CAF - Long-Term Effects of a Commodity Boom. Rubber Slavery in the Amazon

C A FW O R KI N GPA P E R# 2 0 2 6 / 0 2

F i r s t v e r s i o n : F e b r u a r y 2 4 , 2 0 2 6 ( c u r r e n t ) Long-Term Effects of a Commodity Boom: Rubber Slavery in the Amazon

Daniel Araujo1 | Humberto Laudares2 | Dafne Murillo3 | Hector Paredes4 | Felipe Valencia Caicedo5 1University of British Columbia. danielmendonca190@gmail.com 2University of Geneva. hlaudares@gmail.com 3Columbia University. dm3160@columbia.edu 4Paris School of Economics. hector.paredes@psemail.eu 5Brown University. valencia@brown.edu Can a brief period of economic prosperity leave a legacy of longterm adversity? This study examines the lasting impact of the Amazon Rubber Boom (1870-1914) on contemporary income, inequality, Indigenous presence, and forest conservation. Empirically we combine variation in historical rubber distribution with an instrumental variable strategy using FAO-based rubber suitability and a Regression Discontinuity design around concession boundaries. Municipalities with greater rubber presence experienced short-term gains in 1920 but long-run reversals by 2010, showing lower income, population density, and higher inequality and Indigenous extinction. Grid-level analyses across the Amazon further show that historical rubber suitability is associated with lower economic activity and sparser population today, alongside greater deforestation. The findings, consistent across Brazil, Colombia, Peru, and Bolivia, indicate that the rubber boom’s short-lived wealth reinforced extractive institutions and violence against Indigenous peoples, leaving long-lasting

economic, social, and environmental scars across the Amazon. K E Y W O R D S Commodity Booms, Forced Labor, Indigenous Peoples, Amazon 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. ©2026 Corporación Andina de FomentoC A FD OC UM E NT O D E T R AB AJ O # 2 0 2 6 / 0 2

E s t a v e r s i ó n : 2 4 d e f e b r e r o d e 2 0 2 6

Efectos a Largo Plazo de un Auge de Materias

Primas: La Esclavitud del Caucho en la Amazonía

Daniel Araujo1 | Humberto Laudares2 | Dafne Murillo3 | Hector Paredes4 | Felipe Valencia Caicedo5 1University of British Columbia. danielmendonca190@gmail.com 2University of Geneva. hlaudares@gmail.com 3Columbia University. dm3160@columbia.edu 4Paris School of Economics. hector.paredes@psemail.eu 5Brown University. valencia@brown.edu ¿Puede un breve período de prosperidad económica dejar una herencia de adversidad a largo plazo? Este estudio examina el impacto duradero del auge del caucho en la Amazonía (18701914) sobre el ingreso, la desigualdad, la presencia indígena y la conservación forestal contemporáneos. La estrategia empírica combina la variación en la distribución histórica del caucho con una estrategia de variables instrumentales basada en el potencial para el cultivo de caucho según la FAO y un diseño de Regresión Discontinua alrededor de los límites de las concesiones. Los municipios con mayor presencia histórica de caucho experimentaron incrementos de corto plazo en 1920, pero reversiones de largo plazo hacia 2010, mostrando menores niveles de ingreso y densidad poblacional, y mayores desigualdad y extinción indígena. Los análisis a nivel de grilla para toda la Amazonía muestran además que la idoneidad histórica del caucho se asocia con menor actividad económica y población actuales, junto con una mayor deforestación. Los resultados, consistentes en Brasil, Colombia, Perú y Bolivia, indican que la riqueza efímera del auge del caucho reforzó instituciones extractivas y la violencia contra los pueblos indígenas, dejando profundas huellas económicas, sociales y ambientales en la región amazónica. K E Y W O R D S Auge de materias primas, trabajo forzado, pueblos indígenas, Amazonía 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. ©2026 Corporación Andina de FomentoARAUJO ET AL. 2 1|INTRODUCTION The extraction of commodities and natural resources has been part and parcel of Latin America’s economic reality throughout history. This legacy of extraction was especially important during colonial times and was later exacerbated by industrial demand for natural products. Notable among these was Amazon rubber, which experienced a boom from 1870 to 1914. In this paper, we scrutinize the shortand long-term impact of this historical episode. Although rubber extraction led to a temporary increase in wealth accumulation in cities such as Iquitos and Manaus, its broader development effects were devastating. Areas that historically extracted rubber are now poorer and have higher levels of inequality. We document this dramatic development reversal and hypothesize that the use of forced Indigenous labor underpinned these deleterious long-run effects. To investigate this, we build a new dataset and run a series of empirical tests. To the best of our knowledge, this is the first empirical economics study of the effects of Indigenous slavery in Latin America, following pioneering work on Indigenous welfare in the Americas by Feir et al. (2024a) and Barsanetti and Ferreira (2022). In addition, we advance the literature on commodities in Latin America by going further back in time than did Méndez and Van Patten (2022). We also examine the impact of rubber in the Americas, complementing the work of Lowes and Montero (2021) for Africa.

