EQUIVALENCE SCALES, WELL-BEING, INEQUALITY, AND POVERTY: SENSITIVITY ESTIMATES ACROSS TEN COUNTRIES USING THE LUXEMBOURG INCOME STUDY (LIS) DATABASE

The Luxembourg Income Study (LIS) database on which this article is based offers researchers exciting new possibilities for international comparisons based on household income microdata. Among the choices the LIS microdata allows a researcher, e.g. income definition, income accounting unit, etc., is the choice of family equivalence scale, a method for estimating economic well-being by adjusting income for measurable differences in need.
The range of potential equivalence scales that can and are being used in the ten LIS countries and elsewhere to adjust incomes for size and related differences in need span a wide spectrum. The purpose of this paper is to review the available equivalence scales and to test the sensitivity of various income inequality and poverty measures to choice of equivalence scale using the LIS database. The results of our analysis indicate that choice of equivalence scale can sometimes systematically affect absolute and relative levels of poverty; and inequality and therefore rankings of countries (or population subgroups within countries). Because of these sensitivities, one must carefully consider summary statements and policy implications derived from cross-national comparisons of poverty and/or inequality.     

INTERNATIONAL COMPARISONS OF EARNINGS INEQUALITY FOR MEN IN THE 1980s

In this paper we present a comparative analysis of earnings inequality during the 1980s among prime age men who headed households and worked year-round, full-time from five industrialized countries-Canada, Sweden, Australia, West Germany, and the United States. The data were obtained from the Luxembourg Income Study (LIS) database, a multinational collection of microdata sets from various countries which have been assembled for the primary purpose of making cross-national comparisons of economic and social well-being. The results of the comparison indicated that during the mid-1980s, the United States had the most unequal distribution of earnings and Sweden the least unequal. Between the early 1980s and mid-1980s, however, the earnings distributions in all five countries showed evidence of becoming more unequal, especially in the United States, Canada, and Sweden.     

Factor Components of Inequality: A Cross‐Country Study

This paper uses data from the Luxembourg Income Study to examine some of the forces that have driven changes in household income inequality over the last three decades of the twentieth century. We decompose inequality for six countries (Canada, Germany, Norway, Sweden, the U.K., and the U.S.) into the three sources of market income (earnings, property income, and income from self‐employment) and taxes and transfers. Our findings indicate that although changes in the distribution of earnings are an important force behind recent trends, they are not the only one. Greater earnings dispersion has in some cases been accompanied by a reduction in the share of earnings which dampened its impact on overall household income inequality. In some countries the contribution of self‐employment income to inequality has been on the rise, while in others, increases in inequality in capital income account for a substantial fraction of the observed distributional changes.     

INTERNATIONAL COMPARISONS OF WEALTH INEQUALITY

This study presents reasonably comparable estimates of the size distribution of household or personal wealth for eight OECD countries—Australia, Canada, France, Germany, Japan, Sweden, the United Kingdom, and the United States. In the mid-1980s, the U.S. ranked as the most unequal and Japan the least, while the other six countries had roughly comparable levels of wealth inequality. Moreover, while wealth inequality rose sharply in the U.S. during the 1980s, it increased modestly in Sweden and showed little change or a slight decline in Canada, France, and the U.K. A comparison of time trends for the U.K. and the U.S. suggests that the relatively high wealth inequality in the U.S. in the 1980s represents a marked turnaround from the 1950s, when the U.S. was considerably more equal in terms of wealth ownership than the U.K. Comparative results for the two countries hold for both conventional (marketable) wealth and for augmented wealth, which includes a valuation of public and private pension wealth.     

