New Frontiers in Regional Science: Asian Perspectives 21 Moriki Hosoe Editor Regional Economic Analysis of Power, Elections, and Secession New Frontiers in Regional Science: Asian Perspectives Volume 21 Editor in Chief Yoshiro Higano, University of Tsukuba Managing Editors Makoto Tawada (General Managing Editor), Aichi Gakuin University Kiyoko Hagihara, Bukkyo University Lily Kiminami, Niigata University Editorial Board Yasuhiro Sakai (Advisor Chief Japan), Shiga University Yasuhide Okuyama, University of Kitakyushu Zheng Wang, Chinese Academy of Sciences Yuzuru Miyata, Toyohashi University of Technology Hiroyuki Shibusawa, Toyohashi University of Technology Saburo Saito, Fukuoka University Makoto Okamura, Hiroshima University Moriki Hosoe, Kumamoto Gakuen University Budy Prasetyo Resosudarmo, Crawford School of Public Policy, ANU Shin-Kun Peng, Academia Sinica Geoffrey John Dennis Hewings, University of Illinois Euijune Kim, Seoul National University Srijit Mishra, Indira Gandhi Institute of Development Research Amitrajeet A Batabyal, Rochester Institute of Technology Yizhi Wang, Shanghai Academy of Social Sciences Daniel Shefer, Technion - Israel Institute of Technology Akira Kiminami, The University of Tokyo Jorge Serrano, National University of Mexico Advisory Board Peter Nijkamp (Chair, Ex Officio Member of Editorial Board), Tinbergen Institute Rachel S Franklin, Brown University Mark D Partridge, Ohio State University Jacques Poot, University of Waikato Aura Reggiani, University of Bologna New Frontiers in Regional Science: Asian Perspectives This series is a constellation of works by scholars in the field of regional science and in related disciplines specifically focusing on dynamism in Asia Asia is the most dynamic part of the world Japan, Korea, Taiwan, and Singapore experienced rapid and miracle economic growth in the 1970s Malaysia, Indonesia, and Thailand followed in the 1980s China, India, and Vietnam are now rising countries in Asia and are even leading the world economy Due to their rapid economic development and growth, Asian countries continue to face a variety of urgent issues including regional and institutional unbalanced growth, environmental problems, poverty amidst prosperity, an ageing society, the collapse of the bubble economy, and deflation, among others Asian countries are diversified as they have their own cultural, historical, and geographical as well as political conditions Due to this fact, scholars specializing in regional science as an inter- and multi-discipline have taken leading roles in providing mitigating policy proposals based on robust interdisciplinary analysis of multifaceted regional issues and subjects in Asia This series not only will present unique research results from Asia that are unfamiliar in other parts of the world because of language barriers, but also will publish advanced research results from those regions that have focused on regional and urban issues in Asia from different perspectives The series aims to expand the frontiers of regional science through diffusion of intrinsically developed and advanced modern regional science methodologies in Asia and other areas of the world Readers will be inspired to realize that regional and urban issues in the world are so vast that their established methodologies still have space for development and refinement, and to understand the importance of the interdisciplinary and multidisciplinary approach that is inherent in regional science for analyzing and resolving urgent regional and urban issues in Asia Topics under consideration in this series include the theory of social cost and benefit analysis and criteria of public investments, socio-economic vulnerability against disasters, food security and policy, agro-food systems in China, industrial clustering in Asia, comprehensive management of water environment and resources in a river basin, the international trade bloc and food security, migration and labor market in Asia, land policy and local property tax, Information and Communication Technology planning, consumer “shop-around” movements, and regeneration of downtowns, among others Researchers who are interested in publishing their books in this Series should obtain a proposal form from Yoshiro Higano (Editor in Chief, higano@jsrsai.envr.tsukuba.ac.jp) and return the completed form to him More information about this series at http://www.springer.com/series/13039 Moriki Hosoe Editor Regional Economic Analysis of Power, Elections, and Secession 123 Editor Moriki Hosoe Faculty of Economics Kumamoto Gakuen University Kumamoto, Japan ISSN 2199-5974 ISSN 2199-5982 (electronic) New Frontiers in Regional Science: Asian Perspectives ISBN 978-4-431-55895-8 ISBN 978-4-431-55897-2 (eBook) https://doi.org/10.1007/978-4-431-55897-2 Library of Congress Control Number: 2017964609 © Springer Japan KK 2018 This work is subject to copyright All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed The use of general descriptive names, registered names, trademarks, service marks, etc in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations Printed on acid-free paper This Springer imprint is published by Springer Nature The registered company is Springer Japan KK The registered company address is: Shiroyama Trust Tower, 4-3-1 Toranomon, Minato-ku, Tokyo 1056005, Japan Preface The main object of this book is to explain some of the mechanisms of integration and secession among regions from the point of view of game theory To attain this goal, the incentives of regions in a country to unite or to secede and the conditions in the way for the member countries of a federation to leave peacefully or by way of violence are