Comprehensive Exams Comparative Politics 2020

Humphreys and Weinstein 2008

Who Fights? The Determinants of Participation in Civil War - Survey of fighters in Sierra Leone. Fighters were either in insurgent group (RUF) or civilian defense group (CDF). Those in RUF mostly abducted, while those in CDF mostly there voluntarily. - Grievance theory: Poverty and lack of ID with a party predicts both sides joining - Olsonian cost/benefits: Material offers make joining more likely, as does the belief that one would be safer inside the group - Social ties: Village alienation doesn't predict joining, but social ties to other group members do - The best model is one in which these three measures are considered jointly: They are complements not substitutes - Can't forget that involuntary joining is a major reason for rebellion

Jackson and Rosberg 1982

Why Africa's Weak States Persist - Many political scientists employ a Weberian concept of states which involve a monopoly of force, but African states don't qualify as this - One cannot explain the persistence of some states by using a concept of the state that doesn't give sufficient attention to the juridical properties of statehood - Capacity of Africa's governments to exercise control hinges on domestic authority (personal not instituitonal), apparatus of power (alienated elites), economic circumstances (reliance on exports, small workforce) - Juridical statehood maintained in Africa for 3 reasons: (1.) Pan-Africanism expressed in acceptance of inherited jurisdictions - Common interest in support of international rules because of common vulnerability of states - African states became independent when international system was highly organized

Cederman et. al. 2010

Why Do Ethnic Groups Rebel? - Past theories have written off grievance approach, but has been tested with ethnic fractionalization as opposed to fine data - Collect new dataset on ethnic groups and their access to state (executive) power - Frequency of conflict increases with the degree of exclusion, and excluded groups are much more likely to experience a rebellion in their name. Larger groups are more likely to experience a conflict in their name. Past conflict predicts future conflict.

Gurr 1970

Why Men Rebel - The greater the frustration and deprivation, the greater the scope of targeted political action, intensity of violence, and duration of conflict - Three steps: -- Development of discontent: Stems from relative deprivation- the difference between expectations and capabilities -- Politicization of discontent: Function of societal variables -- Actualization of political violence: Depends on existing institutions

Arrow 1994

- "Methdological individualism": Recognition that individuals shape economic analyses - For example, when economists talk about the "market", all transactions are conducted through preferences of individuals - All rules of game theory are social - Information is supplied socially

Manin, Przeworski, and Stokes 1999

- "Representation": Government acts in the best interests of the public - "Responsiveness": The relationship between signals and policies - "Accountability": The relationship between outcomes and sanctions

Hall and Taylor 1996

- About the 3 new institutionalisms - Historical institutionalism: Conflict among rival groups for scarce resources is the basis for politics. Social/cultural traits are not driving. Institutions embody a Nash equilibrium. - Rational choice institutionalism: Arose from studies of American congressional behavior. Stable majorities exist because of institutional features such as agenda control. Politics as a set of collective-action dilemmas - Sociological institutionalism: Many institutional forms are not adopted because they are the most efficient, but because they are culturally-specific practices

Kasara and Suryanarayan 2015

- Although scholars of inequality have concluded that the poor are less likely to vote, empirically this is not always true - Authors argue that the rich will turn out when they are threatened with being taxed, which is determined by two factors: (1.) political salience of redistribution and (2.) extractive capacity of the state - IV: Salience of redistribution (degree to which people in each income quintile support different political parties), bureaucratic capacity (risk scores from a consumer group) - DV: Turnout of different income groups - They find that the turnout of the rich is positively correlated with bureaucratic equality and salience - Last, they use an IV approach and use war experience to instrument for extractive capacity. The relationship between bureaucratic capacity, political salience, and turnout of the rich holds.

Fisman and Miguel 2005

- Although social norms are often mentioned as a contributor to corruption, there is little evidence beyond cross-country empirics - Background: Prior to 2002, diplomats were protected from legal charges including parking violations in the US - Authors examine parking tickets in NYC by diplomats - Number of parking tickets is strongly correlated with other country corruption measures - Diplomats from low corurption countries behave well even when they could get away with it, while those from high corruption countries violate away - Contrary to models of norm convergency, there is a positive interaction between home country corruption levels and tenure in NYC - Diplomats from countries where popular attitudes toward the US are worse have more parking violations - Isolates corruption culture away from institutions

Fearon and Laitin 2003

- Argue that the prevalence of civil war in the 1990s was not due to the Cold War - Greater degree of ethnic or religious diversity does not make a country more prone to civil war - Main factors that determine civic violence are conditions that favor insurgency - Find linear accumulation of civil wars over time (Cold War alone not responsible) - For any level of diversity, as one moves up the income scale the # of civil wars decreases- evidence against the hypothesis that diversity increases civil war or that ethnic strife is activated by modernization - After controlling for income, civil wars are no less frequent in democracies but are more frequent in anocracies - Causal mechanism behind fewer civil wars is a well-financed and administratively competent government

Thies 2005

- Argues that the current application of Tilly's model is too narrow to be of use in LA - External rivals should allow the state to augment its extractive capacity while internal rivals should have the opposite effect - DV: Tax ratio (tax revenue as percent of GDP) - Interstate war does not have a significant effect on extraction, but civil wars have a negative effect on extraction (in line with Centeno 2002) - The interaction term bewteen interstate and civil wars is positive: States experiencing both were able to extract better - Democracy has a positive and significant impact, higher external debt reduces extraction - Most importantly, and in line with an expanded, bellicist theory, long-term interstate rivalries increase extraction by the state

Lieberman 2007

- Argues that the relationship between ethnicity and public policy is channeled largely through political competition over the social status of ethnic groups and the propensity to view risks as pooled - When countries have strong internal boundaries dividing society into groups, the AIDS epidemic was likely to be understood in ethnic terms. - Studies AIDS policies in Brazil, India, and South Africa - Brazil had the most outstanding response, which he credits to weakly institutionalized race - In South Africa, discussions of risk and politics of blame broke down along racial lines and lowered demand for aggressive AIDS policies. In India, caste and religious boundaries also framed the politics of risk.

Golden and Chang 2001

- Ask when and why political competition may fail to inhibit rent seeking on the part of elected officials - Test the hypothesis that competition among candidates from the same party is associated with corruption - Examine three origins of political corruption: Intraparty (as a result of the characteristics of the electoral system), interparty competition, and specifically for the Italian case, the legal change in the 1970's over public financing that criminalized preexisting financing of campaigns - Use legal data from the Italian parlimanet about charges of corruption against politicians - Hypothesize that Senators would be less corrupt because they are not elected by open-list PR compared to other reps. They find that intraparty competition drove corruption, while interparty competition had little effect. - These results were not due to culture: When they map corruption in Italy, they find that corruption was not concentrated in the South early on, and that charges of wrongdoing are well distributed across Italy - They argue that it is political institutions, not culture, that drives corruption

Acemoglu and Robinson 2001

- Assumption: The poor want democracy. - Under dictatorships, the poor are excluded from politics but they pose a threat - Elites will try to make concessions by promising redistribution, but they are unable to make credible commitments for the future - Elites have to implement democratic institutions to successfully make credible commitments and quell revolution - However, coups are likely to happen if democratization is costly - Transitions are more likely in times of recessions, when the costs of political change are low for both elites and masses.

Inglehart 1988

- Author argues for the conception that political culture exists. He shows that countries have a relatively stable level of life satisfaction across time, and that democratic institutions depend on enduring cultural traits like life satisfaction and trust. Culture is enduring and leads to democracy across nations. - He concludes that "civic culture", tapped by life satisfaction, intrapersonal trust, and support for revolutionary change all link to the number of years a country has been democratic. Viable democracy does not depend on economic factors alone.

Depeteris-Chauvin et. al. 2020

- Authors argue that shared collective experiences such as sporting events can be effective at priming national unity and reducing ethnic misturst - Empirics: Gathered data on all official matches played by men's national teams of sub-Sarahan countries from 1990-2015 and utilized Afrobarometer surveys administered within 15 days of each match - DV: Measure of ethnic ID that measures relative importance to national identity. Also use measures of trust in "fellow countrymen", trust in other ethnic groups, and ethnic neighbor preferences. - Individuals interviewed after victories are 5.3 pp less likely to report a strong sense of ethnic identity. Victories in friendly matches do not matter, while matches against traditional rivals matter more. After a national win, intrapersonal trust with compatriots increases and respondents are more welcoming of ethnic outsiders of neighbors. - Use RDD to test the effects in teams that barely won or lost entry into global championships. Teams that barely qualified experienced a reduction of 8.6 pp in the number of conflict episodes

Miller 2015

- Autocrats suffer from a lack of information about citizen opposition, but they need that information to prevent upheaval. Elections provide that information and allow autocratic leaders to make policy concessions that prevent revolt. - Semi-competitive elections are a near-universal feature of autocracies - Recent literature has focused on two reasons for elections: Shows of strength and opportunities for patronage - Autocrats want to also know which groups are disaffected so they can be targeted with policy - Patronage actually allows for a signal of credibility because it forces opposition to have some cost - Qualitative case study: Singapore. The PAP never loses elections, but the relative loss is a signal of discontent. In 1991, three seats were lost in Chinese working-class districts so they were targeted with policy - Cross national studies. IV: Electoral change. DV: Education and social spending. Greater losses lead to increased education and social spending. However, this only happens in resource-poor states. Resource rich states are unresponsive.

Calvo and Murillo 2004

- Both supply and demand influence the type of policies that parties pursue - Supply: Political biases in fiscal and electoral institutions that regulate the distribution of resources - Demand: Patronage as a distributive mechanism that provides different returns to voters with different skills - Three elements helped the Peronists in Argentina to achieve a partisan advantage in their access to fiscal resources: (1.) geographic distribution of the Peronist vote, (2.) majoritarian bias in electoral rules which restricts third parties in provinces where PJ vote is, and (3.) fiscal federal institutions that favor PJ providences - Three steps: - First, they show that geographic distribution provides an advantage for PJ. DV: Expenditures received by province, IV: Peronist vote share - Second, They show that public employment provides better returns to Peronists. DV: % of congressional votes obtained, IV: Incumbent gov, median income, public employment, public expenditures per capita. PJ benefits more than competitors from patronage spending. - Finally, they show that Peronist supporters benefit with a higher public wage premium

Lust-Okar 2006

- Case study of Jordan - Argues that there is competition in elections, but the competition is not over policymaking but over access to state resources and rules of the game - Voters vote based on access and efficacy, not based on personal views - Electoral observers are less important than otherwise assume - If state resources decline, the distribution will be hard to maintain and elections have the potential to be highly contested battles

Olson 1965

- Challenges pluralist view of politics - Collective action is difficult when k is small and n is large, and collective action is most difficult in large groups - Collective action faces three problems: anonymity, ability to claim one's action does not make a difference, and enforcement - Small groups overcome these obstacles because they are personal - Large groups can overcome these problems by offering selective incentives

Shepsle and Bonchek 1996

- Chapters 8 and 9 discuss coordination and collective action - Review of prisoner's dilemma: Individuals will not cooperate unless there are an uncertain number of games - But cooperation is made more or less likely due to internalized values, external enforcement - Collective action happens when individuals need to reach a common goal - Reinforce Olson 1965

Aldrich 1993

- Classic equation of voting: V = pB - C + D - Author says that the p and B term will always be minuscule, so the calculus will be carried by the D term. - Turnout is not a particularly good example of collective action problems since it will always be low-cost, low-benefit - The actions of strategic politicians explain why turnout is higher in close elections even though people ignore the P term - The D term has been overly narrowly interpreted in ways that remove it from politics - We should see the D term as relating to longer term political values and preferences

Auerbach and Thachil 2018

- Clients play a meaningful role in selecting brokers and select them on the basis of (1.) efficacy and (2.) distribution potential - Challenge conventional wisdom that coethnicity is most important - Forced-choice conjoint survey in Indian slums for slum president - Ethnic identity is salient and residents prefer coethnics, however, this can be overridden with extremely strong preferences for capability (proxied by education) - Mixed education for how much residents favor bureaucratic connectedness (occupation)

Lipset 1959

- Compares average level of wealth, degree of industrialization and urbanization, and level of education between democratic and nondemocratic countries. More developed = more democracy. - Mechanisms: Higher status for low income individuals means longer time horizon. Development changes the class distribution. Development may change norms. - Democratic systems are most stable when citizens face cross-pressures by multiple social forces - Democratization looks unlikely in Asia and Africa because of poverty, low education, class structure, and role of leftist parties. Latin America is similar to Europe therefore more successful.

Geddes 1990

- Concerned with the phenomenon of selecting on the DV: Without looking at cases where the event in question did not happen, you cannot get a full picture of why a phenomenon occurred - Two mistaken inferences when looking at subsample of cases: (1.) mistake of assuming that any shared variable is a cause, (2.) one can assume there is a relationship in the same that is not reflected in the population as a whole - Reanalyzes Skocpol (1979) and find that it does not hold in LA

Brady and McNulty 2011

- Costs matter in the traditional voting equation, but by how much? - In LA County, precincts were reduced and randomly redistricted - Authors hypothesize that changing polling places increases transportation and search costs - Obtained polling data from 2002 and 2003 elections - There is a 3.03% decline in polling place voting when respondents have their voting location changed - Those who had polling places moved further away voted less, suggesting transportation costs were a large portion of the costs - Democrats were more affected than Republicans, but partisan change is about .22%. Although small, this could impact about 1 in 200 elections.

