¹ See Browder, John O., ‘The Limits of Extractivism’, BioScience, vol. 42, no. 3 (1992)CrossRefGoogle Scholar; Nepstad, Daniel C. and Schwartzman, Stephan (eds.), Non-Timber Products from Tropical Forests: Evaluation of a Conservation and Development Strategy, Advances in Economic Botany, vol. 9 (New York, 1992)Google Scholar; Plotkin, Mark and Famolare, Lisa (eds.), Sustainable Harvest and Marketing of Rain Forest Products (Covelo, 1992)Google Scholar; Redford, Kent H. and Padoch, Christine (eds.), Conservation of Neotropical Forests: Working from Traditional Resource Use (New York, 1992)Google Scholar; Anderson, A. B., May, P. H. and Balick, M. J., Tie Subsidy from Nature: Palm Forests, Peasantry, and Development on an Amazon Frontier (New York, 1991)Google Scholar; Clay, Jason W., ‘A Rainforest Emporium’, Garden, vol. 14, no. 1 (01&02. 1990)Google Scholar; Fearnside, Philip M., ‘Extractive Reserves in Brazilian Amazon’, BioScience, vol. 59, no. 6 (1989)Google Scholar; and Bunker, Stephen G., Underdeveloping the Amazon: Extraction, Unequal Exchange, and the Failure of the Modern State (Chicago, 1985)Google Scholar.

² See Schwartzman, Stephan, ‘Extractive Reserves: The Rubber Tappers' Strategy for Sustainable Use of the Amazon Rainforest’, in Browder, John O. (ed.), Fragile hands of Latin America: Strategies for Sustainable Development (Boulder, 1989)Google Scholar; and Alegretti, Mary Helena, ‘Extractive Reserves: An Alternative for Reconciling Development and Environmental Conservation in the Development of the Brazilian Amazon’, in Anderson, Anthony (ed.), Alternatives to Deforestation (New York, 1990)Google Scholar.

³ See Barham, Bradford L. and Coomes, Oliver T., ‘Reinterpreting the Amazon Rubber Boom: Investment, the State and Dutch Disease’, Latin America Research Review (forthcoming, vol. 29, no. 1, 1994)Google Scholar; and, Coomes, Oliver T. and Barham, Bradford L., ‘The Amazon Rubber Boom: Labor Control, Resistance, and Failed Plantation Development Revisited’, Hispanic American Historical Review (forthcoming, vol. 74, no. 1, 1994)CrossRefGoogle Scholar.

⁴ Katzman, Martin T., ‘Review Article: Ecology, Natural Resources, and Economic Growth: Underdeveloping the Amazon’, Economic Development and Cultural Change, vol. 3;, no. 1 (1987)Google Scholar; Katzman, Martin T., ‘Paradoxes of Amazonian Development in a “Resource-starved” World’, The journal of Developing Areas, vol. 10, no. 4 (1976)Google Scholar; and Norgaard, Richard B., ‘Sociosystem and Ecosystem Coevolution in the Amazon’, Journal of Environmental Economics and Management, vol. 8, no. 3 (1981)CrossRefGoogle Scholar.

⁵ Recent regional studies on the Rubber Boom are available for: Brazil [Weinstein, Barbara, The Amazon Rubber Boom, 1850–1920 (Stanford, 1983)Google Scholar] Peru [Bonilla, Heraclio, ‘El Caucho y la Economía del Oriente Peruano’, in Gran Bretaña y el Perú: Los Mecanismos de un Control Económico, vol. V (Lima, 1977)Google Scholar; Pennano, Guido, La Economia del Caucho (Iquitos, 1988)Google Scholar; Marín, José Antonio Flores, La Explotación del Caucho en el Perú (Lima, 1987)Google Scholar]; Bolivia [Fifer, Valerie J., ‘The Empire Builders: A History of the Bolivian Rubber Boom and the Rise of the House of Suárez’, Journal of Latin American Studies, vol. 2, no. 1 (1970)CrossRefGoogle Scholar] Colombia [Domínguez, Camílio and Gómez, Augusto, La Economía Extractiva en la Amazonía Colombiana 1850–1930 (Bogotá, 1990)Google Scholar]; and Venezuela [Iribertegui, Ramón, Amazonas: El Hombre y el Caucho, Monografía No. 4 (Caracas, 1987)Google Scholar].

