2017年-世界发展银行全球_The_Impact_of_Electricity_Shortages_on_Firm_Productivity___Evidence_from_Pakistan_26页_971kb

Summary of "The Impact of Electricity Shortages on Firm Productivity: Evidence from Pakistan"

Core Content

This policy research working paper examines the impact of electricity shortages on firm productivity in Pakistan using data from 4,500 manufacturing firms surveyed in 2010-2011. The study highlights the significant economic consequences of unreliable power supply, particularly for energy-intensive industries, and provides empirical evidence on how power outages affect firm revenues and value-added.

Main Findings

• Impact on Productivity: A 10 percent increase in the average duration of outages leads to a 0.14 percent decrease in total revenue and a 0.36 percent decrease in value added, holding other factors constant.
• Sectoral Heterogeneity: The effects of power shortages vary across sectors. Energy-intensive industries such as metal, wood, and paper manufacturing are most severely impacted.
• Reliability Measures: Two key reliability indices are used: the System Average Interruption Frequency Index (SAIFI) and the System Average Interruption Duration Index (SAIDI). The study finds that duration of outages has a more significant impact on firm productivity than frequency.
• Economic Consequences: The paper discusses the broader economic and institutional challenges contributing to power shortages, including underpricing, circular debt, and inefficient infrastructure.

Key Information

Electricity Shortages in Pakistan

• Frequency and Duration: On average, firms in Pakistan experience nearly one outage per day, with durations of around 5.3 hours. In some areas, outages can last up to 18-20 hours per day.
• National Context: Power shortages have become a chronic issue in Pakistan, with the gap between projected demand and actual supply reaching 26 percent of total demand by 2013.
• Systemic Issues: Systematic rolling blackouts, underpricing, and circular debt have led to inefficiencies and high losses in the electricity network. Transmission and distribution losses in 2014 reached 21 percent of all electricity generated.

Data Sources

• Firm-Level Data: From the 2010-2011 Census of Manufacturing Industries, covering 4,499 firms across 23 sectors.
• Power Shortage Data: From the National Electric Power Regulatory Authority (NEPRA) reports, providing district-level outage data for 2010-2015.

Empirical Approach

• Baseline Regression Model:
  $$
  Y_{i} = \alpha + \beta * \ln(\text{shortage}{i}) + \gamma * X{i} + \delta S_{i} + \epsilon_{i}
  $$
  Where $Y_{i}$ is the outcome (revenue or value-added), $\text{shortage}{i}$ is the outage index, $X{i}$ are input costs and firm characteristics, and $S_{i}$ are sector fixed effects.
• Sectoral Analysis: The study allows for heterogeneous impacts by two-digit PSIC sectors, showing that certain sectors (e.g., wood, printing, and metal manufacturing) are more vulnerable to power shortages.

Results

Product Revenues

• A 10 percent increase in outage duration leads to a 0.14 to 0.28 percent decrease in firm revenues.
• The preferred specification (including sector fixed effects) estimates an elasticity of -0.014.
• Frequency of outages has a smaller and less significant impact on revenues.

Value-Added

• A 10 percent increase in outage duration leads to a 0.36 percent decrease in value-added.
• The elasticity of value-added to outage duration is -0.036, more than twice as large as the elasticity for revenue.
• Frequency of outages has a positive but insignificant effect on value-added, suggesting that more frequent outages may lead to shorter individual outages, which firms may prefer.

Sectoral Heterogeneity

• Sector 16 (Wood and products of wood and cork): Has the largest effect on revenue, with an elasticity of -0.24, though it is imprecisely estimated.
• Sector 18 (Printing and reproduction of recorded media): Has the largest effect on value-added, with an elasticity of -0.4, again imprecisely estimated.
• Sectors 15 (Leather and related products) and 17 (Paper and paper products): Show significant and moderate impacts, with elasticities of -0.15 and -0.06, respectively.
• Sectors 24 and 25 (Metal products): Have larger-than-average elasticities in the range of -0.15 to -0.20.

Limitations and Implications

• Causal Identification Challenges: The study acknowledges limitations in causal identification due to potential omitted variable bias, endogeneity, and measurement error.
• Self-Generation Not Captured: The data do not include information on self-generation, which may mitigate some impacts of outages.
• Conservative Estimates: The estimated impacts are likely conservative due to the inability to account for all factors, such as self-generation and location-specific variations.
• Future Work: The paper suggests that future research should use panel data and more precise measures of outages at the firm level to better estimate the causal impact of power shortages on productivity.

Conclusion

The study concludes that electricity shortages have a measurable and negative impact on firm productivity in Pakistan, particularly on energy-intensive industries. While the effects of frequency are less significant, the duration of outages is a more critical factor. The paper emphasizes the need for improved electricity supply reliability to support economic growth and reduce productivity losses.