Skip to main content

State Power and the Economic Origins of Democracy

An Erratum to this article was published on 06 September 2016


Recent comparative politics scholarship on regime change has not taken state capacity seriously. Prominent works on the relationship between democracy and economic inequality center on the expectation by economic elites that democratization will lead to economic redistribution. But state capacity is necessary for redistribution, and where extractive capacity is lacking, rational economic elites should not fear that suffrage expansion would lead to effective redistribution, nor should the masses expect to gain economically from democratization. State capacity thus acts as a scope condition for the effect of inequality on regime outcomes. This prediction is confirmed through replication and extension of the analysis in Boix (2003), with the addition of the presence of a regularly implemented national census as a proxy for state capacity. In strong states, the effect of inequality on regime change is confirmed. But where the state is weak, inequality is shown to have no effect on regime outcomes. Thus, including state capacity in theories of regime change calls into question general claims about the “economic origins” of dictatorship and democracy.

This is a preview of subscription content, access via your institution.


  1. On the failure of these theories to distinguish between the divergent effects of income and land inequality, see Ansell and Samuels (2010).

  2. This paper uses the terms state “strength” and state “capacity” interchangeably to capture the ability of the state to effectively implement its chosen policies. Given the focus on redistribution, the emphasis below is on the extractive facet of state capacity.

  3. Lipset (1959, 51) also claimed that increased inequality reduced the prospects for democracy, but he linked the two via an attitudinal argument rather than one based on the relative costs of tolerating and suppressing democracy. As inequality increases, Lipset argues that elites see political rights for the lower classes as “absurd and immoral.” Thus Dahl, rather than Lipset, is an appropriate point of departure for recent scholarship on the economic origins of democracy.

  4. As will be discussed in more detail below, redistribution is used broadly to refer to the wide range of policies that affect the distribution of assets in society, ranging from confiscation of wealth to social welfare policies, and including primary education and—most prominently in terms of its salience and its effects—progressive taxation. This broad conception of redistribution is based on Meltzer and Richard (1981).

  5. The language of the will and capacity of governments to redistribute is drawn from Bellin (2004), who argues that authoritarianism is stable where governments are willing and able to repress opponents.

  6. Because he provides a statistical analysis of these predictions, I center my discussion on his account rather than on the more complex theory proposed by Acemoglu and Robinson (2006). For a critique of the empirical evidence in that work, see Smith (2008).

  7. Redistribution entails both the extraction of resources from their holders and the allocation of those resources to other social actors—and the state’s ability to extract and to allocate may not co-vary. I thank Daniel Gingerich for clarifying this point. In this preliminary examination of the effect of state power on regime outcomes, I focus only on the power to extract (which may be more salient in the calculation of social actors since it represents a cost directly imposed by the government) and set the issue of allocation (which is moot where the state cannot extract) aside. Thus, allocation of wealth—whether through distribution of assets or through pro-poor spending—is not considered in the following discussion. O’Neil (2007) shows that allocation of wealth is a real challenge for the state in her discussion of the crisis of Chile’s privatized pension system. As a result of the privatization of pensions, the Chilean state no longer has accurate records of pension holders, and it therefore faces significant infrastructural challenges in its efforts to provide the legally mandated minimum pension to many retired workers.

  8. If the power of the state has grown steadily over time, as claimed by scholars ranging from Weber (1958) to Mann (1993) and Tilly (1992), then the test of the relationship between inequality and democracy on data from contemporary states is evaluating the theory in a set of cases where it is most likely to hold because state capacity is particularly high. Care should be taken not to make unjustified generalizations from the findings of analyses of contemporary cases to earlier periods in history where states were much weaker. The distinct but parallel methodological error of oversampling from developed countries (where state capacity is higher) in cross-national studies, and generalizing to all countries in the world also characterizes this literature, as discussed in detail below.

  9. Among many arguments to this effect, see the essays by Rueschemeyer and Evans in Lange and Rueschemeyer 2005.

  10. One might also argue that state leaders respond to the difficulty in building capacity in the short run by choosing to bear the fiscal and political costs of maintaining extractive and coercive capacity even during peacetime.

  11. An additional assumption is that in deriving their preferences about regime types based on the prospects for redistribution, all social actors possess full information about the strength of the state. This assumption might be relaxed to complicate the account provided here.

