Institutions 1/35 Institutional Design, Quality, and Conflict H˚avard Mokleiv Nyg˚ard Peace Research Institute Oslo March 11, 2015 Institutions Introduction 2/35 Outline Introduction What are institutions? What is an institution? The civil democratic peace Good governance Power-sharing Institutions What are institutions? 3/35 Conceptual framework: Institutions The Big Question: (How) Can we design institutions to avoid conflict? The Big Question II: Do institutions have an independent effect or are they just ‘codifications’ of behaviour? The Big Question III: How do we deal with endogeneity? Institutions both cause of conflict and caused by conflict Institutions What is an institution? 4/35 What is an institution? What is an institution? Douglas North (1991): Institutions are the humanly devised constraints that structure political economic and social interaction. They consist of both informal constraints (sanctions, taboos, customs, traditions and codes of conduct), and formal rules (constitutions, laws, property rights). Our literature is mostly concerned with the latter, or a mix of both formal and informal Institutions What is an institution? 5/35 Formal and informal institutions Literature mostly looks at formal political institutions Problem: How to measure institutions? Study constitutions (Normative vs. empirical reality) Study behaviour (How do we measure the effect of an institution then?) Most datasets (e.g Polity, Freedom House) are a mix of both Some are mostly de jure (e.g IAEP) Some are mostly about behaviour (e.g World Governance Indicators, from World Bank) Institutions What is an institution? 6/35 Conceptual framework: Institutions Four ‘things’ that national (political) institutions regulate: 1 Participation 2 Competition 3 Selection of the executive 4 Constraints on the executive (Horizontal constraints) Feature of these institutions: Their quality, The degree to which the rules are followed, and ‘good’ This lecture will be concerned with institutions that fall along one of these institutional dimensions Institutions What is an institution? 7/35 Conceptual framework: Theoretical Foundations Most theories about how institutions reduce conflict, either focus on: Grievances (institutions produce political motivations for conflict) Opportunities (some institutions make conflict easier to organize) Bargaining (institutions reduce or increase commitment problems) Be aware of this when going through the readings! Institutions The civil democratic peace 8/35 The democratic peace Institutions The civil democratic peace 9/35 The Liberal Peace Three kinds of arguments from the classical literature: 1 Non-democratic leaders are more aggressive than democratic leaders 2 Liberal institutions has a pacifying effect on people 3 Democracy ‘individualizes’ and ‘rationalizes’ people Four (five) kinds of arguments from the modern literature: 1 Democratic leaders are responsible to the public and can be ousted (structural) 2 Democratic leaders are used to handling disputes peacefully (institutional) 3 Democracies are on average wealthier than non-democracies → they have more to loose from a war (structural) 4 International Organizations can hinder conflict and democracies are more likely to join such organizations (structural) 5 Trade (structural) Institutions The civil democratic peace 10/35 A civil democratic peace? Key question: is there a civil democratic peace? Is democracy pacifying also inside of states? Are democratic states less likely to see internal armed conflict Special case of semi democratic, or inconsistent, regime Not discussed: democracy and repression Institutions The civil democratic peace 11/35 Inconsistent Regimes Institutions The civil democratic peace 12/35 Murder in the Middle Consistent democracies and autocracies both have much lower risk of conflict (civil war) then regimes in the middle Autocracies do not have higher probability of conflict than full democracies Change in regime type, in either direction, increases the risk of conflict Possible prediction: The world is likely to become more democratic: changes are more likely in the democratic direction, and over time this will lead to more stable democracies Institutions The civil democratic peace 13/35 Murder in the Middle Institutions Good governance 14/35 Governance and Open-Access Orders Three aspects of institutions that are inadequately captured in most studies: The formal-informal divide Institutions have histories In developing countries there are both ‘modern’ and ‘traditional’ institutions Institutions and the organization of violence shape each other, and (according to some) part of the same social order Two papers that address this: Hegre and Nyg˚ard, 2014 (empirical), and North, Wallis and Weingast, 2009 (theoretical) Institutions Good governance 15/35 Governance Hegre and Nyg˚ard, 2014 Studies the impact of ‘good governance’ (i.e institutional quality) on internal armed conflict Separate between formal (the institutional setup) and informal (the quality of institutions) institutions Argue that ‘good governance’ is important, beyond democratic institutions Draw explicitly on ‘grievance’ theory Look at conflict onset and recurrence in