Abstract
Decentralization is meant to improve access to public services, but relatively few studies examine this question empirically. We explore the effects of decentralization on access to health and education in Colombia using an original database covering over 95 % of Colombian municipalities. We show that decentralization improved enrollment rates in public schools and access of the poor to public health services. In both sectors, improving access was driven by the financial contributions of local governments. Small increases in own-shares of spending led to surprisingly large increases in the access of the poor in both sectors. Our theoretical model implies that where local information dominates productive efficiency, elected local governments will provide services better tailored to local needs. Decentralizing such services should increase their use by the public. Together, theory and empirics imply that decentralization made the Colombian state more accountable. It provided local officials with the information and incentives they need to allocate resources in a manner responsive to voters’ needs and improve the impact of public expenditures.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.Notes
Colombia’s public accounts classify such items as teachers’ and health workers’ salaries as investments, and not running costs.
Which could allow central government to serve a given population with fewer facilities, and hence, lower cost than numerous local governments would provide.
We consider only non-negative values of g l∗ and g c∗.
Reversing the order of play and solving for local government as the Stackelberg leader yields symmetric results in which all provision is local unless α<τ. We consider this order of play less realistic.
Health data are available for the period 1997–2004.
More data on a wider variety of local characteristics are collected in Colombia than any other country in the region bar Brazil.
Meaning that we do not combine information from different sources into a single variable.
According to Law 617 of 2000, local taxes and other revenues levied locally belong to municipalities and may be disposed of freely by them. Such revenues differ from central transfers, which must be spent in specific ways on specific public goods. The sources of revenues available to municipalities to finance education and health services combine own resources, including tax and non-tax revenues, and transfers from central government. The latter are divided into transfers for education, health and general purposes. Up to 28 % of transfers for general purposes may be spent on running costs in the smallest and poorest municipalities, and less in larger ones (Articles 78 and 79, Law 715 of 2001). For more on this see Sánchez and Zenteno (2010).
Law 60 of 1993 regulated the transfer system between 1994–2001. According to that law, transfers to municipalities increased with population and relative poverty as measured by Unsatisfied Basic Needs: an index measuring the proportion of the local population for whom a predetermined level of basic needs are unmet.
The proportion of local educational spending undertaken by departmental governments is omitted in order to avoid perfect collinearity.
Fondo de Solidaridad y Garantía (literally the Solidarity and Guarantee Fund).
Departmental health spending in the municipality as a proportion of total health spending is left out of the regression so as to avoid perfect collinearity.
Total funds available locally for education and health include direct transfers from central government to municipalities, indirect spending through departments (states) for non-certified municipalities, and municipalities’ own resources.
Hausman test results for education: chi-square(1)=0.02 with Prob>chi-square=0.8997; for health: chi-square(1)=0.21 with Prob>chi-square=0.6449.
Wooldridge test results for education: F(1,1075)=58.66 with Prob=0.00; for health: F(1,1056)=8.660 with Prob=0.0033.
According to Law 141, those who exploit a non-renewable natural resource must pay a percentage of the value of production in royalties. Percentages vary according to the resource exploited and the amount extracted. Royalties are paid to the Treasury, which distributes these amongst the departments and municipalities where exploitation took place, including exporting ports and municipalities crossed by pipelines.
Note that the dependent variable is not censored/truncated. Observed “zeros” are real zeros, and not failures of measurement or excluded negative values. A 2SLS panel estimation therefore is appropriate. As a check, we also estimated the IV model with a Tobit first stage. The findings did not change.
The second instrument used in the education equation is the land Gini coefficient, under the neoinstitutionalist assumption that the concentration of economic power affects the level of investment in public goods. In the case of health, the second instrument used is FARC guerrilla activity, under the assumption that a municipality confronted with illegal armed groups must divert resources from social to other types of spending, such as security and infrastructure reconstruction (Sánchez and Diaz 2007). Both instruments have the expected sign.
References
Barankay, I., & Lockwood, B. (2007). Decentralization and the productive efficiency of government: evidence from Swiss cantons. Journal of Public Economics, 91, 1197–1218.
Batina, R. G., & Ihori, T. (2005). Public goods: theories and evidence. New York: Springer.
Besley, T., & Coate, S. (2003). Centralized versus decentralized provision of local public goods: a political economy approach. Journal of Public Economics, 87, 2611–2637.