examined Since 1990, the Soviet Union drastically split into 15 independent countries; Yugoslavia was divided into countries after the bloody civil war In Asia, tensions over secession have become more and more serious, including separatist tensions in China, India, Iraq, Myanmar, and Sri Lanka Why such regional conflicts occur? Are there any devices to solve them? In this book, we try to answer such questions, and study the movement of regional secession and integration in each region, like culture, ethnicity, or religion There are several important resources of power in regional conflicts One of the most important factors in the bargaining power of conflicting parties is ethnicity Ethnicity is an element that dominates the common feelings of people in many cases Therefore, committed power in regional conflicts is very important Another factor is religious fervor A religious mind and religious enthusiasm are different things However, in countries or regions in ethnic crisis situations, the two are often united We also investigate the influence of power of a third region on secession We can raise the recent consolidation by Russia of the Crimean Peninsula as an example of this third party effect Election is also an important power to solve regional conflicts Therefore, how the formats of regional election or referendum are designed is critical as some commitment devices of secession The decentralization of power in tax-subsidy policy and in the political governance form in two regions is shown to be important as a way to avoid wasteful conflict for the secession incentive of a minority region We point out that the natural resources in secessionist are often critical as bargaining power over secession conflict How those powers depend on heterogeneity costs associated with different preferences over the type of region and the relative size of the two regions is shown v vi Preface Lastly, we would like to thank the editorial board at the Japan Section of the Regional Science Association for compiling this volume as one of the series of New Frontiers in Regional Science: Asian Perspectives, in particular, Professor Yoshiro Higano, who kindly reviewed the contents of this book as editor in chief Kumamoto, Japan August 2017 Moriki Hosoe Contents The Issue of Regional Conflict and Secession Moriki Hosoe Secession, Representative Elections, and Political Governance Moriki Hosoe 13 Regional Integration and Secession with Rent-Seeking Moriki Hosoe 41 Integration and Secession Under Threat of Third Region Moriki Hosoe 61 Integration, Third Region Effect, and Policy Commitment Shigeharu Sato 79 Secession and Fiscal Policy Tadahisa Ohno 89 Voting and Secession 117 Shigeharu Sato Resource Sharing, Civil Conflict, and Secession 143 Moriki Hosoe Empirical Analyses of Secession and Integration: What Determines National Size? 161 Takashi Kihara Index 193 vii Chapter The Issue of Regional Conflict and Secession Moriki Hosoe 1.1 The Issue of Regional Conflict and Secession Regional conflicts, which have persisted from ancient times to modern times, have included ethnic wars and wars of conquest Secession has been the cause and result of regional conflicts During the last few decades, many national breakouts have occurred Aleksandar and Radan [2] provide a comprehensive guide to the secession issues The most dramatic examples were the big bang in Eastern Europe after the disintegration of the Soviet Union Even today, from Catalonia and Scotland to South Sudan and Kashmir, secessionist movements demanding independence are occurring throughout the world under all kinds of political and economic circumstances A short look at two histories in Asia is illustrative of the causes of the breakout and formation of states One is related to Bangladesh The other is related to Timor 1.1.1 Bangladesh and Timor In the sixteenth century, the Bengal region developed into the center of commerce and industry under the Mughal Empire The UK controlled most of the Bengal region as England’s East India Company was formed at the end of the eighteenth century After Britain withdrew from colonial India in 1947, districts in which Muslims constituted a majority separated and became independent as Pakistan The district in which a majority of inhabitants believed in Hinduism was separated as M Hosoe ( ) Faculty of Economics, Kumamoto Gakuen University, Kumamoto, Japan e-mail: hosoe@kumagaku.ac.jp © Springer Japan KK 2018 M Hosoe (eds.), Regional Economic Analysis of Power, Elections, and Secession, New Frontiers in Regional Science: Asian Perspectives 21, https://doi.org/10.1007/978-4-431-55897-2_1 M Hosoe India What is now Bangladesh became East Pakistan Partition was followed by war with India over Kashmir, mass migration of Muslims, and resettlement of Hindus and Sikhs within the new borders This led to violence, financial loss, and death on a large scale Pakistan therefore comprised East Pakistan and West Pakistan, geographically separated by 1600 km Muslims lived in both places, but economic and political disparities between the two regions spread As a result, two regions had been under serious conflict Since the Urdu language used in West Pakistan Province became the national language, the civil war began The Awami League, based in East Pakistan with a high population, was eventually victorious In December 1974, the independence of Bangladesh was acknowledged Several million people were killed or displaced by the India-Pakistan war and the Bangladesh war for independence Given this history, the main causes of independence and secession of Bangladesh are the differences in language, geographical remoteness, economic disparity, and Western