## wp17183 - References

## Source details

**Canonical URL:** [wp17183 - References](https://www.imf.org/-/media/files/publications/wp/2017/wp17183.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2017/wp17183.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2017/wp17183.pdf.json)

---

### References
- Aguiar, M., 2005, “Investment, devaluation, and foreign currency exposure: The case of Mexico,” Journal of Development Economics, Vol. 78(1), pp.95-113.
- Beck, T., Demirguc‐Kunt, A., Laeven, L. and Levine, R., 2008, “Finance, firm size, and growth,” Journal of Money, Credit and Banking, Vol. 40(7), pp.1379-1405.
- Berg, A., Ostry, J.D. and Zettelmeyer, J., 2012. “What makes growth sustained?” Journal of Development Economics, 98(2), pp.149-166.
- Bleakley, H. and Cowan, K., 2008, “Corporate dollar debt and depreciations: much ado about nothing?” The Review of Economics and Statistics, Vol. 90(4), pp. 612-626.
- Bleakley, H. and Cowan, K., 2010, “Maturity mismatch and financial crises: Evidence from emerging market corporations,” Journal of Development Economics, Vol. 93(2), pp.189-205.
- Chen, H. J. and Chen, S. J., 2012, “Investment-cash flow sensitivity cannot be a good measure of financial constraints: Evidence from the time series,” Journal of Financial Economics, Vol. 103(2), pp.393-410.
- Chen, R. and M. Dao, 2011, “The real exchange rate and employment in China,” IMF Working Paper No. 11/148, Washington, DC: International Monetary Fund.
- Claessens, C., Tong, H., and S.-J. Wei, 2012, “From the financial crisis to the real economy: Using firm-level data to identify transmission channels,” Journal of International Economics, Vol. 88, pp. 375–387.
- Desai, M. A., Foley, C. F., and K. J. Forbes, 2008, “Financial constraints and growth: Multinational and local firm responses to currency depreciations,” Review of Financial Studies, Vol. 21, pp. 2857–2888.
- Di Giovanni, J. and A. A. Levchenko, 2010, “Putting the parts together: Trade, vertical linkages, and business cycle comovement,” American Economic Journal: Macroeconomics, Vol. 2, pp. 95–124.
- Elsby, Michael W. L., Hobijn, B., and A. Sahin, 2013, “The decline of the US labor share,” Brookings Papers on Economic Activity, Fall 2013, pp. 1–52.
- Fazzari, S. M., Hubbard, R.G., and B. Petersen, 2000, “Investment-cash flow sensitivities are useful: A comment on Kaplan and Zingales,” Quarterly Journal of Economics, Vol. 115(2), pp. 695–705.
- Forbes, K., 2002, “How do large depreciations affect firm performance?” IMF Staff Papers, Vol. 49, pp. 214–238.
- Forbes, K., 2002, “Cheap labor meets costly capital: The impact of devaluations on commodity firms,” Journal of Development Economics, Vol. 69, pp. 335–365.
- Gilchrist, S. and Himmelberg, C.P., 1995, “Evidence on the role of cash flow for investment,” Journal of Monetary Economics, Vol. 36(3), pp.541-572.
- Hsieh, C.-T. and P. J. Klenow, 2009, “Misallocation and manufacturing TFP in China and India,” Quarterly Journal of Economics, Vol. CXXIV, Issue 4, pp. 1403–1448.
- Kalemli-Ozcan, S., Kamil, H., and Villegas-Sanchez, C., 2015, “What Hinders Investment in the Aftermath of Financial Crises? Insolvent Firms or Illiquid Banks?,” NBER Working Paper 16528.
- Kaplan, S. N. and L. Zingales, 1997, “Do investment-cash flow sensitivities provide useful measures of financing constraints?” Quarterly Journal of Economics, Feb. 1997, pp. 170–215.
- Kaminsky, G.L. and Schmukler, S.L., 2008, “Short-run pain, long-run gain: Financial liberalization and stock market cycles,” Review of Finance, 12(2), pp.253-292.
- Laeven, L. and F. Valencia, 2013, “Systemic banking crises database,” IMF Economic Review, Vol. 61, pp. 225–270.
- Lewellen, J. and K. Lewellen, forthcoming, “Investment and cash flow: New evidence,” Journal of Financial and Quantitative Analysis.
- Levy-Yeyati, E., Sturzenegger, F., and P. A. Gluzmann, 2013, “Fear of appreciation,” Journal of Development Economics, Vol. 101, pp. 233–247.
- Levy-Yeyati, E., Sturzenegger, F., and P. A. Gluzmann, 2012, “Exchange rate undervaluation and economic growth: Diaz Alejandro (1965) revisited,” Economics Letters, Vol. 117, pp. 666–672.
- Pagano, M. and Pica, G., 2012, “Finance and employment,” Economic Policy, Vol. 27, Issue 69, pp.5-55.
- Pratab, S. and C. Urrutia, 2004, “Firm Dynamics, Investment & Debt Portfolios: Balance Sheet Effects of the Mexican Crisis of 1994,” Journal of Development Economics, 2004, 2, pp. 535-563.
- Rajan, R. and L. Zingales, 1998, “Financial dependence and growth,” American Economic Review, Vol. 88, pp. 559–586.
- Rauh, J.D., 2006, “Investment and financing constraints: Evidence from the funding of corporate pension plans,” The Journal of Finance, Vol. 61(1), pp.33-71.
- Reinhart, R. and K. Rogoff, 2014, “This time is different: A panoramic view of eight centuries of financial crises,” Annals of Economics and Finance Society for AEF, Vol. 15(2), pp. 1065–1188.
- Rodrik, D., 2008, “The real exchange rate and economic growth,” Brookings Papers on Economic Activity, Fall 2008, pp. 365–412.
- Tirole, J., 2006, The Theory of Corporate Finance, Princeton University Press.

