## _wp06294

## Source details

**Canonical URL:** [_wp06294](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp06294.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp06294.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp06294.pdf.json)

---

### Summary of purpose and methods
- Objective: Analyze role of factor-biased technological progress, openness to trade, and changes in employment protection in explaining movements in labor’s share over 1960–2000.
- Sample and periodization:
  - Panel of 18 industrial countries over the 1960-2000 period.
  - Data averaged over successive five-year periods to eliminate cyclical effects.
  - Sample split in 1985 into “preglobalization/pre-IT revolution” and “postglobalization/post-IT revolution.”
- Estimation approach:
  - Basic specification: panel regression with country-fixed effects (level specification): Yit = β0i + β1Xit + uit.
  - Robustness check: first-difference (delta) regressions: ΔYit = β0 + γt + β1ΔXit + uit (fixed effects dropped; time effects included when significant).
- Dependent variables:
  - Compensation share in national income (CS): wages and salaries, employer-financed benefits, unemployment insurance, social security, workmen’s compensation.
  - Employment share in national income (ES): includes self-employment income with two-thirds allocated to labor earnings and one-third to capital income.
  - Gini coefficient (Gini): standard measure of income inequality (data from Dollar-Kraay dataset and WIID).
- Key explanatory variables and proxies:
  - Productivity and technology: Labor productivity (LProd = GDP per hour worked); Productivity per worker (ProdW = GDP / total employment); lagged versions LagLProd, LagProdW.
  - Openness to trade: Open = (Exports+Imports)/GDP; TradeDev = trade share with developing countries; FDI = (FDIinward+FDIoutward)/GDP; KFlow = capital flows / GDP.
  - Bargaining power of labor: Union Density (UN); Employment Protection (EP) ranging 0 to 2.

### Key empirical findings (overall)
- Long-run trend:
  - Labor’s share rose through the 1960s and 1970s and experienced a decline in recent decades in industrial countries.
  - Decline likely driven by structural changes (capital-augmenting technological progress and globalization), not only cyclical factors.
- Technology / productivity:
  - Preglobalization (pre-1985): technological progress appears labor-augmenting.
    - A 1 percentage point increase in lagged productivity per worker increased compensation share by 0.34 percentage point (text statement) and employment share by 0.29 percentage point (text statement).
    - Table 1 (Compensation Share, Level, Pre-IT/Globalization Era): Lagged productivity per worker coefficient = 0.35 *** (0.02).
    - Table 1 (Employment Share, Level, Pre-IT/Globalization Era): Lagged Productivity Per Worker coefficient = 0.29 *** (0.01).
  - Postglobalization (after 1985/IT revolution): technological progress appears capital-augmenting.
    - A 1 percentage point increase in lagged productivity per worker resulted in a decline of compensation and employment shares.
    - Table 1 (Compensation Share, Level, IT/Globalization Era): Lagged productivity per worker coefficient = -0.24 *** (0.06).
    - Table 1 (Employment Share, Level, IT/Globalization Era): Lagged Productivity Per Worker coefficient = -0.11 ** (0.04).
- Trade and globalization:
  - Openness to trade and increasing trade with developing countries had a negative effect on labor’s share in industrial countries (consistent with Heckscher-Ohlin).
  - Table 1 (Compensation Share, Level):
    - (Exports+Imports)/GDP coefficient: Pre-IT: -0.08 (0.08); IT Era: -0.14 *** (0.02).
  - Table 1 (Employment Share, Level):
    - (Exports+Imports)/GDP coefficient: Pre-IT: -0.21 *** (0.07); IT Era: -0.17 *** (0.04).
- Bargaining power / employment protection:
  - Employment protection continues to move income toward labor, but globalization appears to have dampened this effect.
  - Table 1 (Compensation Share, Level): Employment protection coefficient: Pre-IT: 0.01 (0.01); IT Era: 0.018 (0.017).
  - Table 1 (Employment Share, Level): Employment Protection coefficient: Pre-IT: 0.05 *** (0.00); IT Era: 0.10 *** (0.03).
- Inequality (Gini) results:
  - If technological progress is capital-augmenting, inequality tends to worsen; IT revolution likely favored highly skilled workers.
  - Table 1 (Gini Coefficient, Level, IT/Globalization Era): Lagged Productivity Per Worker coefficient = 0.15 ** (0.06); Employment Protection coefficient = -0.01 (0.01).
  - Table 1 (Gini Coefficient, Level, Pre-IT/Globalization Era): Employment Protection coefficient = -0.09 ** (0.03).

### Quantitative highlights (selected coefficients and model fits)
- Compensation Share (Level):
  - Pre-IT/Globalization Era:
    - (Exports+Imports)/GDP = -0.08 (0.08)
    - Lagged productivity per worker = 0.35 *** (0.02)
    - Employment protection = 0.01 (0.01)
    - N = 48, R2 = 0.95
  - IT/Globalization Era:
    - (Exports+Imports)/GDP = -0.14 *** (0.02)
    - Lagged productivity per worker = -0.24 *** (0.06)
    - Employment protection = 0.018 (0.017)
    - N = 72, R2 = 0.94
- Employment Share (Level):
  - Pre-IT/Globalization Era:
    - (Exports+Imports)/GDP = -0.21 *** (0.07)
    - Lagged Productivity Per Worker = 0.29 *** (0.01)
    - Employment Protection = 0.05 *** (0.00)
    - N = 31, R2 = 0.98
  - IT/Globalization Era:
    - (Exports+Imports)/GDP = -0.17 *** (0.04)
    - Lagged Productivity Per Worker = -0.11 ** (0.04)
    - Employment Protection = 0.10 *** (0.03)
    - N = 68, R2 = 0.81
- Gini Coefficient (Level):
  - Pre-IT/Globalization Era:
    - Employment Protection = -0.09 ** (0.03)
    - N = 45, R2 = 0.82
  - IT/Globalization Era:
    - Lagged Productivity Per Worker = 0.15 ** (0.06)
    - N = 70, R2 = 0.75

### Detailed findings by theme

- A. Technology / Productivity (detailed ranges and significance)
  - Preglobalization (prior to 1985):
    - For a percentage point increase in labor productivity (LProd), compensation share increases by 0.22–0.44 percentage point.
    - For a percentage point increase in LProd, employment share increases by 0.11–0.48 percentage point.
    - For a percentage point increase in ProdW, compensation share increases by 0.25–0.58 percentage point.
    - For a percentage point increase in ProdW, employment share increases by 0.13–0.61 percentage point.
    - Effects always significant at the 99 percent confidence level.
  - Postglobalization (after 1985):
    - For a percentage point increase in LProd, compensation share falls by 0.13–0.19 percentage point.
    - For a percentage point increase in LProd, employment share falls by 0.05–0.21 percentage point.
    - For a percentage point increase in ProdW, compensation share decreases by 0.25–0.36 percentage point.
    - Effects significant at the 99 percent confidence level and robust.
  - Lagged productivity variables produce very similar regression results.

