## Capital Account Liberalization and Wage Inequality: Evidence from Firm Level Data

## Source details

**Canonical URL:** [Capital Account Liberalization and Wage Inequality: Evidence from Firm Level Data](https://www.imf.org/-/media/files/publications/wp/2023/english/wpiea2023048-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2023/english/wpiea2023048-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2023/english/wpiea2023048-print-pdf.pdf.json)

---

### Research questions and scope
- Does capital account liberalization induce an increase in wage inequality?
- What are the main channels?
- Is there heterogeneity across countries and industries?
- Sample: firm-level data from Orbis for ASEAN5 countries over the period 1995–2019.
- Main outcome: salary per employee (log). Main empirical focus: between-firms wage inequality.

### Data and key measures
- Firm-level:
  - salary per employee (Log).
  - firm size (number of employees).
  - total factor productivity (TFP) estimated following Ackerberg et al. (2015) / Diez et al. (2021).
  - profit-to-wage ratio.
- Country-level:
  - Gini (market and disposable income) from SWIID.
  - capital account liberalization indices: Chinn-Ito, FKRSU (overall, inflows, outflows, FDI, equity, bonds); FARI (robustness).
  - Financial development index from IMF (Sahay et al., 2015; Svirydzenka, 2016).
  - Financial crises from Nguyen et al. (2022).
  - Informal sector size from Medina and Schneider (2020).
  - Collective bargaining coverage from ILO.
- Industry-level:
  - export orientation (share of domestic value added in foreign final demand) from OECD TiVA.
  - external financial dependence (EFD) computed as median firm-level EFD (firm external finance use over ten years divided by capital expenditure).
- Stylized fact:
  - Between-firms wage dispersion (standard deviation of log wage per employee) accounts for 17 percent of variation in market inequality.
  - Table 1 result examples: Between‑Firms Wage Dispersion coefficient 0.475*** for Gini – market income; Within R2 0.169; and 0.346*** for Gini – disposable income; Within R2 0.074.

### Empirical strategy
- Baseline specification: log wage per employee at firm f, industry i, country c, year t regressed on KAL index interacted with firm initial wage rank (rank scaled 0–1 based on initial salary per employee), controls X (including firm size, TFP), and firm, country, year fixed effects.
- Coefficient of interest: β1 on (KAL index # Initial salary rank). β1>0 implies wages grow faster in initially high-paying firms (increases between-firm wage inequality); β1<0 implies the opposite.

### Main findings
- Baseline effects:
  - Capital account liberalization increases between-firms wage inequality (KA Openness # Initial salary rank positive and highly significant across Chinn-Ito and FKRSU specifications).
  - Examples of reported interaction coefficients (KA Openness # Initial salary rank): 2.140***; 1.647***; 1.712***; 0.910***. FKRSU columns include 8.324***; 7.516***; 6.810***; 9.306***.
  - Interpreted magnitude: column (4) implies a one standard deviation increase in capital account openness leads to about a 25 percent wage differential between the top‑ranking firm and lowest ranking firm in the sample.
- Robustness checks:
  - Results robust to reform dummies, top/bottom 20% distributional checks, GMM dynamic specification, country-year fixed effects, and alternative KAL index (FARI).
  - Table 4 examples: KA openness # Top 20% 0.680***; 0.434***; 1.945***. KA openness # Bottom 20% -1.212***; -1.110***; -6.339***.
- Channels and heterogeneity:
  - Direction and type of flows:
    - Both inflow and outflow liberalization increase between-firm inequality, with larger effects for inflows. Example: KA Openness # Initial salary rank 5.919*** for inflows (Table 8).
    - Equity flow liberalization has larger inequality effects than bonds. Example: KA Openness # Initial salary rank 5.389*** for equity (Table 9).
    - FDI inflow liberalization increases between-firm inequality; effect amplified when directed to firms with initially high TFP. Examples: KA Openness # Initial salary rank 0.360* and KA Openness # Initial salary rank # High initial TFP 0.699* (Table 10).
  - Profit-to-wage channel:
    - Capital account liberalization induces an increase in Profit-to-Wage ratios, especially at firms with initially high profit-to-wage ratios. Examples: KA Openness # Initial salary rank # High initial profit-wage 1.062**, 1.070***; KA Openness # Initial salary rank positive 1.530*** (Table 14).
    - Profit-to-wage ratio regressions confirm increases: KA Openness-lagged and KA Openness-lagged # High initial profit-wage positive and significant in some specifications (Table 15).
  - Country and industry heterogeneity:
    - Financial development mitigates the inequality effect of liberalization. Example: KA Openness # Initial salary rank # High FD -4.353*** in (1) of Table 12.
    - Large informal sector amplifies the effect. Example: KA Openness # Initial salary rank # CCj 0.098*** in the column for Informal Sector Size (Table 13).
    - Collective bargaining coverage mitigates the effect. Example: collective bargaining coverage coefficient CCj -0.202*** and offsetting interaction signs in Table 13.
    - Industries with high external financial dependence and high export orientation tend to mitigate the impact (interactions in Table 13 show mitigation).
  - Financial crises:
    - No statistically significant moderating role for financial crises in the sample (Table 11), attributed to scarcity of crisis episodes—most around the Asian financial crisis 1997–2001.