  12. Notably, Côte d’Ivoire is coded as a strong state for most years (1975–84, 1988–) according to the census measure of state capacity developed below; thus, the high stakes of electoral conflict in terms of land rights in this case do not challenge the logic of my argument. Other cases cited by Boone are coded as weak states below; land wealth remains (as this section describes) the most propitious avenue for redistribution even under these circumstances.

  13. Salience refers not to the effects of a policy, but to the visibility of those effects to voters. (Pierson and Hacker 2005) The salience of redistributive efforts is taken for granted in the literature linking inequality and democracy. The effects on these models of incorporating variation in policy salience remains an area for future research.

  14. I do not extend the analysis past 1990 for two reasons: first, my focus in this paper is on a crucial omitted variable in “economic origins” analyses rather than on developing a new theory of the determinants of regime outcomes. Thus, my focus is on replicating Boix’s analysis as closely as possible. Second, recent scholarship such as that of Levitsky and Way (2010) shows that regime dynamics changed with the end of the Cold War.

  15. Measures of the difficulty of terrain are certainly exogenous from political dynamics, and are therefore at first glance appealing as proxies for state capacity. However, their time-invariant nature makes them inappropriate for this analysis.

  16. The ability to tax was one of the eight benefits of the census offered by medieval state theorist Jean Bodin, arguing that the census “provides for the just grievances of the poor against the rich (Alterman 1969, 60).

  17. The case of Brazil, where despite an effective and thorough census, the government’s ability to collect taxes is limited, shows the imperfections of the use of the census as a measure of state extractive capacity. Nevertheless, the census provides a systematic, if imperfect, assessment of the ability of the state to collect costly information from society. I thank Zachary Elkins for bringing the example of Brazil to my attention.

  18. This data is available in the electronic supplementary material for this paper (see note below on electronic supplementary material)

  19. Given the vagaries of census administration, this index contains some surprising variation. For example, the gap between 1970 and 1981 in census administration in Belgium leads the year 1980 to be coded a “0” while all other years between 1961 and 1990 are coded “1”. To smooth the data and capture a longer-term sense of the power of the state, I build a second measure, which assesses whether there have been at least two qualifying census events in the previous 25 years. Using this second measure does not change any of the substantive findings, but it reduces the number of cases covered by nearly 2,300. All results discussed in the remainder of the paper use the 10-year measure. Results for the 25 year measure are available from the author.

  20. This claim is based on a slightly different operationalization of the census measure: number of census iterations between 1950 and 2000, and using values for other measures for the year 2000. The correlation between census and tax ratio is 0.586, between the census measure and the male literacy rate 0.399, and between the census measure and the primary school completion rate 0.533.

  21. Other structural characteristics appear to have little effect on census administration: country size is correlated with census administration at 0.059, and the standard measure of mountainousness (taken from Fearon and Laitin 2003) is correlated with census administration at −0.007.

  22. The fact that domestic factors have a muted impact on the implementation of the census are consonant with the findings of Ventresca (1995) that international factors are central in explaining the dynamics of census administration.

  23. A list of all codings for state strength based on the method described above is included as Table A in the electronic supplementary materials.

  24. I thank Carles Boix for generously providing his dataset for replication.

  25. The dataset is available for download from I use the “suggested best fit” measure from the dataset in the analysis below.

  26. Boix (2003, 76 fn9) dismisses an earlier version of this data for methodological reasons. Babones and Alvarez-Rivadulla (2007) describe the version used here as a vast improvement over both earlier versions and other available data. The stakes of using different inequality datasets are described in more detail below.

  27. This model does not include other control variables, such as resource dependence or level of development, that are common in studies of regime dynamics. Boix (2003, Chapter 2) addresses these choices, and provides a discussion of the results from alternative specifications that do include measures of these variables.

  28. Additionally, in my replication, the beta coefficient for religious fractionalization surpasses the 5 % threshold, while it is only significant at the 10 % level for Boix.

  29. As Herrera and Kapur (2007) show, this pattern is not unique to this particular dataset.

  30. To properly model the interaction, the census variable itself must be added to the regression, as well as the interaction of census and lagged regime (which must be added because inequality is interacted with the lagged regime variable). The latter interaction is included in the results as the alpha value for the census.

  31. Using Boix’s original data and the Deininger and Squire operationalization of inequality shown in the leftmost columns of Table 4, I find that where the state is strong, the marginal effect of inequality on democracy is estimated to be −0.043, with the 95 % confidence interval bounded at {−0.063, −0.023}. In other words, the relationship is significant and negative. Where the state is weak, the point estimate is 0.006, with a 95 % confidence interval bounded at {−0.049, 0.060}. The relationship, in other words, cannot be statistically distinguished from zero. Graphical representations of these results, using the technique developed by Braumoeller (2004) are deleted for lack of space but are available from the author on request.