- B. Trade openness (detailed magnitudes)
  - General openness (Open = (Exports+Imports)/GDP):
    - For every percentage point increase in Open, compensation share falls by 0.13–0.15 percentage point.
    - For every percentage point increase in Open, employment share falls by about 0.16 percentage point.
    - Effects significant at the 99 percent confidence level for all combinations and robust.
  - Trade with developing countries (TradeDev):
    - Preglobalization: a percentage point increase in TradeDev raised compensation by 0.15–0.18 and employment by 0.23–0.26 percentage point (99 percent significance).
    - Postglobalization: generally a negative effect on compensation and employment; during globalization the Gini rises with TradeDev (not always robust/significant).
  - FDI/GDP:
    - Preglobalization: each percentage point increase in FDI/GDP decreases employment share by 0.46–0.50 percentage point (99 percent significance).
    - Postglobalization: each percentage point increase in FDI/GDP decreases employment share by 0.10–0.15 percentage point (99 percent significance).
    - Multivariate postglobalization regressions suggest Gini rises by about 0.07 percentage point for every percentage point increase in FDI/GDP.
  - KFlow regressions produce results similar to Open and FDI.

- C. Bargaining power of labor (detailed magnitudes)
  - Preglobalization:
    - A 1 percentage point increase in EP increases compensation share by 0.02–0.04 percentage point and employment share by 0.03–0.06 percentage point (robust).
    - Union density (UN) effect on compensation and employment is insignificant.
  - IT/globalization era:
    - Employment share increases with EP by about 0.08–0.10 percentage point (effect not always significant).
    - EP effect generally positive and significant but less robust than preglobalization.
    - Inequality decreases with increased bargaining power of labor, but results are not always robust or significant.

- D. Inequality and Gini (summary from appendices)
  - Pre-IT era: higher productivity and stronger labor protection generally associated with lower Gini (examples: Lagged Labor Productivity: -0.124*** (0.035) in Table A6.1; Employment Protection: -0.108*** (0.023)).
  - IT era: lagged productivity often associated with higher Gini in level regressions (example: Lagged Labor Productivity: 0.093** (0.037) in Table A6.1); FDI/GDP positive and significant in several IT-era specifications (example: Table A6.3 IT Level: FDI/GDP Ratio = 0.073*** (0.016)).
  - Appendix V summary coefficients preserve mixed and specification-dependent results; several productivity measures switch sign between eras.

### Robustness and econometric notes
- Bivariate regressions align with expected signs, but multivariate regressions required choosing one proxy per concept due to high collinearity.
- Preferred multivariate specification uses trade openness, lagged productivity per worker, and employment protection, with unemployment rate included when accounting for job creation effects.
- Two-stage least-squares using lagged productivity as an instrument produced results similar to noninstrumented regressions; Hausman tests indicate endogeneity problems are not particularly serious for productivity variables.
- Robustness checks included:
  - Using alternative proxies for openness, productivity, and labor protection.
  - Dropping countries or years from the sample.
  - Including the inflation rate in regressions.
- Productivity-related results are described as especially strong and robust; detailed robustness checks reported in Appendix D and Appendix IV.

### Subsample: non-continental-European countries
- Six-country subsample (Australia, Canada, Japan, the United States, the United Kingdom, Norway):
  - Results broadly similar but not as strong or significant.
  - Openness decreases labor share in the subsample.
  - A percentage point increase in productivity per worker raises labor share by 0.35 percentage point prior to the IT revolution (significant at the 95 percent confidence level).
  - Labor’s share declines with productivity increases during the IT/globalization era, but results not always significant.

### Interpretation and conclusions
- The decline in labor’s share in OECD countries over recent decades appears largely an equilibrium response to:
  - Capital-augmenting technological progress (post-1985 IT revolution),
  - Greater openness to trade and increased trade with developing countries,
  - Increased capital mobility reducing labor’s bargaining power.
- Employment-protection policies and unionization still favor labor, but globalization has likely weakened their effectiveness in sustaining labor’s share.
- Empirical evidence consistent with Heckscher-Ohlin predictions: increasing openness leads industrial (capital-rich) countries to specialize in capital-intensive production, reducing returns to relatively scarce labor and lowering labor’s share.
- Suggested further work: explore cross-country variations that remain unexplained and extend analysis to developing countries (data limitations noted).

### Appendix highlights

- Appendix III — Correlation matrix (selected entries preserved exactly):
  - Variables order: Open, TradeDev, FDI, LProd, ProdW, LagLProd, LagProdW, EP, UN.
  - Open row: 1.000, -0.668, 0.312, 0.026, -0.041, 0.003, -0.040, 0.064, 0.344
  - TradeDev row: -0.668, 1.000, -0.268, -0.014, 0.199, -0.031, 0.157, 0.082, -0.374
  - FDI row: 0.312, -0.268, 1.000, 0.406, 0.225, 0.405, 0.266, -0.344, -0.042
  - LProd row: 0.026, -0.014, 0.406, 1.000, 0.715, 0.979, 0.759, -0.003, -0.269
  - ProdW row: -0.041, 0.199, 0.225, 0.715, 1.000, 0.710, 0.984, -0.082, -0.173
  - EP row: 0.064, 0.082, -0.344, -0.003, -0.082, -0.065, -0.119, 1.000, 0.026
  - UN row: 0.344, -0.374, -0.042, -0.269, -0.173, -0.262, -0.171, 0.026, 1.000

- Appendix IV — Selected multivariate specification results (preserving exact coefficients and significance; examples):
  - Table A4.1 — Compensation Share on (Exports+Imports)/GDP, Lagged Labor Productivity, Employment Protection:
    - Post-IT/Globalization Era (Level): (Exports+Imports)/GDP: -0.147*** (0.016); Lagged Labor Productivity: -0.143*** (0.036); Employment Protection: 0.017 (0.016); N: 72; R2: 0.938
  - Table A4.5 — Compensation Share: (Exports+Imports)/GDP, Productivity per Worker, Employment Protection:
    - IT/Globalization Era (Level): (Exports+Imports)/GDP: -0.132*** (0.018); Productivity per Worker: -0.252*** (0.043); Employment Protection: 0.016 (0.016); N: 72; R2: 0.938
  - Table A4.9 — Compensation Share: FDI/GDP Ratio, Labor Productivity, Employment Protection:
    - Pre-IT/Globalization Era (Level): FDI to GDP Ratio: 0.141*** (0.013); Labor Productivity: 0.398*** (0.027); Employment Protection: 0.036*** (0.002); N: 38; R2: 0.971

