Marco Méndez-Atienza

This paper evaluates Mexico City’s voluntary gun buyback program, Sí al desarme, sí a la paz, launched in January 2019. Using state-month data for 2017–2023, we estimate program effects with synthetic control and synthetic difference-in-differences designs. We find a statistically significant decline in our composite outcome, which combines recorded firearm-related crimes with unclassified deaths to conservatively guard against possible outcome reclassification. The estimated decline is equivalent to roughly 17% of the composite outcome’s pre-treatment mean. Because unclassified deaths need not be firearm-related, this specification prioritizes robustness to misclassification over direct substantive interpretation. The more interpretable disaggregated estimates show that the effect is concentrated in intentional firearm injuries, while effects on firearm homicides are statistically indistinguishable from zero across specifications. An illustrative Phase 1 calculation implies approximately 11,354 MXN ($560 USD) in spending per estimated averted event in the composite outcome. We also find no detectable substitution into non-firearm injuries. Mechanism evidence indicates that the program primarily collected civilian handguns and is consistent with a situational opportunity-reduction channel in civilian settings: reducing immediate access to firearms can limit escalation and lower the likelihood that routine conflicts translate into gun injuries, rather than directly disrupting organized-crime arsenals. Robustness checks, including placebo timing tests around COVID lockdown and triangulation with INEGI vital statistics, suggest that the results are not driven by reporting or reclassification. The findings contribute the first causal evidence on a voluntary gun buyback in Mexico and, by focusing on non-lethal firearm injuries rather than solely on homicides, identify a margin of impact that prior Latin American evaluations have not examined.

This paper evaluates Mexico City’s Utopías, a network of large community complexes designed to activate situational crime-prevention mechanisms. Using monthly georeferenced crime data (2017–2024) and a staggered difference-in-differences estimator, we compare AGEBs within 300 metres of each Utopía to not-yet-treated areas. The results show clear heterogeneity: significant reductions in intentional injuries, simple drug possession, mass event theft, public-service vehicle theft without violence, and RTP robbery with violence, but increases in shoplifting, property damage, intimidation, violent vehicle theft, pickpocketing on public transport, and homicide by beating. These mixed effects are consistent with Situational Crime Prevention predictions about differential mechanism activation under conditions of inequality and institutional fragility, as spatial activation strengthens guardianship while simultaneously expanding opportunities for opportunistic crime. The study shows how environmental design performs under inequality and highlights the need for sustained management to mitigate criminogenic spillovers.

This paper analyzes the spatial distribution of emergency calls reporting public brawls and disturbances (riña) around Mexico City Metro stations in 2019. The outcome is reported public disorder, 911 calls registered by the city's C5 system, rather than verified incidents, and the analysis is descriptive and associational throughout. Using global and local spatial autocorrelation statistics and Getis–Ord hotspot analysis under transit-network and geographic spatial-weight structures, and across 100, 200, and 400 m buffers, the paper reports three main findings. First, the geography of reported disorder is denominator-dependent: call counts cluster in the central commercial core, while calls per Metro rider concentrate in a compact north-eastern corridor around Lines 4, 5, and B. Second, this per-rider corridor is not merely a denominator artifact: hotspot stations show above-average call counts and above-average calls per surrounding resident, although low ridership amplifies their per-rider rate relative to the system average. Third, aggregate neighborhood marginalization does not appear to be the main correlate of hotspot intensity. The population-weighted 2020 urban marginalization index at the colonia level is weakly to moderately negatively associated with hotspot intensity, and hotspot catchments contain no population in high-marginalization colonias. Instead, specific dimensions of residential vulnerability, especially overcrowding, limited digital access, and educational deficit, co-vary more consistently with the corridor. Because the outcome depends on reporting behavior as well as on underlying events, these associations characterize reported disorder, not its full latent distribution. The findings suggest that reported station-area disorder is better understood as the spatial overlap of mobility exposure, urban activity, and selected dimensions of residential vulnerability than as a simple function of aggregate neighborhood deprivation.

