CAF - Women’s representation and the governance of the commons

C A F - W O R KI N G PA P E R # 2 0 2 4 / 0 1

D e c e m b e r 1 9 , 2 0 2 5 Women’s representation and the governance of the commons Sandra Aguilar-Gomez1 | Francesca Eustacchi 2 | Laura Aguirre-Duran 3 1Tecnologico de Monterrey, Mexico City, Mexico. saguilargomez@tec.mx 2University College Dublin, School of Economics, Dublin, Ireland. 3Universidad de los Andes. Roughly half of the world’s land is held collectively by indigenous and local communities, often governed through common-property institutions. Despite national efforts to expand political inclusion, women remain largely excluded from these local bodies. This paper examines how gender inclusion affects the management of common property resources. We exploit a 2016 reform to Mexico’s Agrarian Law, which mandated gender quotas in agrarian committees, and estimate the causal effects of women’s participation on environmental, economic, and organizational outcomes using a staggered eventstudy design. Drawing on data from over a decade of agrarian community records (2012–2023), we find that greater participation by women reduces annual deforestation rates by about 6 percent without diminishing local economic activity. Inclusion also reshapes communities’ interactions with the state: applications to empowerment-oriented programs rise, while applications to production subsidies decline. Although aggregate application volumes remain constant, approval rates fall in communities with more women in office, suggesting persistent institutional barriers. These findings reveal how formal inclusion of women can reshape the governance of shared resources, but highlight systemic barriers faced by women in community leadership positions.

JEL codes: D71, J16, O13, Q23

K E Y W O R D S Gender quotas, Political representation, Collective governance, Deforestation, Staggered difference-in-differences, Mexico. Small sections of text that are less than two paragraphs may be quoted without explicit permission as long as this document is acknowledged. Findings, interpretations and conclusions expressed in this publication are the sole responsibility of its author(s) and cannot be, in any way, attributed to CAF, its Executive Directors or the countries they represent. CAF does not guarantee the accuracy of the data included in this publication and is not, in any way, responsible for any consequences resulting from its use. ©2025 Corporación Andina de FomentoC A F - D O C U M E NT O D E T R AB A J O # 2 0 2 4 / 0 1

E s t a v e r s i ó n : 1 9 d e d i c i e m b r e d e 2 0 2 5

La representación de las mujeres y la gobernanza de los bienes comunes Sandra Aguilar-Gomez1 | Francesca Eustacchi 2 | Laura Aguirre-Duran 3 1Tecnologico de Monterrey, Mexico City, Mexico. saguilargomez@tec.mx 2University College Dublin, School of Economics, Dublin, Ireland. 3Universidad de los Andes. Aproximadamente la mitad de la tierra del mundo es de propiedad colectiva de pueblos indígenas y comunidades locales, con frecuencia administrada mediante instituciones comunitarias de gestión de recursos. A pesar de los esfuerzos nacionales por ampliar la inclusión política, las mujeres siguen estando en gran medida excluidas de estos espacios de decisión. Este documento analiza cómo la inclusión de género en estas

instituciones afecta la gestión de los recursos de propiedad común. Para ello, utilizamos la reforma de 2016 a la Ley Agraria de México, que estableció cuotas de género en los comités agrarios, y estimamos los efectos causales de la participación femenina sobre resultados ambientales, económicos e institucionales mediante un diseño de estudio de eventos escalonado. Con datos de más de una década de registros de comunidades agrarias (2012–2023), encontramos que una mayor participación de las mujeres reduce las tasas anuales de deforestación en alrededor de 6 por ciento sin afectar negativamente la actividad económica local. La inclusión también modifica la forma en que las comunidades interactúan con el gobierno: aumentan las solicitudes a programas orientados al fortalecimiento de capacidades y disminuyen las solicitudes a subsidios productivos. Aunque el volumen total de solicitudes se mantiene estable, las tasas de aprobación caen en comunidades con mayor presencia femenina en cargos de autoridad, lo que sugiere la persistencia de barreras institucionales. En conjunto, los resultados muestran que las reformas de inclusión pueden transformar la gobernanza de los recursos compartidos, pero también revelan restricciones sistémicas que enfrentan las mujeres en posiciones de liderazgo comunitario.