⁶ According to one authority on the boom, Amazonia is ‘unusually uniform in its physical and economic characteristics’ and ‘most of the conclusions [of the work] can be generalized for the Amazon as a whole, since the repercussions of the rubber boom were much the same for tappers and traders working just a few miles inland and their counterparts working a thousand miles upstream’, Weinstein, Amazon Rubber Boom, p. 4. Despite this view, Weinstein's richly descriptive account of the boom at various locales in Brazil provides considerable evidence of striking spatial differences in relations of production across the region.

⁷ Weinstein, Amazon Rubber Boom, p. 9. For regional and world production data during this later period, see LeCointe, Paul, L'Amazpnie Brésilienne, vol. I (Paris, 1922), p. 451Google Scholar.

⁸ For statistics on wild rubber production by country during the boom see Pearson, Henry C., The Rubber Country of the Amazon (New York, 1911), pp. 214–15Google Scholar; LeCointe, Paul, L' Amazonie Brésilienne, vol. I (Paris, 1922), pp. 433–44Google Scholar; Pennano, La Economía del Caucho, pp. 178–9; and Iribertegui, Amazpnas, pp. 168–70.

⁹ See India Rubber World, vol. 31, no. 5 (1 March 1905), p. 180.

¹⁰ For example, see Flores Marfn, ha Explotación del Caucho.

¹¹ See Baumol, William J., Panzer, John C. and Willig, Robert D., Contestable Markets and the Theory of Industry Structure (New York, 1982)Google Scholar for a formal treatment of the question of contestable markets.

¹² For example, at Manaus in 1909, 45 firms received more than 100,000 kg of rubber, accounting for 83.3 % of all rubber received at the port; the remainder was received by over 900 small firms. Loureiro, Antônio José S., A Grande Crise (1908–1916) (Manaus, 1986), pp. 222Google Scholar, 230–5. A small steamer in a single trip from the Islands district near the mouth of the Amazon could collect rubber from 200–300 shippers destined for 7; or 80 different rubber receivers in Belém. Pearson, The Rubber Country, p. 41.

¹³ The Brazilian trading house of Vianni repeatedly attempted to corner the Belém rubber exchange over more than a decade in the 1870s and 1880s without success. Weinstein, The Amazon Rubber Boom, pp. 139–55. The government of Brazil attempted later to valorise rubber by purchasing rubber on the open market in Belém. For a discussion of impediments to the formation of a rubber monopoly during the boom, see India Rubber World, vol. 23, no. 5 (1 Feb. 1901), pp. 135–6.

¹⁴ The determinants of monopoly and strategic behaviour by firms in scarce resource industries are explored elsewhere. See Barham, Bradford L., ‘Strategic Capacity Investments and the Alcoa-Alcan Monopoly: 1888–1945’, in Barham, Bradford L., Bunker, Stephen, and O'Hearn, Denis (eds.), States, Firms, and Raw Materials: The World Economy and Ecology of Aluminum (Madison, forthcoming, 1994)Google Scholar; Bradford L. Barham, Stephen Bunker and Denis O'Hearn, ‘The Organization and Performance of Raw Materials Industries in Resource-rich Regions’, In ibid.; and Bradford L. Barham, ‘Strategic Commitments in Scarce Resource Industries: Microfoundations of Multinational Monopolies’ (Madison, unpublished mimeo). Relative scarcity and a highly concentrated distribution of the material are necessary conditions if a few firms are going to obtain market power by controlling access to the scarce resource. In the case of wild rubber, its naturally dispersed nature made this type of strategy impossible.

¹⁵ Hevea latex, from which the highly appreciated Pará fine rubber was formed, dominated the Amazon rubber trade. Hevea rubber exports from the Brazilian Amazon during 1890–1909 exceeded caucho exports by approximately 5–10:1. Derived from Pearson, The Rubber Country, p. 215.

¹⁶ Weinstein. The Amazon Rubber Boom, p. 17.

¹⁷ Santos, Roberto, História Econômica da Amazônia (1800–1920) (São Paulo, 1980), pp. 66Google Scholar, 83. 84–18.

¹⁸ LaRue, Carl D., The Hevea Tree in the Amazon Valley, USDA Dept. Bulletin No. 1422 (Washington, DC, 1926), p. 60Google Scholar.

¹⁹ Fuentes, Hildebrando, Loreto: Apuntes Geográficos, Históricos, Estadistícos, Políticosy Societies, Tomo I (Lima, 1908), p. 214Google Scholar; Woodroffe, Joseph F., The Upper Reaches of the Amazon (New York, 1914) p. 105Google Scholar.