  32. These findings are robust to using the Beck et al. (1998) approach to time-dependent binary TSCS data, and to a split-sample analysis using the SIDD inequality data. Results are included as Table B and C in the electronic supplementary material. The split-sample analysis is marked by very unbalanced sample sizes because the census measure codes most cases as strong states. Yet the coefficient on inequality for the weak state analysis is not only statistically indistinguishable from zero, but also oppositely signed from what the Boix model predicts—thus the rejection of the Boix hypothesis in the weak state cases is almost certainly not an artifact of the small number of cases coded as having weak states. Nevertheless, a more nuanced, multichotomous, measure of state capacity, or one with a different cut-point, would be useful for further exploring this issue. Replicating and extending Model 1A in Boix (2003), which includes wealth as an independent variable, yields similar results for the conditional effect of state capacity on the relationship between inequality and regime dynamics.

  33. On external pressures as fundamental to regime outcomes in the developing world see Huntington 1991 and Levitsky and Way 2010.

  34. Hagopian (1990) argues that the ability of regime-associated elites to maintain their hold on political positions after the transition to democracy might also underlie their surprising willingness to accept regime change. This mechanism is distinct from the role of weak enforcement emphasized in this paper.

  35. A related but distinct claim is common in the literature on state formation, where many argue that social actors are only willing to accept increased state capacity if political representation is offered in return. See for example Tilly (1992) and Levi (1988).

  36. On the lack of a relationship between democracy and redistribution in Latin America, see Menaldo (2008).


  • Acemoglu D, Robinson JA. Economic origins of dictatorship and democracy. Cambridge University Press; 2006.

  • Acemoglu D, Robinson JA. Why did the West extend the franchise? Growth, inequality, and democracy in historical perspective. Q J Econ. 2000;CXV:1167–99.

    Article  Google Scholar 

  • Acemoglu D, Robinson JA. A theory of political transitions. Am Econ Rev. 2001;91:938–63.

    Article  Google Scholar 

  • Alesina A, Baqir R, Easterly W. Public goods and ethnic divisions. QJ Econ (November). 1999;CXIV(4):1243–84.

    Article  Google Scholar 

  • Alterman H. Counting people: the census in history. New York: Harcourt, Brace and World; 1969.

    Google Scholar 

  • Ansell B, Samuels D. ‘Inequality and democratization: a Contractarian approach’ Comparative Political Studies (December). 2010.

  • Babones S, Alvarez-Rivadulla MJ. Standardized income inequality data for use in cross-national research. Sociol Inq. 2007;77(1 (February)):3–22.

    Article  Google Scholar 

  • Baland J-M, Robinson JA. ‘Land and power: theory and evidence from Chile’ (Unpublished Manuscript). 2004.

  • Barnett MN. Confronting the Costs of War: Military Power, State, and Society in Egypt and Israel. Princeton University Press; 1992.

  • Beck N, Katz JN, Tucker R. Taking time seriously: time series cross section analysis with a binary dependent variable. American Journal of Political Science. 1998;42(4 (October)):1260–88.

    Article  Google Scholar 

  • Bellin E. The robustness of authoritarianism in the Middle East in comparative perspective. Comparative Politics. 2004;36(2 (January)):139–57.

    Article  Google Scholar 

  • Boix C. Democracy and redistribution. Cambridge University Press; 2003.

  • Boix C. and Sebastián R. ‘A Complete Data Set of Political Regimes 1800-1999’. The University of Chicago; 2001.

  • Boone C. ‘Electoral populism where property rights are weak’ Comparative Politics (January). 2009. pp.183–201.

  • Braumoeller B. Hypothesis testing and multiplicative interaction terms. Int Organ. 2004;58(Fall):807–20.

    Google Scholar 

  • Centeno MA. Blood and debt: war and the nation state in Latin America. Penn State University Press; 2002.

  • Colburn FD, Rahmato D. Rethinking socialism in the Third World. Third World Q. 1992;13(1):159–72.

    Article  Google Scholar 

  • Collier D. Squatters and oligarchs: authoritarian rule and policy change in Peru. Johns Hopkins University Press; 1976.

  • Dahl R. Polyarchy: participation and opposition. Yale University Press; 1971.

