Income Distribution Data Research Articles

Global and regional inequality in the distribution of income: Estimation with limited and incomplete data

The paper examines the nature and extent of global and regional inequality using the most recent country level data on inequality drawn from World Bank studies, and real per capita income from the Penn World Tables, for the period 1980–1990. The methodology employed in the paper is based on a mixture of parametric and non-parametric approaches to inequality measurement. It is designed to handle the limited and incomplete nature of income distribution data from different countries. Empirical results show a very high degree of global inequality, but with some evidence of catch-up and convergence between regions.

Should the Kuznets effect be relied on to induce equalizing growth: Evidence from post-1950 development

The large body of cross-sectional statistical research assessing 1. (a) Kuznets inverted-U for less-developed countries (LDCs), 2. (b) a turning point, is analyzed. An alternative methodology is presented. Those countries which started in 1950 well below the “turning point” in RGDP and which reached an RGDP well above the “turning point” by 1980 are utilized as key cases. Income distribution data and assessments from individual country experts are combined with historical-comparative microanalyses for each key case — Brazil, Costa Rica, Greece, Japan, Malaysia, South Korea, Taiwan, and Turkey — to construct patterns of income distribution during this crucial growth period. It is demonstrated that the jury is still very much out on the Kuznets inverted-U for LDCs, and the notion of a turning point should be rejected. Copyright © 1997 Elsevier Science Ltd

World-Economic Trends in the Distribution of Income, 1965-1992

This article assesses trends in world income inequality between 1965 and 1992. The study first decomposes Theil's T to show that between-country inequality is the most significant component of overall world income inequality, which facilitates fruther research, as between-country longitudinal data are more easily available than national income distribution data. Next, the study uses a larger set of nations and finds a signigicant increase in world income inequality as measured by both Gini and Theil coefficients, particularly during the 1980s. The findings are robust even after controlling for differential rates of population growth or using alternative sources of data. By indentifying these trends, the article is able to explain past discrepancies and recent shifts in the relevant empirical and theoretical literature.

Growth, income distribution, and democracy: What the data say

This paper investigates the relationship between income distribution, democratic institutions, and growth. It does so by addressing three main issues: the properties and reliability of the income distribution data, the robustness of the reduced form relationships between income distribution and growth estimated so far, and the specific channels through which income distribution affects growth. The main conclusion in this regard is that there is strong empirical support for two types of explanations, linking income distribution to sociopolitical instability and to the education/fertility decision. A third channel, based on the interplay of borrowing constraints and investment in human capital, also seems to receive some support by the data, although it is probably the hardest to test with the existing data. By contrast, there appears to be less empirical support for explanations based on the effects of income distribution on fiscal policy.

The Kuznets curve: A reassessment

This paper revisits the Kuznets hypothesis. An extensive data set containing the most recent estimates of income distribution is utilized to test the Kuznets hypothesis. In addition, the data comparability problem, with respect to income distribution data, when conducting econometric analysis is addressed. The results indicate that the data comparability problem is not severe enough to nullify the estimates obtained from pooled regressions. The Kuznets hypothesis holds for this data set.

Poverty and Inequality in Latin America: A Neostructural Perspective

Two facts stand out from even the most cursory inspection of income distribution data in Latin America. First, income is highly concentrated, making the region one of the most unequal in the world. Second, despite the fact that the region has now reached an intermediate position among the world's nations in terms of per capita income (some US$2,000 per year on average), a level which, by historical standards, would presuppose a more equitable distribution of income, this has not happened. In fact, the distribution of income in Latin America has become, if anything, more highly skewed than ever. The question then is: why is this so? The explanation forms the heart of this paper. Precisely because neither of these statements would hold if the countries to which they apply were either “average” or “well-behaved,” the explanation thatfollows assumesa “neostructuralist perspective” (Lustig, 1991) — i.e., one that ascribes causality to a number of regional charactristics related to its history and the interaction among its factor, cultural, and institutional endowments.

Testing for unit roots with income distribution data

In this paper I test the unit root hypothesis for US log GNP using the information available in income distribution data. The percentile data of an income distribution are shown to follow the same autoregressive pattern as does mean income. Under the null hypothesis of a unit root log GNP is cointegrated with the percentile data. A sequence of augmented HEGY-Tests, however, presents strong evidence against the unit root hypothesis for the distribution data and hence for log GNP. Using a full information estimation procedure for the percentiles under the alternative yields an estimate of the autoregressive coefficient which is in principle testable by an approximate Dickey-Hasza-Fuller test. The appropriate critical values are found by bootstrap methods. Again, inference is clearly unfavorable for the unit root hypothesis.

