This section presents four sets of poverty data for the LAC region. While these data do not always agree, if interpreted cautiously, they present a useful picture of poverty and extreme poverty in the LAC region.
Data on poverty and extreme poverty in the LAC region should be interpreted with a great deal of caution. Estimates of the incidence of poverty generally are not comparable among countries. Even for the same country estimates are sometimes not comparable over time, and annual changes in particular are often suspect. Several efforts have been made to adjust country-level data in order to obtain internationally comparable figures. Even in these cases, the assumptions required to make the adjustments should make readers wary of the outcomes. Following are the principal reasons for the lack of comparability:
1) The concept of income used to determine poverty levels differs among countries. Sometimes excluded from the income base is income derived from remittances--not only money sent home by family members living abroad but also intra-country transfers, especially from urban residents to poorer relatives living in rural areas. Also sometimes excluded, or at least underestimated, is "imputed" income, particularly the value of agricultural goods consumed by the households that produce them, and the rental value of housing units owned (and often constructed) by the households occupying them. For instances in which these and other types of income are excluded or underreported, the incidence of poverty tends to be overestimated.
2) Underreported income: adjustments to take into account underreported income (if made at all) yield uncertain outcomes, partly because patterns of underreporting by decile differ from country to country.
3) Survey coverage is sometimes restricted to major urban areas. Rural areas tend to have higher poverty rates than urban areas, even after adjusting for the lower cost of living in rural areas. Moreover, rural-urban differentials vary across countries.
4) Some poverty estimates are based on household incomes, while others are based on individual incomes. As such, calculations can differ depending on which type of income is analyzed.
5) Definitions of what constitutes poverty vary among countries.
6) The amount of income needed to purchase a "food basket" that meets minimum nutritional standards may be overestimated if the composition of the food basket is based on an ideal nutritional standard.
7) The quality of the design, implementation, and analysis of household surveys varies among countries.
8) In some countries, annual surveys have not always been conducted during the same month of the year. Since agricultural incomes in particular exhibit marked seasonal variations, many people in most LAC countries derive much of their income from agriculture, lack of uniformity in the survey months introduces distortions into the annual series.
9) Methodological changes in the design and implementation of household surveys can make annual data for a given country non-comparable over time.
Another problem with the data available from standard published sources (mainly documents of the international financial institutions) is that they are current. This makes it particularly difficult to determine the extent to which economic recovery in the 1990s may have reduced the incidence of poverty and extreme poverty. Considerable evidence suggests that policy reform-induced economic recovery during the current decade has been associated with only a modest reduction of the incidence of poverty in most countries. Some of the more current data sets, however, show greater reductions in poverty rates.
Table 3.1 ¼ ¼ .
Average Poverty in Latin America
Source Juan Luis Londoño and Miguel Székely, "Persistent Poverty and Excess Inequality in Latin America, 1970-1995," Working Paper 357, Inter-American Development Bank (IDB), Washington, DC October 1997.
Definitions Headcount ratio is the percentage of the population living in total or extreme poverty. The total poverty line is set at $2 per day per person in 1985 PPP (purchasing power parity) dollars; the extreme poverty line is $1 per day per person in 1985 PPP dollars. Poverty gap is defined as the average shortfall of the poor with respect to the poverty line, multiplied by the head count ratio. The Foster-Greer-Thorbecke Index (2) is a variation of the poverty gap indicator. By squaring the poverty gap, this version of the FGT index assigns greater weights to lower incomes. An increase in the FGT index indicates greater poverty.
Notes This study, using what the authors judged to be good-quality time-series data for 13 LAC, sought to calculate the incidence of poverty for the LAC region, and for individual countries, by applying a common methodology to the data. The 13 countries for which comparable data could be calculated are: The Bahamas, Brazil, Chile, Colombia, Costa Rica, the Dominican Republic, Guatemala, Honduras, Jamaica, Mexico, Panama, Peru, and Venezuela. These countries comprise 83% of the region's total population. The authors readily admit that "complete cross-national comparability is not attainable". Furthermore, in moving toward greater comparability, simplifying assumptions have been made that may affect individual-country data in different ways. As such, the results need to be interpreted cautiously and in conjunction with other poverty indicators. The study contains a detailed methodological overview.
The data for this study are somewhat more current than the ECLAC data, and the average number of annual observations per country is larger (approximately 7). Nevertheless, for six countries the most recent data are only for 1993 or earlier years. Only for Brazil, Costa Rica, and Venezuela are data available on a (nearly) annual basis since 1980. For the LAC region, this study estimates that the incidence of poverty fell modestly from 35.2% to 33.1%, while the absolute numbers of those living in poverty and extreme poverty rose by 4.6 million and 1.1 million, respectively.