Swidler 1986

- Culture isa "took kit" of symbols, stories, and rituals that people use to solve problems - Culture's causal significance is not in defining ends of actions, but in providing cultural components used to construct strategies of actions - In the example of the "culture of poverty", values are not enough to causally explain outcomes

Bisin and Verdier 2001

- Culture manifests through childhood socialization - Formal model - Parents have less incentives to socialize their children the more widely dominant their values are in the population - Parents prefer children with their own cultural traits and will attempt to socialize them with this trait

Boix 2003

- Democracy prevails when either economic equality or capital mobility are high - Economic equality promotes democracy because redistribution becomes less costly to elites - Capital mobility drives down taxes, which again makes democracy less threatening. - This theory of capital mobility contributes to the literature in other ways: (1.) clarifies why the "resource curse" is there, (2.) explains why economic development is associated with democratization: as the country industrializes, it depends less on agriculture - Calculates the yearly probability of democratic transition/breakdown as a function of income distributions. Democratization and democratic consolidation are bolstered by high levels of income equality and fair distribution of property.

Lake and Baum 2001

- Descriptively, democracy has a positive impact on well-being - Model: States are forms that produce public services in exchange for revenue. The extent of public goods depends on contestability, which is derived from the availability of alternative rulers and costs of political participation - Empirics: - DV: Various measures of public health and education - IV: Constraints on executive and level of autocracy - Most influential aspect of public services spending is whether the country was a democracy

Hellman et. al. 2003

- Develop a cross-country measure of the capture economy across transition economies - In high-capture economies officials have created a private market for the provision of typically public goods, in low capture economies there are limited advantages - State capture appears more likely in an environment of partial economic reforms - Captor firms must compete against large, influential incumbents - Although captor firms benefit disproportionately, state capture negatively impacts the economy as a whole

Nunn 2008

- Development of slave trade degraded many preexisting institutions in Africa due to increased instability - Author argues that these institutional changes can explain underdevelopment today - IV: Total number of slaves taken from country. DV: Contemporary income. There is a significant relationship between slave exports and income. - Results were not due to less successful countries participating in the slave trade or domestic slavery - IV regression: Instrument is distance from each African country to slave-demanding location - Mechanism: Strong relationship between ELF and slave trade. Slave trade likely caused precolonial state underdevelopment.

Guriev and Treisman 2019

- Dictatorships that depend on violence and repression to survive are dying and in their place are "informational autocrats" who maintain power by manipulating information - Key to their theory is the gap in political knowledge between the "informed elites" and general public - Mechanism: Economic modernization (especially increased education), which makes it more difficult to rule through repression alone. - Descriptively: -- There is less imprisonment and state political killings -- New autocrats are much less likely to have state ideologies -- Democratic institutions such as legislatures and elections are more common - Conduct text analysis on dictator speeches and find that new informational autocrats are less likely to use violent rhetoric and more likely to talk about policy - Using Gallup, public is much more likely to say media is unbiased when freedom is low. In flawed democracies, elites are much less likely to approve of leadership than the general public.

Htun 2004

- Different political remedies exist that are logically appropriate for different groups - Quotas suit groups whose boundaries crosscut partisan divisions - Reservations create incentives for the formation of group-specific parties and suit groups whose boundaries coincide with political cleavages - Empirics: Profiles systems of all countries and shows that overwhelmingly, quotas exist for women and reservations exist for ethnic minorities

Banerjee and Iyer 2005

- Districts in India where the collection of land revenue was assigned to landlords underperform districts where this system was avoided. - Under landlord systems, inequality increased and even presently have lower Gini measures of land equality. - Places that didn't have a landlord were easier for colonists to extract from, so they also received greater investments

Ross 2001

- Does oil hinder democracy? - There are three potential causal mechanisms: -- Rentier effect: Resource-rich governments use low tax rates and patronage to reduce accountability. Three subcategories: (1.) taxation effect- less taxes. (2.) spending effect- greater patronage. (3.) group formation effect- government rents to prevent group formation -- Repression effect: Resource wealth enables greater internal security -- Modernization effect: Growth based on oil fails to bring about social changes - Empirics: -- DV: Regime type -- IV: Oil and minerals -- Oil and minerals have strong antidemocratic effects, and the effect is larger per additional barrel in countries with less oil -- To test the rentier hypothesis, he uses indicators for taxes, government consumption, and GDP from government. These coefficients confirm the theory. -- To test repression, they use variables on military spending. Oil leads to higher military spending, but minerals do not. -- To test modernization, he uses classic modernization variables. No evidence for relation between oil and any but occupational specialization.

Collier and Hoeffler 2004

- Empirical measure of economic opportunity: (1.) extortion of natural resources, (2.) donations from diasporas, (3) subventions from hostile government - Measure costs of rebellion by income per capita, male schooling, growth rate of economy - Measures of grievances: EF, political repression, political exclusion, inequality - DV: Civil war - Primary commodity exports are significant, as are foregone earnings, military advantage - Grievance variables do not matter

Huber 2017

- Ethnicity and income based promises solve the credible commitment problem because they promise group retribution - Parties representing larger minorities will be disadvantaged because they can offer less to their supporters than can parties representing smaller majorities (because the share of spoils is determined by the denominator) - As the number of nonrich grows smaller holding ethnicity constant, class parties should be more likely. As the number of those in the ethnic majority grows smaller relative to the nonrich, ethnic parties will be more likely - Contrary to models of redistribution that say it is most costly to elites when inequality is high, he says that elite expectations about the cost of democracy depend on the interaction between inequality and ethnic diversity

Wedeen 2010

- Ethnography has long been considered too unruly in comparison with other methods, but it is important to understand the lives of people under study - Four characteristics that all interpretative social scientists share: (1.) view knowledge as historically situated and entangled in power relationships, (2.) they are also constructivists in that they see the world as socially made, (3.) eschew individualistic assumptions behind the RC and behavioralist literature, (4.) particularly interested in language and symbolic systems

Gerber, Green, and Larimer 2008

- Experiment on social pressure and turnout - Treatment arms: Civic duty flier, researchers studying turnout flier, sent personal turnout flier, neighbors sent turnout flier - Social pressure boosted turnout the most - Social networks matter for collective action

Ansell and Samuels 2014

- Explanations based on the Meltzer-Richards model can't explain the many cases of democratization when inequality is high - They argue that there is a tension between autocracy and property rather than a threat to property under democracy - Democracy leads to property rights - Regime change results when rising elites seek to reign In the power of autocratic elites to expropriate their income and assets. - In terms of land, equality supports democracy. High landowning inequality is a proxy for the strength of a conservative elite who don't want to share power. - In terms of income inequality, democracy is more likely to emerge when rising groups accumulate income, therefore inequality will be relatively high - Holding land equality constant, increasing income inequality leads to higher prob of democratization. Holding income inequality constant, the more unequal the land distribution the less likely a country will democratize. - Using surveys from the WVS, they show that cross nationally, as inequality increases demand for redistribution declines

Wantchekon 2003

- Field experiment in Benin - Treatment: Type of message received from party (clientelistic or public goods message) - DV: Vote for type of candidate - On average, the clientelist treatment had a positive impact, while the programmatic one had a negative impact - Women prefer public policy appeals more than men, which might be due to the fact that they are more connected - Public policy appeals are less effective in the poor North - In conclusion, the party strategy is determined partly as a function of demand

Olken 2007

- Field experiment in Indonesia - Treatments: Localities told they would be audited, individuals received community meeting invitations, or they received invitations plus comment cards - DV: Difference between what villages claim they spent and what they actually spent, as assessed by a team of engineers. - On average, 24% of expenditures could not be accounted for - Audits had substantial negative effects on unaccounted expenditures by about 8 pp - Almost all of unaccounted expenditures were due to differences between reported and actual quantities, not misreported price of goods - One reason audits might have not done more is that they did not bring criminal charges - Audits also missed nepotism, which increased in audited areas - Community meeting invitations also reduced corruption, but not as much, and mostly when it came to wages, suggesting that community monitoring was driven by self interest

Robinson and Seim 2018

- Field experiment in Malawi on driver bribes - Treatment: Indicators of wealth or political power - Political connections insulate citizens from bribery, particularly among low SES drivers. It impacts both frequency and size of bribes requested - Signals of wealth alone have no effect on the likelihood of bribe solicitation or size, but high-class vehicles are stopped less often, potential due to the conflation of political power - Study suggests that corruption is strategically applied by political actors

Chong et. al. 2015

- Field experiment in Mexico where districts received fliers that told them percent of resources the mayor spent for corrupt purposes, the percent spent on the poor, or no flier - IV: Dummy for treatment, dummies for level of corruption uncovered - DV: Various measures of beliefs about institutional dishonesty - Overall, corruption information didn't change people's beliefs. Only when the level of corruption is high was there a change in beliefs because voters already believed that mayors were corrupt. - Informing citizens about high levels of turnout also made them 7 pp less likely to vote - Story about why people turn out

Bjorkman and Svensson 2009

- Field experiment on public health clinics in Uganda - Treatment: Series of community meetings where facilitators encourage community members to develop a plan on how to improve service delivery - DV: Whether the intervention increased the quantity and the quality of health care provision - Those in the treatment group more likely to have suggestion boxes, numbered waiting cards, information on free services. Intervention changed provider performance, and doctors were more likely to use equipment, patients didn't wait as long, there was less absenteeism, and treated clinics were cleaner. This suggests that bureaucrats may perform better when they are monitored - Better community outcomes as well: citizens more likely to use the clinic, less likely to self-treat, and had better health outcomes including lower child mortality

Mauro 1995

- First empirical test of the impact of corruption - Uses expert survey data on corruption and institutional efficiency - All indicators of institutional efficacy are correlated: It's likely that this is because corruption happens where red tape slows down bureaucratic procedures - Strong relationship between bureaucratic efficiency and political stability - Uses ELF as an Instrument to escape endogeneity concerns - Corruption is strongly negatively associated with the investment rate - Bureaucratic efficiency index is negatively correlated with low growth and the main channel through which this occurs is that bad institutions effect the growth rate by lowering the investment rate

North and Weingast 1989

- For economic growth to occur, the government must make a credible commitment to property rights. To do so credibly, he must either set a precedent of responsible behavior or bound himself by institutions. - Case study of the development of property rights in England - The Crown used to violate property rights because the immediate payoff was higher than the consequences. However, after the Civil War, it the Crown could no longer renege because of institutional threat. - To show that the commitment was successful, they show that the willingness to lend to the Crown greatly increased with the curbing of arbitrary powers.

Acemoglu and Robinson 2005

- Four main paths to political development: (1.) Transition to democracy then consolidation: Example: Britain. Following the Civil War, Parliament was strengthened and monarchy was constrained. Since 1832, strategic suffrage concessions were made to prevent unrest. (2.) Transition to democracy, but quick collapse with subsequent cycling. Example: Argentina. Democracy created in 1912, destroyed 1943, created 1946, destroyed 1955, etc, etc. Dictatorships collapsed because of social protests, democracies collapsed because the radical policies they adopted induced coups. (3.) Democracy never occurs because of prosperity. Example: Singapore. Single party dominance since first elections in 1948. Does not have aristocracy, most capitalists are foreigners. (4.) Democracy never occurs because it is too costly. Example: S. Africa. Elites maintained apartheid because economy depended on it. Model: - Democracy promises concessions not only today but tomorrow. Both democratization and coups transfer de facto power to de jure power. Factors that influence democratization: - Civil society necessary for both - Shocks increase the chance of both democratization and coups - Land-based elites make democratization and consolidation more difficult. Coups are less costly when elites are landowners v. human capitalists. - Democratic institutions can be structure to make the majority less threatening, which increases chances of both. - Inverted U shape between democratization and inequality. Democracy has best chance at middle levels. - Middle class can either drive up democracy or halt it depending on their position relative to the median voter.

Kitschelt 2000

- From Aldrich (1995), parties have to solve both a collective action problem and a social cost problem. To solve the first, politicians have to band together to advertise. To address the second, parties produce a policy platform to make citizen choices easier. - When neither are solved, parties depend on charisma. When they solve only the second, they build legislative caucuses and factions. When they address the first, they are clientelistic. When they address both, they are programmatic. - Contrary to other claims, clientelism creates close accountability - Clientelism and programmism are difficult to combine because programmatic focal points that solve social choice problems are usually grounded in universal ideology that goes against particular benefits. Clientelistic politicians also have little incentive to implement programs - Modernization theory holds that parties are clientelist during early stages of modernization because poor and uneducated citizens discount the future, rely on short causal chains, and prize instant advantages - Electoral laws and the personalization of candidate competitions can facilitate clientelism, but there is not a tight empirical link between them - Presidential systems more likely to be clientelistic because they personalize competition, deemphasize programs, constrain legislatures, and maintain loyalty through inducements - Ethnocultural parties more likely to be clientelist

Anderson and Ross 2013

- Haber-Menaldo analysis from 1800-1970s is misleading because oil production didn't start en masse until 1940s. Oil production mostly occurred privately until the nationalization wave of the 1970s, where there was the creation of "rentier states" - In a regression, when they include only an oil income variable there is a positive relationship between oil and democracy. However, when they interact total oil with a post-1980 dummy variable, they find that there was a negative effect of oil after 1980.

