Search

International audience; This paper assesses the impact of remittances and climate variability on households' food security in Burkina Faso. It relies on an original database from the World Bank survey on migration and remittances conducted in 2009. A principal component analysis allows elaborating of a food security index. The Standardised Precipitation and Evapotranspiration Index (SPEI) makes it possible to characterise the climate situation throughout the country relative to long-term records. The econometric results corroborate the harmful effect of deteriorating climatic conditions on households' food security. Remittances, however, enhance food security and dampen the negative impact of the SPEI on food security. These findings are robust to a potential endogeneity bias using distance from migrants' households to railway lines and migrant education as instruments.; Cet article évalue l’impact des transferts des migrants et de la variabilité climatiquesur la sécurité alimentaire des ménages au Burkina Faso. Il s’appuie surune base de données originale construite à partir de l’enquête 2009 de la Banquemondiale sur les migrations et les transferts. Une analyse en composantes principalespermet d’élaborer un indice de sécurité alimentaire. L’indice standardisé deprécipitation et d’évapotranspiration (SPEI) caractérise la situation climatique dansles différenes régions. Les résultats économétriques corroborent l’effet négatif dela détérioration des conditions climatiques sur la sécurité alimentaire des ménages.En revanche, les transferts de fonds renforcent la sécurité alimentaire et atténuentl’effet négatif du SPEI sur la sécurité alimentaire. Ces résultats sont robustes à unbiais d’endogénéité potentiel des transferts en utilisant la distance au chemin defer des ménages et l’éducation des migrants comme instruments.

International audience; This literature review covers recent works dealing with the ultimate causes of economic development. The article first considers the driving role of the rules of the social game. Then the focus shifts to the role of historical circumstances. Finally, we seek to understand how geography can influence development paths. These studies are multidisciplinary and use new databases. The results should not imply that there is a cultural, historical or geographical determinism. That is, historical contingencies and economic policy decisions can foster lasting changes in development trajectories.

The paper aims to highlight the impact of youth resentment on violent conflicts in Burkina Faso. This work takes advantage of one of the latest nationwide UNICEF-sponsored surveys conducted in Burkina Faso before some parts of the country became inaccessible because of attacks. Among other information, this survey collected data on youth resentment towards the ability of their household to fulfill their needs. This resentment is closely related to perceived relative deprivation. We merge this survey into an original dataset that gathers data from the Armed Conflict Location and Event Data Project (ACLED), mining data from the MINEX project, and distances data computed using Burkina Faso’s roads information. The results of an event count model show that youth resentment explains the occurrence of conflicts. Moreover, the presence of mining companies, the remoteness from infrastructures, ethnic diversity, and polarization also significantly affect violence against civilians.

The paper aims to highlight the impact of youth satisfaction (or dissatisfaction) on the occurrence of violent conflicts in Burkina Faso. This work takes advantage of one of the latest nationwide UNICEF-sponsored survey conducted in Burkina Faso before some parts of the country became inaccessible because of attacks. Among other pieces of information, this survey collected data on youth resentment towards the ability of their household to fulfill their needs. Alongside data from the Armed Conflict Location and Event Data Project (ACLED), mining data from the MINEX project, and distances data computed using Burkina Faso's roads information are mobilized. Relying on the negative binomial regressor, our results show that youth resentment has a positive and significant impact on the occurrence of conflicts. Moreover, the presence of mining companies, the remoteness from infrastructures, ethnic diversity, and polarization also affect significantly the occurrence of violence against civilians.

This paper assesses the impact of remittances and climate variability on the food security of households in Burkina Faso. It draws from the World Bank 2010 survey on migration and remittances in Burkina Faso and uses a database from Burkina Faso’s Department of Meteorology regarding rainfall recorded in the ten weather stations throughout the country between 2001 and 2010. We build a food security index using principal component analysis that encompasses the accessibility and utilization dimensions of the concept. We also compute an inter-annual rainfall index and the latter is found to have a negative impact on food security. After controlling for potential endogeneity issues using distance variables and migrant characteristics as instruments, remittances are found to enhance food security. Results are robust to alternative measures of food security and alternative calculations of rainfall variability. The paper also highlights that remittances dampen the negative effect of rainfall variability on food security.

