Skip to main content

Democracy, populism and hyperinflation: some evidence from Latin America


In this paper we test for the populist view of inflation in South America during the eventful period between 1970 and 2007, a period which captures the latest democratic transition in the continent, episodes of hyperinflation and finally macroeconomic stabilisation. The results—based on panel time-series data and analysis—confirm the prediction which suggests that recently elected governments coming into power after periods of political dictatorship, and which are faced with demand for redistribution, end up engaging in populist (or redistributive) policies, which tend to lead to high inflation and overall poor macroeconomic performance. All in all, we suggest that the implementation of democracy as such requires not only the “right political context”—or an appropriately constrained executive—to work well, but it also must come with certain economic institutions (central bank independence and a credible and responsible fiscal authority), institutions that were (coincidentally) absent in South America right after re-democratisation, but which would presumably raise the costs of pursuing populist policies in the first place.

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

Fig. 1
Fig. 2


  1. For instance, Bittencourt (2009) investigates the case of the Brazilian hyperinflation of the 1980s and 1990s, and he suggests that the high rates of inflation seen at the time contributed to increase earnings inequality. Moreover, Easterly and Fischer (2001) suggest that the poor from 38 countries consider inflation to be a more pressing problem than the rich, which suggests that the poor are the ones suffering more with higher inflation.

  2. For instance, Aisen and Veiga (2006) make use of GMM estimators, however they are not able to find any significant effect of democracy on inflation when they make use of a sample only of developing countries, which includes South America. In addition Acemoglu et al. (2008) suggest that in countries with intermediate levels of development (represented by a dummy which includes South America) the implementation of central bank independence has been beneficial to macroeconomic performance. The problem is that most South American countries implemented central bank independence well after their macroeconomic stabilisations, towards the end of the 1990s, which indicates that central bank independence in its own is probably not the sole reason for the stabilisation of the early 1990s in the region.

  3. In addition, Al-Marhubi (1997) suggests that higher inequality is positively associated with higher inflation rates in a cross-section of countries, Desai et al. (2005) suggest that inequality affects inflation, but conditional on the political structure, and Dutt and Mitra (2008) suggest that excessive inequality leads to political instability, which in turn leads to policy volatility, and therefore lower investment and economic growth.

  4. Although we report estimates using the baseline \(DEMOC\) alongside \(POLITY\), in the “Appendix”, for robustness sake, we also report results using the variable constraints on the executive \((XCONST)\).

  5. An alternative to IPS (2003) is the test by Levin et al. (2002). However, this test assumes parameter homogeneity, and therefore does not consider a possible heterogeneity bias present in the data. Moreover, given that these countries shared some macroeconomic characteristics in the 1980s and early 1990s, some would argue that there is between-country dependence present. An alternative that considers the existence of between-country dependence is proposed by Pesaran (2007), the cross-section IPS (CIPS) test. However, CIPS assumes that \(N>10\) and we have \(N=4\) in our data set. In addition, one would argue that, given the structure of the data, structural breaks are a possibility. The test proposed by Im et al. (2005) takes that into account. However, this test also assumes large \(N\), which is not entirely the case here. Basically, the IPS test presents more flexibility in terms of sample size and asymptotics, and is therefore informative and the best alternative available at this stage.

  6. The Mean Group (MG) estimator, proposed by Pesaran and Smith (1995), which is the simple time-series averages of all countries, is also an alternative. However, this estimator is sensitive to outliers, a problem not faced by the RC estimator. Nevertheless, for robustness sake, we report in the “Appendix” the dynamic MG estimates. Moreover, the Random Effects estimator is not needed in this instance because when \(T\rightarrow \infty \) (which is the case here) it is equivalent to the FE estimator (Arellano 2003). In addition, Bond (2002) argues that GMM-type estimators are not an alternative under \(T>N\) because of the overfitting problem.

  7. In addition, we make use of a variable that counts the number of years after democratisation and its square term to account for the fact that perhaps democratic maturity comes in the long run. We report those alternative estimates in the “Appendix”.