Kalyvas 2000

- How should religious parties be incorporated in democracies? - Process: Democracies face religious parties that want to subvert democracies. Incumbents must accept results or subvert election. Challenger must make a credible commitment that they will not end democracy. - Two cases: Algeria and Belgium - In Algeria, FIS was set to win election and threatened to win Islamist rule. Military coup intervened. Consistent conflict between moderates and radicals meant that there wasn't a credible signal from party. - In Belgium, Catholic party won elections and results were allowed to stand. Although there were moderate/radical conflicts, the Church intervened on the side of the radicals and the party sent a clear signal of compliance. - Difference in religious responses due to three factors: (1.) strong organizational organizations are good at ingroup policing (2.) young parties easier to control (3.) religious messages contingent on party leadership - Religious moderation is a path dependent process

Nunn and Wantchekon 2011

- In areas heavily exposed to the slave trade, norms of mistrust toward others were more beneficial than norms of trust - Mistrust developed because individuals turned on each other during the slave trade - IV: Slave trade, Dv: Norms of trust in relatives, neighbors, coethnics, local institutions. Slave trade participation reduces trust. Also report IV estimates using distance from coast. - Two potential mechanisms: (1.) slave trade could have caused cultural rules of thumb to evolve, or (2.) slave trade could have resulted in deterioration of political institutions, which leads to lower trust. They find evidence for both mechanisms. - Norms of mistrust persist across ethnicities rather than geographic locations, as tested by groups that have moved

Huber 2012

- In majoritarian systems, small geographic groups have stronger incentives to vote cohesively in an attempt to have influence. - In PR, forming a party is easy enough that multiple parties may appeal to the same ethnic group - Ease of party formation decreases ethnicization - Uses international surveys to find that measures of group based polarization and fractionalization decrease in PR systems

Gandhi and Przeworski 2006

- In order to thwart rebellion by elites, dictators will use legislatures to distribute rents and policy concessions - Formal model: Dictator chooses policies and distribution of rents, then opposition decides whether to rebel. -- Magaloni (2008) has a problem with this because there's a credible commitment issue -- Three equilibria: (1.) when the opposition is weak, the dictator has leeway to choose his own policy/rents (2.) when the opposition is strong, the dictator makes extensive compromises and enough rents to prevent rebellion (3.) when the opposition is weak but the dictator cannot inflict much damage, he will offer few compromises and the opposition rebels - Argue that the number of parties is an indicator of policy concessions. - Rentier states don't need as much cooperation and therefore don't need as many parties - Empirics: - DV: Number of parties - IV: Mineral rentier state, type of government (civilian, military, monarch), policy polarization (religious frac.), probability opposition overthrows (# parties inherited and # of past transitions) - Availability of minerals reduces need for parties, monarchs need fewer parties, dictators tolerate more parties when inherited, past transitions make dictatorship more repressive

Alesina et. al. 2013

- In places that practiced plow agriculture there developed a specialization of production along gender lines, which cemented attitudes about the role of women for decades to come - Culture might persist for three reasons: (1.) underlying cultural traits are reinforced by institutions, (2.) culture can remain intact because of complementarity between cultural beliefs and industrial structure, (3.) cultural beliefs are inherently sticky. - First, associate plow use with women's roles: DV: Traditional participation in agriculture. IV: Traditional plow use. They find a negative relationship between plow use and women's participation. - Then, they connect plow use with contemporary outcomes: Traditional plow use is associated with less female labor force participation and less female firm ownership - Reaffirm results using suitability for plow use as an instrument - Finally, they measure attitudes among children of immigrants: There is a negative relationship between tradition of plow use in the home country and participation in the labor force

Greif 2002

- In the 12th century, institutions developed that allowed for trade despite lack of a centralized authority known as the community responsibility system - If a member of community A had defaulted on a contract with a member of community B, each member of community A was held liable - Uses historical documents to prove the existence of the CRS. Exchange with a separation between the quid and quo were common in Europe. Merchants were easy to identify with comunities - However, the CRS was its own downfall. It became easier to falsify community ID, and the system attracted risky traders - Article indicates that economic institutions supporting market exchange can be based on and positively reinforce a particular social structure

Amat and Beramendi 2016

- In traditional accounts of inequality, inequality is a barrier for poor people to vote, but descriptively this is mostly true among middle income countries - They examine the conditions under which parties choose mobilization via clientelism or programmatism - Descriptively, there is a strong negative relationship between clientelism and turnout inequality, and a strong positive relationship between programmatic competition and turnout inequality - Utilize the natural experiment of Brazilian anti-corruption reports - Treatment: Whether corruption made public before elections - They hypothesize that when audits are released, clientelism will cease to be viable and there will be a switch to programmatism and there will be a reduction in the level of turnout - DV: Change in levels of turnout between 2000-2004 - IV: Degree of inequality within municipality, type of political strategy adopted, treatment dummy - Audit release had a negative impact of turnout especially when mismanagement revealed and inequality were high - Story about turnout: Clientelism increases the level of turnout.

Lindbeck and Weibull 1987

- Individuals have probabilistic party distributions known by parties - One aspect of model is that party politics differ according to social identities and that groups with more uniform party preferences are more influential than those with dispersed party preferences - Party strategy is not informed by the median voter, but by the distribution of preferences in the electorate due to social identities - Parties will target swing voters because they can be persuaded for the least cost

Ferraz and Finan 2008

- Information sharing among voters is an important way to constrain politicians - Natural experiment in Brazil where the results of an audit were released before or after an election - The electoral performance of mayors about whom information was released prior to election was no worse than those whose performance was not shared, but this was conditional on the level of corruption revealed - Interpretation of information about corruption was influenced by prior beliefs of voters - Information release about corruption had the largest impact when the municipality had a local radio station

Miguel 2004

- Kenya and Tanzania were divided somewhat randomly for administrative colonist purposes - Since then, they have pursued radically different policies. In Tanzania, there is a national language, the educational system emphasizes history and culture, overhaul of local government. In Kenya the colonial government remains and presidents tend to be coethics. - Tanzania might have better collective action outcomes for two reasons: (1.) as collective identity increases, taste-based preferences are less important, (2.) reforms increase social interactions and increase ingroup policing - Empirics: Use ELF to measure ethnic diversity. Estimated relationship between ethnic diversity and local provision of goods is negative in Kenya, but positive in Tanzania - Qualitatively, schools in Kenya struggle to be funded because of ethnic competition while stronger social sanctions are imposed in Tanzania

Granovetter 1978

- Knowing norms and preferences not enough to predict collective action - Must know distribution of norms throughout group - Individuals have threshhold preferences - Change the distribution of threshhold preferences and you change the collective action - People might also have different thresholds when they know those who are rioting

Gans-Morse et. al. 2014

- Machines engage in different clientelistic strategies depending on the electoral environment - Types of clientelism: Vote buying, turnout buying, abstention buying, double persuasion, rewarding loyalists - Each individual has a different buyoff strategy and cost - The machine should only target citizens who offer the most net votes per dollar, so they will pay twice as much for vote buying than other forms - Machines tailor their mix to five aspects of the political environment: (1.) compulsory voting, (2.) strength of ballot secrecy, (3.) salience of political preferences, (4.) political polarization, and (5.) level of machine support

Rauch and Evans 2000

- Making entry to the bureaucracy conditional on passing an exam or getting a university degree, or paying people the same as the private sector will produce capable individuals - Empirics: Expert surveys on a sample of 25 countries. IV: Merit based promotion, public salary equal to private salary. DV: Bureaucratic performance, corruption - Salary doesn't determine performance as much - The element that is most important for improving performance is meritocratic recruitment - Causal interpretation: Meritocratic recruitment causes good bureaucratic performance

Richards 1986

- Military authoritarian breakdown in LA was due to the failure of stabilization policies implemented by the military - When the open-market reforms in LA failed, the military stepped asside - Includes Peru, Argentina

Dixit and Londregon 1996

- Model where two parties vie for support. Individuals within groups have heterogeneous preferences but the distribution of preferences is different for each group. - When groups have the same ability to redistribute, three groups will be targeted: (1.) groups with relatively many moderates who can be persuaded with economic benefits (2.) groups that are relatively indifferent to party ideology relative to private consumption benefits (3.) low-income groups more willing to compromise political preferences - However, the ability of parties to redistribute will sometimes differ based on the amount of political knowledge, eg with party machines. In this case, core voters will be targeted.

Fujiwara 2015

- Natural experiment with electronic voting machine access in Brazil. Only some districts had access to voting machines. - 23% of Brazilians are illiterate and only written ballots are used so many votes aren't counted - Main outcome: Number of votes that were valid ballots - Partisan vote share moves left when eV is used - The effect of EV is stronger in municipalities with less educated populations - Places with EV spent more on health care and had better health outcomes, including prenatal visits and fewer underweight babes

Dunning 2008

- Natural experiments are experiments in which the treatment comes from naturally occurring phenomena and the researcher can make a credible claim that the assignment is "as-if random" - Author advocates for creating a "continuum of plausibility" that spans from pure experiments to observational studies. Natural experiments can be placed on this continuum based on the plausibility that their treatments are "as-if" random.

Chattopadhyay and Duflo 2004

- Policymaking gender bears on different policy outcomes and leaders will invest more in infrastructure that is relevant to the needs of their own - Natural experiment in India where some women randomly assigned to be leaders - In places where the highest leader is a woman, there is more participation by women in the lower political system - In both states, the gender of the Pradhan impacted the provision of the public goods - Takeaway: Gender of leaders matters for policy provision

Burgess et. al. 2015

- Presidents allocate more to coethnics, however, they will become more constrained in this bias as constraints on the executive increase. - Democracy will bring less ethnic bias because there are freer flows of information, a vocal civil society, and an independent parliament - They use data on road provisions in Kenya to show that in times of democracy, there is less ethnic favoritism in road building

Cantoni et. al. 2019

- Previous literature on collective action reviews free rider problem versus "strategic complementarity", where participation is more likely when turnout is higher because there are lower costs to participate - Field experiment in Hong Kong - Treatment: Individuals given information about anticipated protest turnout - Among those who thought turnout would be higher, turnout was less likely. Those who believed turnout would be lower were more likely to turn out after the new information. This suggests a "strategic substitutability" attitude toward collective action.

Grossman and Michelitch 2018

- Previous scholars have highlighted the importance of information in making politicians accountable - Competitive environments are important because the important will not have an incentive to change if there are no real challengers -Field experiment in Uganda - Treated districts had information widely disseminated about incumbent performance - Outcome measures: politician scores on NGO scorecards, performance evaluations from fellow politicians and bureaucrats, school grant outcomes - The transparency ID program significantly increased politician performance, but only in competitive constituencies. There was no overall treatment effect

Brierley 2020

- Previous studies assume politicians act to constrain corrupt bureaucrats - They argue that in many countries politicians use oversight (transfers) of bureaucrats to get bureaucrats to engage in corruption on their behalf - In Ghana, this happens through awarding building contracts to political allies through restricting access to applications or giving inside information - DV: Level of corruption reported by bureaucrats - IV: Political discretion, proxied by politicians' ability to transfer bureaucrats - Finds that the more discretion there is, the more corruption - Uses a list experiment to confirm the mechanism behind the corruption-discretion mechanism: Bureaucrats don't speak out against corruption because they fear being transferred - Potential policy responses: Place strict movements on transfers, and incentivize public servants to work hard to develop a longer time horizon

Huber and McCarty 2004

- Previous theories of bureaucrats and legislatures emphasize the tradeoff between bureaucratic expertise and political control, but in many contexts these bureaucratic characteristics do not exist - Uses a formal model - Politicians are harmed by low capacity bureaucrats because of efficiency losses - The best action that a bureaucrat can be induced to take improves for the politician as capacity increases - when we ignore the impact of capacity, the politician always prefers a bureaucrat with an ideal point closer to the politician, but if we take capacity into account, the politician might prefer an opposite but capable bureaucrat because they are easier to control

Lupu 2013

- Previous theories of partisan attachments are that they are durable social identities over time. However, this might not be reflective of developing democracies, where political party identities are more fluid - Develops model of partisanship based on "comparative fit": Individuals self-categorize not only based on resemblance but also based on differences from other groups - Suggests that when party platforms converge, partisan attachments should decrease - Survey experiment - Voters given no info, party platforms, interparty alliances, or both types of information - DV: Whether individual has PID - Those in the policy group were much more likely to have PID compared to control, and those in alliances group had much less - Mechanism behind results is the perceived differences in party platforms, as evaluated by a 0-10 left-right rating by respondents

Posner 2004

- Previously, the most common indicator of ethnic fractionalization was ELF, which produces the likelihood that two people chosen at random will be from different ethnic groups. There are three shortcomings to this approach: (1.) underlying data is suspect, (2.) summarizing country-wide data obscures regional diversity, and (3.) there is a mismatch between the measure and what researchers are actually trying to estimate. - The problem is that ELF includes dozens of groups in each country that might be culturally different that are politically irrelevant - A good index of fractionalization must reflect groups that are actually competing over policy - Author constructs a new measure (PREG) for 42 Afriacn countries, to be used specifically for macroeconomic DVs - Replicates a previous paper that used ELF to claim that ethnic diversity leads to macroeconomic policies that lead to growth outcomes. He notes that when Africa is subsetted, there is no relationship between ethnic diversity and macroecon outcomes. However, using the PREG measure the relationship reappears.