Rubber latex was mainly extracted through private concessions in the Amazon basin and exported worldwide to fuel the appetite for bicycle pneumatics and the nascent demand for car tires during the late nineteenth and early twentieth centuries. 1 The geographic coverage of this colossal enterprise spanned Brazil, Peru, Colombia, Ecuador, and Venezuela. The Hevea tree was never domesticated in South America, barring the establishment of plantations. Instead, rubber entrepreneurs relied on exploitative labor arrangements with Indigenous people, which effectively amounted to modern slavery. 2 Although slavery had at the time been formally abolished in Latin America, rubber extraction depended on coercive labor arrangements, particularly debt peonage, that trapped Indigenous people in conditions of forced labor amounting to enslavement. Official reports and literary accounts detail the horrors the local population underwent in stations such as Casa Arana in the Putumayo region.3 Another key aspect of the Amazon rubber boom is that it was suddenly dominated by rubber plantations in Southeast Asia (Hanson, 1944). 4 This production explosion on the other side of the globe led to a sharp price drop and the sudden collapse of the Amazon rubber industry (see Figure 1). What was the long-term impact of this natural resource boom? In the short run, the “rubber fever” drew capital to Amazonian hubs; at its peak, from 1870 to 1910, Manaus, the capital of Amazonas state in northern Brazil, was the richest city in South America. With its grand opera house and a city-electrification project among the earliest in the world, it was dubbed the “Paris of the Tropics.” But did this prosperity trickle down to the local population over time in terms of development outcomes and public-goods provision? Or was the rubber boom an archetypal enclave economy built on Indigenous exploitation that left enduring economic, social, and environmental scars? The Amazon

