2014年-世界发展银行全球_New_Coincident_and_Leading_Indicators_for_the_Lebanese_Economy_30页_1mb

Summary of "New Coincident and Leading Indicators for the Lebanese Economy"

Core Content

This paper introduces two new economic indicators for Lebanon: the World Bank Coincident Indicator (WB-CI) and the World Bank Leading Indicator (WB-LI). These indicators aim to provide timely and accurate measures of economic activity in the absence of reliable and comprehensive official statistics. The methodology is based on the National Bureau of Economic Research (NBER) and Conference Board (CB) approach, with modifications to improve statistical properties through minimization and calibration techniques.

Main Points

Weakness of Economic Statistics in Lebanon

• Lebanon suffers from weak and infrequent economic statistics, which hinders timely economic analysis and decision-making.
• Key areas with statistical deficiencies include national accounts, balance of payments, prices, inflation, and labor and social measures.
• Reliable GDP data is only available annually up to 2011, and the Consumer Price Index (CPI) was updated in December 2013.
• Official unemployment data is outdated, with the last update from 2009.

Existing Indicators and Their Limitations

• The BdL-CI, a coincident indicator developed in 1993, uses eight variables but has fixed weights and does not account for all economic sectors.
• The IIF-CI, developed in 2010, includes five additional variables but still has issues with accuracy.
• Both indicators have been criticized for lacking statistical rigor and not being fully objective in weight selection.

Purpose of WB-CI and WB-LI

• The WB-CI and WB-LI are designed to offer more accurate and timely proxies for GDP growth.
• The WB-CI is a monthly coincident indicator that closely tracks real GDP growth with a time lag of four to five months.
• The WB-LI is a leading indicator that aims to forecast GDP growth one year ahead, providing early signals of economic turning points.

Key Information

WB-CI Construction

• Variables: 21 potential variables were considered, covering real, external, monetary, and fiscal sectors.
• Selection: Only 13 variables were ultimately used, chosen based on both statistical and economic significance.
• Methodology:
  • Seasonal trends are removed using the X-12-ARIMA technique.
  • Variables in monetary terms are deflated using the CPI.
  • Monthly data is smoothed using moving averages.
  • Symmetric percentage changes are calculated for each variable.
  • Weights are calibrated to minimize the difference between the WB-CI growth rate and the actual GDP growth rate.
• Time Frame: Data spans from December 2004 to December 2011 (85 observations).

WB-LI Construction

• Purpose: To forecast GDP growth one year ahead and detect early turning points in the economy.
• Methodology:
  • Based on the NBER-CB approach, but with a focus on minimizing the forecast error between WB-LI and WB-CI one year later.
  • Weights are selected to ensure that the WB-LI growth rate aligns as closely as possible with the WB-CI growth rate 12 months ahead.
• Variables: 17 potential variables were used, selected based on both statistical and economic significance.
• Statistical Significance: Variables must have a correlation coefficient with WB-CI at time $ t + 12 $ greater than 0.5 in absolute value.
• Economic Significance: Variables should reflect expectations or responses to economic shocks.

Policy Implications

• The WB-CI and WB-LI can serve as valuable tools for policymakers, investors, and businesses to monitor and forecast economic activity.
• These indicators may help in making more informed decisions by providing timely and accurate data in the absence of robust official statistics.
• The WB-LI is particularly useful for anticipating changes in the business cycle and planning accordingly.

Limitations and Caveats

• The accuracy of these indicators is currently limited due to the short sample period and lack of comprehensive data.
• The WB-CI and WB-LI may not fully capture all economic sectors, especially those with limited data availability.
• The effectiveness of the indicators may improve as more data becomes available over time.

Conclusion

• The WB-CI and WB-LI offer promising alternatives to existing indicators, improving the timeliness and accuracy of economic analysis in Lebanon.
• These indicators are based on a modified NBER-CB approach, with weights calibrated to better align with GDP growth patterns.
• While the current results are encouraging, the indicators require further refinement and validation with more extensive data.

References and Appendices

• The paper includes several tables and figures to support the methodology and results.
• Tables present the potential variables, their significance, and the final set used in constructing the indicators.
• Figures illustrate the performance of the indicators against GDP growth and their ability to detect turning points in economic activity.

Tables Summary

• Table 1: Lebanon's GDP decomposition from the supply side.
• Table 2: Potential candidates for the WB-CI.
• Table 3: Economic and statistical significance of WB-LI variables.
• Table 4: Potential candidates for the WB-LI.
• Table 5: Final variables used in WB-CI.
• Table 6: Final variables used in WB-LI.
• Table 7: Error comparison between GDP growth and the indicators.
• Table 8: Impact of shocks on WB-CI growth rate.
• Table 9 and 10: Monthly data for WB-CI and WB-LI.
• Table 11: Unit root test results.
• Table 12: Granger causality test results.

Figures Summary

• Figure 1: Performance of existing coincident indicators in Lebanon.
• Figure 2: Growth rate of WB-CI aligns with GDP growth.
• Figure 3: WB-LI detects turning points 12 months ahead.
• Figure 4: WB-LI is an accurate forecast tool.
• Figure 5: Economic activity is volatile and influenced by political and security shocks.
• Figure 6 and 7: Impulse response functions and variance decomposition of WB-CI growth rate.