- Appendix V & VI — Inequality (selected exact coefficients):
  - Appendix V (Table A5.1): Lagged Labor Productivity — IT Level: 0.035*** (0.011); Employment Protection — Pre-IT Level: -0.098*** (0.006)
  - Appendix VI (Table A6.1): IT/Globalization Era (Level) — Lagged Labor Productivity: 0.093** (0.037); Employment Protection: -0.001 (0.014); N: 70; R2: 0.751
  - Appendix VI (Table A6.3): Pre-IT Level — FDI/GDP Ratio: 0.246*** (0.0293); Lagged Labor Productivity: -0.209*** (0.007); Employment Protection: -0.193*** (0.000); N: 34; R2: 0.859

- Appendix VII — Production function and implications (exact expressions preserved)
  - Production function: Y = F(K,L) = [ α(AK)^( (σ-1)/σ ) + (1- α) L^( (σ-1)/σ ) ]^( σ/(σ-1) )
  - Capital’s share: KShare = K * MP_K / Y = α A^( (σ-1)/σ ) (K/Y)^( (σ-1)/σ )
  - Labor’s share: LShare = L * MP_L / Y = 1 - KShare = 1 - α A^( (σ-1)/σ ) (K/Y)^( (σ-1)/σ )
  - Derivatives:
    - ∂LShare/∂(K/Y) = - α ( (σ-1)/σ ) A^( (σ-1)/σ ) (K/Y)^( -1/σ )
    - ∂LShare/∂A = - α ( (σ-1)/σ ) A^( -1/σ ) (K/Y)^( (σ-1)/σ )
  - Key implications:
    - Cobb-Douglas (Y = A K^α L^(1-α)) cannot explain changing factor shares; CES with capital-augmenting A allows A and K/Y to affect factor shares.
    - If σ < 1 (complements), as K/Y increases, capital share decreases and labor share increases.
    - If σ > 1 (substitutes), as K/Y increases, labor’s share decreases.
    - Positive technology shock A (holding K/Y constant) increases capital share and decreases labor share if σ > 1.

*Source: IMF staff estimates and analyses from “Determinants of Labor’s Share in National Income,” based on OECD; IMF; United States Department of Commerce, Bureau of Economic Analysis; and related datasets as reported in the chapter.*

### References..............................................................................................................

### _wp06294 - References..............................................................................................................

### References
- References................................................................................................................................32

### Figures
- 1.            Cross-Country      Average      Labor Share in National Income .............................................4
- 2.            
Gini Coefficient in Mostly English-Speaking and Other Countries.......................................8
- 3.    Labor Share in Mostly English-Speaking and Other Countries.....................................9
- A1.1    Compensation Share: Actual and Fitted Values ..........................................................18

### Tables
- Tables

*Source: _wp06294 - References..............................................................................................................*

### 1. Determinants of Labor’s Share in National Income ....................................................12

### 1. Determinants of Labor’s Share in National Income

### Summary of purpose and methods
- Objective: Analyze role of factor-biased technological progress, openness to trade, and changes in employment protection in explaining movements in labor’s share over 1960–2000.
- Sample and periodization:
  - Panel of 18 industrial countries over the 1960-2000 period.
  - Data averaged over successive five-year periods to eliminate cyclical effects.
  - Sample split in 1985 into “preglobalization/pre-IT revolution” and “postglobalization/post-IT revolution.”
- Estimation approach:
  - Basic specification: panel regression with country-fixed effects (level specification): Yit = β0i + β1Xit + uit.
  - Robustness check: first-difference (delta) regressions: ΔYit = β0 + γt + β1ΔXit + uit (fixed effects dropped; time effects included when significant).
- Dependent variables:
  - Compensation share in national income (CS): wages and salaries, employer-financed benefits, unemployment insurance, social security, workmen’s compensation.
  - Employment share in national income (ES): includes self-employment income with two-thirds allocated to labor earnings and one-third to capital income.
  - Gini coefficient (Gini): standard measure of income inequality (data from Dollar-Kraay dataset and WIID).
- Key explanatory variables and proxies:
  - Productivity and technology:
    - Labor productivity of total economy (LProd) = GDP per hour worked.
    - Productivity per worker (ProdW) = GDP / total employment.
    - Lags used: LagLProd, LagProdW.
  - Openness to trade:
    - (Open) Ratio of trade to GDP = (Exports + Imports)/GDP.
    - (TradeDev) Trade share with developing countries = (Exports to developing countries + Imports from developing countries)/Total trade.
    - (FDI) FDI to GDP ratio = (FDIinward + FDIoutward)/GDP.
    - (KFlow) Capital flows / GDP.
  - Bargaining power of labor:
    - Union Density (UN) = percentage of unionized workforce.
    - Employment Protection (EP) = variable ranging from 0 to 2, increasing with strictness.

### Key empirical findings
- Long-run trend:
  - Labor’s share rose through the 1960s and 1970s and experienced a decline in recent decades in industrial countries.
  - The decline is likely driven by structural changes (capital-augmenting technological progress and globalization), not only cyclical factors.
- Nature of technological change:
  - Preglobalization (pre-1985): technological progress appears labor-augmenting.
    - A 1 percentage point increase in lagged productivity per worker increased:
      - Compensation share by 0.34 percentage point (text statement).
      - Employment share by 0.29 percentage point (text statement).
    - Table 1 (Compensation Share, Level, Pre-IT/Globalization Era):
      - Lagged productivity per worker coefficient = 0.35 *** (standard error (0.02)).
    - Table 1 (Employment Share, Level, Pre-IT/Globalization Era):
      - Lagged Productivity Per Worker coefficient = 0.29 *** (standard error (0.01)).
  - Postglobalization (after 1985/IT revolution): technological progress appears capital-augmenting.
    - A 1 percentage point increase in lagged productivity per worker resulted in a decline of compensation and employment shares (text summarizes).
    - Table 1 (Compensation Share, Level, IT/Globalization Era):
      - Lagged productivity per worker coefficient = -0.24 *** (standard error (0.06)).
    - Table 1 (Employment Share, Level, IT/Globalization Era):
      - Lagged Productivity Per Worker coefficient = -0.11 ** (standard error (0.04)).
- Trade openness and globalization effects:
  - Openness to trade and increasing trade with developing countries had a negative effect on labor’s share in industrial countries (consistent with Heckscher-Ohlin).
  - Table 1 (Compensation Share, Level):
    - (Exports+Imports)/GDP coefficient:
      - Pre-IT/Globalization Era: -0.08 (standard error (0.08)).
      - IT/Globalization Era: -0.14 *** (standard error (0.02)).
  - Table 1 (Employment Share, Level):
    - (Exports+Imports)/GDP coefficient:
      - Pre-IT/Globalization Era: -0.21 *** (standard error (0.07)).
      - IT/Globalization Era: -0.17 *** (standard error (0.04)).
- Bargaining power / employment protection:
  - Employment protection continues to move income toward labor, but globalization appears to have dampened this effect.
  - Table 1 (Compensation Share, Level):
    - Employment protection coefficient:
      - Pre-IT/Globalization Era: 0.01 (standard error (0.01)).
      - IT/Globalization Era: 0.018 (standard error (0.017)).
  - Table 1 (Employment Share, Level):
    - Employment Protection coefficient:
      - Pre-IT/Globalization Era: 0.05 *** (standard error (0.00)).
      - IT/Globalization Era: 0.10 *** (standard error (0.03)).
- Inequality (Gini) results:
  - If technological progress is capital-augmenting, inequality tends to worsen; IT revolution likely favored highly skilled workers.
  - Table 1 (Gini Coefficient, Level, IT/Globalization Era):
    - Lagged Productivity Per Worker coefficient = 0.15 ** (standard error (0.06)).
    - Employment Protection coefficient = -0.01 (standard error (0.01)).
  - Table 1 (Gini Coefficient, Level, Pre-IT/Globalization Era):
    - Employment Protection coefficient = -0.09 ** (standard error (0.03)).
- Robustness and econometric notes:
  - Bivariate regressions align with expected signs, but multivariate regressions required choosing one proxy per concept due to high collinearity.
  - Preferred multivariate specification uses trade openness, lagged productivity per worker, and employment protection, with unemployment rate included to account for job creation effects.
  - Two-stage least-squares using lagged productivity as an instrument produced results similar to noninstrumented regressions; Hausman tests indicate endogeneity problems are not particularly serious for productivity variables.