### Dynamic evidence
- Event-study design:
  - Event defined as largest change in capital account openness index within a country.
  - Effects on between-firms wage inequality are persistent in both short and medium term (Figure 2 shows coefficients plotted with 90% confidence intervals indicating persistence).
  - Interpretation: supports a plausibly causal effect and persistent treatment pattern.

### Magnitudes and selected statistics (preserved as reported)
- Between-Firms Wage Dispersion accounts for 17 percent of variation in market inequality (text).
- Table 1 coefficients: Between‑Firms Wage Dispersion 0.346*** (Gini – disposable income), 0.475*** (Gini – market income). Within R2: 0.074 and 0.169.
- Table 2 interaction examples (KA Openness # Initial salary rank): 2.140***; 1.647***; 1.712***; 0.910***; FKRSU columns include 8.324***; 7.516***; 6.810***; 9.306***.
- Table 4 interaction examples: KA openness # Top 20% 0.680***; 0.434***; 1.945***. KA openness # Bottom 20% -1.212***; -1.110***; -6.339***.
- Table A1 summary statistics highlights:
  - Gini coefficient (disp. income) mean 41.66, Std. Dev. 2.39, Min 37.50, Max 47.40 (Obs. 124).
  - Chinn-Ito Index mean 0.55, Std. Dev. 0.27 (Obs. 125).
  - FKRSU index – overall flows mean 0.45, Std. Dev. 0.26 (Obs. 125).
  - Financial development index mean 0.50, Std. Dev. 0.16 (Obs. 125).
  - Informal sector size mean 30.74, Std. Dev. 13.33 (Obs. 115).
  - Salary per employee (Log) Obs. 27,868; Mean 9.01; Std. Dev. 1.07; Min -0.22; Max 16.02.
  - Profit-to-wage ratio Obs. 27,868; Mean 2.13; Std. Dev. 63.42; Min -1914.96; Max 8170.

### Policy implications and recommendations
- Policies that protect workers and mitigate distributional consequences of financial globalization:
  - Increase collective bargaining coverage.
  - Promote financial development.
  - Promote trade liberalization (distributional effect that may concentrate workers in larger firms, reducing wage dispersion).
  - Reduce the size of the informal sector.
- Rationale:
  - These policies can limit pass-through of firm-level differences to wages, improve access to finance for broader firms/households, and reduce mechanisms by which liberalization disproportionately benefits initially high‑paying firms.

### Contributions and scope for future research
- Contributions:
  - First firm-level evidence (to authors’ knowledge) on distributional consequences of capital account liberalization with a focus on between-firms wage inequality.
  - Bridges literature on firm wage inequality and capital account liberalization; employs a novel firm-level empirical strategy.
- Future research:
  - Explore within-firm wage inequality dynamics when employer-employee matched data become available.

*Italic source: IMF Working Paper — Capital Account Liberalization and Wage Inequality: Evidence from Firm Level Data (References and chapter content provided).*

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

### Capital Account Liberalization and Wage Inequality: Evidence from Firm Level Data

### Research questions and scope
- Does capital account liberalization induce an increase in wage inequality?
- What are the main channels?
- Is there heterogeneity across countries and industries?
- Sample: firm-level data from Orbis for ASEAN5 countries over the period 1995–2019.
- Main outcome: salary per employee (log). Main empirical focus: between-firms wage inequality.

