## wp18137 - 2016. In advanced economies, markups have increased by an average of 39 percent since 1980.

## Source details

**Canonical URL:** [wp18137 - 2016. In advanced economies, markups have increased by an average of 39 percent since 1980.](https://www.imf.org/-/media/files/publications/wp/2018/wp18137.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2018/wp18137.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2018/wp18137.pdf.json)

---

### Key findings
- Markups of publicly traded firms have generally increased during 1980-2016, especially in advanced economies (AEs).
- For AEs, markups have increased by a GDP-weighted average of 39 percent since 1980.
- Increase is broad-based across industries and countries and driven by the highest-markup firms in each economic sector; the distribution of markups has widened.
- For emerging markets and developing economies (EMDEs), there is less evidence of a rise in markups.
- Positive relation between firm markups and other indicators of market power, such as profits and industry concentration.
- Firm-level association between markups and the labor share is generally negative.

### Data, scope, and methodology
- Sample coverage:
  - 74 economies total: 33 AEs and 41 EMDEs.
  - Period: 1980-2016 (37 years).
  - More than 631,000 estimates of firm markups obtained.
- Data sources:
  - Thomson Reuters Worldscope for international publicly traded firms.
  - Compustat approach of De Loecker and Eeckhout (2017) extended internationally.
- Measurement:
  - Markups computed as the ratio of firm sales to the cost of variable inputs, scaled by the output elasticity of variable inputs.
  - Baseline variable input measure: cost of goods sold (COGS).
  - Robustness: findings hold when controlling for additional operational costs such as selling, general, and administrative expenses (SGA).
- Coverage of economic significance:
  - For 2016, U.S. firms in the sample have sales equivalent to 79 percent of U.S. GDP.
  - For the other 73 economies in the sample, firms have sales equivalent, on average, to 75 percent of their respective economy’s GDP.

### Evolution of markups across countries and industries
- Increase concentrated in AEs; less evidence of rise in EMDEs.
- Rise driven by firms able to extract especially high markups within sectors.
- Widening distribution of firm markups between 1980 and 2016.
- Robustness: evolution estimates across countries are robust to alternative estimation approaches.

### Relation between markups, investment, and innovation (overview)
- Section III investigates firm-level markups and firm performance: capital expenditure (investment), R&D spending (innovation), and the labor share.
- Sample focus: United States (most complete data) and extension to 32 other AEs.

### Investment — specification and main findings
- Dependent variable: investment rate = capital expenditure as a share of the previous year’s capital stock (CapExp / lagged PPE).
- Main explanatory variable: log of firm-level markup (ln(markup)), with firm and time fixed effects and controls including Tobin’s Q, sales rate, and lagged R&D rate.
- Nonlinear specification: interaction terms included for ln(markup) × ln(markup) and ln(markup) × Concentration (adjusted HHI at 115 ICB sub-sector level).
- Main empirical patterns:
  - Estimated inverted-U (non-monotonic) relation between markups and investment: higher markups initially associated with increasing investment; at higher markup levels, increases in markups associated with lower investment.
  - Industry heterogeneity: inverted-U not driven by any single industry; coefficients for quadratic markup interacted with industry dummies are negative for all 10 ICB industries.
  - Concentration effect: as market concentration rises, higher markups are more likely to be associated with lower investment. For concentration index values above 0.5, higher markups are associated with lower investment.
  - Sub-sector market concentration index in sample ranges from 0.1 to 1.0.
- Macroeconomic significance (U.S. firms):
  - Computed share of U.S. publicly listed firms for which rising markups are associated with lower investment in 2016:
    - 8 percent, 17 percent, and 6 percent across the three specifications reported in Table 2 (columns 1–3), respectively.
  - In contrast, no firms had a negative association between higher markups and investment as of 1980.
- Key coefficients (Table 2, U.S. data):
  - Markup: 0.061*** (0.007); 0.072*** (0.012); 0.107*** (0.012)
  - Markup × markup: -0.045*** (0.005); -0.043*** (0.005)
  - Markup × concentration: -0.152*** (0.031); -0.131*** (0.031)
  - Number of observations: 57,371; 52,319; 52,319
  - R²: 0.228; 0.221; 0.22

