically inconclusive for roughly a quarter o f those countries analyzed, namely Austria (p< .335), Bulgaria (p< .917), Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 103 China (p< .909), Indonesia (p< .066), Kuwait (p< .469), Mexico (p< .315), Norway (p< .589), UAE (p< .933) and Yemen (p< .614). Nevertheless, given the widely dispersed nature of these “exceptional” countries across diverse geographic and cultural contexts, it is safe to assume that the results of the model appear generally conclusive worldwide. Increased taxation and decreased nontax revenue resources do not statistically appear to reduce autocracy, either in the Arab or non-Arab countries, as commonly assumed in rentier literature. Evaluating Democratic Determinants Worldwide The second series of models, results also included in Table 5, were generated using the Polity III indicator of democracy (DEMOC). Unlike the most specified model of the determinants of autocracy, however, the most significant model using democracy as the dependent variable improved significandy when factored over time. This series of regressions describes the generally inverse relationship that exists between taxation and democracy, an observation that, not only sharply contradicts with frequent generalizations made in rentier literature, but also with commonplace assumptions in liberal democratic thought captured in the adage, “no taxation without representation.” When assuming a nearly simultaneous influence of the most meaningful indicators on democracy in the global sample of countries (DEMOC), the model is significant in describing 89.5% of the variance in the pattern of democracy with 1.444 degrees of error. This model is primarily comprised of all tax revenue indicators, except social welfare taxation, but also factors relatives state size (SIZE), relative change in GDP per capita (CGDPCAP), relative change in tax impact (CTAXIMP), and the nonadditivity indicator for relative tax impact times relative state size (STAXIMP). All o f the tax revenue indicators included in the model are considered highly significant (P values o f < .012) and negatively Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 104 correlated with democracy, except the aggregate tax revenue indicator (TAXIMP). This model finds increased domestic and international taxes as a share of state revenue serve not to foster democratic patterns, as commonly assumed, but actually hinder them. Only relative tax impact on state expenditures among the tax revenue indicators is positively correlated with democracy (coefficient of .108). This could be possibly be explained by a strong correlation between democracy and state expenditures not addressed in within the model but similarly experienced when regressed against autocracy. This same, highly significant positive correlation exists for relative state impact when magnified by relative state size (STAX IMP). However, relative state size itself is found to be negatively correlated with democracy, albeit with less statistical significance (P value of .134). Relative change in GDP per capita (CGDPCAP) and relative tax impact (CTAXIMP), while they contribute to the explanatory power of the model itself, are not particularly significant enough to describe with any certainty the nature of their effect (P values of .974 and .821 respectively). Together, these rentier indicators regressed on democracy, holding time static, appear to present statistical evidence that is in sharp contrast to the assumptions typically associated with liberal democratic theory and which form some of the core rentier assumptions. When factoring a delayed impact in the effects of these indicators on democratic patterns, this observation becomes only more profound. The Delayed Effects of Democratic Determinants While the most significant model for determining the fiscal influences on national political behavior appears highly explanatory, even if the results of which appear contrary to common assumptions, it only improves over time. With each added year of delayed impact of the independent variables on democracy, the model becomes of greater explanatory power and reduced fallibility. The third through fifth models presented in Table 5 describe the Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 105 results of the preferred democratic model (DEMOC) when factored with a five (DEMOCP5) and eleven year time lag (DEMOCP11) in the independent variables affecting the dependent variable, democracy. While not presented, with each added year of time lag or delayed effect to the model, its statistical significant increases and associated error decreases. This is true only up to the eleventh year o f delay. At that point the relative explanatory power of the model peaks at 92.5% with a root mean