Table 3.2 ¼ ¼ .
Incidence of Poverty, 1978-1995
Source Juan Luis Londoño and Miguel Székely, "Persistent Poverty and Excess Inequality in Latin America, 1970-1995," Working Paper 357, Inter-American Development Bank (IDB), Washington, DC October 1997.
Definitions The poverty line represents a level of household income just adequate to meet the basic nutritional and non-food needs of all its members. The extreme poverty line (or indigence) represents a level of household income that can not adequately cover the nutritional needs of all its members.
Notes Please refer to Table 3.1 for technical notes.
Table 3.3 ¼ ¼ .
Incidence of Poverty
Source Social Panorama of Latin America 1997, Economic Commission for Latin America and the Caribbean (ECLAC), United Nations, Santiago, Chile, 1997.
Definitions The poverty line represents the level of income required by a household in order to meet the basic needs of all its members. The poverty lines for each country were determined on the basis of the estimated cost of a basket of staple foods taking into account existing eating habits, actual availability of food in the country and relative prices which meets the nutritional needs of the population. The value of this basket was then added to the total estimated sum required by households to meet the basic non-food needs of their members. The indigence line (or extreme poverty) represents a level of household income that can not adequately cover the nutritional needs of all its members. The poverty line in urban areas works out to be twice the indigence line, while in rural areas, it is 75% higher than the respective basic food budget. In calculating indigence lines, the differences between food prices in metropolitan areas and those in the remaining urban and rural areas were taken into account.
Notes ECLAC's poverty estimates are based on special tabulations of household survey data. The percentages of households and persons classified as poor and as indigent were obtained by comparing the per capita monthly value of the respective lines with the total income of each household, also expressed on a per capita basis.
The ECLAC data set covers 14 countries, including the five most populous (Brazil, Mexico, Colombia, Argentina, and Peru), and provides an average of 4 annual observations per country. ECLAC also calculates region-wide estimates of poverty and indigence covering a total of 19 countries. The major shortcoming of the ECLAC data is their age. Except for Chile, the most recent figures are for 1994. In some cases, the data cover urban areas only.
For the LAC region, the data reveal a modest reduction in the incidence of poverty from 41% in 1990 to 39% in 1994. The incidence of extreme poverty declined from 18% to 17% over the same period. The next edition of ECLAC's Social Panorama is likely to show a greater reduction in poverty in individual countries and for the region as a whole as more recent figures are taken into account.
Table 3.4 ¼ ¼ .
Incidence of Poverty
Source Samuel A. Morley, "Poverty during Recovery and Reform in Latin America: 1985-1995" Inter-American Development Bank (IDB), Washington, DC, May 1997.
Definition The poverty line represents a level of household income just adequate to meet the basic nutritional and non-food needs of all its members.
Notes The poverty data contained in Table 3.4 appear to be the most comprehensive and current figures available. They cover 17 countries, representing about 90% of the LAC population. For three countries, data are provided through 1996, and for another seven the last survey year is 1995. Only for the Dominican Republic (1992) and Nicaragua (1993) is the last survey year prior to 1994.
Of the 17 countries in this data set, 12 show reductions, often significant, in the incidence of poverty since the late 1980s. These 12 countries include Argentina and Peru, for which the figures are for urban areas only. However, national-level estimates also provided for these countries show increases in poverty rates. The initial survey years for these countries was in the mid 1980s, before the initiation of hyperinflation and the subsequent adoption of major economic-policy reforms. The data for Argentina and Peru show that, despite recent improvements, the incidence of poverty in these countries remains higher than it was in the mid 1980s.
Although no regional poverty estimates are provided, the source concludes that it is highly probable that the absolute numbers of poor in these seventeen countries has declined, despite the methodological dilemma of summing poverty estimates from different countries with different poverty lines and periods of coverage. Given the growth in population, this implies that the incidence of poverty for the region as a whole must have fallen by at least ten percent.
Table 3.5 ¼ ¼ .
Chile
Source IMF Staff Country Report No. 98/26, International Monetary Fund (IMF), Washington, DC, April 1998. Primary source: Caracterización Socioeconómica Nacional (CASEN), Ministry of Cooperation and Planning (MIDEPLAN), Santiago, Chile, 1997.
Definitions The relative poverty line represents a level of household income just adequate to meet the basic nutritional and non-food needs of all its members. The indigence line (or extreme poverty) represents a level of household income that can not adequately cover the nutritional needs of all its members. Total poverty is the sum of relative poverty and indigence.
