Hostname: page-component-78c5997874-94fs2 Total loading time: 0 Render date: 2024-11-11T13:59:01.119Z Has data issue: false hasContentIssue false
  • English
  • Français

The Dynamics of Poverty in Chile*

Published online by Cambridge University Press:  29 April 2008

CHRISTOPHER NEILSON ,
DANTE CONTRERAS ,
RYAN COOPER  and
JORGE HERMANN
CHRISTOPHER NEILSON
Affiliation:
Christopher Neilson works in the Central Bank of Chile
DANTE CONTRERAS
Affiliation:
Dante Contreras works in the Centro de Microdatos, University of Chile
RYAN COOPER
Affiliation:
Ryan Cooper works in the OPTE Foundation, Chile
JORGE HERMANN
Affiliation:
Jorge Hermann works in the Department of Economics, University of Chile.
Get access Rights & Permissions [Opens in a new window]

Abstract

This paper uses the 1996–2001 National Socioeconomic Survey panel database to analyse poverty dynamics in Chile, drawing a distinction between chronic and transient poverty. We found that while 20 per cent of the population was living below the official poverty line both in 1996 and 2001, only 9 per cent of the population was poor at both dates. We also found that when the poverty line was raised, the amount of households which could be considered chronically poor rose steadily, whereas the transitory component of poverty remained more or less stable. Analysis of the direct reasons for changes in household poverty status leads us to the conclusion that labour dynamics are far more relevant than demographic changes. Household heads who suffered health problems are significantly less likely to leave poverty. Household human and physical capital are also relevant, as well as the sector in which the household head works. Simulating this exercise using different poverty lines reveals that some variables are not robust to changes in the definition of poverty, while others which originally appeared to be insignificant become so for most other possible poverty lines.

Resumen:

Este artículo utiliza los datos de la Encuesta Nacional Socioeconómica de 1996-2001 para analizar las dinámicas de la pobreza en Chile, marcando una distinción entre la pobreza crónica y la transitoria. Encontramos que mientras el 20 por ciento de la población se encontraba viviendo por debajo de la línea oficial de pobreza tanto en 1996 como en 2001, sólo el 9 por ciento de la población era pobre en ambas fechas. También encontramos que cuando la línea de pobreza fue elevada, la cantidad de hogares que se podían considerar crónicamente pobres creció de forma consistente, mientras que el componente transitorio de la pobreza permaneció más o menos estable. El análisis de las causas directas de los cambios en el estatus de la pobreza de los hogares nos llevan a la conclusión de que las dinámicas laborales son mucho más relevantes que los cambios demográficos. Los jefes de hogares que sufren de problemas de salud son significativamente menos capaces de dejar la pobreza. El capital humano y físico de los hogares también es relevante, así como el sector en el que el jefe de los hogares trabaja. Al simular este ejercicio utilizando diferentes líneas de pobreza se revela que algunas variables no son resistentes al cambio de definición de la pobreza, mientras que otras que originalmente parecían insignificantes se fortalecieron para la mayoría de las otras líneas posibles de pobreza.

Palabras clave: Chile, dinámicas de pobreza, pobreza crónica, pobreza transitoria, simulación

Resumo:

Este artigo utiliza o banco de dados do Levantamento Socioeconômico Nacional para analizar dinâmicas da pobreza no Chile, distinguindo entre pobreza transitória e crônica. Descobrimos que enquanto 20% da população vivia abaixo da linha de pobreza oficial em 1996 e em 2001, somente 9% da população vivia em condições de pobreza em ambos as datas. Também percebemos que quando elevou-se a linha de pobreza, a quantia de lares que poderiam ser considerados cronicamente pobres subiu constantemente enquanto o componente transitório da pobreza permaneceu mais ou menos estável. Análises dos motivos diretos de mudanças no status de pobreza dos lares nos leva à conclusão que as dinâmicas do trabalho são muito mais relevantes do que as mudanças demográficas. Chefes de família que sofreram problemas de saúde têm muito menos chance de deixar a pobreza. Também relevante é o capital humano e físico, assim como o setor onde no qual o chefe de família trabalha. Simulando esse exercício utilizando diferentes linhas de pobreza demonstra que algumas variáveis não resistem a mudanças na definição de pobreza, enquanto outras que originalmente pareciam insignificantes tornam-se significativas perante a maioria das outras possíveis linhas de pobreza.

Palavras-chave: Chile, dinâmicas da pobreza, pobreza crônica, pobreza transitória, simulação.