### Appendix A. Proof of Proposition
- Optimal capital in second period given by Equation 2: K1 = f(RER, Π1) (presented as 1 ( , ( )) K f RER RER Π in source).
- Change in RER affects K through:
  - direct effect (shifting the F’(.) schedule), and
  - indirect effect via RER’s effect on Π1.
- Total derivative expression (as given): 1 1 dK dRER KKK dRER RER RER Π   Π 
- Implicit function theorem applied to Equation 2 yields (source presentation):
  - ( ) [ ] 1 ' ( , ) (1) '( ) / 1 (1 ) '' ( ) '' ( ) '' ( ) '' ( ) 1 C K f K F K RER K RER C K F K C K F K RER φ α γ α − Π  − −     −  − −
- Normalizations used:
  - Normalize the RER prior to perturbation to 1.
  - Scale the consumption wage w so profits prior to the RER change equal 1 by setting 1 (1 ) w α α αα − = − .
- Combining with sensitivities (Equation 3) and profits’ RER sensitivity:
  - ( ) ( ) 1 ln 1 ln 1 d d RER α γ α Π = − −
- Derived condition (as presented):
  - [ ] , 1 1 1 (1 ) ln 1 ' ( , ) (1) '' ( , ) 0 ln '' ( , ) '' ( ) K RER dK C K f K C K f dRER C K f F K ε φ α γ α − − = = −Π + − + −Π > −Π − −
- Immediate conclusion:
  - , 0. K RER d d ε α >
- Further result when '( ) 0 F K RER  =  (marginal return constant):
  - [ ] , 2 1 '' ( ) '' ( . ) (1 ) 0 . (1) '' ( , ) '' ( ) K RER d F KC dff C KfF K ε α γ α −  = − > −  −Π − −
- Empirical note:
  - Second prediction tested by controlling for expected return to investment, or Tobin’s q, in the investment equation.

### Appendix B. Figures and Tables

#### Figures: Correlations and Firm-Level Responses
- Figure 4: Correlation between the median firm-specific labor share (expenses related to salaries and benefits scaled by total sales) and the US labor share at the 3-digit (NAICS) industry level. Firm-specific labor share available for one third of firm-year observations. Data sources: Elsby, Hobijn, and Sahin (2013), Worldscope.
- Figure 5: Firm Growth and the Real Exchange Rate
  - Panel A: Firms in advanced economies (about 25,000 tradable sector firms from 66 countries during 2000-2011; data: Worldscope, IFS).
  - Panel B: Firms in emerging market countries (same sample definition).

#### Sample Composition (Table 1)
- Sample based on firms with non-missing total asset information. Data sources: Worldscope, WEO, IFS.
- Country-level snapshot (selected entries):
  - Argentina: 213 firms; # traded sector firms 112; % traded sector firms 83%
  - China: 9,241 firms; # traded sector firms 2,404; % traded sector firms 81%
  - India: 534 firms; # traded sector firms 11,171; % traded sector firms 85% (note: as printed)
  - United States: 11,106 firms; # traded sector firms 9,359; % traded sector firms 84%
- Total: 31,190 firms; 25,416 tradable sector firms; 81% traded sector firms.

#### Descriptive Statistics (Table 2)
- Firm-level variables (2000-2011), winsorized:
  - Assets t (USD million): Obs. 132,200; Mean 1,260.06; St. Dev. 9,631.95; Min 0.00; p25 30.85; Median 114.37; p75 415.55; Max 797,800.00
  - Capital Stock: K t (USD million): Obs. 121,205; Mean 737.02; St. Dev. 5,345.27; Min 0.00; p25 9.33; Median 47.35; p75 213.66; Max 374,094.22
  - Employment t: Obs. 121,833; Mean 3,923; St. Dev. 18,610; Min 1; p25 127; Median 515; p75 2,016; Max 2,333,000
  - Asset Growth t: Obs. 132,200; Mean 0.144; St. Dev. 0.384; Min -0.390; p25 -0.058; Median 0.072; p75 0.226; Max 1.513
  - Investment Rate (Capex t/K t-1): Obs. 117,722; Mean 0.130; St. Dev. 0.156; Min 0.000; p25 0.035; Median 0.077; p75 0.161; Max 1.000
  - Tobin's Q t (market value+total debt/assets): Obs. 124,471; Mean 1.713; St. Dev. 2.278; Min 0.000; p25 0.653; Median 0.998; p75 1.768; Max 16.414
- Country-level and industry-level variables:
  - Log(Real exchange rate) t: Obs. 665; Mean 0.311; St. Dev. 0.451; Min -0.664; p25 -0.032; Median 0.291; p75 0.575; Max 1.612
  - Credit t/GDP t-1: Obs. 661; Mean 0.882; St. Dev. 0.564; Min 0.088; p25 0.394; Median 0.815; p75 1.216; Max 3.195
  - Labor share (industry-level): Obs. 608; Mean 0.637; St. Dev. 0.199; Min 0.052; p25 0.556; Median 0.647; p75 0.748; Max 0.010 (note: as printed)