²⁰ See LaRue, The Hevea Tree, p. 60; and Yungjohnann, John C., White Gold. The Diary of a Rubber Cutter in the Amazon. 1906–1916, Prance, Ghillean T (ed.) (Oracle, 1989), p. 61Google Scholar. Data from the Putumayo and other regions where native peoples worked caucho under abusive overseers suggest much lower yields, between 40 and 120 kg per collector per season. UCongress, ‘Slavery in Peru’, House Documents, vol. 3, 62D Congress, 3d Session, Doc. No. 1366 (1913), pp. 68, 86Google Scholar.

²¹ Whereas the journey from the export houses of Belém at the mouth of the Amazon River to rubber exchanges in Europe or the United States is indeed appreciated as long in both distance and time, it was the voyage between the rubber fields and Belém that reflected the true remoteness and degree of geographical separation between the gatherer and the buyer. A return trip from Belém to the rubber forests of the Beni River would take 270 days (barring misfortune). Pearson, The Rubber Country, p. 121. Such a duration would be the equivalent then of 13 return voyages from Belém to New York.

²² For example, if land is relatively abundant in an agricultural region, then production will tend to be organised in an extensive manner, with choices of crops and techniques of cultivation that take advantage of the abundant land and economise on other more scarce resources. Moreover, the more abundant a factor, the lower a relative return it is likely to command in the sector. So, if land is abundant and labour is scarce, then there should be a premium on labour.

²³ Large areas of untapped trees remained in the basin, particularly in the more remote regions of Brazil, such as Grosso, Matto. Schurz, William L., Hargis, O. D., Marbut, C. F. and Manifold, C. B., Rubber Production in the Amazon Valley, U.S. Department of Commerce Trade Promotion Series, No. 23 (Washington, DC, 1925), p. 39Google Scholar.

²⁴ See Denevan, William M., ‘The Aboriginal Population of A'mazonia’, in Denevan, W. M. (ed.), The Native Population of the Americas, second edition (Madison, 1992)Google Scholar; and Bunker, Stephen G., ‘Modes of Extraction, Unequal Exchange, and the Progressive Underdevelopment of an Extreme Periphery: The Brazilian Amazon, 1600–1980’, American Journal of Sociology, vol. 89, no. 5 (1984), pp. 1029–32CrossRefGoogle Scholar.

²⁵ An informative discussion of labour scarcity is presented by Akers, C. E., Report on the Amazon Valley, Its Rubber Industry and Other Resources (London, 1912), pp. 67–70Google Scholar, 81–2.

²⁶ Santos, História Econômica da Amazônia, p. 111.

²⁷ The extent of the labour scarcity problem can be appreciated by comparing the relative labour market conditions in the Amazon and Asia: in 1912, real wages in the Amazon were slightly more than eight times those in Asia, and this is after the price decline of the 1910s was well under way. Akers, Report on the Amazon, p. 102.

²⁸ Burns, E. Bradford, ‘Manaus 1910: Portrait of a Boom Town’, Journal of Inter-American Studies, vol. 7, no. 3 (1965), p. 402CrossRefGoogle Scholar.

²⁹ Hardenburg, W. E., The Putumayo: The Devil's Paradise (London, 1912)Google Scholar.

³⁰ Murphy, R. and Steward, J., ‘Tappers and Trappers: Parallel Process in Acculturation’, Economic Development and Cultural Change, vol. 4, no. 4 (1956)CrossRefGoogle Scholar.

³¹ See Weinstein, The Amazon Rubber Boom, pp. 35–68.

³² The Madeira River rapids reportedly swallowed u p between 5 and 2 5 % of the rubber shipped and foreign companies refused to insure boats that tempted fate in the rapids. See Fifer, ‘The Empire Builders’, p. 130; India Rubber World, vol. 24, no. 5 (1 Aug. 1901), p. 327; and, U.S. Consular Reports, vol. 66, no. 248 (May, 1901), p. 148.

³³ Schurz et al., Rubber Production, pp. 29–30.

³⁴ See Pearson, The Rubber Country, pp. 38–42.

³⁵ For reports of the number of days actually worked during the tapping season, see Jumelle, Henri, Les Plantes à Caoutchouc et à Gutta: Exploitation, Culture et Commerce dans tous les Pays Chauds (Paris, 1903), p. 86Google Scholar; Fuentes, Loreto, vol. I, p. 212; Lange, Algot, ‘The Rubber Workers of the Amazon’, bulletin oj the American Geographical Society, vol. 43, no 1 (1911), P 34Google Scholar; and Schurz et al., Rubber Production, p. 13.