Epstein et. al. 2006

- Przeworski et. al. (2000) find that modernization is not a causal factor in democracy, it just prevents democracies from backsliding. - First, they critique Przeworskii's interpretation of their coefficients and show that a correct interpretation leads to the conclusion that GDP influences transitions to democracy but not transitions to autocracy - They also claim that using only a dichotomous measure of democracy leaves out an important category: Partial democracies. They show confirm with this new measure that modernization impacts the probability of transitions both to and from democracy.

Gonzalez-Ocantos et. al. 2019

- Qualitative methods have long documented widespread clientelism, but surveys about experiences reveal little. The authors argue that this discrepancy is due to social desirability bias - To circumvent these concerns, they used a list experiment in Nicaragua - Asked how many activities occurred out of a list of campaign activities, treatment was whether vote buying was included - Demonstrated that there are strong norms against vote buying: 70% said people would not report, 75% said they felt it was unethical - 24% of respondents revealed corruption through list experiment compared to 2.4% who reported when researchers asked directly

Kuran 1991

- RR doesn't predict upheavals well, but explains why they are rare - Structuralism treats as significant the structural side of the revolution - Individuals hold both private and public preferences (difference between is preference falsification) - Whether you participate is due to a tradeoff between external payoffs and internal payoffs, switching point is revolutionary threshold - Every revolution is a surprise as a result of preference falsification - In E Europe, few individuals spoke up and were repressed, alienated individuals did not know how wide alienation was shared, once some opposition to government was allowed then the dam broke

Tversky and Kahneman 1986

- Rational choice assumptions often do not hold: -- Cancellation: No preference between states of the world that yields the same outcome regardless of choice -- Transitivity: One's preferences is not relative to other choices -- Dominance: If one option is better than another in one state and at least as good in all other states, the dominant option should be chosen -- Invariance: Different representations of the same choice should yield the same preference - Demonstrate that invariance and dominance can be violated through framing effects

Wedeen 2002

- Reconceptualization of culture as "semiotic practices" in which symbols are mutually created and interpreted - Traditional treatments of culture underemphasize heterogeneity - Ethnicity should not be taken for granted, but as a construction. Authors could recognize this by constructing a database in which degrees of ethnicity are noted - Culture is not primordial

Putnam 1993

- Regional variations in culture drive the success of institutions. People from civil regions support equality, compromise, belong to unions, feel less powerless, and say their leaders are less corrupt - Quant evidence: Civil structure is a strong predictor of subsequent economic performance, holding civic transitions constant - Networks of civil engagement overcome collective action problems by increasing potential costs to defectors, fostering norms of reciprocity, facilitating communication

Siegel 2009

- Relationship between network size and aggregate participation is conditional on the distribution of individual motivations in the population - Power of social elites depends on the structures in which they reside - Typology of four network type: Small World, Village, Opinion Leader, Hierarchical Network -In Small World, individuals are highly connected and high levels of participation - Village: Clusters of individuals connected by single ties, less efficiency and participation - Opinion Leader: Increasing elites increases participations - Hierarchy model: Homogeneity of elite motivations determines participation

Calmfors et. al. 1988

- Relationship between real wage and centralization is non-monotonic: Real wages will be low and high rates of centarlization and high for medium levels of centralization - High centralization: Finland, Norway, Sweden - Low centralization: US, Japan, UK - Centralization is defined as "the extent of interunion and interemployer cooperation in wage bargaining with the other side" - Middling levels of centralization lead to the worst economic outcomes - They also try to replicate studies substituting their corporatism measure in for centralization and find that centralization is not an empirical drive of the results around corporatism

Berman 1997

- Robust civil society in Germany helped the rise of the Nazi Party - Critiques Tocquevilliant theories of democracy that say that civic culture is inherently good - Case study of Weimar Germany - Nazis used Germany's rich associational network to capture important sectors of the German electorate

Calvo 2009

- Rokkan (1970) and Boix (1999) argued that PR rose in places with a Socialist threat, but it fails to explain PR reform in places without socialism - Author wants to provide a self-interested rationale for the move away from majoritarian rules with or without an emerging Socialist threat - Multiparty competition under majoritarian rules increases the severity of partisan biases. When a new party arrives and draws votes, there will be a more severe partisan bias. Territorially concentrated minority parties will obtain more seats than territorial dispersed ones - When new parties enter majoritarian systems, there are large shifts in how many seats are obtained even with small vote changes. But under PR, each new party reduces massive swings. - Data: Allocation of seats and votes for every national legislative election in W European countries - After the intro of PR, there was higher predictability of seat share in comparison to vote share

Fearon and Laitin 1996

- Scholars of interethnic relations tend to emphasize violent - Two instances in which much violence has been profiled and little has occurred: Russia and Africa - Two potential solutions emerge that limit violence - Spiral regime: Members of group A indiscriminately punish members of group B because of member B - In-group policing: Individuals of group A keep interacting with group B, but group B punishes their own - Individuals cooperate for fear of losing future payoffs if they defect and cause a breakdown of relations - The more that individuals interact outside their group, the less effective the threat of sanction - When group A is much larger than group B, threat of spiral punishment cannot be used effectively by group B, but the threat is credible from small groups to large ones

Dunning et. al. 2019

- Seven independent research teams coordinated on the design of randomized trials in developing countries that exposed information about incumbent performance - Hypothesized that the effect of information would depend on what voters already believed - Instead, they did not find that either positive or negative news had any treatment effect

Skocpol 1979

- Social revolutions are rapid transformations of simultaneous societal and political transformations - Four approaches to revolutions: (1.) Marxist- revolutions come out of class-divided modes of production. They are not isolated episodes, but class based movements coming out of structural conditions. (2.) Aggregate psychological theories: A la Gurr 1970. Political violence occurs when there is widespread relative deprivation. (3.) Systems-value consensus: Stable society is one in which institutions reflect values. When new values or technologies challenge the system, it must resynch through revolution. (4.) Political conflict theories: Level of discontent doesn't matter, what matters is that people are members of organized groups. Empirics: - Case study of France, Russia, and China with comparative historical analysis - In all revolutions, externally mediated crises combined with internal conditions to produce (1.) the incapacitation of the central state machineries of the Old Regimes, (2.) widespread rebellions by the lower classes, most crucially peasants, and (3.) attempts by mass-mobilizing political leaderships to consolidate revolutionary state power.

Trejo 2009

- Spread of religious competition in poor, unequal multiethnic societies traditionally dominated by a single religious supplier is an important causal mechanism in why religions mobilize poor and minority citizens - Empirics: Dataset of indigenous insurgent protests. Religious competition is a strong predictor of indigenous mobilization. Religious competition also drives ethnic claims and overall ethnic mobilization - Qualitative evidence: 3 religious community members that experienced or didn't experience need to recruit indigenous members

McMillan and Zoido 2005

- Study of Montesinos bribery records of politicians, judges, media, etc - The most expensive actors to bribe were media outlets, while the least important were judges - Although no single judge or politician could bring the system down, a single media actor could (and in fact, they did in the end) and thus they had a stronger negotiating position

Besley and Burgess 2002

- Study on food provisions and flood relief in India - Positive correlation between newspaper circulation levels and measures of government responsiveness - Greater political competition is associated with higher levels of public food distribution - Local language newspapers have an effect on government responsiveness, but not English or Hindu newspapers

Chang and Golden 2007

- Study whether different electoral systems have systematically different effects on the degree of corruption observed - hypothesize that political corruption becomes more severe as district magnitude increases under open-list PR systems because in open-list politicians need a large amount of financial resources in order to compete against others in the party - Examine this relationship both cross-nationally and in Italty - Cross-nationally, use the CPI as DV. IV: District magnitude and open/closed electoral system - Closed list PR is more corrupt than open-list PR when district magnitude is small, but once magnitude gets sufficiently large, open-list PR systems are more corrupt - Presidential systems are associated with greater perceived corruption - Subnational evidence from Italy: Until 1994, Italy used open-list system. DV: Difference between money allocated to public infrastructure and money actually spent. IV: District magnitude. Larger districts are associated with more corruption.

Alt et. al. 2015

- Survey experiment in Denmark to test the impact of different sources of information on economic voting - Treatment: Treated groups were told that unemployment would either be 5/7 percent and that info came from Central Bank, opposition, or government party - Political sophistication measured as respondent's estimate of the current unemployment rate - All treatments caused respondents to update. All voters updated less from political parties than from Central Bank. - Sophisticated and unsophisticated voters responded differently. Unsophisticated do not distinguish between parties. Sophisticated voters distinguish between parties - Strong evidence for economic voting, but only among those who are politically sophisticated

Haber and Menaldo 2011

- The "resource curse" literature finds that oil and minerals negatively impact democratization - However, these studies do not consider the counterfactual: Without oil, what would happen in these countries? - Empirics: Notably, they only look at cases where countries significantly depend on oil. First, they study cases in which countries were democratic in the first year and compare them to after they found oil. 2/3 remain democratic. Next, they run a regression in which they compare countries with counterfactual countries in the same region that did not discover oil. They find no evidence that oil hinders democracy. Finally, they run a matched regression with countries very similar to those that discovered oil and find that the discovery of oil doesn't greatly impact polity score. - In conclusion, there is a lack of evidence that resources either hurt or help democracy.

Haggard and Kaufman 2012

- The distributive conflict approach conceptualizes democratization in terms of dynamics of inequality. However, the authors say that there is less of a relationship between distributional conflict and authoritarianism as one might think. - Acemoglu and Robinson (2006) do not present empirical evidence, while Boix (2003) runs regressions on land equality and democratization - They conduct a medium-n analysis looking at causes of democratization. To find whether a country had a democratization based on distributional conflict, they argue that (1.) there must be mobilization due to redistributive grievances and (2.) the rising costs of repressing those demands appears to force elites to make concessions. They find that distributive conflicts appears to play a causal role in only about 55 percent of transitions, while 45 percent is attributed to other reasons. - They also look at democratic breakdowns and find that only half occurred in order to prevent distribution - They suggest a number of other causal mechanisms, including elite challenges, international actors, or elite pivots to democratization because they believe they can hold power.

Gandhi and Lust-Okar 2009

- The empirical realities are that authoritarian elections are not rare and autocracies with elections are more durable than those without them - Competing explanations for elections: Coopt party members, signal dominance, coopt opposition, mitigate risk of violent removal from office, informational - Elections can establish policy congruence, but they can also lead to coercion of voters - Research is based on a limited set of studies- Mexico, China, Egypt - Elections could lead to democratization via norms influence, victory for democratic opposition, mobilization outside of electoral arena

King et. al. 2013

- The purpose of Chinese censorship programs is not to suppress criticism but to reduce the probability of collective action - Gathered social media posts before the censors could get to them - Posts with any collective action potential were blocked, whether they were pro or anti government -- Those that relate to protests, crowd formation, related to individuals who have organized in the past, relate to nationalism - Even topic areas that were highly controversial but policy specific were not blocked

Hall and Soskice 2003

- There are two types of capitalist institutions: Coordinated Market Economies (CME; Germany, Scandanavia) and Liberal Market Economies (LME; US, Australia, Canada) - Those two types can be distinguished by the primary way in which firms coordinate with each other and other actors, such as trade unions. In LMEs, firms primarily coordinate by way of hierarchies and market mechanisms. Coordinated market economies rely more heavily on non-market forms of interaction in the coordination of their relationships with other actors. - Focus on five areas in which they differ: industrial relations, vocational training, corporate governance, inter-firm relations, relations with employees - Places an emphasis not just on firms and institutions but also on "common knowledge" (history and traditions)

Akerlof and Kranton 2000

- They argue that (1.) people have identity-based payoffs derived from their own actions; (2.) people have identity-based payoffs derived from others' actions; (3.) third parties can generate persistent changes in these payoffs, and (4.) some people may choose their identity, but choice may be proscribed for others. - The concept of identity explains economic behavior in several ways. It can explain behavior that appears detrimental, it can explain a new type of externality, it reveals a new way that preferences can be changed, and choice of identity may be the most important "economic" decision people can make. - Stems from a model in which identity is based on social categories with prescriptive behavior - Examples of how identity can influence behavior: Piercings, feelings toward gendered work, donations to alumni associations

