My research centers on housing economics and affordability, with related work on institutions, inequality, and digital economy in developing economies.
This paper estimates the causal effect of social media on gender inequality. To tackle endogeneity, we use an instrumental variable approach with accessibility variables (cost of media use and internet coverage). Based on data from 164 countries between 2013 and 2021, we find that social media penetration has a diminishing effect on gender inequality. The results have passed tests for over-identification and are robust to alternative measures, changes in the sample and changes in the estimation method. Multiple channels through which social media impact on gender have been identified, including women’s political emancipation, education and human capital development, and enforcing the rule of law. From our heterogeneity analysis, it is evident that the causal effect on gender inequality is heightened in countries (i) with high human capital levels, (ii) with high-quality institutions, and (iii) in low-income and middle-income economies.
Sub-Saharan African (SSA) nations face a dual fiscal challenge: the need to increase tax revenue mobilization and to manage revenue instability. This paper examines the causal impact of institutional quality on tax revenue stability in SSA from 2000 to 2020. Using UNU-WIDER data and the System GMM technique, the study finds that institutional quality reduces tax revenue instability, with a stronger effect on indirect tax revenues than on direct taxes. This result is robust to changes in the measure of tax revenue instability, the measure of quality of institutions, the sample, and the time horizon. The effect is particularly pronounced in resource-rich countries and in countries experiencing economic growth. The article's results are relevant to tax policy, specifically the necessary adjustments to improve tax revenue stability through institutional quality.
This paper studies residential vacancy as a by-product of housing construction driven by the urbanization process. We study the effect of urban sprawl – measured as the amount of newly developed land per inhabitant – on vacancy rates from a panel of about 35,000 French municipalities from 1995 to 2019. Since urban sprawl is not randomly distributed across municipalities, we use average slope and soil stability as instrumental variables that are assumed to affect vacancy rates only through differentiated sprawling capacities. We find that median annual sprawl (+27 m2/inhab.) increases the average vacancy rate by +0.33 percentage points, equivalent to +2.85 vacant dwellings per municipality. This result passes the over-identification test and is robust to spatial spillovers between municipalities that are accounted for by data aggregation. Heterogeneity analysis shows that the estimated effect is more pronounced for short-term vacancies, for apartments rather than houses, for older dwellings, and in municipalities with declining employment or higher initial vacancy rates. Our results highlight the importance of taking household preferences into account when designing more effective public policies to tackle the housing affordability crisis.
Housing affordability has deteriorated markedly across OECD countries, yet the role of income inequality in shaping housing-market outcomes remains insufficiently understood. This paper examines whether increases in income inequality predict subsequent declines in housing affordability through a demand-composition channel, whereby rising income concentration shifts purchasing power toward high-income households and increases pressure on housing demand. Using an annual panel of 35 OECD countries from 2000 to 2021 and local projection methods, I trace the dynamic response of the house price-to-income ratio following changes in inequality. The results show that increases in inequality systematically precede declines in housing affordability. A 0.01 increase in the Gini coefficient is associated with a 0.85--1.1 percent increase in the house price-to-income ratio over the subsequent one to three years, with effects building gradually over time. The response is stronger when inequality is measured using top income shares and is amplified in high-credit and high-inflation environments, consistent with the proposed mechanism. Overall, the findings suggest that income inequality is an important predictor of future housing affordability pressures and highlight the interaction between distributional dynamics and macro-financial conditions in shaping housing-market outcomes.
Persistent housing vacancy reduces the effective supply of housing in many markets, where substantial numbers of dwellings remain unused despite growing housing demand. Although vacancy taxes and renovation subsidies have become central policy instruments, limited evidence exists on how property owners trade off financial incentives against the non-monetary barriers associated with housing reactivation. This paper investigates these trade-offs using a discrete choice experiment conducted among owners of vacant dwellings in France. Respondents evaluated hypothetical policy scenarios characterised by vacancy taxation, renovation requirements, subsidy schemes, and local housing market conditions. The analysis estimates the relative importance of these determinants and explores preference heterogeneity across owner characteristics. The results indicate that major renovation requirements constitute the principal barrier to housing reactivation. While higher vacancy taxes significantly increase the likelihood of bringing vacant dwellings back onto the market, their effect remains considerably smaller than the deterrent associated with extensive renovation needs. Owners' preferences also vary systematically according to socio-demographic characteristics, previous rental experiences, and awareness of existing policy instruments, highlighting the importance of behavioural and institutional factors in shaping housing reactivation decisions. These findings suggest that policies relying primarily on vacancy taxation are unlikely to mobilise the existing housing stock effectively. More successful strategies should combine fiscal incentives with measures that reduce renovation complexity, transaction costs, administrative burdens, and perceived risks. By providing new stated-preference evidence on owners' housing reactivation decisions, this paper contributes to the housing economics literature by identifying the relative importance of financial and non-financial barriers to housing reactivation and offers policy insights for improving the mobilisation of the existing housing stock.