³⁶ Attrition rates due to death, illness, and desertion were very high indeed. A report from US Consul Kenneday at Belém noted that of 100 workers recruited and sent to the rubber fields, 75 would die, desert or leave because of illness. U.S. Consular Reports, vol. 59, no. 220 (Jan. 1899), p. 70. In one specific case, 300 tappers sent upriver from Belem resulted in a net increase at the end of the year of only 5 tappers to the upriver estate. India Rubber World, vol. 23, no. 2 (1 Nov. 1900), p. 34. Of 350 workers engaged at Belém to work the Madeira–Marmore railway, only 65 reached Pto. Velho. Santos, História Econômica da Amazônia, p. 96. Typically a patron in the Upper Brazilian Amazon could expect to lose 5 out of 25 tappers shortly after arrival on the estate. Woodroffe, The Upper Reaches, pp. 221–2. Other reports speak of high tapper mortality. See Lange, In the Amazon Jungle, pp. 89–90; Woodroffe, The Upper Reaches, pp. 221–2; LeCointe, L'Amazonie Brésilienne, vol. I, pp. 101, 106; India Rubber World, vol. 4 (1891), p. 209 and vol. 23 (1900), p. 34; and Yungjohnann, White Gold, pp. 45, 49, 61. Death rates of 25–30% (and as high as 50%) have been suggested for rubber workers in Bolivia during the boom. Fifer, ‘The Empire Builders’, p. 130. A common rule of thumb used by rubber estate owners when gauging labour requirements was to engage 80 men where the work could be done by 50 able bodies. Schidrowitz, Philip, Rubber (London, 1911), p. 26Google Scholar.

³⁷ India Rubber World ran regular articles attempting to explain the major price swings of rubber and why speculation was probably a minor force relative to the supply shocks that wild rubber extraction experienced. Th e very inelasticity of wild rubber supply (and demand) made the likelihood of large price swings rather high. In Brazil, exchange rates were particularly volatile during the rise of Republicanism in the 1890s.

³⁸ Weinstein, The Amazon Rubber Boom, p.30.

³⁹ Tappers changed estates for a variety of reasons, from seeking the protection of a new patron from an old patron bent on reclaiming a debt to the prospect of working on better estates. Estate owners had difficulty attracting tappers where trees were no longer highly productive. India Rubber World, vol. 25, no. 2 (1 Nov. 1901), p. 46.

⁴⁰ Each year Cearense would leave their homes and go up to rubber fields, gather and sell rubber, and take home pay to their families. Over time, many Cearense, wh o originally migrated annually, settled in upriver areas. India Rubber World, vol. 30, no. 3 (1 Jun e 1904), p. 298. India Rubber World, vol. 41, no. 2 (1 Nov. 1909), p. 45. Nevertheless, seasonal migration from estate to town continued. One observer wrote, ‘En toda la Amazonía las exigencias de la industria gomera‥hace que las poblaciones de las ciudades principales como el Pará, Manaos e Iquitos: y de otras secundarias como Santaren [sic], Obydos, Teffé‥Contámana, etc …tengan dos períodos bien distintos, que son: uno de completo lleno en su població y otro de mengua en ella: coincidiendo el primero con‥las lluvias frecuentes, las crecientes de los ríos y las innundaciones de los terrenos bajos‥saliendo entonces todo el personal seringuero, y cauchero, después la yafra, del fondo de las selvas a las cuidades‥’. Herrera, Genaro, ‘Censo Urbano de Iquitos’, Boletín de la Sociedad Geográfica de Lima, Tomo XXX, Trim. I & II (1914), p. 49Google Scholar. Another writer spent the off-season (five months) with tappers in Remate de Malas (Benjamin Constant) on the Yavari River observing their habits and accompanied them back upriver to work on their respective estates. Lange, In the Amazon Jungle, p. 97. Reports of the size of Amazon river towns and cities often present a range of populations, depending on whether the tappers were in town. US Consul Fuller wrote, ‘[t]he population of Iquitos is variously estimated at 12,000 to 15,000. In March, April, and May, when the rubber gatherers are not at work in the forest, it will run up to 20,000’. Fuller, Stuart J., ‘Trade in Peruvian Rubber District’, U.S. Daily Consular and Trade Reports, no. 274 (20 11. 1912), p. 913Google Scholar. The population of Contámana (Ucayali River, Peru) would swell from about 500 during low water to 1500 when the tappers returned. Fuller, ‘Trade in Peruvian Rubber District’, p. 926.

⁴¹ Weinstein, The Amazon Rubber Boom, p. 30.