#### Firm Growth and Real Exchange Rate (Table 3)
- Median firm growth rates during 2000-2011 split by country-year RER movement:
  - All firms:
    - Appreciation: Total Assets growth 4.7%; Depreciation: 10.3%; Difference (2)-(1) 5.6%; p-value test (2)>(1) 0.000
    - Appreciation: Capital Expenditure 2.0%; Depreciation: 8.0%; Difference 6.0%; p-value 0.000
    - Appreciation: Market Capitalization -0.5%; Depreciation: 5.2%; Difference 5.7%; p-value 0.000
  - Firms in advanced economies:
    - Total Assets: Appreciation 3.7%; Depreciation 9.2%; Difference 5.5%; p-value 0.000
    - Capital Expenditure: Appreciation 1.4%; Depreciation 6.5%; Difference 5.1%; p-value 0.000
    - Market Capitalization: Appreciation 0.3%; Depreciation 4.5%; Difference 4.3%; p-value 0.000
  - Firms in emerging market countries:
    - Total Assets: Appreciation 8.4%; Depreciation 13.8%; Difference 5.5%; p-value 0.000
    - Capital Expenditure: Appreciation 4.0%; Depreciation 14.3%; Difference 10.3%; p-value 0.000
    - Market Capitalization: Appreciation -3.6%; Depreciation 7.3%; Difference 10.9%; p-value 0.000

#### Baseline Results — Tradable Sector Firms (Table 4)
- Dependent variables: total asset growth (cols 1-2), log-investment ratio (cols 3-6), and cash flow (cols 7-8). Sample: tradable sector firms. Interactions with country-time fixed effects absorbed.
- Selected coefficient estimates (standard errors in parentheses; significance stars preserved in source):
  - Log(RER) t * Labor share:
    - 0.413*** (cols 1-2), 0.960** (col 3), 0.880* (col 4), 0.935** (col 5), 0.896* (col 6), 0.107*** (col 7), 0.082** (col 8).
  - Log(RER) t * Labor share * (Credit/GDP) t-1:
    - 0.290*** (cols 1-2), 0.766** (col 3), 0.731** (col 4), 0.713** (col 5), 0.718** (col 6).
  - Labor share * (Credit/GDP) t-1: coefficients include -0.062 (cols 1-2) and other negative estimates in other columns.
- Observations and fit:
  - Observations: examples include 132,200; 93,271; 95,988; 137,484.
  - R-squared values: examples 0.369, 0.371, 0.628, 0.631, 0.624, 0.627, 0.719, 0.720.
- Fixed effects: Country-year FE, Firm FE, Year FE always included; Industry-year FE included in some columns.

#### Placebo — Non-Tradable Sector Firms (Table 5)
- Dependent variables: log-investment ratio (cols 1-4) and cash flow (cols 5-6). Sample: non-tradable sector firms.
- Key estimates:
  - Log(RER) t * Labor share: coefficients generally negative and imprecise (e.g., -1.812, -1.833, -1.633, -1.625); cash flow coefficients small (-0.063, -0.087).
  - Log(RER) t * Labor share * (Credit/GDP) t-1: -1.777** (col 1), -1.197 (col 2), -1.759** (col 3), -1.113 (col 4).
- Observations: examples 24,631; 25,736; 33,515. R-squared examples 0.651, 0.653, 0.646, 0.648, 0.782.

#### Firm Value and Stock Returns (Table 6)
- Dependent variables: net sales growth (cols 1-2), total market capitalization growth (cols 3-5), and annual stock market return (cols 6-8). Sample: tradable sector firms.
- Selected coefficients:
  - Log(RER) t * Labor share: 0.324*** (col 1), 0.228** (col 2), 0.181 (col 3), 0.432** (col 4), 0.908** (col 5), 0.069 (col 6), 0.593*** (col 7), 0.823*** (col 8).
  - Log(RER) t * Labor share * (Credit/GDP) t-1: 0.223*** (col 1), 0.129** (col 2), 0.236*** (col 3), 0.244* (col 4), 0.429* (col 5), 0.224* (col 6), 0.563*** (col 7), 0.660*** (col 8).
  - Labor share * (Credit/GDP) t-1: e.g., -0.266*** (col 7), -0.200** (col 8).
- Observations: e.g., 126,279; 105,727; 123,893; 97,197; 75,256; 123,632; 106,019; 80,707.
- R-squared range: 0.396 to 0.506.