Adjustment methods for non-response and their application to Finnish income data

Non-response is an increasing rather than a decreasing problem in surveys, but there are also many other deficiencies in them. The non-response rates in Finnish income surveys have been 15-30070 on average. Since these values do not go missing randomly, obvious biases exist in estimates. This paper discusses their effects and the opportunities available for adjusting for them presenting statistical methods developed or applied in Finland. Applications to income distribution data are demonstrated.

Multinational Corporations in Less Developed Countries: An Alternative Strategy

In a world characterized by increasing transnationalization of business, the multinational corporations (MNCs) are destined to play a central role in global economic integration. By focusing on the dynamic trilateral relationship between MNCs, home and host countries, this study critically explores certain aspects of MNC operations in less developed countries (LDCs) and proposes an alternative strategy aimed at the achievement of long-term stability in MNC-LDC relations. The dimensions of MNCs' impact are documented through case studies based on income distribution data, the analysis of political events and consumption patterns – representing the economic, political and cultural aspects of MNC influences respectively. The proposed alternative strategy is designed to ameliorate tensions between MNCs and LDCs, based on enlightened self-interest, sound business practices, cultural sensitivity and social responsibility.

A PROTOTYPE MICRO‐MACRO LINK FOR THE CANADIAN HOUSEHOLD SECTOR

This paper seeks to forge a link between Canadian macro and micro data relating to the household sector. The analysis is in three parts. The first part begins with National Accounts data on the personal sector. These data are adjusted to remove transactions relating to non‐biological persons, so that the result is income and expenditure for the household sub‐sector. The second part starts with the annual household survey used to collect income distribution data. These survey data are augmented in various ways to account for under‐reporting and to add information from other micro data sets particularly the periodic survey of household expenditure patterns and a sample of individual income tax returns. The result is a comprehensive, albeit partially synthetic, household micro data set. In the final part of the paper these two largely independent data sets are compared, and the general quality of the results is discussed.

PARETO EXTRAPOLATION USING GROUPED INCOME DATA*

ABSTRACT. The distribution of above‐median income groups can be described accurately by OLS regression of the Pareto distribution. The excellent fit of log‐linear regression has led some researchers to conclude that extrapolated income measures in the open end of the distribution are reliable. The analysis of income distribution data for metropolitan areas reveals: (1) that OLS tends to consistently overestimate the number of families in the tail, and (2) that the overestimation can have serious implications for hypothesis testing.

Liberalization and inequality in Sri Lanka: A comment

Paul Glewwe has recently argued against the widely held view that the liberalizing policy reform in Sri Lanka in the late 1970s increased inequality. We contend here that Glewwe has not convincingly discredited the household income distribution data, that he has misinterpreted the expenditure distributions, and that he has not considered the welfare distributional effects of the relative price changes induced by the policy reforms. Furthermore, we argue that a sound basis remains for the contention that the policy reforms of the late 1970s increased inequality.

An empirical study of selection and estimation of alternative models of the Lorenz curve

Five different models of the Lorenz curve are estimated using the Australian grouped income distribution data. Performances of these models are evaluated in regard to their adequacy in modelling the Lorenz curve based on empirical results.

On Measuring Intergroup Inequality

This article develops an alternative measure for use in comparing two income distributions. The measure builds on the argument that under certain conditions inequality and dissimilarity are not the same thing, and this difference can be used to develop a measure of “crossover.” As far as we know, no analyst has developed a statistical measure that can determine the size of crossover (overlap) between two cumulative distribution functions. The measure is then applied to racial income distribution data in the United States. The results show some of the “styles” of crossover that vary from bottom-up to top-down patterns. We argue that the overlap information generated by this measure can provide insight into who benefits by trends in intergroup convergence that is not gained when using existing intergroup inequality measures.