⁴² More generally, transaction costs are considered to be the costs of running an economic system, ‘the economic equivalent of friction in physical systems’. Oliver Williamson, E., The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting (New York, 1985), p. 18–9Google Scholar. Transaction costs are distinguished from direct production costs as the costs that come, in some sense, prior to and after production. Prior costs are associated with finding the agent(s) with whom exchange or transactions will occur and then drafting, negotiating, and safeguarding an agreement with them. Costs that arise after production include the process of correcting misalignments, the set up and maintenance costs of structures to govern disputes, and the bonding costs of securing commitments.

⁴³ DeKalb, Courtenay, ‘The Business of Rubber Gathering in the Amazon Valley, and its Development by American Capitalists’, India Rubber World and Electrical Trades Review, vol. 2, no. 3 (15 06 1890), p. 192Google Scholar.

⁴⁴ Russan, Ashmore, ‘Working Rubber Estates in the Amazon’, India Rubber World, vol.27, no. 1 (1 10 1902), p. 6Google Scholar.

⁴⁵ Pearson, The Rubber Country, pp. 77, 142, 160.

⁴⁶ For example, the Peruvian Amazon Company (J. C. Arana) held definitive title to 5.75 million hectares in the Putumayo region. Schurz et al., Rubber Production, p. 364.

⁴⁷ For a map of property holdings in one Amazonian river basin during the boom, see Coomes, Oliver T., ‘Making a Living in the Amazon Rain Forest: Peasants, Land, and Economy in the Tahuayo River Basin of Northeastern Peru’, unpubl. PhD. diss., University of Wisconsin-Madison, 1992, p. 134Google Scholar.

⁴⁸ Land holdings in the Brazilian districts of Breves, Melgaço and Anajas (near Belém) consisted of a large number of small estates (i.e. less than 600 ha, or equivalent to not more than a few estradas) apparently worked by independent tappers of humble origins. Weinstein, The Amazon Rubber Boom, pp. 45–7. Estates along the Amazon River in the vicinity of Iquitos typically contained only 21 estradas. Derived from Fuentes, Loreto, Tomo II, pp. 8I–3.

⁴⁹ Fifer, ‘The Empire Builders’, p. 141.

⁵⁰ For example, see Wolf, Howard and Wolf, Ralph, Rubber: A Story of Glory and Greed (New York, 1936)Google Scholar; Collier, Richard, The River that God Forgot: The Story of the Amazon Rubber Boom (London, 1968)Google Scholar; and Bakx, Keith, ‘From Proletarian to Peasant: Rural Transformation in the State of Acre, 1870–1986’, The Journal of Development Studies, vol. 24, no. 1 (1988)CrossRefGoogle Scholar.

⁵¹ The contrast between upriver and downriver experiences of hevea rubber tappers, particularly with respect to tenure patterns and tapper welfare, is noted in Barbara Weinstein, ‘The Persistence of Precapitalist Relations of Production in a Tropical Export Economy: The Amazon Rubber Trade, 1850–1920’, in Hanagan, Michael and Stephenson, Charles (eds.), Proletarians and Protest: The Roots of Class Formation in an Industrializing World (New York, 1986), pp. 63–5Google Scholar; and Weinstein, The Amazon Rubber Boom, pp. 45–8.

⁵² For example, see Yungjohnann, White Gold.

⁵³ In Malaysia, over-exploitation of hevea trees – known as ‘slaughter tapping’ – is most common on plantations where tappers pay a fixed rent. Chew, Tek-Ann, ‘Share Contracts in Malaysian Rubber Smallholdings’, hand Economics, vol. 67, no. 1 (1991), p. 86Google Scholar.

⁵⁴ Moral hazard refers to ‘the possibility that insureds will fail to take appropriate loss-mitigating actions in the insurance interval and will not candidly accept accountability’. Williamson, The Economic Institutions, p. 51. The same author suggests that a broader definition would include ex ante and ex post efforts to ‘lie, cheat, steal, mislead, disguise, obfuscate, feign, distort, and confuse’. Ibid., p. 51.

⁵⁵ In addition to moral hazard, patrons had to be concerned for their very security. One report notes the skill and propensity of Cearense to use a knife; any estate manager that treated them unfairly put himself at great risk. India Rubber World, vol. 41, no. 2 (1 Nov. 1909), P. 45.

⁵⁶ See Fuentes, Loreto, Tomo I, p. 214.

