Can welfare be measured with a preference-satisfaction index?

Welfare in economics is generally conceived of in terms of the satisfaction of preferences, but a general, comparable index measure of welfare is generally not taken to be possible. In recent years, in response to the usage of measures of subjective well-being as indices of welfare in economics, a number of economists have started to develop measures of welfare based on preference-satisfaction. In order to evaluate the success of such measures, I formulate criteria of policy-relevance and theoretical success in the context of preference-satisfaction measures of welfare. I present a detailed case study of the methodological choices put forward in a prominent generalized proposal for measuring welfare through preferences recently published in the American Economic Review. I contrast this with an alternative welfare measure which also uses preferences to weight aspects of welfare: the ICECAP-A measure. I assess the methodology of both approaches in detail and argue that the two goals of a preference measure of welfare can only be satisfied at the expense of making a measure prohibitively costly.     

COMPARING ECONOMIC WELL-BEING AMONG ELDERLY AMERICANS

Income-only and income-net worth measures of elderly economic well-being are derived from a single public welfare function-optimizing model of household production and intertemporal resource allocation. These measures are estimated with data on United States elderly in 1984 from the Survey of Income and Program Participation. The income-net worth measure also incorporates independently estimated work-life expectancies and earnings replacement rates. Under both measures, minority, moderately disabled, and unmarried female elderly are the poorest sub-populations. Increasing poverty with age is found under the income-only measure, but not the income-net worth measure.     

Environmental risk and welfare valuation under imperfect information

Consumers are often uninformed, or unsure, about the ambient level of environmental risk. An optimal policy must jointly determine efficient levels of self-protection, information provision, and public risk mitigation efforts. Unfortunately, conventional welfare measures are not amenable to welfare analysis in the presence of imperfect information. We develop a theoretical welfare measure, called quasi-compensating variation, that is a natural extension of compensating variation (CV). We show that this welfare measure offers not only a money metric of the “value of information,” but also a means to appropriately evaluate the welfare effects of various policies when consumers are imperfectly informed about ambient risk. This welfare measure allows us to obtain a number of results that the traditional CV measure fails to offer. In particular, we show that the consumer’s willingness to pay for a (small) environmental risk reduction is higher for those who underestimate ambient risk than for those who overestimate or are perfectly informed if the marginal return to self-protection increases with ambient risk.     

Consumers' Surplus As an Exact and Superlative Cardinal Welfare Indicator

This article shows that the Bennet-Bowley consumer surplus measure is an exact measure of Allais' disposable surplus if the consumer's utility function is of the translation-homothetic generalized quadratic form. The Bennet-Bowley consumer surplus measure, therefore, is a superlative cardinal welfare measure for the entire class of translation homothetic preferences. Because the exactness results in this article apply for cardinal welfare measures, they can be meaningfully aggregated across consumers to make aggregate welfare comparisons.     

EXPERIENCE IN MEASUREMENT OF WELFARE COMPONENTS AND THEIR REGIONAL IMPLICATIONS

The paper relates the Swedish discussion and criticism of national accounting statistics, especially GNP, as a measure of welfare. It describes some results of recent Swedish attempts to find practicable measures of welfare components, i.e. the investigations of the Low Income Commission and of the Expert Group on Regional Development Research. In both cases the regional aspects of welfare are emphasized in the paper.
The results are presented as a sign of important needs for new methods and new systems of concepts in measuring welfare. The Expert Group has for instance in different ways tried to map and compare the "service structure" of separate parts of Sweden. The Low Income Commission has principally studied the position in Swedish society of low income recipients and has not been working particularly on the illumination of regional differences, but since different types of region are included as a background variable, the investigations also give certain measures of the regional aspects of welfare.
In the last part of the paper some of the risks that seem to be difficult to avoid in trying to use welfare measurements are pointed out.     

PANEL DATA HEDONICS: ROSEN'S FIRST STAGE AS A “SUFFICIENT STATISTIC”

Traditional cross-sectional estimates of hedonic price functions can recover marginal willingness to pay for characteristics, but face endogeneity problems for estimating nonmarginal welfare measures. This article shows that when panel data on household demands are available, one can construct a second-order approximation to nonmarginal welfare measures using only the first-stage marginal prices. With repeated cross sections of product prices, the measure can be set identified or, under a single-crossing restriction, point identified. Bounds also can be constructed when there are mobility costs. Finally, a variant remains valid when individual preferences shift over time.     

HOW DO EUROPEANS EVALUATE INCOME DISTRIBUTIONS? AN ASSESSMENT BASED ON HAPPINESS SURVEYS

The purpose of this paper is to measure the evaluation of income inequality by European citizens. Starting from the concept of a social welfare function defined on income distributions the paper estimates the degree and nature of inequality aversion of Europeans. It uses subjective well-being (SWB) as an empirical measure of welfare and estimates how SWB is related to average income and measures of income inequality (from an appropriate class). The estimated relationship is used to determine those inequality measures which qualify as proper representations of people's inequality aversion.     