  8. An alternative to SUR is the Common Effects Estimator proposed by Pesaran (2006). However, N is assumed to be large and in our data set \(N=4\). Furthemore, Kapoor et al. (2007) propose an estimator that also works best under the \(N\rightarrow \infty \) assumption.

  9. For a more thorough discussion about panel time-series analysis in general, see Smith and Fuertes (2008).

  10. For the sake of space we do not report the dynamic SUR estimates, nevertheless, the results are quantitatively and qualitatively similar to the ones in Table 5. Available on request.

  11. When we estimate static and dynamic equations with a dummy for the last 4 years of dictatorship, the estimates are negative and statistically significant, indicating lower inflation rates during the period. Available on request.

  12. For instance, Santiso (2006) highlights the importance of the much improved macroeconomic performance in Latin America recently to produce better economic outcomes from the late 1990s onwards.


  • Acemoglu D, Egorov G, Sonin K (2011) A political theory of populism (unpublished)

  • Acemoglu D, Johnson S, Querubín P, Robinson JA (2008) When does policy reform work? The case of central bank independence. Brookings Pap Econ Act 1:353–418

    Google Scholar 

  • Acemoglu D, Johnson S, Robinson J, Thaicharoen Y (2003) Institutional causes, macroeconomic symptoms: volatility, crises and growth. J Monet Econ 50(1):49–123

    Google Scholar 

  • Aisen A, Veiga FJ (2006) Does political instability lead to higher inflation? A panel data analysis. J Money Credit Banking 38(5):1379–1389

    Google Scholar 

  • Alesina A, Drazen A (1991) Why are stabilizations delayed? Am Econ Rev 81(5):1170–1188

    Google Scholar 

  • Al-Marhubi F (1997) A note on the link between income inequality and inflation. Econ Lett 55:317–319

    Article  Google Scholar 

  • Arellano M (2003) Panel data econometrics. Oxford University Press, Oxford

    Book  Google Scholar 

  • Beetsma Roel MWJ, Van Der Ploeg F (1996) Does inequality cause inflation? The political economy of inflation, taxation and government debt. Public Choice 87:143–162

    Google Scholar 

  • Bittencourt M (2009) Macroeconomic performance and inequality: Brazil, 1983–94. Dev Econ 47(1):30–52

    Google Scholar 

  • Bond SR (2002) Dynamic panel data models: a guide to micro data methods and practice. Portuguese Econ J 1(2):141–162

  • Brender A, Drazen A (2007) Why is economic policy different in new democracies? Affecting attitudes about democracy. NBER working paper series

  • Desai RM, Olofsgård A, Yousef TM (2003) Democracy, inequality, and inflation. Am Political Sci Rev 97(3):391–406

    Google Scholar 

  • Desai RM, Olofsgård A, Yousef TM (2005) Inflation and inequality: does political structure matter? Econ Lett 87:41–46

    Google Scholar 

  • Dornbusch R, Edwards S (1990) Macroeconomic populism. J Dev Econ 32:247–277

    Article  Google Scholar 

  • Dutt P, Mitra D (2008) Inequality and the instability of polity and policy. The Economic Journal 118:1285–1314

    Article  Google Scholar 

  • Easterly W, Fischer S (2001) Inflation and the poor. J Money Credit Banking 33(2):160–178

    Article  Google Scholar 

  • Gasiorowski M (1995) Economic crisis and political regime change: an event history analysis. Am Political Sci Rev 89(4):882–897

    Article  Google Scholar 

  • Im K-S, Lee J, Tieslau M (2005) Panel LM unit-root tests with level shifts. Oxf Bull Econ Stat 67:393–419

    Google Scholar 

  • Im KS, Pesaran MH, Shin Y (2003) Testing for unit roots in heterogeneous panels. J Econ 115(1):53–74

