巴德学院利维经济研究所-气候变化与财政策略（英）-2023.7-19页_481kb

报告摘要

Summary of "Climate Change and Fiscal Marksmanship: Evidence From an Emerging Country, India"

Introduction and Context

The paper analyzes fiscal marksmanship, defined as the accuracy of budgetary forecasting, in the context of climate change-related public spending in India. Climate change spending includes adaptation and related programs. Fiscal rules, such as the Fiscal Responsibility and Budget Management Act (FRBMA), stipulate a 3% fiscal deficit-to-GDP ratio, with a post-COVID consolidation path aiming for 4.5% by 2025-26. Fiscal marksmanship is crucial for rational expectations and effective fiscal policy.

Key Objectives

Assess budget forecast errors for climate change spending from 2017-18 to 2020-21 across various ministries.
Decompose errors into bias, systematic variation, and randomness to identify improvable aspects.
Evaluate subnational fiscal space for climate change commitments.

Literature Review Summary

Prior studies on fiscal marksmanship in India show higher systematic errors before 1990, shifting to random errors afterward, primarily using federal data. Subnational analyses are limited, with most empirical work focused on national levels. European studies highlight political, institutional, and economic factors influencing forecast errors under fiscal rules like the Stability and Growth Pact.

Methodology

Errors are calculated using budget estimates (BE), revised estimates (RE), and actuals.
Key measures include:
- Mean Error (ME): Average difference between forecast and actual (indicator of bias).
- Root Mean Square Error (RMSE): Magnitude of errors, scaled to actuals.
- Theil's Inequality Coefficient (U): Decomposes errors into bias proportion, unequal variation, and random components.
Data sources include Government of India finance accounts and Reserve Bank of India analyses.

Key Findings

Forecast Errors: For climate adaptation spending, ministries like Consumer Affairs, Food and Public Distribution, and Science and Technology show high Theil's U values (e.g., 0.496 for bias in the former). Systematic errors are prevalent and can be reduced through better forecasting methods.
Error Decomposition: Systematic errors (bias and unequal variation) dominate in most ministries, while random errors are less significant but harder to predict.
- BE and RE errors highlight areas with field-specific issues, such as tax autonomy in states affecting subnational forecasting.
Subnational Fiscal Space: State-level developmental spending constitutes ~60% of total public expenditure, with potential discretionary space for climate commitments identified at the aggregate state level.
Policy Implications: Fiscal marksmanship can be improved via policy innovations within fiscal rules, such as better data use and forecasting techniques. Subnational analyses require resolving intertemporal comparability issues.

Conclusion

The paper emphasizes that enhancing fiscal marksmanship is vital for effective climate policy under tight fiscal constraints. Higher systematic errors should be targeted for reduction through refined fiscal models. Future research should focus on subnational fiscal space decomposition and expanding analyses to include post-COVID fiscal strategies.