colonialism Let us raise the case of Timor-Leste as another example of regional conflict in Asia Since the sixteenth century, Timor Island was occupied by the Netherlands in the western half and Portugal in the eastern half After World War II, East Timor continued as a Portuguese colony after the former Dutch territory became independent as Indonesia The independence movement also strengthened in East Timor, as a result of the democratization of Portugal East Timor declared its independence in 1975 However, the independence faction and a conspiracy with Indonesia conflicted, erupting into civil war Reportedly, about two million people were killed The Indonesian Soeharto administration came to consolidate this region However, in 1999, following the United Nations-sponsored act of selfdetermination, the referendum had a majority in favor of independence, and East Timor established the independence through intervention of a multinational force of the United Nations In this case of Timor-Leste, ethnic conflict or religious conflict was not necessarily a main cause of regional conflict Development of nationalism was promoted since Portuguese colonization Severe poverty under Indonesian repression provided the main momentum driving the push toward independence 1.1.2 Quebec and Crimea In modern democratic societies, regional conflicts that arise with secession are resolved in most cases by peaceful means, contrasting against bloody conflict in regions in Asia These controversies are subjected to ardent parliamentary deliberations and become important issues in parliamentary elections or referendums These are often brought to judicial decisions For example, regarding the independence issue in Quebec, Canada, the central government had asked the court to rule on the legality of a unilateral Quebec declaration of independence after Quebec implemented the referendum The Supreme Court of Canada ruled in 1998 that neither the Canadian constitution nor international law allows Quebec to secede from Canada unilaterally It also showed Empirical Analyses of Secession and Integration: What Determines National Size? 179 in contrast, significant but negative As were the results of preceding literature, increases in both “openness” and “country size” have positive effects on the per capita GDP growth rate, whereas the interactive terms of “openness” and “country size” have negative effects on growth These results are consistent with those reported by Alesina, Spolaore, and Wacziarg [4] and the theoretical model explained in Sect 9.3 Eq (9.21) “The effect of country size on economic growth should be mediated by the degree of openness.10 ” The coefficients of other control variables are consistent with neoclassical growth theory The “initial level of GNI per capita” and “total fertility rate” have significant and negative effects on the growth rate, whereas “gross fixed capital formation per GDP” has a significant and positive effect on growth.11 9.5.3.2 Total Population as the Country Size Table 9.7 shows another set of estimation results of the “the growth rate of GDP per capita” (%) The country size variable used in these estimations is Ln (total population) The estimation method is the same as that used for the previous regression model in the subsection above Unbalanced panel data are from 162 to 192 countries, depending on data availability, during 1960–2014, the data of which are averaged over 11 nonoverlapping 5-year periods The signs and significances of all relevant coefficients in the estimations presented above are the same as those of the growth estimations by Ln (real GDP) as the variable for the country size The relation between country size (in terms of real GDP or population) and growth (of GDP per capita) depends on the trade regime Small countries can prosper in the world of free trade if they adopt open policy, while in a world of trade restrictions, larger countries have an advantage, because they have larger markets 9.6 Empirical Analyses of Territorial Armed Conflicts 9.6.1 Introduction Territorial armed conflicts have interesting characteristics and effects The total number of armed conflicts aimed at the independence of a territory leads to different developments from the total number of armed conflicts, including civil and international wars 10 Alesina, Spolaore, and Wacziarg [4], p.1278 Contrary to the empirical results of Alesina, Spolaore, and Wacziarg [4] where government size is negatively related to per capita growth, the government consumption per GDP is not negatively but positively and significantly related to the per capita GDP growth rate in specification 11 180 T Kihara Table 9.7 2SLS panel estimation for the growth rate of GDP per capita (country size is measured by Ln (total population)) Spec 9.404*** ( 2.71) Constant Ln (GNI per capita (one-period lag) Openness (trade/GDP ratio) Openness country size Country size (Ln (total population)) Total fertility rate Gross fixed capital formation/GDP Government consumption/GDP Adjusted R2 Countries/obs 0.095*** (2.63) 0.005** ( 2.49) 0.663*** (3.36) Spec 7.438* ( 1.74) 0.550*** ( 3.17) 0.111** (2.56) 0.006** ( 2.24) 0.767*** (3.68) Spec 8.000* (1.79) 1.493*** ( 10.87) 0.094** (2.34) 0.005** ( 2.12) 0.519** (2.27) 1.155*** (8.19) Spec 5.155 (1.40) 1.220*** ( 6.45) 0.050*** (2.65) 0.003** ( 2.36) 0.401*** (2.69) 0.970*** ( 5.37) 0.103*** (2.59) 0.050 192/1409 0.135 186/1165 0.215 186/1159 0.335 179/1075 Spec 3.162 (0.77) 1.186*** ( 6.16) 0.053** (2.51) 0.003** ( 2.13) 0.442*** (3.00) 0.929*** ( 4.84) 0.114*** (3.26) 3.456* (1.75) 0.361 162/995 (Note) t statistics based on heteroskedastic-consistent (White-robust) standard error are in parentheses *, **, and *** represent significant levels at 10%, 5%, and 1%, respectively Number of Armed conflicts 60 50 40 30 20 10 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 00 03 06 09 12 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20 20 20 Year Territorial Incompatibility(excl extrasystemic armed conflicts) Total No of conflicts Fig 9.3 Number of armed conflicts (total and territorial incompatibility) (Source: UCDP/PRIO [15] and author’s calculation) Figure 9.3 above presents the number of armed conflicts beginning or continuing each year “Total No of conflicts” shows all armed conflicts listed in UCDP/PRIO Armed Conflict Dataset (Version 4-2014a) [15] “Territorial Incompatibility (excl extra-systemic armed conflicts)” are those civil conflicts aimed at occupation of a Empirical Analyses of Secession and Integration: What Determines National Size? 181 territory The number of all armed conflicts increased toward the end of the Cold War (1991) but then decreased up to 2013, although the number of “territorial conflicts” has remained stable at around 10–20 cases per year throughout post-World War II period In this section, we present our investigation of the factors causing “territorial conflicts,” which might engender secession and increase the number of nations of the world Tobit (censored) regression technique is used with panel data from more than 200 countries/territories over more than 50 years Regression results obtained from an estimation of territorial conflicts are compared with those for all armed conflicts to elucidate factors that are peculiar to civil wars aimed at occupying territory After reviewing the relevant literature in the following Sect 9.6.2, we explain in Sect 9.6.3 the data and regression method used in this section Regression results and observations are presented in Sect 9.6.4 9.6.2 Literature on Territorial Armed Conflicts An increasing amount of the theoretical and empirical literature has described examination of the determinants of secessionism and/or civil wars aimed at occupation of territories in a country (1) In a recent study, Morelli and Rohner [9] highlighted the importance of natural resource concentration and ethnic group concentration related to ethnic conflicts Theoretical models identified that bargaining failure occurs (a) when the share of natural resources located in the powerless region is large and (b) when the powerless group has a substantially large winning probability in secessionist rather than in all-out conflicts The results of OLS and logit regressions conducted both at the country level and at ethnic group level confirm the fundamentally important role of geographic concentration variables, i.e., oil GINI and regional resource concentration, as well as interaction term with variables which increase the winning probability Civil war onset is also more likely in countries having (c) lower per capita GDP, (d) less democracy, (e) more population, and (f) a higher degree of ethnic fractionalization (2) Data for the subnational level have been examined recently, as described in the literature Sorens [11] analyzed secessionism risk factors at the substate or regional level in advanced democratic countries Using cross-sectional and panel data of regions, Tobit regression revealed that (a) regional language, (b) history of independence, (c) relative affluence, (d) population, (e) multiparty political system, and other factors explain secessionist strength in voting (3) As another example of empirical analyses using data for the subnational level, Brown [6] contributed to the disaggregation of the econometric study of civil war by particularly addressing secessionist conflict as a distinct subset of civil wars and using a subnational dataset The dependent variable is armed conflict coded as “territorial incompatibility,” with exclusion of extra-systemic anti-colonial wars, which corresponds to the armed conflicts for secession Using panel data, multilevel 182 T Kihara logit regressions and the Cox proportional hazard regression revealed that increases in measures of (a) ethnic diversity, (b) hydrocarbon deposits, (c) democracy (Polity IV index), (d) horizontal inequality, and (e) previous attempts at secession raise the risk of secession, whereas an increase in (f) GDP per capita reduces the risk of secession (4) As an example of theoretical models used to explain secession, Spolaore [13] presented the ideas and concepts related to the economics-based approach to conflict and national borders within an analytical framework in which populations engage in conflict over borders and resources and where the territory is divided according to military strength Game-theoretic analyses revealed that (a) a bilateral nonaggression pact allows net saving in defense spending but at the cost of less effective protection against external aggression, (b) a political union provides cheaper and better protection, (c) political unions might break up with higher heterogeneity costs and less economic returns to defend, and (d) domestic conflict multiplies heterogeneity costs and requires higher returns to form a political union The analyses explained by Spolaore [13] suggest that (i) an increase in military capacity facilitates conflicts, (ii) the relation between defense spending and country size is complex because of the existence of international alliances, and (iii) an increase in heterogeneity costs or a decrease in economic returns (e.g., per capita income or per capita GDP growth) prompts the breakup of a political union and territorial conflicts Spolaore [12] provided a study of the interconnections among country openness, international conflicts, and the