Estimate Equivalence Scales Research Articles

Equivalence scales for continuous distributions of expenditure

Research background: In the actual sizable populations of households, the standard microeconomic concept of equivalence scales is intractable since its necessary condition of equality of household welfare levels is unlikely to be fulfilled. Purpose of the article: This paper aims to develop a concept of an equivalence scale, which can be suitable for continuous distributions of expenditures in the population. Methods: Using household welfare intervals, we get the random equivalence scale (RES) as the ratio of expenditure distributions of the compared populations of households. Findings & value added: We derive the parametric distribution of RES for the lognormal distributions of expenditures. The truncated distribution of RES is applied to account for possible economies of scale in the household size. A society?s inequality aversion can be helpful when selecting a single equivalence scale. We estimate RES for Poland using microdata on expenditures and subjective assessments of household welfare intervals. The estimated equivalence scales turned out to be very flat and dependent on welfare.

Estimating equivalence scales and non-food needs in Egypt: Parametric and semiparametric regression modeling.

This paper investigated the appropriate specifications of Engel curves for non-food expenditure categories and estimated the deprivation indices of non-food needs in rural areas using a semi parametric examination of the presence of saturation points. The study used the extended partial linear model (EPLM) and adopted two estimation methods—the double residual estimator and differencing estimator—to obtain flexible shapes across different expenditure categories and estimate equivalence scales. We drew on data of the Egyptian Household Income, Expenditure, and Consumption Survey (HIEC). Our paper provides empirical evidence that the rankings of most non-food expenditure categories is of rank three at most. Rural households showed high economies of scale in non-food consumption, with child’s needs accounting for only 10% of adult’s non-food needs. Based on semi-parametrically estimated consumption behavior, the tendency of non-food expenditure categories to saturate did not emerge. While based on parametrically estimated consumption behavior, rural areas exhibited higher deprivation indices in terms of health and education expenditure categories, which indicates the need to design specific programs economically targeting such vulnerable households.

Equivalence Scale and Income Poverty: Two Approaches to Estimate Country-specific Scale for the Czech Republic

The at-risk-of-poverty rate, the relative income poverty indicator applied in the EU, can be highly sensitive to the equivalence scale used to transform household income to an equivalent for individuals. This study applies two well-established approaches to estimate the equivalence scale: an ‘objective’ one, based on consumption expenditures available in the national Household Budget Survey, and a ‘subjective’ one, based on the Minimum Income Question available in EU–Statistics on Income and Living Conditions data. The aim is to contrast the two estimated equivalence scales in the Czech Republic in the 2006–2016 period with the OECD-modified scale applied uniformly for decades across the EU countries. Our findings suggest that the adult weight in the equivalence scale is decreasing over time, while the child weight is relatively stable under both approaches. The estimated weights are lower than the officially applied ones, with the exception of the expenditure-based adult weight, which is very close to the OECD-modified weight. Applying the estimated scales affects the income poverty rate and leads to different rates than the official ones: while the trends of the rates are similar when the two estimated scales are used, the official income poverty rate trend deviates from those two.

Differences across countries and time in household expenditure patterns: implications for the estimation of equivalence scales

ABSTRACT When comparing economic well-being using income or expenditures, an equivalence scale is often used to adjust for differences in characteristics that affect needs. For example, a family of two is assumed to need more income than a single person, but not twice as much due to the economies of scale in consumption. In this study, we ask whether it is appropriate to use a common equivalence scale when comparing economic well-being across countries and/or time if consumption expenditure patterns differ? Based on an Engel methodology, we estimate equivalence scales for a diverse set of countries (Canada, France, Israel, Poland, South Africa, Switzerland, Taiwan, United States) in different time periods (1999–2012). We find considerable differences in economies of scale across countries, as well as increases over time. Notably, we find that economies of scale are larger than those implied by the widely accepted ‘square root of household size’ equivalence scale. Our results indicate that using a common equivalence scale to compare economic well-being across countries and/or time is misleading. Specifically, if economies of scale are understated (as is the case when using the ‘square root of household size’), the relative poverty experienced by larger versus smaller families is being overstated.

