Search results

Hypothesis tests for dominance in income distributions has received considerable attention in recent literature. See, for example, Barrett and Donald (2003a, b), Davidson and Duclos (2000) and references therein. Such tests are useful for assessing progress towards eliminating poverty and for evaluating the effectiveness of various policy initiatives directed towards welfare improvement. To date the focus in the literature has been on sampling theory tests. Such tests can be set up in various ways, with dominance as the null or alternative hypothesis, and with dominance in either direction (X dominates Y or Y dominates X). The result of a test is expressed as rejection of, or failure to reject, a null hypothesis. In this paper, we develop and apply Bayesian methods of inference to problems of Lorenz and stochastic dominance. The result from a comparison of two income distributions is reported in terms of the posterior probabilities for each of the three possible outcomes: (a) X dominates Y, (b) Y dominates X, and (c) neither X nor Y is dominant. Reporting results about uncertain outcomes in terms of probabilities has the advantage of being more informative than a simple reject/do-not-reject outcome. Whether a probability is sufficiently high or low for a policy maker to take a particular action is then a decision for that policy maker.

The methodology is applied to data for Canada from the Family Expenditure Survey for the years 1978 and 1986. We assess the likelihood of dominance from one time period to the next. Two alternative assumptions are made about the income distributions – Dagum and Singh-Maddala – and in each case the posterior probability of dominance is given by the proportion of times a relevant parameter inequality is satisfied by the posterior observations generated by Markov chain Monte Carlo.

We use stochastic dominance to test whether investors should prefer riskier securities as the investment horizon lengthens. Simulated return distributions for stocks, bonds, and U.S. Treasury bills are generated for holding periods of one to 20 years and stochastic dominance tests are run to establish preferences among the alternative portfolios. With independent returns, we find no evidence that high-risk securities (stocks) dominate low-risk securities (bonds) as the investment horizon lengthens. Under the assumption that security returns are correlated across time, we find that common stocks dominate corporate bonds and U.S. Treasury bills for sufficiently long investment horizons.

In an evaluation of a job training program, the causal effects of the program on wages are often of more interest to economists than the program's effects on employment or on income. The reason is that the effects on wages reflect the increase in human capital due to the training program, whereas the effects on total earnings or income may be simply reflecting the increased likelihood of employment without any effect on wage rates. Estimating the effects of training programs on wages is complicated by the fact that, even in a randomized experiment, wages are truncated by nonemployment, i.e., are only observed and well-defined for individuals who are employed. We present a principal stratification approach applied to a randomized social experiment that classifies participants into four latent groups according to whether they would be employed or not under treatment and control, and argue that the average treatment effect on wages is only clearly defined for those who would be employed whether they were trained or not. We summarize large sample bounds for this average treatment effect, and propose and derive a Bayesian analysis and the associated Bayesian Markov Chain Monte Carlo computational algorithm. Moreover, we illustrate the application of new code checking tools to our Bayesian analysis to detect possible coding errors. Finally, we demonstrate our Bayesian analysis using simulated data.

This paper deals with poverty decompositions into subgroups defined with respect to intervals of income and the robustness of comparisons of the absolute contribution of such groups to poverty. For instance, world poverty estimates by the World Bank often distinguish between the extreme poor whose incomes are lower than $1.25 a day (in PPP terms) and the other poor with incomes between $1.25 and $2.5 a day. Existing dominance conditions can tell whether overall poverty and extreme poverty have declined in a robust manner when comparing countries at two points of time, but they cannot say anything for the contribution of the non-extreme poor to overall poverty. In the present paper we propose stochastic generalized dominance criteria to perform robust poverty ordering when the focus is placed on some interval of the poverty domain. Using generated data based on grouped data from World Bank’s PovcalNet tool, the paper finally investigates whether the robust decline of extreme poverty around the world during the last decades was also accompanied by a decline of the contribution of non-extreme poverty.

Dominance techniques are used to analyze and rank inequality, welfare, and poverty across regions in Thailand in the 1990s. Inference-based dominance methods are applied to consumption expenditure microdata from the Household Socio-Economic Surveys (SES) of 1992, 1994, 1996, 1998 and 2000. Attention is focused on the period immediately before and after the economic contraction of 1996–1997. Lorenz dominance is employed to assess inequality, while first-order Engel food share dominance is applied to rank welfare across time and among regions. Poverty is evaluated by comparing truncated food-share quantile functions. The evidence reveals that the economic crisis in 1997 seems to affect inequality in Bangkok (the richest region) more than the Northeast (the poorest region), and most dramatic changes occur in the North and South. Welfare in Bangkok is unambiguously higher than in other regions before and after economic contraction. In fact, the great economic contraction changes the rankings of economic well-being and poverty only in the North, South, and Northeast.

This chapter conducts a systematic comparison of behavioral economics’s challenges to the standard accounts of economic behaviors within three dimensions: under risk, over time, and regarding other people. A new perspective on two underlying methodological issues, i.e., inter-disciplinarity and the positive/normative distinction, is proposed by following the entanglement thesis of Hilary Putnam, Vivian Walsh, and Amartya Sen. This thesis holds that facts, values, and conventions have inter-dependent meanings in science which can be understood by scrutinizing formal and ordinary language uses. The goal is to provide a broad and self-contained picture of how behavioral economics is changing the mainstream of economics.

In response to common criticisms on the appropriateness of mean-variance in asset allocation decisions involving hedge funds, we offer a mean-Gini framework as an alternative. The mean-Gini framework does not require the usual normality assumption concerning return distributions. We also evaluate empirically the differences in allocation outcomes between the two frameworks using historical data. The differences turn out to be significant. The evidence thus confirms the inappropriateness of the mean-variance framework and enhances the attractiveness of mean-Gini for this asset class.

Anchoring vignettes have become a popular method to adjust self-assessed data for systematic differences in reporting behaviour to aid comparability, for example, of cross-country analyses. The method relies on the two fundamental assumptions of response consistency and vignette equivalence. Evidence on the validity of these assumptions is equivocal. This chapter considers the utility of the vignette approach by considering how successful the method is in moving self-assessed reports of health mobility towards objective counterparts. We draw on data from the Survey of Health, Ageing and Retirement in Europe (SHARE) and undertake pairwise country comparisons of cumulative distributions of self-reports, their objective counterparts and vignette adjusted reports. Comparison of distributions is based on tests for stochastic dominance. Multiple cross-country comparisons are undertaken to assess the consistency of results across contexts and settings. Both non-parametric and parametric approaches to vignette adjustment are considered. In general, we find the anchoring vignette methodology poorly reconciles self-reported data with objective counterparts.

In a period of political change in the European Union, when the European Constitution is in the centre of the debate, the social convergence among European Union countries is a crucial issue. However, the measurement of welfare, inequality and poverty and the comparisons among countries are issues of great controversy. One of the main reasons for this is that implicit or explicit value judgements have to be made, and it is not easy to determine which of these value judgements are the most appropriate ones. In this paper we apply inference-based stochastic dominance methods to study welfare, inequality and poverty in European Union countries in 2000, applying purchasing power parities from the OECD. There are two main advantages of the methods and data used in this work: on the one hand, the stochastic method uses explicit and widely, though not universally, accepted assumptions, and if this small number of assumptions is accepted, the welfare and poverty ranking that the method provides is unambiguous. On the other hand, the use of the European Community Household Panel permits the comparisons in welfare, inequality and poverty among different countries using harmonised data. In addition, the use of inference tests permits a more precise ranking.