THE EFFECT ON POVERTY STATUS IN ISRAEL OF CONSIDERING WEALTH AND VARIABILITY OF INCOME

This paper assesses the effects of including wealth and the variability of income on the incidence of poverty and the degree of income inequality in Israel. A special survey, which includes data on the wealth and income of a national sample of Israeli families in 1963–64 and 1964–65, allows us to go beyond measures based on current income alone.
The first section reviews earlier studies of poverty in Israel. The next section looks at poverty and inequality in terms of current income, current wealth, and a combined measure of income and wealth. The combined measure is the Hansen-Weisbrod measure (HW), which equals income plus the annuity value of wealth, assuming all wealth is just consumed at the time of death. It is interesting that, in spite of the much higher wealth inequality than income inequality, the HW measure was slightly more equally distributed than income. This result occurred because the annuity component made up a low share of the total HW measure and the correlation between income and wealth was well under 1. Although overall inequality and poverty were similar for income and HW measures, the incidence of poverty by subgroup depended on the measure used.
The final section presents a dynamic view of poverty and inequality. Year-to-year changes in poverty were substantial. Because of the use of a relative poverty concept and the rise in real incomes, the real income poverty line rose by 15 percent between 1963 and 1964. Still, of those in income poverty in 1963, 37 percent managed to escape poverty in 1964. The paper shows how the degree to which poverty was stable or transitory varied substantially by age and country of origin.     

INCOME INEQUALITY IN AUSTRALIA AND NEW ZEALAND: INTERNATIONAL COMPARISONS AND RECENT TRENDS

In this paper we present results on the distribution of income in Australia and New Zealand that can be compared with those for a range of other advanced countries. The framework of analysis, concepts and definitions used have been developed as part of the Luxembourg Income Study (LIS). Using data for the early 1980s, the results indicate that the income distributions in Australia and New Zealand are not, as previous research has suggested, more equal than those in other countries. Neither country has an equivalent net family income inequality ranking in the top half of the eight countries studied. Further analysis indicates increasing inequality in Australia in the first half of the 1980s and, on the basis of some indicators, in New Zealand also. The paper does not investigate the causes of these increases in inequality, although the results indicate that the rise in property income has been a factor behind them.     

Income Inequality and Macroeconomic Volatility: An Empirical Investigation

We explore the impact of macroeconomic volatility on the distribution of income. Using a cross‐section of developed and developing countries, we find that greater output volatility, defined as the standard deviation of the rate of output growth, is associated with a higher Gini coefficient and income share of the top quintile. The coefficients suggest that a strong effect on inequality resulting from a reduction in volatility: the Gini coefficient of a country like Chile would fall by 6 points if it were to reduce its volatility to the same level as Sweden or Norway. Our results seem not to be driven by the high‐inequality/high‐volatility Latin American countries.     

Education and Income Inequality: New Evidence From Cross-Country Data

This paper presents empirical evidence on how education is related to income distribution in a panel data set covering a broad range of countries for the period between 1960 and 1990. The findings indicate that educational factors—higher educational attainment and more equal distribution of education —play a significant role in making income distribution more equal. The results also confirm the Kuznets inverted–U curve for the relationship between income level and income inequality. We also find that government social expenditure contributes to more equal distribution of income. However, a significant proportion of cross–country variation in income inequality remains unexplained.     

INTERGENERATIONAL INCOME MOBILITY IN SWEDEN: WHAT DO TAX‐DATA SHOW?

The aim of this paper is to investigate intergenerational income mobility in Sweden by means of a representative sample drawn from tax-data files. Longitudinal data on actual parent-child pairs spanning 1978–92 are employed. Regression and correlation coefficients are analyzed and transition matrices calculated in order to investigate income mobility over generations. The results achieved show high intergenerational income mobility in Sweden between fathers and sons in comparison to estimations performed in most other countries and more especially compared to the U.S. This indicates that Sweden does not only have lower cross-sectional income inequality, but also higher intergenerational income mobility than those countries. The mother's earnings influence children's earnings less than the father's. However, the mother's earnings correlate more strongly with a daughter's earnings than they do with that of a son. The major indication of immobility across generations is found in the upper income deciles.     