size of nations Smaller countries tend to be more open and less likely to engage in conflict Furthermore, in the world of high openness and low international conflict, the political size of nations will matter less, therefore justifying smaller states (and a larger number of countries by secession) in equilibrium The following sections present an empirical examination of the theoretical propositions above We use panel data for more than 200 countries and territories and during 1961–2014 9.6.3 Data and Regression Method The existence of (a) civil conflicts aimed at territories and (b) armed conflicts in general is regressed on various explanatory variables They include (i) national size in terms of population and GDP, (ii) economic return in terms of income level and growth, (iii) political democracy, political rights, and civil liberty, (iv) heterogeneity in terms of linguistic fractionalization, (v) openness of the country in terms of trade/GDP ratio, and (vi) military strength in terms of military expenditure/GDP Empirical Analyses of Secession and Integration: What Determines National Size? 9.6.3.1 183 Data For the estimation of “civil conflicts aiming at territories,” the dependent variable is the number of armed conflicts coded by “territorial incompatibility,” with the exclusion of “extra-systemic” anti-colonial wars, based on UCDP/PRIO Armed Conflict Dataset (Version 4-2014a) (hereinafter, UCDP/PRIO) [15] Each territorial conflict is assigned If there are three territorial conflicts in the same country-year, that country-year is assigned If no territorial conflict takes place in a country-year, a zero is assigned For the estimation of “all armed conflicts,” the dependent variable is the number of all armed conflicts listed in UCDP/PRIO in respective country-year The means of counting the conflicts are the same as that used for the estimation of territorial conflicts Data for most independent variables used for both regressions, including GNI per capita, population, GDP, trade/GDP, military expenditure/GDP, per capita GDP growth rate, and GINI coefficient, were referred from the World Development Indicators Databank online [16] The independent variable used for the political democracy is the “polity2” index of the Polity IV dataset [10], which combines the polity score, with values of C10 (strongly democratic) to 10 (strongly autocratic) Indices used for “heterogeneity” and for “political rights and civil liberty” are, respectively, “linguistic fractionalization” and “Freedom House index.” These data are explained in Sect 9.4.1 For all variables, the values averaged over the nonoverlapping 5-year period are used for estimation 9.6.3.2 Regression Method Dependent variables are “censored” above zero and have multinomial values, depending on the number of conflicts in the respective 5-year periods The majority of the countries in the dataset have never experienced armed conflict The minimum possible score for the dependent variable is zero “Tobit (censored) regression” is therefore used as the standard procedure for censored data to estimate both “territorial conflicts” and “all armed conflicts”, using panel data In the same context, Sorens [11] used Tobit regression to estimate the voting share for secession 9.6.4 Empirical Results for Armed Conflicts Tables 9.8 and 9.9 respectively present the results of Tobit regressions to estimate for the number of “territorial conflicts” and “all armed conflicts.” Comparing the estimated coefficients on each explanatory variable in the estimation of “territorial conflicts” with those in the estimation of “all armed conflicts,” we observe the following commonalities and differences between Tables 9.8 and 9.9 Military expenditure/GDP Per capita GDP growth rate Freedom House index (PRCCL/2) (FH index)2 (Trade/GDP)2 Linguistic fractionalization Trade/GDP Ln(GDP) Ln(Population) Polity IV(polity2) Ln(GNI per capita) Constant 1.189*** (5.32) Spec 9.023*** ( 10.62) 0.172*** ( 3.53) 0.022** (2.10) 0.504*** (11.46) 0.500*** (11.36) 1.181*** (5.28) Spec 8.967*** ( 10.53) 0.672*** ( 9.76) 0.022** (2.10) 1.104*** (4.85) 0.0154** (2.08) 8.31E-5* ( 1.867) Spec 10.070*** ( 9.63) 0.196*** ( 3.72) 0.021** (2.06) 0.546*** (10.37) 0.065*** (5.00) 1.568*** (4.51) Spec.4 11.432*** ( 8.75) 0.189*** ( 2.69) 0.042*** (2.62) 0.617*** (9.40) 0.010 ( 0.59) 1.170*** (5.26) Spec 9.088*** ( 10.55) 0.189*** ( 3.81) 0.025** (2.39) 0.516*** (11.37) 0.132* (1.69) 1.160*** (4.52) Spec 10.199*** ( 9.52) 0.140** ( 2.35) 0.045** (2.28) 0.524*** (10.64) 0.752*** (3.46) 0.078*** ( 3.09) 1.114*** (4.39) Spec 11.611*** ( 9.62) 0.075 ( 1.22) 0.037* (1.91) 0.525*** (10.67) Table 9.8 Tobit (censored) estimation for Territorial Conflicts(Dependent variable: territorial incompatibility (extra-systemic wars excluded) 1.647*** (4.22) Spec.8 8.550*** ( 5.81) 0.243*** ( 2.89) 0.059*** (3.02) 0.561*** (7.89) 184 T Kihara 557.4 0.863 1306 1961–2014 559.2 0.866 1306 1961–2014 549.7 0.863 1293 1961–2014 322.8 0.839 786 1986–2014 552.4 0.857 1305 1961–2014 485.9 0.856 1152 1971–2014 480.8 0.849 1152 1971–2014 0.035*** ( 2.94) 228.8 0.900 524 1981–2014 (Note) Tobit regressions were conducted using data of 215 countries and territories during 1961–2014 (averaged over 11 nonoverlapping 5-year periods), at maximum z-values are in parentheses *, **, and *** represent significant levels at 10%, 5%, and 1%, respectively Log likelihood AIC Countries/obs Period GINI coefficient Empirical Analyses of Secession and Integration: What Determines National Size? 185 Military expenditure/GDP Per capita GDP growth rate Freedom House index (PRCCL/2) Linguistic fractionalization Trade/GDP Ln(GDP) Ln(Population) (Polity IV)2 Polity IV (polity2) Ln(GNI per capita) Constant 0.890*** (5.61) Spec 6.422*** ( 11.39) 0.170*** ( 4.43) 0.010 (1.23) 0.007*** ( 4.37) 0.420*** (13.62) 0.422*** (13.67) 0.878*** (5.53) Spec 6.428*** ( 11.42) 0.596*** ( 11.45) 0.010 (1.28) 0.007*** ( 4.26) 0.973*** (6.02) 0.005*** ( 2.95) Spec 5.527*** ( 8.98) 0.125*** ( 3.01) 0.010 (1.28) 0.007*** ( 4.72) 0.364*** (10.57) 0.052*** (5.03) 1.090*** (4.77) Spec.4 8.012*** ( 9.78) 0.242*** ( 4.42) 0.025** (2.17) 0.007*** ( 2.79) 0.528*** (11.94) 0.041*** ( 3.48) 0.869*** (5.49) Spec 6.637*** ( 11.61) 0.168*** ( 4.41) 0.010 (1.29) 0.006*** ( 4.21) 0.436*** (13.82) Table 9.9 Tobit (censored) estimation for All Armed Conflicts(Dependent variable: all armed conflicts) 0.265*** (4.41) 0.766*** (4.37) Spec 7.551*** ( 10.92) 0.178*** ( 3.98) 0.063*** (4.31) 0.004*** ( 2.01) 0.419*** (12.64) 1.098*** (3.81) Spec.7 7.856*** ( 6.95) 0.186** ( 2.47) 0.048*** (2.77) 0.008** ( 2.48) 0.520*** (9.42) 186 T Kihara 864.5 1.335 1306 1961–2014 863.7 1.333 1306 1961–2014 851.3 1.329 1293 1961–2014 507.5 1.312 786 1986–2014 858.2 1.328 1305 1961–2014 758.2 1.330 1152 1971–2014 0.011 ( 1.23) 347.0 1.355 524 1981–2014 (Note) Tobit regressions were conducted using data of 215 countries and territories during 1961–2014 (averaged over 11 nonoverlapping 5-year periods), at maximum z-values are in parentheses *, **, and *** represent significant level at 10%, 5%, and 1%, respectively Log likelihood AIC Countries/obs Period GINI coefficient Empirical Analyses of Secession and Integration: What Determines National Size? 187 188 T Kihara “GNI per capita” is one indicator of income level and economic return In both estimations for territorial conflicts and all armed conflicts, an increase in GNI per capita has a significant and negative correlation with armed conflict Not only territorial but all armed conflicts are prone to occur more often in countries with lower income levels This result is consistent with those of the existing literature “Polity IV” index is a measure used to gauge the degree of democratization of the country’s political system For the estimation of territorial conflicts, the coefficients of Polity IV index are significant and positive, which means that territorial conflicts are prone to occur more often in countries with more democratized regimes This result is consistent with those reported in the relevant literature For the estimation of all armed conflicts, however, Polity IV index has a quadratic relation with the occurrences of all armed conflicts, instead of a linear relation, which is observed in regressions for territorial conflicts In most model specifications, the peaks of inverse U-shaped quadratic functions are estimated when the Polity IV indices are from zero to three, close to the middle level of democracy Unlike territorial conflicts, armed conflicts in general increase when autocratic political systems become democratic ones or when democratic political systems revert to less democratic ones This might be a result that is consistent with recent heated conflicts in the Middle East “Population” and “GDP” are indicators used to gauge the “size of a nation.” In both estimations for territorial conflicts and all armed conflicts, the population size has a significant and positive relation to armed conflict Not only territorial but all armed conflicts are prone to occur more often in more populous countries The sign, significance, and magnitude of coefficients estimated using the total GDP as the “size” variable are similar to those of coefficients estimated using the total population The size of nations is related positively to the frequency of armed conflicts in general “Linguistic fractionalization” is an indicator of national “heterogeneity.” Increased national linguistic fractionalization has a significant and positive relation with both territorial conflicts and all armed conflicts Not only territorial but all armed conflicts are prone to occur more often in countries where languages are more fractionalized “Trade/GDP (%)” is a proxy for national “openness.” Trade/GDP has a quadratic relation with the incidences of territorial conflicts.12 The level of trade/GDP ratio which maximizes the incidences of territorial conflicts is 93%, which is close to the sample mean (78.3%) of the ratio It might mean that countries moving from closed to more open trade are prone to have territorial conflicts because some regions of a country might come to exist as sovereign states by enjoying the benefits of international trade Unlike regression coefficients for territorial conflicts, trade/GDP shows a linear and negative relation with the incidence of all armed conflicts Armed conflicts including international wars are less likely to exist when 12 When only the linear term of the trade/GDP ratio is included as an explanatory variable, the estimated coefficients are positive but not significant Empirical Analyses of Secession and Integration: What Determines National Size? 189 countries are more open to trade The benefits enjoyed in the global markets outweigh the possible benefits that might accrue from armed conflict “Military expenditure/GDP (%)” is a proxy for “military capacity.” An increase in military expenditure/GDP has a significant and positive correlation not only with territorial conflicts but also with armed conflicts in general Increased military capacity might lead to more conflicts, which is consistent with results reported in the existing literature “Per capita GDP growth rate (%)” is another indicator of economic return In the estimation for territorial conflicts, the coefficient of per capita GDP growth rate is positive but not significant The economic growth