Powell and Whitten 1993

- They argue that the lack of replication for economic voting in the comparative context occurs because of a failure to pay attention to the electoral context in which citizens choose - Four considerations of political context: (1.) economic performance relative to other industrialized democracies, (2.) swing that governments experience in previous elections, (3.) policymaking context, (4.) ideological image of government - DV: Gain/loss in incumbent vote - IV: Inflation, unemployment, and growth - First, they use measures of economic variables as compared to other industrialized democracies and find all variables have correct effect, although they are not significant - Next addition: taking account of "swing" from past election by using lagged measure of party vote. Find that incumbents do consistently lose previous electoral gain - Third alteration: Inclusion of political institutions variable. The greater the perceived unified control of policymaking by the incumbent, the more likely the citizen will assign responsibility (using variables like committee system, bicameral system, minority government, coalition government) - Finally, they find that right parties are penalized for inflation, nonright governments are penalized for unemployment

Croke et. al. 2016

- They argue that the positive relationship between education and political participation does not necessarily hold under authoritarianism - Conduct an RDD with education reforms in Zimbabwe - IV: Education access - DV: Political participation (contacted local officials, attended meeting, raised issue at meeting, voting) - Education significantly reduces levels of political participation - Citizens who received education are better off in terms of employment, that education leads to greater interest in politics, and that educated respondents are more likely to support democratic institutions in theory - Thus, the mechanism is that education in competitive autocracies reduces participation through both a change in resources as well as a value change

Finan and Schecter 2012

- They argue that vote-buying is sustained in part due to psychological factors such as reciprocity. - Consistent with three theories of clientelism: (1.) reciprocity norms alleviate credible commitment problems, (2.) middlemen choose reciprocal individuals because there is less uncertainty, (3.) vote buying as the outcome of a repeated game - Use surveys in Paraguay - Reciprocity is measured in trust games - 26% of the sample claimed to have been offered something in exchange for their vote, while middlemen claimed that 33% of those surveyed were offered something (and 46% of voters overall) - Middlemen know other village members extremely well. Scored high on knowledge of spouse name, political attributes, political leanings, and social preferences (reciprocity) - A 1 SD increase in reciprocity increases the likelihood of being targeted by 9.6 pp

Acemoglu, Johnson, and Robinson 2001

- Three conditions: (1.) Different types of colonization created different sets of institutions, (2.) colonization strategy was influenced by the feasibility of settlements, and (3. ) colonial state and institutions persisted after independence - Could be a potential explanation for property rights: Setting up institutions to enforce them Is costly and elites want to stick to the status quo. Institutions are sticky. Furthermore, colonization impacts future property rights. - Empirics: They find a strong correlation between institutions and economic performance. Later, they use settler mortality as an instrument.

Habyarimana et. al. 2007

- Three main theorized mechanisms that connect diversity to goods: preferences ("commonality of tastes" and "other regarding preferences"), technology ("efficacy" of communication and "findability"), and strategy selection (in this case, social sanctioning) - Series of lab and lab-in-the-field experiments in Uganda - For "commonality of tastes", they conduct a survey that asks about preferences for collective foods and find no evidence that groups differ in taste. - To test "other regarding preferences", they set up an anonymous dictator game and do not find non-coethnic discrimination - To test "efficacy" mechanism, they have individuals solve a puzzle together, and coethnics do no better together - To test "findability", they have participants "find " a coethnic in the field, and respondents are much better able to find coethnics - To test social sanctioning, they test dictator game when individuals are not anonymous and find discrimination does occur - Findings suggest that coethnics cooperate because they adhere to ingroup reciprocity norms (social sanctioning), which is enforced by increased findability - Coethnics may be better able to achieve collective action

Stepan and Linz 1996

- Three minimal conditions for democratization: (1.) state must exist (2.) democratic transition must be completed with elections (3.) executives cannot violate constitution/rights - "Consolidated democracy": Situation in which democracy is the "only game in town" - Conditions for consolidation: (1.) free civil society (2.) autonomous political society (3.) political actors must be subject to rule of law (4.) must be a bureaucracy (5.) must be institutionalized economic society (not command or free market economy) - Two major obstacles for consolidation: (1.) ethnic conflict (2.) too much hope for economic reform when there are simultaneous economic and political reforms

Moore 1966

- Three routes to modernity: Bourgeoisie revolution, facism, communism - Things that help democratization: (1.) a balance between monarchy and parliament, (2.) a turn toward commercial agriculture, (3.) there must be a revolutionary break from the past, (4.) there must be a weakening of landed aristocracy, (5.) there must not be an upper class revolution against workers - Peasant revolutions will happen when there is an absence of commercialization of agriculture, the survival of peasant social institutions, and weak links between peasants and the upper class.

Simpser 2017

- To isolate the influence of culture away from institutions, the author looks at second-generation European individuals who have been exposed to culture but not to institutions - DV: Attitudes toward corruption (whether bribes are "seriously wrong"), IV: Attitudes compared to non-second-gen country residents - Respondents with ancestors from countries with higher average disapproval of bribery also disapprove more. This result is driven by mothers' attitudes and are not part of a larger package of attitudes toward the law.

Coase 1960

- Traditional economic involvements regarding harm surmised that those who are inflicting the harm should be the ones to pay - Author says that the ultimate result of the payment should consider the marginal costs and benefits to both parties - When transaction costs are low, bargaining between parties will lead to a Pareto efficient outcome regardless of the initial allocation of property - When there are transaction costs, social arrangements might be less clear and might come down to the legal system of rights

Francois et al. 2015

- Traditional models of Africa have assumed a "strongman" theory of the president in which president dominates the political structure - They argue that politics in Africa are actually relatively well-shared, as evidenced by ethnic representation in cabinets - Leaders have a self-interest to represent other groups in cabinets in order to main support from other ethnic groups in the masses and to prevent topple from inside the cabinet - Examine the cabinets of 15 sub-Saharan African countries - African cabinet allocations tend to match population shares - The authors reject the theory that African autocracies are "one man shows"

Campello and Zucco 2015

- Traditionally, voters vote according to the economic circumstances in the country - Argue that this differs in LA because economies are driven by exogenous economic circumstances - Voters vote based on economy but are incapable of recognizing that the economy is driven by outside factors, thus some LA leaders are unfairly punished

Przeworski and Limongi 1997

- Two explanations of the positive relationship between democracy and economic development: (1.) Endogenous (modernization theory): Democracy is more likely to emerge because of economic development. (2.) Exogenous: Democracies develop independent of economic development, but they are more likely to survive in developed countries. - They critique modernization theory, saying that if it were true there must be some threshold under which there is no dictatorship. Empirically, they cannot find it. Furthermore, few authoritarian regimes meet the theory of growing slowly over time. Lipset (1959) says quick growth leads to destabilization, but they find that quick growth is more likely to lead to democracy. - Argue instead for the exogenous theory. Empirically, the probability that democracies die after average income exceeds $6k is almost zero, but they are quite likely to die under that point.

Treisman 2007

- Two major revolutions in the study of corruption: Development of new data and increased studies on the causes of corruption - Lower perceived corruption correlates closely with higher economic development - Democratic Institutions, and in particular political rights, are related to corruption. Presidentialism is also related to corruption. Golden and Chang (2006) open v. closed PR system results are not robust. - May be greater corruption when there are larger economic rents available for capture- fuel export indicator predicts corruption - In terms of data, there are new expert survey indexes but these do not necessarily line up well with actual experiences of corruption

Alesina et. al. 1999

- Two reasons might cause more diversity to lead to fewer public goods: (1.) different ethnic groups could have different preferences over which type of public good to produce, or (2.) each ethnic group's utility is reduced if other groups also use a good - When public goods are not desired, politicians will commit more to private patronage - Empirics: Ethnic fragmentation = ethnic fractionalization index. The share of city spending on roads decreases with higher ethnic diversity.

Chandra 2004

- Voters are motivated by a desire for material or psychic goods and also have information constraints. Thus they favor coethnics in the delivery of benefits and votes. - PR systems with several parties allow small ethnic groups a greater degree of efficacy than FPP systems, which favor large groups - Examines multiple parties in India - Ethnic parties are most likely to succeed in patronage-democracies when they have competitive rules for intraparty advancement and the ethnic group they seek to mobilize is larger than the winning threshhold.

Gilli and Li 2015

- When dictators face credible threats by citizens or by the selectorate, they choose efficient economic policies instead of appropriating private benefits - Two-stage game: (1.) Dictator pays patronage to selectorate, selectorate obtains utility, decides whether to remove autocrat before citizens revolt (2.) Masses either don't revolt, successfully revolt, or unsuccessfully revolt - Different types of regimes respond differently to threats - At very high and very low levels of revolutionary threat, there will not be responsiveness. Autocrats will be responsive to intermediary threats.

Diaz-Cayeros et. al. 2000

- Why do citizens support regimes of which they obviously disapprove? - They argue that narratives of repression are not enough. They do not explain why autocrats are more likely to fall when the economy is poor. - Autocrats distribute to resources to supporters and deny opposition so that it is more rewarding for citizens to support the autocrat and get resources than to oppose the party and get nothing - Argue that the PRI lost in Mexico despite this mechanism for three reasons: (1.) The economic collapse of the 1980s lowered potential rewards (2.) Growing opportunities in the international market raised the opportunity cost of remaining under the PRI (3.) Economic liberalization and greater openness allowed many localities to prosper and become less dependent on the PRI.

Svolik 2019

- Why do voters support undemocratic candidates? - Electoral competition confronts voters with crosspressures: Party or democracy. - Cross-national survey experiment - Randomized candidate qualities, including whether the candidate violated a democratic principle - Ordinary people support democracy and will punish them by up to 35 percent, but the magnitude of punishment decreases when partisan differences are large or society is divided. Respondents more likely to tolerate democratic violations when the individual is of their party - Centrists are a key democratic force. Political moderates punish undemocratic candidates more severely than strong partisans.

Nichter 2008

- With the secret ballot, the issue is not vote buying, it is turnout buying - Implications of turnout buying: Machines focus rewards on strong supporters, they will target the poor, and they offer rewards where they can effectively monitor. - Machines gain votes by offering rewards to unmobilized supporters in repeat interactions - Evidence: Party sympathizers have a higher probability of receiving rewards, individuals with more favorable opinions more likely to get rewards, as are those who have voted for the party before, those with low incomes, and those in smaller municipalities

Kydd and Walter 2002

-Extremists commit violence to create mistrust between moderate groups and governments. The purpose of violence is to exacerbate uncertainty that the deal will not lead to peace. - They present a formal model and test the model using data on the Palestinian- Israeli peace process. Hamas's attacks against Israel have not been random, instead, they have concentrated attacks around major events in the peace process.

Olken and Pande 2012

-Lit review of corruption - Three broad themes: (1.) Although there has been a revolution in measurement, estimated levels of corruption are heterogeneous and there are few conclusions about magnitude, (2.) Corruption is bad for efficiency and equity, (3.) Corrupt officials act in line with economic principles re: monitoring, yet their ability to substitute different forms of corruption makes limiting corruption difficult in the long run

Shayo 2009

-People don't just vote their economic self interest, they also vote their identity. - Support for redistribution decreases with national identification among the nonwealthy - The poor are more likely than the rich to identify with their nation - Across democracies, there is a negative correlation between levels of national identification and levels of redistribution. - Uses the World Values survey to confirm these results

Almond and Verba 1989

-The authors argue that the founding of a stable democracy requires not just institutions, but also a civic culture. -There are three types of political culture: Parochial culture, subject culture, and participant culture - Study culture through surveys in a variety of Western countires

Ross 2006

A Closer Look at Oil, Diamonds, and Civil War - Attempts to estimate the relationship between civil war and resources by using a non-endogenous measure of resources not linked to GDP - Oil and diamond production increase civil war risk through two mechanisms: (1.) fostering insurgencies in resource-rich regions (resources make separatism more valuable, as measured by the fact that onshore oil is linked with conflict but offshore oil is not), (2.) through processes linked to trade shocks (as measured by trade shock interaction variable)

Wood 2001

An Insurgent Path to Democracy - Sustained mobilization from below can transform key interests of economic elites, leading to pressure on the state to compromise with insurgents - Three steps: Mobilization fostered by exclusion deeps to insurgency and produces an "insurgent counterelite" who can help solve a regime crisis -> ongoing mobilization leads some elites to consider compromise b/c of increasing costs -> bargain succeeds when insurgents accept inclusion at the cost of economic moderation - Empirics: Interviews with political actors in El Salvador and South Africa - In ES, labor obtained by coersion. Repression by military caused civil war, which shaped the economic structure of the country by decreasing agriculture as % of GDP and decreasing output. Party for economic elites began to moderate and compromised. International pressure also helped. - In SA, colonial structures led to labor coersion. After increased repression, protests occurred, unions grew and became militant. International invstment declined, as did econ. performance. Business elites banded together to lobby for political reform and negotiated the transition. - In both cases, international pressure reinforced domestic pressure

Haber 2009

Authoritarian Government - Dictators have three options when it comes to controlling launching institutions: terror, cooptation, and organizational proliferation. - Terror: Dictator terrorizes the party organization. Undermines the ability of the government to function. But, dictator can't make credible property rights promises to investors, he will expropriate, and there will be low growth rates - Cooptation: Buying loyalty. Allocate property rights to a selected group of people to create rents, usually done through barriers to entry to limit competition. Impressive short run growth, but not productive in the long run because of misallocation. - Organizational proliferation: Dictator creates many organizations so they are forced to coordinate. Property rights are broadly given to the rank and file. Example: Mexico. This system has the most economic growth in the long run.