Nouvelle version mise à jour en avril 2020. Titre de la première version : Forest spirits. What we know - and don't know - about the effectiveness of policies against deforestation [Mars 2019]; Studies addressing the effectiveness of policies aiming at combatting deforestation have produced mixed results, showing no obvious and undebated correlations between a certain type of forest policy instrument and its success in preventing or deterring deforestation. Hence, why anti-deforestation policies succeed or not still remains unclear. This paper proposes a new reading grid of the effectiveness of anti-deforestation policies, by mapping the causal mechanisms at stake from its design to its objective of avoiding deforestation. 264 empirical evaluations are collected and reviewed in order to reveal the theory of change (ToC) that emerges from the current practice of forest policy evaluation. This mapping sheds light on the different causal steps necessary for anti-deforestation policies to be effective, and on the conditions at stake at the various stages of the causal chain, according to the existing literature. Doing so, it allows visualising the reasons for the success of anti-deforestation policies -or lack of, as per the literature corpus analysed. It also provides guidelines with regard to the elements of context to look at when designing and implementing such anti-deforestation policies. Finally, it exposes what is most researched by evaluators, as well as observes the main apparent evaluation blind spots.Our results highlight that the effectiveness of anti-deforestation policies is context-specific on political, economic, social, cultural, environmental and structural dimensions: The political willingness, the nature of available funding, the governance structure, the existence of forest-related traditions, the social and economic situation of local populations, and the nature and spatial scale of the deforestation drivers are elements that influence the success of the policy. They may play this role at different stages of the policy design and implementation: at decision-making stage, during its implementation, or when evaluating its results and outcomes. The majority of conditions reported from the literature concern the implementation phase, at the local level. In other words, according to evaluators, the success or failure of anti-deforestation policies mostly lies in its effective implementation in the field, i.e. in its ability to generate a social acceptance of and compliance to the policy rules. However, some studies also show that a successfully implemented policy does not avoid deforestation if the actual deforestation driver is not properly addressed, thus resulting in deforestation being displaced or unchanged. This underlines the importance of prior risk assessments and field studies to design an adapted policy instrument to combat deforestation.

Most recent studies addressing the effectiveness of policies aiming at combatting deforestation produce mixed results, showing no consistency between a certain policy design and its success in preventing or deterring deforestation. Hence, why anti-deforestation policies succeed or not remains unclear. Against this background, this paper studies the literature evaluating the effectiveness of anti-deforestation policies. 264 empirical evaluations are reviewed and synthetized in order to reveal the theory of change (ToC) that emerges from the current practice of forest policy evaluation. This allows visualising what is described in the forest policy evaluation literature in terms of the causal relations and the conditions at stake for a policy instrument to combat deforestation successfully. It also reveals those conditions that are under-researched. In other words, the paper depicts what makes forest policies successful in evaluators’ minds. Our results expose the context-specificity of the effectiveness of anti-deforestation policies and confirm the mixed-success that has been characterizing them. They also indicate that policy evaluators tend to focus on the policy implementation phase rather on than on its design, and correspondingly tend to focus their evaluations on downstream outcomes. More specifically, how implementation generates a social acceptance of and compliance to the policy rules is among the most commonly reported conditions. Likewise, results also indicate that conditions needed to deliver environmental outcomes differ from those needed to deliver social and economic outcomes, which suggests considering multi-purposes policies with care. Among most notable under-developed fields of investigation are the role of improved information on local populations’ behaviour with respect to deforestation, the effect of generating social and economic benefits on environmental outcomes, and the effectiveness of anti-deforestation policies in terms of final environmental impacts (biodiversity, carbon sequestration and watershed protection). These analyses aim at providing researchers with directions for research programming, as well as providing policy designers with indications about conditions for policy effectiveness.

This paper analyses the determinants of the optimal size of protected areas and what conducts neighboring effects. We investigate in which measure the infrastructure effect and the scarcity effect matter. We obtain several results. The size of protected area mainly depends on preferences toward forest, on the firms’ production costs and on the relation between municipalities. As far as total deforestation is concerned asymmetric regulation is better than no regulation. The infrastructure effect always leads to smaller protected areas than the scarcity effect. Under the infrastructure effect, centralized decisions do not always work in favor of larger protected areas than decentralized decisions contrary to the scarcity effect. We also show that decentralized decisions can reach the first best under the infrastructure effect without public intervention. A study of protected areas in the Brazilian Legal Amazônia corroborates our theoretical results.

International audience; While mitigation efforts in developed and emerging economies are necessary in order to meet ambitious climate targets, the international community strives to explore strategies to help the most vulnerable populations to cope with the short-term and long-term impacts of climate change. In the perspective of the 21st COP of the UNFCCC (Paris, December 2015), this Special Issue on ‘Climate change mitigation and adaptation in developing and transition countries’ addresses two complementary topical issues. On the one hand, migration – international and internal – and remittances are analyzed as adaptation strategies for vulnerable households and individuals. On the other hand, climate policies in emerging economies are examined in light of their distributional impacts for households and of the strategic issues they may raise. This special issue introduces five papers with a diversity of approaches, e.g., game theory, econometric modeling and computable general equilibrium (CGE) modeling.

