2009年-世界发展银行全球_Poverty_in_Latin_America___Sources_Of_Welfare_Disparities_in_Ecuador_32页_723kb

Summary of "Sources Of Welfare Disparities in Ecuador"

Core Content

This policy research working paper investigates the sources of welfare disparities in Ecuador, focusing on the role of portable characteristics (human capital) and geography in explaining differences in living standards between urban and rural areas. The paper contributes to the broader understanding of spatial poverty in Latin America and provides an applied framework to analyze the relative importance of these factors.

Main Views

The paper presents two contrasting views on the causes of welfare disparities:

1. Concentration View: Poor areas are the result of persistent concentration of individuals with personal attributes that inhibit growth in living standards. According to this view, geography does not play a causal role in explaining poverty levels or growth.
2. Geography View: Geographic factors themselves are the cause of high poverty and weak growth in living standards. Local public goods and infrastructure can influence the ability of poor households to improve their welfare.

The paper argues that the geography view is more relevant in explaining the welfare differential in Ecuador, especially in rural areas, where geographic externalities such as access to infrastructure and services play a significant role.

Key Findings

• Education is the most important factor in explaining differences in living conditions between urban and rural areas. It accounts for a large portion of the welfare differential.
• At the national level, portable characteristics (non-geographic variables) explain 72% of the welfare gap between urban and rural areas, while returns to these characteristics explain 28%.
• When comparing urban and rural areas within regions, the results show:
  • In urban areas, portable characteristics explain 90% of the welfare differential between the leading (coast) and lagging (Amazon) regions.
  • In rural areas, returns to characteristics explain 30% of the welfare differential between these regions.
• Demographics and education are the primary contributors to the urban-rural welfare gap, with education accounting for 68.5% and demographics for 32%.
• House ownership and occupation also play a role, though less significant than education.
• Pichincha province (which includes Quito) is found to have higher welfare levels than other provinces, both in urban and rural areas, after controlling for non-geographic variables.

Data and Methodology

• The study uses data from the Living Conditions Survey (ECV) for 2005-2006, collected by Ecuador's National Statistics and Census Institute (INEC).
• The dataset is representative at the national level and includes urban and rural areas.
• A linear regression model is used to estimate the contribution of non-geographic characteristics and their returns to welfare.
• The Oaxaca-Blinder decomposition is applied to break down the differences in welfare between urban and rural areas into contributions from characteristics and returns.

Regional Analysis

• The paper extends the analysis to three regions: Sierra, Costa, and Oriente.
• Sierra and Costa are found to be more similar in terms of characteristics and returns compared to Oriente, which is mainly rural and shows different patterns.
• The Oriente region has a higher proportion of rural households (61%) and exhibits significant differences in welfare due to both characteristics and returns.

Conclusion

The study concludes that portable characteristics such as education and demographics play a dominant role in explaining welfare disparities in Ecuador, especially in urban areas. However, geographic factors are also important, particularly in rural areas where access to infrastructure and services affects the returns to characteristics. The paper highlights the need for targeted poverty reduction strategies that consider both human capital and geographic conditions.

Key Information

• Population: Over 13.75 million in 2008, with nearly 60% in urban areas.
• Economic structure: Heavily dependent on oil exports, which account for an average of 35% of GDP.
• Poverty rate: Increased to 45% in 2001 due to the 1999 economic crisis, then decreased to around 36% by 2004.
• Urban-Rural Poverty Gap: In 2004, rural poverty rate was more than twice the urban rate.
• Methodology: Uses a linear model and Oaxaca-Blinder decomposition to assess the impact of non-geographic variables and geographic dummies on welfare.
• Geographic Dummies: Includes 21 provinces (excluding Galápagos), with Pichincha as the base category.
• Variables Analyzed:
  • Demographics (number of babies, children, teenagers, adults; age of head; marital status; ethnicity)
  • Education (levels of education for head and spouse)
  • House ownership
  • Occupation status

Summary of Decomposition Results

• Demographics account for 32% of the urban-rural welfare differential.
• Education accounts for 68.5% of the differential.
• Geographic variables have a smaller impact, with the difference between Pichincha urban and rural areas being of similar magnitude.
• Occupation and house ownership contribute less than education but still play a role in explaining welfare differences.