Income Shocks and Ethnic Group Bias
| Název práce v češtině: | Příjmové šoky a etnická předpojatost |
|---|---|
| Název v anglickém jazyce: | Income Shocks and Ethnic Group Bias |
| Klíčová slova: | příjmové šoky, endogenita, ethnicita, instrumentální proměnné |
| Klíčová slova anglicky: | income shock, endogeneity, ethnicity, instrumental variables |
| Akademický rok vypsání: | 2017/2018 |
| Typ práce: | diplomová práce |
| Jazyk práce: | angličtina |
| Ústav: | Institut ekonomických studií (23-IES) |
| Vedoucí / školitel: | doc. PhDr. Michal Bauer, Ph.D. |
| Řešitel: | skrytý - zadáno vedoucím/školitelem |
| Datum přihlášení: | 23.05.2018 |
| Datum zadání: | 23.05.2018 |
| Datum a čas obhajoby: | 19.06.2019 09:00 |
| Místo konání obhajoby: | Opletalova - Opletalova 26, O105, Opletalova - místn. č. 105 |
| Datum odevzdání elektronické podoby: | 05.05.2019 |
| Datum proběhlé obhajoby: | 19.06.2019 |
| Oponenti: | prof. Roman Horváth, Ph.D. |
| Kontrola URKUND: | |
| Seznam odborné literatury |
| Alesina , Alberto & Stelios Michalopoulos & Elias Papaioannou, 2016. "Ethnic Inequality," Journal of Political Economy, University of Chicago Press, vol. 124(2)
Blundell, Richard & Costa Dias, Monica, 2008."Alternative Approaches to Evaluation in Empirical Microeconomics," IZA Discussion Papers 3800, Institute for the Study of Labor (IZA). Available at: https://ideas.repec.org/p/iza/izadps/dp3800.html Burke, Marshall and Hsiang, Solomon and Miguel, Edward, Climate and Conflict 2015. Annual Review of Economics, Vol. 7, pp. 577-617, 2015. Available at SSRN https://ssrn.com/abstract=2640071 Collier, Paul and Anke Hoeffler. 1998. “On Economic Causes of Civil War.” Oxford Econ. Papers 50 (October): 563-73. Esteban, Joan and Ray, Debraj, 1994, On the Measurement of Polarization, Econometrica, 62, issue 4, p. 819- 51, Available at https://EconPapers.repec.org/RePEc:ecm:emetrp:v:62:y:1994:i:4:p:819-51. Fearon, James D. and David D. Laitin. 2003. “Ethnicity, Insurgency, and Civil War.” American Political Science Review 97 (March): 75-90. Henderson, Vernon & Adam Storeygard & David N. Weil, 2012. "Measuring Economic Growth from Outer Space," American Economic Review, American Economic Association, vol. 102(2), pages 994-1028, April. Miguel, Edward, Shanker Satyanath, and Ernest Sergenti. 2004. “Economic Shocks and Civil Conflict: An Instrumental Variables Approach.” Journal of Political Economy 112 (4): 725-53 Rohner, Dominic, Thoenig, Mathias and Zilibotti, Fabrizio, (2012), Seeds of Distrust: Conflict in Uganda, No 8741, CEPR Discussion Papers, C.E.P.R. Discussion Papers, https://EconPapers.repec.org/RePEc:cpr:ceprdp:8741. Sarsons, Heather, 2015. "Rainfall and conflict: A cautionary tale," Journal of Development Economics, Elsevier, vol. 115(C), pages 62-72 Shayo, Moses, A Theory of Social Identity with an Application to Redistribution (2007). Available at SSRN: https://ssrn.com/abstract=1002186 |
| Předběžná náplň práce |
| Motivation:
Group bias might increase the likelihood of conflict and limit cooperation among members of different ethnic groups. This is especially dangerous if one group has power over the others. Conflict, as Rohner, Thoenig, and Zilibotti (2012) show, results in lower inter-ethnic trust. Such situation can then end up in a vicious circle: ethnic bias fuels ethnic conflict, which, again, strengthens ethnic bias etc. Miguel, Satyanath, and Sergenti (2004) relate economic shocks to the incidence of conflict. They conclude that negative economic shocks are strongly associated with an increase in the number of conflicts, but the impact is not significantly different in richer, more democratic, or more ethnically diverse countries, which is rather surprising. Sociological theories propose several alternative explanations for ethnic bias. A key distinction between them is whether they explain group identification as a tool to achieve certain social status, or to get an informal insurance network. The abovementioned conclusion of Miguel, Satyanath, and Sergenti (2004) suggests that status could be more important, otherwise adverse economic shocks should impact ethnically diverse countries more strongly (provided that ethnic bias increases the chances of ethnic conflict). In my thesis, I will estimate the first part of this relation, i.e. from economic shocks to ethnic bias. This could help clarify which of the sociological approaches explains ethnic group bias better. My results might also be useful for assessing potential risks arising from negative income shocks. Hypotheses: 1. Negative income shocks increase ethnic group bias. 2. The effect of income on group bias is weaker in highly ethnically polarized areas. 