Housing affordability has become a major challenge across OECD countries, yet little is known about the role of monetary policy frameworks in shaping housing outcomes. This paper examines the causal effect of inflation targeting on housing affordability in a panel of OECD countries over 1990--2020. We exploit the staggered adoption of inflation targeting and use entropy balancing with weighted least squares to address selection bias. Housing affordability is measured using the house price-to-income ratio, where higher values indicate lower affordability. We find that inflation targeting has a statistically significant effect on housing affordability, and the results remain robust to placebo and falsification tests, alternative treatment definitions, additional controls, and alternative samples excluding major financial crises. Heterogeneity and mechanism analyses indicate ...
This paper examines the relationship between income inequality and housing affordability in 35 OECD countries over the period 2000--2021. To address endogeneity concerns, we implement a two-stage least squares (2SLS) instrumental variables approach. Specifically, we use two instruments: tax progressivity, proxied by the marginal tax rate at 100% of the average wage, and R&D expenditures as a share of GDP. The results provide evidence consistent with the view that greater inequality worsens housing affordability by increasing the house price-to-income ratio. This effect appears to be stronger in low-growth environments, where weak income dynamics make it more difficult for households to absorb rising housing costs. We also examine several conditions under which this relationship becomes more severe. The findings suggest that the adverse effect of inequality is amplified by stronger housing price pressures, more expansionary credit conditions, and higher inflation, while the role of urbanization appears less systematic. The main results remain robust across alternative specifications, inequality measures, and subsample analyses. Overall, the paper suggests that housing affordability should be understood not only as a housing-market issue, but also in relation to broader distributional and macro-financial conditions.
This study investigates the dynamic and heterogeneous effects of land subdivision on parcel-level land prices, leveraging a unique micro-dataset that merges exhaustive cadastral and notarial records for the Bourgogne–Franche-Comté region and Haute-Marne department, over the period 1995–2017. Using a two-stage difference-in-differences (2S-DiD) design—tailored for staggered treatment timing and heterogeneous effects, the analysis identifies a substantial and robust price premium associated with formal lot division, both for undeveloped plots and developed plots. The results show that subdivision significantly increases the price per square meter, with the effect being most pronounced in high-demand, central, and rural locations. Event study models reveal that these price gains are immediate and persistent, with no evidence of anticipatory effects prior to subdivision. The magnitude of the premium is strongest in the tightest housing markets, for parcels near commune centres, and in rural areas. Furthermore, the study demonstrates that for undeveloped plots, the reduction in lot size outweighs the unit-price increase—thereby lowering total transaction costs and enhancing affordability—whereas on developed plots the effect on affordability appears negligible. These findings provide new insights into the mechanisms through which subdivision shapes land market dynamics and offer actionable guidance for policymakers seeking to enhance housing supply, affordability, and efficient land use.
This study explores residential vacancy in the Communauté de Communes Côte Ouest Centre Manche in Normandy, combining data from the "Zero Logement Vacant" platform (ZLV), the LOVAC database, and responses from a sample of owners of vacant dwellings. The theoretical section defines vacancy and the policy tools used for its reactivation, based on a review of recent literature. The results reveal a pattern of structural vacancy, particularly concentrated in town centers. Vacant homes are typically old and require major renovations. The main causes include renovation needs, high costs, complex property transitions, and fears related to rental risks. Policy recommendations include increasing renovation subsidies, streamlining administrative procedures, promoting rental intermediation programs, and conducting awareness campaigns.
