2015年-IMF国际货币组织全球_The_Sources_of_Business_Cycles_in_a_Low_Income_Country_34页_416kb

Summary of "The Sources of Business Cycles in a Low Income Country"

Core Content

This paper investigates the sources of macroeconomic fluctuations in Ghana and South Africa, focusing on the role of global and domestic shocks. It employs a Bayesian Vector Auto-Regressive (BVAR) model to identify the impact of credit supply shocks, productivity shocks, and commodity price shocks on business cycles. The study is particularly relevant for low-income countries (LICs) and aims to contribute to the literature on macroeconomic stabilization in developing economies.

The authors compare Ghana and South Africa, two countries with similar economic structures but different levels of development, to understand how the domestic vs. global nature of shocks influences macroeconomic outcomes. They argue that while global shocks are more significant in South Africa, they have a relatively smaller impact on Ghana, which suggests a lower degree of financial integration.

Main Points

• Global Shocks are more dominant in South Africa than in Ghana.
• These shocks affect the economy through three channels: trade, credit, and commodity prices.
• Credit supply shocks have a negative but modest effect on global output.
• Productivity shocks have sharp and persistent effects on global output.
• Commodity price shocks are more significant for South Africa than for Ghana.
• The inflation targeting framework in Ghana has increased the role of commodity price shocks in macroeconomic volatility.
• The paper uses sign and zero restrictions within BVAR models to identify structural shocks, following the methodology of Uhlig (2005) and Mountford and Uhlig (2009).
• The G7 countries are used as a proxy for global economic conditions, with G7 factors identified using the first principal component.
• The variance decomposition analysis shows that global shocks account for a larger share of macroeconomic variability in South Africa compared to Ghana.

Key Information

Shocks Studied

• Credit supply shocks: Exogenous changes in the availability of credit.
• Productivity shocks: Exogenous changes in total factor productivity.
• Commodity price shocks: Exogenous changes in the prices of primary goods, particularly gold in the case of Ghana and South Africa.

Methodology

• A BVAR model is used to estimate the effects of these shocks.
• The model includes 16 variables for the credit and productivity shocks, and 6 variables for commodity price shocks.
• Sign restrictions are applied to distinguish between exogenous and endogenous responses.
• Zero restrictions are used to separate global shocks from domestic ones.
• Recursive identification is used for commodity price shocks, with commodity prices ordered first in the VAR.

Empirical Findings

• Global credit shocks have a significant impact on South Africa but not on Ghana, indicating less financial integration in Ghana.
• Global productivity shocks have persistent and strong effects on output in both countries.
• Commodity price shocks have a larger effect on South Africa than on Ghana, but increasingly important in both countries in recent years.
• The inflation targeting regime in Ghana has made commodity price shocks a more significant source of volatility.
• Domestic credit shocks have a larger impact on inflation in Ghana and South Africa than global shocks.
• Domestic productivity shocks lead to output declines and inflation increases, with interest rates responding differently in the two countries.

Policy Implications

• The monetary policy response in Ghana is more focused on inflation than on output stabilization.
• The structural characteristics of the economy, such as the dominant primary goods sector, make commodity price shocks an important consideration for policy modeling.
• The study suggests that structural DSGE models should incorporate credit and productivity shocks, as well as commodity price shocks, to better understand business cycles in low-income countries.

Structure of the Paper

1. Introduction: Sets the context and outlines the objective of the study.
2. Background on Macroeconomic Conditions in Ghana: Describes Ghana’s historical macroeconomic conditions and the implementation of the Structural Adjustment Program (ERP) and inflation targeting.
3. Methodology: Explains the BVAR model, the identification strategy, and the data used.
4. Empirical Results: Presents the impulse response functions (IRFs) and variance decomposition for the three types of shocks.
5. Conclusion: Summarizes the findings and discusses the policy implications for stabilization in low-income countries.

Tables and Figures

• Tables 1 and 2 provide detailed data information and identification restrictions.
• Tables 3 and 4 report the median percentage variance shares attributed to each type of shock.
• Figures 1–12 illustrate the IRFs and variance decomposition results for Ghana and South Africa.

Keywords and JEL Classifications

• Keywords: Credit Shocks, Developing Countries, Macroeconomic Stabilization Policies, Sign Restrictions, Bayesian VAR.
• JEL Classifications: C51, C33, C15, C53, E3, E43, E52, N17.