### Innovation (R&D) — specification and main findings
- Dependent variable: R&D expenditure as a share of the previous year’s total assets (RD / lagged TA).
- Controls: Tobin’s Q and level of sales as a share of previous year’s total assets, plus firm and time fixed effects.
- Main empirical patterns:
  - Non-monotonic (inverted-U) relation between markups and R&D, similar to physical investment: higher markups initially associated with increasing R&D; at higher markups or higher concentration levels, the marginal relation becomes negative.
  - Results consistent with Aghion et al. (2005): firms’ incentives to innovate decline as market power strengthens beyond a point.
- Key coefficients (Table 3, U.S. data):
  - Markup: 0.027*** (0.004); 0.036*** (0.006); 0.043*** (0.007)
  - Markup × markup: -0.009** (0.003); -0.008** (0.004)
  - Markup × concentration: -0.048*** (0.016); -0.045*** (0.016)
  - Number of observations: 59,470; 54,627; 54,627
  - R²: 0.055; 0.055; 0.055

### Robustness checks (investment and innovation)
- Instrumental variables:
  - Instrument: median markup of other firms in the same ICB sub-sector (excluding the firm), and analogous instrument for the interaction term.
  - First-stage strength: each first-stage equation has an F-statistic on the excluded instruments with a p-value well below 0.001 percent.
  - Second-stage results (Table A4): quantitatively similar to baseline OLS results.
  - IV estimates (Table A4): Investment IV Markup 0.202*** (0.060); Investment IV Markup × markup -0.228*** (0.054); R&D IV Markup 0.045 (0.027); R&D IV Markup × markup -0.055** (0.027).
- Additional fixed effects:
  - Adding sector-year fixed effects has little effect on results (Table A5).
- Alternative concentration measure:
  - Using alternative HHI definitions yields similar estimation results (Table A6).
- SGA / production input concern:
  - Re-estimating production functions controlling for SGA produces markups strongly correlated with baseline; sales-weighted average markup increase controlling for SGA is 35 percent over 1980-2016 versus 42 percent in baseline (35/42).

### Profitability and concentration links to markups
- Positive relation between markups and broad measures of profitability:
  - On average, for firms in the United States and other AEs, a 10 percentage point rise in markups is associated with, respectively, 19 and 13 percentage point increases in the ratio of dividends to sales (Table 1).
- Table 1 coefficients (selected):
  - Dividend/sales: 1.923*** (0.233) for USA; 1.309*** (0.056) for Other AEs.
  - Market cap/sales: 0.019*** (0.001) for USA; 0.016*** (0.001) for Other AEs.
  - Concentration: 0.654*** (0.037) for USA; 0.135*** (0.014) for Other AEs.

### Extensions — distance to technological frontier
- Distance to frontier: firm’s TFP gap to the maximum TFP in the sector-year (distance positive except for frontier firm).
- Expanded specification includes ln(markup) × distance and ln(markup) × ln(markup) × distance.
- Findings (Table 4):
  - The cross-derivative relevant to Aghion et al. (2005) is, on average, negative.
  - The cross-derivative is negative if the markup is smaller than 1.87 (a value well above the sample mean), implying the inverted-U is steeper in more neck-and-neck industries.
- Table 4 coefficients (selected):
  - Investment: Markup 0.104*** (0.010); Markup × markup -0.078*** (0.009); Markup × Technology distance -0.035*** (0.011); Markup × markup × Technology distance 0.028*** (0.010).
  - Number of observations: 49,921 (Investment); 51,095 (R&D).

### Extensions — labor share
- Quasi-firm-level labor share constructed from industry-level average wage per employee (OECD STAN) combined with firm-level employment and sales.
- Main findings:
  - Relation between markups and this firm-level labor share is generally negative and monotonic (Table 5).
  - As market concentration increases, the negative relation between markups and the labor share becomes stronger.
  - Interpretation: consistent with view that rising market power reduces the labor share (consistent with Autor et al. (2017)).
- Table 5 coefficients (selected):
  - Markup: -0.095*** (0.026); 0.128*** (0.038); 0.100** (0.044)
  - Markup × concentration: -0.654*** (0.104); -0.657*** (0.104)
  - Number of observations: 87,129; 80,888; 80,888
  - R²: 0.035; 0.036; 0.03

### Cross-country extension — other advanced economies
- Re-estimation for 32 other AEs combined in a panel with country-time fixed effects.
- Results (Table 6): qualitatively similar relations between markups, investment, innovation, and labor share for other AEs as for the United States.
- Table 6 selected coefficients:
  - Investment Markup: 0.107*** (0.012) USA; 0.091*** (0.011) non‑US AE.
  - R&D Markup: 0.043*** (0.007) USA; 1.028*** (0.122) non‑US AE.
  - Labor share Markup: 0.100** (0.044) USA; 0.065 (0.045) non‑US AE.
  - Observations (examples): Investment 52,319 (USA) / 72,616 (non‑US AE); R&D 54,627 / 70,661; Labor share 80,888 / 64,799.