Keywords

Chilepoverty dynamicschronic povertytransient povertysimulation
Type
Research Article
Information
Journal of Latin American Studies , Volume 40 , Issue 2 , May 2008 , pp. 251 - 273
Copyright
Copyright © Cambridge University Press 2008

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 Dante, Contreras, ‘Poverty and Inequality in a Rapid Growth Economy: Chile 1990–1996’, Journal of Development Studies, vol. 39, no. 3 (2003), pp. 181–200Google Scholar; David, Hojman, ‘Poverty and Inequality in Chile: Are Democratic Politics and Neoliberal Economics Good for You?’, Journal of Interamerican Studies and World Affairs, vol. 38, no. 2/3 (1996), pp. 73–96.Google Scholar

2 Details of the CASEN surveys can be found at the website of the Ministerio de Planificación at www.mideplan.cl/casen/index.html

3 For a review of the methods for measuring poverty in Chile, see Juan Carlos Feres, ‘Notas sobre la medición de la pobreza según el método del ingreso’, Revista de la CEPAL, no. 61 (1997), pp. 119–33.

4 This decomposition is of course dependent on the time frame, and also the number and frequency of the observations available in the datasets used for analysis. This limits the comparability of magnitudes of poverty between countries or studies and should be taken into account in any comparative studies.

5 In 2001 the first Unemployment Insurance (UI) system was created in Chile. Its main features are the following: individual saving accounts plus a common fund; a coverage of all private sector workers over the age of 18; compulsory enrolment for all workers starting a job after October 2002; funding by three actors, workers (0.6 percent of their wage), government (an annual contribution of US$10 million) and employers (2.4 per cent out of which 0.8 per cent goes towards the common fund).

6 Bob, Baulch and John, Hoddinott, ‘Economic Mobility and Poverty Dynamics in Developing Countries’, Journal of Development Studies, vol. 36, no. 6 (2000), pp. 1–24.Google Scholar A related group of literature is that concentrating overall income mobility. Gary Fields, Distribution and Development: A New Look at the Developing World (Cambridge, Mass: 2001), presents a thorough review of theory and empirical evidence in emerging economies.

7 Jyotsna Jalan and Martin Ravallion, ‘Determinants of Transient and Chronic Poverty: Evidence from Rural China’ (World Bank Policy Research Working Paper Series, no. 1936, 1998).

8 See Orazio Attanasio and Miguel Szekely, Portrait of the Poor: An Assets-Based Approach (Johns Hopkins University Press, 2001), and Julie Litchfield, ‘Update: Income Distribution and Poverty Measure for Chile, 1987–98’, in World Bank, Chile's High Growth Economy: Poverty and Income Distribution, 1987–1998 (Washington, 2002), pp. 42–72, for discussions using cross-section data on the role of household capital and poverty in Latin America and Chile respectively.

9 See the survey of empirical results in Baulch and Hoddinott, ‘Economic Mobility and Poverty Dynamics’.

10 Chronic Poverty Research Centre, The Chronic Poverty Report, 2004–05 (Manchester, 2005).

11 Dante, Contreras, Osvaldo, Larrañaga, Julie, Litchfield, Alberto, Valdés, ‘Poverty and Income Distribution in Chile, 1987–1998: New Evidence’, Cuadernos de Economía, vol. 38, no. 114 (2001), pp. 191–208.Google Scholar

12 Litchfield, ‘Update’.

13 David Bravo et al., ‘Chile: Trade Liberalization, Employment and Inequality’, in Rob Vos et al. (eds.), Economic Liberalization, Distribution and Poverty. Latin America in the 1990s (Aldershot, 2002).

14 Christopher Scott and Julie Litchfield, ‘Inequality, Mobility and the Determinants of Income among the Rural Poor in Chile, 1968–1986’ (London School of Economics, Development Economics Research Programme, Discussion Paper no. 53, 1994); Christopher, Scott, ‘Mixed Fortunes: A Study of Poverty Mobility among Small Farm Households in Chile, 1968–86’, Journal of Development Studies, vol. 36, no. 6 (2000), pp. 155–80.Google Scholar

15 The authors mention difficulties with the definition, coverage, and value of income, as well as the setting of the poverty line in both years. Because only 146 households are available, with observations almost twenty years apart, any conclusions must be tentative, especially those referring to poverty dynamics.

16 70 per cent of the poor in 1968 was still poor 28 years later and 65 per cent of those not poor in 1968 was still not poor in 1986: see Scott, ‘Mixed Fortunes’.