3. In-group bias of ethnic minority members in an otherwise ethnically homogenous areas is less sensitive to a change in income. Methodology: Estimating ethnic bias of individuals directly from income would likely be methodologically incorrect due to endogeneity of the income variables. To overcome this pitfall, I will instrument for income with exogenous variables, such as rainfall variation.data from the Climate Prediction Center database. Similar IV methods have been used in the past, although not for ethnic group bias estimation. Rainfall variation is not determined by family or economic background of individuals, but as Sarsons (2015) points out, this does not necessarily mean that the exclusion restriction is satisfied. She demonstrates that even if the effect of rainfall variation on income is attenuated by construction of dams in selected areas, rainfall variation still correlates with riot incidence in rural India, and thus, there must be other channels through which it influences violence. I will therefore test several different instruments to see if they yield similar results. Building on the empirical approach employed by Rohner, Thoenig, and Zilibotti (2012), I will work with individual level Afrobarometer data for Uganda, whose geographical and economic conditions give a reasonable guarantee that rainfall will have strong impact on incomes of people. Afrobarometer survey asks participants a series of questions related to their opinions on topics spanning living conditions, education, political preferences etc. Group bias can be measured as a difference between individuals’ attitudes towards certain group and her usual attitudes. Survey results also include geographical information on participants‘ (regional and village level) and their ethnicity. Income is proxied by questions related to ownership of assets (TV, radio, …), self-assessed economic well-being and employment status. Wide range of individual personal characteristics is covered, which makes it possible to control for other factors related to group bias. I will supplement Afrobarometer’s data with regional statistical data (similarly to the Rohner, Thoenig, and Zilibotti (2012) paper). The statistical data can be further complemented by nighttime light intensity measurements in the area (see Hendorson, Storeygard and Weil (2012)) as another proxy for economic development. I will use the Geo-Referenced Ethnic Group dataset to calculate ethnic fractionalization within individual regions. To achieve reliable results, I will employ both instrumental variable regression and difference-in-differences estimation. So far, there have been six rounds of the Afrobarometer survey (the first was held in 1999), and individual level data can, therefore be compared across time. Expected Contribution: Effects of income shocks on various aspects of human behaviour have been studied heavily in the past. Miguel, Satyanath, and Sergenti (2004) discuss the relation of income shocks and conflict incidence. According to their results, negative income shocks increase probability of conflict. Ethnic group bias might catalyse such events. It disintegrates societies and makes peaceful coexistence more difficult. I will estimate the impact of income shocks on individuals’ ethnic group bias, which, to my knowledge, has not been done before. This could be useful in predicting ethnic tensions in areas affected by adverse income shocks. It could also help to explain whether ethnic groupiness is an adaptive strategy to cope with difficult living conditions. In addition to that, I will examine exogeneity of instrumental variables for income in developing countries, which might be useful for future research. Outline: 1. Introduction 2. Literature review 3. Afrobarometer data description 4. IV methodology description and testing 5. IV estimation using rainfall data 6. Difference-in-differences estimation 7. Discussion and robustness checking 8. Conclusion |
| Předběžná náplň práce v anglickém jazyce |
| Motivation:
Group bias might increase the likelihood of conflict and limit cooperation among members of different ethnic groups. This is especially dangerous if one group has power over the others. Conflict, as Rohner, Thoenig, and Zilibotti (2012) show, results in lower inter-ethnic trust. Such situation can then end up in a vicious circle: ethnic bias fuels ethnic conflict, which, again, strengthens ethnic bias etc. Miguel, Satyanath, and Sergenti (2004) relate economic shocks to the incidence of conflict. They conclude that negative economic shocks are strongly associated with an increase in the number of conflicts, but the impact is not significantly different in richer, more democratic, or more ethnically diverse countries, which is rather surprising. Sociological theories propose several alternative explanations for ethnic bias. A key distinction between them is whether they explain