#### Alternative Channels — Aggregate Lending and Savings (Table 7)
- Dependent variables: total asset growth (cols 1-2) and log-investment ratio (cols 3-6). Sample: tradable sector firms.
- Key interaction estimates:
  - Log(RER) t * Labor share: 0.374*** (col 1), 0.388*** (col 2), 0.995** (col 3), 0.966* (col 4), 0.946** (col 5), 0.967** (col 6).
  - Log(RER) t * Labor share * (Credit/GDP) t-1: 0.301*** (col 1), 0.312*** (col 2), 0.874** (col 3), 0.855** (col 4), 0.812** (col 5), 0.844** (col 6).
  - Log(RER) t * Ext. Finance Dependence * (Credit/GDP) t-1: negative coefficients in some specifications (e.g., -0.359*; -0.389** in cols 5-6).
- Observations: e.g., 121,375; 86,681; 89,152. R-squared examples 0.365, 0.368, 0.627, 0.630.

#### Alternative Channels — Firms’ Balance Sheet (Table 8)
- Dependent variables: total asset growth and log-investment ratio. Sample: tradable sector firms.
- Controls include FX share (country-level) and firm net worth (common equity/total assets).
- Selected coefficients:
  - Net worth t-1: 0.008*** (cols shown).
  - Log(RER) t * Labor share: 0.414***, 0.438*** in baseline columns.
  - Log(RER) t * Labor share * (Credit/GDP) t-1: 0.291***, 0.302***, and larger (e.g., 1.200** in one subsample).
  - Log(RER) t * Labor share * FX share: coefficients vary by subsample (-0.003, -0.041, 1.321*, 1.256, -5.737**).
  - Log(RER) t * (Net worth) t-1: 0.004***.
- Observations: e.g., 132,200; 93,271; 15,611 for FX>50% subsample. R-squared range 0.369 to 0.672.

#### Robustness — Excluding Currency and Banking Crises (Table 9)
- Dependent variables: total asset growth and log-investment ratio. Currency crisis dates from Laeven and Valencia (2013); Banking crisis dates from Reinhart and Rogoff (2014).
- Key interaction estimates remain positive and significant:
  - Log(RER) t * Labor share: 0.415***; 0.404***; 0.386*** across robustness samples.
  - Log(RER) t * Labor share * (Credit/GDP) t-1: 0.295***; 0.285***; 0.234*** across robustness samples.
- Observations: e.g., 132,033; 131,880; 114,485 after dropping crisis years. R-squared examples 0.372, 0.631, 0.410, 0.646.

#### Robustness — Tobin’s Q Variations (Table 10)
- Dependent variable: log-investment ratio. Tobin’s Q measured as lagged, average over (t-1,t), and market/book.
- Tobin’s Q coefficients: 0.050***, 0.049***, 0.061***, 0.059***, 0.049***, 0.047*** across specifications.
- Interaction terms retain positive coefficients:
  - Log(RER) t * Labor share: 0.859*; 0.988**; 0.976**.
  - Log(RER) t * Labor share * (Credit/GDP) t-1: 0.687**; 0.758**; 0.760**.

#### Robustness — Measures of Financial Constraints (Table 11)
- Dependent variable: log-investment ratio.
- Financial constraints proxied by log(Credit/GDP) t-1, firm log-employment t-1, and log-assets t-1.
- Selected results:
  - Log(RER) t * Labor share: large and sometimes imprecise coefficients (e.g., 2.644**, 2.860**).
  - Log(RER) t * Labor share * Financial Constraints: positive and significant in many specifications (e.g., 0.569**, 0.607**, 0.078***, 0.062**, 0.056**, 0.060**).
  - Labor share * Financial Constraints: examples -0.449*** and -0.370*** when financial constraint = log-assets.

#### Robustness — Country-Industry-Year Fixed Effects (Table 12)
- Dependent variables as in baseline regressions. Double interaction terms absorbed by country-industry-time fixed effects.
- Selected coefficients:
  - Log(RER) t * Labor share: 0.491*** (asset growth), 1.369** (investment), 1.256** (investment), 0.133*** (cash flow).
  - Log(RER) t * Labor share * (Credit/GDP) t-1: 0.291***, 1.018**, 0.927** in various specifications.
  - Labor share * (Credit/GDP) t-1: -0.163*** in one specification.

#### Robustness — Four-Year Averages Panel (Table 13)
- Regressions run on panel of four-year averages over 2000-2003, 2004-2007, 2008-2011. Firm-level variables at beginning of period; Tobin’s Q contemporaneous (period average).
- Selected coefficients:
  - Log(RER) t * Labor share: 0.309***, 0.336***, and larger coefficients in four-year aggregations (e.g., 1.794**, 1.665*, 1.854**, 1.726*, 1.227***, 0.948*** across different dependent variables).
  - Log(RER) t * Labor share * (Credit/GDP) t-1: 0.393***, 0.420***, 1.743***, 2.014***, 1.687***, 1.973***.
- Observations: e.g., 46,726; 24,491; 24,796; 34,672. R-squared values in averaged panels: 0.728 to 0.897.

