Remittances and varieties of democratization in developing countries

ABSTRACT

How do remittances affect democratization in developing countries? In this paper we reconcile divergent findings in the literature by examining the effect of remittances on procedural and liberal democracy in developing countries at various stages of their democratic development. Remittances are primarily sent to middle-class individuals and bypass government control. Yet, governments in countries receiving remittances want to tax this remittance income. Government officials therefore need to incentivize participation of the middle class in the formal economy by reducing the threat of expropriation. Improving procedural democracy, which assures citizens of improved property and rule of law protections, is one way to accomplish this. We argue that this relationship should only be present in mixed regime types, with the democratizing effect of remittances waning as the country’s level of democracy or autocracy strengthens. Further, we expect elements of liberal democracy, such as civil rights and equality under the law, to remain unchanged in all remittance-receiving countries. The middle class and governing elite are less incentivized to improve liberal democracy to limit the power and mobilization capacity of the poor and prevent increased redistribution. We test our theory on a dataset of developing countries from 1975 through 2011.

KEYWORDS:

• Remittances
• regime change
• authoritarianism
• developing countries
• taxation

Acknowledgements

The authors wish to thank Faisal Ahmed, Katrina Burgess, Jennifer Gandhi, Steve Miller, James M. Scott, and Joel Simmons for their excellent suggestions on this project throughout its development.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Daniel C. Tirone http://orcid.org/0000-0002-3961-6959

Notes

1 Bates and Lien, “Taxation, Development and Representative Government”; Timmons “Taxation and Representation”.

2 Ahmed, “Perils of Unearned Foreign Income”; Pfutze, “Clientelism Versus Social Learning”.

3 See Acemoglu and Robinson, Economic Origins of Dictatorship and Democracy. Haggard and Kaufman, “Inequality and Regime Change”, challenge the role of distributional conflict on democratization.

4 Mukand and Rodrik, “Political Economy of Liberal Democracy”; Zakaria, “Rise of Illiberal Democracy”; Merkel, “Embedded and Defective Democracies”.

5 Smith and Zeigler, “Liberal and Illiberal Democracy”; Bugaric, “Populism, Liberal Democracy, Law”; Rose and Shin, “Democratization Backwards”.

6 Escribà-Folch, Meseguer and Wright, “Remittances and Democratization”; Bearce and Park, “Why Remittances are a Political Blessing and Not a Curse.”

7 Bates and Lien, “Taxation, Development and Representative Government”; North and Weingast, “Constitutions and Credible Commitments”; Meltzer and Richard, “Rational Theory of the Size of Government”; Timmons, “Taxation and Representation”.

8 Boix, Democracy and Redistribution.

9 Ross, “Does Taxation Lead to Representation?”.

10 Cheibub, “Extractive Capacity of Governments”.

11 Levi, Of Rule and Revenue; Kenny and Winer, “Tax Systems in the World”.

12 For a discussion on the differential impact of remittances on spending see Easton and Montinola, “Remittances, Regime Type, and Government Spending Priorities.”

13 Doyle, “Remittances and Social Spending”.

14 Abdih et al., “Remittances and Institutions”.

15 Ahmed, “Perils of Unearned Foreign Income”. An alternative hypothesis is that public good spending could also be essential to political survival (see Bueno de Mesquita and Smith, “Leader Survival, Revolutions, and the Nature of Government Finance.”).

16 Tyburski, “Resource Curse Revisted”. Tyburski, “Curse or Cure?”.

17 Pfutze, “Clientelism Versus Social Learning”; Bearce and Park, “Why Remittances are a Political Blessing and Not a Curse.”

18 Escribà-Folch, Meseguer and Wright, “Remittances and Democratization”.

19 Escribà-Folch, Meseguer and Wright, “Remittances and Protest in Dictatorships.”

20 Reinke, “Seminar on Remittance Statistics”.

21 Mosley and Singer, “Migration, Labor and the International Political Economy”, 294.

22 Levi, Of Rule and Revenue.

23 Boix, Democracy and Redistribution.

24 Adams and Cuecuecha, “Impact of Remittances on Investment and Poverty”; Bui, le and Daly, “Microlevel Impacts of Remittances”.

25 Wright, “Do Authoritarian Institutions Constrain?”. Notably, this effect varies across authoritarian regimes.

26 Catrinescu et al., “Remittances, Institutions and Economic Growth”.

27 Bearce and Park, “Why Remittances are a Political Blessing and Not a Curse”.

28 Timmons, “Taxation and Representation”.

29 Freund and Spatafora, “Remittances, Transaction Costs and Informality”; Kosse and Vermeulen, “Migrants’ Choice of Remittance Channel”. See also Ratha, “Workers’ Remittances”. Aggarwal et al., “Do Remittances Promote Financial Development?”, find that remittances are associated with financial development in a global sample of developing countries. We would expect democracies to be driving this result. See Ambrosius and Cuecuecha, “Remittances and Use of Financial Services”, for an assessment of remittances and informal financial institutions.

30 Ross, “Does Taxation Lead to Representation?”.

31 Gandhi, Political Institutions under Dictatorship.

32 Cheibub, “Extractive Capacity of Governments”; Bastiaens and Rudra, Democracies in Peril.

33 Escribà-Folch et al., “Remittances and Democratization”.

34 See Leviskty and Way, “Competitive Authoritarianism” and Gandhi, “Political Institutions under Dictatorship.”

35 Mukand and Rodrik, “Political Economy of Liberal Democracy”, use the term ‘electoral’ democracy. We employ the term ‘procedural’ democracy to encompass more than elections, such as checks and balances and the rule of law.

36 Property right protections are stronger in democracies relative to nondemocracies; World Bank, World Governance Indicators.