Códigos JEL: D71, J16, O13, Q23

K E Y W O R D S Cuotas de género, Representación política, Gobernanza colectiva, Deforestación, Diferencias en diferencias escalonadas, México. Pequeñas secciones del texto, menores a dos párrafos, pueden ser citadas sin autorización explícita siempre que

se cite el presente documento. Los resultados, interpretaciones y conclusiones expresados en esta publicación son de exclusiva responsabilidad de su(s) autor(es), y de ninguna manera pueden ser atribuidos a CAF, a los miembros de su Directorio Ejecutivo o a los países que ellos representan. CAF no garantiza la exactitud de los datos incluidos en esta publicación y no se hace responsable en ningún aspecto de las consecuencias que resulten de su utilización. ©2025 Corporación Andina de FomentoAGUILAR -GOMEZ ET AL . 2 1 | INTRODUCTION Women’s political voice has expanded worldwide, yet the pace of change remains uneven, especially outside urban centers. Today, women hold just 25% of parliamentary seats and 36% of local government positions worldwide (UN Women, 2022). Electoral reforms have attempted to address this gap: 132 countries have adopted gender quota reforms, with Latin America leading the way through legislated candidate quotas. A deeper gender representation divide tends to predominate in rural areas. Globally, only one in five agricultural landowners or rights-holders is female (UN Women, 2024), and nearly half of rural residents (47%) still believe men make better political leaders, while only a third (34%) of urban residents do (UN Women, 2022). Persistent underrepresentation thus constrains women’s ability to champion policies that directly affect their livelihoods and well-being, from land rights to local public goods (Chattopadhyay and Duflo, 2004; Rehavi, 2007; Funk et al., 2015). Approximately half of the world’s land, particularly rural and forested areas, is governed through communal or collective land tenure systems (World Resources Institute, 2017). 1

Institutional efforts to improve women’s representation within these collective governance structures are prevalent worldwide, but we still know little about how effective such reforms are. The urgency of exploring gender dynamics in environmental management is underscored by ongoing global deforestation. Each year, approximately 10.2 million hectares of forest—an area comparable in size to Panama–is lost (FAO, 2024). Forests are essential for biodiversity conservation, hosting approximately 80% of terrestrial species and playing a vital role in mitigating climate change by sequestering carbon emissions; deforestation alone accounts for nearly 15% of global greenhouse gas emissions (WWF, 2024; IPCC, 2019). Furthermore, forest loss severely affects local livelihoods, with around 1.6 billion people globally depending on forests for agriculture, forestry, and medicinal resources (World Bank, 2021). Understanding the impact of enhanced participation of women in collective land management is particularly relevant given the extensive evidence highlighting gender differences in leadership and management, with some positive outcomes associated with the presence of women in power. For instance, women in governance roles prioritize community-oriented investments, reduce corruption, and foster enhanced collaboration.2 Similarly, at the household level, women managing financial resources consistently achieve greater efficiency and place a stronger emphasis on the welfare of households and communities, an insight leveraged extensively in conditional cash transfer programs and microfinance initiatives (Duflo, 2012; Lechene, 2010; Hoffmann, 2008; Davis et al., 2002; Doepke and Tertilt, 2019; Rubalcava 1In terms of formal land ownership rigths, (World Resources Institute, 2017) reports that this number is reduced to one-fifth of global land, with the rest held under customary tenure. Most of this communal land lies in