  • Deininger K, Squire L. A new data set measuring income inequality. The World Bank Econ Rev. 1996;10(3):565–91.

    Article  Google Scholar 

  • Dunning T. Crude democracy: natural resource wealth and political regimes. Cambridge University Press; 2008.

  • Durand F, Thorp R. Reforming the State: A Study of the Peruvian Tax Reform. Oxford Agrarian Studies. 1998;26(2):133–51.

    Article  Google Scholar 

  • Eckstein S. The poverty of revolution: the state and the urban poor in Mexico. Princeton University Press; 1988.

  • Epstein DL, Bates R, Goldstone J, Kristensen I, O’Halloran S. Democratic transitions. American Journal of Political Science. 2006;50(3 (July)):551–69.

    Article  Google Scholar 

  • Fearon J, Laitin D. Ethnicity, Insurgency, and Civil War. American Political Science Review. 2003;97(1):75–90.

    Article  Google Scholar 

  • Hagopian F. Democracy by undemocratic means? Comparative Political Studies. 1990;23(3):147–70.

    Article  Google Scholar 

  • Herrera Y, Kapur D. Improving data quality: actors, incentives, and capabilities. Polit Anal. 2007;15(4):365–86.

    Article  Google Scholar 

  • Herring RJ. Zulfikar Ali Bhutto and eradication of feudalism in Pakistan. Comp Stud Soc Hist. 1979;21(4):519–57.

    Article  Google Scholar 

  • Hidalgo FD, et al. (2010) ‘Economic Determinants of Land Invasions’. Rev Econ Stat. 2010;29(3):505–23.

    Article  Google Scholar 

  • Hoffman K, Centeno MA. The lopsided continent: inequality in Latin America. Annu Rev Sociol. 2003;29:363–90.

    Article  Google Scholar 

  • Hunter W. ‘Contribution to symposium: big unanswered questions in comparative politics’ APSA-CP Newsletter vol. 19 #1 (Winter). 2008.

  • Huntington S. The Third Wave: Democratization in the Late Twentieth Century. University of Oklahoma Press; 1991.

  • Kain RJP. and Baigent E. The cadastral map in the service of the state: a history of property mapping. University of Chicago Press; 1992.

  • Knight A. The Mexican Revolution vol. 1: Porfirians, liberals, and peasants. Lincoln: University of Nebraska Press; 1986.

    Google Scholar 

  • Kohl J. Peasant and revolution in Bolivia, April 9, 1952–August 2, 1953. Hisp Am Hist Rev. 1978;58:238–59.

    Article  Google Scholar 

  • La Porta R, Lopez de Silanes F, Shleifer A, and Vishny R. ‘The quality of government’ NBER Working Paper no.6727. 1998.

  • Lange M, Rueschemeyer D. States and development: historical antecedents of stagnation and advance. Palgrave Macmillan; 2005.

  • LeGrand C. Frontier expansion and peasant protest in Colombia 1850–1936. University of New Mexico Press; 1986.

  • Levi M. Of rule and revenue. Berkeley: University of California Press; 1988.

    Google Scholar 

  • Levitsky S, Way L. Competitive Authoritarianism: Hybrid Regimes after the Cold War. Cambridge University Press; 2010.

  • Lieberman ES. Race and regionalism in the politics of taxation in Brazil and South Africa. Cambridge: Cambridge University Press; 2003.

    Book  Google Scholar 

  • Linz J. and Stepan A. Problems of Democratic Transition and Consolidation: Southern. 1996.

  • Lipset SM. Political Man: the Social Bases of Politics. Johns Hopkins University Press; 1959.

  • Loveman B. Struggle in the countryside: politics and rural labor in Chile, 1919–1973. Bloomington IN: University of Indiana Press; 1976.

    Google Scholar 

  • Mainwaring S, Pérez-Liñan A. Level of Development and Democracy: Latin American Exceptionalism 1945-1996. Comparative Political Studies. 2003;36(9):1031–67.

    Article  Google Scholar 

  • Mann M. The autonomous power of the state: its origins, mechanisms and results. Archives Européenes de Sociologie. 1984;25:185–213.

    Article  Google Scholar 

  • Mann, M. The sources of social power volume 2: the rise of classes and Nation-States 1760-1914. New York: Cambridge University Press; 1993.

  • Mares I. Social protection around the world: external insecurity, state capacity, and domestic political cleavages. Comparative Political Studies. 2005;38(6 (August)):623–51.