National product related welfare measures in the presence of technological change: Externalities and uncertainty

The purpose of this paper is to discuss under what conditions welfare can be measured by observables related to the national product (or Hamiltonian along the optimal trajectory). Under nonattributable technological or environmental change, welfare will depend on time itself, meaning that the Hamiltonian along the optimal trajectory will be a biased measure of welfare. This result will also hold if we make the time dependence of welfare endogenous, by replacing technological change will externalities that are not internalized during optimization. On the other hand, if we take the externalities fully into account, then the Hamiltonian will represent the appropriate measure of welfare. Similar results also hold in the case of uncertainty, where we show that a generalized Hamiltonian provides a welfare measure, and that the deterministic measures are special cases of their stochastic counterparts.     

Measuring the contribution of durable goods to the welfare cost of inflation

In this paper, we measure the contribution of durable goods to the welfare cost of inflation, in the context of an endogenous growth model with durable and nondurable goods, where purchases of the latter require only a partial cash payment compared with the former. Unlike existing measures, our proposed welfare measure is computationally efficient and relatively easy to implement.We find that durability adds a significant component to the welfare costs of inflation.     

AN EMPIRICAL ESTIMATE OF THE RELATION BETWEEN TWO MEASURES OF ECONOMIC WELFARE*

The main objective of this paper is to explore and quantify the difference between two measures of comparative economic welfare: (a) the more or less conventional measure of per capita national income, and (b) the capitalized value of expected future income per capita. The paper begins with a brief summary of the argument in favor of the present value of expected future income per capita as a measure of economic welfare. This is followed by an examination of the empirical relationship of the ratio of the suggested alternative measure to per capita income and an analysis of the variables used to compute the present value of expected income per capita. The main conclusion drawn from the calculations is that very substantial differences occur in the measurement of relative economic well-being depending on which measure is used. A final section discusses the implications of this finding for international comparisons of economic welfare.     

A SOCIAL WELFARE VIEW OF THE MEASUREMENT OF INCOME EQUALITY

Economists' use of the term "equality" in reference to a distribution of incomes has historically been in the sense of a consensus for some statistical characteristic(s) of the distribution rather than a firm concept of equality. Of course such a concept rests on appropriate welfare assumptions about income and its distribution, assumptions which, for the most part, have been left implicit (and unknown) in discussions of income equality in the literature.
Our purpose in this paper is dual: first, we wish to discover an unambiguous, welfare-related equality measure. This we accomplish through suitable assumptions on a social welfare function. What is produced is an "index" of equality which describes the performance of a given distribution relative to the maximum welfare derivable from the total income it represents. The measure thus depends functionally on the welfare attributes of income, something which in reality we know little about.
This impasse leads us to inquire into the sensitivity of the index over specifications of the welfare function, which is done by comparing equality ranks for the states of the United States for 1960 under various functional forms and among curves within a given form. As an interesting secondary issue, the performance of traditional equality measures is tested relative to the welfare-oriented index to discover implications about their welfare content.
It is found that the equality index is, in certain ranges for the welfare function, insensitive to its specification. The findings lead directly to conclusions concerning traditional equality measures, their usefulness in correctly accounting for equality differences among alternative income distributions and, concomitantly, their implicit welfare inputs.     

Measuring Attitudes Towards Inequality

Individuals' attitudes to inequality aversion are measured using survey data, based on the leaky-bucket experiment, for several groups of students in Australia and Israel. Three forms of social welfare function are estimated. It is found that measures of inequality aversion can be obtained with some precision and that these estimates are substantially lower than the values typically used by those measuring inequality and examining optimal tax structures. Furthermore, a welfare function based on the Gini inequality measure is generally found to give a better fit than forms based on constant relative or constant absolute inequality aversion.
JEL Classification : C 91; D 63     

Aggregate productivity and aggregate technology

Aggregate productivity and aggregate technology are meaningful but distinct concepts. We show that a slightly modified Solow productivity residual measures changes in economic welfare, even when productivity and technology differ because of distortions such as imperfect competition. Our results imply that aggregate data can be used to measure changes in welfare, even when disaggregated data are needed to measure technical change. We then present a general accounting framework that identifies several new non-technological gaps between productivity and technology, gaps reflecting imperfections and frictions in output and factor markets. Empirically, we find that these gaps are important, even though we abstract from variations in factor utilization and estimate only small average sectoral markups. The evidence suggests that the usual focus on one-sector DSGE models misses a rich class of important propagation mechanisms that are present only in multi-sector models.     

