Predicting Household Poverty: A Methodological Note with a Kenyan Example

Household surveys provide data that is used for identifying and measuring the poverty status of households and individuals. However, carrying out such surveys is expensive, especially in poor developing countries. Thus it is important to make maximum use of the available survey data in developing countries, especially in sub‐Saharan Africa, where such data are expensive to collect and analyse. This paper develops a simple method for using poverty indices derived from survey data for a given year, to predict poverty rates for subsequent periods without having to conduct a new household survey. We illustrate the workings of the method with data from Kenyan household surveys for 1994 and 1997.     

Existence of Sparsely Supported Correlated Equilibria

We show that every N-player K ₁ × ... × K _N game possesses a correlated equilibrium with at least zero entries. In particular, the largest N-player K × ... × K games with unique fully supported correlated equilibrium are two-player games. We thank an anonymous referee for most useful comments. The first author acknowledges financial support from Spanish Ministry of Science and Technology, grant SEJ2004-03619, and in form of a Ramón y Cajal fellowship. The second author acknowledges support by the PASCAL Network of Excellence under EC grant no.506778, as well as from Spanish Ministry of Science and Technology and FEDER, grant BMF2003-03324. Both authors also acknowledge financial support from BBVA grant “Aprender a jugar.”     

DETERMINANTS OF SCHOOL ENROLMENT AND EDUCATION ATTAINMENT: EMPIRICAL EVIDENCE FROM KENYA1

This paper investigates the determinants of demand for schooling in Kenya. Probit and ordered probit methods are used to model enrolment and attainment respectively. The results show that child characteristics, parental education and other household characteristics, quality and cost of schooling are important determinants of demand for education services in Kenya. The results further show that girls would be more affected by policy changes than boys. The findings call for targeting in efforts to boost and sustain demand for schooling in Kenya. The study recommends immediate policy interventions focusing on improving quality of education and poverty alleviation.     

A Model of Occupational Choice Applied to Rural Kenya

Abstract: The paper studies occupational patterns in rural Kenya using a large cross-sectional data collected from farm households in seven districts. Consistent with the agrarian nature of the Kenyan economy, we find that 74 per cent of rural households have farming as their principal occupation. The remaining 26 per cent are engaged in nonfarm activities. The existing rural occupational structure is shown to be a consequence of rational career decisions of households. Education and proximity to market centers are the key factors in the transformation of occupational structure in rural Kenya. Résumé: Cet article analyse les schémas de choix professionnels dans les zones rurales du Kenya en se basant sur un vaste esnemble de données recueillies auprès des ménages ruraux de sept districts. Comme le laissait présager le caractère agricole de l'économie kényane, 74 pour cent des ménages ruraux ont l'agriculture pour activité principale; les 26 pour cent restants sont engagés dans des activités non agricoles. Il est démontré que la structure actuelle des choix de professions dans les zones rurales découle de décisions rationnelles prises par les ménages. L'éducation et la proximité des centres de commercialisation sont les facteurs prépondérants qui conditionnent la transformation des schémas professionnels dans les zones rurales du Kenya.     

Financing Medical Care through Insurance: Policy Lessons from Household‐ and Community‐Level Analysis in Kenya

Medical insurance is an important feature of a health care system in which patients pay user charges to get medical treatment. Without insurance, many people would not afford acceptable care in a fee‐for‐service system. Since health is a merit good, making insurance broadly available in communities is a major policy issue in countries where user fees finance medical treatments. The paper analyses data from facility and household surveys in Kenya and shows that policies which popularize medical insurance can be inefficient because there exist community and household level factors that inhibit its use. The results further reveal substantial variations in the way the variables that influence the use of insurance affect different population sub‐groups. In this regard, it is important to ensure that vulnerable groups in communities are not excluded from insurance schemes in which they invest. The policy value of the paper is to call attention to factors such as place of residence, gender, income, education, community institutions, transaction costs and facility quality – that hinder or facilitate use of medical insurance so that these factors can be considered when institutionalizing insurance in communities. We note that using the community as a unit of analysis in field surveys may strengthen policy conclusions usually obtained from such surveys. L’assurance médicale constitue un élément important de tout système de soins de santé dans lequel les patients paient pour obtenir des soins médicaux. Sans assurance, de nombreuses personnes seraient dans l’incapacié de faire face aux coûts d’un système de paiement à l’acte. La santéétant un bien tutélaire, généraliser l’accès à l’assurance au sein des communautés est une question politique majeure dans les pays où les soins médicaux sont financés par les utilisateurs. S’appuyant sur l’analyse des données d’enquêtes dans les institutions sanitaires et auprès des ménages kenyans, l’article montre que les politiques qui visent à généraliser l’assurance médicale peuventêtre inefficaces parce que certains facteurs aux niveaux des communautés et des ménages en inhibent l’utilisation. Il révèle également de fortes disparités dans la faµon dont les variables qui influencent l’utilisation de l’assurance affectent différentes sous‐sections de la population. A cet égard, il est important de veiller à ce que les groupes vulnérables ne soient pas exclus des plans d’assurance dans lesquels les communautés investissent. S’agissant des politiques à mettre en oeuvre, l’intérêt de cet article est qu’il attire l’attention sur des facteurs tels que le lieu de résidence, le genre, le revenu, l’éducation, les institutions communautaires, les coûts de transaction et la qualité des infrastructures, lesquels entravent ou facilitent l’utilisation de l’assurance médicale, et pourraient être pris en compte lorsque l’on institutionnalise l’assurance dans les communautés. Il est à noter qu’utiliser la communauté comme unité d’analyse des enquêtes de terrain peut renforcer les conclusions qu’engendrent généralement ce type d’enquêtes.     

A comment on Kenyan migration movements

The author comments on an article by Richard E. Bilsborrow, John O. Oucho, and John W. Molyneaux concerning the economic, geographic, and ethnic factors affecting migration in Kenya. In particular, he criticizes the model of migration developed in the original study.     

Some consequences of the unknottedness of the Walras correspondence

Two basic properties concerning the dynamic behavior of competitive equilibria of exchange economies with complete markets are derived essentially from the fact that the Walras correspondence has no knots.     

An innovative feature selection method for support vector machines and its test on the estimation of the credit risk of default

Support vector machines (SVM) have been extensively used for classification problems in many areas such as gene, text and image recognition. However, SVM have been rarely used to estimate the probability of default (PD) in credit risk. In this paper, we advocate the application of SVM, rather than the popular logistic regression (LR) method, for the estimation of both corporate and retail PD. Our results indicate that most of the time SVM outperforms LR in terms of classification accuracy for the corporate and retail segments. We propose a new wrapper feature selection based on maximizing the distance of the support vectors from the separating hyperplane and apply it to identify the main PD drivers. We used three datasets to test the PD estimation, containing (1) retail obligors from Germany, (2) corporate obligors from Eastern Europe, and (3) corporate obligors from Poland. Total assets, total liabilities, and sales are identified as frequent default drivers for the corporate datasets, whereas current account status and duration of the current account are frequent default drivers for the retail dataset.     

Trade union membership and earnings in Kenyan manufacturing firms

This study analyses the effect of trade unions on male earnings in the Kenyan manufacturing sector using a regression method, which takes into account endogeneity of the union status of workers. In contrast to earlier studies of the Kenyan labour market that report a negative effect of unions on wages, a positive effect is found. The study further shows that elite workers tend to abstain from union membership.