Oil Extraction and Local Social Development in Ecuadorian Amazon

Abstract

Large oil reserves were discovered in northern Ecuadorian Amazon region in 1967. In 1972 the country became an oil exporter, and ever since crude petroleum has been the backbone of the national economy. According to different studies, the national effects of oil extractivism have been detrimental for socially inclusive and sustainable development. Economic growth has been unstable and low, achieved economic diversification is minimal, poverty still affects to about a third of the population, underemployment accounts for about half of the labor force, and the environmental effects of oil extraction (deforestation, pollution, biodiversity loss, oil spills) have been severe. Most of the research about oil and development in Ecuador has been done from national or internationally comparative perspectives, and some studies were focused on local social conflict in extractive areas. This article complements former research, by focusing on local social effects of oil extraction, using census data and other geographical sources with a high level of spatial resolution. An initial analysis was based on the 1990, 2001 and 2010 population censuses as the main social information source. A social development index (SDI) was elaborated, using principal components analysis, from 19 social indicators on education, health, housing and gender differences, broken down by parishes, the smallest administrative division in Ecuador. The Amazon consistently has been the most socially deprived region in Ecuador, both in the urban and rural areas. Moreover, differentiating between oil extracting zones and the rest of the Amazon, the social index remains lower in the oil extraction region, both in urban and rural areas. The most critical social indicators regard child mortality and access to health services, while differences in education are less severe. A deeper analysis was done on explanatory factors leading to social distribution, breaking down the SDI at the census track level, and performing a spatial autoregressive multiple regression model with the SDI as the dependent variable, and selecting as independent variables oil extraction proximity, soil fertility, access to markets, proportion of deforested areas, a dummy for rural tracks, and 3 indicators of employment structure (proportions of agriculture, wage earners, and tourism in the labor force). All independent variables reached statistically significant coefficients. The oil proximity indicator had a negative effect on social wellbeing, and deforestation, included in quadratic form, presents an initial small positive effect on social conditions, which vanishes as deforestation increases over 65% of the area. All the remaining regression coefficients had the expected signs. The main regression findings are: a. After controlling for all relevant variables with available information, oil has a net negative effect on the local social development index. b. The social improvement linked to deforestation, other things being equal, is small, subject to decreasing returns, and disappear in advanced stages of the process. c. Economic diversification towards tourism has a high social redistribution effect. As remaining oil reserves in Ecuador are low, and the environmental impact of deforestation is severe, a strategy towards alternative development paths in Ecuador’s Amazon is a priority.