### Quantitative highlights from Table 1 (selected coefficients and significance)
- Compensation Share (Level):
  - Pre-IT/Globalization Era:
    - (Exports+Imports)/GDP = -0.08 (0.08)
    - Lagged productivity per worker = 0.35 *** (0.02)
    - Employment protection = 0.01 (0.01)
    - N = 48, R2 = 0.95
  - IT/Globalization Era:
    - (Exports+Imports)/GDP = -0.14 *** (0.02)
    - Lagged productivity per worker = -0.24 *** (0.06)
    - Employment protection = 0.018 (0.017)
    - N = 72, R2 = 0.94
- Employment Share (Level):
  - Pre-IT/Globalization Era:
    - (Exports+Imports)/GDP = -0.21 *** (0.07)
    - Lagged Productivity Per Worker = 0.29 *** (0.01)
    - Employment Protection = 0.05 *** (0.00)
    - N = 31, R2 = 0.98
  - IT/Globalization Era:
    - (Exports+Imports)/GDP = -0.17 *** (0.04)
    - Lagged Productivity Per Worker = -0.11 ** (0.04)
    - Employment Protection = 0.10 *** (0.03)
    - N = 68, R2 = 0.81
- Gini Coefficient (Level):
  - Pre-IT/Globalization Era:
    - Employment Protection = -0.09 ** (0.03)
    - N = 45, R2 = 0.82
  - IT/Globalization Era:
    - Lagged Productivity Per Worker = 0.15 ** (0.06)
    - N = 70, R2 = 0.75

### Interpretation and conclusions
- The decline in labor’s share in OECD countries over recent decades appears largely an equilibrium response to:
  - Capital-augmenting technological progress (post-1985 IT revolution),
  - Greater openness to trade and increased trade with developing countries,
  - Increased capital mobility reducing labor’s bargaining power.
- Employment-protection policies and unionization still favor labor, but globalization has likely weakened their effectiveness in sustaining labor’s share.
- The empirical evidence is consistent with Heckscher-Ohlin predictions: increasing openness leads industrial (capital-rich) countries to specialize in capital-intensive production, reducing returns to relatively scarce labor and lowering labor’s share.

*Source: IMF staff estimates and analyses from “Determinants of Labor’s Share in National Income,” based on OECD; IMF; United States Department of Commerce, Bureau of Economic Analysis; and related datasets as reported in the chapter.*

### 0.11 percentage point, respectively. These coefficients are statistically significant. In the later

### _wp06294 - 0.11 percentage point, respectively. These coefficients are statistically significant. In the later

### A. Key empirical findings (overview)
- Regression evidence indicates faster productivity growth has been associated with widening income inequality in the later period.
- Both compensation and employment share decrease with trade openness; the effect on the compensation share is not as significant during the preglobalization era.
- For a percentage point increase in the trade-to-GDP ratio, the compensation and employment share fall by 0.14 and 0.17 percentage point, respectively.
- No impact from trade on inequality was found in some specifications; this is consistent with earlier studies using a range of techniques.
- A higher degree of employment protection benefits labor more than capital, raising compensation and labor share; globalization may dampen this effect by decreasing the bargaining power of labor.

### B. Robustness checks
- Results are robust and broadly similar across specifications.
- Robustness checks included:
  - Using alternative proxies for openness, productivity, and labor protection.
  - Dropping countries or years from the sample.
  - Including the inflation rate in regressions.
- Productivity-related results are described as especially strong and robust; detailed robustness checks are reported in Appendix D and Appendix IV for alternative specifications.

### C. Technology / Productivity (detailed results)
- Preglobalization/pre-IT era (prior to 1985):
  - For a percentage point increase in labor productivity of the total economy (LProd), the compensation share increases by 0.22–0.44 percentage point.
  - For a percentage point increase in LProd, the employment share increases by 0.11–0.48 percentage point.
  - For a percentage point increase in productivity per worker (ProdW), the compensation share increases by 0.25–0.58 percentage point.
  - For a percentage point increase in ProdW, the employment share increases by 0.13–0.61 percentage point.
  - These effects are always significant at the 99 percent confidence level.
  - Interpretation: productivity increases prior to 1985 have been labor-augmenting; inequality should decrease with increases in productivity, but results on Gini are not always robust or significant.
- Postglobalization/post-IT era (after 1985):
  - For a percentage point increase in LProd, the compensation share falls by 0.13–0.19 percentage point.
  - For a percentage point increase in LProd, the employment share falls by 0.05–0.21 percentage point.
  - For a percentage point increase in ProdW, the compensation share decreases by 0.25–0.36 percentage point.
  - Effects are significant at the 99 percent confidence level and robust.
  - Interpretation: productivity increases after 1985 have been capital-augmenting; inequality should increase with increases in productivity, though results are not always significant or robust.
- Lagged productivity variables produce very similar regression results.