### Key numeric magnitudes and sample facts
- Sample: publicly traded firms in 74 countries; data span 1980-2016; 19 ICB super-sectors estimated.
- U.S. markups: sales-weighted average markup (factor) rises from 1.12 in 1980 to 1.59 in 2016, implying a rise of 42 percent (1.59/1.12).
- Industry range (U.S.) of sales-weighted average markup increases across 10 broad ICB industries: between 7 and 137 percent over 1980-2016.
- Largest sub-sector increase: “Biotechnology” (Health Care) with a 419 percent increase in markups over the period.
- Controlling for SGA: increase over 1980-2016 is 35 percent (17 percent smaller than baseline 42 percent; 35/42).
- AEs excluding U.S. (32 AEs): GDP-weighted average estimated markups increased by 35 percent since 1980.
- Relation to profitability: a 10 percentage point rise in markups associated with 19 percentage point increase in dividends-to-sales (U.S.) and 13 percentage point increase (other AEs).
- Share of U.S. firms with negative marginal association between markups and investment in 2016: 8 percent, 17 percent, and 6 percent across the three specifications; none in 1980.
- Instrumental variables first-stage p-value: well below 0.001 percent.
- Threshold markup for negative cross-derivative in distance-to-frontier test: 1.87.
- Market concentration index (adjusted HHI) in sample: ranges from 0.1 to 1.0.
- Country-level sample coverage examples (Table A1): USA 133,231 firm-year markups; JPN 80,690; CHN 36,000; IND 27,350; TWN 24,427; KOR 24,103; CAN 29,142; GBR 35,989.

### Contribution and open questions
- Contribution:
  - Provides firm-level evidence on the evolution of market power for a large set of countries over several decades.
  - Investigates how changes in firm-level markups relate to firm investment and innovation decisions in a multi-country framework.
- Limitations and avenues for future research:
  - Paper does not assess the underlying causes of the rise in markups (examples: technological change, changes in antitrust regulation) and leaves this for future analysis.
  - Focus limited to publicly traded firms; related work suggests privately held companies may have experienced smaller rises in markups.

*Source: IMF Working Paper (wp18137), 1980–2016 firm-level analysis using Thomson Reuters Worldscope and Compustat-based markup estimation.*

### 2016. In advanced economies, markups have increased by an average of 39 percent since 1980.

### wp18137 - 2016. In advanced economies, markups have increased by an average of 39 percent since 1980.

### Key findings
- Markups of publicly traded firms have generally increased during 1980-2016, especially in advanced economies (AEs).
- For AEs, markups have increased by a GDP-weighted average of 39 percent since 1980.
- The increase is broad-based across industries and countries and is driven by the highest-markup firms in each economic sector; the distribution of markups has widened.
- For emerging markets and developing economies (EMDEs), there is less evidence of a rise in markups.
- Positive relation found between firm markups and other indicators of market power, such as profits and industry concentration.
- Firm-level association between markups and the labor share is generally negative.

### Data, scope, and methodology
- Sample coverage:
  - 74 economies total: 33 AEs and 41 EMDEs.
  - Period: 1980-2016 (37 years).
  - More than 631,000 estimates of firm markups obtained.
- Data sources:
  - Thomson Reuters Worldscope for international publicly traded firms.
  - Compustat approach of De Loecker and Eeckhout (2017) extended internationally.
- Measurement:
  - Markups computed as the ratio of firm sales to the cost of variable inputs, scaled by the output elasticity of variable inputs.
  - Baseline variable input measure: cost of goods sold (COGS).
  - Robustness: findings hold when controlling for additional operational costs such as selling, general, and administrative expenses (SGA).
- Coverage of economic significance:
  - For 2016, U.S. firms in the sample have sales equivalent to 79 percent of U.S. GDP.
  - For the other 73 economies in the sample, firms have sales equivalent, on average, to 75 percent of their respective economy’s GDP.

### Evolution of markups across countries and industries
- Increase concentrated in AEs; less evidence of rise in EMDEs.
- Rise driven by firms able to extract especially high markups within sectors.
- Widening distribution of firm markups between 1980 and 2016.
- Robustness: evolution estimates across countries are robust to alternative estimation approaches.

