Decentralized Despotism: How Indirect Colonial Rule Undermines Contemporary Democratic Attitudes Evidence from Namibia Marie Lechler 1 Lachlan McNamee 2 1 University of Munich 2 Stanford University June 20th, 2016 M.Lechler and L.McNamee Decentralized Despotism: June 20th, 2016 1 / 27
Idea During colonial times, southern and central Namibia were directly ruled by the German and South African authorities while northern Namibia was indirectly ruled. As a consequence, the role of traditional authorities was much more important in northern Namibia than in southern and central Namibia. This influence of traditional authorities in northern Namibia persists to the present day. Idea: use division of Namibia as natural experiment to identify the long-term effects of indirect colonial rule. M.Lechler and L.McNamee Decentralized Despotism: June 20th, 2016 2 / 27
Main result We find that individuals living in the formerly indirectly ruled part of the country have lower support for democracy and the rule of law. ⇒ Hypothesis: local governance structures, which are organized on a ’despotic’ basis (i.e. traditional leadership) undermine democratic attitudes M.Lechler and L.McNamee Decentralized Despotism: June 20th, 2016 3 / 27
Table of contents Motivation and historical background 1 Results 2 Conclusion 3 M.Lechler and L.McNamee Decentralized Despotism: June 20th, 2016 4 / 27
Motivation Our project contributes to understanding of determinants of political attitudes and of viability of democratic systems. We are able to document an important effect of the ongoing influence of traditional authorities in Sub-Saharan Africa, who play a uniquely important role in the governance of these countries. M.Lechler and L.McNamee Decentralized Despotism: June 20th, 2016 5 / 27
Motivation Adds to literature about the importance of ’civic culture’ and individual political attitudes for the viability of democratic institutions Almond & Verba (1963); Inglehart (1988,1990); Persson & Tabellini (2009) and Fuchs-Sch¨ undeln & Sch¨ undeln (2015) literature about long-term effects of direct vs indirect colonial rule Iyer (2010), Acemoglu et al (2000, 2001) literature about role of traditional authorities in Sub-Sahara Africa Mamdani (1996); Dusing (2002); Acemoglu, Reed & Robinson (2013); Baldwin (2013, 2015); De Kadt & Larreguy (2014) M.Lechler and L.McNamee Decentralized Despotism: June 20th, 2016 6 / 27
History of the border Figure : Precolonial map of ethnicities (Murdock, 1967) M.Lechler and L.McNamee Decentralized Despotism: June 20th, 2016 7 / 27
History of the border 1884: Namibia becomes German colony Germans settle in central and southern Namibia Rinderpest epidemic devastated cattle-dependent communities, which were concentrated in central and southern Namibia in the 1890s Strength tribes living in the North and inaccessibility of the North hamper settlements in the north ⇒ Germans easily conquered southern and central Namibia 1905: Introduction of the ‘’Police Zone” (police protection cannot cover entire protectorate) Due to financial constraints German activities (administrative, economic and military) focus on Police Zone M.Lechler and L.McNamee Decentralized Despotism: June 20th, 2016 8 / 27
History of the border Source: Miescher, Namibia’s Red Line, 2012 M.Lechler and L.McNamee Decentralized Despotism: June 20th, 2016 9 / 27
History of the border 1915: Police Zone boundary and German policies of restricting movement were adopted by the South Africans. 1964: Police Zone boundary is formalized by Odendaal Commission 1990: Namibian independence. Transition to parliamentary democracy and homogenization of policies north and south of the border. M.Lechler and L.McNamee Decentralized Despotism: June 20th, 2016 10 / 27
Direct vs indirect rule Indirect colonial rule in northern Namibia Traditional authorities in the north were given a lot of political autonomy Despotic, hereditary and authoritarian system Influence persists even after independence and constitutes a parallel system of governance Direct colonial rule in central and southern Namibia German (and later South African) colonial administration directly ruled over this part of the country Traditional authorities were given little or no political role Since independence, people only experience governance via elected and accountable local representatives at all levels of government M.Lechler and L.McNamee Decentralized Despotism: June 20th, 2016 11 / 27