group identification as a tool to achieve certain social status, or to get an informal insurance network. The abovementioned conclusion of Miguel, Satyanath, and Sergenti (2004) suggests that status could be more important, otherwise adverse economic shocks should impact ethnically diverse countries more strongly (provided that ethnic bias increases the chances of ethnic conflict). In my thesis, I will estimate the first part of this relation, i.e. from economic shocks to ethnic bias. This could help clarify which of the sociological approaches explains ethnic group bias better. My results might also be useful for assessing potential risks arising from negative income shocks. Hypotheses: 1. Negative income shocks increase ethnic group bias. 2. The effect of income on group bias is weaker in highly ethnically polarized areas. 3. In-group bias of ethnic minority members in an otherwise ethnically homogenous areas is less sensitive to a change in income. Methodology: Estimating ethnic bias of individuals directly from income would likely be methodologically incorrect due to endogeneity of the income variables. To overcome this pitfall, I will instrument for income with exogenous variables, such as rainfall variation.data from the Climate Prediction Center database. Similar IV methods have been used in the past, although not for ethnic group bias estimation. Rainfall variation is not determined by family or economic background of individuals, but as Sarsons (2015) points out, this does not necessarily mean that the exclusion restriction is satisfied. She demonstrates that even if the effect of rainfall variation on income is attenuated by construction of dams in selected areas, rainfall variation still correlates with riot incidence in rural India, and thus, there must be other channels through which it influences violence. I will therefore test several different instruments to see if they yield similar results. Building on the empirical approach employed by Rohner, Thoenig, and Zilibotti (2012), I will work with individual level Afrobarometer data for Uganda, whose geographical and economic conditions give a reasonable guarantee that rainfall will have strong impact on incomes of people. Afrobarometer survey asks participants a series of questions related to their opinions on topics spanning living conditions, education, political preferences etc. Group bias can be measured as a difference between individuals’ attitudes towards certain group and her usual attitudes. Survey results also include geographical information on participants‘ (regional and village level) and their ethnicity. Income is proxied by questions related to ownership of assets (TV, radio, …), self-assessed economic well-being and employment status. Wide range of individual personal characteristics is covered, which makes it possible to control for other factors related to group bias. I will supplement Afrobarometer’s data with regional statistical data (similarly to the Rohner, Thoenig, and Zilibotti (2012) paper). The statistical data can be further complemented by nighttime light intensity measurements in the area (see Hendorson, Storeygard and Weil (2012)) as another proxy for economic development. I will use the Geo-Referenced Ethnic Group dataset to calculate ethnic fractionalization within individual regions. To achieve reliable results, I will employ both instrumental variable regression and difference-in-differences estimation. So far, there have been six rounds of the Afrobarometer survey (the first was held in 1999), and individual level data can, therefore be compared across time. Expected Contribution: Effects of income shocks on various aspects of human behaviour have been studied heavily in the past. Miguel, Satyanath, and Sergenti (2004) discuss the relation of income shocks and conflict incidence. According to their results, negative income shocks increase probability of conflict. Ethnic group bias might catalyse such events. It disintegrates societies and makes peaceful coexistence more difficult. I will estimate the impact of income shocks on individuals’ ethnic group bias, which, to my knowledge, has not been done before. This could be useful in predicting ethnic tensions in areas affected by adverse income shocks. It could also help to explain whether ethnic groupiness is an adaptive strategy to cope with difficult living conditions. In addition to that, I will examine exogeneity of instrumental variables for income in developing countries, which might be useful for future research. Outline: 1. Introduction 2. Literature review 3. Afrobarometer data description 4. IV methodology description and testing 5. IV estimation using rainfall data 6. Difference-in-differences estimation 7. Discussion and robustness checking 8. Conclusion |