Cassette, A., & Paty, S. (2010). Fiscal decentralization and the size of government: a European country empirical analysis. Public Choice, 143, 173–189.
Ceballos, M., & Hoyos, D. (2004). Tendencias del comportamiento electoral y descentralización en los municipios de Colombia, 1988–2000. Crisis States Programme Working Paper No. 57. London School of Economics.
Clark, D. (2009). The performance and competitive effects of school autonomy. Journal of Political Economy, 117(4), 745–782.
Devarajan, S., Khemani, S., & Shah, S. (2009). The politics of partial decentralization. In E. Ahmad & G. Brosio (Eds.), Does decentralization enhance service delivery and poverty reduction? Cheltenham: Edward Elgar.
Dowding, K., & John, P. (1994). Tiebout: a survey of the empirical literature. Urban Studies, 31(4/5), 767–797.
Escaleras, M., & Register, C. A. (2012). Fiscal decentralization and natural hazard risks. Public Choice, 151(1–2), 165–183.
Faguet, J. P. (2012). Decentralization and popular democracy: governance from below in Bolivia. Ann Arbor: University of Michigan Press.
Faguet, J. P., & Sánchez, F. (2008). Decentralization’s effects on educational outcomes in Bolivia and Colombia. World Development, 36(7), 1294–1316.
Galiani, S., Gertler, P., & Schargrodsky, E. (2008). School decentralization: helping the good get better, but leaving the poor behind. Journal of Public Economics, 92, 2106–2120.
Gonçalves, S. (2013). The effects of participatory budgeting on municipal expenditures and infant mortality in Brazil. World Development. doi:10.1016/j.worlddev.2013.01.009.
Ladino O., W. H. (2008). Organización del estado Colombiano y formas organizativas del estado a nivel territorial. Bogotá:Escuela Nacional de Administración Pública.
Manor, J. (1999). The political economy of democratic decentralization. Washington: World Bank.
Oates, W. (1972). Fiscal federalism. New York: Harcourt Brace.
Ostrom, E., & Whitaker, G. P. (1973). Does local community control of police make a difference? Some preliminary findings. American Journal of Political Science, 17, 48–76.
Ostrom, E., Schroeder, L., & Wynne, S. (1993). Institutional incentives and sustainable development: infrastructure policies in perspective. Boulder: Westview Press.
Prohl, S., & Schneider, F. (2009). Does decentralization reduce government size? A quantitative study of the decentralization hypothesis. Public Finance Review, 37(6), 639–664.
Republica de Colombia (2001). Ley 715 de 2001. Bogotá: Gobierno Nacional.
Rodden, J. (2003). Reviving leviathan: fiscal federalism and the growth of government. International Organization, 57, 695–729.
Sánchez, F., & Diaz, A. (2007). Los efectos sociales del conflicto armado interno. In F. Sánchez (Ed.), Las cuentas de la violencia, Bogotá: Editorial Norma.
Sánchez, F., & Zenteno, J. (2010). Descentralización y desempeño fiscal: el caso de Colombia. In M. del Valle & A. Galindo (Eds.), Descentralización y sostenibilidad fiscal subnacional: los casos de Colombia y Perú. Región Andina: Banco Interamericano de Desarrollo.
Tiebout, C. M. (1956). A pure theory of local expenditures. Journal of Political Economy, 64, 416–424.
Treisman, D. (2007). The architecture of government: rethinking political decentralization. New York: Cambridge University Press.
Varian, H. (1994). Sequential contributions to public goods. Journal of Public Economics, 53, 165–186.
World Bank (2004). World development report 2004: making services work for poor people. New York: Oxford University Press.
Wu, A. M., & Lin, M. (2012). Determinants of government size: Evidence from China. Public Choice, 151(1–2), 255–270.
Acknowledgements
This paper was written while Faguet was on research sabbatical at the Center for Latin American Studies, UC Berkeley, to whom he is grateful for warm hospitality and support. The research was financed by the Corporación Andina de Fomento’s Research Papers Program, a STICERD/LSE New Researcher Award, and the British Academy. We are very grateful to Patricia Rincón, Camila Torrente and Victoria Soto for expert research assistance, and to Robin Burgess, Maitreesh Ghatak, Alain de Janvry, Asim Khwaja, Dilip Mookherjee, Daniel Ortega, Pablo Sanguinetti, Daniel Treisman, Hernan Vallejo, three anonymous reviewers, and seminar participants at STICERD and the LACEA 2008 meetings for their thoughtful suggestions. All remaining errors are ours.