rate is not related to the incidence of territorial conflict Contrary to results of the regression analyses for territorial conflicts, its coefficients are significant and negative in the estimations for all armed conflicts Better economic conditions might reduce the incidence of armed conflict in general “Freedom House index” shows the degree of political rights and civil liberty It is one indicator used to evaluate policy and a nation’s institutional environment The index is a value from (best) to (worst) The Freedom House index has a significant and positive linear relation with the incidence of territorial wars and with that of armed conflict in general Deterioration of policy and institutional environments might increase the probability of armed conflict “GINI coefficient” is an indicator of income inequality among individuals in a country Contrary to intuition and theoretical literature, the GINI coefficient has a significant but negative correlation with territorial conflict Greater income inequality among people makes minorities less capable of fighting for territory In the estimation for all armed conflicts, its coefficients are negative but not significant Income inequality might not increase the incidence of armed conflict, including international wars In sum, lower income, larger national size in terms of population and total product, greater linguistic heterogeneity, larger military force, and deterioration of policy and institutional environments are associated with more frequent armed conflicts, including territorial as well as international conflicts These factors commonly increase the probability of the occurrence of armed conflict.13 The peculiarities of “territorial conflicts” are apparent in its estimated coefficients: “Polity IV,” “trade/GDP,” “per capita GDP growth rate,” and “GINI coefficient.” Territorial conflicts, but not armed conflicts in general, are prone to occur more often in countries with more democratized regimes, in countries moving from a closed economy to a more open economy and in countries where income is more equally distributed 13 Tobit estimations for the armed conflicts except territorial conflicts also reveal almost the same results as those for all armed conflicts in significances and signs of the estimated coefficients of the independent variables, although the results for which are not listed here 190 T Kihara 9.7 Conclusion The secession and integration of nations is related closely to how the increase or decrease in national size improves or degrades the economic efficiency and performance Alesina and Spolaore [3] and other reports from the literature describe that (i) country size derives from a trade-off between economies of scale in large countries and costs of cultural and ethnic heterogeneity, which might be increasing in the size of countries Additionally, (ii) small countries tend to be more open to international trade for securing economies of scale Moreover, (iii) economic performance measured in terms of per capita GDP growth rates is enhanced by the “country size” and the “trade openness” of a nation A smaller nation can reap economic benefits from a more open trade regime (i.e., the returns with respect to country size become less and less important as countries become more open) With reference to related literature, and using panel data for more than 160 countries during 1960–2014, this chapter described an empirical examination of the effects of the country size measured by total population or real GDP on government expenditure per GDP as a proxy for the inefficient provision of public goods, on the trade volume per GDP as a proxy for trade openness, and on the growth rate of GDP per capita as a proxy for economic performance Data are averaged over 11 nonoverlapping 5-year periods Empirical results presented in this chapter are mostly consistent with those reported in the literature obtained using cross-section OLS or data before 2000 Increased country size, population density, and the urban population ratio raise the efficiency of the public goods provision and have significant and negative effects on government consumption per GDP Increases in both the country size and trade openness have significant and positive effects on the growth rates of GDP per capita, whereas the coefficient reflecting the interaction of the country size and the trade openness in the growth regression is significant but negative As globalization and free trade proceeds, the secession of large countries and the creation of small countries are expected to become increasingly common This study also finds empirically that religious and linguistic fractionalizations raise costs of heterogeneity in a nation and have significant and positive impacts on government consumption and public educational expenditure per GDP Moreover, results show that economies of scale in supplying public goods can be an important benefit to countries with populations as large as 570 million people, which means that every nation, except for China and India, can benefit from population expansion Additional empirical studies are necessary to identify factors causing civil conflict and defense expenditures which “have historically played a central role in the determination of national borders,14 ” secession, and the size of nations Some determining factors to cause civil conflicts aimed at occupying territories in a county 14 Spolaore [13] p.763 Empirical Analyses of Secession and Integration: What Determines National Size? 191 are also examined in this chapter.15 Tobit regressions revealed that, in addition to the common factors causing armed conflicts in general, territorial conflicts are more