China, Canada, Brazil, Australia, and Mexico. 2Evidence includes, e.g., Chattopadhyay and Duflo (2004); Clots-Figueras (2011); Duflo and Topalova (2004); Hicks et al. (2016); Clayton and Zetterberg (2018); Wängnerud (2009); Brollo and Troiano (2016); Alatas et al. (2009); Epstein et al. (2005); Brito Rebolledo et al. (2024).AGUILAR -GOMEZ ET AL . 3 et al., 2009; Pitt et al., 2003; Swaminathan et al., 2010; Ssewamala et al., 2010). Research highlights that gendered traits and structural constraints—such as compliance, risk aversion, prosocial behavior, and vulnerability to environmental degradation—generate distinct environmental attitudes and practices (Brañas-Garza et al., 2018; Croson and Gneezy, 2009; Aguilar Gómez et al., 2024; Aguilar-Gómez and Salazar-Díaz, 2025). Empirical findings from South Asia further reinforce this notion, demonstrating that women’s active participation in land management correlates with more sustainable forest-use practices (Agarwal, 2009, 2010; Leone, 2019). Given the extensive evidence of women’s positive impacts in governance, a critical policy question emerges: Can enhancing women’s representation also effectively address environmental challenges? This paper connects the literature on gender equality and environmental governance by examining whether institutional reforms that expand women’s political representation can also enhance collective resource management. Specifically, it asks: Did Mexico’s legislated gender quotas succeed in increasing women’s participation within communal land institutions, and did this greater inclusion contribute to more sustainable forest outcomes?

We focus on Mexico, where local decision-making bodies, formally known as agrarian communities, represent the most widespread form of collective land tenure; 60% of the country’s forest area (Comisión Nacional Forestal, 2020) is managed by agrarian communities who depend on forest use for their livelihoods. We employ data on the universe of agrarian communities with forest cover between 2012 and 2022 to analyze the causal impact of women’s participation in the decision-making bodies of Mexican agrarian communities on deforestation rates and land management. To answer our research questions, we leverage the introduction of the 2016 Agrarian Law, which mandated that no more than 60% of candidates for agrarian community governance committees be of the same gender. Agrarian communities hold regular elections every three years but on staggered calendars. The nonuniform timing of their first post-2016 elections gave rise to a natural experiment that we take advantage of with a staggered-adoption difference-in-differences (DID) design (Callaway and H.C., 2021; Sunderland et al., 2014; Freyaldenhoven et al., 2021). Because not all the communities that became subject to the quota immediately saw an increase in women’s representation, all the results except those on women’s participation itself should be interpreted as intention-to-treat (ITT) effects—that is, the average impact of exposure to the quota reform regardless of actual changes in representation. To assess the causal effect of the election of women on outcomes such as deforestation, we further construct local average treatment effect (LATE) estimates by dividing the reduced-form effect of the reform on each outcome by its first-stage effect on women’s representation, following a Wald estimator approach. Our analysis reveals that the quota reform significantly increased representation of women by 5.9 percentage points (pp) in the first postreform election, with this figure growing to 7.6 pp by the third year—an increase of nearly one-third relative to the pre-reform average of 24%. However, the reform did not reshape power evenly within communityAGUILAR -GOMEZ ET AL . 4 governance structures. Agrarian community boards have six different posts with varying responsibilities and attributions. The most significant gains occurred in administrative and supporting roles—positions centered on record-keeping, oversight, and auxiliary tasks—where women’s participation rose by 13–21 pp over three years. In contrast, in the most influential posts, which control community resources and executive decisions, there was little to no change. The quota thus opened the door to women’s inclusion, albeit primarily at the lower tiers of authority, revealing a stepped glass ceiling within agrarian governance. Were these mixed representation gains enough to foster social and environmental change? Our results suggest they were. We find that exposure to the quota reform led to a 6% reduction in annual deforestation rates, a clear environmental dividend from the greater inclusion in local governance. The effect was strongest in the year following the first postreform election and was driven by small agrarian communities—settings where collective rules are easier to enforce and local oversight is most effective. In contrast, no significant changes are observed for medium or large communities, where coordination is more difficult and land use pressures are higher. The reductions were also concentrated in areas without formal environmental protection, indicating that women’s leadership makes a difference precisely where local discretion matters most. Scaling the estimates by the reform’s first-stage effect on women’s representation, the LATE implies that a 1 pp increase in the number of women elected reduced deforestation rates by about 1.2 pp from their level before treatment. We also explore the mechanisms behind these results. First, we find that the reduction in deforestation did not come at the expense of economic activity: Nighttime lights, a

