“Parents and Children Talk: English Language Proficiency within Immigrant Families”

This paper extends the analysis of the acquisition of destination language proficiency among immigrants by explicitly incorporating interactions among family members—mother, father and children. Single equation, bivariate, and four-equation (multivariate) probit analyses are employed. Immigrant English language skills are greater the younger the age at migration, the longer the duration of residence, the higher the level of education, and for immigrants not from Asia. Large positive correlations in the unmeasured determinants of proficiency exist between spouses, between siblings, and between parents and children, although the latter relationship is stronger for the mother. The findings imply that learning takes place within the household.     

Assortative Mating by Education and Postponement of Couple Formation and First Birth in Britain and Sweden

This paper focuses on postponement of couple formation and first birth. We analyze couples born 1930–1979 who have experienced both events. First, we estimate a spouse choice equation based on individual human capital and family of origin characteristics as well as marriage market indicators, using a multinomial logit model. We create nine educational categories using predicted probabilities. Secondly, we estimate the rates of union formation and first birth using Weibull hazard models with individual unobserved heterogeneity, predicted educational categories from the first step, and other fixed and time varying variables. Higher education of one of the spouses, duration of education, and unfavorable labor market conditions delay couple formation and first birth. Swedish women form their unions later than British women, but once the union has been formed, they have their first birth sooner.JEL Classification: D1, J1An earlier paper, co-authored by Eiko Kenjoh, was presented at the Scholar seminar on Education and Postponement of Maternity, University of Amsterdam. We are thankful for comments from Shoshana Grossbard-Shechtman, Michael Lindahl, Adriaan Kalwij, Andrey Launov and other seminar participants at the University of Groningen, the University of Lund and ESPE 2004. We thank Howard Yourow for improving substantially our English and Sebastiene Postma and Robert Helmink for typing several versions of the paper. Further, we thank Eiko Kenjoh for giving us the country comparable education variable, which she constructed for earlier work (see Kenjoh, 2004).     

Childhood Abuse and Welfare Use

Policies aimed at reducing welfare use focus solely on adults, yet welfare users very often report experiences of childhood abuse. Such abuse is known to have long-term psychological effects and may set the stage for later welfare use. This study uses a random sample of poor women to determine how a history of childhood abuse relates to the probability of receiving cash and in-kind assistance over a five-year period. It also investigate whether childhood abuse correlates with the length of receipt among program users. Women experiencing both physical and sexual abuse during childhood were 16–25 percentage points more likely than others to use both cash and in-kind programs as adults. Conditional on program use, there was no relation of childhood abuse to the extent of program use during the study period.     

Informal Care Intensity and Caregiver Drug Utilization

Providing informal care has negative health consequences for informal caregivers. If these health consequences increase drug utilization among caregivers, estimates of health care savings from informal care—mainly realized through reductions in utilization among care recipients—should consider the increased drug costs incurred by informal caregivers. This paper evaluates whether more intensive informal caregivers have higher drug utilization than less intensive caregivers, controlling for initial health status and other factors. We find that informal care intensity is associated with higher drug consumption. An increase of 10% of total informal care per day is associated with a 0.7% increase in drugs. The small magnitudes indicate that, in this application, it is not important to consider caregiver drug utilization when quantifying the net savings to the health care system of informal care. For individual caregivers, such as those who take multiple drugs per month and/or have no drug coverage, the increase in drug utilization associated with intensive caregiving is likely to be costly.     

The Serbian Banking System: Recent Development and Major Players

The paper analyzes the banking system of Serbia. First, we briefly overview the institutional and legal framework of the banking system and describe the main structural features of the sector. Then we discuss the evolution in traditional banking intermediation, and outline the main features of the banking competitive environment, focusing on the transformations in ownership structure of banks, especially on the increasing importance of foreign ownership, and summarising the key characteristics of major players.     

