|
on Discrete Choice Models |
| By: | Marie-Hélène Durand (DIAL, IRD, Paris) |
| Abstract: | (english) This cross-country study examines the determinants of school attendance through household surveys conducted the same year in seven West African countries. The aim of this comparison, based on the relatively homogeneous population of urban households living in the capitals, is to look at national differences. We used a tree-level logit model allowing to decompose the unobserved heterogeneity between a within family effect, a between families effect and a neighbourhood effect. The results show that the generalized pattern that explains the out of schooling for all these countries is the discrimination within the family towards some children, namely girls. Although this study has been based on comparable populations with similar rates of schooling the disparities in the impact of household characteristics on child schooling among the countries highlight the importance of the economical and cultural context. Children living in the poorest households are less likely to go to school except for great size families, which seems to indicate different types of family behaviour to cope with limited resources. Large households in Dakar tend to treat their children in a more egalitarian way regarding school attendance but their well being depends on the family income. On the contrary in Cotonou where families are of small size, children of very poor families have the same probability to go to school as wealthier ones and it is above all due to discrimination between children within the family that some of them do not attend school. _________________________________ (français) Cette étude analyse les déterminants de la non-scolarisation des enfants à partir d’enquêtes ménage réalisées la même année dans sept pays francophones ouest africains. La comparaison, restreinte à la population relativement homogène des ménages urbains de la capitale, a pour objectif d’évaluer l’existence de dissimilarités nationales. Nous utilisons un modèle hiérarchique à composantes inobservables permettant de décomposer l’hétérogénéité inobservée entre un effet de discrimination individuelle à l’égard de certains enfants au sein des familles, un effet de demande d’éducation de la part des familles et un effet de voisinage relevant soit de phénomènes d’interactions sociales ou de déficience des infrastructures dans le quartier de résidence des familles. Pour l’ensemble des pays, les résultats montrent que c’est surtout la discrimination à l’encontre de certains enfants au sein de la famille, essentiellement les filles, qui explique l’exclusion de l’école. Bien que cette étude ait été réalisée sur des populations comparables avec des taux de scolarisation moyens assez similaires, il existe entre les différentes capitales une certaine disparité des déterminismes familiaux de la scolarisation des enfants montrant l’importance du contexte économique ou culturel plus général. Les impacts de la pauvreté et de la taille des familles sont alternatifs et semblent relever de deux modes de gestion des contraintes de ressources de la part des familles. Les enfants qui vivent au sein de familles dakaroises sont traités de façon plus égalitaire qu’ailleurs mais leur bien-être est aussi plus qu’ailleurs sensible au revenu de ces familles. Par contre à Cotonou, c’est essentiellement en raison de discriminations qu’une partie des enfants n’est pas scolarisée. |
| Keywords: | School attendance, household survey, West Africa, random coefficient models, Scolarisation, enquêtes ménage, Afrique de l’Ouest, Modèle à coefficients aléatoires. |
| JEL: | C25 O12 O55 I21 |
| Date: | 2006–01 |
| URL: | http://d.repec.org/n?u=RePEc:dia:wpaper:dt200602&r=dcm |
| By: | Mette Gørtz (Department of Economics, University of Copenhagen) |
| Abstract: | This paper discusses the determinants of the retirement decision and the implications of retirement on economic well-being. The main contribution of the paper is to formulate the role of individual heterogeneity explicitly. We argue that individual heterogeneity in 1) productivity of market work versus housework, 2) preferences for leisure compared to consumption, and 3) marginal utility of wealth, is correlated with the retirement decision. Based on US consumption and time use data for 2001 and 2003 from the Consumptions and Activities Mail Survey (CAMS), we study the patterns of individual choices of expenditure, household production and leisure for people in and around retirement. The unobserved individual heterogeneity factor is isolated by comparing cross-sectional evidence and panel data estimates of the effects of retirement on consumption and time allocation. Based on cross-section data, we can identify a difference in consumption due to retirement status, but when the panel nature of the data is exploited, the effect of retirement on consumption is small and insignificant. Moreover, the analyses point at a large positive effect of retirement on household production. Our results therefore contribute to the discussion of the so-called retirement-consumption puzzle. Many analyses of the retirement-consumption drop assume that the retirement decision is exogenous. However, the individual decision on when to retire may depend on expected changes in consumption and time allocation. This suggests that the retirement decision is endogenous. To test this, we apply an instrumental variables method in the treatment effects tradition. |
