The use of quantile regressions in estimating gender wage differentials: a case study of the Philippines

Quantile regressions are used to analyse the size, components, and nature of the gender-earnings differentials as well as the returns to education by gender in the Philippines, over the entire conditional distribution of earnings, using data from the 1999 APIS survey. This approach allows more focused conclusions about the nature of gender-earnings differentials. Incorporating the quantile regression approach to the Oaxaca–Blinder decomposition allows the localization of gender discrimination to particular quantiles of the earnings distribution and permits more targeted policy intervention.     

Old Wine in a New Bottle: Growth Convergence Dynamics in the EU

In this paper, we explore convergence of real per capita output across the European Union (EU) countries, as well as the transitional behavior of possible underlying factors that are responsible for any convergence or divergence pattern. The new panel convergence methodology developed by Phillips and Sul (2007) is employed in a production function growth accounting approach and data from the Total Economy Database and the Total Economy Growth Accounting Database. The empirical findings suggest that the EU countries form two distinct convergent clubs, exhibiting considerable heterogeneity in the underlying growth factors. These findings should help policy makers in designing appropriate growth-oriented programs as well as in setting priorities in their implementation.     

Data Aggregation Issues for Crop Yield Risk Analysis

With increased emphasis on risk management in agriculture and a lack of disaggregated or farm‐level yield time series, decision makers are often faced with having to make adjustments to temporal yield risk measures obtained from readily available but aggregated yield data. This paper provides some empirical evidence on what type of aggregation bias to expect when measuring temporal yield risk using yield observations averaged across a region relative to yield risk estimated from quarter‐section yield time series in wheat. This study highlights some of the challenges faced when estimating aggregation distortions in measuring yield risk defined by temporal variance, especially given the nature of the empirical data set used. Cluster analysis, visual examination of relative frequency distributions and mapping of yield risk clusters suggest that using a readily available, aggregate temporal yield risk measure has the tendency to underestimate yield risk observed at the quarter‐section level and that clear, geographic yield risk boundaries do not exist in municipalities or across larger areas in this study. Further research on crops more risky than wheat appears promising. Avec un plus grand intéret sur la gestion du risk dans l'agriculture et un manque de données détaillees ou bien de collections de séries temporelles sur les rendements, les décideurs sont souvent tenus d'apporter des correctifs aux measures du risk obtenues a partir des données de rendements qui sont disponibles. Cet artcle apporte une preuve empirique du type de biais lie a l'agrégation qui peut être présent dans le calcul du risk de rendement temporel obtenu a partir de rendements moyens de blé observés au niveau régional en comparaison du risk de rendement qui est estimé a partir de données basées sur des quart‐de‐sections. Cette étude met en exergue quelques uns des obstacles qui se présentent dans l'estimation de distosions liées a l'aggrégation dans le calcul du risk de rendement défini par la variance temporelle, speciallement étant donne la charactère empirique des données utilisées. L'analyse de groupe, l'examen visual de la distribution des fréquences relatives, et la cartographie de classes de risk de rendement suggèrent que l'utilisation de la measure du risk de rendement basée sur des données disponibles de risk aggrège temporel a tendence a sousestimer le risk de rendement observe au niveau des quart‐de‐sections et qu'il n'y a pas de frontières de risk de rendement certaines, géographiques qui existent entre les municipalités ou bien a travers les zones plus larges examinées dans cette etude.     

VOLUNTARY DISCLOSURE OF SEGMENT INFORMATION: FURTHER AUSTRALIAN EVIDENCE

This study provides further empirical evidence on incentives for Australian firms to voluntarily report segment information. Various economic reasons why firms may elect to present segment information have been hypothesised in previous research. Bradbury [1992] and McKinnon and Dalimunthe [1993] found firm size, minority interest and industry membership as significant identifiable characteristics motivating voluntary segmental disclosure. Variables found to be insignificant in Bradbury [1992] which were not examined by McKinnon and Dalimunthe [1993] are tested in this paper. Hypotheses relating to size, financial leverage, assets-in-place, earnings volatility, ownership diffusion, outside equity (minority) interest, overseas association as well as diverse and mining and oil classification hypotheses are empirically examined. Univariate tests and multivariate logit analysis suggest that for a extensive sample of diversified firms, voluntary segment disclosure is significantly related to size, leverage and involvement in mining or oil activities.     

Bottom Marx

Much of the economic analysis published in recent years has been Marxist in flavour. The publication of a dictionary of Marxist thought offers a guide to its various schools. Chris Tame dissects the contrasting interpretations of its numerous contributors.