2000年-世界发展银行全球_Measuring_Growth_in_Total_Factor_Productivity_4页_410kb

报告摘要

Summary of "Measuring Growth in Total Factor Productivity"

Core Content

Total Factor Productivity (TFP) growth is a key indicator of economic performance, reflecting more efficient use of inputs. It is crucial for understanding past and potential economic growth, yet measuring it is complex due to the sensitivity of estimates to assumptions and the difficulty in interpreting the results.

Main Points

1. Importance of TFP Growth

TFP growth is a major driver of income and welfare improvements across countries.
Cross-country differences in income and growth rates are largely attributed to differences in productivity.
Estimating TFP growth is essential for evaluating economic performance.

2. Challenges in Measuring TFP

Sensitivity to assumptions: Small changes in assumptions about the production function can lead to significant variations in TFP estimates.
Interpretation issues: TFP growth may not solely reflect technical change but also include factors like increasing returns to scale, markups due to imperfect competition, and sectoral reallocations.

3. Data and Methodology

TFP Estimation Formula:
$$
g_{A} = g_{Y} - \gamma [\alpha g_{K} + (1 - \alpha)g_{H}]
$$
where $g_{Y}$, $g_{K}$, and $g_{H}$ are growth rates of output, physical capital, and human capital-adjusted labor, respectively.
Human Capital Adjustment: Human-capital-adjusted labor input $H$ is calculated using:
$$
H = L P e^{0.1 S}
$$
where $L$ is the working-age population, $P$ is the participation rate, and $S$ is the average years of schooling.

4. Sensitivity of TFP Estimates

Table 1 shows that TFP growth estimates in Korea (1960-1997) vary significantly based on the assumed values of $\alpha$ and $\gamma$.
- With $\gamma = 1.0$, TFP growth ranges from 1.6% to 0.3% depending on $\alpha$.
- With $\gamma = 1.2$, TFP growth becomes negative, indicating that output growth may be attributed to scale economies rather than productivity improvements.
- With $\gamma = 0.8$, TFP growth is higher, suggesting decreasing returns to scale.
Elasticity of substitution (σ) also affects TFP estimates, as shown in the second table. For example, with $\alpha = 0.3$ and $\sigma = 0.8$, TFP growth is 3.3%, but it drops to -0.7% with $\sigma = 1.2$.

5. Interpretation of TFP Growth

Endogeneity of inputs: TFP growth is not just a result of technical change, but also influences the growth of physical and human capital.
Naive vs. accurate estimation: A simple ratio of TFP growth to output growth can be misleading. A better method involves calculating the covariance between productivity growth and output growth, divided by the variance of output growth.
In Korea, this more accurate method shows that TFP growth accounts for over 80% of output growth, compared to a naive estimate of 18%.

6. Estimating the Production Function

Two main approaches are used:
1. Assuming constant returns to scale and perfect competition: In this case, $\alpha$ can be interpreted as 1 minus the share of wages in value added.
2. Econometric estimation: This involves regressing output growth on input growth, but requires addressing endogeneity and finding appropriate instrumental variables.
The econometric approach can provide insights into scale economies and market imperfections, which are important for policy analysis.

7. Policy Implications

The econometric approach has been used in the Korea Economic Report (World Bank, 1999) to analyze the impact of competition on productivity growth in key manufacturing sectors.
Understanding the sources of TFP growth is crucial for designing effective economic policies.

Key Information

TFP Growth Formula: $g_{A} = g_{Y} - \gamma [\alpha g_{K} + (1 - \alpha)g_{H}]$
Human Capital Adjustment: $H = L P e^{0.1 S}$
TFP Growth in Korea (1960-1997):
- Output growth ($g_Y$): 8.5%
- Physical capital growth ($g_K$): 11.6%
- Human capital growth ($g_H$): 4.9%
Sensitivity to assumptions:
- $\alpha = 0.3$ and $\gamma = 1.0$ yield 1.6% TFP growth.
- $\alpha = 0.5$ and $\gamma = 1.0$ yield 0.3% TFP growth.
- $\gamma = 1.2$ can lead to negative TFP growth.
Endogeneity of inputs: Factor inputs (capital and labor) are not exogenous and can respond to TFP growth, affecting estimates.