the same model Institutions Good governance 16/35 Governance Empirical setup: Good governance Six subconcepts: Rule of law, corruption, bureaucratic quality,economic policies, military in politics, political exclusion and repression Formal political institutions: Use the SIP index of democracy (combination of Polity and Polyarchy (Vanhanen) Informal political institutions: The residual ‘good governance’ not explained by formal institutions Institutions Good governance 17/35 Formal vs. informal institutions Canada Cuba Dominican Republic Jamaica Trinidad and Tobago Honduras Nicaragua Costa Rica Panama Guyana Ecuador Brazil Bolivia Paraguay Chile Argentina Uruguay Ireland Netherlands Belgium Luxembourg France Switzerland Portugal Germany Poland Austria Hungary Czech Republic Italy Albania Greece Cyprus Bulgaria Romania Finland Sweden Norway Denmark Iceland Equatorial Guinea Gambia, The Mauritania Burkina Faso Ghana Cameroon Kenya Tanzania Zambia Zimbabwe Malawi South Africa Namibia Botswana Swaziland Madagascar Mauritius Morocco Tunisia Libya Syrian Arab Republic Jordan Saudi Arabia Bahrain Qatar United Arab Emirates Oman China Mongolia Taiwan, China Korea, Dem. Rep. Korea, Rep. Japan Bhutan Lao PDR Vietnam Malaysia Singapore Australia New Zealand Solomon Islands Fiji United States Mexico Serbia Guinea-Bissau Niger Cote d'Ivoire Guinea Sierra Leone Nigeria Central African Republic Congo, Rep. Burundi Djibouti Lesotho Algeria Egypt, Arab Rep. Nepal Haiti Guatemala El Salvador Colombia Venezuela, RB Peru United Kingdom Spain Russian Federation Mali Senegal Togo Chad Congo, Dem. Rep. Uganda Rwanda Somalia Ethiopia Angola Mozambique Comoros Sudan Iran, Islamic Rep. Turkey Iraq Israel Afghanistan India Pakistan Bangladesh Myanmar Sri Lanka Thailand Philippines Indonesia Papua New Guinea -3-2-1012 Informalinstitutionsindex -1.5 -1 0 1 2 Formal institutions No Conflict New Conflict Conflict Recurrence Institutions Good governance 18/35 Governance Empirical findings: ‘Good governance’ index is negatively associated with conflict relapse ‘Formal’ institutions (SIP score), also negatively associated with conflict relapse ‘Informal’ institutions (residual good governance not explained by formal inst) negatively related to conflict relapse Good informal institutions matter more Also attempt to handle endogeneity Institutions Good governance 19/35 Governance and conflict recurrence .01.02.03.05.1.2.5 Probabilityofrecurrence 0 5 10 15 20 25 Years after Conflict Low Gov Low Gov, 10/90 pctile High Gov High Gov, 10/90 pctile Institutions Good governance 20/35 Open-Access Orders Largely un-addressed: Institutions and patterns of organized violence evolve together That some countries with good institutions have less violence and economic development is due to same factors ‘Development clusters’ (Peter Evans) Condensed version of a book: ‘Violence and Social Orders: A Framework for Interpreting Recorded Human History’ Institutions Good governance 21/35 Open-Access Orders Condensed: ‘over the last ten millennia, societies have used institutions to limit and contain violence. These institutions simultaneously give individuals control over resources and social functions, and limit the use of violence by shaping the incentives that individuals and groups face. We call these patterns of social organization social orders.’ Institutions and violence part of the same integrated process Peaceful, inclusive societies part of one social order: Open access orders Violent, corrupt and poor societies: Limited access orders Institutions Good governance 22/35 Open-Access Orders Closed access orders: Access to organizations and economic activity limited Personal relationships basis for political organization Violence is (often) privatized (many organizations; paramilitaries, army, police, ethnic groups etc.) Those who can produce violence form the dominant coalition Access to organizations and other privileges distributed to organizations with violence potential ‘the threat of violence becomes part of the arrangement that controls the actual use of violence’ Open Access orders: Open access to organizations and economic activity Unbiased rule of law Contracts are credible Violence monopolized, and credibly enforced by the state Impersonal identity (your formal position determines status) Perpetually lived organizations Institutions Good governance 23/35 Open-Access Orders Take-home points: North, Wallis and Weingast see organized violence as part of a closed access order ‘Good institutions’ and peace are two sides of the same coin! The threat of violence determine who will gain access to organizations Fundamental dilemma: Countries are undemocratic, and have ‘bad governance’ because they have to to control violence We cannot change the institutions without changing the distribution of violence potential In open-access orders competition through violence is not profitable Institutions Power-sharing 24/35 The power-sharing debate Widespread idea: Power-sharing institutions reduce conflict Power-sharing institutions are more than democracy: Institutions that increase the majority needed to make a decision Institutions that increase the number of veto-players in the