The origins and strengths of regional parties

British Journal of Political Science

Volumn 38, Issue 1, 2008, Pages 135-159

  Brancati, Dawn ^a  

^a HARVARD UNIVERSITY   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 36949028158     PISSN: 00071234     EISSN: 14692112     Source Type: Journal    
DOI: 10.1017/S0007123408000070     Document Type: Article

References (89)

1. Lieven de Winter, ed., Regionalist Parties in Western Europe (New York: Routledge, 1998);
2. Michael Keating, The New Regionalism in Western Europe (Northampton, Mass. : Edward Elgar, 1998) ;
3. Pieter Van Houten, 'Regional Assertiveness in Western Europe: Political Constraints and the Role of Party Competition' (doctoral dissertation, University of Chicago, 2000).
4. Dawn Brancati, 'Pawns Take Queen: The Destabilizing Effects of Regional Parties in Europe', Constitutional Political Economy (Special Issue) 16 (2005), 143-59;
5. Dawn Brancati, 'Decentralization: Fueling the Fire or Dampening the Flames of Ethnic Conflict and Secessionism.', International Organization, 60 (2006), 65.1-85.
6. De Winter, ed., Regionalist Parties in Western Europe;
7. Stein Rokkan and Derek W. Urwin, eds, The Politics of Territorial Identity (Beverly Hills, Calif.: Sage Publications, 1982);
8. Joanne Bay Brzinski, 'The Changing Forms of Federalism, and Party Electoral Strategies: Belgium and the European Union', Publius, 29 (1999), 45-70;
9. James Fearon and Pieter van Houten, 'The Politicization of Cultural and Economic Differences' (Fifth Meeting of the Laboratory in Comparative Ethnic Processes (LiCep), Stanford University, 2002);
10. Derek J. Hearl, Ian Budge and Bernard Pearson, 'Distinctiveness of Regional Voting: A Comparative Analysis Across the European Community Countries (1979-93)', Electoral Studies, 15 (1996), 167-82;
11. Michael Hechter and Margaret Levi, 'A Rational Choice Approach to the Rise and Fall of Ethnoregional Political Parties', in A. Tiryakian and R. Rogowski, eds, New Nationalism in the Developed West (Boston, Mass.: Allen and Unwin, 1985), pp. 128-46.
12. William M. Chandler, 'Federalism and Political Parties', in W. M. Chandler, ed., Federalism and the Role of the State (Toronto: University of Toronto Press, 1987), pp. 149-70;
13. David Truman, 'Federalism and the Party System', in A. W. MacMahon, ed., Federalism: Mature and Emergent (New York: Doubleday, 1955), pp. 115-36;
14. Aaron B. Wildavsky, 'Party Discipline Under Federalism: Implication of the Australian Experience', in A. B. Wildavsky, ed., American Federalism in Perspective (Boston, Mass.: Little, Brown, 1967), pp. 162-81;
15. Scott Desposate, 'The Impact of Federalism on National Political Parties in Brazil', Legislative Studies Quarterly, 29 (2004), 259-85.
16. Pradeep Chhibber and Ken Kollman, 'Party Aggregation and the Number of Parties in India and the United States', American Political Science Review, 92 (1998), 329-4;
17. Pradeep Chhibber and Ken Kollman, The Formation of National Party Systems: Federalism and Party Competition in Britain, Canada, India and the United States (Princeton, N.J.: Princeton University Press, 2004).
18. Political decentralization, as it is defined here, is sometimes known by different names, including federalism, policy decentralization, or decision-making decentralization. See William Riker, Federalism: Origin, Operation, Significance (Boston, Mass.: Little, Brown, 1964);
19. Jonathan Rodden, 'Comparative Federalism and Decentralization: On Meaning and Measurement', Comparative Politics (2004), 481-500;
20. Daniel Treisman, 'Defining and Measuring Decentralization: A Global Perspective' (unpublished manuscript, University of California, Los Angeles, 2002). Increasingly, scholars are replacing the term 'federalism' with the term 'decentralization' for various reasons, including the desire to include countries that do not describe themselves as federal, such as Spain or Italy, but which have regional governments with independent decision-making powers.