### Data and key measures
- Firm-level: salary per employee; firm size (number of employees); total factor productivity (TFP); profit-to-wage ratio. TFP estimated following Ackerberg et al. (2015) / Diez et al. (2021).
- Country-level: Gini (market and disposable income) from SWIID; capital account liberalization indices: Chinn-Ito, FKRSU (overall, inflows, outflows, FDI, equity, bonds); FARI (robustness). Financial development index from IMF (Sahay et al., 2015; Svirydzenka, 2016). Financial crises from Nguyen et al. (2022). Informal sector size from Medina and Schneider (2020). Collective bargaining coverage from ILO.
- Industry-level: export orientation (share of domestic value added in foreign final demand) from OECD TiVA; external financial dependence (EFD) computed as median firm-level EFD (firm external finance use over ten years divided by capital expenditure).
- Stylized fact: between-firms wage dispersion (standard deviation of log wage per employee) accounts for a non-trivial 17 percent of variation in market inequality (Table 1 result: Between-Firms Wage Dispersion coefficient 0.475*** for Gini – market income; Within R2 0.169; other column: 0.346*** for Gini – disposable income; Within R2 0.074).

### Empirical strategy
- Baseline specification: log wage per employee at firm f, industry i, country c, year t regressed on KAL index interacted with firm initial wage rank (rank scaled 0–1 based on initial salary per employee), controls X (including firm size, TFP), and firm, country, year fixed effects.
- Coefficient of interest: β1 on (KAL index # Initial salary rank). β1>0 implies wages grow faster in initially high-paying firms (increases between-firm wage inequality); β1<0 implies the opposite.

### Main findings
- Baseline:
  - Capital account liberalization increases between-firms wage inequality. (Table 2, Chinn-Ito and FKRSU specifications: KA Openness # Initial salary rank positive and highly significant; examples: 2.140***, 1.647***, 1.712***, 0.910*** in columns reported).
  - Interpreted magnitude: column (4) implies a one standard deviation increase in capital account openness leads to about a 25 percent wage differential between the top‑ranking firm and lowest ranking firm in the sample (text statement).
- Robustness:
  - Results robust to reform dummies (Table 3), top/bottom 20% distributional checks (Table 4), GMM dynamic specification (Table 5), country-year fixed effects (Table 6), and alternative KAL index (FARI, Table 7).
  - Table 4: salary per employee grows faster at the top 20% and decreases at the bottom 20% following liberalization (examples: KA openness # Top 20% 0.680***; KA openness # Bottom 20% -1.212***).
- Channels and heterogeneity:
  - Direction and type of flows matter:
    - Both inflow and outflow liberalization increase between-firm inequality, with larger effects for inflows (Table 8: KA Openness # Initial salary rank examples 5.919*** for inflows vs smaller for outflows).
    - Equity flow liberalization has larger inequality effects than bonds (Table 9: KA Openness # Initial salary rank 5.389*** for equity; bond-related interactions smaller or less robust).
    - FDI inflow liberalization increases between-firm inequality; effect amplified when directed to firms with initially high TFP (Table 10: KA Openness # Initial salary rank and KA Openness # Initial salary rank # High initial TFP positive and sometimes significant; example KA Openness # Initial salary rank 0.360* and KA Openness # Initial salary rank # High initial TFP 0.699* in panel).
  - Profit-to-wage channel:
    - Capital account liberalization induces an increase in Profit-to-Wage ratios, especially at firms with initially high profit-to-wage ratios (Table 14: KA Openness # Initial salary rank # High initial profit-wage examples 1.062**, 1.070***; and KA Openness # Initial salary rank positive 1.530***).
    - Profit-to-wage ratio regressions confirm increases (Table 15: KA Openness-lagged and KA Openness-lagged # High initial profit-wage positive and significant in some specifications).
  - Country and industry heterogeneity:
    - Financial development mitigates the inequality effect of liberalization (Table 12: interactions KA Openness # Initial salary rank # High FD show negative significant coefficients, e.g., -4.353*** in (1)).
    - Large informal sector amplifies the effect (Table 13: interactions show amplification; example KA Openness # Initial salary rank # CCj 0.098*** in column for Informal Sector Size).
    - Collective bargaining coverage mitigates the effect (Table 13: KA Openness # CCj and KA Openness # Initial salary rank # CCj show offsetting signs; collective bargaining coverage coefficient CCj -0.202***).
    - Industries with high external financial dependence and high export orientation tend to mitigate the impact (Table 13: EFD and export dependence interactions show mitigation; examples KA Openness # Initial salary rank coefficients and KA Openness # CCj).
  - Financial crises:
    - No statistically significant moderating role for financial crises in the sample (Table 11; scarcity of crisis episodes—most around Asian financial crisis 1997–2001).