system A veto player is an actor (collective or individual) whose approval is necessary to change the status quo Key idea in power-sharing theory: To reduce conflict, give ethnic groups veto-player status Institutions Power-sharing 25/35 The power-sharing debate Intellectual history Goes back to Aristotle, via Montesquieu In modern form, chief proponent is Arend Lijphart Lijphart (1977). ‘Democracy in Plural Societies’ Lijphart (1999). ‘Patterns of Democracy: Government Forms and Performance in Thirty-Six Countries’ Institutions Power-sharing 26/35 The power-sharing debate Lijphart‘s main idea: Some societies are deeply divided along ethnolinguistic and religious lines, but have avoided conflict Identifies elite cooperation and ‘consociational’ institutions (also known as power-sharing institutions) These institutions are institutions that ‘includes rather than excludes, and that tries to maximize the ruling coalition instead of being satisfied with a bare majority‘ Lijphart thought these institutions explained why these societies avoided conflict Institutions Power-sharing 27/35 The power-sharing debate What kind of institutions was Lijphart talking about: Proportional representation (that creates grand-coalition governments) Mutual veto (legislative-executive balance) Judicial independence Balanced bicameralism Legislative quotas for minorities An independent central bank Federalism Corporatism Institutions Power-sharing 28/35 The power-sharing debate All of these institutions constrain majority power They give minorities a stake in the system Notice how this fits with central insights in bargaining theory: Commitment problems can be reduced by institutional guarantees Breaking commitments to share power in these institutional settings are more costly Puts the whole institutional framework into play, and this increases costs Bargaining theory is currently the most sophisticated theoretical framework for power-sharing theory (see e.g Walter 1999) Also notice the connections with the literature on ethnic exclusion (Cederman et.al) Inclusion of ethnic groups will reduce political grievances Institutions Power-sharing 29/35 The power-sharing debate Which countries have power-sharing institutions? Many modern democracies: Belgia and the Netherlands are classic examples (drawn heavily on by Lijphart) But many developing countries too: South Africa (Liphart adviced the constitutional council) El Salvador Kenya (after riots in 2007/2008 elections) Do these countries have PS institutions because they are conflictual? Institutions Power-sharing 30/35 The power-sharing debate 1975 1980 1985 1990 1995 2000 2005 0.60.70.80.91.0 year Institutionalindices Polity Consociation index rel. excluded population (EPR) Institutions Power-sharing 31/35 The power-sharing debate Residuals (POLITY score and power−sharing index) −10 −5 0 5 Sudan Algeria United Arab Emirates Gambia Azerbaijan Laos Oman Iraq Uzbekistan Guinea Uganda Kyrgyszstan Congo Saudi Arabia Liberia Tajikistan Upper Volta / Burkina Faso Tunisia Kenya Rwanda Burma Kuwait Swaziland Togo Vietnam, Dem. Rep. Ukraine Niger Tanzania Bolivia Taiwan Venezuela Yugoslavia/Serbia Moldova Georgia Ghana Zambia Chad Armenia Japan Uruguay Nicaragua Turkey/Ottoman Empire Dominican Republic Namibia Bangladesh Macedonia Malaysia South Africa India Estonia Ivory Coast Denmark Costa Rica Switzerland New Zealand Thailand Ireland Greece Mongolia Finland Institutions Power-sharing 32/35 The power-sharing debate The pro-power-sharing position: Power sharing reduces conflict Brings peace to post-conflict societies Should be advocated and adopted by developing countries The anti-power-sharing position: Power-sharing hardens ethnic divisions and increases ethnic voting Power-sharing does not make parties gravitate towards the center, but towards the extremes This increases conflict potential Donald Horowitz (1985) Phillip Roeder (2005) Przeworski: ‘democracy is a system in which parties lose elections’ → power-sharing is fundamentally un-democratic Institutions Power-sharing 33/35 The power-sharing debate What‘s the evidence? Studies on curriculum are a (small) sample. Literature is huge! Major contribution: Hartzell and Hooddie 2003: Looks at power-sharing institutions following negotiated settlements 38 civil wars ended this way in period 1945-1998 Four ‘dimensions’ of power-sharing: Political, economic, military, territorial Find that extensive power-sharing along many dimension decreases risk of return to conflict Find much of the same in a book from 2007 Institutions Power-sharing 34/35 The power-sharing debate Cammett and Malesky 2012 Studies countries that have had a negotiated settlement Dissaggregate power-sharing provisions Find that closed list PR is robustly linked to decline in risk of recurrence Closed list PR encourages non-personalistic voting and better governance (because it discourages patronage) They also address the issue of endogeneity Institutions Power-sharing 35/35 The power-sharing debate Discussion: There are problems with the sample Post-conflict countries Is ‘selection’ properly handled in these studies? These are studies of effects of medicine in those who have already been sick