Author information
Authors and Affiliations
Corresponding author
Appendix: Data summary
Appendix: Data summary
Variable* | Obs. | Mean | Std. Dev. | Min. | Max. |
|---|---|---|---|---|---|
Education (1994–2004) | |||||
Increase in student enrollment in public schools (ln) | 10553 | 0.0238 | 0.1091 | −0.5214 | 1.5224 |
Own resources/Total education expenditures | 10559 | 0.0406 | 0.0731 | 0 | 0.9938 |
Municipal independence | 10559 | 0.0117 | 0.1045 | 0 | 1 |
Statutory transfers (poverty)/Total education expenditures | 10559 | 0.1744 | 0.1344 | 0 | 0.6580 |
Statutory transfers (No. of students)/Total education expenditures | 10559 | 0.0351 | 0.0732 | 0 | 0.9960 |
Municipal expenditure growth | 10559 | 0.0537 | 0.1688 | −0.7263 | 1.8967 |
Per capita expenditure on public education (ln) | 10559 | 13.7073 | 0.3643 | 11.8198 | 15.5100 |
Student/Teacher ratio (lagged) | 10559 | 22.3558 | 9.1899 | 0 | 446.0 |
University graduates as a share of municipal personnel | 10559 | 0.0696 | 0.1008 | 0 | 0.8937 |
Population (ln) | 10559 | 9.6355 | 1.0456 | 6.3297 | 15.7657 |
Unsatisfied basic needs | 10559 | 45.3104 | 22.2733 | 1.1293 | 105.2663 |
Displaced population, receiving municipalities | 10559 | 0.0014 | 0.0087 | 0 | 0.3503 |
Displaced population, expelling municipalities | 10559 | 0.0056 | 0.0245 | 0 | 0.7788 |
Unemployment rate (departmental) | 10559 | 0.1270 | 0.0849 | 0.03 | 0.9990 |
Public-school gross enrollment rate (lagged) (% of school-age population) | 10559 | 0.9720 | 0.3906 | 0.1 | 2.5000 |
Private enrollment rate (% school-age pop. in private schools) (ln, lagged) | 10559 | 0.0133 | 0.0266 | 0 | 0.4532 |
Per capita local taxes (ln) | 10559 | −4.4642 | 1.2134 | −9.3527 | −0.5045 |
Health (1997–2004) | |||||
Increase in health insurance amongst the poor (ln) | 6266 | 0.0829 | 0.1732 | −0.4976 | 2.4013 |
Own resources/Total health expenditures | 6267 | 0.0090 | 0.0321 | 0 | 0.8778 |
Municipal independence | 6267 | 0.0070 | 0.0800 | 0 | 1 |
Statutory transfers (poverty)/Total health expenditures | 6267 | 0.1340 | 0.1500 | 0 | 0.7360 |
Statutory transfers (No. of insured poor)/Total health expenditures | 6267 | 0.2600 | 0.0900 | 0 | 0.9180 |
Municipal expenditure growth | 6267 | 0.0308 | 0.2179 | −0.6820 | 3.1490 |
Per capita expenditure on public health (ln) | 6267 | 11.9546 | 0.5395 | 8.9660 | 14.5126 |
University graduates as a share of municipal personnel | 6267 | 0.0695 | 0.1010 | 0 | 0.8937 |
Population (ln) | 6266 | 9.6477 | 1.0588 | 6.3297 | 15.7657 |
Unsatisfied basic needs | 6267 | 42.8152 | 22.6682 | 1.1293 | 104.2634 |
Displaced population, receiving municipalities | 6267 | 0.0023 | 0.0112 | 0 | 0.3503 |
Displaced population, expelling municipalities | 6267 | 0.0090 | 0.0313 | 0 | 0.7788 |
Unemployment rate (departmental) | 6267 | 0.1417 | 0.0795 | 0.050 | 0.9990 |
% coverage of public health insurance | 6267 | 0.5915 | 0.4981 | 0.025 | 6.8081 |
Per capita local taxes (ln) | 6262 | 0.0071 | 0.0239 | 0.000 | 0.5450 |
Rights and permissions
About this article
Cite this article
Faguet, JP., Sánchez, F. Decentralization and access to social services in Colombia. Public Choice 160, 227–249 (2014). https://doi.org/10.1007/s11127-013-0077-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11127-013-0077-7