⁵⁷ Discussed in some detail by Hardenburg, The Putumayo: The Devil's Paradise; ‘Correspondence Respecting the Treatment of British Colonial Subjects and Native Indians Employed in the Collection of Rubber in the Putumayo District’, House of Commons Sessional Papers (Session 14 Feb. 1912 to 7 March 1913) vol. 68, misc. no. 8; Taussig, Michael, ‘Culture of Terror-space of Death: Roger Casement's Putumayo Report and the Explanation of Terror’, Comparative Studies in Society and History vol. 26, no. 3 (1984)CrossRefGoogle Scholar; and Taussig, Michael, Shamanism, Colonialism, and the Wild Man: A Study in Terror and Healing (Chicago, 1987)CrossRefGoogle Scholar.

⁵⁸ Woodroffe, The Upper Reaches, p. 160.

⁵⁹ To finance a team of 15–50 caucheros c. 1906 in Peru required 20,000–50,000 Soles (US $9,750–24,333) Fuentes, Loreto, Tomo I, p. 214.

⁶⁰ The Peruvian Amazon Company invested £1,000,000 ($4,866,500) in the Putumayo estates in establishing villages, outposts, trails, and steamer stops as well as in the steamers that plied the waters of this closed river. On the Ucayali River, C. F. Fitzcarrald is attributed with creating an extensive, though relatively short-lived, system of some 200 steamer stops and villages linked by his own steamships.

⁶¹ Plane, Auguste, A Trovers l'Amérique Equatoriale: Le Pérou (Paris, 1903), p. 317Google Scholar.

⁶² About £1,000 (US $4,867) was required typically to open a 50 estrada estate. Woodroffe, The Upper Reaches, pp. 215–16. The cost of labour recruitment and transport ranged from £20 to £100 per tapper. Akers, Report on the Amazon, p. 69. Santos, História Econômica da Amazônia, p. 166. Each worker would receive rubber extraction tools valued at $100–250, and supplies for the six-month season worth $300–500. Derived from Santos, História Econômica da Amazônia, pp. 149, 166; Loureiro, A Grande Crise, pp. 109–10; and Woodroffe, Joseph F. and Smith, Harold H., The Rubber Industry of the Amazon, and How its Supremacy can be Maintained (London, 1915), p. 52Google Scholar.

⁶³ In the case of caucho extraction, advanced provisions were not essential where the product was collected by native people who had a secure source of subsistence foods and materials, and trade was based on the barter of manufactured goods for rubber. In some cases, extra-economic forces bound tribal people to the patron or trader. On the Ucayali River (c. 1880–98), natives worked caucho under C. F. Fitzcarrald who they regarded as ‘El hijo del Sol’. Reyna, Ernesto, Fitzcarrald, el Key del Caucho (Lima, 1942), p. 22Google Scholar. On the Putumayo, mutual fear and extreme violence linked native collectors and overseers. Taussig, ‘Culture of Terror-space’; Taussig, Shamanism, Colonialism, and the Wild Man. Nevertheless, in both cases economic relations persisted as tribal peoples exchanged caucho for goods.

⁶⁴ See DeKalb, ‘The Business of Rubber’, p. 192; India Rubber World, vol. 24, no. 5 (1 Aug. 1901), p. 328; Russan, ‘Working Rubber Estates’, pp. 6–7; Cooper, Clayton S., The Brazilians and their Country (New York, 1917), p. 327Google Scholar; Melby, J. F., ‘Rubber River: An Account of the Rise and Collapse of the Amazon Boom’, Hispanic American Historical Review, vol. 22, no. 3 (1942), pp. 464–7CrossRefGoogle Scholar; and Weinstein, ‘The Persistence of Precapitalist Relations’, pp. 67–8. For accounts of the difficulties encountered by more recent efforts in Amazonian rubber development, see Dean, Warren, ‘The Rubber Development Schemes of the United States in the Brazilian Amazon, 1939–1956’, in Anais (Rio de Janeiro, 1992)Google Scholar; and Dean, Warren, Brazil and the Struggle for Rubber (New York, 1987)Google Scholar.

⁶⁵ The credit contract is often described as a form of ‘debt peonage’ or ‘slavery’. See Wolf and Wolf, Rubber, p. 34; Chirif, Alberto and Mora, Carlos, ‘La Amazonia Peruana’, in Historia del Perú, Baca, Juan Mejía (ed.), Tomo XII (Lima, 1980), p. 285Google Scholar; Coates, The Commerce in Rubber, pp. 94–6; Hecht, Susanna and Cockburn, Alexander, The Fate of the Forest: Developers, Destroyers and Defenders of the Amazon (New York, 1989), p. 62Google Scholar; Bakx, ‘From Proletarian to Peasant’, pp. 147–8; and Fearnside, ‘Extractive Reserves’, p. 388. The trading (‘truck’) system is also ill viewed, being ‘‥nothing more than organized robber on a huge and particularly cruel scale’. Woodroffe and Smith, The Rubber Industry, p. 137. Even today, river traders draw the scorn of producers and domestic consumers of agricultural and forest products in the port cities of Amazonia. Padoch, Christine, ‘Risky Business’, Natural History, vol. 96, no. 10 (1987), P. 59Google Scholar.