21. Jonathan Rodden, 'Comparative Federalism and Decentralization'.
22. Robert Agranoff and Juan Antonio Ramos Gallarín, 'La evolución hacia una democracia federal en España: un examen del sistema de relaciones intergubernamentales', in R. Agranoff and R. Bañon i Martinez, eds, El estado de las Autonomías ¿ Hacia un nuevo federalismo? (Bilbao: Instituto Vasco de Administración Pública; 1998);
23. Pradeep Chhibber, Democracy Without Associations: Transformation of the Party System and Social Cleavages in India (Ann Arbor: University of Michigan Press, 1999).
24. Tulia G. Faletti, 'A Sequential Theory of Decentralization: Latin American Cases in Comparative Perspective', American Political Science Review, 99 (2005), 327-46.
25. Chhibber, Democracy Without Associations.
26. Kanchan Chandra, Why Ethnic Parties Succeed: Patronage and Head Counts in India (New York: Cambridge University Press, 2004).
27. See Samuel C. Patterson and Anthony Mughan, eds, Senate: Bicameralism in the Contemporary World (Columbus: Ohio State University Press, 1999).
28. In Germany the upper house is composed of the prime minister and the cabinet ministers of each of the country's regional (Länder) legislatures. The upper house is, thus, appointed rather than elected because the regional legislatures do not conduct separate elections to determine their representatives in the upper house, although the regional legislatures do elect these ministers in the first place.
29. De Winter, ed., Regionalist Parties in Western Europe;
30. Rokkan and Urwin, eds, The Politics of Territorial Identity;
31. Bonnie Martha Meguid, 'Competition between Unequals: The Role of Mainstream. Party Strategy in Rising Party Success in Western Europe' (doctoral dissertation, Harvard University, 2002) ;
32. Van Houten, 'Regional Assertiveness in Western Europe'.
33. Alberto Alesina and Enrico Spolare, 'On the Number and Size of Nations,' Quarterly Journal of Economics, 112(1997),.1027-56;.
34. Patrick Bolton and Gerard Roland, 'The Breakupof Nations: A Political Economy Analysis', Quarterly Journal of Economics, 112 (1997), 1057-90;
35. Walter Mattli, The Logic of Integration: Europe and Beyond (New York: Cambridge University Press, 1999).
36. Alesina and Spolare, 'On the Number and Size of Nations'.
37. See Riker, Federalism: Origin, Operation, Significance.
38. Alexander Hamilton, James Madison and John Jay, The Federalist Papers, reprint edn (Harmondsworth, Midx.: Penguin Classics, 1987).
39. Kathleen O'Neill, 'Decentralization as an Electoral Strategy', Comparative Political Studies, 36 (2003), 1068-91;
40. Kent Eaton, Politics Beyond the Capital: The Design of Subnational Institutions in South America (Stanford, Calif: Stanford University Press, 2004);
41. Maria Escobar-Lemmon, 'Political Support for Decentralization: An Analysis of the Colombian and Venezuelan Legislatures', American Journal of Political Science, 47 (2003), 683-97.
42. Treisman, 'Defining and Measuring Decentralization'; Alesina and Spolare, 'On the Number and Size of Nations'; Ugo Panizza, 'On the Determinants of Fiscal Centralization: Theory and Evidence', Journal of Public Economics, 74 (1999), 97-139.
43. Crawford Young, The African State in Comparative Perspective (New Haven, Conn.: Yale University Press, 1994);
44. Jeffery Herbst, States and Power in Africa: Comparative Lessons in Authority and Control (Princeton, N.J.: Princeton University Press, 2000).
45. For example, in the 2000 Congress of Deputy elections in Spain, regional parties won 29 per cent of the seats in the Canary Islands compared to 35 per cent in Catalonia, and about 8-9 per cent of the seats in Aragon and Madrid, compared to 12 per cent in Galicia.
46. Countries must hold two democratic elections for there to be an opportunity for a changeover of power.
47. In a separate analysis, I have measured regions according to the geographic regions of a country as a robustness check. No substantively different conclusions can be drawn about the relationship between decentralization and regional parties under this different specification.
48. These countries are: Cyprus, Czech Republic, Czechoslovakia, Israel, Moldova, Netherlands, New Zealand, Slovakia and Trinidad and Tobago. Portugal is also excluded from the analysis because its official election results (as reported on the National Election Commission's website) lack data on the Azores. Yugoslavia is also excluded because only regional (not national) elections were held in 1989.