    Google Scholar 

  • Kapoor M, Kelejian HH, Prucha IR (2007) Panel data models with spatially correlated error components. J Econ 140(1):97–130

    Google Scholar 

  • Levin A, Lin C-F, James Chu C-S (2002) Unit root tests in panel data: asymptotic and finite-sample properties. J Econ 108(1):1–24

    Google Scholar 

  • Paldam M (1987) Inflation and political instability in eight Latin American countries 1946–83. Public Choice 52:143–168

    Article  Google Scholar 

  • Pesaran MH (2006) Estimation and Inference in large heterogeneous panels with a multifactor error structure. Econometrica 74(4):967–1012

    Article  Google Scholar 

  • Pesaran MH (2007) A simple panel unit root test in the presence of cross-section dependence. J Appl Econ 22(2):265–312

    Google Scholar 

  • Pesaran MH, Smith R (1995) Estimating long-run relationships from dynamic heterogeneous panels. J Econ 68(1):79–113

    Google Scholar 

  • Przeworski A, Limongi F (1997) Modernization: theories and facts. World Politics 49(2):155–183

    Article  Google Scholar 

  • Sachs JD (1989) Social conflict and populist policies in Latin America. In: NBER working paper series

  • Santiso J (2006) Latin America’s political economy of the possible. The MIT Press, Cambridge

    Google Scholar 

  • Sargent T, Williams N, Zha T (2009) The conquest of South American inflation. J Political Econ 117(2):211–256

    Google Scholar 

  • Smith R, Fuertes A-M (2008) Panel time-series. In: Centre for microdata methods and practice. Institute for Fiscal Studies, London

  • Swamy PAVB (1970) Efficient inference in a random coefficient regression model. Econometrica 38(2):311–323

    Article  Google Scholar 

  • Veiga FJ (2000) Delays of inflation stabilizations. Econ Politics 12:275–295

    Article  Google Scholar 

  • Zellner A (1962) An efficient method of estimating seemingly unrelated regressions and test for aggregation bias. J Am Stat Assoc 57(298):348–368

    Google Scholar 

  • Zellner A (1969) On the aggregation problem: a new approach to a troublesome problem. In: Fox KA et al (eds) Economic models, estimation and risk programming: essays in honor of Gerhard Tintner, Springer-Verlag, Berlin, pp 365–378

Download references

Author information

Authors and Affiliations


Corresponding author

Correspondence to Manoel Bittencourt.

Additional information

I thank the Editor, an anonymous referee, Johannes Fedderke, Reneé van Eyden, an ERSA referee, and seminar participants at Pretoria, UCT, Göttingen, Stellenbosch, ESSA 2009 in Port Elizabeth and Verein für Socialpolitk 2010 in Hannover for comments, and ERSA for financial support.



In Table 6 we report the estimates of a variable that counts the number of years \((YEARS)\) after democratisation and its respective squared term. The static estimates suggest a non-linear relationship, which indicates that in the long run democracy matures and with that maturity comes not only better institutions, but also better macroeconomic performance. Moreover, in Table 7 we report estimates using the Polity IV variable constraints on the executive, and the estimates confirm the ones reported in the paper, which suggest that macroeconomic performance deteriorated after democratisation. Finally, in Table 8 we report the dynamic pooled estimates provided by the MG estimator, and the results, once again, confirm the prediction that those countries experienced higher inflation rates after re-democratisation.

Table 6 Estimates of YEARS and \(\text{ YEARS}^{2}\) on inflation, 1970–2007
Table 7 Static and dynamic estimates of XCONST on inflation, 1970–2007
Table 8 Dynamic estimates of DEMOC, POLITY, DUMMY, XCONST and YEAR on inflation, 1970–2007

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Bittencourt, M. Democracy, populism and hyperinflation: some evidence from Latin America. Econ Gov 13, 311–332 (2012).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:


  • Democracy
  • Populism
  • Hyperinflation
  • Latin America

JEL Classification

  • E31
  • E65
  • N16
  • O23
  • O54