*IMF Working Paper wp17183 - Appendix B. Figures and Tables (content excerpt).*

### References

### wp17183 - References

### References
- Aguiar, M., 2005, “Investment, devaluation, and foreign currency exposure: The case of Mexico,” Journal of Development Economics, Vol. 78(1), pp.95-113.
- Beck, T., Demirguc‐Kunt, A., Laeven, L. and Levine, R., 2008, “Finance, firm size, and growth,” Journal of Money, Credit and Banking, Vol. 40(7), pp.1379-1405.
- Berg, A., Ostry, J.D. and Zettelmeyer, J., 2012. “What makes growth sustained?” Journal of Development Economics, 98(2), pp.149-166.
- Bleakley, H. and Cowan, K., 2008, “Corporate dollar debt and depreciations: much ado about nothing?” The Review of Economics and Statistics, Vol. 90(4), pp. 612-626.
- Bleakley, H. and Cowan, K., 2010, “Maturity mismatch and financial crises: Evidence from emerging market corporations,” Journal of Development Economics, Vol. 93(2), pp.189-205.
- Chen, H. J. and Chen, S. J., 2012, “Investment-cash flow sensitivity cannot be a good measure of financial constraints: Evidence from the time series,” Journal of Financial Economics, Vol. 103(2), pp.393-410.
- Chen, R. and M. Dao, 2011, “The real exchange rate and employment in China,” IMF Working Paper No. 11/148, Washington, DC: International Monetary Fund.
- Claessens, C., Tong, H., and S.-J. Wei, 2012, “From the financial crisis to the real economy: Using firm-level data to identify transmission channels,” Journal of International Economics, Vol. 88, pp. 375–387.
- Desai, M. A., Foley, C. F., and K. J. Forbes, 2008, “Financial constraints and growth: Multinational and local firm responses to currency depreciations,” Review of Financial Studies, Vol. 21, pp. 2857–2888.
- Di Giovanni, J. and A. A. Levchenko, 2010, “Putting the parts together: Trade, vertical linkages, and business cycle comovement,” American Economic Journal: Macroeconomics, Vol. 2, pp. 95–124.
- Elsby, Michael W. L., Hobijn, B., and A. Sahin, 2013, “The decline of the US labor share,” Brookings Papers on Economic Activity, Fall 2013, pp. 1–52.
- Fazzari, S. M., Hubbard, R.G., and B. Petersen, 2000, “Investment-cash flow sensitivities are useful: A comment on Kaplan and Zingales,” Quarterly Journal of Economics, Vol. 115(2), pp. 695–705.
- Forbes, K., 2002, “How do large depreciations affect firm performance?” IMF Staff Papers, Vol. 49, pp. 214–238.
- Forbes, K., 2002, “Cheap labor meets costly capital: The impact of devaluations on commodity firms,” Journal of Development Economics, Vol. 69, pp. 335–365.
- Gilchrist, S. and Himmelberg, C.P., 1995, “Evidence on the role of cash flow for investment,” Journal of Monetary Economics, Vol. 36(3), pp.541-572.
- Hsieh, C.-T. and P. J. Klenow, 2009, “Misallocation and manufacturing TFP in China and India,” Quarterly Journal of Economics, Vol. CXXIV, Issue 4, pp. 1403–1448.
- Kalemli-Ozcan, S., Kamil, H., and Villegas-Sanchez, C., 2015, “What Hinders Investment in the Aftermath of Financial Crises? Insolvent Firms or Illiquid Banks?,” NBER Working Paper 16528.
- Kaplan, S. N. and L. Zingales, 1997, “Do investment-cash flow sensitivities provide useful measures of financing constraints?” Quarterly Journal of Economics, Feb. 1997, pp. 170–215.
- Kaminsky, G.L. and Schmukler, S.L., 2008, “Short-run pain, long-run gain: Financial liberalization and stock market cycles,” Review of Finance, 12(2), pp.253-292.
- Laeven, L. and F. Valencia, 2013, “Systemic banking crises database,” IMF Economic Review, Vol. 61, pp. 225–270.
- Lewellen, J. and K. Lewellen, forthcoming, “Investment and cash flow: New evidence,” Journal of Financial and Quantitative Analysis.
- Levy-Yeyati, E., Sturzenegger, F., and P. A. Gluzmann, 2013, “Fear of appreciation,” Journal of Development Economics, Vol. 101, pp. 233–247.
- Levy-Yeyati, E., Sturzenegger, F., and P. A. Gluzmann, 2012, “Exchange rate undervaluation and economic growth: Diaz Alejandro (1965) revisited,” Economics Letters, Vol. 117, pp. 666–672.
- Pagano, M. and Pica, G., 2012, “Finance and employment,” Economic Policy, Vol. 27, Issue 69, pp.5-55.
- Pratab, S. and C. Urrutia, 2004, “Firm Dynamics, Investment & Debt Portfolios: Balance Sheet Effects of the Mexican Crisis of 1994,” Journal of Development Economics, 2004, 2, pp. 535-563.
- Rajan, R. and L. Zingales, 1998, “Financial dependence and growth,” American Economic Review, Vol. 88, pp. 559–586.
- Rauh, J.D., 2006, “Investment and financing constraints: Evidence from the funding of corporate pension plans,” The Journal of Finance, Vol. 61(1), pp.33-71.
- Reinhart, R. and K. Rogoff, 2014, “This time is different: A panoramic view of eight centuries of financial crises,” Annals of Economics and Finance Society for AEF, Vol. 15(2), pp. 1065–1188.
- Rodrik, D., 2008, “The real exchange rate and economic growth,” Brookings Papers on Economic Activity, Fall 2008, pp. 365–412.
- Tirole, J., 2006, The Theory of Corporate Finance, Princeton University Press.