Differences across Place and Time in Household Expenditure Patterns: Implications for the Estimation of Equivalence Scales

When comparing economic well-being using income or expenditures, an equivalence scale is often used to adjust for differences in characteristics that affect needs. For example, a family of two is assumed to need more income than a single person, but not twice as much due to the economies of scale in consumption. In this study, we ask whether it is appropriate to use a common equivalence scale when comparing economic well-being across countries and/or time if consumption expenditure patterns differ?Based on an Engel methodology, we estimate equivalence scales for a diverse set of countries (Canada, France, Israel, Poland, South Africa, Switzerland, Taiwan, United States) in different time periods (1999-2012). We find considerable differences in economies of scale across countries, as well as increases over time. Notably, we find that economies of scale are larger than those implied by the widely accepted 'square root of household size' equivalence scale.Our results indicate that using a common equivalence scale to compare economic well-being across countries and/or time is misleading. Specifically, if economies of scale are understated (as is the case when using the 'square root of household size'), the relative poverty experienced by larger versus smaller families is being overstated.

Assessing differences in household needs: a comparison of approaches for the estimation of equivalence scales using German expenditure data

Equivalence scales are routinely applied to adjust the income of households of different sizes and compositions. Because of their practical importance for the measurement of inequality and poverty, a large number of methods for the estimation of equivalence scales have been proposed. Until now, however, no comprehensive comparison of current methods has been conducted. In this paper, we employ German household expenditure data to estimate equivalence scales using several parametric, semiparametric, and nonparametric approaches. Using a single dataset, we find that some approaches yield more plausible results than others while implausible scales are mostly based on linear Engel curves. The results we consider plausible are close to the modified OECD scale, and to the square root scale for larger households.

Catastrophic health payments: does the equivalence scale matter?

We present a revised method for estimating equivalence scales. Such scales are used to adjust household welfare to account for the size of the household, and are used extensively in the application of the World Health Organization's (WHO) methodology for the evaluation of catastrophic health payments. Applications of the WHO method are underpinned by early estimates that do not control for household income, and, therefore, are likely to overstate equivalence. Thus, in addition to revising the method, we update the scale estimates for one country, South Africa, using more recent data. South Africa is considered, because the end of Apartheid has led to extensive social and economic changes that have influenced household structure and, presumably, equivalence. We also present information on the possible degree to which earlier estimates are overstated, as well as the effect that has on other components of the WHO method, especially the determinants of out-of-pocket expenditures and catastrophic health payments. We find that, in the worst case, initial estimates could be overstated by as much 46%, leading to the understatement of poverty lines by as much as 17%. Despite these large differences, the average incidence of catastrophe in health expenditure was largely unaffected. Instead, differences in scales affect conclusions related to the determinants of out-of-pocket payments and catastrophic health expenditures, as well as the distribution of catastrophe across household size. Given that South Africa has low levels of catastrophic health expenditure, the effect could be even larger in other countries, and, therefore, we recommend that researchers consider a range of scales, when examining catastrophic health expenditures.

Welfare analysis when people are different

AbstractInequality and poverty estimation (indeed, all welfare analysis) must deal with the fact that people are heterogeneous. Equivalence scales and indifference scales are tools that may be used for this. An equivalence scale gives the relative costs faced by different people; an indifference scale gives the relative cost of living for people in different types of households. Equivalence scales and indifference scales can be estimated using off‐the‐shelf household‐level consumer expenditure data and standard econometric techniques for nonlinear equation systems. I offer a short introduction to the identification, estimation and use of equivalence scales and indifference scales and argue that these are complementary tools in the analysis of inequality and poverty. The methods are illustrated with Canadian household expenditure data from the Surveys of Household Spending 2004–2009. Estimated equivalence scales for disability are presented, along with estimated household model parameters and an analysis of consumption poverty.Résumé Analyse du niveau de bien‐être quand les gens sont différents. Les mesures d’inégalité et de pauvreté (en fait toutes les mesures et analyses du niveau de bien‐être) doivent prendre en compte le fait que les gens sont hétérogènes. Les échelles d’équivalence et les échelles d’indifférence sont des outils qu’on peut utiliser pour ce faire. Une échelle d’équivalence établit les coûts relatifs auxquels les différentes personnes doivent faire face; une échelle d’indifférence établit le coût relatif différent de gens qui font partie de différents types de ménages. Les échelles d’équivalence et les échelles d’indifférence peuvent être estimées à l’aide de données sur les dépenses de consommation des ménages pour les produits et services disponibles sur le marché, et des techniques économétriques conventionnelles pour calibrer les systèmes d’équations non‐linéaires. L’auteur présente une courte introduction aux problèmes d’identification, d’estimation, et d’usage des échelles d’équivalence et des échelles d’indifférence, et propose que ce sont des outils complémentaires dans l’analyse de l’inégalité et de la pauvreté. Les méthodes sont illustrées à l’aide de données sur les dépenses des ménages canadiens fournies par les Enquêtes sur les dépenses des ménages 2004‐2009. Les échelles d’équivalence sont estimées pour l’incapacité, ainsi que les paramètres estimés du modèle des ménages et une analyse de la pauvreté en termes de consommation.