Notes Although the Londoño and Székely study (Tables 3.1 and 3.2) used poverty data on Chile from a different source, the Encuesta Nacional de Empleo (ENE), they consider the data set employed in this study to be of good quality.
Table 3.6 ¼ ¼ .
Costa Rica
Source Encuesta de Hogares de Propositos Múltiples, Dirección General de Estadística y Censos (DGEC), online database, San Jose, Costa Rica, 1998.
Definitions The relative poverty line represents a level of household income just adequate to meet the basic nutritional and non-food needs of all its members. The indigence line (or extreme poverty) represents a level of household income that can not adequately cover the nutritional needs of all its members. Total poverty is the sum of relative poverty and indigence.
Notes Many poverty estimates have been made for Costa Rica since 1980--most on the basis of annual household surveys. While these estimates do not always agree, both with respect to the level of poverty and to short-term trends, the survey methodology has been reasonably consistent over time, and the overall quality of the data is among the highest in the LAC region. Although important methodological changes were introduced in 1987, and again in 1996, the DGEC has revised the data from earlier years to reflect these modifications.
Table 3.7 ¼ ¼ .
El Salvador
Source Ministerio de Planificación y Coordinación del Desarrollo Económico y Social (MIPLAN) and, subsequently, Ministerio de Economía, Encuestas de Hogares de Propositos Múltiples.
Definitions The indigence line (or extreme poverty) represents a level of household income that can not adequately cover the nutritional needs of all its members. Relative poverty indicates a level household income between one and two times the indigence level. Total poverty is the sum of relative poverty and the indigence level.
Notes Regular multipurpose household surveys were begun in 1988. The first four covered urban areas only; subsequent surveys have been national in scope. The quality of the data reported in Table 3.7 is not ideal due to incomplete coverage of income (especially for the top and bottom income deciles), differences in the months during which the surveys were conducted, and errors in expanding the sample. Furthermore, the implementation of the annual surveys has been uneven (outside experts consider the 1990 survey to have been especially faulty). However, data quality has improved since 1990.
Table 3.8 ¼ ¼ .
Honduras
Source Annual multipurpose household surveys, Government of Honduras, 1998.
Definitions The relative poverty line represents a level of household income just adequate to meet the basic nutritional and non-food needs of all its members. The indigence line (or extreme poverty) represents a level of household income that can not adequately cover the nutritional needs of all its members. Total poverty is the sum of relative poverty and indigence.
Notes The methodology for household surveys in Honduras has been revised on several occasions. Data for 1991-94 have been recalculated in accordance with the 1995 revision, but for prior years the data are not sufficiently comparable to report in Table 3.8. The quality of the data has improved over time, but much room for improvement remains. Overall levels of poverty tend to be overestimated, partly due to incomplete coverage of income, as well as a high standard for minimum nutritional requirements.
Table 3.9 ¼
Peru
Source Encuestas Nacionales de Niveles de Vida (ENNIV) and Encuesta Nacional de Hogares, Instituto Cuanto, 1998.
Definition The poverty line represents a level of household income just adequate to meet the basic nutritional and non-food needs of all its members.
Notes The rural poverty figures for 1996 are from the Encuesta Nacional de Hogares; all other figures are from the ENNIV. As such, the rural poverty figures for 1996 are not comparable to those for earlier years.
Table 3.10 .
Distribution of Income or Consumption
Source World Development Indicators 1998 (CD-ROM version), World Bank, Washington, DC, 1998.
Definitions Gini index measures the extent to which the distribution of income (or, in some cases, consumption expenditures) among individuals or households within an economy deviates from a perfectly equal distribution. A Lorenz curve plots the cumulative percentages of total income received against the cumulative number of recipients, starting with the poorest individual or household.
The Gini index measures the area between the Lorenz curve and a hypothetical line of absolute equality, expressed as a percentage of the maximum area under the line. Thus a Gini index of zero represents perfect equality while an index of 100 implies perfect inequality. Percentage share of income or consumption is the share that accrues to subgroups of population indicated by deciles or quintiles. Percentage shares by quintiles may not add up to 100 because of rounding.
Notes a) refers to expenditure shares by percentiles of population. b) ranked by per capita expenditure. c) refers to income shares by percentiles of population. d) ranked by per capita income.
Inequality in the distribution of income is reflected in the percentage share of either income or consumption accruing to segments of the population ranked by income or consumption levels. The Gini index provides a convenient summary measure of the degree of inequality.
Income and consumption data come from nationally representative household surveys. As these surveys differ in method and in the type of data collected, the distribution indicators are not strictly comparable across countries.
[an error occurred while processing this directive]