### Appendix A. Proof of Proposition
- The optimal capital level in the second period, according to Equation 2 in the text, is a function of the RER and first-period profits:
  - K1 = f(RER, Π1)  (presented in text as 1 ( , ( )) K f RER RER Π)
- A change in RER affects K through:
  - direct effect (shifting the F’(.) schedule) and
  - indirect effect via RER’s effect on Π1.
- Total derivative expression (as given):
  - 1 1 dK dRER KKK dRER RER RER Π   Π 
- Implicit function theorem applied to Equation 2 yields the expression (as presented in the source):
  - ( ) [ ] 1 ' ( , ) (1) '( ) / 1 (1 ) '' ( ) '' ( ) '' ( ) '' ( ) 1 C K f K F K RER K RER C K F K C K F K RER φ α γ α − Π  − −     −  − −
- Normalizations used in the derivation:
  - Normalize the RER prior to perturbation to 1.
  - Scale the consumption wage w such that profits prior to the RER change are equal 1, by setting 1 (1 ) w α α αα − = − .
- Combining with sensitivities (Equation 3 in text) and of profits with respect to RER:
  - ( ) ( ) 1 ln 1 ln 1 d d RER α γ α Π = − −
- Derived condition (as presented):
  - [ ] , 1 1 1 (1 ) ln 1 ' ( , ) (1) '' ( , ) 0 ln '' ( , ) '' ( ) K RER dK C K f K C K f dRER C K f F K ε φ α γ α − − = = −Π + − + −Π > −Π − −
- Immediate conclusion:
  - , 0. K RER d d ε α > 
- Further result when marginal return to investment is constant, i.e., '( ) 0 F K RER  =  :
  - [ ] , 2 1 '' ( ) '' ( . ) (1 ) 0 . (1) '' ( , ) '' ( ) K RER d F KC dff C KfF K ε α γ α −  = − > −  −Π − −
- Empirical note from the source:
  - In the data, the second prediction is tested by controlling for expected return to investment, or Tobin’s q, in the investment equation.

*IMF Working Paper wp17183 - References (content excerpt)*

### Appendix B. Figures and Tables

### Appendix B. Figures and Tables

### Figures: Correlations and Firm-Level Responses
- Figure 4: Correlation between the median firm-specific labor share (expenses related to salaries and benefits scaled by total sales) and the US labor share at the 3-digit (NAICS) industry level. Firm-specific labor share available for one third of firm-year observations. Data sources: Elsby, Hobijn, and Sahin (2013), Worldscope.
- Figure 5: Firm Growth and the Real Exchange Rate
  - Panel A: Firms in advanced economies (about 25,000 tradable sector firms from 66 countries during 2000-2011; data: Worldscope, IFS).
  - Panel B: Firms in emerging market countries (same sample definition).

### Sample Composition (Table 1)
- Sample based on firms with non-missing total asset information. Data sources: Worldscope, WEO, IFS.
- Country-level snapshot (selected entries):
  - Argentina: 213 firms; # traded sector firms 112; % traded sector firms 83%
  - China: 9,241 firms; # traded sector firms 2,404; % traded sector firms 81%
  - India: 534 firms; # traded sector firms 11,171; % traded sector firms 85% (note: as printed)
  - United States: 11,106 firms; # traded sector firms 9,359; % traded sector firms 84%
- Total: 31,190 firms; 25,416 tradable sector firms; 81% traded sector firms.

### Descriptive Statistics (Table 2)
- Firm-level variables (2000-2011), winsorized:
  - Assets t (USD million): Obs. 132,200; Mean 1,260.06; St. Dev. 9,631.95; Min 0.00; p25 30.85; Median 114.37; p75 415.55; Max 797,800.00
  - Capital Stock: K t (USD million): Obs. 121,205; Mean 737.02; St. Dev. 5,345.27; Min 0.00; p25 9.33; Median 47.35; p75 213.66; Max 374,094.22
  - Employment t: Obs. 121,833; Mean 3,923; St. Dev. 18,610; Min 1; p25 127; Median 515; p75 2,016; Max 2,333,000
  - Asset Growth t: Obs. 132,200; Mean 0.144; St. Dev. 0.384; Min -0.390; p25 -0.058; Median 0.072; p75 0.226; Max 1.513
  - Investment Rate (Capex t/K t-1): Obs. 117,722; Mean 0.130; St. Dev. 0.156; Min 0.000; p25 0.035; Median 0.077; p75 0.161; Max 1.000
  - Tobin's Q t (market value+total debt/assets): Obs. 124,471; Mean 1.713; St. Dev. 2.278; Min 0.000; p25 0.653; Median 0.998; p75 1.768; Max 16.414
- Country-level and industry-level variables:
  - Log(Real exchange rate) t: Obs. 665; Mean 0.311; St. Dev. 0.451; Min -0.664; p25 -0.032; Median 0.291; p75 0.575; Max 1.612
  - Credit t/GDP t-1: Obs. 661; Mean 0.882; St. Dev. 0.564; Min 0.088; p25 0.394; Median 0.815; p75 1.216; Max 3.195
  - Labor share (industry-level): Obs. 608; Mean 0.637; St. Dev. 0.199; Min 0.052; p25 0.556; Median 0.647; p75 0.748; Max 0.010 (note: as printed)