rubber boom provides a window into the long-term legacies of forced labor. Rubber barons 1For the history of private colonization in Brazil, see Martins Secco Luce (2024). 2By the time of the rubber boom, slavery had been abolished in all producing countries, the last of which was Brazil, in 1888.Hevea brasiliensisis a species native to the Amazon rainforest and the primary source of natural rubber, which is extracted from its bark (Dean, 1987). The exploitation of rubber in the Amazon required an understanding of the suitability of different types of rubber trees. 3A particularly vivid picture is provided by Werner Herzog’sFitzcarraldoand the reports of Roger Casement, cited above and fictionalized by Mario Vargas Llosa. 4Rubber tree seeds were stolen and smuggled by British explorer Henry Wickham to this end.ARAUJO ET AL. 3 and traders operated with near-total impunity, enforcing brutal conditions through violence. We document this traumatic historical reality and, combining historical and contemporary sources, provide causal evidence on a first-order socioeconomic and policy issue. To this end, we create a novel dataset comprising a diverse range of archival records, current administrative datasets, and highly disaggregated information from satellite missions, national records, and Indigenous community presence. Sources such as HydroSHEDS, BIOCLIM, and NASA Nighttime Lights allow us to analyze the enduring effects of the Amazon rubber boom on local communities while accounting for geographic and weather conditions. In addition to traditional economic outcomes, we assess deforestation and illegal activities to explore the complex interactions between exploitation, cultural assimilation, and environmental degradation. We employ two complementary empirical strategies. First, we use an instrumental variable approach that exploits exogenous variation in rubber suitability from the FAO’s Global Agro-Ecological Zones dataset to instrument for the historical distribution of rubber presence documented by Schurz et al. (1925). This strategy allows us to isolate the effect of historical rubber activity from potential confounding factors related to geography, settlement, or accessibility. In this first set of analyses, we compare areas with similar environmental characteristics but differing in their predicted suitability for rubber cultivation. Second, we leverage a Regression Discontinuity (RD) design around the boundaries of historical rubber concessions, such as those established by Casa Arana, to capture sharp spatial contrasts in exposure to extractive institutions and coerced labor. Concession boundaries are digitized from maps originally commissioned by the rubber firms themselves. This paper contributes to the literature on the long-run economic effects of agricultural commodity booms. A close study is Méndez and Van Patten (2022), which documents the positive effects of the United Fruit Company on development outcomes in twentieth-century Costa Rica. For identification, they exploit an RD design around the concessions granted to this multinational. Their main channel is the ability of this monopsonistic firm to attract workers by investing in local amenities, a channel that was not present in the current context. In contrast to the Central American case, we examine an earlier concessions regime in Latin America and test whether the Amazonian rubber boom had adverse long-run consequences. Dell and Olken (2020) also finds a positive effect of the Dutch Cultivation System in Java, Indonesia, where areas with colonial-era sugar factories have better infrastructure and are more industrialized today.5 We hypothesize the opposite pattern for the Amazon rubber boom due to precarious working conditions and minimal investments in rubber regions. Across the Americas, a large literature links exploitative regimes, natural endowments, and commodity booms to long-run development. Sokoloff and Engerman (2000) provides a theory for underdevelopment based on natural resources and subsequent colonial institutions. In that vein, Bobonis and Morrow (2014) shows that after the abolition of forced labor in Puerto Rico in 1874, increased coffee prices led to a decline in literacy rates on the island. This evidence is consistent with the findings of Uribe-Castro (2019) on coffee production, human-capital investments, and structural transformation in Colombia. Musacchio et al. (2014) examine the interaction between commodity booms and education in Brazil. Related work shows that ranching specialization in Argentina resulted in persistently slower industrialization compared to cereal-growing regions (Droller and Fiszbein, 2021). On the institutions channel, Naritomi et al. (2012) documents the importance of colonial commodity booms for Brazilian institutional development, focusing on sugar and gold. To the best of our knowledge, no economics paper has examined the impact of the Amazon rubber boom. We also contribute to the literature on the economic consequences of slavery and coerced labor. Nunn (2008) and Nunn and Wantchekon (2011) document the negative impact of 5Similarly, Kok and Lim (2024) finds a positive effect of rubber estates on human capital in Malaysia.ARAUJO ET AL. 4 slavery on exporting nations in Africa in terms of income and decreased trust. More closely related, Lowes and Montero (2021) finds negative long-run effects of rubber concessions granted during the Congo Free State: concessions reduced education, wealth, and health outcomes through an institutional-persistence channel in which village chiefs in former concession areas are less likely to be elected and provide fewer public goods. We examine Indigenous slavery during a period of formal abolition in Latin America. Although we share a focus on rubber, the Amazon’s geographic, ethnographic, and historical context differs markedly from that in Africa. Our emphasis is on the exploitation of Indigenous people

rather than traditional chief authority. We also analyze mechanisms related to environmental outcomes and illegal activities, such as coca cultivation. More broadly, evidence from Latin America shows that coercive labor arrangements left enduring marks on development. Laudares and Caicedo (2023) studies the long-term impact of transatlantic slavery on Brazilian inequality, while Dell (2010) demonstrates how the miningmitalabor system decreased consumption and increased stunting in Peru’s Southern Highlands. Rivadeneira (2024) highlights the intergenerational legacy ofconcertaje, a debt-based system of forced labor in Ecuadorian farms orhaciendas, showing adverse effects on descendants’ educational and labor market outcomes. More recently, Espinoza and Paredes (2025), drawing on new data onhaciendaexposure across Peru, finds that districts with larger shares of their population employed on these estates accumulated less human capital throughout the twentieth century. Relatedly, Araujo et al. (2024) shows that illegal mahogany markets in the Amazon perpetuate contemporary forms of forced labor, revealing the persistence of coercive institutions in the region. While these studies highlight the economic legacies of different forms of coerced labor, ours is the first to provide evidence on Indigenous slavery’s long-run economic consequences in Latin America. In doing so, we complement the work on Indigenous welfare in North America (Feir, 2016; Feir et al., 2024a) with evidence from South America.6 Recent work on Indigenous issues considers colonial treaties (Feir et al., 2023) and American Indian wealth (Feir et al., 2024b). In Latin America, research has focused on Indigenous communities in Mexico (Elizalde, 2020), Indigenous groups in Argentina (Dal Bó and Lopez, 2024), and the Mapuche community in Chile (Jaimovich and Toledo, 2021; Jordán, 2021). Still, more research is needed to understand the