proxy for local production, remained unchanged after the reform. Second, drawing on detailed administrative data, we examine whether women’s leadership altered how agrarian communities interact with government support programs—a key channel of resources and capacity building. We find no change in the total number of applications submitted but a marked decline in the number of awards granted, suggesting that structural barriers—be they limitations in leadership experience or discrimination—also play a role. When grouping program applications and allocations by domain, distinct patterns emerge across categories. In programs related to women’s empowerment, applications increased, but awards did not. In programs that promote productive activities such as management plans, technical assistance, and entrepreneurship, both applications and awards declined, while for conservationand technology-oriented programs, which fund forest maintenance and infrastructure, we detect no effect. These results indicate that while women’s entry into collective governance shifted decision-making priorities, entrenched institutional barriers continued to limit their ability to translate formal representation into material influence. Furthermore, the contrast between rising interest in empowerment programs and declining participation in production-oriented programs underscores how gendered roles persist within ANs: women may seek greater involvement in domains tied to inclusion, but men’sAGUILAR -GOMEZ ET AL . 5 control over land, income, and production decisions continues to shape which applications are ultimately successful. Our paper contributes to three strands of literature. First, we add to research on women’s underrepresentation in decision-making bodies. By leveraging a binding candidate quota, we show that formal rules can substantially increase women’s election rates in collective land governance but that these gains are concentrated in lower-ranking positions. This pattern aligns with evidence that while quotas are effective in opening access, they do not necessarily dismantle gender hierarchies within institutions (Bagues and Campa, 2021; Fujiwara et al., 2024). Our results highlight how formal inclusion can coexist with persistent

structural barriers to substantive power. Second, we contribute to the relatively unexplored economic literature on women’s roles in communal land management. Prior studies from South Asia find that higher participation of women in forest user groups improves forest outcomes and promotes more sustainable resource use. 3 We extend this evidence to a new context—Mexico’s collective lands—using causal identification and high-resolution data. We also show that the environmental gains under women’s leadership do not come at the expense of local economic activity, pointing to a potential pathway for sustainable development rather than a trade-off between conservation and production. Finally, we advance the literature on gender and local governance by analyzing how women’s participation shapes interactions with the state. Using comprehensive administrative data on applications to and allocations from public support programs, we show that women’s entry into collective governance changes how communities engage with government resources but that persistent structural and bureaucratic barriers limit their ability to translate representation into material influence. This provides a rare, institutional perspective on how gender inclusion interacts with state capacity and program delivery in rural settings. 2 | INSTITUTIONAL BACKGROUND 2.1 | Collective land tenure and governance Collective land tenure is the dominant ownership form in rural Mexico. More than 32,000 agrarian communities occupy approximately 46% of the national territory (RAN, 2024). Agrarian communities are custodians of Mexico’s forests. At a canopy-cover threshold of 30% or more, over 60% of the national forests (roughly 70.6 million hectares) lie within the boundaries of agrarian communities (Comisión Nacional Forestal, 2020). More than half of agrarian communities actively manage forests as a source of livelihoods; forests directly support about 11.87 million people (9.3% of Mexico’s population), providing timber, nontim3See, for example, Agarwal (2009, 2010); Leone (2019), who document these effects for India and Nepal, showing that forests fare better when more women are involved and that active participation in decision-making is essential for translating representation into conservation gains.AGUILAR -GOMEZ ET AL . 6 ber products, and ecosystem services. Committees can decide which stands are opened for logging, where crops or pasture can expand, and whether environmentally sensitive zones should be protected. They also negotiate timber contracts, enroll in conservation programs, and engage with nongovernmental organizations. Figure 1 illustrates this territorial and environmental relevance. Panel (a) shows the national distribution of agrarian communities, highlighting the extent of collectively owned land parcels (ejidos and comunidades) across Mexico. Panel (b) shows only those agrarian communities with more than 10% forest cover, our analysis sample, revealing their concentration in the central and southern regions where communal forest management is most prevalent, but also their presence in the northern highlands. These entities emerged from Mexico’s postrevolution land-redistribution program. The key institutional distinction between these two forms of collective governance—ejidos and comunidades—is that ejido parcels may, with assembly approval, be converted to private property, whereas land inside acomunidad is inalienable. In practice, de facto privatization occurs in both. In all other respects, however, their governance structures and legal obligations are the same under the Agrarian Law, and the same rules regulate their elections and decision-making procedures. Throughout the paper, we therefore refer to both as agrarian communities4 and do not analyze them separately, as the reform applied to both and their institutional behavior is otherwise equivalent for the purposes of this study. Authority within each unit is distributed across three bodies: the General Assembly and