Transition from unemployment benefits to social assistance in seven European OECD countries

The topic of this paper is the transition from unemployment benefit schemes to social assistance in seven European OECD countries. The unemployment benefit schemes are formally quite different in the seven countries. Most are mandatory but Denmark and Sweden have voluntary unemployment insurance and Finland has a mandatory basic scheme with a voluntary income related component on top of that. Self employed people can join the U.B. schemes in the 3 Scandinavian countries. All the U.B. schemes have working or contribution conditions to be met by the members in order to obtain eligibility for benefits. These conditions are relatively tight in the Netherlands, France and Sweden and relatively easy in Denmark and Great Britain with Germany and Finland in between (based on rules in 1994–95). Recent developments in several countries have been to tighten the access conditions to the U.B. schemes and from 1997 Denmark will be ‘on line’ with Germany. The Netherlands have tightened the access criteria very significantly in 1995. There is a considerable variation as far as the duration of the unemployment benefit period is concerned. Sweden has a benefit period in fact without effective time limitations. The Danish benefit period is also very long, 7 years now being reduced to 5 years. 5 years is also the maximum duration in the Netherlands and in France, but only after many years of work and after a relatively high age has been reached. The maximum period in Germany, 22/3 years, also requires a long work history and a relatively high age. Finland and Great Britain have uniform benefit period (just as in Denmark and Sweden), in Finland it is appr. 2 years (longer for elderly unemployed just as in Sweden and Denmark) and in Great Britain it was 1 year but from October 1996 it was reduced to 1/2 year. The differences in the duration of the benefit periods between the seven countries are very considerable. As already mentioned, there has been a tendency to reduce the benefit period in several countries. Such a change is also being considered for the ‘never ending’ benefit period in the Swedish U.B. scheme. The benefit formula is purely flat rate in Great Britain and income related in the other countries. There is a maximum benefit level in 5 of these countries, but not in Finland, where the compensation is stepwise decreasing with increasing income. In the 5 countries with income related benefits and a maximum benefit level, this maximum level is reached at a relatively low income in Denmark (2/3 APW income) and Sweden (close to APW income) and at a relatively high income in the Netherlands (appr. 1.5 APW income) and Germany (appr. 1.7 APW income) and at a very high income level in France. France is the only country among the 7, where the benefits after an initial period are being reduced regularly (every 4 months) in the benefit period down to a minimum level. Sweden and Germany have reduced the benefit levels in recent years. Denmark has the highest gross compensation percentage, 90, in relation to lost income, but it is only effective over a relatively narrow income interval, from approx. 133,000 DKK to 162,000 DKK (1996), ‘between’ the min. and max. U.B. rates. According to the 3 institutional criteria applied here, access to the schemes, duration of the benefit period and the type of benefit formula, the U.B. schemes of the seven countries studied are very different. The exit scheme from U.B is social assistance in most of the countries, but not in Germany, where it is possible to continue in a scheme with lower compensation but still income related. For Sweden it is hardly meaningful to speak of an ‘exit scheme’ when the U.B. insurance is without effective time limitations. The exit schemes in the other countries are all characterized by having flat rate benefits. All the exit schemes are means tested and this is a crucial difference to the U.B. schemes, and they are without time limitations. Means testing and no effective time limitations are usual characteristics for social assistance and social assistance like schemes. In all the countries, except in Denmark and Sweden, there is ‘topping-up’ from social assistance to a guaranteed minimum level disregarding the income sources. In the two Scandinavian countries mentioned, a ‘social event’, i.e. illness or unemployment, is required in order to be eligible for social assistance benefits. Net replacement rates are used to illustrate the levels of compensation within the U.B. and the S.A. schemes, to identify possible incentives problems, and to illustrate the economic implications of the transition from U.B. to S.A. schemes. The net replacement rates presented are calculated by using the ‘disposable income after net housing costs’ income concept. The calculations include several family types, singles and couples with and without children and for the couples with one or two incomes. A general result but with some modifications, cf. the following, could be that the U.B. based replacement rates usually are higher than the S.A. based, but that the difference is minor when the U.B. scheme is flat rate or income related with a maximum benefit level being reached at a relatively low income, for Great Britain there are in several cases no difference at all. The ‘topping-up’ has the implication that the U.B. and S.A. based replacement rates are often identical at the lower end of the income scale. The very high S.A. based replacement rates (well over 100 per cent) often seen for Denmark and Sweden at low income levels do not necessarily imply, that social assistance in those two countries is more generous than in the other countries, it is very much an effect of not having more or less automatic ‘topping-up’ to a guaranteed minimum income level when earned income is low. The single parent family type seems to have incentives problems at relatively low income levels in most of the countries, especially when receiving U.B., she may temporarily be caught in the ‘unemployment trap’. The one earner couple with children may also be exposed to the ‘unemployment trap’ but on a more permanent basis. The S.A. based replacement rates for this family type are extraordinarily high (and higher than the U.B. based) in Denmark and Sweden, where this family type, however, is very rare. For the two earner family (where one of the spouses always has earned income) the means testing of S.A. makes an impact on the S.A. based net replacement rates, they are in most cases substantially lower than the U.B. based. The results of the net replacement calculations indicate no or only minor economic implications by a transition from U.B. to S.A. in the lower end of the income scale while the effect in most cases will be more substantial in the higher end of the income scale. This is not always the case in Sweden and Denmark where S.A. for some family types are preferable to U.B. and where the difference in other cases may be so small, that it implies incentives problems for joining the voluntary U.B. scheme. The calculation of ‘long term’ (5 years) net replacement rates for families at a low income level (the point in the income distribution where only 5 per cent have lower income) and only including one earner families, reveals that either the long benefit period in the U.B. scheme (Sweden and Denmark) or ‘topping-up’ (Germany, the Netherlands and Great Britain) generate ‘constant’ net replacement rates, and quite high ones, at low income levels. Only in France and to some extent in Finland will there be a decrease in net replacement rates over time. Most of the differences concerning the duration of unemployment benefit periods and to some extend the benefit formulas have no effect on the long term net replacement rates. At higher income levels the time limitations in the U.B. schemes will be visible again, except in Great Britain where the U.B. and S.A. benefits are almost identical. It is not possible to point out a ‘worst’ country with respect to ‘incentives’ problems, but the Danish, and to some degree also the Swedish, U.B. schemes with their high net replacement rates at lower income levels and long duration periods ‘stand’ out in many cases, and the S.A. schemes in those two countries may also contribute to permanent ‘unemployment traps’ as well as lack of incentives to join the voluntary insurance schemes. The two Scandinavian countries are, however, not alone with ‘incentives problems’. The section on long term net replacement rates showed that for low income levels the replacement rates were high and constant in most countries. This result is often due to ‘topping-up’ to a guaranteed minimum standard.     