conditions of the almost 10% of Latin Americans who identify as Indigenous7. The rest of the paper is organized as follows. Section 2 covers the background in terms of rubber exploitation in the Amazon and the rubber boom of 1870–1910. Section 3 describes the datasets used in the empirical analysis. Section 4 lays out our empirical strategies, including reduced form and RD estimates. Section 5 presents our results. Finally, Section 6 concludes. 2|BACKGROUND 2.1|Rubber in the Amazon The Spanish Crown arrived in the Amazon region between 1499 and 1500 during expeditions led by the Pinzón brothers and Diego de Lepe. However, it was not until forty years later that Francisco de Orellana successfully completed the first recorded expedition across the Amazon River, reaching Ecuador. In 1616, during the Iberian Union, Francisco Caldeira Castelo Branco’s arrival marked Portugal’s takeover of the Amazon region. Portugal began distributing donatary captaincies to implement its colonial system in the Americas (Freire 6See Reséndez (2016) on the slavery of Indigenous people in North America. 7Laudares et al. (2025) explores the relationship between deforestation and Indigenous Peoples’ health during the COVID-19 pandemic in Brazil.ARAUJO ET AL. 5 et al., 1994; Martins Secco Luce, 2024). Lacking significant deposits of gold or silver, colonial explorers turned to collect forest products and medicinal plants—referred to as “backlands drugs" (drogas do sertão)—such as cinnamon, cloves, cocoa, and copaifera, which were exported to Europe. Indigenous peoples, often enslaved, were responsible for extracting these resources throughout the colonial period.8 The extraction of latex from theHevea brasiliensistree became a significant economic

activity in the Amazon during the nineteenth century. Once coagulated, this latex produced rubber. The Olmecs, Mayas, and Aztecs were among the first to use rubber for balls and other objects. News of its existence in the Americas reached Europe during Columbus’s second voyage in 1493. In 1736, La Condamine introduced a milky, elastic substance called caoutchouc to the French Royal Academy of Sciences. This sparked further experimentation: in 1764, Fresnau dissolved rubber in turpentine, and in 1779, Fabbroni did so using naphtha. Early uses included erasing pencil marks, manufacturing elastic bands, and waterproofing. Nonetheless, the transformative moment in rubber production was Charles Goodyear’s invention of the vulcanization process in 1839. Thanks to this technological breakthrough and with rising demand for rubber in industrialized nations, the Amazon became the world’s primary supplier until the 1880s. The world needed rubber for tires, surgical gloves, raincoats, and boats. Accordingly, rubber production in the Brazilian Amazon soared, from 31,365 kilograms in 1827 to 2,673,000 kilograms in 1860. Despite this growth, the extraction methods remained similar to those used for “backlands drugs" during colonial times (Weinstein, 1983). Theseringueiros(rubber tappers) collected latex by making incisions in Hevea trees and attaching small bowls to collect the sap. This process was labor-intensive, as the trees were scattered across large areas. On weekends, rubber tappers transported the latex to landowners or local merchants (aviadores), who sold it in cities like Belém and Manaus. The rubber tappers typically received only half of the sales revenue and were often required to