⁶⁶ See Weinstein, The Amazon Rubber Boom, pp. 96, 263–5; Weinstein, Barbara, ‘Capital Penetration and Problems of Labor Control in the Amazon Rubber Trade’, Radical History Review, vol. 27 (1983), pp. 135–6Google Scholar; and Schmink and Wood, Contested Frontiers, p. 45.

⁶⁷ Fuller, ‘Trade in Peruvian Rubber’, pp. 924–7.

⁶⁸ Only modest capital barriers faced entrants to the rubber trade (i.e. a boat and provisions), and itinerant traders flourished around the port cities, towns and steamer stops. Vociferously denounced as pirates and unscrupulous interlopers, efforts to curb itinerant traders were generally unsuccessful in part because of support given by urbanbased merchants. Weinstein, The Amazon Rubber Boom, pp. 51–2. Numerous accounts from the period note the success of itinerant traders in exchanging their wares for hevea and caucho, and of the frustration of foreign and local concerns in limiting the leakage of rubber out of the traditional aviamento system. See for example, DeKalb, ‘The Business of Rubber Gathering’, p. 192. Competition stimulated by itinerant traders rather than brute force was probably the more successful strategy for maintaining flows of rubber and merchandise within credit lines.

⁶⁹ In caucho extraction, the debt-merchandise contract often included a share of the final product for the gatherer. Unlike hevea, caucho could not be sustainably tapped and the tree was felled to extract the latex. Credit was advanced by the patron in the form of supplies to team members and the harvested caucho was sold or exchanged for goods along the rivers or in the port cities. Whereas a team of international fortune seekers may have kept a high share of the product, the indebted local worker probably retained a smaller portion of the surplus. The ‘credit plus share’ contract enabled the caucho worker to travel and subsist for long periods in the forest and provided incentives to maximise the harvest and ensure delivery of all of his product.

⁷⁰ See Chew, ‘Share Contracts’.

⁷¹ When rubber prices plummeted in the 1910s and 1920s, other arrangements for rubber extraction became more feasible. Reports on conditions in the rubber fields during the 1920s indicate that the debt-merchandise contract in many areas was replaced by share and fixed rent contracts. See Schurz et al., Rubber Production, p. 27. When rubber prices fell, other forest and agricultural products, such as Brazil nuts (Brazil), cotton (Peru), and even subsistence agricultural products, became competitive and estates diversified production. See Coomes, ‘Making a Living in the Amazon’, pp. 138–44, for a case study from northeastern Peru. Diversification and the rise of agriculture effectively redefined the relative importance of the factors of production: land became the primary determinant of returns and control over access to land, rather than capital or labour, now shaped labour relations. On many estates, debt-merchandise contract tappers became tenants or share tappers and this transition was a response to major changes in the prevailing economic environment, specifically the preciptous decline of rubber prices.

⁷² Calculated as i=(d+u/I–u) where i is the nominal interest rate, d is the desired rate of return on investment, and u is the expected rate of loan default. Administrative costs associated with the loan (i.e. opportunity cost on lender's time), assumed here to be zero, would raise the interest rate beyond the level set by the risk premium. For example, an administrative cost of 20% would raise the rate from 37.5% to 62.5%. For an explanation of rate determination that incorporates both the risk premium associated with default and administrative costs of lending and collecting, see Bottomley, A., ‘Interest Rate Determination in Underdeveloped Rural Areas’, American Journal of Agricultural Economics, vol. 57, no. 1 (1975), p. 282CrossRefGoogle Scholar.

⁷³ This notion underlies the design of insurance systems. When a large number of independent alternatives are pooled, the risk of large losses is reduced to a small probability, because the joint probability of independent events is the product of the individual probabilities. In our example, if each tapper has a one in five chance of dying or deserting, and if each loss is viewed as an independent phenomenon, then the risk to the patron of losing his entire working capital investment is 0.2 to the tenth power or one in ten million. For the trader with 100 tappers in his scheme, the probability of total default will be minuscule. The greater the number of independent events, the more the range of outcomes will be squeezed around the mean, by the reduction in probability of big losses or big gains.

