2005年-世界发展银行全球_Growth_Trends_in_the_Developing_World___Country_Forecasts_and_Determinants_62页_668kb

Summary of WPS3775: Growth Trends in the Developing World

Core Content

This working paper presents real per capita GDP growth forecasts for all developing countries from 2005 to 2014. It also identifies the main growth determinants and evaluates their impact on growth. The study uses a cross-country regression framework, which is based on the neoclassical growth model, to forecast growth trends and assess the effects of various growth factors.

Main Points

• Growth Forecasts: The paper provides real per capita GDP growth forecasts for all developing countries over the next decade (2005-14). For 55 countries representing 80% of the developing world's GDP, it forecasts the growth effects of key determinants assuming they follow past trends.

• Growth Determinants: The main growth determinants include:

  • Public infrastructure (most significant)
  • Secondary school enrollment
  • Trade openness
  • Financial deepening

• Estimated Growth Contributions:

  • The joint contribution of these four determinants to average, annual per capita GDP growth is estimated to be 1 percentage point.
  • Public infrastructure alone contributes 0.5 percentage points.
  • Failure to improve public infrastructure could reduce the overall growth dividend by 50%.

• Regional Variations:

  • East Asia has had the highest growth rates, with China leading at above 3% per capita growth over 40 years.
  • South Asia has seen significant growth, particularly in India and Sri Lanka.
  • Middle East and North Africa had two countries (Egypt and Tunisia) that grew at above 3% annually.
  • Latin America experienced negative growth in the 1980s but recovered in the 1990s.
  • Sub-Saharan Africa has had mixed results, with Botswana as a notable success and Haiti as a case of prolonged negative growth.

• Volatility Trends:

  • Developed countries have the least output volatility, followed by South Asia.
  • Output volatility in Africa and the Middle East has been declining since the 1970s.
  • East Asia saw an increase in volatility in the 1990s compared to the 1960s.
  • Latin America experienced increased volatility in the 1980s and reduced volatility in the 1990s.

Key Information

• Methodology:

  • The paper uses a cross-country growth regression model to forecast growth.
  • The model incorporates variables such as education, financial depth, trade openness, government burden, public infrastructure, and governance.
  • It also accounts for the initial output gap to improve the accuracy of growth forecasts.

• Forecasting Approach:

  • The model is based on Loayza et al. (2005) and uses a panel dataset.
  • It forecasts growth by using univariate models for the growth determinants in the model.
  • The model does not account for policy interactions or policy effectiveness, as it focuses on the growth effects of indicators rather than the policies themselves.

• Limitations and Caveats:

  • Forecasts are estimates, not predictions, based on assumptions that may not reflect current or future conditions.
  • The model uses proxies for growth determinants, and changes in these proxies should not be interpreted literally.
  • The accuracy of forecasts depends on the specification of the model, the stability of regression coefficients, and the quality of the explanatory variables.

• Data Sources:

  • The data is sourced from the World Bank and other international databases.
  • Education is measured as the ratio of secondary school enrollment to the population in the corresponding age group.
  • Financial depth is measured as the ratio of claims on the private sector by deposit money banks and other financial institutions to GDP.
  • Trade openness is a residual from a regression of trade to GDP ratio on area, population, and structural characteristics.
  • Government burden is measured as the ratio of government consumption to GDP.
  • Public infrastructure is measured using the number of telephone mainlines per capita.
  • Governance is measured using the first principal component of four indicators from the ICRG: law and order, bureaucracy quality, corruption, and public accountability.

Conclusion

The paper emphasizes that while the model provides useful insights into growth possibilities, it should be used in conjunction with country-specific information to form more accurate consensus forecasts. The model is not intended to guide policy design, but rather to assess the growth effects of changes in growth determinants. The results highlight the importance of public infrastructure in driving growth, followed by education, trade openness, and financial deepening.