37 Mukand and Rodrik, “Political Economy of Liberal Democracy”.

38 Freedom House, Freedom in the World Methodology.

39 Mukand and Rodrik, “Political Economy of Liberal Democracy”.

40 Stark, Taylor and Yitzhaki, “Migration, Remittances and Inequality”.

41 Acostam et al., “Remittances and Household Behavior”; Bearce and Park, “Why Remittances are a Political Blessing and Not a Curse”; Goce-Dakila and Dakila, “Impact of Overseas Filipino Workers Remittances”. See Stark et al., “Migration, Remittances and Inequality”, for barriers the poorest households face to high-paying migrant work.

42 Ansell and Samuels, Inequality and Democratization. Remittance-receiving individuals engage in more protest activity; Dionne et al., “Another Resource Curse?”. However, protests have not been associated with fostering liberal democracy; Smith and Ziegler, “Liberal and Illiberal Democracy”. Remittances do not improve voter turnout - Dionne et al., “Another Resource Curse?” - and migrant households do not have a statistically significant impact on elections results in municipalities with a history of competitive elections; Pfutze, “Does Migration Promote Democratization?”.

43 Gandhi and Lust-Okar, “Elections Under Authoritarianism”.

44 Bravo, “Credit where credit is due?”, and Ahmed, “Perils of Unearned Foreign Income.” An alternative perspective is that citizens are less dependent on the state and may vote for opposition parties (see Perez-Armendariz and Crow, “Do Migrants Remit Democracy?”). However, even if this would occur, the strengthened electoral infrastructure increases the likelihood that the previous government could regain power.

45 Mukand and Rodrik, “Political Economy of Liberal Democracy”.

46 Mukand and Rodrik, “Political Economy of Liberal Democracy”, 8.

47 Rudra, “Globalization and Decline of the Welfare State”. While our argument suggests a middle-upper class alliance, this is not a necessary condition for our argument.

48 Timmons, “Taxation and Representation”. Acemoglu and Robinson, Economic Origins of Dictatorship and Democracy, and Boix. Democracy and Redistribution, associate a richer middle class with democratization.

49 It is beyond the scope of this analysis to examine if advances in procedural but not liberal democracy are associated with regime survival. However, we do assume that political leaders are motivated by regime survival and act in a manner they perceive will increase their tenure.

50 World Bank, “World Development Indicators”.

51 See Freund and Spatafora, “Remittances, Transaction Costs, and Informality”. Although using formal remittances in our panel estimations is less than fully desirable, the operationalization of our data are consistent with other scholars such as Ratha, “Workers’ Remittances”.

52 Marshall et al., Polity IV Project.

53 Freedom House, “Freedom in the World Methodology”.

54 Tierney et al., “More Dollars than Sense”.

55 For measures with a minimum of zero, the log is taken by adding one to the value before taking the natural log.

56 Gleditsch et al., “Armed Conflict 1946-2001”; Themnér and Wallensteen, “Armed Conflict, 1946–2011”.

57 PEW, Global Attitudes Project.

58 We limit our sample here to low income countries to control for between-country income effects. We present this relationship across middle income countries, all surveyed countries, and various case examples in Appendix A.

59 Freund and Spatafora, “Remittances, Transaction Costs, and Informality”.

60 Singh, “From Shock Absorber to Shock Transmitter”; Ratha and Plaza, “Harnessing Diaspora”; Freund and Spatafora, “Remittances, Transaction Costs, and Informality”.

61 Unless noted, all models lag all regressors two years. Our primary results are robust to one- and three-year lags; see Appendix C. We limit our samples to countries for which the Polity measures are non-missing to address concerns of sample selection due to missing data.

62 A Hausman test indicates that fixed effects were preferred over random. Our 25 year time series reduces Nickell bias concerns, Nickell, “Biases in Dynamic Models”.

63 Polity entering the interaction twice allows us to plot the effect across only the range of that measure, with the influence of the Polity squared term observed in the curvilinear estimated marginal effect of remittances.

64 Results are robust to using total or per capita U.S. dollars for our financial flow measures, with some variation in the range in which the effect is statistically significant.

65 Arellano and Bond, “Tests of specification for panel data”. We use the xtabond2 Stata command from Roodman, “How to Do xtabond2”.

66 The Difference-in-Hansen test failed to reject the null assumption that the instrument is exogenous.

67 This measure addresses concerns of the imprecision involved in utilizing Polity as a measure of democracy; Trier and Jackman, “Democracy as a Latent Variable”.

68 Factor analysis results identified a common factor among the six measures, as to be expected. However, our measures also loaded along a second dimension as we would also anticipate, with the procedural democracy measures achieving the same sign which was different than those representing liberal democracy.

69 Given the anticipated linear relationship, we use a single interaction between remittances and Polity rather than the curvilinear construction used earlier.

70 Post-estimate testing confirms that the difference between the estimated marginal effects at the highest and lowest values of Polity is statistically significantly.

71 To assess the respondents’ perceptions of the current state of elections and the judiciary in their country, we use questions with a similar format as noted in the text but asks respondents if these describe current conditions “very well”, “somewhat well”, “not too well”, or “not well at all”. We code respondents who choose “not too well” as living in intermediate regimes.

Additional information

Notes on contributors

Ida Bastiaens

Ida Bastiaens is an Assistant Professor of Political Science at Fordham University. She specializes in international political economy and international development. Her research analyses questions on the political determinants of integration in the global economy, the impact of international integration on fiscal and social welfare in developing countries, and citizen preferences for global capital flows.

Daniel C. Tirone

Daniel C. Tirone is an Assistant Professor of Political Science at Louisiana State University, with a specialization in international political economy. His research interests include the impacts of foreign aid and remittances, civil conflict and terrorism, and the power associated with state financial capabilities.