likely to occur in the countries with more democratized regime, in transition from closed to open economy, and/or with more equal income distribution References Alesina, A (2003) The Size of Countries: Does it Matter? Journal of European Economic Association (2–3), 301–316 Alesina, A., Devleeschauwer, A., Esterly, W., Kurlat, S., & Wacziarg, R., (2003) Fractionalization Journal of Economic Growth 8, 155–194 Alesina, A and Spolaore,E., (2003) The Size of Nations, Cambridge, MIT Press Alesina, A., Spolaore, E., & Wacziarg, R., (2000) Economic Integration and Political Disintegration American Economic Review, 90 (5), 1276–1296 Alesina, A, & Wacziarg, R., (1998) Openness, Country Size and the Government Journal of Public Economics 69, 305–321 Brown, G K., (2010) The Political Economy of Secessionism: Inequality, Identity, and the State, Bath Papers in International Development 9, 1–29 Feenstra, R C., Inklaar R., & Timmer, M P., (2013) The Next Generation of the Penn World Table (Penn World Table, Version 8.0) available for download at www.ggdc.net/pwt Freedom House (2015) Freedom in the World Country Rating (1972–2014) (online) Morelli, M., & Rohner, D., (2015) Resource Concentration and Civil Wars Journal of Development Economics 117, 32–47 10 Polity IV Project (2014) Polity IV dataset version 2013, Center for Systematic Peace 11 Sorens, J., (2005) The Cross-sectional Determinants of Secessionism in Advanced Democracy Comparative Political Studies 38(3), 304–326 12 Spolaore, E., (2004) Economic Integration, International Conflicts and Political Unions Rivisita di Politica Economica 94, 3–15 13 Spolaore, E., (2009) National Borders, Conflict and Peace In M.R Girfinkel & S Skaperdas The Oxford Handbook of the Economics of Peace and Conflict Chapter 30, (pp.763–786) New York, Oxford University Press 14 United Nations (2015) United Nations Homepage 15 Uppsala Conflict Data Program (UCDP) & International Peace Research Institute, Oslo (PRIO) (2014) the UCDP/PRIO Armed Conflict Dataset version 4-2014a 16 World Bank (2015) World Development Indicators Databank (online) 15 The estimated model for territorial conflicts in Sect 9.6 does not assume any strategic behavior Strategic game settings may be assumed to be the background theory of the estimated models Index A a natural resource, 92 a secession economy, 93 an integrated economy, 92 an internal exit, 90 anti-separation law, armed conflict, armed conflicts, 179 B bargaining, 16, 53, 67, 145 bliss point, 15, 63, 80, 83, 85–87 bliss point , 85 C Catalonia, 117 centralized integration, 8, 29 citizen candidate model, 5, 10, 13, 44 civic war, 144 civil conflicts, 180 civil war, 9, 13, 145 coase theorem, 42 coase thiorem, 25 commitment device, conflict cost, 154 conflict expenditure, 145 conflict regime, 145 conflict-preventing condition, 144, 149, 152 contest model, cost padding, 44, 55 country size, 162, 165, 166, 170, 177, 178 D Dawns–Hotelling model, 118 decentralized integration, 8, 35 democratic procedure, 117, 118, 141 E economies of scale, 162 education, 172 election, ethnicity, 4, EU, 117 F federalism, 7, 14, 35, 144 first-mover, 66, 76 fiscal federalism, fractionalization, 183, 188 Freedom House, 183, 189 G GDP, 188 GINI coefficient, 189 GNI per capita, 188 government, 117–119 government consumption, 170 government expenditures, 170 government size, 165 growth, 166 © Springer Japan KK 2018 M Hosoe (eds.), Regional Economic Analysis of Power, Elections, and Secession, New Frontiers in Regional Science: Asian Perspectives 21, https://doi.org/10.1007/978-4-431-55897-2 193 194 H heterogeneity, 162, 183 I income tax policy, 90 integration, 13, 41, 118, 122, 131, 134, 135 integration conflict, 80, 83, 86, 87 L local public good, 119 local public goods, 118 M majority region, 15, 80 median voter theorem, 19, 21 military expenditure, 175, 189 minority region, 16, 80 monitoring accuracy, 48 N Nash equilibrium, 21, 64, 81, 131, 132 natural resource, non conflict regime, 144, 148 Index referendum, 2, referendums, 117, 119, 125, 130, 131, 136, 139–141 regional economics, 118 regional transfer policy, 90 rent seeking, 42, 57 rent-seeking, 10 representative election, 19, 20 resource abundance, 143 resource sharing, 143 S scale economies, 161 scale merit, 76, 151 Scotland, 117, 119 secession, 13, 41, 119, 121–125, 127–141, 145 second-mover, 66, 76 single-peaked preferences, 80 Spain, 117 spillover effect, 49 strategi complement, 148 T territorial armed conflicts, 179 territorial incompatibility, 180 the UK, 117 threat point, 81 three-region model, 62, 79, 80 trade, 161, 188 trade openness, 163, 177, 178 transferrable utility, 81 P penalty sharing rate, 150 per capita tax, 119–121, 126, 132 per-capita GDP growth rates, 178, 189 policy commitment, 79, 80, 83, 86 policy cordination, 18 policy externality, 14, 93 policy preference, 13, 15, 19 policy-cordination, 39 political economy, 118 political governance, 8, 13 Polity IV, 183, 188 population, 172, 179, 188 public policy, 90 V vote, 131, 137, 138 voting, 117–119, 122–128, 131, 132, 136–138, 140, 141 R real GDP, 178 W winning probability, 144, 148 U UK, 117 ... 1.2 Power of Conflicting Parties in a Secession Movement The title of this book is The Regional Economic Analysis of Power, Elections, and Secession What is power in the context of secession? ... Issue of Regional Conflict and Secession Moriki Hosoe Secession, Representative Elections, and Political Governance Moriki Hosoe 13 Regional Integration and Secession. .. development and refinement, and to understand the importance of the interdisciplinary and multidisciplinary approach that is inherent in regional science for analyzing and resolving urgent regional and