squared error (Root MSE) of 1.2253. In year twelve, however, statistical meaningfulness within the model decreases to 92.3% and the error associated rises to 1.2461. There clearly appears to be a time lag of eleven years in maximizing the affect and meaningfulness of the determinants on democratic patterns. However, at the eleventh year, the model can be further refined to produce even more significant results. By removing the indicators for relative state size (SIZE), relative tax impact (TAXIMP) and change in relative tax impact (CTAXIMP), the model further improves in explanatory power to 92.6% and decreases in error (Root MSE of 1.2196). Those tax indicators retained in the model still remain negatively correlated with democracy and of relatively high statistical significance (P value < .039), except taxes on expenditure (EXPTAX). While this indicator too is also relatively meaningful (P value < .047), it is found to be positively correlated with democracy. Relative tax impact, when magnified by relative state size, is also positively correlated but with much higher degree of significance (P value < .000). The change in GDP per capita improves the model but is limited in producing any certain effect.120 Substantive tax revenue components are found to be largely directionaliy consistent and relatively statistically certain in their effect on democracy, even 120 The results of change in GDP per capita (CGDPCAP) can be said to only be 63.8% conclusive (P value of .638) which is to high to be statistically significant. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 106 when factored over time. This is particularly the case for domestic and international sources of taxation. As with regressions against autocracy, payroll and property taxes appear to be the most consistendy substandve determiners of democracy. Unlike the “best fit” model for autocracy, however, this refined model for democracy is highly applicable to all of the remaining 37 countries included.121 The model proves extremely statistically applicable to twenty-seven of those countries regressed (P values < .000), highly significant for eight others (P values of < .032), and only moderately significant in the remaining two countries (P values < .162). The model appears suspect only in the case of Nepal and Singapore. Considering the broad scope of this analysis and the relative certainty of the core determinants of taxation, this model remains of relatively high certainty, significant explanatory power, and cross-national applicability. Juxtaposed to common assumptions about the relationship between taxation and democracy, this study finds little statistical evidence to support increased revenue extraction as a causal force in encouraging democratization. Rather, statistically speaking, evidence points much to the contrary. Most indicators for taxation, particularly those that are considered more direct, appear to be inversely related to democracy. Nontax revenues, while of lesser overall statistical significance, appear to be inversely related to autocracy. Furthermore, when considering the determinants of autocracy, as the antithesis of democracy, taxation appears to foster, not restrict, political authoritarianism. The countries of the Arab world display surprisingly similar patterns with non-Arab countries with regards to “rentier” or fiscal indicators. In this too this particular study offers challenges to common assumptions of Arab exceptionalism with regards to democratic political development. 121 Since the span of data availability is limited for bach Bulgaria and China over the last two and a half decades, they are automatically dropped from the model when a time lag o f eleven years is factored. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 5 CONCLUSION Summary of Analysis This particular inquiry sought to discover the extent of the statistical significance of the Rentier State Theory. After tracing the evolution of the theory’s primary hypotheses, quantifying its core determinants and contextual framework, and analyzing the results of multiple statistical regressions done on a number of explanatory models, no empirical evidence was found to support the core components of the Rentier State Theory. To the contrary, chere is every quantitative indication that many of the assumptions about the relationship between various sources of state revenue and national political behavior are not only statistically unverifiable but actually opposite of empirical evidence. In short, petroleum profits were not found in this study to be highly negatively correlated with participatory politics as is hypothesized and qualitatively depicted in rentier literature. When evaluating a data sample containing financial statistics on a representative sample of