pay commissions on the money or goods they received. In the cities, the latex was prepared for export, primarily by American or British companies. Landowners, as the sole providers of credit, exploited theseringueirosextensively. Many workers became indebted to rubber barons through a system that trapped them in continuous labor to repay loans. Journalist Euclides da Cunha described them as “debt slaves" (Cunha, 1955). Living conditions for rubber tappers were harsh. In other parts of the Amazon, particularly in the Colombian and Peruvian region of Putumayo, rubber extraction involved even greater exploitation of Indigenous labor. The Arana family enslaved Indigenous people to extractcaucholatex from theCastilla elasticatree, often employing extreme measures such as torture, mutilation, and killing to enforce labor discipline. UnlikeCastilloa elastica, which was destroyed during latex extraction,Hevea brasiliensis trees could be harvested for up to fifty years. According to Weinstein (1983), Brazilian landowners extracting Hevea latex had stronger economic incentives to maintain productive relationships with rubber tappers. Additionally, power was more decentralized in the Brazilian Amazon compared to the Arana family’s dominance in Peru. The de facto absence of the central state in the Peruvian Amazon gave enormous power to the rubber barons. 2.2|The Rubber Boom: 1870–1910 Technological advancements in steam navigation and communication facilitated the initial growth of the rubber economy in the 1870s. Steamships significantly reduced transportation costs, while the expansion of telegraph networks connected Amazonian exporters to global 8As a matter of fact, Brazilwood orpau brasilgave the name to the country.ARAUJO ET AL. 6 markets. Rubber exports from Pará and Amazonas, the two primary production hubs, began to dominate Brazil’s trade balance, becoming second only to coffee.

The 1890s marked the height of the Amazon rubber boom. By this period, the global demand for rubber had expanded dramatically, fueled by the bicycle craze and the emergence of the automobile industry. The industrialization of transportation systems in Europe and the United States further accelerated demand for rubber, as car tires became an essential product. Figure 1 shows the evolution of wild and plantation rubber production in Brazil (on the left axis), and the value per ton in British pounds (on the right axis) from 1825 to

1925. The main sources of information are Schidrowitz and Dawson (1952); Hotchkiss (1924); Benchimol (2010) and Vayagool (1967).

Brazil’s geographical advantage and monopoly on wild rubber extraction supported skyrocketing prices and rapid economic growth. Urban centers like Manaus and Belém transformed into symbols of opulence, with infrastructure projects such as the Manaus Opera House, funded by rubber profits. Population growth in the Amazon surged as immigrants arrived to capitalize on the economic opportunities. By 1910, the population of the Amazon had grown more than 400% and regional incomes had increased by 800% (Barham and Coomes, 1994; Fernandes, 2010). In that year, the country was responsible for 74% of the world’s wild rubber production (Fonseca, 1950). Yet, this growth came with deep structural vulnerabilities. The industry’s reliance on wild rubber extraction meant that production was labor-intensive and geographically limited. Additionally, the profits of the boom were unevenly distributed. While local elites and foreign investors benefited, rubber tappers lived under harsh conditions, often facing disease, malnutrition, and exploitation. Barham and Coomes (1994) argue that local surplus retention during the boom was significant, with private and state investments focusing on infrastructure and trade. However, this investment pattern created a fragile economic structure heavily

dependent on rubber exports. The authors also highlight the phenomenon of “Dutch Disease," whereby the booming rubber sector led to the neglect of other industries, such as agriculture, exacerbating regional inequalities. On the other side of the Brazilian border, Casa Arana—the notorious rubber enterprise of Julio César Arana—played a brutal and pivotal role in the Amazon rubber boom. In the Putumayo region of Peru and Colombia, Arana ruled a private fiefdom, allegedly as large as Belgium, through his British-registered Peruvian Amazon Company, subjecting Indigenous populations to a systematic regime of forced labor, exploitation, and violence (Goodman, 2010; Tully, 2011; Pineda, 2000). Under this coercive arrangement, nearly 30,000 Indigenous individuals were enslaved, with approximately 4,000 tons of rubber extracted under conditions of extreme brutality (Tully, 2011). Initially highly profitable, Arana’s operations contributed substantially to the prosperity of Amazonian ports such as Iquitos, but the wealth generated depended heavily on severe human rights abuses (Pineda, 2000). The exposure of these atrocities through eyewitness accounts, notably by journalist

W. E. Hardenburg and British consul Roger Casement between 1909 and 1912, sparked international outrage and led to increased scrutiny by humanitarian organizations such as the Anti-Slavery and Aborigines’ Protection Society (Goodman, 2010). Casa Arana remains emblematic of the profound human cost underlying extractive booms and the enduring social and economic legacies of historical exploitation in the Amazon.