### Firm Growth and Real Exchange Rate (Table 3)
- Median firm growth rates during 2000-2011 split by country-year RER movement:
  - All firms:
    - Appreciation: Total Assets growth 4.7%; Depreciation: 10.3%; Difference (2)-(1) 5.6%; p-value test (2)>(1) 0.000
    - Appreciation: Capital Expenditure 2.0%; Depreciation: 8.0%; Difference 6.0%; p-value 0.000
    - Appreciation: Market Capitalization -0.5%; Depreciation: 5.2%; Difference 5.7%; p-value 0.000
  - Firms in advanced economies:
    - Total Assets: Appreciation 3.7%; Depreciation 9.2%; Difference 5.5%; p-value 0.000
    - Capital Expenditure: Appreciation 1.4%; Depreciation 6.5%; Difference 5.1%; p-value 0.000
    - Market Capitalization: Appreciation 0.3%; Depreciation 4.5%; Difference 4.3%; p-value 0.000
  - Firms in emerging market countries:
    - Total Assets: Appreciation 8.4%; Depreciation 13.8%; Difference 5.5%; p-value 0.000
    - Capital Expenditure: Appreciation 4.0%; Depreciation 14.3%; Difference 10.3%; p-value 0.000
    - Market Capitalization: Appreciation -3.6%; Depreciation 7.3%; Difference 10.9%; p-value 0.000

### Baseline Results — Tradable Sector Firms (Table 4)
- Dependent variables: total asset growth (cols 1-2), log-investment ratio (cols 3-6), and cash flow (cols 7-8). Sample: tradable sector firms. Interactions with country-time fixed effects absorbed.
- Selected coefficient estimates (standard errors in parentheses; significance stars preserved in source):
  - Log(RER) t * Labor share:
    - Columns reporting positive, significant coefficients include 0.413*** (cols 1-2), 0.960** (col 3), 0.880* (col 4), 0.935** (col 5), 0.896* (col 6), 0.107*** (col 7), 0.082** (col 8).
  - Log(RER) t * Labor share * (Credit/GDP) t-1:
    - 0.290*** (cols 1-2), 0.766** (col 3), 0.731** (col 4), 0.713** (col 5), 0.718** (col 6).
  - Labor share * (Credit/GDP) t-1: coefficients range including -0.062 (cols 1-2) and larger negative estimates in other columns (not statistically significant in many specifications).
- Observations and fit:
  - Observations range across specifications (e.g., 132,200; 93,271; 95,988; 137,484).
  - R-squared values: examples 0.369, 0.371, 0.628, 0.631, 0.624, 0.627, 0.719, 0.720.
- Fixed effects: Country-year FE, Firm FE, Year FE always included; Industry-year FE included in some columns.

### Placebo — Non-Tradable Sector Firms (Table 5)
- Dependent variables: log-investment ratio (cols 1-4) and cash flow (cols 5-6). Sample: non-tradable sector firms.
- Key estimates:
  - Log(RER) t * Labor share: coefficients generally negative and imprecise (e.g., -1.812, -1.833, -1.633, -1.625) and small coefficients for cash flow (-0.063, -0.087).
  - Log(RER) t * Labor share * (Credit/GDP) t-1: -1.777** (col 1), -1.197 (col 2), -1.759** (col 3), -1.113 (col 4).
- Observations: example counts 24,631; 25,736; 33,515. R-squared examples 0.651, 0.653, 0.646, 0.648, 0.782.

### Firm Value and Stock Returns (Table 6)
- Dependent variables: net sales growth (cols 1-2), total market capitalization growth (cols 3-5), and annual stock market return (cols 6-8). Sample: tradable sector firms.
- Selected coefficients:
  - Log(RER) t * Labor share: e.g., 0.324*** (col 1), 0.228** (col 2), 0.181 (col 3), 0.432** (col 4), 0.908** (col 5), 0.069 (col 6), 0.593*** (col 7), 0.823*** (col 8).
  - Log(RER) t * Labor share * (Credit/GDP) t-1: e.g., 0.223*** (col 1), 0.129** (col 2), 0.236*** (col 3), 0.244* (col 4), 0.429* (col 5), 0.224* (col 6), 0.563*** (col 7), 0.660*** (col 8).
  - Labor share * (Credit/GDP) t-1: mixed signs and significance, e.g., -0.266*** (col 7), -0.200** (col 8).
- Observations: e.g., 126,279; 105,727; 123,893; 97,197; 75,256; 123,632; 106,019; 80,707.
- R-squared range: 0.396 to 0.506.