oil-rich and oil-poor Arab states over the course of the last two and a half decades, particularly statistical information regarding state revenue sources, little empirical evidence was found to exist to support the core hypotheses of the Rentier State Theory. Nontax revenues, particularly those commonly associated with external rent, were found to be either statistically insignificant or negatively correlated with autocratic patterns and increased suppression of individual civil liberties. Increased tax revenues, on the other hand, were 107 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 108 found co generally f o s t e r autocracy and civil liberty suppression. These findings are direcdy opposite of what would have been anticipated in prior literature concerning the role of external renc in shaping national political behavior in the Arab world. Furthermore, these empirical patterns regarding the revenue determinants of national political behavior in the Arab world were found to be similar in the non-Arab world. Challenging the uniqueness of the rentier pattern in the Arab world, an analysis of a broader sample of 39 Arab and non-Arab countries over the same time period, this pattern was found to be even more statistically significant. The few nontax revenues found to be meaningful among this broader survey similarly exhibited a negative influence upon the development of autocracy patterns. Meanwhile, almost all state revenue originating from taxation, be they direct, indirect, domestic or international, were found to be positively correlated with autocracy and negatively correlated with democracy. When factoring an eleven year time lag in caxation producing an effect on democratic patterns, this negative correlation was found to be even more significant. This study detected only slight variation between Arab and non-Arab countries with regards to these general patterns. What then can be said about the Rentier State Theory? In the words of one of its chief architects, “the dynamic relationship between rent and political order is far from being automatic .. [and].. cannot be proven in general, and is in fret not supported by experience.”122 This inquiry found Giacomo Luciani’s observation to be entirely correct. While access to external rent may dominate Arab public finance and be a modern phenomenon within the Arab world worthy of analysis, it can not be quantitatively substantiated as the principal cause of limited democratization within the region. Rather, 122 Luciani, “Resources, Revenues and Auchoritarianism in the Arab World,” in Brynen et al.. Political Liberalization and Democratization in the Arab World _, pp. 211-212. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 109 this inquiry finds empirical evidence of a much more complex relationship. Whether this relationship is the result of the simultaneous interpenetration of both state and society, as Krimley observed to be the case in Saudi Arabia, or the effect of rising public expectations, as described by Gause, external rent is not found to give license for state autonomy and the autocratic political nature autonomy inspires. Nevertheless, the limits of theory construction solely around qualitative evidence, as is the case of the Rentier State Theory, and not in conjunction with empirical evidence, is demonstrated in this inquiry to be readily apparent. Areas for Improvement and Further Study As with most empirical analyses undertaken on social scientific subjects, this study was severely restricted in a number of important areas. First and foremost, limited data availability reduced the desired scope of the inquiry to a little less than half of the Arab world, representational as it may have been. A number of important countries, such as Algeria, Libya, Saudi Arabia, Iraq and Qatar, in which rent forms a sizable share of state income, were excluded because they lacked comparable data available for those countries included in the study. Even when available, international state financial data is often not detailed enough to allow for precise analysis of narrow subjects, like external rent-derived, state income. Rather, general indicators of external rent, such as state income and grants, were required to serve as imprecise proxies in lieu of more desirable statistical data. Furthermore, as is also the case in all empirical analyses of social scientific phenomena, quantification of inherently qualitative subject matter proves a rough measurement at best. Since the analysis of several previously assumed determinants o f national political behavior patterns was the principal focus of this inquiry, at best only general observations could be Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 110 made. Despite the empirical shortcomings experienced in