    Article  Google Scholar 

  • Meltzer AH, Richard SA. A rational theory of the size of government. J Polit Econ. 1981;89(5):914–27.

    Article  Google Scholar 

  • Menaldo V. ‘Banking on redistribution: financial institutions and fiscal redistribution in Latin American democracies’ (Paper Presented at APSA). 2008.

  • O’Donnell G. On the state, democratization and some conceptual problems: a Latin American view with some glances at post-Communist countries. World Dev. 1993;21(8):1355–69.

    Article  Google Scholar 

  • O’Donnell G. and Schmitter P. Transitions from Authoritarian Rule: tentative Conclusions about Uncertain Democracies. Johns Hopkins University Press; 1986.

  • O’Neil, S. ‘Fixing social security in Latin America (Again)’ Americas Quarterly (Fall). 2007.

  • Onoma KA. The politics of property rights institutions in Africa. Cambridge University Press; 2010.

  • Pierson P. and Hacker JS. Off center: the Republican Revolution and the Erosion of American Democracy. Yale University Press; 2005.

  • Powelson J, Stock R. The peasant betrayed: agriculture and land reform in the Third World. Cato Institute Press; 1990.

  • Przeworski A, Alvarez ME, Cheibub J, and Limongi F. Democracy and development. Cambridge University Press. 2000.

  • Rueschemeyer D, Huber Stephens E, and Stephens J. Capitalist development and democracy. University of Chicago Press; 1992.

  • Scott JC. Seeing like a state. Yale University Press; 1998.

  • Seligmann LJ. Between reform and revolution: political struggles in the Peruvian Andes, 1969–1991 (Stanford). Stanford, CA: Stanford University Press; 1995.

    Google Scholar 

  • Skocpol T. States and social revolutions. Cambridge: Cambridge University Press; 1979.

    Book  Google Scholar 

  • Slater D. Ordering power: contentious politics and authoritarian leviathans in Southeast Asia. Cambridge University Press; 2010.

  • Smith B. ‘Rethinking the economic origins of dictatorship and democracy: the continuing value of cases and comparisons’ APSA-CP Newsletter. Winter; 2008. pp.16-21.

  • Thies CG. ‘State Building, Interstate and Intrastate Rivalry: A Study of Post-colonial Developing Country Extractive Efforts 1975–2000. Int Stud Quart. 2004;48(1):53–72.

    Article  Google Scholar 

  • Tilly C. Coercion, capital, and European states AD 990–1992. Blackwell; 1992.

  • Ventresca M. When states count: institutional and political dynamics in modern census establishment 1800–1993. Stanford University: PhD Dissertation; 1995.

    Google Scholar 

  • Weber M. From Max Weber: essays in sociology H. H. Gerth and C. Wright Mills, eds. Oxford University Press 1958.

  • Wright TC. Landowners and reform in Chile. Urbana: University of Illinois Press; 1982.

    Google Scholar 

Download references


Earlier versions of this paper were presented at the 2008 meetings of the Midwest Political Science Association and the American Political Science Association, Temple University, and Dartmouth University, and an early version was published as Brooks World Poverty Institute Working Paper #93 in 2009. I thank Carles Boix for sharing his data, and Mark Copelovitch, Kevin Arceneaux, Megan Mullin, Ryan Vander Wielen, and Ben Ansell for answering queries about methodological issues. Carles Boix, John Carey, Mark Copelovitch, Jorge I. Domínguez, Thad Dunning, Zach Elkins, Daniel Gingerich, Kimuli Kasara, Steve Levitsky, Juan Pablo Luna, Victor Menaldo, Dan Reiter, Rachel Beatty Riedl, James Robinson, David Samuels, Prerna Singh, Maya Tudor, Milan Vaishnav, Matthias vom Hau, Daniel Ziblatt, and the graduate students in my seminar ‘The State in Comparative Politics’, and colleagues in the Political Science department at Temple read early versions and provided helpful comments. I thank them as well as the reviewers and journal editors for suggestions that greatly improved the paper.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Hillel David Soifer.

Additional information

An erratum to this article can be found online at

An erratum to this article is available at

Electronic supplementary materials

Below is the link to the electronic supplementary material.


(DOCX 61 KB)

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Soifer, H.D. State Power and the Economic Origins of Democracy. St Comp Int Dev 48, 1–22 (2013).

Download citation

  • Published:

  • Issue Date:

  • DOI:


  • Democratization
  • State capacity
  • Redistribution
  • Inequality