Hypothesis 1. Indirect and direct colonial rule have persistent effects on democratic attitudes (test with OLS). 2. The key mechanism underlying this relationship is the ongoing influence of traditional authorities, which acts as a parallel despotic governance structure (test with 2SLS). M.Lechler and L.McNamee Decentralized Despotism: June 20th, 2016 12 / 27
Data M.Lechler and L.McNamee Decentralized Despotism: June 20th, 2016 13 / 27
Data Political attitude data: Afrobarometer (1999-2008, 4 survey rounds) Covers 166 EAs and 1426 individuals Questions Contact traditional authority: During the past year, how often have you contacted a traditional ruler? (Scale 0-3) Support for democracy (Scale: 1-3) Courts make binding decisions (Scale: 1-5) How likely is law enforcement by authorities (Scale: 1-4) Questioning leaders vs respect for authority (Scale: 1-4) M.Lechler and L.McNamee Decentralized Despotism: June 20th, 2016 14 / 27
Data Table : Balancing table for observations from buffer zone (1) (2) (3) Direct rule Indirect rule Difference Contact traditional leader 0.24 0.79 0.55*** Age 33.4 36.3 2.90*** Gender 1.49 1.49 0.00 Education 3.80 3.82 0.02 Without food 1.31 1.10 -0.20** Observations 254 1,164 1,418 M.Lechler and L.McNamee Decentralized Despotism: June 20th, 2016 15 / 27
The effect of indirect rule on political attitudes Baseline estimation equation: ′ Y ider = α + β · Indirectrule d + X ider γ + η e + µ r + ǫ ider Y expresses the outcome variables (the measures for democratic attitudes) for individual i, living in enumeration area d, belonging to the ethnic group e, being surveyed in round r. Indirectrule is a dummy variable indicating whether the individual lives in an enumeration area which belonged to the indirectly or the directly ruled part of Namibia X is a set of individual-level control variables η e are ethnicity fixed effects µ r are survey-round fixed effects. M.Lechler and L.McNamee Decentralized Despotism: June 20th, 2016 16 / 27
Identification Treatment: indirect vs direct colonial rule independent of other factors affecting political attitudes close to the colonial border (location of border exogenous to pre-colonial political attitudes) Pre-treatment attitudes were the same within each ethnic group M.Lechler and L.McNamee Decentralized Despotism: June 20th, 2016 17 / 27
OLS estimates Table : Effect of indirect rule on political attitudes (1) (2) (3) (4) Support Support Courts make Courts make VARIABLES for democracy for democracy binding decisions binding decisions Indirect colonial rule -0.178** -0.189** -0.169* -0.178* (0.0746) (0.0757) (0.102) (0.105) Constant 2.493*** 2.660*** 3.557*** 3.844*** (0.0908) (0.237) (0.130) (0.265) Observations 1,347 1,329 1,392 1,375 R 2 0.019 0.036 0.089 0.103 Ethnicity FE yes yes yes yes Survey round FE yes yes yes yes Controls no yes no yes # clusters 165 165 165 165 Results from OLS regressions including ethnicity and survey round fixed effects. Control variables are age, education dummies and income dummies. The sample consists of buffer zone observations only. Standard errors (clustered by Enumeration Area) in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1. M.Lechler and L.McNamee Decentralized Despotism: June 20th, 2016 18 / 27
OLS estimates Table : Effect of indirect rule on political attitudes (1) (2) (3) (4) Law Law Respect Respect VARIABLES enforcement enforcement authorities authorities Indirect colonial rule -0.272** -0.298** 0.122 0.131 (0.115) (0.116) (0.0890) (0.0899) Constant 3.507*** 3.134*** 1.950*** 1.978*** (0.127) (0.304) (0.121) (0.279) Observations 1,026 1,009 1,396 1,379 R 2 0.043 0.061 0.096 0.110 Ethnicity FE yes yes yes yes Survey round FE yes yes yes yes Controls no yes no yes # clusters 128 128 165 165 Results from OLS regressions including ethnicity and survey round fixed effects. Control variables are age, education dummies and income dummies. The sample consists of buffer zone observations only. Standard errors (clustered by Enumeration Area) in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1. M.Lechler and L.McNamee Decentralized Despotism: June 20th, 2016 19 / 27
Robustness Applying an ordered probit model Link Using observations for the entire country (not only buffer zone) Link Clustering observations on the constituency level (60 clusters instead of 165) Link M.Lechler and L.McNamee Decentralized Despotism: June 20th, 2016 20 / 27
Channel of causality Identification of contact to traditional leaders as channel of causality by applying 2SLS estimations. → instrument contact with traditional authorities with indirect colonial rule dummy M.Lechler and L.McNamee Decentralized Despotism: June 20th, 2016 21 / 27
Recommend
More recommend