this analysis, however, the direction of the results and relative significance associated with them make this study o f at least marginal descriptive value. Furthermore, since challenging the underlying assumptions of the Rentier State Theory involves simply disproving the direction of the presumed determinants and not their relative strength, these quantitative limitations were of ultimately only marginal consequence for this study. This said, there is still plenty of opportunity to further analysis of the relative importance of petroleum profits in determining national political patterns in the Arab world. Expanding the spatial and temporal scope of the inquiry to include an even greater sampling of Arab and non-Arab countries over an increased range of time would no doubt significantly strengthen the universality of these results. Similarly, further specifying tax and nontax state revenue sources, including corporate verses private income taxes and external rent verses other forms of nontax revenue, should also improve the quality and accuracy of the findings. Finally, analyzing these results for problems of nonlinearity and heteroskedasticity and correcting for these empirical complexities would further improve their reliability and statistical significance. However quantitatively limited, this study proves nonetheless groundbreaking on several important fronts. Firstly, it traced the development o f the Rentier State Theory in topic literature over the last two and a half decades. Second, it isolated the core assumptions and operationalized the primary determinants involved in the theory. And finally it brought an empirical dimension to a heretofore exclusively impressionistic study of the relationship between petroleum profits and participatory politics within the Arab world. While the ‘third wave’ of global democratization may have crested without profoundly disturbing the oil-rich region Arab world, the prospects of declining oil reserves Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. I l l and increasing globalization within the region will make it less likely to effectively withstand the next. Whatever the ultimate cause, or more likely causes, of persistent authoritarian patterns within the Arab world, the rapid socio-economic changes currendy taking place within the region on all fronts at the close of this century will no doubt spill-over into the polidcal sphere. Petroleum profits giving way to participatory politics in the next century may ultimately prove more coincidental or tangential than causal in an historical perspective. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. BIBLIOGRAPHY Abdel-Fadil, Mahmoud, “The Pure Oil-Rentier States: Problems and Prospects for Development,” Oil and Arab Cooperation, vol. 5, no. 3 (1979). Almond, Ggariel A. James S. Coleman, eds. The Politics of Developing Areas. Princeton: Princeton University Press, I960. Amuzegar, Jahangir, “OPEC’s Seventh Life,” Middle East Policy, vol. 5, no. 3 (September 1997), pp. 25 -39. Anderson, Lisa, “Absolutism and the Resilience of Monarchy in the Middle East,” Political Science Quarterly, vol. 106, no. 1 (1991). Anderson, Lisa. The State and Social Transformation in Tunisia and Libya. 1830-1980. Princeton, New Jersey: Princeton University Press, 1986. Arat, Zehra, “Democracy and Economic Development: Modernization Theory Revisited,” Comparative Politics, vol. 21 (1988), pp. 21-36. Arat, Zehra. Democracy and Human Rights in Developing Countries. Boulder, Colorado: Lynne Rienner, 1991. Ayubi, Nazih, “Arab Bureacracies: Expanding Size, Changing Roles,” in Adeed Dawisha and I. William Zartman, eds.. Beyond Coercion, the Durability of the Arab State. London: Croom Helm, 1988. pp. 16 -29. Barkey, Henri, ed. The Politics of Economic Reform in the Middle East. New York: St. Martin’s Press, 1992. Bates, Robert and Da-Hsiang Donald Lien. "A Note on Taxation, Development and Representative Government," Politics and Society, vol. 14., no. 1 (1985), pp. 53-70. Beblami, Hazem and Giacomo Luciani, eds. The Rentier State. New York: Croom Helman, 1987. Bianchi, Robert. Unruly Corporatism: Associational Life in Twentieth-Century Egypt. New York: Oxford University Press, 1989. 