Welfare from a benefit viewpoint

Summary Individual welfare is most naturally measured in terms of individual utility but this has the well-known disadvantage that utility levels of different consumers cannot be meaningfully compared. This difficulty is traditionally avoided by using various willingness-to-pay measures, such as compensating and equivalent variation. These measures are based on price changes. This paper develops alternative welfare measures using willingness-to-trade concepts as originally proposed by Dupuit (1844). These measures are based directly on commodity bundle changes. These welfare measures can be represented as integrals under certain inverse demand functions.An important property of the proposed welfare measures studied here is that they can be meaningfully aggregated to form overall welfare measures. These measures in turn directly quantify a compensation criterion. It is shown that competitive prices provide a first-order approximation to the welfare measures. Furthermore a second-order approximation can be found by forming a suitable aggregation of the individual second-order effects.Finally, it is shown that the representations for consumer welfare as integrals under inverse demand curves can be extended to the aggregate measures as well. This then provides a complete complement to traditional measures based on price changes.This research was supported by Grant SES-9022881 from the National Science Foundation. The author wishes to thank Andrew Yates for valuable suggestions on this paper.     

SOCIAL CHALLENGES TO ECONOMIC ACCOUNTING AND ECONOMIC CHALLENGES TO SOCIAL ACCOUNTING

The question addressed in this paper is: Why can't we have a good measuring rod of the economic and social performance of our society? The answers are basically positive but lie mostly in the direction of (1) avoiding simplistic solutions such as turning the national income accounts into a measure of social welfare and (2) providing the elements of an information strategy to obtain such a measure or more accurately such a set of measures. The proposed information strategy highlights five activities: (1) the presentation and analysis of welfare outcomes, an activity which is analogous to but broader than “social indicators”; (2) social accounting which includes economic accounting, demographic accounting, and time-use accounting; (3) model building and operation which, unlike accounting, are concerned with behavioral or causal relationships used to explain and project welfare outcomes; (4) hypothesis testing to develop new insights into economic and social behavior; and finally (5) the building and maintenance of a data base required for carrying on the aforementioned four activities.     

Exact nonradial input, output, and productivity measurement

Summary. The use of measures originally suggested by Bennet, Bowley, and Hicks in the context of cost of living, welfare, and consumer surplus measurement to measure inputs, outputs, and productivity is examined. Suitably normalized versions of the Bennet-Bowley measures are shown to be exact and superlative measures of input, output, and productivity indicators. Received: April 2, 2001; revised version: August 8, 2001     

Trade,GMOs and Environmental Risk: Are Current Policies Likely to Improve Welfare?

Food with inputs from genetically modified organisms (GMOs) has met considerable skepticism among European Union (EU) consumers. The EU import ban on GM food has triggered a great deal of controversy and has been partly replaced by a mandatory labeling scheme. Although there is no measure in the General Agreement on Tariffs and Trade that directly addresses the use of product labeling, WTO and others have been skeptical to mandatory product labeling on the grounds that they may be used as hidden protectionism hampering global welfare. This study has two foci. First, we examine how different policies for the production and use of GMOs might influence the market outcome in consumer food markets. Second, we evaluate the welfare effects of the policy measures. We find that mandatory labeling often increases domestic welfare and, may also enhance global welfare. On the other hand, a trade ban is more likely to decrease global welfare.     

Evaluation of Welfare Change in Quantity-Constrained Regimes*

There are many circumstances where individuals are subject to constraints other than income and prices. Evaluation of welfare change from a price-taking position to a quantity-constrained position presents special problems. While remaining firmly in the partial-equilibrium tradition, indifference curves in the price-quantity space are applied to the identification of an equivalent variation. a modified Marshallian measure and a compensating variation in quantity-constrained situations. These measures satisfy the condition of CV < MS < EV for a normal good. A Marshallian measure that has been used—the ‘simple Marshallian measure’—is shown to have a considerable drawback.     

The inadequacy of cost of living indices based on subjective preferences: an ethical and methodological critique

In this paper we argue that there is scant justification for replacing the traditional fixed‐basket Laspeyres price index with so called ‘true cost of living indices’. We begin with a discussion of the possible explanations for some empirical results for inflation found for different social groups in Ireland in the late 1990s. Our arguments concerning appropriate inflation indices are primarily ethical and are not dependent on these results being interpreted in a non‐neoclassical vein. They do however gain extra force if one accepts non‐neoclassical explanations for the empirical results. We go on to draw conclusions as to how best to measure the welfare effects of changes in the price of goods. This links in to the broader debate regarding objective versus subjective measures of welfare.     

WELFARE MEASUREMENT,INVOLUNTARY UNEMPLOYMENT,AND HETEROGENEITY

This paper concerns welfare measurement in an economy with union wage setting, where the equilibrium is characterized by unemployment. Contrary to results derived in the first best, the current value Hamiltonian is not an exact welfare measure in an economy with unemployment. Instead, the welfare measure also depends on “employment effect,” which are caused by the discrepancy between supply and demand in the labor market. In addition, since unemployment gives rise to heterogeneity, distributional effects will also characterize the welfare measure.