Gandhi and Przeworski 2007

Authoritarian Institutions and the Survival of Autocrats - Cooptation model - When autocrats need to neutralize threats from larger groups within society and solicit the cooperation of outsiders, they rely on nominally democratic institutions which incorporate potential opposition forces and give them a stake in ruler survival - Empirics: -- Monarchies and militaries coexisted with legislatures for about 60 percent of the time, while civilians ruled with legislatures 92.5 percent of the years - DV: # of parties/legislature presence - Autocrats when institutionalize survive for longer than those who underinstitutionalize - Autocrats construct several institutional trenches. When they face threats from civil society, they need a legislature that contains opposition.

Albertus 2015

Autocracy and Redistribution - Most redistributive land reform happens under autocratic rule, not democratic rule. Democracy has not been threatening to landowners. - Empirical focus: LA - For land redistribution to occur, it requires a coalitional split between landed elites and ruling political elites that gives political elites an incentive to cut landed elites out of power. Once there is a split, there must also be low instituitonal constraints (few veto points) to carry out redistribution - Democracies have too many veto points to effectively carry out redistribution - Also looks empirically at places outside of LA and finds that 80 percent of land reforms beyond LA occurred when there was a coalitional split and low institutional constraints

Centeno 1997/2002

Blood and Debt - Previous theories of war state that war supports the institutional development of the state, but author says that there are three requisites in order for this to occur: (1.) relevant states must be forced to turn inward to meet the financial challenges of war - Adequate administrative mechanisms must already be in place - Central state must have sovereignty over territory and supported by enough local actors to make extraction possible - Wars are a *potential* stimulus, they do not determine state growth - A significant amount of LA states had war in the 19th century, but did not develop a central state - How do we explain the variation? (1.) States had other sources of finance that allowed them to evade extraction, like printing money and international sources (2.) Wars came at the wrong time and states were too weak to take advantage (3.) LA states didn't have any internal elite unity for the extraction process, elites weren't bound to the state

Levitsky and Murillo 2013

Building Institutions on Weak Foundations - Many LA countries have weak institutions - Cannot be explained by punctuated equilibrium or models of gradual change- it is both frequent and radical - Causes of replacement lie in institutional origins- they are born weak, uncertain, and incongruent w/ actors of veto players - Why is uncertainty worse in LA? -> regime instability (uncertainty of power distribution), electoral volatility, social inequality, instituitonal borrowing, rapid design - Discretion over rule enforcement is much wider than in other democracies. Altering the level of enforcement is a model of institutional change. - Easier to create durable/effective institutions when informal power holders are included in rule-writing

Laver and Shepsle 1990

Coalitions and Cabinet Government - Under coalition governments, cabinets are divided into "portfolios" of policy jurisdictions - Formal model - Coalition makes proposed allocation of cabinet portfolios

Tilly 1992

Coercion, Capital, and European States - States make war and war makes states - States are the product of capital accumulation plus the concentration of coercive means - When states mobilize to create wars, they often create an extraction apparatus that becomes an administrative structure

Dube and Vargas 2013

Commodity Price Shocks and Civil Conflict - Opportunity cost theory: With a price shock to labor intensive goods, worker wages rise and participation in the criminal sector becomes less appealing - Rapacity theory: With a price shock to non-labor intensive goods, municipal revenues will rise and participation in the criminal sector will be more appealing - Data on the Colombian civil war, price history for oil and coffee - When the price of coffee decreases, all types of violence increase in areas with coffee growing - When the price of oil increase, paramilitary violence increases in areas with oil revenues

Meguid 2005

Competition Between Unequals - Niche parties are different in three ways: (1.) they reject traditional class-based divisions of politics, (2.) issues they raise to do not coincide with existing lines, and (3.) limit issue appeals to one or two issues - Traditional models don't take into account the role of salience - Actions of mainstream parties determine the viability of niche parties - Mainstream parties can adopt several strategies: -- Dismissive strategy: Signal that issue is not important, don't take a position -- Accomodative: Recognize issue importance by taking a position -- Convergence: Parties adopt the same position, funnel votes away -- Adversarial: Parties adopt opposite positions, can give votes to niche party - Empirics: Study legislative elections in Europe 1970-2000 - DV: % of votes received by a given niche party in a national election - Mainstream party tactics effect niche party votes

Levitsky and Way 2010

Competitive Authoritarianism - CA is a post-Cold War phenomenon - Three factors determine whether incumbents hold on to power: Western leverage, linkage, and organizational power - Western leverage: A state's vulnerability to democratizing pressure, rooted in the size of state economies and presence of "black knights" such as Russia. Higher leverage increases the costs of authoritarianism. High leverage alone usually leads to electoralization, not democratization - Linkage: The transmission of international influence, triggers the probability that govt abuse will lead to intl responses - Organizational power: State strength (high intensity v low intensity coercion and party strength

Shleifer and Vishny 1993

Corruption - Basic model: One government produced good is sold by a government official with a monopsony. Without theft, the official turns over the price of the good and adds a bribe, with theft the official only gets the bribe. The cost of the good can be lower in the case with theft - To complicate the model, agents collect bribes on complementary goods. In one case, different agencies all set their bribes independently rather than maximize joint revenue, but when there are low barriers to entry the bribe will go to infinity and sales go to zero, leaving everyone worse off. In a different case, sellers collude to get the best joint payoff. - When will each organization occur? - Collusion is most likely to be enforced when price cutting can easily be detected - Authoritarian governments with little responsiveness have high levels of corruption - The best arrangement to reduce corruption is to produce competition among bureaucrats - Bribes are worse than taxes because secrecy distorts markets. Officials will import only corruptible goods, distorting the market from what consumers want.

Magaloni 2008

Credible Power-Sharing and the Longevity of Authoritarian Rule -Although many authors say that the use of side payments is enough to generate elite support (Bueno de Mesquita et. al. 2003), the author says that this is problematic because there is a credible commitment problem. Other elites could still overthrow the regime or the dictator could repress them. Simple cooptation suffers from a credible commitment problem. - To survive in office, dictators need to establish power-sharing agreements with ruling coalitions, however, these arrangements are not credible - A solution to the commitment problem is to delegate control to a political party, which guarantees a share of power to the ruling elite over the long run. - Dictators are able to solve the commitment problem when their political parties (a.) control access to power positions and spoils, and (b.) promote those in organizations - Economic growth boosts electoral dictatorships because as citizens rally in support of the dictator, there are fewer incentives for elites to defect - Descriptive data on the prevalence of different types of autocracies- after the cold war, the overwhelming majority are hegemonic party rule - Performs survival analysis on regimes- party dictatorships are more enduring than military regimes, most of all single-party dictatorships

Huber and Shipan 2002

Deliberate Discretion - Two ways politicians design statutes: (1.) long statutes with detailed language to manage the policy process or (2.) vague statues that leave details unspecified - Legislators delegate in order to make more informed policies: bureaucrats have comparative advantage in knowledge. However, they might have different preferences. Therefore, more delegation means more informed policy, but less delegation means more discretion - In the model, an increased cost to specify statutes means more bureaucratic discretion, an increase in the probability of detecting bureau noncompliance increases discretion, and higher conflict reduces discretion - Two case studies: Medicaid in US and labor-related legislation internationally. Finds that policy conflict (divided government) leads to lower discretion, greater legislative capacity leads to lower discretion, legislative vetoes increase discretion

Scheve and Stasavage 2012

Democracy, War, and Wealth: Lessons from Two Centuries of Inheritance Taxation - Traditional account of inheritance tax is that it rose with universal male suffrage - Instead, they argue that war mobilization drives taxation for two reasons: (1.) if wealthier Individuals are less likely to fight, those who fight might demand they bear the cost, and (2.) if wealth holders benefit more from war, they should pay more - Empirics: Dataset of top income tax rate in 19 countries. No evidence that the expansion of male suffrage increased taxation, but there is a significant and positive effect of war mobilization - State capacity and extraction are determined by expectations about who should pay what - State capacity expands during periods of war in line with Tilly (1992)

De la O and Rodden 2008

Does Religion Distract the Poor? - Traditional models of income assume that low-income voters want more redistribution (Meltzer and Richards 1981). However, this is often not true. - Rleationship between income and vote choice has been weak since 1960, but the relationship between church attendance and voting is strong - Moral concerns cause religious voters to vote more conservatively - DV: Whether respondent votes for right or left part - IV: Issue scales and churchgoing - Data from developed European countries - The difference between low and high income groups is small among the secular, but much more pronounced among the religious - The relative moral conservativeness of the religious poor creates an offsetting push away from the left

Berkowitz et. al. (2003)

Economic Development, Legality, and the Transplant Effect - The way that a country receives its formal law is more much meaningful determinant of current effectiveness than of type of legal family - Based on two assumptions: (1.) for the law to be effective, it must be meaningful in the context which it is applied so citizens/elites have a desire to enforce and develop the law, and (2.) the judges, etc that are responsible for developing the law must increase the quality of law in a way that is responsive to demand - Empirics: Divide countries into origins and receptive/unreceptive transplants and classify legal system - To measure outcomes, they use survey data on effectiveness of judiciary, corruption, risk of government appropriation, economic development - It is worst to be an unreceptive transplant who has received French law - There is no direct Impact of transplantation, but there is an indirect effect through legality

Cusack et. al. 2007

Economic Interests and the Origins of Electoral Systems - Suggest that in contrast with Rokkan (1970) and Boix (1999), PR is an economic choice - Organized groups in the economy wanted PR because of the committee and consensus based regulatory policymaking - DV: Type of electoral system - IV: Index of (1.) strong guild tradition and local economics, (2.) widespread rural cooperatives, (3.) high employer coordination, (4.) industry based or nationally centralized unions, and (5.) a large-skill based export sector - The coordination index is the only variable that helps explain electoral systems

Levitsky and Way 2002

Elections Without Democracy - Post-Cold War regimes are not simply struggling democratic regimes- reflects a pro-democracy bias - Competitive authoritarianism: Formal democratic institutions are the means of exercising authority, but incumbents violate rules to such an extent that it is not democratic -Elections are bitterly fought but often feature abuses of state power, biased media, harassment, but generally free of massive fraud. Major election manipulation is a risk for the incumbent - Legislatures are relatively weak but can be focal points of opposition activity - Media are legal and often influential - Regions closer to the west have been more likely to transition to democracy - Since the 1990s, competitive authoritarian regimes were most likely to emerge where conditions were unfavorable to the consolidation of either democratic or authoritarian regimes -- Liberal hegemony, global economic change, developments in media, growth of international networks make it more difficult for autocracies to sustain themselves

Blaydes 2011

Elections and Redistributive Politics in Mubaraks Egypt - Elections in Egypt serve two goals: (1.) a device for distribution of rents and promotions, and (2.) to provide information on the competence of the competing candidates & provide information about regional support - Regime depended on opportunistic budget cycles to spead spoils before the election (examined empirically through election year inflation, higher per capita calorie consumption) - Why does MB participate? Becuase they can establish themselves as the main opposition in the long run, capitalize on growing support for Islamism, and attract like-minded individuals - Foreign actors promote electoralization rather than democratization - States use elections when they do not have large amounts of natural resources/need a mechanism for distributing rents

Gelbach and Svolik 2016

Formal Models of Nondemocratic Politics - Two approaches to institutions: (1.) Take a particular institution as given and examine incentives created by it, or (2.) treat institutions as equilibria outcomes to strategic outcomes - Example of first approach: Bueno de Mesquita (2003) - Example of second: Theories of legislatures in autocracies

Bates 1974

Goods of modernity are desired but scarce and modernization creates new patterns of stratification. In turn, this creates ethnic competition in Africa.

Kreuzer 2010

Historical Knowledge and Quantitative Analysis - Rails against quant methods for not paying attention to history - Attempts to replicate both Cusack 2007 and Boix 1999 - Little luck with data gathering and replication for Cusack. Recodes using historical evidence, and when he reruns analysis, finds that coordination no longer possesses a significant coefficient. Picks out three implied causal mechanisms: (1.) labor markets shape institutional preferences of business and labor, (2.) in PR-adopting countries electoral preferences are consistent across all major parties, and (3.) institutional preferences should be stable in the lead up to adoption. He doesn't find evidence for these- no history of business interests expressed - Boix is able to be replicated - History is important for evaluating the plausibility of causal mechanisms, ensuring the quality of data, untangling multiple causality

Garcia-Ponce and Pasquale 2015

How Political Repression Shapes Attitudes Toward the State - Following the war of independence in Zimbabwe, Mugabe created the ZANU-PF, which has consistently repressed citizens - Violence is used instrumentally and consistently follows elections as a form of punishment for the opposition - Empirics: Looks at respondents of Afrobarometer before and after the area was exposed to conflict - DV: Trust in president. Recent indirect exposure to violent political repression leads to an increase in reported trust for the president, ruling party, and police - Recent exposure to violent repression more strongly predicts presidential support closer to an election, but this change does not happen among supporters of the president - DV: Fear of being a victim of violence. Treated respondents are much more likely to fear that they will be a victim of violence, consistent with preference falsification - DV: Question about who sent researchers to do the survey. Treated respondents are more likely to say that the survey was sponsored by the state, consistent with preference falsification.