### D. Trade openness (detailed results)
- General openness (Open = (Exports+Imports)/GDP):
  - For every percentage point increase in openness (Open), the compensation share falls by 0.13–0.15 percentage point.
  - For every percentage point increase in openness, the employment share falls by about 0.16 percentage point.
  - Effects are significant at the 99 percent confidence level for all combinations of explanatory variables and are robust.
  - Employment is more responsive to openness than compensation, suggesting self-employment income may also drop with increased openness.
- Trade with developing countries (TradeDev):
  - Preglobalization/pre-IT era: for every percentage point increase in trade share with developing countries, the compensation share rises by 0.15–0.18 percentage point; the employment share rises by 0.23–0.26 percentage point. Results significant at the 99 percent level and robust.
  - Postglobalization era: generally a negative effect on compensation and employment from increasing trade share with developing countries; during the globalization era the Gini coefficient rises with the trade share with developing countries (not always robust/significant).
  - Possible explanation: Heckscher-Ohlin implications may fail when trade barriers, transportation costs, and technology differences are large (more relevant pre-1985).
- FDI/GDP ratio (FDI) as proxy for openness and capital mobility:
  - Preglobalization/pre-IT era: every percentage point increase in FDI/GDP decreases employment share by 0.46–0.50 percentage point.
  - Postglobalization/post-IT era: every percentage point increase in FDI/GDP decreases employment share by 0.10–0.15 percentage point.
  - Effects significant at the 99 percent confidence level.
  - Multivariate regressions for the postglobalization era suggest the Gini coefficient rises by about 0.07 percentage point for every percentage point increase in the FDI/GDP ratio.
- Capital flow ratio (Kflow) regressions produce results similar to Open and FDI regressions.

### E. Bargaining power of labor (detailed results)
- Preglobalization era:
  - For a 1 percentage point increase in employment protection (EP), the compensation share increases by 0.02–0.04 percentage point.
  - For a 1 percentage point increase in EP, the employment share increases by 0.03–0.06 percentage point.
  - Effects are robust; union density (UN) effect on compensation and employment share is insignificant.
  - Inequality decreases with increased bargaining power of labor, but results are not always robust or significant.
- IT/globalization era:
  - Employment share increases with EP by about 0.08–0.10 percentage point (effect not always significant).
  - Prior to globalization the effect of EP was almost always significant; in the globalization era the effect is positive and significant but less robust.
  - Inequality decreases with increased bargaining power of labor in expected direction, but results are not always robust or significant.

### F. Are the results strictly European?
- Regressions run on a subsample of six non-continental-European countries (Australia, Canada, Japan, the United States, the United Kingdom, and Norway) show:
  - Results are broadly similar but not as strong or significant.
  - Openness to trade decreases labor share in the subsample.
  - A percentage point increase in productivity per worker raises labor share by 0.35 percentage point prior to the IT revolution (significant at the 95 percent confidence level).
  - Labor’s share declines with productivity increases during the IT/globalization era, but results not always significant.
  - Prior to globalization, trade was associated with lower inequality (negative coefficient) in the subsample; effects of FDI on inequality prior to globalization are mostly positive but often insignificant and not robust.
  - During the globalization era, increases in FDI lead to increases in inequality, while increases in labor productivity and labor protection have negative effects on inequality.

### G. Conclusions (summary)
- Analysis covers impact of technology, openness, and employment protection on labor share in national income for 18 OECD member countries during 1960–2000.
- Main conclusions:
  - Technological progress has been capital-augmenting during the globalization era (post-1985); prior to the mid-1980s productivity growth increased labor’s share.
  - Openness to trade and increasing trade with developing countries had a negative effect on the labor share in industrial countries, consistent with Heckscher-Ohlin model predictions.
  - Labor-protection policies still shift income toward labor, but globalization has made this effect less significant.
  - Pre-globalization changes in labor’s share were mostly driven by productivity increases and to a lesser extent by trade; post-globalization changes have been equally driven by productivity and openness.
- Inequality findings:
  - IT revolution likely favored highly skilled workers at expense of less skilled workers; technological progress appears skill-biased.
  - Increasing openness is associated with higher income inequality as industrial countries specialize in skill-intensive goods.
  - Employment-protection policies lower inequality.
- Interpretation:
  - The decline in labor’s share in OECD countries over past decades may be an equilibrium adjustment to capital-augmenting technological progress and increased globalization, rather than purely cyclical.
  - Workers’ wealth effects may be muted by increasing direct and indirect ownership of equities by households.
- Suggested further work:
  - Explore cross-country variations that remain to be explained.
  - Extend analysis to developing countries; data limitations may complicate long-term trend analysis.

*Source: IMF staff calculations and analysis from the supplied content.*

### Appendix III. Correlation Matrix of Explanatory Variables

### Appendix III. Correlation Matrix of Explanatory Variables

### Correlation matrix (Table A3.1)
- Variables: Open, TradeDev, FDI, LProd, ProdW, LagLProd, LagProdW, EP, UN
- Correlations (row = variable on left; columns in order Open, TradeDev, FDI, LProd, ProdW, LagLProd, LagProdW, EP, UN):
  - Open: 1.000, -0.668, 0.312, 0.026, -0.041, 0.003, -0.040, 0.064, 0.344
  - TradeDev: -0.668, 1.000, -0.268, -0.014, 0.199, -0.031, 0.157, 0.082, -0.374
  - FDI: 0.312, -0.268, 1.000, 0.406, 0.225, 0.405, 0.266, -0.344, -0.042
  - LProd: 0.026, -0.014, 0.406, 1.000, 0.715, 0.979, 0.759, -0.003, -0.269
  - ProdW: -0.041, 0.199, 0.225, 0.715, 1.000, 0.710, 0.984, -0.082, -0.173
  - LagLProd: 0.003, -0.031, 0.405, 0.979, 0.710, 1.000, 0.782, -0.065, -0.262
  - LagProdW: -0.040, 0.157, 0.266, 0.759, 0.984, 0.782, 1.000, -0.119, -0.171
  - EP: 0.064, 0.082, -0.344, -0.003, -0.082, -0.065, -0.119, 1.000, 0.026
  - UN: 0.344, -0.374, -0.042, -0.269, -0.173, -0.262, -0.171, 0.026, 1.000

*Source: Appendix III. Correlation Matrix of Explanatory Variables*

### Appendix IV. Multivariate Regression Results for Alternative Specifications

### Table A4.1 — Compensation Share on (Exports+Imports)/GDP, Lagged Labor Productivity, Employment Protection
- Pre-IT/Globalization Era (Level):
  - (Exports+Imports)/GDP: 0.079 (0.078)
  - Lagged Labor Productivity: 0.254*** (0.044)
  - Employment Protection: 0.005 (0.011)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 51
  - R2: 0.918
- Pre-IT/Globalization Era (Delta):
  - (Exports+Imports)/GDP: -0.128 ** (0.073)
  - Lagged Labor Productivity: 0.005 (0.022)
  - Employment Protection: 0.015*** (0.003)
  - Country Effects: none
  - Time (Period) Effects: fixed
  - N: 51
  - R2: 0.454
- Post-IT/Globalization Era (Level):
  - (Exports+Imports)/GDP: -0.147*** (0.016)
  - Lagged Labor Productivity: -0.143*** (0.036)
  - Employment Protection: 0.017 (0.016)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 72
  - R2: 0.938
- Post-IT/Globalization Era (Delta):
  - (Exports+Imports)/GDP: -0.146*** (0.029)
  - Lagged Labor Productivity: 0.007 (0.005)
  - Employment Protection: 0.009 (0.009)
  - Country Effects: none
  - Time (Period) Effects: none
  - N: 55
  - R2: 0.261

