2002年-世界发展银行全球_The_Distributional_Impacts_of_Indonesias_Financial_Crisis_on_Household_Welfare___A_Rapid_Response_Methodology_27页_752kb

Summary of "The Distributional Impacts of Indonesia's Financial Crisis on Household Welfare: A 'Rapid Response' Methodology"

Core Content

This article examines the distributional effects of the 1997 Asian currency crisis on household welfare in Indonesia using a "rapid response" methodology. The authors focus on estimating the compensating variation—the amount of money needed to maintain pre-crisis utility levels after price changes—without relying on post-crisis household data, which is often delayed.

The methodology is based on pre-crisis consumption data and province-level commodity price changes. It uses nonparametric techniques, specifically locally weighted least squares, to explore how price changes affected different income and geographic groups. The approach allows for immediate analysis during economic crises, when timely data are critical for policy intervention.

Main Points

1. Impact of the Crisis on Household Welfare

• The Indonesian rupiah collapse led to a 12% GDP decline and rampant inflation.
• Food prices nearly tripled in 18 months, with nonfood prices also rising, though less sharply.
• The urban poor were most affected, while poor rural households were somewhat protected due to their ability to produce food.

2. Compensating Variation Measures

• Two methods are used to estimate compensating variation:
  • Expression (2): A first-order approximation that considers budget shares and proportional price changes.
  • Expression (5): A second-order approximation that includes substitution effects via price elasticities.
• The first-order method is simpler but does not account for substitution behavior, potentially overestimating the impact.
• The second-order method is more accurate as it incorporates Slutsky derivatives and compensated price elasticities.

3. Data and Methodology

• SUSENAS (1996 National Socio-Economic Survey) provides pre-crisis consumption data for 61,965 households.
• BPS (Indonesian Central Statistical Office) provides province-level price data for 44 cities from January 1997 to October 1998.
• The methodology uses unit values (expenditure per unit consumed) as a proxy for prices, and clustered data to estimate price elasticities.
• The Deaton method is used to estimate own-price and cross-price elasticities from the unit value data, assuming constant prices within clusters.

4. Key Findings

• Rice was the most important commodity for households, with budget shares ranging from 4.8% to 26.9%.
• Rice prices increased by 195.2%, with a standard deviation of 29.2%, indicating high variation in price changes.
• Housing and utilities had the least price increase (23.8%).
• Households with young children were disproportionately affected.
• Geographic location significantly influenced the impact of price changes, even within urban or rural areas and income categories.
• Self-produced food and owner-occupied housing are treated as negative expenditures to account for their benefits.

Key Information

• Data Sources:

  • SUSENAS: Pre-crisis consumption data.
  • BPS: Monthly price data for 44 cities.
  • Unit values: Used as proxies for prices.
  • Clustered data: Used to estimate price elasticities.

• Methodological Innovations:

  • Nonparametric analysis (locally weighted least squares) to assess compensating variation across income levels.
  • Rapid response approach: Avoids reliance on delayed post-crisis data.
  • Use of budget shares and price changes: To estimate welfare impacts without post-crisis consumption data.

• Limitations and Considerations:

  • The methodology is limited to pre-crisis data, which may not capture all post-crisis behaviors.
  • Price data are city-based, which may not fully represent rural price changes.
  • Aggregation is necessary due to limited data, but it may obscure heterogeneity in consumption patterns.

• Policy Implications:

  • The urban poor were most vulnerable to the crisis.
  • Substitution effects are important in mitigating the impact of price shocks.
  • Geographic and demographic factors must be considered when analyzing distributional impacts.

Conclusion

The study demonstrates that the Indonesian financial crisis had widespread negative impacts on household welfare, with the urban poor suffering the most. The rapid response methodology offers a timely and practical approach for analyzing the distributional consequences of economic crises, even in the absence of post-crisis household data. It highlights the importance of consumption patterns, price changes, and substitution behavior in understanding how crises affect different segments of the population.