| Keywords: | retirement; consumption; household production; heterogeneity |
| JEL: | C23 D91 J14 |
| Date: | 2006–01 |
| URL: | http://d.repec.org/n?u=RePEc:kud:kuieca:2006_01&r=dcm |
| By: | Giovanni B. Concu (Risk and Sustainable Management Group, University of Queensland) |
| Abstract: | This article tests for the effect of distance on non-use values using a Choice Modelling (CM) experiment. Estimating a distance decay relationship for non-use values (NUVs) is important because it would define the market area for an environmental good, i.e. identify the limits for aggregating individual benefit estimates. In contrast to the common definition of NUVs as non-usersÕ values, the CM experiment designs the environmental attributes so that NUV changes can be disentangled from Use Value (UV) changes. The experiment also allows for testing different specification of the distance covariates. Data are obtained from a geographically representative sample. Results show that NUVs do not depend on distance. Aggregation of NUVs is based on income and individualsÕ environmental attitudes. |
| Keywords: | choice modelling, non-use values, aggregation, distance, geographical sampling. |
| JEL: | Q51 Q58 |
| Date: | 2005–12 |
| URL: | http://d.repec.org/n?u=RePEc:rsm:murray:m05_6&r=dcm |
| By: | Giovanni B. Concu (Risk and Sustainable Management Group, University of Queensland) |
| Abstract: | This paper describes a Choice Modelling experiment set up to investigate the relationship between distance and willingness to pay for environmental quality changes. The issue is important for the estimation and transfer of benefits. So far the problem has been analysed through the use of Contingent Valuation-type of experiments, producing mixed results. The Choice Modelling experiment allows testing distance effects on parameters of environmental attributes that imply different trade-offs between use and non-use values. The sampling procedure is designed to provide a Ògeographically balancedÓ sample. Several specifications of the distance covariate are compared and distance effects are shown to take complex shapes. Welfare analysis also shows that disregarding distance produces under-estimation of individual and aggregated benefits and losses, seriously hindering the reliability of costbenefit analyses. |
| Keywords: | choice Modelling techniques, distance, aggregation, sampling, functional forms. |
| JEL: | Q51 Q58 |
| Date: | 2005–12 |
| URL: | http://d.repec.org/n?u=RePEc:rsm:murray:m05_7&r=dcm |
| By: | Fazia Pusterla; Laura Resmini (ISLA, Universita' Bocconi, Milano) |
| Abstract: | In this paper, we utilize firm-level data on foreign firm manufacturing plants in four Central and Eastern European Countries, in order to trace their location and estimate the determinants of their choice processes. At this purpose, we use a nested logit model as developed by McFadden (1974). Several reasons explain the need for mapping FDI. First of all, this would help to understand the real competitiveness of regions and countries in providing location advantages able to complement MNEs’ organisational and internalisation advantages. Secondly, it would directly show the distribution of the benefits usually associated with FDI. Finally, an understanding of FDI location decisions is important for policy makers who believe that MNEs may offer a positive contribution to the economic development of disadvantage areas and thus correct potential regional imbalances. Main results show that sector specific factors affect the choice of final location. These unobserved characteristics influence both the determinants and the structure of MNEs location choice process. |
| Keywords: | foreign direct investment, location choice, transition countries |
| JEL: | F23 R38 |
| Date: | 2005–06 |
| URL: | http://d.repec.org/n?u=RePEc:slp:islawp:islawp20&r=dcm |
| By: | Rob Euwals (CPB, The Hague, IZA, CEPR); Daniel J. van Vuuren (CPB, and Vrije Universiteit Amsterdam); Ronald P. Wolthoff (Vrije Universiteit Amsterdam) |
| Abstract: | In the early 1990s, the Dutch social partners agreed upon transforming the generous and actuarially unfair PAYG early retirement schemes into less generous and actuarially fair capital funded schemes. The starting dates of the transitional arrangements varied by industry sector. In this study, we exploit the variation in starting dates to estimate the causal impact of the policy reform on early retirement behaviour. We use a large administrative dataset, the Dutch Income Panel 1989-2000, to estimate hazard rate models for early retirement. We conclude that the policy reform induced workers to postpone early retirement. In particular, both the price effect (reducing implicit taxes) and the wealth effect (reducing early retirement wealth) are shown to have a positive impact on the early retirement age. Yet, we show that model specifications including the most commonly used financial incentive measures are open to further improvements, given that these are outperformed by a simple specification with dummy variables. |
| Keywords: | early retirement; intertemporal choice; duration analysis |
| JEL: | C41 D91 J26 |
| Date: | 2006–02–21 |
| URL: | http://d.repec.org/n?u=RePEc:dgr:uvatin:20060021&r=dcm |