### Table A4.2 — Employment Share: (Exports+Imports)/GDP, Lagged Labor Productivity, Employment Protection
- Pre-IT/Globalization Era (Level):
  - (Exports+Imports)/GDP: -0.180** (0.078)
  - Lagged Labor Productivity: 0.233*** (0.013)
  - Employment Protection: 0.045*** (0.001)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 31
  - R2: 0.974
- Pre-IT/Globalization Era (Delta):
  - (Exports+Imports)/GDP: -0.128** (0.052)
  - Lagged Labor Productivity: 0.038* (0.019)
  - Employment Protection: 0.089*** (0.024)
  - Country Effects: none
  - Time (Period) Effects: none
  - N: 29
  - R2: 0.195
- Post-IT/Globalization Era (Level):
  - (Exports+Imports)/GDP: -0.166*** (0.040)
  - Lagged Labor Productivity: -0.066*** (0.025)
  - Employment Protection: 0.099*** (0.025)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 68
  - R2: 0.807
- Post-IT/Globalization Era (Delta):
  - (Exports+Imports)/GDP: -0.185*** (0.045)
  - Lagged Labor Productivity: 0.011* (0.006)
  - Employment Protection: 0.052 (0.042)
  - Country Effects: none
  - Time (Period) Effects: none
  - N: 51
  - R2: 0.203

### Table A4.3 — Compensation Share: Trade Shares w/Dev. Countries, Labor Productivity, Employment Protection
- Pre-IT/Globalization Era (Level):
  - Trade Share w/ Dev. Countries: 0.176*** (0.033)
  - Labor Productivity: 0.221*** (0.033)
  - Employment Protection: 0.016*** (0.004)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 66
  - R2: 0.925
- Pre-IT/Globalization Era (Delta):
  - Trade Share w/ Dev. Countries: 0.141** (0.060)
  - Labor Productivity: 0.121** (0.056)
  - Employment Protection: 0.023*** (0.006)
  - Country Effects: none
  - Time (Period) Effects: fixed
  - N: 67
  - R2: 0.472
- Post-IT/Globalization Era (Level):
  - Trade Share w/ Dev. Countries: 0.024 (0.071)
  - Labor Productivity: -0.189*** (0.009)
  - Employment Protection: 0.011 (0.013)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 72
  - R2: 0.920
- Post-IT/Globalization Era (Delta):
  - Trade Share w/ Dev. Countries: 0.015 (0.069)
  - Labor Productivity: -0.373*** (0.047)
  - Employment Protection: 0.010 (0.031)
  - Country Effects: none
  - Time (Period) Effects: none
  - N: 54
  - R2: 0.272

### Table A4.4 — Employment Share: Trade Shares w/Dev. Countries, Labor Productivity, Employment Protection
- Pre-IT/Globalization Era (Level):
  - Trade Share w/ Dev. Countries: 0.234*** (0.034)
  - Labor Productivity: 0.115*** (0.004)
  - Employment Protection: 0.031*** (0.010)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 33
  - R2: 0.966
- Pre-IT/Globalization Era (Delta):
  - Trade Share w/ Dev. Countries: 0.355*** (0.116)
  - Labor Productivity: 0.118 (0.296)
  - Employment Protection: 0.066*** (0.012)
  - Country Effects: none
  - Time (Period) Effects: none
  - N: 30
  - R2: 0.257
- Post-IT/Globalization Era (Level):
  - Trade Share w/ Dev. Countries: -0.048 (0.111)
  - Labor Productivity: -0.138*** (0.019)
  - Employment Protection: 0.090*** (0.025)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 68
  - R2: 0.784
- Post-IT/Globalization Era (Delta):
  - Trade Share w/ Dev. Countries: -0.016 (0.026)
  - Labor Productivity: -0.224*** (0.083)
  - Employment Protection: 0.069** (0.027)
  - Country Effects: none
  - Time (Period) Effects: fixed
  - N: 51
  - R2: 0.323

### Table A4.5 — Compensation Share: (Exports+Imports)/GDP, Productivity per Worker, Employment Protection
- Pre-IT/Globalization Era (Level):
  - (Exports+Imports)/GDP: -0.036 (0.057)
  - Productivity per Worker: 0.286*** (0.046)
  - Employment Protection: 0.016*** (0.006)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 64
  - R2: 0.936
- Pre-IT/Globalization Era (Delta):
  - (Exports+Imports)/GDP: -0.016 (0.071)
  - Productivity per Worker: 0.133** (0.059)
  - Employment Protection: 0.016*** (0.005)
  - Country Effects: none
  - Time (Period) Effects: fixed
  - N: 66
  - R2: 0.454
- IT/Globalization Era (Level):
  - (Exports+Imports)/GDP: -0.132*** (0.018)
  - Productivity per Worker: -0.252*** (0.043)
  - Employment Protection: 0.016 (0.016)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 72
  - R2: 0.938
- IT/Globalization Era (Delta):
  - (Exports+Imports)/GDP: -0.123*** (0.013)
  - Productivity per Worker: -0.415*** (0.094)
  - Employment Protection: 0.013 (0.033)
  - Country Effects: none
  - Time (Period) Effects: none
  - N: 54
  - R2: 0.375

### Table A4.6 — Compensation Share: Trade Share s/Dev/ Countries, Productivity per Worker, Employment Protection
- Pre-IT/Globalization Era (Level):
  - Trade Share w/Dev. Countries: 0.183*** (0.034)
  - Productivity per Worker: 0.256*** (0.041)
  - Employment Protection: 0.017*** (0.003)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 64
  - R2: 0.941
- Pre-IT/Globalization Era (Delta):
  - Trade Share w/Dev. Countries: 0.340*** (0.087)
  - Productivity per Worker: 0.274** (0.116)
  - Employment Protection: 0.038*** (0.009)
  - Country Effects: none
  - Time (Period) Effects: none
  - N: 66
  - R2: 0.287
- IT/Globalization Era (Level):
  - Trade Share w/Dev. Countries: 0.013 (0.071)
  - Productivity per Worker: -0.348*** (0.019)
  - Employment Protection: 0.013 (0.013)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 72
  - R2: 0.919
- IT/Globalization Era (Delta):
  - Trade Share w/Dev. Countries: -0.010 (0.078)
  - Productivity per Worker: -0.539*** (0.104)
  - Employment Protection: 0.019 (0.034)
  - Country Effects: none
  - Time (Period) Effects: none
  - N: 54
  - R2: 0.209