There and back again – estimating equivalence scales with measurement error

ABSTRACTUsing income satisfaction data from the German Socio-Economic Panel, we find large differences in the equivalence weight of a partner when it is being estimated by direct and reverse regressions. We argue that neither of the two models will produce consistent estimates when there is stochastic error in satisfaction and measurement error in incomes. We propose a correction of mismeasured incomes using a constructed alternative income measure. The corrected results are relatively close to those obtained from direct regression. We do not find evidence that previous studies, using the direct regression method, severely suffer from measurement error in incomes.

Welfare analysis and redistributive policies

Applied welfare analyses of redistributive systems nowadays benefit from powerful tax-benefit microsimulation programs combined with administrative data. Arguably, most of the distributional studies of that kind focus on social welfare defined as a function – typically inequality or poverty indices – of household equivalized income. In parallel, economic research has made considerable progress in the measurement of welfare along several dimensions. Distinct but related branches of the literature have attempted (i) to model different behavior (in a way that matter for incidence and redistribution of tax-benefit policies), (ii) to go beyond income, (iii) to better define and estimate equivalence scales, (iv) to open the household black box and measure welfare at the individual level. I suggest a general framework to critically review these streams of literatures and to discuss whether recent advances in each of these fields have been or could be readily operationalized in welfare analyses and policy simulations.

Matching as Non-Parametric Preprocessing for the Estimation of Equivalence Scales

Abstract: Empirically analyzing household behavior usually relies on informal data preprocessing. That is, before an econometric model is estimated, observations are selected in such a way that the resulting subset of data is sufficiently homogeneous to be of interest for the specific research question pursued. In the context of estimating equivalence scales for household income, we use matching techniques and balance checking at this initial stage. This can be interpreted as a non-parametric approach to preprocessing data and as a way to formalize informal procedures. To illustrate this, we use German micro-data on household expenditure to estimate equivalence scales as a specific example. Our results show that matching leads to results which are more stable with respect to model specification and that this type of formal preprocessing is especially useful if one is mainly interested in results for specific subgroups, such as low-income households.

A Restatement of Equivalence Scales Using Time and Monetary Expenditures Combined with Individual Prices

In the classic equivalence scale estimations based on micro-econometric analysis, only the costs of market consumption are taken into account. We define the concept of full cost equivalence scales as integrating both monetary and the time use costs and measure it on matched French family budget and time use surveys. These data allow us to define full prices and to estimate equivalence scales conditional to possible substitution through prices. The results show higher full scales than the monetary for the preferred Independent of a Base specification. Relative to the cost of the second adult, the full child cost is also larger than the monetary for matching estimators and the Prais-Houthakker model. It shows households' capacity to substitute domestic production to market activities. In consequence the measures of income inequality diminish considerably using full income and full equivalence scales.

Engel's Equivalence Scales: Theory and Practice

The Engel’s approach is largely used to estimate equivalence scales. In this paper, we start by recalling the theoretical framework of this approach as well as the econometric methods which are usually adopted to estimate the Engel’s scales. We point out the different elements that may largely influence the estimated equivalence scales. Also, we discuss some possible pitfalls and how certain country contexts can lead to equivalence scale estimations which violate Engel's laws. For instance, this can be the case where extreme poverty is widespread. We show how the choice of the minimum expenditure threshold, the model’s specification, the demographic characteristics of the reference household and the reference age group can all substantially modify estimates of equivalence scales. The issues are illustrated using household data from Burundi and Mali.

A Nonparametric Partially Identified Estimator for Equivalence Scales

Methods for estimating equivalence scales usually rely on rather strong identifying assumptions. This paper considers a partially identified estimator for equivalence scales derived from the potential outcomes framework and using nonparametric methods for estimation, which requires only mild assumptions. Instead of point estimates, the method yields only lower and upper bounds of equivalence scales. Results of an analysis using German expenditure data show that the range implied by these bounds is rather wide, but can be reduced using additional covariates.

A Nonparametric Partially Identified Estimator for Equivalence Scales

Methods for estimating equivalence scales usually rely on rather strong identifying assumptions. This paper considers a partially identified estimator for equivalence scales derived from the potential outcomes framework and using nonparametric methods for estimation, which requires only mild assumptions. Instead of point estimates, the method yields only lower and upper bounds of equivalence scales. Results of an analysis using German expenditure data show that the range implied by these bounds is rather wide, but can be reduced using additional covariates.