Ross 2004

How do Natural Resources Influence Civil War? - Previous theories about resources have weak theories about mechanisms. - Hypotheses about start: (1.) Primary commodities allow groups to loot to fund wars, (2.) natural resources lead to grievances, (3.) resources incentivize separation, (4.) oil wealth weakens state capacity - Hypotheses about duration: (5.) Resources fund weaker side, enabling longer wars, (6.) Financial incentives to oppose peace settlement, (7.) Governments less likely to settle for autonomy - 13 states chosen on a "most likely" basis: If we should find them anywhere, we should find them here - For start: No evidence of looting or grievance, some evidence of separatism hypothesis - Resource wealth lengthens conflict, looting mechanism is true here - Resource battles occurred in majority of cases, but there was also mutual exploitation of resources - Four unexpected hypotheses: (1.) resource wealth encouraged international intervention, (2.) wealth enabled groups to sell future rights ("booty rights"), (3.) increased duration by selling future exploitation rights, (4.) increased casualty rate through state overreaction

Huber et. al. 2005

Institutional Context, Cognitive Resources, and Party Attachments Across Democracies - A country's social and institutional context should have a systematic impact on party attachments - Institutions that encourage retrospective clarity of responsibility also encourage individuals to form party attachments Empirics: - DV: Whether or not an individual has an attachment to a party - IV: Ethno-religious fractionalization, number of parties, legislative fractionalization, candidate control of ballot, party age - Party attachments increase with the salience of group identities and with the quality of electoral choice. They increase with greater clarity of party responsibility and with institutions that encourage party discipline - System-level factors that discourage party attachments have their strongest effects on low-education individuals

Kalyvas and Balcells 2010

International System and Technologies of Rebellion - Military and geographic shift after the Cold War caused major shift in international system that changed the dynamics of civil war - Studies of civil war: (1.) focus on onset but ignore the way wars are fought, (2.) assume wars are homogenous over time and space, (3.) emphasizes domestic factors - Robust insurgency is linked to the Cold War through three channels: (1.) material support, (2.) revolutionary beliefs, and (3.) military doctrine - End of Cold War associated with (1.) decline of rebel capacity, (2.) decline of state capacity, (3.) emergence of postcommunist states - Empirics: -- Descriptive: While irregular war dominated the Cold War period, it dramatically declined after the Cold War -- Regression: DV: Type of war. IV: Dummy variable for post-1990. Find an impact of the Cold War on the type of war fought.

Glaeser and Shleifer 2002

Legal Origins - Common law tradition originates in England and relies on lay judges, broader legal principles, and oral arguments. - Civil law was adopted by France and relies on professional judges, legal codes, and written records - A central goal of building a national legal system is how to protect law enforcers from being bullied by local interests - They argue that France chose to rely on state-employed judges because local feudal lords were too powerful and there was no possibility of effective justice - England had weaker local lords and so juries were less vulnerable - Empirics: Qualitative accounts of both countries - Consequences of systems: Written evidence In civil law facilitates oversight of courts by higher level officials, reliance on trials in common law makes it more difficult to review decisions. Basically, the laws developed with different monitoring needs. - Social outcomes: French civil law countries have less secure property rights, greater government regulation, greater government ownership of banks, higher levels of corruption - Transplant of civil law into new environments might raise problems for property rights because civil law systems work best when the preferences of the sovereign are close to the community. If civil law is introduced where preferences diverge, the sovereign will use control over judges to politicize disputes - In sum, state outcomes are influenced by the kinds of institutions that develop, and institutions develop in reaction to present balances of power

Truex 2016

Making Autocracy Work - Despite the presence of uncompetitive elections, delegates in authoritarian parliaments appear to feel accountable to their constituents - To stay in power, the autocrat must placate citizens and provide a minimum standard of welfare, however, autocrats have incomplete information about citizen preferences. Delegates reduce uncertainty by passing along grievances while staying silent on sensitive issues - Regimes devise incentives to foster selective empathy and reward deputies with rents to install loyalty - Tests these in China using citizen surveys, interviews with NPC, interviews with financial experts

Humphreys 2005

Natural Resources, Conflict, and Conflict Resolution - Six potential mechanisms behind civil war and resources: (1.) greedy rebels mechanism, (2.) greedy outsiders mechanism, (3.) grievance mechanism, (4.) feasibility mechanism: resources finance rebellions started for other reasons, (5.) weak states mechanism, (6.) sparse networks mechanism: resources impact the relationship between citizens - Mechanisms behind duration: (1.) feasibility mechanism, (2.) booty futures, (3.) fragmented organizational structure, (4.) pork mechanism: divisible rents make settlement easier, (5.) domestic conflict premium: groups benefit from fighting, (6.) international conflict premium: third parties can facilitate fighting - Says that issues w/ econometric work around CI have to do with not using not fine enough data - Collier and Hoeffler use share of exports percent of GDP, but this is not theoreticall valid - Collects new data on diamonds and oil - Finds evidence for weak states hypothesis, and that past oil production relates to conflict onset but not future oil production (negates greedy rebels, greedy outsiders, booty futures mechanisms) - Interaction between political instability and natural resources and finds that in weak states production may be more risky for conflict - Diamonds and oil are associated with shorter conflicts, easier military victory, but no association with easier negotiated settlements - Countries dependent on agricultural commodities are at risk independent of oil and diamonds - Stronger support for weak states structure and grievance hypotheses than for booty futures or state capture

Levi 1988

Of Rule and Revenue - Rulers maximize revenue to the state, but not absolutely. They maximize subject to constraint on relative bargaining power, transaction costs, and discount rates, which determine the choice of revenue system

Collier 1999

On the Economic Consequences of Civil War - Civil wars are more damaging because they are fought on own territory, undermine state institutions, and deteriorate economic environment by encouraging portfolio diversion into foreign assets - Peace reduces the costs of economic activity, but not to prewar levels - Because peace might reverse the exodus of wealth there is the potential for accelerated growth (a "peace dividend") - Empirics: Looks at per capita GDP growth rate of each country 1960-1989 - Short wars cause continued post-war decline, while long wars give rise to rapid growth - Looks at impact of civil war on different sectors of the economy, looking at Uganda as an example - During war, war-invulnerable activity doubled as a share of GDP while the share of war-vulnerable activity halved

Sanchez de la Sierra 2019

On the Origins of the State - Armed actors create "essential functions of a state" to better expropriate - When easily extractable/taxable resources are in demand, armed actors emerge as stationary bandits - When more hidden resources are in demand, armed actors introduce permits and more sophisticated tax systems - Empirics: Variation in actions of armed actors in the Democratic Republic of the Congo. Looks at impact of exogenous shocks (Playstation announcement, recession) in price of gold/coltran - Type of extractive apparatus depends on the nature of the resource

Bueno de Mesquita et. al. 2003

Political Institutions, Policy Choice, and the Survival of Leaders - What circumstances lead to the provision of private v public goods? - Selectorate: Those in society who hold the power to remove the incumbent and select their replacement - Winning coalition: Individuals within the selectorate that allow the individual to win - Selectorate and winning coalition influence the mix of private and public goods -- Public goods become more attractive to provide relative to private goods as the size of the winning coalition increases -- Loyalty of the members of the leaders's coalition increases as the size of the selectorate increases, holding the size of the winning coalition fixed - Game theory - Challenger cannot credibly commit to keeping individuals in long-term coalition, so members of current coalition are reluctatnt to support challenger. The incumbent, however, can guarantee access to future private goods credibly. - When citizens are patient, the incumbent can offer fewer goods and look as good as the challenger - In contrast, when they are impatient the incumbent must offer more in the immediate - Autocrats prefer small winning coalitions and large selectorates

Dahl 1973

Polyarchy - Two dimensions for democracy: Public contestation and right to participate in elections - The greater the conflict between government and opposition, the more likely each will seek to deny the other - The lower the barriers to public contestation and the greater the incorporation of the populace, the more difficult it is for government to repress citizens - Three potential paths to polyarchy: (1.) liberalization precedes inclusiveness: hegemony > oligarchy > polyarchy. (2.) inclusiveness precedes liberalization: closed hegemony becomes inclusive, then increases opportunities for contestation. (3.) shortcut: hegemony suddenly transforms due to universal grant of suffrage and rights - First sequence is most common among old and stable polyarchies - Most polyarchies come from already independent states where incumbent leaders yield peacefully - circumstances most favorable to polyarchy are when access to violence and sanctions is dispersed

Fearon 2005

Primary Commodity Exports and War - First reproduces Collier and Hoeffler (2004) and finds that their parabolic relationship is sensitive to minor model changes - They also delete data listwise. When he retains some data or performs imputation, the results do not hold - C &H measure of primary commodity exports is based mainly on cash crops and exports, which does not match with their story that exports measure rebel financing (they are too difficult to exploit) - Alternative hypothesis: Higher dependence on commodities marks weaker state institutions and those states are seen as less reliable by investors. Oil exporters have much weaker states. - Empirics: (1.) empirical association between primary commodity exports and civil war is not robust, (2.) association due to inclusion of fuel exports in commodity measure is robust, (3.) seems unlikely that oil exports produce more civil war risk because it provides rebel financing, and (4.) there is direct evidence that oil exporters have less reliable states

Fearon 1995

Rationalist Explanations for War - There are two main reasons that states go to war: (1.) private information about relative capabilities and the incentive to misrepresent that information, or (2.) commitment problems in which a solution is unreasonable because one or more states would have an incentive to renege

Bellin 2012

Reconsidering the Robustness of Authoritarianism in the Middle East - Fundamental question in the Arab Spring was whether to shoot or not - The military sees itself as having three primary missions: defend country, maintain security, look out for military institution. - Using lethal force against civilians threatens military legitimacy, especially when protests are peaceful - When a military is patrimonial, it is more invested in regime survival - In Tunisia, the military was small and institutionalized and crowds were large so they did not shoot. Egypt was institutionalized but military had economic links with regime, but they still did not shoot. Bahrain, Libya, and Syria were patrimonial - Other factor: Social mobilization. Grievances, emotional triggers, a sense of impunity, and access to social media caused mobilization. - The more demonstrators mirror the demographic profile of the military and the larger their number, the less likely the military will shoot.

Arjona and Kalyvas 2011

Recruitment into Armed Groups in Colombia - Survey of fighters from FARC and paramilitary groups Micro factors: - Grievances: Rebels no more likely to be ideologically motivated than defense members - Opportunity cost: Individuals were just as employed when they joined. Individuals with jobs often joined non-paying groups anyway. - Material costs: High proportion of individuals mentioned being motivated by goods for joining - In sum, rebels are no worse off economically or politically Macro factors: - Places with weaker state capacity had more rebel participation - Social networks are important: Individuals are more likely to join if their friends did - In sum, there is no clear answer when it comes to individualistic motivations for joining, but macro-level impacts such as social networks and weak states matter

Boix 1999

Setting the Rules of the Game - After suffrage was introduced, parties changed electoral systems based on perceived position among new voters. Majoritarian systems survived when the new socialist party was weak or if one of established parties retained a dominant position - Empirics: Quantitative validation of Rokkan (1970). DV: Effective electoral threshold (how many votes needed for party to be in the political system). IV: Strength of socialism, effective # of old parties, threat - The higher the threat, the lower the new electoral threshold - In small and medium countries, fragmentation increases the chance of adopting a PR system

Soifer 2009

State Power and the Redistributive Threat - We should not assume that redistribution should happen even if the median voter wants it, a certain level of state capacity is necessary for the redistribution of wealth. Taxes on income require a strong state - IV: Ability to collect census (as proxy for state power to extract) -DV: Regime outcomes - Higher levels of inequality make coups more likely - When the state is strong, there is an Impact of inequality on democratic outcomes, but when the state is weak, there is no relationship

Robinson 2002

States and Power Review Essay - Herbst misses the whole picture by saying that only in Africa did lack of population lead to lack of state building. More important is the nature of colonization - Herbst focuses on the colonial period in defining African institutions, but Robinson says that pre-colonial interactions, such as slavery, were also important - Herbst takes for granted that it would be better if African states had more capacity, but Robinson says it matters whether states use the capacity for bad outcomes (like violence)

Herbst 2000

States and Power in Africa - Tilly cannot be applied to all states - Population distribution and space drastically changes the calculation in African states. European states competed over territory, but it doesn't make sense for African states to do the same - Ongoing problem: How to project power over sparsely settled lands - State consolidation in Africa can be understood through three dynamics: (1.) assessment of costs of expansion by individual leaders (2.) nature of buffer mechanisms established by state (3.) Nature of the regional state system (cooperation not conflict on the continent more generally)

Leemann and Mares 2014

The Adoption of Proportional Representation - In Rokkan's original arguments, there are 2 political reasons for systemic changes: (1.) social democratic threat, and (2.) disproportionality in allocation of votes to seats. The authors argue that these theories are complements - Testable implications from Cusack: (1.) where investments in specific skills are higher, economic actors should seek to establish PR, (2.) reigons with high levels of investment in skills experience more consensual relations in the industrial sphere - Rokkan testable hypotheses: (1.) contestation of race affects politician support of PR, (2.) PR support influenced by status quo seat allocation - First, provide a qualitative account of German parliament. Social Democrats had the worst vote to seat ratio and supported PR. Zentrum and Conservatives benefitted from PR and wanted to keep. - Quant analysis: - DV: Support for PR - The more disadvantageous the seat-vote ratio, the more likely party members support PR. Increase in left party strength increases support for PR - No effect for the average skill level of a district on PR support - Members of Zentrum who voted against their party faced stronger left contenders

Blattman and Annan 2010

The Consequences of Child Soldiering - Southern rebels overthrew government and army in 1980s, and some fighters started Lord's Resistance Army headed by Kony. Organization abducted children on a large scale - Survey of War Affected Youth conducted in Uganda in 2005-2006 - Abducted youth obtain .75 years less education on average and are 15 percent less likely to be literate. No differences in % working, but those abducted were working lower skill jobs. - Losses in education and literacy are strongly correlated with long abduction.