⁷⁴ Santos, História Econômica da Amazônia, p. 299.

⁷⁵ Pearson, The Rubber Country, pp. 58, 98, 142, 160, 163, 170.

⁷⁶ Barham and Coomes, ‘Reinterpreting the Amazon Rubber Boom’.

⁷⁷ Labour was drawn even out of activities essential to the rubber trade. The US Consul at Belém wrote early during the boom, ‘‥the inhabitants along the Amazon being so absorbed in the production of rubber, owing to its recent rise in value, that labor enough cannot be had to furnish the steamers with wood’, U.S. Consular Reports, vol. 12, no. 39 (March 1884), p. 204.

⁷⁸ The movement of men upriver to the rubber fields was massive. Consul Kenneday reported on 23 September 1903 that 5,000 men had left Belém over the previous seven days for rubber fields upriver. U.S. Monthly Consular Reports, vol. 74, no. 280 (Jan. 1904), p. 205.

⁷⁹ See Lange, ‘The Rubber Workers’, p. 36. A vivid description of Manaus at the peak of the boom is provided by Burns, ‘Manaus 1910’.

⁸⁰ Santos, Historia Econômica da Amazônia, p. 113.

⁸¹ U.S. Daily Consular and Trade Reports, no. 271 (16 Nov. 1912), p. 856.

⁸² Wolf and Wolf, Rubber, p. 66.

⁸³ Ideally, the atgument that tappers earned sufficient returns to allow surplus accumulation would be based on a full financial accounting of the credit contract over time, from the initial financing to eventual resolution. We encountered no complete tapper-patron financial accounts in the literature (e.g. cost of merchandise, merchandise delivered, rubber credited at relative market prices, possible off-season income or cash sales to itinerant traders, credit advanced for off-season subsistence on the estate or in town) that would allow accurate estimation of tapper returns. Instead, several reports of ‘average’ earnings by tappers were found in which ‘typical’ conditions were not carefully specified. See Loureiro, A Grande Crise, pp. 54–5, 109–10; Woodroffe, The Upper Reaches, p. 221; Jumelle, Les Plantes à Caoutchouc, pp. 96–100; Lange, ‘The Rubber Workers’, pp. 33–6; Santos, História Econômica da Amazônia p. 166; and de Rivière, Baron H. Arnous, ‘Explorations in the Rubber Districts of Bolivia’, Journal of the American Geographical Society of New York, vol. 32 (1900), p. 437CrossRefGoogle Scholar. Few reports are explicit, for example, about prices paid and received among trader, patron, and tapper: was the $300 of goods advanced to the tapper the cost to the patron or the charge levied by the patron to the tapper (i.e. including the 100% markup commonly charged)? Strong spatial and temporal variations in tree yields, goods and rubber prices, off-season activity, cash sales to itinerant traders, and tapper mortality also limit the representativeness of these estimates. With gross tapper earnings of $350–600 per season, assumptions about specific prices, yields, accounting procedures, etc. make the difference between the tapper realising a healthy economic surplus, breaking-even, or spiralling deeper into debt. A rather unique set of primary data is needed to construct reliable estimates of tappers’ returns, preferably from the records of a number of estates or trader zones that would include a clear definition of property relations, prevailing prices, yields, and extractive and productive practices.

⁸⁴ Akers, Report on the Amazon, p. 81.

⁸⁵ One visitor reported tappers earning the equivalent of $8–10 per day on the Yavari River in 1910. Lange, In the Amazon Jungle, p. 47. Another report suggests that a tapper could make £50–60 ($243–292) per season in net revenue. Schidrowitz, Rubber, p. 27. A similar figure is reported by Bonnechaux, M. ($290–347 per 100 day season) in U.S. Consular Reports, vol. 67, no. 255 (12 1901), p. 575Google Scholar.

⁸⁶ For one such case, see Weinstein, The Amazon Rubber Boom, p. 24.

⁸⁷ Loureiro, A Grande Crise, pp. 109–10; Jumelle, Les Plantes à Caoutchouc, pp. 96–100; and Woodroffe, The Upper Reaches, p. 221.

⁸⁸ See India Rubber World, vol. 40, no. 4 (1 July 1909), p. 348.

⁸⁹ Coomes and Barham, ‘The Amazon Rubber Boom’; and Barham and Coomes, ‘Reinterpreting the Amazon Rubber Boom’.

⁹⁰ Barham and Coomes, ‘Reinterpreting the Amazon Rubber Boom’.