2006年-世界发展银行全球_Forecasting_Investment_Needs_in_South_Africas_Electricity_and_Telecommunications_Sectors_24页_410kb

Summary of "Forecasting Investment Needs in South Africa’s Electricity and Telecommunications Sectors"

Core Content

This paper estimates the demand for electricity and telecommunications services in South Africa using a panel-data approach for the period 1980–2002, and projects investment needs through 2010 under two growth scenarios. The study aims to provide a framework for understanding the relationship between infrastructure investment and economic growth, while also informing policy decisions on scaling up infrastructure in South Africa.

Main Objectives

1. To estimate the demand for electricity and telecommunications using an innovative econometric framework.
2. To project investment needs for South Africa under two growth scenarios: the current growth rate of 3.6% per annum and an accelerated growth rate of 6% per annum.
3. To assess the implications of infrastructure investment for growth and equity, particularly in the context of Sub-Saharan Africa.

Key Findings

• Infrastructure and Growth: Infrastructure investment is a key driver of economic growth and welfare. In South Africa, electricity has the strongest and most consistent impact on growth.
• Infrastructure Deficit: Sub-Saharan Africa faces significant infrastructure deficits, which have led to lower growth and poor social outcomes. The region needs about 5% of GDP annually in infrastructure investment to meet growth targets, with additional 4% for operations and maintenance.
• South Africa’s Decline in Investment: South Africa has experienced a long-term decline in infrastructure investment, with per capita investment falling by 72% from 1976 to 2002. Investment as a share of GDP dropped from 8.1% to 2.4%, below international benchmarks.
• Policy Context: The South African government has recognized the need to scale up infrastructure investment, with projections indicating an increase from 5.2% of GDP in 2004/05 to 6.7% in 2008/9, in nominal terms from R72 billion to R135 billion.
• Empirical Framework: The paper uses a three-stage econometric approach, including panel data analysis, to estimate the long-run demand for infrastructure. The Pooled Mean Group (PMG) estimator is employed to account for both long-run homogeneity and short-run heterogeneity across countries.
• Demand Elasticities: The income elasticity for electricity is approximately 1.4, which is consistent with international studies. For telecommunications, the elasticity is around 1.2, also in line with global evidence.

Growth Scenarios and Investment Projections

• Current Growth Scenario (3.6% per annum):

  • Average annual investment needs in electricity: ~0.2% of GDP.
  • Average annual investment needs in telecommunications: ~0.75% of GDP.

• Accelerated Growth Scenario (6% per annum):

  • Investment needs are expected to approximately double.
  • This implies significantly higher annual investment requirements in both sectors.

Methodology and Data

• Data Source: Panel data from 52 low- and middle-income countries for the period 1980–2002, sourced from the World Bank’s World Development Indicators.
• Variables:
  • Dependent Variables: Electricity production per capita (kwh), total telephone lines per 1,000 population.
  • Independent Variables: GDP per capita (PPP), sectoral shares (agriculture, manufacturing, services), and a dummy variable for the structural break in 1996 due to the introduction of mobile phone technology.
• Estimation Technique: The paper uses the PMG estimator, which allows for heterogeneity in short-run dynamics while assuming homogeneity in long-run relationships. The method is robust and tested for validity using Hausman tests.

Sectoral Impact

• Electricity: A higher share of manufacturing in GDP is positively correlated with electricity demand, while a higher share of agriculture is negatively correlated.
• Telecommunications: The share of services in GDP has a positive impact on telephony demand, and the structural break in 1996 (mobile phone technology) significantly accelerated growth in per capita telephone usage.

Conclusion

The study highlights the importance of infrastructure in driving economic growth and reducing inequality in South Africa. It provides a detailed empirical basis for forecasting investment needs and supports the ongoing policy efforts to increase infrastructure spending as part of the country’s accelerated growth strategy. The results suggest that without significant investment, South Africa may not be able to catch up with infrastructure performance of upper-middle-income countries.

Key Figures and Tables

• Table 1: Summary statistics of the panel data, showing means, standard deviations, and other relevant metrics.
• Table 2: Estimation results of long-run demand equations for electricity and telephony, including income elasticity, sectoral shares, and structural break effects.
• Table 3: Unit costs for electricity and telecommunications, with lower and upper bounds for the long-run marginal cost of electricity generation.

Authors and Acknowledgments

• Authors: Željko Bogetić (World Bank, Africa Region) and Johannes W. Fedderke (University of Cape Town, School of Economics).
• Research Assistance: Andrew Hill.
• Earlier Work: The paper builds on and extends the earlier studies by Bogetić and Fedderke (2005a, b), and incorporates findings from Fay (2001) and Fay and Yepes (2003).

References

• Aschauer, D. A. (1989)
• Leipziger, D., et al. (2003)
• Estache, A. (2005)
• World Bank (1994, 2005a, b)
• Easterly, Calderón, and Serven (Figure 1)
• Fay (2001, 2003)
• Pesaran, Shin, and Smith (1999)
• National Treasury (2005b)

Implications for Policy

The paper supports the need for increased infrastructure investment to meet the goals of accelerated and shared growth in South Africa. It also emphasizes the importance of accurate demand estimation and cost analysis in planning for future infrastructure needs.