Inequality, Welfare and Poverty: Theory and Measurement: Volume 9

Recent developments in the economic theory of justice recognize that an individual's income is a function of variables beyond and within the individual's control, called circumstances and effort, respectively. Inequality of opportunities refers to those income inequalities due exclusively to differential circumstances. Assuming that income comparisons across people with different circumstances are made according to Roemer's (1993, 1998) Pragmatic Theory of Responsibility, this interpretative paper discusses: (i) the scope of application of Peragine's (2000, 2002) approach to the partial ranking of income distributions in a number of static and dynamic scenarios; (ii) the use of complete indicators of inequality and social welfare to measure the inequality of opportunities along other types of income inequalities; and (iii) the connection between the measurement of the inequality of opportunities in a dynamic scenario and the existing theory of income mobility.

We establish a general relationship between the standard form of the individualistic social-welfare function and the “reduced-form” version that is expressed in terms of inequality and mean income. This shows the relationship between the property of monotonicity and the slope of the equity-efficiency trade-off. Particularly simple results are available for a large class of inequality measures that includes the Gini. These results do not require differentiability of the social welfare function.

The paper examines social welfare of and inequality within a finite homogeneous population consisting of identical individuals. These individuals form groups. All members of a group possess the same income; groups may differ in size and per-capita income. The framework is a generalization of the usual one to a situation in which the weight assigned to an income level can be arbitrary (but is, of course, fixed). It is in particular relevant for the measurement of welfare and inequality when households having different needs or size are investigated. Several classes of orderings which depend on the ranking of incomes are completely characterized.

This paper aims to clarify the similarities and differences between the concepts of bi-polarization and inequality by proposing an extended measure of bi-polarization, which is consistent with the second polarization curve.

The standard decomposition property of population subgroups for the Gini coefficient can be generalized to the extended Gini coefficients. Then, it is explicitly shown that the Wolfson bi-polarization index can be obtained by subtracting the within-groups from the between-groups Gini coefficients, computed for groups separated by the median value.

Moreover, we demonstrate the existence of a critical interval of the sensitivity parameter values (v) of the extended Gini coefficient, within which the second polarization curve can be consistently expressed as the subtraction of the within-groups inequality component from the between-groups inequality component. This critical interval is defined by v ϵ[2, 3]. This approach has the conceptual advantage of viewing inequality and polarization within the same framework.

When incorporating differences in household characteristics, the choice of equivalence scale can affect the ranking of income distributions. An alternative approach was pioneered by A. B. Atkinson and F. Bourguignon (G. R. Feiwel (Ed.), Arrow and the Foundation of the Theory of Economic Policy, Macmillan, New York, 1987), who derive a sequential Lorenz dominance criterion for comparing zistributions with an identical population structure. In order to make their approach applicable to international comparisons, we extend their criterion to the case of different marginal distributions of household types, and derive a sequential stochastic dominance criterion that highlights the importance of first order dominance of the marginal distribution of household characteristics for obtaining consistent rankings of income distributions. Comparisons of distributions are made using the Luxembourg Income Study database for a number of countries.

Dardanoni's (1993) results on mobility and social welfare in a simple Markov model of “pure” mobility are extended to rank distributions characterized by unequal growth. Partial orders of welfare states are derived and shown to be equivalent to rankings of generalized permanent income Lorenz curves. The results represent a natural extension to mobility of Shorrocks' (1983) well-known generalization of the Lorenz dominance theorem of Kolm (1969) andAtkinson (1970).

In this paper we review alternative measure of intergenerational mobility, emphasizing the distinction between absolute, relative and ordinal mobility. We then compare the performance of various mobility indices using real data. FromTreiman and Ganzeboom (1990) dataset we compare the degree of occupational and educational intergenerational (father-son) mobility in 16 countries in a single year (comprised between 1968 and 1982). From three Bank of Italy surveys (1993, 1995, 1998) we obtain a comparable measure of social prestige and we show that intergenerational mobility in Italy across regions or age cohort exhibits different trends according to different indicators. We suggest that ordinal relative and absolute measures provide divergent indications whenever we compare mobility data with markedly different marginal distributions.

The techniques of simple random sampling are seldom appropriate in the empirical analysis of income distributions. Various types of weighting schemes are usually required either from the point of view of welfare-economic considerations (the mapping of household/family distributions into individual distributions) or from the point of view of sample design. The different types of weights have different implications for the sampling distribution of estimators of welfare indices.

