Jean-Louis Arcand

Teacher training and HIV/AIDS prevention in West Africa: regression discontinuity design evidence from the Cameroon

We assess the impact on teenage childbearing as well as student knowledge, attitudes, and behavior of a typical HIV/AIDS teacher training program in the Cameroon. Applying a regression discontinuity design identification strategy based on the key administrative criterion that determined program deployment, we find that 15-17 year old girls in teacher training schools are between 7 and 10 percentage points less likely to have started childbearing, an objective proxy for the incidence of unprotected sex. They are also significantly more likely to have used a condom during their last sexual intercourse. For 12-13 year old girls, the likelihood of self-reported abstinence and condom use is also significantly higher in treated schools, while the likelihood of having multiple partners is significantly lower.

Does Community Driven Development Work? Evidence from Senegal

This brief summarizes the results of a gender impact evaluation study, entitled Does community driven development work? Evidence from Senegal, conducted between 2003 and 2005, in Senegal. The study observed that using a variety of estimation procedures, including instrumental variables, and working at different levels of aggregation, we find statistically significant and quantitatively important effects of the program on access by villagers to clean water and health services, as well as on standard measures of child malnutrition. The program increases per capita household expenditures by 65 percent. The program significantly improves access to clean water and health facilities. The program also significantly reduces the prevalence of underweight and stunted children and significantly increased household expenditure per capita and nutrition of children. The number of female villagers on the council is increasing with the likelihood of a village receiving a completed project. Funding for the study derived from the World Bank.

The Impact of Land Mines on Child Health: Evidence from Angola

This article estimates the causal impact of land mines on child health in Angola, controlling for conflict exposure. Our identification strategy is based on the geography of the Angolan civil war. We posit that distance between communes and rebel headquarters is an exogenous driver of land mine contamination. We find that land mine intensity is positively correlated with the distance to a set of rebel headquarters. Instrumental variables estimates, based on two household surveys and the Landmines Impact Survey, indicate that land mines have large and negative effects on weight-for-age and height-for-age. We discuss our results with respect to the costs and benefits of land mine clearance, as well as the long-term costs of early malnutrition. We also compare the magnitude of our estimates with those of related studies on the impact of conflict on child health.

Disequilibrium dynamics during the great depression

Abstract We test the flexibility of wages and prices in the U.S. before World War II using a simple two-market disequilibrium model. We test the model for four different tatonnement adjustment mechanisms and we find that the equilibrium restriction is strongly rejected in all cases. Hausman specification tests reject the equilibrium restriction but do not reject three of the disequilibrium specifications. Parameter estimates imply that the persistence of the Great Depression is not attributable to nominal rigidities but was caused by the system becoming dynamically neutral. We compute estimates of excess aggregate demand from 1892 to 1940 and find that a model in which adjustment obtains in prices in the goods market and in quantities in the labor market provides the best description of the data.

The Making of a (vice-) President: Party Politics, Ethnicity, Village Loyalty and Community-Driven Development

African politics are often said to be dominated by ethnic divides, with the ensuing policies implemented by leaders being based almost exclusively on their ethnic power base. In this paper, we demonstrate that the identity of leaders matters for the attribution of development projects in the context of one of the largest Community-Driven Development (CDD) programs in Senegal. After showing that leadership matters, we consider its determinants by focusing on those factors that determine who becomes president and vice-president of a Conseil rural, the smallest administrative unit in Senegal, and which is elected by universal suffrage. We also consider the link between power in the Conseil rural and that in the Conseil de Concertation et de Gestion (CCG), an assembly coopted by the Conseil rural president that is typical of local institutions set up in the context of CDD programs, and which is responsible for the attribution of development projects to individual villages. Using a unique dataset, we show that ethnicity plays almost no role in determining who becomes president (or vice-president) of a Conseil rural and vice-president of the CCG, while party politics, age, political experience, village loyalty, and educational and professional qualifications do. Our results highlight the crucial importance, in terms of development policy, of the local political institutions that are often created alongside CDD programs.

Maternal Height and the Sex Ratio

Abstract This paper tests the generalized Trivers Willard hypothesis in the spirit of Kanazawa (2005), which predicts that parents with heritable traits that increase the relative reproductive success of males compared to females will have relatively more male than female offspring. We test whether taller mothers are more likely to have a male first-born using data on 400,302 mothers in a sample of Demographic Health and Surveys (DHS) from 46 developing countries. Despite using a plethora of statistical models that take into account the multi-level structure of the data, we find no strong evidence in favor of the hypothesis between and within communities, as well as on a country-by-country basis. Conversely, Andrews (1989)s inverse power calculations suggest that the absence of a statistically small effect cannot be rejected.