Etudes & documents; Protected areas are increasingly used as a tool to fight against deforestation. This paper presents new evidence on the spillover effects that occur in the decision to deforest and the creation of protected areas in local administrative entities in Brazilian Legal Amazon over the 2001-2011 period. We also highlight the interdependence between these two decisions. We proceed in two steps. First, we assumed that protected areas are created to stop the negative effects of deforestation on biodiversity. In order to control for the non-random location of protected areas, biodiversity indicators are used as excluded instruments. This model is estimated using a spatial model with instrumental variables. Second, a simultaneous system of spatially interrelated cross sectional equations is used to take into account the interdependence between the decision to deforest and the creation of protected areas. Our results show (i) that deforestation activities of neighboring municipalities are complements and that (ii) there is evidence of leakage in the sense that protected areas may shift deforestation to neighboring municipalities. The net effect of protected areas on deforestation remains however negative; it is moreover stable across two sub-periods. Our results confirm the important role of protected areas to curb deforestation and thereby biodiversity erosion. Moreover, they show that strategic interactions deserve attention in the effectiveness of conservation policies.

Improving access to capital through credit and public spendings is an important step toward development and poverty alleviation. At the same time, deforestation-related activities, like agricultural expansion, can be seen as relying on natural capital, through the depletion of forest resources and the use of land in an extensive way. It is then important to better understand how a better access to capital influences the use of land as a natural capital. This paper assesses the relationship between financial development, public spendings and deforestation. Are they substitute or complement? Our econometric analysis shows that deforestation is positively correlated to access to credit and public spendings, which gives some evidence that natural capital is a complement to credit and public spendings.

Relationships between health and economic prosperity or economic growth are difficult to assess. The direction of the causality is often questioned and the subject of a vigorous debate. For some authors, diseases or poor health had contributed to poor growth performances especially in low-income countries. For other authors, the effect of health on growth is relatively small, even if one considers that human capital accumulation needs also health investments. It is argued in this paper that commonly used health indicators in macroeconomic studies (e. g. life expectancy, infant mortality or prevalence rates for specific diseases such as malaria or HIV/AIDS) imperfectly represent the global health status of population. Health is rather a complex notion and includes several dimensions which concern fatal (deaths) and non-fatal issues (prevalence and severity of cases) of illness. The reported effects of health on economic growth vary accordingly with health indicators and countries included in existing analyses. The purpose of the paper is to assess the effect of health on growth. The augmented Solow model is modified so as to account for human capital depreciation. It is argued that the latter is measured by the so-called disability-adjusted life year (DALY) that was proposed by the World Bank and the WHO in 1993. Income regressions are run on 129 countries over the 2000-2004's period, where the potential endogeneity of the health indicator is dealt for. The negative effect of poor health on development is not rejected thus reinforcing the importance of achieving MDGs.

Although widely studied, deforestation remains a topical and typical issue. The relationship between economic development and deforestation is still at stake. This paper presents a meta-analysis of Environmental Kuznets Curve (EKC) studies for deforestation. Using 71 studies, offering 631 estimations, we shed light on why EKC results differ. We investigate the incidence of choices made by authors (econometric strategy, deforestation measure, temporal coverage, geographical area, measure of economic development...) on the probability of finding an EKC. After a phase of work corroborating the EKC, we find a turning point after the year 2001. Building on our results, we conclude that the EKC story will not fade until theoretical alternatives will be provided.

This paper establishes a link between deforestation and credit cycles in Latin American countries. The latter exhibit rapid deforestation rates as well as macroeconomic instability that is often rooted in credit booms and crunches episodes: data available on the last years show a coincidence between higher macroeconomic instability and deforestation increases. This paper provides a theoretical explanation and econometric investigations of this phenomenon. A key ingredient of the model is the existence of two sectors: a modern agricultural sector and a subsistence one, which are hypothesised to catch the basic features of Latin American agricultural sectors. Agricultural production relies on three production factors: land, capital and labour. Agents clear forested areas in order to increase agricultural lands. Interest rates movements have an effect on agricultural decisions and thus on deforestation since they induce factor movements between the agricultural sectors. It is shown that deforestation occurs in response to interest rates increases or decreases primarily because of the irreversible character of forest conversion. Econometric tests are conducted on the 1948-2005 period on an exhaustive sample of Latin American countries. The database on deforestation is a compilation of FAO censuses and several measures of credit cycles are calculated as well. The main output of the paper is to evidence a link between credit cycles and deforestation. The results are robust to the introduction of usual control variables in deforestation equations.