two committees elected by it. The General Assembly, in which all registered members with land rights (ejidatarios or comuneros) may participate and vote, decides on land use, allocates resources, and adopts internal regulations. This typically implies one assembly member per family since land rights are inherited without subdivision. The first committee, the Commissariat, comprises a president, a first secretary, and a treasurer; it manages communal assets, coordinates productive activities, and represents the community in legal transactions and in dealings with all levels of government, including applications for and receipt of government programs. The second committee, the Oversight Committee, also has three members and audits accounts, oversees compliance with assembly mandates, and may convene extraordinary sessions when irregularities arise. Within the Commissariat, the president functions as the head of the local executive, chairs meetings, executes assembly decisions, and represents the agrarian unit before public and private actors; the treasurer exercises budgetary and financial control, manages bank accounts, and prepares periodic reports; and the first secretary maintains minute books, membership rolls, and land and cadastre records. The Oversight Committee performs ex post oversight: Its president leads monitoring and investigations and may trigger corrective actions or call the assembly when irregularities are detected. The first and second secretaries primarily provide auxiliary support, maintain audit files, and draft supervisory reports 4The literal translation of the legal term in Mexican law would be “agrarian communities,” but because this term is uncommon in English-language scholarship, we use agrarian communities or agrarian units throughout.AGUILAR -GOMEZ ET AL . 7 (a) Agrarian nuclei land (b) Agrarian nuclei with > 10% forest area F I G U R E 1 Collective land in agrarian communities and forested sample Notes: Panel (a) maps the spatial distribution across Mexico of agrarian communities under collective tenure.

Panel (b) shows the subset of communities with more than 10% of their territory covered by forest, the analytical sample used in this paper. Data from the Agrarian Registry (RAN) and INEGI land cover maps. (Morett-Sánchez and Cosío-Ruiz, 2017). Since December 2016, the Agrarian Law has required that no more than 60% of candidates for any committee be of the same gender. The statute does not dictate the gender composition of the winners. Importantly for our identification strategy, election calendars are not synchronized across communities. Women remain underrepresented in both the economic and political life of rural Mexico. Labor force participation among rural women is 42%, compared with 51% among urbanAGUILAR -GOMEZ ET AL . 8 women and 77% among rural men (ENOE 2024 Q4). Production units refer to the set of lands, infrastructure, machinery and equipment, animals and other goods used to carry out productive agricultural or forestry activities. Units headed by women average 8.8 hectares, compared to 14.7 hectares for units headed by men (Instituto Nacional de Estadística y Geografía (INEGI), 2024), and women have markedly lower access to agricultural machinery (Secretaría de Agricultura y Desarrollo Rural, 2023). Only 26% of the holders of formal rights to ejidos or comunidades are women (Comisión Nacional Forestal, 2020). Political gaps are large in local and national rural governance: According to National Agrarian Registry (RAN) records from 2023, fewer than 23% of committee seats are held by women, and women preside over fewer than 7% of agrarian communities’ Commissariats. 2.2 | The 2016 agrarian elections reform Mexico’s 2016 reform establishing gender quotas in agrarian communities’ candidate lists