49. I collected this dataset over a two-and-half year period by contacting every democracy in the world (scoring more than 4 on the Polity IV index for at least two consecutive elections) via email, letter, fax, telephone, etc. to get copies of their official election results. I received responses from fifty (out of a total of seventy-five) countries and included the data from all fifty countries in my dataset. I have no reason to expect that the countries for which I was not able to collect data are different from those for which I was able to collect data in a way that will bias the results. Although many of the excluded countries are less developed than those included, the overall level of development of a country should not increase or decrease the strength of regional parties, although regional variation in levels of development may affect the strength of regional parties. To further ensure that case selection is not driving the results, 1 conducted the same analyses on a smaller subset of the dataset, which produced similar results. I also discarded outliers and inspected the data to determine whether the presence of missing data produces different results, which it does not.
50. The following countries have democratically elected sub-national legislatures with the authority to raise their own taxes, but do not have authority over political issues, or any other fiscal issues, and are considered centralized in this study: Bolivia (tax authority post-1994), Botswana, Colombia, Costa Rica, Estonia, Greece (elected only), Hungary, Iceland, Indonesia, Ireland (elected only), Latvia, Lithuania, Niger, Norway, Poland, Romania, Slovenia, Sweden and Turkey. In this study, there are four countries in which sub-national legislatures are not elected and do not have the authority to levy taxes: Bermuda, Mauritius, Spain (1977-79), the United Kingdom (1974-97).
51. Daniel J. Elazar, Federal Systems of the World: A Handbook of Federal, Confederal and Autonomy Arrangements (London: Longman, 1994);
52. John Gerring, Strom C. Thacker and Carola Moreno, 'Centripetal Democratic Governance: A Theory and Global Inquiry', American Political Science Review, 99 (2005), 567-81.
53. The seventeen countries are: Argentina, Australia, Belgium (since 1978), Bosnia-Herzegovina, Canada, Finland, Germany, India, Italy, Malaysia, Mexico, South Africa, Spain (since 1980), Switzerland, the United Kingdom (1945-74; 1998-2000), the United States and Venezuela.
54. More detailed summary statistics are available from the author upon request.
55. In an election where there are four parties competing and only one is a regional party, the value of the first measure is 25 per cent. If this party wins 10,000 votes out of a possible 40,000 votes and ten out of a possible forty seats, the values of the second and third measures are also 25 per cent.
56. I have also defined regional parties as parties that compete in more than one region of a country, but not in every region, and I have measured regional parties according to these same three measures. Decentralization has the same substantive effect on regional party strength, in these models. However, since my argument is more consistent with an understanding of regional parties as parties competing in only one region of a country, 1 do not rely on this definition or measurement of regional parties in this article.
57. i = the number of votes won by a party in a district (or constituency)/total number of votes won by the party in a district (or constituency) and ND is the number of districts (or constituencies) in a country.
58. 2, where p, represents the vote share of a party participating in an election (Rein Taagepera and Matthew Soberg Shugart, Seats and Votes: Effects and Determinants of Electoral Systems (New Haven, Conn.: Yale University Press, 1989)).
59. Ethno-linguistic heterogeneity, however, is only one of the types of regional cleavages that may affect the strength of regional parties in countries. Other regional cleavages may be religious or economic in nature. See James D. Fearon and Pieter van Houten, 'The Politicization of Cultural and Economic Difference' (unpublished manuscript, Department of Political Science, Stanford University, no date). Unfortunately, cross-national data on these particular cleavages are not available. Ethno-linguistic heterogeneity is nonetheless one of the most important, if not the most important, cleavage associated in the literature with regional parties.
60. Charles Lewis Taylor and Michael C. Hudson, World Handbook of Political and Social Indicators (New Haven, Conn.: Yale University Press, 1976).
61. I have compared the Roeder data (available from Professor Roeder, University of California, San Diego) to census data I have collected. The level of fractionalization is similar in both datasets. I, therefore, rely on the Roeder dataset because it contains data for more countries.
62. The MAR dataset only includes 'at risk' groups, however. At-risk groups are 'all non-state communal groups that collectively suffer or benefit from systematic discriminatory treatment vis-à-vis other groups, and/or groups that collectively mobilize in defense or promotion of their self-defined interests.' Cleavages may be stronger or weaker in countries, depending on the presence of 'non-risk' groups in a country. See Minorities at Risk Project (MAR), .
63. See Shaheen Mozaffar, James R. Scarrit and Glen Galaich, 'Electoral Institutions, Ethnopolitical Cleavages, and Party Systems in Africa's Emerging Democracies', American Political Science Review, 97 (2003), 379-90. It was not possible to create a weighted index because of insufficient data on a group's proportion of a country's total population.