On the Evaluation of Economic Mobility

This paper presents a framework for the evaluation and measurement of "reversal" and "origin independence" as separate aspects of economic mobility. We show that evaluation depends on aversion to multi-period inequality, aversion to inter-temporal fluctuations, and aversion to future risk. We construct "extended Atkinson indices" that allow us to quantify the relative impact of reversal and origin independence on welfare. We apply our approach to the comparison of income mobility in Germany and in the U.S.. When aversion to inequality is the only consideration, the U.S. gains more from mobility than Germany. This reflects similar gains from reversal in the two countries but greater gains in the U.S. from origin independence. The introduction of aversion to intertemporal fluctuations and aversion to future risk makes the impact of mobility in the two countries more similar.     

INCOME INEQUALITY IN URBAN COLOMBIA: A DECOMPOSITION ANALYSIS

The persistence of poverty and income inequality in less developed countries (LDCs) is a source of serious concern to development economists. To understand the structure of inequality, several researchers using a variety of methodologies have measured the importance of various contributory factors to overall income variability. The available literature—which now includes studies of Brazil, Mexico, Iran, the Philippines, Taiwan, Thailand, Pakistan, and Colombia-has been reviewed elsewhere (Fields, forthcoming). This paper presents additional evidence for urban Colombia, in the process raising some important methodological issues which bear on the design of future research studies.
The data set used in this paper is described in Section I. The decomposition of Colombian inequality by functional income source is presented in Section 11 for micro data. Section I11 examines the robustness of source decomposition procedures to data aggregation. Section IV presents inequality decompositions by city, and Section V by other income-determining characteristics. Conclusions appear in Section VI.     

RANK AND QUANTITY MOBILITY IN THE EMPIRICAL DYNAMICS OF INEQUALITY

Horizontal and vertical measures of inequality are related through mobility. The paper draws attention to two types of mobility: quantity mobility, which refers to mobility in income itself, and rank mobility, which refers to mobility in the position in the distribution of income. Individually matched census data for earnings in Israel are used to illustrate these concepts empirically. Mobility is measured between 1983 and 1995. It is shown that earnings in Israel are highly mobile. The high degree of earnings mobility implies that horizontal measures of inequality considerably overstate the underlying level of inequality. The method of errors in variables is used to distinguish between current and permanent mobility and inequality. Permanent earnings are more equal than current earnings and less mobile. Finally, the methodology is applied to PSID. It is shown that earnings were more mobile in Israel than in the United States.     

INEQUALITY OF OPPORTUNITIES VS. INEQUALITY OF OUTCOMES: ARE WESTERN SOCIETIES ALL ALIKE?

This paper analyzes the relationship between income inequality and inequality of opportunities for income acquisition in nine developed countries during the 1990s. Equality of opportunity is defined as the situation where income distributions conditional on social origin cannot be ranked according to stochastic dominance criteria. We measure social origin by parental education and occupation and use the database built by Roemer et al. (2003 ). Stochastic dominance is assessed using nonparametric statistical tests. Our results indicate strong disparities in the degree of equality of opportunity across countries and a strong correlation between inequality of outcomes and inequality of opportunity. The U.S. and Italy show up as the most unequal countries in terms of both outcome and opportunity. At the opposite extreme, income distributions conditional on social origin are almost the same in Scandinavian countries even before any redistributive policy. We complement the ordinal comparison by resorting to an original scalar “Gini” index of opportunities, which can be decomposed into a risk and a return component. In our sample, inequality of opportunity is mostly driven by differences in mean income conditional on social origin, and differences in risk compensate the return element in most countries.     

The Shape of the Income Distribution and Economic Growth – Evidence from Swedish Labor Market Regions

We analyze the association between income inequality and economic growth using 72 labor market regions in Sweden during the period of 1990–2006. Compared with studies of cross‐country data, the regional set‐up reduces problems with omitted variable bias and endogeneity as regions within a country share the same redistributive policies and institutions. Using population register data, highly accurate measures of growth and inequality (gini, Q3, p9075, p5010) are derived. OLS cross‐section and panel estimates imply that inequality between the 90 and 75th percentiles enhances regional growth and that the share of income falling to the third quintile reduces growth. These results no longer hold when we apply regions specific fixed effects and/or system GMM.     

ESTIMATING INCOME INEQUALITY IN CHINA USING GROUPED DATA AND THE GENERALIZED BETA DISTRIBUTION

There are two main types of data sources of income distributions in China: household survey data and grouped data. Household survey data are typically available for isolated years and individual provinces. In comparison, aggregate or grouped data are typically available more frequently and usually have national coverage. In principle, grouped data allow investigation of the change of inequality over longer, continuous periods of time, and the identification of patterns of inequality across broader regions. Nevertheless, a major limitation of grouped data is that only mean (average) income and income shares of quintile or decile groups of the population are reported. Directly using grouped data reported in this format is equivalent to assuming that all individuals in a quintile or decile group have the same income. This potentially distorts the estimate of inequality within each region. The aim of this paper is to apply an improved econometric method designed to use grouped data to study income inequality in China. A generalized beta distribution is employed to model income inequality in China at various levels and periods of time. The generalized beta distribution is more general and flexible than the lognormal distribution that has been used in past research, and also relaxes the assumption of a uniform distribution of income within quintile and decile groups of populations. The paper studies the nature and extent of inequality in rural and urban China over the period 1978 to 2002. Income inequality in the whole of China is then modeled using a mixture of province-specific distributions. The estimated results are used to study the trends in national inequality, and to discuss the empirical findings in the light of economic reforms, regional policies, and globalization of the Chinese economy.     