occurred in the context of a broader, decades-long push toward gender parity in party-based public office. Prior to 2002, the law merely recommended the inclusion of women in party lists. Accordingly, women’s representation in the Chamber of Deputies hovered around 15%. Mandatory quotas introduced in 2002 set a 30% floor and were followed by a rise in the actual share of women in the chamber to about 23%; raising the quota to 40% in 2007 brought women’s representation in the national legislature to approximately 28% (Torres, 2010). Enforcement tightened in 2011 after a ruling by the Superior Chamber of the Electoral Tribunal,5 after which women’s representation reached 37%—surpassing the 30% threshold often cited as the minimum for effective influence in collegiate bodies (Martínez, 2013). The parity requirement was then constitutionalized in 2014,6 with political parties mandated to present candidate lists composed of 50% women and 50% men at the federal and local levels. Agrarian communities—Mexico’s collective-land governance institutions—have historically operated on the margins of party politics and therefore, prior to 2016, were not subject to any gender quotas. Consequently, women’s participation in them lagged markedly: Throughout the 1990s and 2000s, women’s presence as candidates and officeholders in agrarian communities remained about 10% in comunidades and 15% in ejidos. The Agrarian Law was amended in 2016 to address this gap by imposing a candidate rule for agrarian committees: No gender may exceed 60% of nominees for any office. While the statute did not mandate outcomes, qualitative evidence suggests that the increase in women’s representation among candidates set in motion a positive feedback loop, generating further opportunities for their engagement in governance and land management initiatives

(García-Morán and Yates, 2022a). 5Ruling SUP-JDC-12624/2011 of the Superior Chamber of the Electoral Tribunal of the Federal Judiciary. 6Article 41, Section I of the CPEUM (Constitución Política de los Estados Unidos Mexicanos), together with Articles 33.3–33.5 of the LGPP (Ley General de Partidos Políticos) and Article 232.3 of the LEGIPE (Ley General de Instituciones y Procedimientos Electorales).AGUILAR -GOMEZ ET AL . 9 During the period of our analyses, several other reforms to the Agrarian Law were passed, as summarized in Appendix Table A.1. However, only four of them stand out for being related to gender. In 2017, a decree added women’s agroindustrial units (UAIMs) to the Agrarian Law, allowing (not requiring) ejidos and comunidades to allocate productive land for women’s collective initiatives; uptake is discretionary and not tied to the first post-2016 election. Hence, this change does not threaten our identification strategy. More recently, reforms enacted in 2023 to Articles 4, 32, and 37 continued this trajectory. Article 4 now requires that rural development policies incorporate a gender perspective, and Article 32 states that the composition of ejido governance bodies should observe the principle of gender parity. However, this latter article does not impose a binding requirement to reserve seats or ensure balanced outcomes. By contrast, Article 37 introduces a concrete obligation: Candidate lists for agrarian committees and auxiliary bodies must include men and women in equal proportion. This reform strengthened the earlier 2016 provision—which established a 60/40 gender cap—by requiring strict 50/50 gender parity in nominations. Because this 2023 mandate represents a different treatment, we omit observations of our outcomes from

2023 onward from the analysis. Article 37 of the Agrarian Law established that no more than 60% of candidates for agrarian governance bodies may be of the same gender. However, only final election results are digitized, which means that there is no public record of candidate lists that would allow systematic verification of compliance at the community level. In practice, the Procuraduría Agraria (PA)—the federal agency responsible for overseeing and certifying agrarian assemblies—is tasked with monitoring adherence to this quota. Official communications and reports indicate that PA officers review assembly records, advise on candidate registration, and occasionally supervise elections to ensure gender balance, usually upon request. Hence, incentives to comply with the law are influenced by the perceived probability of audit by the PA. The agency has also launched programs such as Mujeres por el Acceso a la Tierra to provide legal assistance and promote women’s participation in agrarian governance (Secretaría de Desarrollo Agrario, Territorial y Urbano (SEDATU), 2021). Qualitative analyses highlight that enforcement remains uneven. Several authors note that female nominees are often listed as substitutes and that social norms or fear of community backlash discourage women from running for executive positions. For instance, Torres-Mazuera (2023) concludes that “gender parity in agrarian representation remains a law on paper, with little application in the daily life of ejidos and communities.” These barriers are reinforced by the fact that women remain a minority among land-right holders, which limits their eligibility and confidence to assume leadership roles. Similarly, GarcíaMorán and Yates (2022b) report that there are few women with recognized agrarian rights; many are only beginning to participate in community assemblies, in the face of resistance and cultural stigma. Against these cultural and economic obstacles, one might a priori expect our quantitative analyses to reveal null effects of the quota reform. However, administrative data establish th