### Table A4.7 — Compensation Share: (Exports+Imports)/GDP, Lagged Labor Productivity, Union Density
- Pre-IT/Globalization Era (Level):
  - (Exports+Imports)/GDP: 0.107* (0.061)
  - Lagged Labor Productivity: 0.261*** (0.035)
  - Union Density: -0.121 (0.083)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 51
  - R2: 0.920
- Pre-IT/Globalization Era (Delta):
  - (Exports+Imports)/GDP: 0.013 (0.061)
  - Lagged Labor Productivity: -0.006 (0.024)
  - Union Density: -0.136*** (0.009)
  - Country Effects: none
  - Time (Period) Effects: fixed
  - N: 66
  - R2: 0.471
- IT/Globalization Era (Level):
  - (Exports+Imports)/GDP: -0.149*** (0.014)
  - Lagged Labor Productivity: -0.140*** (0.030)
  - Union Density: 0.043 (0.039)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 72
  - R2: 0.937
- IT/Globalization Era (Delta):
  - (Exports+Imports)/GDP: -0.151*** (0.019)
  - Lagged Labor Productivity: -0.007 (0.006)
  - Union Density: 0.135 (0.071)
  - Country Effects: none
  - Time (Period) Effects: none
  - N: 54
  - R2: 0.291

### Table A4.8 — Compensation Share: Trade Share s/Dev/ Countries, Labor Productivity, Union Density
- Pre-IT/Globalization Era (Level):
  - Trade Share w/ Dev. Countries: 0.145*** (0.034)
  - Labor Productivity: 0.247*** (0.034)
  - Union Density: 0.007 (0.052)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 66
  - R2: 0.921
- Pre-IT/Globalization Era (Delta):
  - Trade Share w/ Dev. Countries: 0.090 (0.076)
  - Labor Productivity: 0.085 (0.052)
  - Union Density: -0.081** (0.035)
  - Country Effects: none
  - Time (Period) Effects: fixed
  - N: 67
  - R2: 0.462
- IT/Globalization Era (Level):
  - Trade Share w/ Dev. Countries: 0.023 (0.070)
  - Labor Productivity: -0.193*** (0.026)
  - Union Density: 0.009 (0.053)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 72
  - R2: 0.920
- IT/Globalization Era (Delta):
  - Trade Share w/ Dev. Countries: 0.006 (0.068)
  - Labor Productivity: -0.363*** (0.066)
  - Union Density: 0.127 (0.051)
  - Country Effects: none
  - Time (Period) Effects: none
  - N: 54
  - R2: 0.270

### Table A4.9 — Compensation Share: FDI/GDP Ratio, Labor Productivity, Employment Protection
- Pre-IT/Globalization Era (Level):
  - FDI to GDP Ratio: 0.141*** (0.013)
  - Labor Productivity: 0.398*** (0.027)
  - Employment Protection: 0.036*** (0.002)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 38
  - R2: 0.971
- Pre-IT/Globalization Era (Delta):
  - FDI to GDP Ratio: -0.01 (0.086)
  - Labor Productivity: 0.177* (0.089)
  - Employment Protection: 0.026*** (0.001)
  - Country Effects: none
  - Time (Period) Effects: fixed
  - N: 36
  - R2: 0.525
- IT/Globalization Era (Level):
  - FDI to GDP Ratio: -0.025 (0.017)
  - Labor Productivity: -0.170*** (0.012)
  - Employment Protection: 0.002 (0.011)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 72
  - R2: 0.921
- IT/Globalization Era (Delta):
  - FDI to GDP Ratio: -0.011 (0.022)
  - Labor Productivity: -0.369*** (0.034)
  - Employment Protection: 0.005 (0.031)
  - Country Effects: none
  - Time (Period) Effects: none
  - N: 54
  - R2: 0.271

*Source: Appendix IV. Multivariate Regression Results for Alternative Specifications*

### Appendix V. Inequality

### Table A5.1 — Gini Coefficient on Openness, Productivity, Bargaining Power
- Pre-IT/globalization Era (Level and Delta) and IT/Globalization Era (Level and Delta) coefficients by regressor:
  - Openness:
    - (Exports+Imports)/GDP:
      - Pre-IT Level: -0.219 (0.170)
      - Pre-IT Delta: -0.242* (0.131)
      - IT Level: 0.035 (0.020)
      - IT Delta: -0.076 (0.062)
    - Trade w/ Developing:
      - Pre-IT Level: 0.105 (0.084)
      - Pre-IT Delta: -0.065 (0.079)
      - IT Level: 0 .089 (0.095)
      - IT Delta: 0.038 (0.114)
    - FDI/GDP:
      - Pre-IT Level: 0.091 (0.057)
      - Pre-IT Delta: -0.067 (0.069)
      - IT Level: 0.049*** (0.038)
      - IT Delta: -0.010 (0.094)
  - Productivity:
    - Labor Productivity:
      - Pre-IT Level: -0.211*** (0.033)
      - Pre-IT Delta: 0.125 (0.350)
      - IT Level: -0.012** (0.057)
      - IT Delta: -0.088 (0.107)
    - Productivity per Worker:
      - Pre-IT Level: -0.208*** (0.043)
      - Pre-IT Delta: 0.039 (0.251)
      - IT Level: 0.068*** (0.021)
      - IT Delta: -0.119 (0.153)
    - Lagged Labor Productivity:
      - Pre-IT Level: -0.165*** (0.008)
      - Pre-IT Delta: 0.027*** (0.010)
      - IT Level: 0.035*** (0.011)
      - IT Delta: 0.033*** (0.011)
    - Lagged Productivity Per Worker:
      - Pre-IT Level: -0.094*** (0.012)
      - Pre-IT Delta: -0.178 (0.171)
      - IT Level: 0.052*** (0.017)
      - IT Delta: -0.264 (0.183)
  - Bargaining Power:
    - Employment Protection:
      - Pre-IT Level: -0.098*** (0.006)
      - Pre-IT Delta: -0.083 (0.007)
      - IT Level: -0.027 (0.013)
      - IT Delta: -0.006 (0.026)
    - Union Density:
      - Pre-IT Level: -0.423*** (0.149)
      - Pre-IT Delta: -0.014 (0.087)
      - IT Level: -0.330 (0.104)
      - IT Delta: -0.240 (0.344)

*Source: Appendix V. Inequality*

### Appendix VI. Multivariate Regression Results for Alternative Specifications (Inequality)

### Table A6.1 — Gini on (Exports+Imports)/GDP, Lagged Labor Productivity, Employment Protection
- Pre-IT/globalization Era (Level):
  - (Exports+Imports)/GDP: 0.037 (0.179)
  - Lagged Labor Productivity: -0.124*** (0.035)
  - Employment Protection: -0.108*** (0.023)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 47
  - R2: 0.792
- Pre-IT/globalization Era (Delta):
  - (Exports+Imports)/GDP: -0.175 (0.171)
  - Lagged Labor Productivity: 0.002 (0.013)
  - Employment Protection: -0.085 (0.074)
  - Country Effects: none
  - Time (Period) Effects: none
  - N: 45
  - R2: 0.139
- IT/Globalization Era (Level):
  - (Exports+Imports)/GDP: -0.015 (0.041)
  - Lagged Labor Productivity: 0.093** (0.037)
  - Employment Protection: -0.001 (0.014)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 70
  - R2: 0.751
- IT/Globalization Era (Delta):
  - (Exports+Imports)/GDP: -0.028 (0.050)
  - Lagged Labor Productivity: 0.023 (0.020)
  - Employment Protection: -0.013 (0.026)
  - Country Effects: none
  - Time (Period) Effects: none
  - N: 51
  - R2: 0.064