Unemployment Shocks and Income Distribution: How did the Nordic Countries Fare during their Crises?

We study the inequality of disposable income in Denmark, Finland, Norway and Sweden during the late 1980s and early 1990s when unemployment rose dramatically in all four countries. A standard measure of inequality — the Gini coefficient – was surprisingly stable in all countries during this period. By decomposing the Gini into income components, we test hypotheses about the reasons for this stable income distribution. Our most straightforward hypothesis, that rising unemployment benefits counteracted the impact of more unequally distributed earnings, receives only limited support. More complex mechanisms seem to have been at work.
JEL classification: D 30; D 31; J 60     

Politics and Income Inequality: Does Politics Still Matter in New Democracies

Does politics still matter for reducing income inequality in new democracies? The standard explanation is that political institutions, in particular the left government and proportional representation, are negatively associated with income inequality among advanced industrial countries, but there have been so few studies attempting to explain the variation of distributional outcomes across new democracies. This article tests the hypotheses about the effects of political institutions on income inequality with unbalanced pooled time-series cross-sectional data that cover 26 fledgling democracies for 1975–2006. The evidence presented here suggests that, other things being equal, a parliamentary system and PR are substantially more likely to be associated with lower levels of income inequality, but a left government and more years of democracy do not appear to be related to lower income inequality.     

Income Inequality in Australia in an International Comparative Perspective

Earlier comparative work on income distribution has tended to suggest that Australia is characterised by less income inequality than other industrialised economies. Concerns about the quality of the Australian data used in such comparisons have led to the need for more detailed assessment of the situation. The Luxembourg Income Study has been a focus for this work by bringing together microdata sets for a range of countries and reorganising them to conform to standardised concepts and definitions. This paper builds on earlier work undertaken as part of the Luxembourg Income Study by including Australia in an international comparative analysis of income distribution and redistribution. The Australian data are those from the 1981–82 Income and Housing Survey, with income tax imputed onto the data file. Results are presented for the gross and net income distributions between both families and individuals in seven countries. A common set of equivalence scales is also used to adjust for differing family needs. The results indicate that, using several summary measures of inequality, the distribution of income in Australia is less equal than in four of the other six countries studied. Earlier research which placed Australia high on the international league table of income equality is thus not confirmed by the results.     

IRELAND's INCOME DISTRIBUTION IN COMPARATIVE PERSPECTIVE

There are concerns that the unprecedented economic boom which Ireland experienced in the second half of the 1990s has raised only some living standards and has widened income gaps. This paper analyzes Ireland's income distribution in comparative perspective, to understand how Ireland's distribution changed and how it compares to other rich countries. We begin with OECD (Organization for Economic Cooperation and Development) and the Luxembourg Income Study (LIS) data to compare Ireland's degree of well-being and inequality with other advanced countries. We also look in some detail at alternative sources of Irish income and their implications for the trends in income inequality in Ireland from 1994 to 2000. For instance, we examine the top of the distribution using data from the administration of the income tax system. We conclude that the spectacular economic growth in the past decade has seen the gap in average income between Ireland and the richer OECD countries narrow dramatically. However, this growth has not greatly affected the Irish ranking in terms of income inequality. Ireland remains an outlier among rich European nations in its high degree of income inequality, though still falling well short of the level seen in the United States. In the end, we find that Ireland's new-found prosperity provides a "social dividend," and choices about how it is used will fundamentally affect whether the current high level of income inequality persists into the future.