This study examines the causal effect of monetary policy on income inequality in emerging economies using system GMM for dynamic panels, complemented by instrumented local projections (LP-IV) as an alternative identification strategy. The sample consists of 46 emerging economies from 2000 to 2018. The results indicate that tight monetary policies contribute to an increase in income inequality. It is important to note that these policies have a minimal impact on income distribution until the third year after their implementation, indicating a delayed effect on inequality. When considering the transmission channels, it is evident that inflation, exchange rates, and the percentage of credit granted as a proportion of GDP are effective tools for monetary policy to influence income distribution. The study's results are robust to sensitivity analyses that take into account changes in sample composition, time horizon, and inequality measurement methods, and are further corroborated by the alternative identification strategy. Further, heterogeneity analysis highlights that the impact of these policies on inequality depends on a number of factors, including the level of labour income, the existence of social protection policies and the condition of the economy.
This paper examines whether mobile money adoption contributes to lower mortality in low- and middle-income countries using a panel of 122 countries over the period 2002–2023. Employing an entropy-balancing, we find that mobile money adoption reduces mortality by about 9 percent relative to the control-group mean. The results remain robust across alternative estimators, specifications, placebo tests, outcome measures, and alternative indicators of mobile money development. We also find that the mortality-reducing effects of mobile money strengthen over time following adoption. We further explore the mechanisms underlying this relationship and find that financial inclusion, remittances, and health expenditures constitute important channels through which mobile money contributes to lower mortality. The effects are particularly pronounced in low-income, Sub-Saharan African, and less urbanized countries. These findings underscore the broader development benefits of digital financial technologies, extending beyond financial inclusion and poverty reduction to improvements in population survival and sustainable development.
Using a panel of 44 Sub-Saharan African countries (2001–2019), this paper estimates dynamic panel models to examine the relationship between illicit financial flows (IFFs) and the Human Development Index (HDI). Higher IFFs are found to be negatively associated with HDI, with the effect more pronounced in countries with weaker governance, greater exposure to IFFs, and lower human development — suggesting that institutional and developmental conditions shape vulnerability to illicit financial activity.
Digital financial inclusion has become a central pillar of development policy, yet little is known about its implications for land-use decisions and environmental sustainability. This paper examines how the adoption of mobile money affects forest loss across 122 developing countries over the period 2002--2023. Exploiting the staggered rollout of mobile money and using the two-stage difference-in-differences (DID2S) estimator, we find that adoption raises annual forest loss by approximately 0.15 percentage points—equivalent to roughly 40% of the sample mean forest-loss rate. The effect emerges gradually after adoption, is negligible in economies with limited agricultural dependence, is approximately twice as large in rural and forest-rich countries, and is attenuated by stronger institutions. Evidence on potential mechanisms suggests that mobile money relaxes financial constraints, stimulates agricultural expansion, and thereby increases pressure on forests. Our findings reveal an important development trade-off: while digital financial inclusion expands economic opportunities, it can also accelerate environmental degradation where forest governance is weak. More broadly, the paper shows that financial innovation reshapes the allocation of natural resources, highlighting that policies promoting financial inclusion may generate unintended environmental externalities unless accompanied by effective land governance and environmental institutions.
This paper examines whether digital financial infrastructure enhances resilience to climate-related and natural disasters through the adoption of mobile money in Sub-Saharan Africa. We combine data from the GSMA Mobile Money Deployment Tracker and the EM-DAT disaster database for 46 countries over the period 2002--2023. To account for staggered treatment timing and heterogeneous treatment effects, we employ a two-stage difference-in-differences estimator. We find that mobile money adoption significantly reduces disaster vulnerability and enhances resilience to disaster shocks. Event-study estimates reveal no evidence of differential pre-treatment trends and indicate that the effects strengthen over time following adoption. The findings are robust to alternative outcome transformations, the exclusion of major mobile-money adopters, alternative measures of mobile money adoption, and falsification tests. Heterogeneity analyses show that resilience gains are concentrated in countries with stronger institutions and are particularly pronounced for floods and droughts. We further find that mobile money adoption increases financial inclusion, suggesting that improved access to financial services is an important channel through which digital finance enhances disaster resilience. These findings provide new evidence that digital financial infrastructure strengthens resilience to climate and natural hazards in developing economies.