Labour Migration and Competitiveness in the European Union

The study is about the characteristics and phenomena of the labour migration that fluxes from, through, and into the East Central European region. The typical groups of migrant workers are emphasized, like the qualified employees, the commuters, the illegal workers, or the migrants according to the family's income optimization. The brain drain is analyzed as a problem of the absence of experts in the region. The tools and possibilities of migration policy are discussed, too, to find suggestions for the state how to turn the direction of migration to the desired way. Through this topic, the countries of destination are introduced with the competitive position of the eastern migrants. The study also reflects on the problem of asymmetric flux of labour out of the region.     

Community and anti-poverty targeting

The standard theory of anti-poverty targeting assumes individual incomes cannot be observed, but statistical properties of income distribution in broadly defined groups are known. ‘Indicator targeting’ rules are then derived for the forms of transfers conditioned on group membership of individuals. In this literature the motivating notion of a ‘group’ is purely statistical, even when it is groups such as localities and ethnicities. We focus instead on groups which are ‘communities’, meaning thereby collections of individuals who have access to community-specific public goods, from which non-members are excluded. Such differential access constitutes a source of inequality among poor individuals belonging to different communities, which is not captured by monetary earnings. We show that this formulation of what constitutes a group changes many of the basic results of the indicator targeting literature. Optimal targeting for poverty alleviation leads to seemingly paradoxical rules, such as targeting transfers to the community that is richer. Total wealth of non-poor members of a community and its distribution both become relevant for specifying optimal indicator targeting rules. In addition, a poverty measure that is sensitive to the community identities of poor individuals, yet defined on nominal incomes, may be incompatible with some of the basic axioms in the standard literature on poverty measurement.     

A ``European Consensus' on the Growth Agenda? A Discussion of the Sapir Report

This paper offers a discussion of An Agenda for a Growing Europe: Making the Economic System Deliver, a report to the President of the European Commission prepared by a group of independent experts. The paper reflects the results of the 3rd Milan European Economy Workshop (May 28–29, 2004) and my personal views. The workshop was organised by the Department of Economics, University of Milan and the Jean Monnet Chair of Economics of European Integration, with the participation of more than twenty distinguished speakers from several institutions across the European Union.     

Costs and Financing of Maritime Education and Training in Europe: Analysis and Policy Implications

The paper reports some findings of an extensive European research project carried out on costs and funds of MET (maritime education and training) systems. The need to improve the quality of the European MET systems is a relevant political and scientific issue which strongly impacts the competitiveness of the shipping sector in Europe. However, MET is quite an expensive ET system since it requires high investments and incurs in high running costs. The situation is also acerbated by the decreasing number of students at MET institutions. There is then the need of having a closer look at the financial aspects of MET so as to come up with some policy recommendations for improving the current situation in terms of optimal use of resources. In the paper a questionnaire-based methodology is employed aiming at identifying and analyzing different cost and funds structures of MET institutions in Europe (cost models). Then, by considering some policy-sensitive variables of the cost models, some policy recommendations are drawn up to improve the efficiency of current systems. They basically refer to the need to concentrate and integrate resources in order to exploit some degree of economies of scale, on top of achieving some economies of scope and pursuing quality in MET. Finally, further lines of research are indicated.     

Bayesian neural networks for nonlinear time series forecasting

In this article, we apply Bayesian neural networks (BNNs) to time series analysis, and propose a Monte Carlo algorithm for BNN training. In addition, we go a step further in BNN model selection by putting a prior on network connections instead of hidden units as done by other authors. This allows us to treat the selection of hidden units and the selection of input variables uniformly. The BNN model is compared to a number of competitors, such as the Box-Jenkins model, bilinear model, threshold autoregressive model, and traditional neural network model, on a number of popular and challenging data sets. Numerical results show that the BNN model has achieved a consistent improvement over the competitors in forecasting future values. Insights on how to improve the generalization ability of BNNs are revealed in many respects of our implementation, such as the selection of input variables, the specification of prior distributions, and the treatment of outliers.