### Alternative Channels — Aggregate Lending and Savings (Table 7)
- Dependent variables: total asset growth (cols 1-2) and log-investment ratio (cols 3-6). Sample: tradable sector firms.
- Key interaction estimates:
  - Log(RER) t * Labor share: 0.374*** (col 1), 0.388*** (col 2), 0.995** (col 3), 0.966* (col 4), 0.946** (col 5), 0.967** (col 6).
  - Log(RER) t * Labor share * (Credit/GDP) t-1: 0.301*** (col 1), 0.312*** (col 2), 0.874** (col 3), 0.855** (col 4), 0.812** (col 5), 0.844** (col 6).
  - Log(RER) t * Ext. Finance Dependence * (Credit/GDP) t-1: negative coefficients in some specifications (e.g., -0.359*; -0.389** in cols 5-6).
- Observations: e.g., 121,375; 86,681; 89,152. R-squared examples 0.365, 0.368, 0.627, 0.630.

### Alternative Channels — Firms’ Balance Sheet (Table 8)
- Dependent variables: total asset growth and log-investment ratio. Sample: tradable sector firms.
- Controls include FX share (country-level) and firm net worth (common equity/total assets).
- Selected coefficients:
  - Net worth t-1: 0.008*** (cols shown).
  - Log(RER) t * Labor share: 0.414***, 0.438*** in baseline columns; other columns report smaller or imprecise coefficients in subsamples.
  - Log(RER) t * Labor share * (Credit/GDP) t-1: 0.291***, 0.302***, and larger (e.g., 1.200** in one subsample).
  - Log(RER) t * Labor share * FX share: coefficients include -0.003, -0.041, 1.321*, 1.256, -5.737** (varying by subsample).
  - Log(RER) t * (Net worth) t-1: 0.004***.
- Observations: e.g., 132,200; 93,271; 15,611 for FX>50% subsample. R-squared range 0.369 to 0.672.

### Robustness — Excluding Currency and Banking Crises (Table 9)
- Dependent variables: total asset growth and log-investment ratio. Currency crisis dates from Laeven and Valencia (2013); Banking crisis dates from Reinhart and Rogoff (2014).
- Key interaction estimates remain positive and significant:
  - Log(RER) t * Labor share: e.g., 0.415***; 0.404***; 0.386*** across robustness samples.
  - Log(RER) t * Labor share * (Credit/GDP) t-1: 0.295***; 0.285***; 0.234*** across robustness samples (some columns report additional significance for interaction with investment).
- Observations: e.g., 132,033; 131,880; 114,485 after dropping crisis years. R-squared examples 0.372, 0.631, 0.410, 0.646.

### Robustness — Tobin’s Q Variations (Table 10)
- Dependent variable: log-investment ratio. Tobin’s Q measured as lagged, average over (t-1,t), and market/book.
- Tobin’s Q coefficients: 0.050***, 0.049***, 0.061***, 0.059***, 0.049***, 0.047*** across specifications.
- Interaction terms retain positive coefficients:
  - Log(RER) t * Labor share: e.g., 0.859*; 0.988**; 0.976**.
  - Log(RER) t * Labor share * (Credit/GDP) t-1: e.g., 0.687**; 0.758**; 0.760**.

### Robustness — Measures of Financial Constraints (Table 11)
- Dependent variable: log-investment ratio.
- Financial constraints measured as log(Credit/GDP) t-1, firm log-employment t-1, and log-assets t-1.
- Selected results:
  - Log(RER) t * Labor share: large and sometimes imprecise coefficients (e.g., 2.644**, 2.860**, smaller coefficients when using firm-size constraints).
  - Log(RER) t * Labor share * Financial Constraints: positive and significant in many specifications (e.g., 0.569**, 0.607**, 0.078***, 0.062**, 0.056**, 0.060**).
  - Labor share * Financial Constraints: mixed signs including -0.449*** and -0.370*** when financial constraint = log-assets.

### Robustness — Country-Industry-Year Fixed Effects (Table 12)
- Dependent variables as in baseline regressions. Double interaction terms absorbed by country-industry-time fixed effects.
- Selected coefficients:
  - Log(RER) t * Labor share: 0.491*** (asset growth), 1.369** (investment), 1.256** (investment), 0.133*** (cash flow).
  - Log(RER) t * Labor share * (Credit/GDP) t-1: 0.291***, 1.018**, 0.927** in various specifications.
  - Labor share * (Credit/GDP) t-1: -0.163*** in one specification.

### Robustness — Four-Year Averages Panel (Table 13)
- Regressions run on panel of four-year averages over 2000-2003, 2004-2007, 2008-2011. Firm-level variables at beginning of period; Tobin’s Q contemporaneous (period average).
- Selected coefficients:
  - Log(RER) t * Labor share: 0.309***, 0.336***, and larger coefficients in four-year aggregations (e.g., 1.794**, 1.665*, 1.854**, 1.726*, 1.227***, 0.948*** across different dependent variables).
  - Log(RER) t * Labor share * (Credit/GDP) t-1: 0.393***, 0.420***, 1.743***, 2.014***, 1.687***, 1.973***.
- Observations: e.g., 46,726; 24,491; 24,796; 34,672. R-squared values high in averaged panels (e.g., 0.728 to 0.897).

*Source: IMF Working Paper — Appendix B. Figures and Tables (tables and figures as provided in the source PDF).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2017/wp17183.pdf_