The rapid rise in global demand for rubber by the turn of the century exposed the limitations of the Amazonian rubber production. Wild rubber could not meet the insatiable demand due to labor shortages and the natural scarcity ofHevea brasiliensis, leading to the initial price increase (Figure 1). Simultaneously, the development of rubber plantations in

British and Dutch colonies in Southeast Asia introduced a low-cost, labor-intensive production system that out-competed Amazonian producers. Seeds were smuggled from theARAUJO ET AL. 7 Amazon to establish plantations on the other side of the globe. Dean (1987) chronicled how the smuggling ofHevea brasiliensisseeds by Henry Wickham in 1876 allowed the British to establish plantations in Ceylon, Singapore, and Malaysia. By 1910, Asian plantations had matured and began flooding the market with cheap rubber, eroding the South American dominance, and leading to a collapse in the prices. The plantation model was still unsuccessful in the Amazon, mostly due to diseases and soil erosion. The collapse of global rubber prices with the expansion of the Asian rubber plantations (denoted as plantation rubber in Figure 1) had devastating effects on the Amazonian economy. Many export houses and traders went bankrupt, urban real estate values crashed, and service industries dependent on rubber profits withered. The social and economic gains of the boom evaporated, leaving the region impoverished and underdeveloped. We next document this reversal empirically. 3|DATA As in a standard persistence study, we have a three-layered data structure. First, there is the historical information, in this case, about the regional variation in exposure to the rubber boom. Second, we compile data on modern development outcomes such as measures of economic activity (e.g., night lights, income). This information is accessible through standard modern censuses as well as socio-demographic or specialized surveys (e.g., for Indigenous populations).9 Third, we utilize a large set of remote sensing data. This includes, among others, data on modern patterns of economic activity, environmental conservation (deforestation, cattle farming, and coca cultivation) as outcomes of interest, as well as geographic and weather data as control variables to help isolate the effect of historical

variables on modern outcomes. 3.1|Historical Exposure & Indigenous Demographics We make use of the rich historiography about rubber in the Amazon, most notably, the consular report of Robert Casement in 1904 for the British Foreign Office. There are extant official and written testimonies by French explorer Eugene Robuchon, from 1907, and judge Carlos Valcarcel, published in 1915. Rubber was an important commodity, so its price was followed by newspapers and its development by specialized magazines such asIndia Rubber World. There are also a series of high-quality secondary sources including Schurz et al. (1925); Hanson (1944); Le Cointe (1947); Weinstein (1983); Barham and Coomes (1994); Stanfield (1998); Nugent (2017). Two historical sources that measured early rubber exposure and Indigenous demographics are worth highlighting. Historical Rubber Distribution. In order to measure early rubber exposure, we have digitized the historical rubber distribution map developed by Schurz et al. (1925), which encompasses Brazil, Peru, Colombia, Ecuador, and Venezuela. Schurz et al. (1925) assessed rubber distribution using a combination of environmental, ecological, and geographic criteria available at the time (Figure 2). Their evaluation drew upon extensive fieldwork, expert observation, local knowledge, and information from earlier expeditions and reports. Key criteria included (i) climatic conditions—such as relative humidity, rainfall, and temperature conducive toHevea brasiliensis; (ii) vegetation cover, that is, observed or reported wild rubber groves; (iii) proximity to rivers and water bodies for both ecological suitability and 9For the study cases of Peru and Colombia, an additional original survey will be implemented, tracing individuals with family backgrounds linked to Casa Arana to study impacts in all of these dimensions but also

more rubber-specific dynamics linked to cultural change and attitudes/beliefs towards environmental issues.ARAUJO ET AL. 8 extraction logistics; (iv) qualitative assessments of topography and soils; and (v) areas of documented historical rubber extraction. To construct municipality-level estimates, we overlaid the most recent administrative municipality or grid shapefiles for Brazil and neighboring countries onto the digitized Schurz et al. (1925) distribution polygons, calculating for each municipality the fraction of land classified as rubber-present by Schurz et al. As shown in Figure 3, this allows us to construct municipality-level estimates of rubber distribution in 1925.10 While there seems to be no granular