### Relation between markups, investment, and innovation
- Empirical approach:
  - Firm-level Tobin’s Q models for capital expenditure and R&D expenditure augmented with firm-level markups.
  - Estimations include numerous fixed effects to isolate markup relations from firm-specific and time-varying factors.
  - Instrumental variables approach used to address potential reverse causality; results remain robust.
- Main result:
  - Evidence of a non-monotonic (inverted U-shaped) relation between markups and investment and innovation rates:
    - Higher markups are associated initially with increasing investment and innovation rates and then with decreasing investment and innovation rates.
  - This non-monotonicity is more pronounced for firms closer to the technological frontier.
  - More concentrated industries feature a more negative relation between markups and investment and innovation.
- Consistency with theory:
  - Results broadly consistent with the inverted U-shape prediction from the theoretical model of Aghion and others (2005).

### Extensions and additional firm-level evidence
- Technological frontier:
  - The inverted U-shape relation between markups and investment is steeper for firms closer to the technological frontier, in line with theoretical predictions.
- Labor share:
  - Firm-level evidence supports the prediction that the labor share of income declines in industries where market power rises (consistent with Autor and others (2017)).
- External validity:
  - Main estimation results hold for firms in 32 other AEs beyond the U.S.

### Contribution and open questions
- Contribution:
  - Presents firm-level evidence on the evolution of market power for a large set of countries over several decades.
  - Investigates how changes in firm-level markups relate to firm investment and innovation decisions in a multi-country framework.
- Limitations and avenues for future research:
  - Paper does not assess the underlying causes of the rise in markups (examples: technological change, changes in antitrust regulation) and leaves this for future analysis.
  - Focus limited to publicly traded firms; related work suggests privately held companies may have experienced smaller rises in markups.

### Classification and metadata
- JEL Classification Numbers: D2, D4, E2, J3, K2, and L1.
- Keywords: Market power, markup, concentration, investment, innovation, labor share.
- Author’s E-Mail Address: FDiez@imf.org, DLeigh@imf.org, STambunlertchai@imf.org.

*Source: IMF Working Paper (wp18137), 1980–2016 firm-level analysis using Thomson Reuters Worldscope and Compustat-based markup estimation.*

### Section III relates the evolution of the markup estimates to investment, innovation, and the

### Section III relates the evolution of the markup estimates to investment, innovation, and the

### Macroeconomic implications — overview
- The section investigates relations between firm-level markups and firm-level economic performance, focusing on capital expenditure (investment), R&D spending (innovation), and the labor share.
- Sample focus: United States (most complete data) and an extension to 32 other advanced economies (AEs).

### Investment — specification and main findings
- Dependent variable: investment rate = capital expenditure as a share of the previous year’s capital stock (CapExp / lagged PPE).
- Main explanatory variable: log of firm-level markup (ln(markup)), with firm and time fixed effects and controls including Tobin’s Q, sales rate, and lagged R&D rate.
- Nonlinear specification: interaction terms included for ln(markup) × ln(markup) and ln(markup) × Concentration (adjusted HHI at 115 ICB sub-sector level).
- Main empirical patterns:
  - Estimated inverted-U (non-monotonic) relation between markups and investment: higher markups initially associated with increasing investment; at higher markup levels, increases in markups associated with lower investment.
  - Industry heterogeneity: the inverted-U relation is not driven by any single industry; coefficients for quadratic markup interacted with industry dummies are negative for all 10 ICB industries.
  - Concentration effect: as market concentration rises, higher markups are more likely to be associated with lower investment. For concentration index values above 0.5, higher markups are associated with lower investment.
  - Sub-sector market concentration index in sample ranges from 0.1 to 1.0.
- Macroeconomic significance (U.S. firms):
  - Computed share of U.S. publicly listed firms for which rising markups are associated with lower investment in 2016:
    - 8 percent, 17 percent, and 6 percent across the three specifications reported in Table 2 (columns 1–3), respectively.
  - In contrast, no firms had a negative association between higher markups and investment as of 1980.

### Innovation (R&D) — specification and main findings
- Dependent variable: R&D expenditure as a share of the previous year’s total assets (RD / lagged TA).
- Controls: Tobin’s Q and level of sales as a share of previous year’s total assets, plus firm and time fixed effects.
- Main empirical patterns:
  - Non-monotonic (inverted-U) relation between markups and R&D, similar to physical investment: higher markups initially associated with increasing R&D; at higher markups or higher concentration levels, the marginal relation becomes negative.
  - Interpretation: consistent with Aghion et al. (2005): firms’ incentives to innovate decline as market power strengthens beyond a point.