### Dynamic evidence
- Event-study:
  - Event defined as largest change in capital account openness index within a country.
  - Effects on between-firms wage inequality are persistent in both short and medium term (Figure 2; coefficients plotted with 90% confidence intervals show persistence).
  - Interpretation: supports plausibly causal effect and persistent treatment pattern.

### Magnitudes and selected statistics (preserved as reported)
- Between-Firms Wage Dispersion accounts for 17 percent of variation in market inequality (text).
- Table 1 coefficients: Between‑Firms Wage Dispersion 0.346*** (Gini – disposable income), 0.475*** (Gini – market income). Within R2: 0.074 and 0.169.
- Table 2 interaction examples (KA Openness # Initial salary rank): 2.140***; 1.647***; 1.712***; 0.910***; FKRSU columns include 8.324***; 7.516***; 6.810***; 9.306***.
- Table 4 interaction examples: KA openness # Top 20% 0.680***; 0.434***; 1.945***. KA openness # Bottom 20% -1.212***; -1.110***; -6.339***.
- Table A1 summary statistics highlights:
  - Gini coefficient (disp. income) mean 41.66, Std. Dev. 2.39, Min 37.50, Max 47.40 (Obs. 124).
  - Chinn-Ito Index mean 0.55, Std. Dev. 0.27 (Obs. 125).
  - FKRSU index – overall flows mean 0.45, Std. Dev. 0.26 (Obs. 125).
  - Financial development index mean 0.50, Std. Dev. 0.16 (Obs. 125).
  - Informal sector size mean 30.74, Std. Dev. 13.33 (Obs. 115).
  - Salary per employee (Log) Obs. 27,868; Mean 9.01; Std. Dev. 1.07; Min -0.22; Max 16.02.
  - Profit-to-wage ratio Obs. 27,868; Mean 2.13; Std. Dev. 63.42; Min -1914.96; Max 8170.

### Policy implications and recommendations (as stated)
- Policies that protect workers and mitigate distributional consequences of financial globalization:
  - Increase collective bargaining coverage.
  - Promote financial development.
  - Promote trade liberalization (distributional effect that may concentrate workers in larger firms, reducing wage dispersion).
  - Reduce the size of the informal sector.
- Rationale: these policies can limit pass-through of firm-level differences to wages, improve access to finance for broader firms/households, and reduce mechanisms by which liberalization disproportionately benefits initially high‑paying firms.

### Contributions and scope for future research
- Contributions:
  - First firm-level evidence (to authors’ knowledge) on distributional consequences of capital account liberalization with a focus on between-firms wage inequality.
  - Bridges literature on firm wage inequality and capital account liberalization; employs novel firm-level empirical strategy.
- Future research:
  - Explore within-firm wage inequality dynamics when employer-employee matched data become available.

*Italic source: IMF Working Paper — Capital Account Liberalization and Wage Inequality: Evidence from Firm Level Data (References and chapter content provided).*

### References

### References

### Major topics covered by the references
- Firm-level heterogeneity and wage inequality (e.g., Abowd, Kramarz, and Margolis (1999); Card, Heining, and Kline (2013); Card et al. (2018); Song et al. (2019); Criscuolo et al. (2021)).
- Capital account liberalization, financial globalization, and distributional effects (e.g., Furceri and Loungani (2015); Furceri and Loungani (2018); Furceri, Loungani, and Ostry (2019); Li and Su (2021); Bumann and Lensink (2016); Das and Mohapatra (2003)).
- Measures, indices, and datasets for financial openness, capital controls, and financial development (e.g., Chinn and Ito (2008); Fernández et al. (2016); Baba et al. (forthcoming, 2023); Svirydzenka (2016); Nguyen, Castro, and Wood (2022)).
- Labor-market decomposition, earnings dynamics, and on‑the‑job training (e.g., Fortin, Lemieux, and Firpo (2011); Abowd, Kramarz, and Margolis (1999); Almeida and Faria (2014); Engbom et al. (2022); Engbom and Moser (2022, forthcoming)).
- Trade liberalization, firm heterogeneity, and wages (e.g., Krishna et al. (2011); Coşar, Guner, and Tybout (2016); Dorn, Fuest, and Potrafke (2018)).
- Shadow economy, corporate saving, external financing, and financial deepening (e.g., Medina and Schneider (2020); Li (2020); Sahay et al. (2015); Zhang and Naceur (2019)).