Non-Parametric Preprocessing for the Estimation of Equivalence Scales

Empirically analyzing household behavior usually relies on informal data preprocessing. That is, before an econometric model is estimated, observations are selected in such a way that the resulting subset of data can be assumed to be sufficiently homogeneous with respect to the specific research question pursued. For example, households with members above retirement age may be excluded where it seems important that they differ from other households with respect to time use and home production. We propose the use of matching techniques and balance checking at this initial stage. This can be interpreted as a non-parametric approach to preprocessing data and as a way to formalize informal procedures. To illustrate this, we use German micro-data on household expenditure to estimate equivalence scales as a specific example. Our results show that matching leads to results which are more stable with respect to model specification and that this type of formal preprocessing is especially useful if one is mainly interested in results for specific subgroups, such as low-income households.

A Goodness-of-Fit Approach to Estimating Equivalence Scales

We use income satisfaction data in order to estimate equivalence scales. Our method differs from previous attempts to use satisfaction data for this purpose in that it can be used to estimate or evaluate any given parametric equivalence scale. It can also be employed to investigate specific questions related to equivalence scales such as whether these scales are income-dependent or to what extent they depend on specific household characteristics such as whether some household members are in a partner relationship. A further advantage of our approach is that, within a first-differencing nonlinear least-squares framework, we can carry out full statistical inference about the hypotheses under consideration. Our empirical results suggest that household economies of scale are higher than those assumed in commonly used scales such as the OECD or the Luxembourg scale. We also obtain the novel empirical result that economies of scale may decrease (rather than increase) in income, which is what one would expect if the share of consumption goods generating economies of scale (such as housing) decreases with income.

Life Insurance and Household Consumption

Using life insurance holdings by age, sex, and marital status, we infer how individuals value consumption in different demographic stages. We estimate equivalence scales and bequest motives simultaneously within a fully specified model where agents face US demographics and save and purchase life insurance. Our findings indicate that individuals are very caring for dependents, that economies of scale are large, that children are very costly (or yield very high marginal utility), that wives with children produce lots of home goods, and that females display habits from marriage, while men do not. These findings contrast sharply with standard equivalence scales.

The price and utility dependence of equivalence scales: Evidence from Indonesia

The purpose of this research is to estimate equivalence scales and evaluate their price and utility dependencies. To do this we unify two strands of the empirical literature on this topic, one that relies on demand data and one that relies on subjective evaluations of wellbeing. This way we are able to employ the strengths of both kinds of information. Equivalence scales are not identified from demand data alone. This is because changes in demographics could affect wellbeing directly, i.e., in ways that are not revealed by changes in behavior. The demand-based literature rules out such effects a priori and restricts parameters that measure them. In this paper we do not rely on such restrictions and instead use subjective evaluations of wellbeing as an additional source of information. Because demand data are highly informative about some of the other parameters of the model its use increases efficiency. This becomes critical once we specify complex preference structures that allow for the equivalence scale's dependence on price and utility. Our model nests the typical models used in the subjective and the demand literature and we are able to test down both kinds. We find evidence for direct utility effects. This rejects the validity of some of the restrictions applied by the demand literature within the Indonesian context. We estimate equivalence scales that are of reasonable magnitude and decrease in utility. We also find that the scales increase in the food to nonfood price ratio for poor households, whereas we do not find price dependence for more affluent households.

A Subjective Well-being Equivalence Scale for Mexico: Estimation and Poverty and Income-distribution Implications

The estimation of equivalence scales is crucial in cases where a well-being comparison of persons living under different household arrangements is needed: for example, to identify the poor, to calculate poverty rates and to estimate income inequality. In spite of the importance of equivalence scales for economic policy, there is little theoretical guidance on their estimation, and most empirical studies have been carried out in developed countries. Traditional estimation methods have been criticized because of their limitations for making welfare comparisons. This paper uses a subjective well-being approach to estimate equivalence scales. The approach provides an equivalence scale founded on economic satisfaction, which can be used to make welfare comparisons for persons living under different household arrangements—for example, in households of different sizes and with different age composition of household members. The empirical study has been carried out for Mexico using a large database. The implications of the subjective well-being scale for the assessment of poverty and income inequality in Mexico are shown on the basis of a national survey and by comparison with alternative scales.