2016年-世界发展银行全球_Learning_or_Leaning___Persistent_and_Transitory_Growth_Spillovers_from_FDI_28页_900kb

Summary of "Learning or Leaning: Persistent and Transitory Growth Spillovers from FDI"

Core Content

This paper investigates the nature of growth spillovers from foreign direct investment (FDI) to local firms in Jordan, distinguishing between persistent learning externalities and transitory leaning effects. It uses firm-level data from 2006 and 2011 to estimate the impact of FDI inflows and outflows on local firms' employment and capital growth, and finds that the spillovers are largely transitory, meaning they vanish when FDI exits.

Main Views

1. Nature of FDI Spillovers

• The paper identifies two types of spillovers:
  • Learning effects: Persistent spillovers that occur due to knowledge transfer, such as technological or managerial practices, which continue even after the foreign firm leaves.
  • Leaning effects: Temporary spillovers that arise from increased demand for local goods and services due to the presence of MNEs, which disappear when FDI exits.

2. Empirical Findings

• Spillovers are asymmetric: FDI exits have a larger and more significant negative impact on local firms' employment and capital growth than FDI entries.
• Backward linkages (where local firms supply inputs to MNEs) are especially affected by MNE exits, indicating a strong leaning effect.
• Smaller firms are more vulnerable to the negative impacts of MNE exits, particularly in backward linkages, suggesting they benefit more from the temporary demand increases.

3. Policy Implications

• If spillovers are transitory, temporary FDI incentives (e.g., tax holidays) may not be sufficient to sustain long-term growth.
• To ensure sustainable growth, the paper suggests that long-term structural policies (e.g., maintaining low corporate tax rates, strengthening domestic absorptive capacity, and promoting technology sharing) are more effective.
• Retaining FDI is crucial for sustaining the benefits of spillovers, as the productivity gains from MNEs are not long-lasting if the spillovers are leaning in nature.

Key Information

1. Data and Methodology

• The data comes from the Jordanian firm census for 2006 and 2011, covering approximately 100,000 firms in both manufacturing and services.
• The sample includes 15,465 firms in both years, representing 53% of employment in Jordan.
• The paper uses a first-differencing approach to estimate the effect of changes in FDI on local firm growth, distinguishing between net changes and changes in spillover variables due to FDI increases or decreases.

2. Sectoral Analysis

• FDI inflows have increased significantly in downstream industries, which are supplied by local firms.
• The impact of FDI varies by sector, size, and age of the firm. Smaller and younger firms are more affected by FDI changes.
• The global financial crisis (2009-2010) caused a sudden decline in FDI and many MNE exits, creating a quasi-natural experiment to test the effects of FDI changes on local firms.

3. Spillover Variables

• Horizontal spillovers (φ^H): Measure the foreign presence in the same sector.
• Backward spillovers (φ^B): Measure the foreign presence in industries supplied by the local sector.
• Forward spillovers (φ^F): Measure the foreign presence in industries that supply the local sector.
• The paper distinguishes between positive and negative changes in these variables, finding that negative changes (due to exits) have a stronger effect on firm growth.

4. Theoretical Framework

• A simple model is presented to illustrate how asymmetric spillovers can occur.
• The model assumes that spillovers can be transitory (δ = 1) or persistent (δ = 0), with the degree of persistence affecting the long-term impact of FDI on local firms.
• The paper shows that net FDI inflows do not fully capture the asymmetric effects of FDI changes, and that decomposing spillovers into increases and decreases is necessary to understand the true nature of the effects.

Conclusion

The paper concludes that FDI spillovers are primarily transitory, stemming from leaning effects rather than learning effects. This has important implications for FDI policy design, suggesting that long-term structural policies are more effective in promoting sustainable growth than temporary incentives. Understanding the asymmetric nature of spillovers helps policymakers make more informed decisions about how to attract and retain FDI to maximize the benefits to the local economy.