To take into account heterogeneity in a social welfare function, Ebert (1997) and Shorrocks (1995) show that the only consistent way of welfare measurement consists of either constructing an artificial distribution in which each household is weighted by the number of equivalent individuals, or weighting by the number of individuals in the household. Both approaches are not only mutually exclusive on axiomatic grounds, they are also in sharp contrast with many empirical applications where there is no weighting at all. Since ultimately, the choice is a normative one between axioms, and hence not easily envisaged, an empirical test of the sensitivity of welfare evaluations for the choice of the different weighting schemes might prove useful. In this paper we apply the different methods to administrative microdata of the 2000 PIT reform in Belgium, obtained from the microsimulation model SIRe of the Belgian Ministry of Finance. We find indeed sensitivity of our results with respect to the different weighting methods. In addition, using the number of equivalent individuals as weights to perform dominance analysis leads to fanciful results with respect to the choice of equivalence scales.

Subjective minimum income (MIQ) and minimum spending (MSQ) are the study focus. Basic Needs Module (1995) data from the U.S. Survey of Income and Program Participation are analyzed. A regression intersection approach is used to estimate household thresholds. MIQ thresholds are higher than MSQ thresholds. Both are higher than U.S. official poverty thresholds, and thresholds based on a National Academy of Sciences (NAS) methodology. Subjective threshold-based equivalence scales imply greater economies of scale than those in the other two measures but are similar to behavioral scales. This finding suggests that families make trade-offs to meet their minimum needs.

The Social Welfare Function (SWF) is a decision rule to rank alternative social states in a complete fashion in terms of social welfare. This paper questions the philosophy of Paretian Principle as a desirable property of the SWF. It shows that it is possible to generalize the widely used Sen SWF, which can be non-Paretian under special circumstances. Also, it demonstrates the disaggregation method of this SWF by components of income using the Gini decomposition process. The method is applied to Australian Household Expenditure Survey data to estimate the trend of welfare of total income and its components in Australia from 1984 to 1993–1994.

This paper examines the changes in the level of social welfare in New Zealand over the period 1984 to 1998 in the context of the country's economic reform process since the early 1980s. The earlier part of this period was also characterized by a largely policy-induced economic recession in New Zealand. In this paper, we make an attempt to identify the sections of the population that became better off in terms of real income, and those that became worse off during the period chosen. In addition, we examine the changes to the overall level of social welfare. The methods used are both ordinal and cardinal. The ordinal method is based on the criterion of generalized Lorenz dominance, and the cardinal method is based on a social evaluation function that provides complete ordering of all possible social states. The social welfare changes, derived with the help of the cardinal method, and measured in terms of real income, are then attributed to the twin influences of mean income changes and changes in measured inequality. In addition to showing up the dramatic increase in the Gini coefficient of income inequality overall, the results also track the changes in real income of the different income groups over time, and quantify how these changes, coupled with the increased inequality, affected the well-being of New Zealanders over a period of extensive economic reform. The study is based on unit record data from four Household Surveys conducted by Statistics New Zealand in the years 1983/84, 1991/1992, 1995/1996 and 1997/1998.

In this paper we investigate the urban/rural dimension of poverty in developed countries. We provide original estimates for Italy, we gather published statistics for France and the United States, and we produce novel cross-country estimates from the LIS database. We show that the size of urban poverty depends on where the boundaries of metropolitan districts are drawn and we observe that overlooking geographical differences in the cost of living is a particularly relevant hypothesis. We find that in France and the United States post-war economic growth and urbanisation were accompanied by a substantial reduction of the poverty risk for the rural population, while poverty rates improved less, or even sometimes deteriorated, for the urban population. The lack of a standard definition of urban/rural area precludes a rigorous comparative study. Our results indicate, however, that only in few countries (Denmark, the United Kingdom and the United States) the greatest poverty rates are found in central cities, while in all other developed countries poor persons are still relatively more frequent in rural areas. This pattern is stronger in the four non-developed economies examined here.

Using regional incomes as the reference group, disposable income poverty rates are computed for the two most recent waves of Luxembourg Income Study (LIS) data available for the following countries: Australia, Canada, Finland, France, Germany, Italy, the United Kingdom, and the United States. In addition, we aggregate the regions of the five western European countries we examine so that we can better assess the effectiveness of Europe's efforts to reduce the economic gaps between regions. We find that the countries we examine have patterns of regional poverty that help us better understand the national aggregate measures, and we are able identify areas where antipoverty efforts should be made a priority.