The purpose of this paper is to highlight the detrimental impact of land tenure insecurity on deforestation in the Brazilian Amazon. It is related to recent controversies about the detrimental impact of land laws on deforestation, which seem to legitimize land encroachments. The latter is mainly the result of land tenure insecurity which is a key characteristic of this region and results from a long history of interactions between rural social unrest and land reforms or land laws. A simple model is developed where strategic interactions between farmers lead to excessive deforestation. One of the empirical implications of the model is a positive relationship between land tenure insecurity and the extent of deforestation. The latter is tested on data from a panel of Brazilian Amazon municipalities. The negative effect of land tenure insecurity proxied by the number of squatters on deforestation is not rejected when estimations are controlled for the possible endogeneity of squatters. One of the main policy implications is that ex post legalizations of settlements must be accompanied by the enforcement of environmental obligations.

Relationships between health and economic prosperity or economic growth are difficult to assess. The direction of the causality is often questioned and the subject of a vigorous debate. For some authors, diseases or poor health had contributed to poor growth performances especially in low-income countries. For other authors, the effect of health on growth is relatively small, even if one considers that human capital accumulation needs also health investments. It is argued in this paper that commonly used health indicators in macroeconomic studies (e. g. life expectancy, infant mortality or prevalence rates for specific diseases such as malaria or HIV/AIDS) imperfectly represent the global health status of population. Health is rather a complex notion and includes several dimensions which concern fatal (deaths) and non-fatal issues (prevalence and severity of cases) of illness. The reported effects of health on economic growth vary accordingly with health indicators and countries included in existing analyses. The purpose of the paper is to assess the effect of health on growth, by using a global health indicator, the so-called disability-adjusted life year (DALY) that was proposed by the World Bank and the WHO in 1993. Growth convergence equations are run on 159 countries over the 1999-2004’s period, where the potential endogeneity of the health indicator is dealt for. The negative effect of poor health on economic growth is not rejected thus reinforcing the importance of achieving MDGs.

Dans cet article, nous examinons les conséquences de l'insécurité foncière sur la croissance économique au Brésil à partir d'un modèle dynamique d'une économie comportant un secteur agricole et un secteur manufacturier. La terre est à la fois un facteur de production fixe et spécifique au secteur agricole et un support d'épargne alternatif au capital employé dans le secteur manufacturier. L'arbitrage entre la détention de capital ou de terre dépend de coûts de transaction spécifiques à l'actif foncier, qui résultent de l'insécurité foncière. L'insécurité foncière entraîne une baisse du prix de la terre et une modification de la composition de l'épargne favorable au capital. Le modèle permet donc d'établir deux restrictions : un effet négatif de l'insécurité foncière sur le prix de la terre et un effet positif sur la croissance. Ces deux restrictions sont testées sur des données de panel pour les 27 états de la fédération brésilienne. L'insécurité foncière est approchée par le nombre de squatters. Les résultats économétriques n'invalident pas le modèle.

The aim of this paper is to highlight an aspect of devaluation that is generally ignored in the literature, namely its positive impact on the integration of domestic markets of tradable goods. The analysis applies to cattle markets in Burkina Faso where cattle is both a tradable and a capital good that can be held inter-temporally. We develop an exogenous switching regime regression model consistent with spatial and inter-temporal arbitrage conditions which categorise markets in two regimes: autarkic and integrated. When markets are autarkic, prices follow a random walk. Conversely, when two markets are integrated, their prices are closely correlated. The switching between the two regimes is driven by transaction costs which are supposed to be a function of the real effective exchange rate, among other variables. Devaluation is shown to have a negative impact on real transaction costs and thus to promote cattle market integration. Copyright 2005, Oxford University Press.

This article investigates whether protected areas are efficient instruments against deforestation in the Brazilian Amazonia. A Dynamic Spatial Durbin Model taking into account both the location bias and the spatial spillover effects between municipalities allows to assess the impact of the different types of protected areas (integral protected areas, sustainable protected areas and indigenous lands) on deforestation. We show that deforestation decisions are strategic complements. The econometric results differ according to the type of protected area. It is shown that: i) integral protected areas and indigenous lands allow for reducing deforestation; ii) sustainable use areas do not help to reduce deforestation; and iii) the spillover effects generated by integral protected areas and indigenous lands lead a reduction in deforestation in their vicinity. A 10% increase in the surface area of integral protected areas (indigenous lands) allows an estimated 9.32 sq. km (10.08 sq. km) of avoided deforestation.