64. I do not include a control for upper houses because I do not have data on upper house elections that are either elected or appointed by regional legislatures in this study.
65. The correlation coefficients between political decentralization and all four measures of fiscal decentralization in this article are less than 0.5.
66. Rodden, 'Comparative Federalism, and Decentralization'.
67. Chhibber and Kollman, 'Party Aggregation and the Number of Parties in India and the United States'; Chhibber and Kollman, The Formation of National Party Systems.
68. In addition to total sub-national expenditure (as a percentage of GDP), Chhibber and Kollman use several qualitative indicators of fiscal decentralization in their book-long treatment of the subject (The Formation of National Party Systems), which include the proportion of public sector employees at the sub-national level of government and the degree of autonomy granted to sub-national governments for social welfare policies and the establishment of public corporations.
69. Gary Cox, Making Votes Count: Strategic Coordination in the World's Electoral Systems (New York: Cambridge University Press, 1997).
70. Warren F. Miller, 'Presidential Coattails: A Study in Myth and Methodology', Public Opinion Quarterly, 19 (1955/56), 353-68.
71. Maurice Duverger, Political Parties: Their Organization and Activity in the Modern State (New York: John Wiley, 1954).
72. Gian Maria Milesi-Ferretti, Roberto Perotti and Massimo Rostagno, 'Electoral Systems and Public Spending', Quarterly Journal of Economics, 117 (2002), 609-57.
73. Cox, Making Votes Count.
74. Breusch-Pagan and White Tests indicate that heteroscedasticity is not a problem in the analysis. I have also explored fixed-effects for regions of the world. Models including fixed-effects for regions of the world produce the same substantive conclusions regarding decentralization as those without world region fixed-effects. I also cluster the data according to country. The size of the coefficient for decentralization is the same in these models and significant at the 0.1 level or better. I also estimate the relationship between decentralization and regional party vote using Feasible Generalized Least Squares (FGLS), which allows variance to vary across countries, and Seemingly Unrelated Regression (SUR) analysis, which allows mean and variance to vary across countries. In FGLS models allowing for panel heteroscedasticity, the effect of decentralization is still positive and significant, but about half the size of that in the OLS models. In the SUR models, the effect of decentralization on regional party vote is positive, significant and about the same size as in the OLS models. The robustness tests are available from the author upon request. Note that it is not possible to use country fixed-effects in this analysis since the main variables of interest do not vary across time.
75. There are fewer observations in Model 2 than in Model 1 because I was unable to determine the concurrency of a number of elections in Venezuela.
76. Chhibber and Kollman, 'Party Aggregation and the Number of Parties in India and the United States'.
77. Chhibber and Kollman, The Formation of National Party Systems.
78. Unfortunately, there is no alternative source of data for fiscal decentralization that separates out transfers from all other sources of funding for these countries.
79. The ELF index lacks data for Bermuda, Finland, Indonesia and Sweden. The exclusion of these countries from the analysis does not change the conclusions of this study about decentralization. Excluding these countries from Models 1-2 does not change the substantive or statistical effect of any of the variables in the models.
80. See Treisrnan, 'Defining and Measuring Decentralization'.
81. Young, The African State in Comparative Perspective;
82. Herbst, States and Power in Africa.
83. Model 1 and Model 2 do not have 296 observations because the population measure only covers the period 1950-2002.
84. I exclude decade fixed-effects from these models.
85. To ensure the validity of these conclusions, I replicate all of the models in the tables (except Model 3 because of the small number of observations in these analyses), excluding decade-fixed effects from the models. I then replace decentralization with the aforementioned instrumental variables. I compare the models with instrumental variables to models without instrumental variables using Hausman specification tests. None of the models with instrumental variables are statistically different from the models without instrumental variables, indicating that the relationship between decentralization and regional parties and the relationship between decentralization and denationalization are not endogenous.
86. Dawn Brancati, 'Decentralisation: Fueling the Fire or Dampening the Flames of Ethnic Conflict and Secessionism,' International Organization, 60 (2006), 651-85;
87. Ugo M. Arnoretti and Nancy Bermeo, eds, Federalism and Territorial Cleavages (Baltimore, Md.: John Hopkins University Press, 2004);
88. Arend Lijphart, Democracy in Plural Societies: A Comparative Exploration (New Haven, Conn.: Yale University Press, 1977);
89. Donald Horowitz, A Democratic South Africa? Constitutional Engineering in a Divided Society (Berkeley: University of California Press, 1991).