### Table A6.2 — Gini on Trade Share w/Developed Countries, Lagged Labor Productivity, Employment Protection
- Pre-IT/globalization Era (Level):
  - Trade Share w/Developed Countries: -0.195** (0.094)
  - Lagged Labor Productivity: -0.075*** (0.011)
  - Employment Protection: -0.110*** (0.036)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 47
  - R2: 0.796
- Pre-IT/globalization Era (Delta):
  - Trade Share w/Developed Countries: -0.158*** (0.052)
  - Lagged Labor Productivity: 0 (0.007)
  - Employment Protection: -0.096 (0.074)
  - Country Effects: none
  - Time (Period) Effects: none
  - N: 45
  - R2: 0.120
- IT/Globalization Era (Level):
  - Trade Share w/Developed Countries: 0.039 (0.087)
  - Lagged Labor Productivity: 0.016** (0.030)
  - Employment Protection: 0.001 (0.017)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 70
  - R2: 0.752
- IT/Globalization Era (Delta):
  - Trade Share w/Developed Countries: 0.024 (0.070)
  - Lagged Labor Productivity: 0.092*** (0.018)
  - Employment Protection: -0.009 (0.040)
  - Country Effects: none
  - Time (Period) Effects: none
  - N: 51
  - R2: 0.060

### Table A6.3 — Gini on FDI/GDP Ratio, Lagged Labor Productivity, Employment Protection
- Pre-IT/globalization Era (Level):
  - FDI/GDP Ratio: 0.246*** (0.0293)
  - Lagged Labor Productivity: -0.209*** (0.007)
  - Employment Protection: -0.193*** (0.000)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 34
  - R2: 0.859
- Pre-IT/globalization Era (Delta):
  - FDI/GDP Ratio: -0.026 (0.015)
  - Lagged Labor Productivity: -0.003* (0.002)
  - Employment Protection: -0.177*** (0.056)
  - Country Effects: none
  - Time (Period) Effects: none
  - N: 32
  - R2: 0.259
- IT/Globalization Era (Level):
  - FDI/GDP Ratio: 0.073*** (0.016)
  - Lagged Labor Productivity: 0.023 (0.033)
  - Employment Protection: 0.027* (0.015)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 70
  - R2: 0.762
- IT/Globalization Era (Delta):
  - FDI/GDP Ratio: 0.013 (0.076)
  - Lagged Labor Productivity: 0.024 (0.022)
  - Employment Protection: -0.008 (0.013)
  - Country Effects: none
  - Time (Period) Effects: none
  - N: 51
  - R2: 0.061

### Table A6.4 — Gini on FDI/GDP Ratio, Lagged Productivity per Worker, Employment Protection
- Pre-IT/globalization Era (Level):
  - FDI/GDP Ratio: 0.013 (0.014)
  - Lagged Productivity per Worker: -0.188*** (0.003)
  - Employment Protection: -0.201*** (0.002)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 33
  - R2: 0.873
- Pre-IT/globalization Era (Delta):
  - FDI/GDP Ratio: 0.256** (0.110)
  - Lagged Productivity per Worker: -0.302 (0.193)
  - Employment Protection: -0.180** (0.067)
  - Country Effects: none
  - Time (Period) Effects: none
  - N: 31
  - R2: 0.318
- IT/Globalization Era (Level):
  - FDI/GDP Ratio: 0.074*** (0.013)
  - Lagged Productivity per Worker: 0.037 (0.051)
  - Employment Protection: 0.021* (0.015)
  - Country Effects: fixed
  - Time (Period) Effects: none
  - N: 70
  - R2: 0.762
- IT/Globalization Era (Delta):
  - FDI/GDP Ratio: -0.000 (0.080)
  - Lagged Productivity per Worker: -0.334 (0.224)
  - Employment Protection: 0.001 (0.011)
  - Country Effects: none
  - Time (Period) Effects: none
  - N: 51
  - R2: 0.074

*Source: Appendix VI. Multivariate regression results for alternative specifications*

### Appendix VII. Capital-Augmenting Technological Progress

### Appendix VII. Capital-Augmenting Technological Progress

### Production function and definitions
- Production function used:
  - Y = F(K,L) = [ α(AK)^( (σ-1)/σ ) + (1- α) L^( (σ-1)/σ ) ]^( σ/(σ-1) )
- Variables and parameters:
  - Y: output
  - K: capital stock
  - L: labor stock
  - A: capital-augmenting technology
  - σ: elasticity of technical substitution between capital and labor
  - α: distribution parameter in the CES function

### Factor shares (exact expressions)
- Capital’s share:
  - KShare = K * MP_K / Y = α A^( (σ-1)/σ ) (K/Y)^( (σ-1)/σ )
- Labor’s share:
  - LShare = L * MP_L / Y = 1 - KShare = 1 - α A^( (σ-1)/σ ) (K/Y)^( (σ-1)/σ )

### Derivatives: how shares respond to K/Y and A
- Partial derivative of labor share with respect to K/Y:
  - ∂LShare/∂(K/Y) = - α ( (σ-1)/σ ) A^( (σ-1)/σ ) (K/Y)^( -1/σ )
- Partial derivative of labor share with respect to A:
  - ∂LShare/∂A = - α ( (σ-1)/σ ) A^( -1/σ ) (K/Y)^( (σ-1)/σ )

### Key implications and scenarios (exact logical conditions preserved)
- Constant-returns-to-scale Cobb-Douglas production function Y = A K^α L^(1-α) cannot explain changes in labor and capital shares because it assumes constant factor shares; technology shocks (A) cannot affect income shares in that specification.
- Using a CES production function with capital-augmenting technological progress allows technology (A) and the capital-output ratio (K/Y) to affect factor shares.
- Effects of movements in K/Y on factor shares depend on σ:
  - If labor and capital are technical complements (σ < 1):
    - As K/Y increases, capital share will decrease, and labor share will increase.
  - If labor and capital are technical substitutes (σ > 1):
    - As K/Y increases, labor’s share will decrease.
- Effect of a positive technology shock A, holding K/Y constant:
  - A positive technology shock A will increase capital share and decrease labor share as long as σ > 1.

*Source: _wp06294 - Appendix VII. Capital-Augmenting Technological Progress*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp06294.pdf_