112 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 113 Bienen, Henry and Mark Gersovitz, “Consumer Subsidy Cuts, Violence, and Polidcal Stability,” Comparative Politics, vol. 19, no. 1 (October 1986), pp. 25-44. Bienen, Henry and Mark Gersovitz, “Economic Stabilization, Condiuonality, and Political Stability,” International Organization, vol. 39 (Autumn 1985), pp. 729-754. Bill, James A., “The Study of Middle East Polidcs, 1946-1996: A Stocktaking.” Middle East Tournal. vol. 50, no. 4 (Autumn 1996), pp. 501-512. Bina, Cyrus. The Economics of Oil Crisis: Theories of Oil Crisis. Oil Rent, and Internationalization of Capital in the Oil Industry. London: Merlin Press, 1985. Black, Cyril E. The Dynamics of Modernization. New York: Harper and Row, 1966. Brynen, Rex. "Economic Crisis and Post-Rentier Democratization in the Arab World: the Case o f Jordan," Canadian Tournal of Political Science, vol. 25, no. 1 (March 1992), pp. 69-97. Brynen, Rex, Bahgat Korany, and Paul Noble, eds. Political Liberalization and Democratization in the Arab World: Volume 1. Theoretical Perspectives. Boulder, Colorado: Lynne Rienner Publishers, 1995. Boulding, Elise, ed. Building Peace in the Middle East. Boulder, Colorado: Lynne Rienner Publishers, 1994. Bollen, Kenneth, “‘If You Ignore Outliers, Will They Go Away?’: A Response to Gasiorowski,” Comparative Political Studies, vol. 20, no. 4 (January 1988), pp. 516- 522. Bollen, Kenneth, “Issues in the Comparative Measurement of Political Democracy*! American Sociological Review, vol. 45, no. 2 (June 1980), pp. 370-390. Bollen, Kenneth, “Liberal Democracy: Validity and Method Factors in Cross-National Measures,” American Tournal of Political Science, vol. 37, no. 4 (November 1993), pp. 1209-1230. Bollen, Kenneth, “Political Democracy and Timing of Development.” American Sociological Review, vol. 44 (1979), pp. 572-587. Bollen, Kenneth, “World System Position, Dependency, and Democracy: The Cross- National Evidence, American Sociological Review, vol. 48 (August 1983), pp. 468- 479. Bollen, Kennethand Robert Jackman, “Political Democracy and the Size Distribution of Income,” American Sociological Review, vol. 50 (1985), pp. 438-457. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 114 Chaudhry, Kiren Aziz, “The Price of Wealth: Business and State in Labor Remittance and Oil Economies,” International Organization, vol. 43, no. 1 (Winter 1989), pp. 101- 145. Chatelus, Michael and Y. Schmeil, “Towards a New Political Economy of State Industrialization in the Arab Middle East,” International Tournai of Middle Eastern Studies, vol. 16, no. 2 (1984). Civil Society and the Prospects for Political Reform in the Middle East. Conference Report of the Civil Society in the Middle East Project (New York University). Conference held September 30 - October 2, 1994 at the Aspen Institute Wye Conference Center, Queenstown, Maryland. Coppedge, Michael and Wolfgang Reinicke, “Measuring Polyarchy, Studies in Compartative International Development, vol. 25, no. 1 (Spring 1990), pp. 51-72. Crystal, Jill. Oil and Politics in the Gulf Rulers and Merchants in Kuwait and Qatar. New York: Cambridge University Press, 1995. Cutright, Phillips, “National Political Development: Measurement and Analysis,” American Sociological Review, vol. 28 (1963), pp. 253-264. Dahl, Robert A Polyarchy: Participation and Opposition. New Haven: Yale University Press, 1971. Deegan, Heather. The Middle East and Problems of Democracy. Boulder, Colorado: Lynne Rienner Publishers, 1994. al-Dekhayel, Abdulkarim. The State and Political Legitimation in an Oil-Rentier Economy: Kuwait as a Case Study. Ph.D. dissertation, University o f Exeter (Britain), 1990. Delacroix, Jacques, “The Distributive State in the World System,” Studies in Comparative International Development, vol. 15, no. 3 (Fall 1980), pp. 3-21. Deutsch, Karl, “Social Mobilization and Political Development.” The American Political Science Review, vol. 55 (1961). Diamond, Larry, Juan Linz and Seymour Martin Lipset. Politics in Developing Countries: Comparing Experiences with Democracy. Boulder, Colorado: Lynne Rienner Publishers, 1990. Esposito, John. The Islamic Threat: Myth or Reality? Oxford: Oxford University Press, 1992. Farsoun, Samih K., “Oil, State and Social Structure in the Middle East.” Arab Studies Quarterly, vol. 10, no. 2 (Spring 1988), pp. 155-175. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 115 Finkle, Jason and Richard Gable, eds. Political Development and Social Change. New York: John Wiley, 1966. Gasiorowski, Mark J., “Economic Dependence and Political Democracy: A Cross-National Study, Comparative Political Studies, vol. 20, no. 4 (January 1988), pp. 489-515. Gasiorowski, Mark J., “The Political Regimes Project,” Studies in Comparative International Development, vol. 25, no. 1 (Spring 1990), pp. 109-125. Gastil, Raymond D. Freedom in the World: Political Rights and Civil Liberties, various years. New York: Freedom House, 1981-1994. Gause, F. Gregory III. Oil Monarchies: Domestic and Security Challenges in the Arab Gulf States. New York: Council on Foreign Relations Press, 1994. Gellner, Ernest, “Civil Society in Historical Context,” International Social S c i e n c e Tournal. vol. 129 (August 1991). Ghabra, Shafeeq, “Democratization in a Middle Eastern State: Kuwait, 1993.” Middle East Policy, vol. Ill (1994), no. 