### Methodologies and empirical approaches cited
- Matched employer‑employee data and firm‑level analyses (e.g., Abowd, Kramarz, and Margolis (1999); Card et al. (2013); Krishna et al. (2011); Moser et al. (2021)).
- Production function estimation and identification properties (Ackerberg, Caves, and Frazer (2015)).
- Dynamic panel data methods and moment restrictions (Blundell and Bond (1998)).
- Decomposition techniques in labor economics (Fortin, Lemieux, and Firpo (2011)).
- Construction and use of broad-based indices and new datasets for capital control measures, financial openness, and financial crises (Chinn and Ito (2008); Fernández et al. (2016); Baba et al. (forthcoming, 2023); Nguyen, Castro, and Wood (2022)).

### Representative citations (as listed)
- Abowd, J. M., Kramarz, F., and Margolis, D. N. (1999). High wage workers and high wage firms. Econometrica, 67(2), 251-333.
- Ackerberg, D. A., K. Caves, and G. Frazer (2015). “Identification Properties of Recent Production Function Estimators.” Econometrica 83(6):2411-2451.
- Almeida, R. K., and Faria, M. (2014). The wage returns to on-the-job training: evidence from matched employer-employee data. IZA Journal of Labor and Development, 3(1), 1-33.
- Card, D., Heining, J., and Kline, P. (2013). Workplace heterogeneity and the rise of West German wage inequality. The Quarterly Journal of Economics, 128(3), 967-1015.
- Card, D., Cardoso, A. R., Heining, J., and Kline, P. (2018). Firms and labor market inequality: Evidence and some theory. Journal of Labor Economics, 36(S1), S13-S70.
- Chinn, M. D., and Ito, H. (2008). A new measure of financial openness. Journal of Comparative Policy Analysis, 10(3), 309-322.
- Fernández, A., Klein, M. W., Rebucci, A., Schindler, M. and Uribe, M. (2016). ‘Capital control measures: a new dataset’, IMF Economic Review, Vol. 64, pp. 548–574
- Furceri, D., and Loungani, M. P. (2015). Capital account liberalization and inequality. International Monetary Fund.
- Furceri, D., and Loungani, P. (2018). The distributional effects of capital account liberalization. Journal of Development Economics, 130, 127-144.
- Furceri, D., Loungani, P., and Ostry, J. D. (2019). The aggregate and distributional effects of financial globalization: Evidence from macro and sectoral data. Journal of Money, Credit and Banking, 51, 163-198.
- Svirydzenka, K. (2016). Introducing a New Broad-based Index of Financial Development. IMF Working Papers, 2016(005).
- Nguyen, T. C., Castro, V., and Wood, J. (2022). A new comprehensive database of financial crises: Identification, frequency, and duration. Economic Modelling, 108, 105770.
- Song, J., Price, D. J., Guvenen, F., Bloom, N., and Von Wachter, T. (2019). Firming up inequality. The Quarterly journal of economics, 134(1), 1-50.
- Engbom, N., Gonzaga, G., Moser, C., and Olivieri, R. (2022). Earnings inequality and dynamics in the presence of informality: The case of Brazil. (Forthcoming, Quantitative Economics)
- Engbom, N., and Moser, C. (2022, forthcoming). Earnings inequality and the minimum wage: Evidence from Brazil. (Forthcoming, American Economic Review).
- Baba, C., Cervantes, R., Darbar, S. M., Kokenyne, A., and Zotova, V., (forthcoming, 2023). “New Measures of Capital Flow Restrictions—AREAER Indices.” International Monetary Fund, Washington, DC.
- Li, X., and Su, D. (2021). Does Capital Account Liberalization Affect Income Inequality? Oxford Bulletin of Economics and Statistics, 83(2), 377-410.

*References section — Capital Account Liberalization and Wage Inequality: Evidence from Firm Level Data, Working Paper No. WP/23/48*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2023/english/wpiea2023048-print-pdf.pdf_