This paper examines whether administrative boundaries generate discontinuities in crime and public safety by exploiting the border between Mexico City and the State of Mexico. Although neighborhoods located on opposite sides of the boundary can be geographically close and share similar urban and socioeconomic characteristics, they are governed by different state and local institutions, police forces, public security strategies, and criminal justice systems. We propose a geographic regression discontinuity design that compares neighborhoods located immediately on either side of the administrative border, using distance to the boundary as the running variable. The analysis will combine geocoded crime or emergency-call data with neighborhood-level demographic and socioeconomic characteristics to estimate whether crossing the administrative boundary is associated with a discrete change in the incidence and composition of crime. We will examine heterogeneous effects across crime categories and different segments of the border, as well as potential mechanisms related to policing, institutional capacity, and urban characteristics. By focusing on highly comparable locations separated by an administrative boundary, the study seeks to provide evidence on the extent to which differences in public security institutions and policies translate into local differences in crime within a continuous metropolitan area.

This paper examines the relationship between agricultural cultivation and criminal activity using high-resolution satellite imagery. We use remotely sensed data to identify and map cultivated areas at a fine spatial scale and combine these measures with geocoded information on crime and violence. This approach allows us to study whether the spatial distribution and expansion of cultivation are systematically associated with changes in criminal activity, including whether particular types of cultivation generate distinct patterns of violence and territorial control. We further examine spatial and temporal heterogeneity in this relationship and explore potential mechanisms linking agricultural production to criminal activity, including competition over land, control of productive territories, and the presence of organized criminal groups. By integrating satellite-based measures of land use with granular crime data, the study seeks to provide new evidence on the geography of illicit activity and on the value of remote sensing for studying criminal behavior in settings where conventional administrative data provide only a partial picture.

This article examines whether the 2026 FIFA World Cup changed the frequency and composition of emergency calls in Mexico. We use administrative data from the national 911 emergency call system and exploit temporal and geographic variation associated with the organization of World Cup matches in host cities. Our empirical strategy compares trends in emergency calls in exposed and non-exposed locations before, during, and after the tournament. We analyze both total call volume and specific categories of incidents, including domestic violence, fights, injuries, property crime, and disturbances of public order. The study contributes to the literature on mega-events, crime, interpersonal violence, and demand for emergency services.

Does the productive structure of a recipient country affect the effectiveness of World Bank lending? This paper provides causal evidence that higher economic complexity, a summary measure of the diversity and sophistication of a country's export basket, significantly improves project outcomes as evaluated by the World Bank's Independent Evaluation Group (IEG). We instrument lagged Economic Complexity Index (ECI) values with a Bartik-style shift-share variable constructed from the trade-partner complexity of each country's pre-sample (1995–2000) export network. The first-stage F-statistic exceeds 235, ruling out weak-instrument concerns. Our preferred IV specification, using a PCA-based composite effectiveness index that combines IEG project outcomes with Bank performance ratings, yields a coefficient of 0.785 (SE = 0.400, p = 0.052), rising to 1.022 (SE = 0.435, p < 0.05) under a parsimonious control set. Heterogeneity analysis reveals that the effect is concentrated in Development Policy Loans (DPLs), consistent with an absorptive capacity mechanism: institutional and policy reforms financed by DPLs require the distributed organizational capabilities that ECI proxies, whereas investment projects impose more modular capacity requirements. These findings suggest that productive structure is a first-order determinant of development lending effectiveness, with implications for project design and country selectivity.

This paper studies the relationship between economic complexity and the gender wage gap across Mexican states. We combine microdata from the National Survey of Occupation and Employment (ENOE) with the state-level Economic Complexity Index (ECI) published by Data México to construct a semiannual panel beginning in 2015. We estimate both raw and adjusted gender wage gaps, with the latter obtained through an Oaxaca–Blinder decomposition. We then examine whether changes in economic complexity are systematically associated with changes in the wage gap, using state and period fixed effects. We also study four possible mechanisms: sectoral composition, formality, occupational segregation, and human capital. Our main hypothesis is that greater economic complexity is associated with a smaller gender wage gap.