Walter 1997

The Critical Barrier to Civil War Settlement - Civil war negotiations rarely end in successful peace agreements because credible guarantees are difficult - Third party guarantors can change the level of fear and insecurity that accompanies treaty implementation by changing the payoff - To be credible, guarantors must: (1.) have self interest in upholding the promise, (2.) be willing to use force and have sufficient military, and (3.) must be able to signal resolve - Empirics: Tests every civil war 1940-1990 - Single most important explanation is the credible-commitment argument- in every case where a third party stepped in there was a successful settlement - Costs did influence civil war outcomes- longer wars were more likely to end in settlements, negotiations succeeded if a military stalemate existed, wars with greater deaths reached settlements more often. However, credible commitment was more often

Duch and Stevenson 2010

The Global Economy, Competency, and the Economic Vote - Unexpected shocks to the economy inform the economic vote, but voters face a problem whether they do not know whether to credit shocks to incumbent versus global economy - Authors hypothesize that voters know whether to attribute shocks to incumbent by observing how fluctuations in the domestic economy deviate from global economy - First, in the survey, the authors prove that voters pay attention to economic outcomes by asking whether the economy in the respondent's country has experienced stable growth. Those in unstable countries successfully identify this. - Next, they want to know if voters use economic information to inform their vote - DV: Incumbent vote - Respondents who perceive differences in the fluctuations of their national economy are more likely to exercise an economic vote

Chen and Yang 2019

The Impact of Media Censorship - Experiment where some Chinese students are given VPN access, and some subset is encouraged to visit Western news sites - Only 55 percent of those who accessed and a large % did not actively use the tool- there was little demand for information - Among those who received access and encouragement, there was more interest in Western media and they were more likely to discuss anti-government topics

Kalyvas 2006

The Logic of Violence in Civil War - Political actors use violence to achieve multiple goals, but expressive goals are overstated. Can also be used to control a population (becomes resource rather than final product). Violence performs a communicative function with a clear deterrent dimension. - Four types of violence (2 x2): State terror, genocide, civil war, and reciprocal extermination - Biases in the study of civil war: (1.) partisan bias: scholars take sides, potentially contaminating data (2.) political bias: failure to distinguish between political competition and armed combat (3.) urban bias: civil war treated as an urban affair when in reality it is mostly rural (often due to data constraints (4.) selection bias: Instances of violence cannot be considered independently of instances where violence does not occur (5.) data problems and overaggregation: indicators of violence are unreliable and inconsistent across nations

Scott 1977

The Moral Economy of the Peasant - Subsistence-oriented peasants prefer to avoid economic disaster rather than take risks to maximize their average income - The fear of food shortages causes a "subsistence ethic" where institutions and social arrangements are crafted to minimize risk, which has also brought a set of moral rights and expectations - Conducts two case studies of peasant rebellions in Burma and Vietnam in the 1930s. Both appeared to be spontaneous and were caused by group indignation and rage over what claims on their product by the state were tolerable - Violation of moral rights provokes resentment and forms the standard against which state extraction is tolerated - Two major changes caused the stability and greater extraction that led to grievances: "North Atlantic capitalism" and the development of the modern state w/o a safety net

Hudson and Matfess 2017

The Neglected Linkage Between Brideprice and Violent Conflict - In many societies, brideprice is an expensive undertaking but required for young men to pay in order to receive societal benefits. - Case study of Boko Haram, South Sudan, Saudi Arabia - In Boko Haram, the organization has been preoccupied with marrying their members. They frequently kidnap women to enable marriages and reward loyalty with women. - In South Sudan, cow raiding is extremely common and is driven by the increasing cow-based price of brideprice - In Saudi Arabia, brideprice is an important policy area. The state has made effort to cap brideprice. Terrorist recruitment is high, but they say it could be higher if there was not a capped brideprice.

Svolik 2012

The Politics of Authoritarian Rule - Two problems in authoritarian rule: (1.) the problem of authoritarian control (leaders v. masses- but only leads to downfall in 1/5 of cases), and (2.) the problem of authoritarian power-sharing (2/3 lost power this way) - The second dynamic is shaped by two dynamics: (1.) credible commitment issues, (2.) violence is the only other solution - Formal institutions help the problem of power sharing by establishing formal rules that facilitate the exchange of info among elites -- Allows for greater transparency- misperceptions will not lead to destabilizing confrontations -- Provides public signal of commitment to power sharing - In China, Mao Zedong and Deng accumulated power while their predecessors were constrained by institutions - Empirics- using ruling coalition spells, dictatorships with parties and legislatures are more durable and less likely to be overthrown in a coup - The functions of institutions in autocracies is to resolve collective action problems among elites and establish formal rules for information-sharing and compliance

Young 2019

The Psychology of State Repression - Lab-in-the-field survey - IV: "Fear" psychological inducement - DV: (1.) index of 12 behavioral questions, (2.) whether or not respondent took pro-democracy wristband, (3.) survey on protest repression risk, (4.) financial risk games - Those who received the fear treatment report a lower likelihood of expressing dissent and are less likely to take the wristband. They have increased expectations that they will experience violence and are more pessimistic that others will turn out. They also exhibited less risk-taking in financial games.

Bellin 2004

The Robustness of Authoritarianism in the Middle East - Authoritarianism has been robust in MENA because the coercive apparatus in many states has been exceptionally able and willing to crush reform. Four elements are essential to this robustness: maintencance of fiscal health, maintenance of international support networks, inversely related to degree of instituitonalization, degree to which it faces a high level of political mobiilization -- Fiscal health: In MENA, few countries face economic collapse and most have sufficient revenue derived from rents -- International support: Western continued support driven by double threat of oil supply and Islamist threat -- Patromonialism: Ethnic ties used to guarantee loyalty and security establishment is highly patrimonial -- Popular mobilization: No where is there cross class mobilization -- Potential 5th variable: Credible threat from Israel

Kurtz 2009

The Social Foundations of Institutional Order - The role of wars and wealth in state building is conditional. What really matter are the distribution of prior social and political variables that determine path development - 2 most important variables: (1.) absence of labor-repressive agriculture and (2.) exclusionary but collective elite dominance - Where labor repressive systems predominate, local control is essential to elite survival- when labor is servile, individuals can't go join an army - Empirics: Comparative historical comparison of Peru and Chile - Peru was divided by race and class and peasants were tied to lands. Their loss would mean the downfall of local control. Elites were not integrated - In Chile employment was voluntary and there was an oversupply, leading to easy mobilization. There was an "oligarchic republic" with buy-in from elites - Chile engaged in wars as an aggressor to justify the expansion of the state

Darden and Grzymala-Busse 2006

The authors argue that there was communist exit in some countries but not others as a result of precommunist education. They find that in countries where pre-communist schooling was implemented, there was a quicker exist from communism. The propose that schooling brought literacy, which allows for the transmission of information across time. Mass schooling can convey nationalist ideas.

Mares and Queralt 2015

The non-democratic origins of income taxation - Adoption of income tax happened because owners of different assets attempted to minimize their own tax burden. Agricultural elites wanted to push taxes onto manufacturers. - In addition, the introduction of taxes was tied to the ability to block the masses from voting (the "vote-tax link") - Empirics: (1.) Quantitative approach- IV: power of landowning elites (measure of land concentration). Positive relationship between IV and tax. Countries where individuals had to pay tax before voting were more likely to pass additional income taxes (2.) Qualitative- collect information on parliamentary passage of British income tax in 1842. Data is on deliberations and roll call votes. Deliberation shows evidence of agri-manu divide. Roll call shows that urban districts opposed, while rural supported. The higher the share of agri workers in the district, the more likely the MP opposed - Against war-based theories of taxation

Kasara 2007

Traditional beliefs about Africa say that coethnics benefit from their representatives in power. Kasara finds that this is not necessarily true. Her DV is the tax rate on agricultural goods. Coethnics are actually more likely to tax coethnics. Leaders in competitive regimes tax less against their home regions.

Levitsky and Murillo 2009

Variation in Institutional Strength - Institutional strength consists of enforcement and stability. Enforcement= degree to which parchment rules are complied with in practice. Stability = institutions survive changes in political system - Many institutions are weak because actors who create them do not tend to enforce them- there is a divergence between political actors real and state goals. Could be adopted because of international demands, pursuit of legitimacy - Instability is often rooted in a disjuncture between rule writers and power holders. New arrangements are likely to endure when rule writers gain the acceptance of actors outside of process or defeat opponents - Time matters: Slowly created institutions are more stable - Instability is path-dependent - When institutions are stable, actors assume that others will play by the rules and there will be less uncertainty, otherwise there are credibility and commitment problems

Wilkinson 2006

Votes and Violence - Ethnic riots are not spontaneous- they are planned by politicians for electoral purposes to change the salience of identities - Town level electoral incentives account for Muslim-Hindu violence, but state level electoral incentives ultimately determine when police come in - Town leaders will stoke violence when elections are more competitive - States protect minorities when minorities are an important part of support case, or when electoral system is so competitive that the party will likely have to negotiate with minority parties int he future - Empirically: Hindu-Muslim riots in India. Probability that a town has a riot is highly related to electoral competition, but state-level patterns of law enforcement dominate local factors - Looks at a comparative perspective: Ethnic riots take place when political competition is fiercest and state's reaction to violence is determined by support base/party competition

Magaloni 2006

Voting for Autocracy - Case study of the PRI in Mexico - Elections are important because they establish power-sharing among ruling elites, disseminate public opinion about regime "invincibility", provide information about supporters and opponents, and trap the opposition so they invest in autocratic instituitons rather than challenging them by violent means - Elite unity is sustained by (1.) creating an aura of invincibility, (2.) distributing spoils, (3.) raising the cost of entry to potential challengers - PRI began in the 1920s and won elections until 2000, but committed massive fraud in the 1988 elections - Empirical evidence: Before the 1980s when the party was competitive, they flooded districts at elections w/ government spending. Found relationship between low spending and splits within ruling party. - Voters don't vote for opposition because (1.) they are uncertain because they've never ruled before, (2.) the ruling party monopolizes state resources, (3.) ruling party threatens electoral fraud - Hegemonic party support depends on (a.) long term economic growth, (b.) vote buying and the distribution of transfers through a "punishment regime" and (c.) electoral fraud and force - Empirics: Places with more opposition support will be punished in later distribution of resources

Geddes 1999

What Do We Know about Democratization After 20 Years? - One solid fact is that democracy is more likely to happen in more developed countries - Previously, one of the most widely accepted generalizations was that splits in the old regime led to transformatioin, but this wasn't true in post-Soviet/Africa - These theories fail because different types of authoritarianism differ from each other just as much as they differ from democracy - Classifies regimes as personal, military, or single party -- Military: Soldiers place highest value on survival of military, not who is in power, will join coups only if the civilian government prevents main goals. Military action is a coordination game- need to know others will join -- Single party: Stag Hunt game. Everyone must coordinate to get the desired outcome -- Personalist: in comparison to single party the payoff to a minority faction excluded from office is lower - Military regimes have the seeds of their own destruction- fall due to factional splits. They are the most fragile but can best negotiate orderly transition. - Leaders of single party regimes face competition, but the benefits of coordination win. Leadership struggle does not result in transition. Resistant to economic shocks. - Personalist are less vulnerable to internal splits, but more vulnerable than single party. Most likely to be violently overthrown. - Empirically, survival rates for regimes are shortest for military rule and longest for single party regimes

Sambanis 2004

What is Civil War? - The most famous project on war classification, the CoW project, has had little review over the year - Difficult to measure civil war because: (1.) difficult to distinguish civil war from intrastate war, (2.) unclear what degree of organization is required to distinguish from state-sponsored violence, (3.) how we should deal with unreliable data, (4.) how to determine when wars stop and start - Some empiric improvements: (1.) should count per capita deaths, (2.) should count civilian deaths - Disagreements in coding rules can change quant analyses. Income level and population are robust, while mountainous terrain, Muslim population, and economic growth are not. Significance of ethnic fractionalization depends on coding rules.