1, pp. 102-136. Goldberg, Ellis, Rasat Kasaba, and Joel Migdal, eds. Rules and Rights in the Middle East: Democracy. Law, and Society. Seattle: University o f Washington Press, 1993. Goldstone, Jack A., Ted Robert Gurr and Farrokh Moshiri , eds. Revolutions of the Late Twentieth Century. San Fransisco: Westview Press, 1991. Gorden, Raymond L. Unidemensional Scaling of Social Variables. New York: The Free Press, 1977. Hadenius, Axel. Democracy and Development. Cambridge: Cambridge University Press, 1992. Harik, Iliya, “Rethinking Civil Society: Pluralism in the Arab World.” Tournal of Democracy, vol. 5, no. 3 (July 1994), pp. 43-56. Harik, Iliya and Denis J. Sullivan, eds. Privatization and Liberalization in the Middel East. Bloomington, Indiana: Indiana University Press, 1992. Haggard, Stephan and Robert Kaufman, eds. The Politics o f Economic Adjustment. Princeton, New Jersey: Princeton University Press, 1992. Hirschman, Albert O. Exit. Voice and Loyalty: Response to Decline in Firms. Organizations and States. Cambridge: Harvard University Press, 1970. Hirschman, Albert O . The Passions and Interests: Political Arguments for Capitalism Before its Triumph. Princeton: Princeton University Press, 1970. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 116 Hudson, Michael C., “After the Gulf War: Prospects for Democratization in the Ajrab World,” Middle East Tournal. v. 45, no. 3 (Summer 1991). Hudson, Michael C. Arab Politics: The Search for Legitimacy. New Haven: Yale University Press, 1977. Huntington, Samuel, “How Countries Democratize,” Political Science Quarterly, vol. 106. Huntington, Samuel, “Political Development and Political Decay,” World Politics, vol. 17, no. 3 (April 1965), pp. 386-430. Huntington, Samuel. Political Order in Changing Societies. New Haven, Connecticut: Yale University Press, 1968. Huntington, Samuel. The Third Wave: Democratization in the Late Twentieth Century. Oklahoma City: University o f Oklahoma Press, 1991. Huntington, Samuel and Myron Weiner, eds. Understanding Polidcal Development. Boston: Little, Brown and Company, 1987. Huntington, Samuel, “Will More Countries Become Democratic? Political Science Quarterly, vol. 99 (1984), pp. 193-218. Ibrahim, Saad Eddin. "Crises, Elites, and Democratization in the Arab World," Middle East Tournal. vol. 47, no. 2 (Spring 1993). Inkeles, Alex, “Transitions to Democracy,” Society, vol. 28, no. 4 (May/June 1991), pp. 67- 72. Inkeles, Alex, ed. On Measuring Democracy: Its Consequences and Concomitants. New Brunswick, New Jersey: Transaction Publishers, 1991. International Monetary Fund, Government Financial Statistics Yearbook. 1980-1993 editions. Washington, DC: IBRD/Word Bank, 1981-1994. International Monetary Fund. International Financial Statistics. 1980-1993 Yearbooks. Washington, DC: IBRD/World Bank, 1981-1994. Ismael, Tareq Y. and Jacqueline S. Ismael, eds. Politics and Government in the Middle East and North Africa. Miami, Florida: Florida International University Press, 1991. Jaggers, Keith and Ted Robert Gurr. Polity III Code Book. Reference # ICPSR 6695. Ann Arbor: Inter-university Consortium for Political and Social Research, 1995. Jaggers, Keith and Ted Robert Gurr, " Tracking Democracy’s Third Wave with the Polity III Data,” Tournal of Peace Reseach. vol. 32, no. 4 (1995), pp. 469-482. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 117 Krimly, Rayed Khalid. The Political Economy of Renner States: A Case Scudv o f Saudi Arabia in the Oil Era. 1950-1990. Unpublished dissertation submitted to the Columbian College and Graduate School of Aits and Sciences of The George Washington University, February 14, 1993. Lerner, Daniel. The Passing of Traditional Society. Glencoe: Free Press, 1958. Levi, Margaret. O f Rule and Revenue. Los Angeles: University of California Press, 1988. Lindenberg, Mark and Shantayanan Derarajan, “Prescribing Strong Economic Medicine: Revising the Myths about Structual Adjustment, Democracy and Economic Performance in Developing Countries,” Comparative Politics, vol 25, no. 2 (1993), pp. 169-182. Lipset, Seymour Martin, ed. Political Man: The Social Bases of Politics. New York: Doubleday. Lipset, Seymour, “Some Social Requisits of Democracy: Economic Development and Political Legitimacy,” The American Political Science Review, vol. 53 (1959). Luciani, Giacomo, ed. The Arab State. Berkely, California: University of California Press, 1990. Luciani, Giacomo. "Economic Foundations of Democracy and Authoritarianism: The Arab World in Comparative Perspective," in Arab Studies Quarterly, vol. 10, no. 4 (Fall 1988), pp. 457-475. MA.. Cook, ed.. Studies in the Economic History o f the Middle East: From the Rise of Islam to the Present Day. London: Oxford University Press, 1970. Makram-Ebeid, Mona, “Democratization in Egypt: The ‘Algeria Complex'.” Middle East Policy, vol. Ill (1994), no. 3, pp. 119-124. Marx, Karl. Capital. Volume 3. London: Lawrence and Wisart, 1972. McLaughlin, Sara, et. al., Timing the Changes of Political Structures: A new Policy Database,” Tournal o f Conflict Resolution, vol. 42, no. 2 (April 1998), p. 232. Mohamedi, Farced, “The Saudi Economy: A Few More Years till Doomsday,” Middle East Report, vol. 185 (November-December 1993). Mousseau, Michael, “Militarized Interstate Conflicts,” Tournal of Conflict Resolution, vol. 42, no. 2 (April 1998). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 118 Muller, Edward N., “Dependent Economic Development, Aid Dependency on the United States, and Democratic Breakdown in the Third World.” International Studies Quarterly, vol.29, pp. 445-470. Nelson, Joan M., ed. Fragile Coalitions: Politics o f Economic Adjustment. New Brunswick: Transaction Books, 1989. Neubauer, Deane E., “Some Conditions of Democracy,” The American Political Science Review, vol. 44 (1979), pp. 1002-1009. Niblock, Tim and Emma Murphy, eds. Economic and Political Liberalization in the Middle East. London: British Academic Press, 1993. Nonneman, Gerd. Development. Administration and Aid in the Middle East. New York: Roudedge, 1988. Nore, Peter and Terisa Turner, eds. Oil and Class Struggle. London: Zed Press, 1981. Norton, Augustus Richard, “The Future of Civil Society in the Middle East.” Middle East Tournal, vol. 47, no. 2 (Spring 1993). O’Donnell, Guillermo and Philippe Schmitter, eds. Transitions from Authoritarian Rule: Tentative Conclusions about Uncertain Democracies. Baltimore, Maryland: Johns Hopkins University Press, 1986. Pennock, Roland J. Democratic Political Theory. Princeton, New Jersey: Princeton University Press, 1979. Pipes, Daniel. In the Path of God: Islam and Political Power. New York: Basic Books, 1983. Pryce-Jones, David. The Closed circle: An Interpretation of the Arabs. New York: Harper and Row, 1989. Przeworski, Adam. Democracy and the Market Political and Economic Refors in Eastern Europe and Latin America. Cambridge, United Kingdom: Cambridge University Press, 1991. Quandt, William B., “The Middle East on the Brink: Prospects for Change in the 21st Century,” Middle East Tournal. vol. 50, no. 1 (Winter 1996), pp. 9-17. Rath, Kathrine, “The Process o f Democratization in Jordan,” Middle Eastern Studies, vol. 30, no. 3 (July 1994), pp.530-557. Ricardo, David. The Works and Correspondence of David Ricardo, vol. 1. Cambridge: The University Press, 1993. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 119 Richards, Alan, “Economic Imperatives and Political Systems,” Middle East Journal, vol. 47, no. 2 (Spring 1993), pp. 217-227. Richards, Alan and John Waterbury. A Political Economy of the Middle East. Boulder, Colorado: Westview Press, 1990. Rossant, John, “The Prince: Inside the $10 billion empire of powerful player Alwaleed,” Business Week. September, 25, 1995, pp. 89-99. Rugh, William A., “The United Arab Emirates: What are the Sources of Its Stability?,” Middle East Policy, vol. 5, no. 3 (September 1997), pp. 14 -24. Russet, Bruce M. Trends in World Politics. New York: Macmillan, 1965. Said, Edward. Orientalism. New York: Pantheon, 1978. Salamd, Ghassan, ed. Democracy without Democrats: The Renewal of Politics in the Muslim World. London: I.B. Tauris 1994. “Sand Under Kadafi is Shifting,” Los Angeles Times. September 19, 1996. Sirhan, Bassam. Modernization and Underdevelopment: A Capital-Surplus Country: Kuwait. Washington, DC: The American University, Ph. D. dissertation, 1980 (p. 150) Shaw, Paul R., “The Political Economy of Inequity in the Arab World.” Arab Studies Quarterly, vol. 6, no. 1-2 (Winter-Spring 1984), pp. 124-154. Simmons, Andre. Arab Foreign Aid. East Brunswick, New Jersey: Associated University Presses, Inc., 1981. Skocpol, Theda. "Rentier State and Shi'a Islam in the Iranian Revolution," Theory and Society, vol. II (1982), pp. 265-283. Smith, Adam. An Inquiry into the Nature and Causes of the Wealth of Nations. London: T. Nelson and Sons, 1884. Steinmo, Sven. Taxation and Democracy. New Haven, Connecticut: Yale University Press, 1993. Vandewalle, Dirk, “Qadhafi’s “Perestroika”: Economic and Political Liberalization in Libya,” Middle East Toumal, vol. 45, no. 2 (Spring 1991), pp. 216-231. Vanhanen, Tatu, “The Democratization Related to Socioeconomic Variables in 147 States in 1980-85,” Scandanavian Political Studies, vol. 12 (1987), pp. 95-127. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 120 Vanhanen, Tatu, The Process oF Democratization: A Comparative Study of 147 States. 1980-1988. Washington, D.C.: Crane Russak, 1990. Wesson, Robert, ed. Democracy: A Worldwide Survey. New York: Praeger, 1987. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.