17 For a discussion on the dangers of making too strong a distinction for policy purposes, see Armando, Barrientos and David, Hulme, ‘Chronic Poverty and Social Protection: Introduction’, European Journal of Development Research, vol. 17, no. 1 (2005), pp. 1–7.Google Scholar

18 Measuring poverty solely in terms of monetary income is an important limitation and the literature is moving towards a multi-dimensional view of poverty. The lack of information (for example, data on consumption) limits the analysis in most research done in Chile. Our main assumption in mapping income poverty and poverty dynamics to consumption and utility is that households are not able or are limited in their capability of smoothing consumption, because they have limited access to capital markets and/or do not produce themselves. For more details and definitions on poverty measures in Chile, see the website of the Ministerio de Planificación y Cooperación (www.mideplan.cl).

19 Ingrid, Woolard and Stephen, Klasen, ‘Determinants of Income Mobility and Household Poverty Dynamics in South Africa’, Journal of Development Studies, vol. 41, no. 5 (2005), pp. 865–97.Google Scholar

20 Neil, McCulloch and Bob, Baulch, ‘Simulating the Impact of Policy Upon Chronic and Transitory Poverty in Rural Pakistan’, Journal of Development Studies, vol. 36, no. 6 (2000), pp. 100–30.Google Scholar

21 The classification of a household as rural or urban corresponds to the definition of the Instituto Nacional de Estadísticas (INE) following the census of 1992. A rural household is defined as belonging to a community with fewer than 1000 inhabitants or one where there are between 1000 and 2000 inhabitants with 50 per cent or less of the population classified as economically active.

22 Guillermo, Cruces and Quentin, T. Wodon, ‘Transient and Chronic Poverty in Turbulent Times: Argentina 1995–2002’, Economics Bulletin, vol. 9, no. 3 (2003), pp. 1–12Google Scholar, provide evidence that the sector in which the household head works is an important variable explaining poverty dynamics in Argentina.

23 The survey question asked is: ‘Who would the household go to for help if they had a serious economic problem?’ Those that answered that they had no one to whom they could turn were categorised as having no social capital.

24 These were defined as ages 0–5, 6–15, 16–65, and more than 65.

25 The CASEN dataset that was used did not provide information for any other type of shocks.

26 The country is divided in 15 administrative regions.

27 Many authors have insisted on the necessity of defining poverty as a multi-dimensional concept rather than relying on income or consumption expenditures per capita. Unfortunately there is no dataset in Chile that allows us to study poverty dynamics in a multidimensional manner. We assume that income per capita is a good measure of poverty. For a discussion on multidimensional poverty dynamics, see David Hulme, Karen Moore and Andrew Shepherd, ‘Chronic Poverty: Meanings and Analytic Frameworks’ (Chronic Poverty Research Centre, Working Paper no. 2, 2001).

28 This panel is also subject to aging of the sample since new young households are not included, and thus the overall representativeness of the sample in 2001 is limited by this fact. This is not fundamental for our conclusions since we are concerned with trajectories conditional on the information in the base year, where the sample is representative.

29 A survey of these can be found in Harold Alderman et al., ‘Attrition in Longitudinal Household Survey Data’ (International Food Policy Research Institute, Discussion Paper no. 96, 2001) for developing countries; see also Franco, Peracchi, ‘The European Community Household Panel: A Review’, Empirical Economics, vol. 27, no. 1 (2002), pp. 63–90.Google Scholar

30 This procedure led to the elimination of 4 per cent of the sample, and results that were consistent with standard literature.

31 The only variable that changes in significance is college education of the household head. This is no longer significant.

32 The average exchange rate in 2001 was 569 pesos to the US dollar.

33 It should be noted that these results are conditional on the time frame spanned by the panel, and as noted above, in this case it includes the Asian crisis. An in-depth study of how growth was distributed across the income distribution during this period in Chile is presented in Dante Contreras, Ryan Cooper and Christopher Neilson, ‘Crecimiento Pro Pobre en Chile’, El Trimestre Económico (forthcoming, 2008).

34 Woolard and Klasen, ‘Determinants of Income Mobility’.

35 For a review of the poverty dynamics related to high school versus higher technical education, see Jere Behrman, Nancy Birdsall and Miguel Székely, ‘Economic Policy and Wage Differentials in Latin America’ (Center for Global Development, Working Paper no. 29, 2003).

36 Cruces and Wodon, ‘Transient and Chronic Poverty’, find a similar result for the case of Argentina, where working in the public sector is associated with transient poverty.

37 See Annie Duflo, ‘Health Shocks and Economic Vulnerability in Rural India: Break the Vicious Circle’ (Centre for Micro Finance Research, Working Paper Series, 2005), for an interesting review of the extent and nature of health shocks in India.