Inequality and Poverty in Malaysia: Measurement and Decomposition

This report gives a perspective on Malaysia, by tracing the development and importance of ethnic pluralism in the country. This pluralism has led the government to show special concern for racial income, distribution and, more generally, for racial economic disparities. The income distribution data used in this study were collected through the Post-Enumeration Survey (PES) of 1970. A comparison shows that the PES is the best source of income data to date on household and individual incomes in Malaysia. The broad features of the PES household income distribution show overall inequality in Malaysia to be fairly high. The individual income distributions show large inequalities within the racial groups. These suggest that racial income disparities may be only part of a much wider problem of income inequality in the country. A profile of poverty in Malaysia is constructed, which identifies the poor in terms of socioeconomic variables such as race, location, employment status, occupation, and education. Such information is useful not only in understanding better the correlates and circumstances of poverty, but also in identifying areas of government intervention for the redress of poverty. The reports concludes if there are no data on the distribution of physical wealth among individuals in Malaysia, the breakdown of the personal income distribution by employment status does shed some light on the distribution of capital assets.

GENERATING INTERNATIONALLY COMPARABLE INCOME DISTRIBUTION DATA: EVIDENCE FROM THE FEDERAL REPUBLIC OF GERMANY (1974), MEXICO (1968) AND THE UNITED KINGDOM (1979).

In this article an attempt is made to generate internationally comparable income distribution data for the Federal Republic of Germany (1974), Mexico (1968) and the United Kingdom (1979). To that end, the same income concept and income unit were adopted for each country, i.e. respectively household available income and the household. Moreover, incomes from various sources were adjusted for inconsistency with National Accounts according to Altimir's methodology. The paper finds that the distribution of persons by household income per equivalent unit is probably the best way of looking at the distribution of economic welfare. It further demonstrates that the distribution of persons by household available income per capita is much closer to this ‘ideal’ distribution than the distribution of households by household available income. Finally, the paper discusses some of the problems arising from the fact that one normally works with grouped data. It is found that in the case of the three countries under study, grouping is likely to have had only a small impact on the results.

Decomposition of inequality measures and a comparative analysis

The main objectives of the paper are to decompose Atkinson's measure into within-group and between-group inequality and to compare it with decompositions of inequality measures, like the Gini coefficient, variance of logarithms, Theil's entropy index and the square of the coefficient of variation. Atkinson's inequality index is given an interpretation of mean order β and then it is decomposed for population subgroups. Various other measures are considered for decompositions and a variety of decomposition schemes for Gini coefficient and other measures in the literature is reviewed and commented upon. Empirical analysis using U.S.A. and U.K. data are performed on the basis of income distribution data classified by family size. For various decomposition schemes, contribution to between size-group and within size-group inequalities are compared in percentage terms. Comparisons are also made over time for the years 1964 and 1974 both for U.S.A. and U.K. and the result indicates that within size-group income inequality has declined while between size group and total income inequality shows an increase. Three important properties of Atkinson's inequality index are proved in the appendix.

Optimal Grouping of Income Distribution Data

Abstract We consider the problem of grouping income distribution data into a given number of groups such that the concealed income differences due to grouping are minimized. When income differences are measured by Gini's pairwise differences, this means minimizing the area between the grouped and ungrouped Lorenz curves. In this case, the necessary condition for optimal grouping is that each group limit be equal to the average income in its two adjacent groups. An iterative procedure for computation and applications including fractile groupings are discussed. Alternative optimal groupings based on other measures of income differences are also considered.

Optimal Grouping of Income Distribution Data

Abstract We consider the problem of grouping income distribution data into a given number of groups such that the concealed income differences due to grouping are minimized. When income differences are measured by Gini's pairwise differences, this means minimizing the area between the grouped and ungrouped Lorenz curves. In this case, the necessary condition for optimal grouping is that each group limit be equal to the average income in its two adjacent groups. An iterative procedure for computation and applications including fractile groupings are discussed. Alternative optimal groupings based on other measures of income differences are also considered.

Spatial differences in poverty: The case of Peru

This paper examines the implications of spatial differences in living costs for measures of poverty. Taking Peru as a case study, spatial price indices are used to calculate location-specific poverty lines, based on the local cost of a basket of basic needs. Applied to income distribution data, these poverty lines yield measures of the extent and location of poverty. The paper shows that these measures are significantly different from, and much more accurate than those obtained using a single countrywide poverty line; the latter can give a misleading picture of poverty by over- stating spatial differences, in general overestimating rural and underestimating urban poverty.