### Robustness checks (investment and innovation)
- Instrumental variables:
  - Instrument: median markup of other firms in the same ICB sub-sector (excluding the firm), and analogous instrument for the interaction term.
  - First-stage strength: each first-stage equation has an F-statistic on the excluded instruments with a p-value well below 0.001 percent.
  - Second-stage results (Table A4): quantitatively similar to baseline OLS results.
- Additional fixed effects:
  - Adding sector-year fixed effects (to capture sector-specific technological change) has little effect on results (Table A5).
- Alternative concentration measure:
  - Using an alternative HHI that computes squared shares of firm sales relative to sector mean sales yields similar estimation results (Table A6).
- SGA / production input concern (related robustness earlier in paper):
  - Re-estimating production functions controlling for SGA produces markups strongly correlated with baseline; sales-weighted average markup increase controlling for SGA is 35 percent over 1980-2016 versus 42 percent in baseline (35/42).

### Profitability and concentration links to markups
- Positive relation between markups and broad measures of profitability:
  - On average, for firms in the United States and other AEs, a 10 percentage point rise in markups is associated with, respectively, 19 and 13 percentage point increases in the ratio of dividends to sales (Table 1).
  - Markups are positively related with market concentration (adjusted HHI constructed on deciles of sector sales); this supports interpretation that rising markups reflect rising market power rather than solely cost-recovery of fixed investment.

### Extensions — distance to technological frontier
- Distance to frontier measure: firm’s TFP gap to the maximum TFP in the sector-year (distance is positive except for frontier firm).
- Expanded specification: adds interactions ln(markup) × distance and ln(markup) × ln(markup) × distance.
- Findings:
  - The cross-derivative relevant to Aghion et al. (2005) is, on average, negative.
  - The cross-derivative is negative if the markup is smaller than 1.87 (a value well above the sample mean), implying support for the model prediction that the inverted-U shape is steeper in more neck-and-neck industries (smaller technological gaps).

### Extensions — labor share
- Quasi-firm-level labor share constructed from industry-level average wage per employee (OECD STAN) combined with firm-level employment and sales.
- Main findings:
  - Relation between markups and this firm-level labor share is generally negative and monotonic (Table 5).
  - As market concentration increases, the negative relation between markups and the labor share becomes stronger.
  - Interpretation: consistent with view that rising market power reduces the labor share (consistent with Autor et al. (2017)).

### Cross-country extension — other advanced economies
- Re-estimation for 32 other AEs combined in a panel with country-time fixed effects.
- Results (Table 6): qualitatively similar relations between markups, investment, innovation, and labor share for other AEs as for the United States, suggesting broadly comparable implications of rising market power across AEs.

### Key numeric magnitudes and sample facts (preserved)
- Sample: publicly traded firms in 74 countries; data span 1980-2016; 19 ICB super-sectors estimated.
- U.S. markups: sales-weighted average markup (factor) rises from 1.12 in 1980 to 1.59 in 2016, implying a rise of 42 percent (1.59/1.12).
- Industry range (U.S.) of sales-weighted average markup increases across 10 broad ICB industries: between 7 and 137 percent over 1980-2016.
- Largest sub-sector increase: “Biotechnology” (Health Care) with a 419 percent increase in markups over the period.
- Controlling for SGA: increase over 1980-2016 is 35 percent (17 percent smaller than baseline 42 percent; 35/42).
- AEs excluding U.S. (32 AEs): GDP-weighted average estimated markups increased by 35 percent since 1980.
- Relation to profitability: a 10 percentage point rise in markups associated with 19 percentage point increase in dividends-to-sales (U.S.) and 13 percentage point increase (other AEs).
- Share of U.S. firms with negative marginal association between markups and investment in 2016: 8 percent, 17 percent, and 6 percent across the three specifications; none in 1980.
- Instrumental variables first-stage p-value: well below 0.001 percent.
- Threshold markup for negative cross-derivative in distance-to-frontier test: 1.87.
- Market concentration index (adjusted HHI) in sample: ranges from 0.1 to 1.0.

*Source: wp18137 - Section III relates the evolution of the markup estimates to investment, innovation, and the (IMF Working Paper).*

### REFERENCES

### wp18137 - REFERENCES

### References
- Ackerberg, D., K. Caves, and G. Frazer, 2015, “Identification Properties of Recent Production Function Estimators,” Econometrica, 83(6): 2411–51.
- Aghion, P., N. Bloom, R. Blundell, R. Griffith, and P. Howitt, 2005, “Competition and Innovation: an Inverted-U Relation,” The Quarterly Journal of Economics, 120(2): 701–28.
- Autor, D., D. Dorn, L. Katz, C. Patterson, and J. Van Reenen, 2017, “The Fall of the Labor Share and the Rise of Superstar Firms,” NBER Working Paper No. 23396.
- Baqaee, D. R. and Farhi, E. (2017), “Productivity and Misallocation in General Equilibrium,” NBER Working Paper 24007.
- Baumol, W. (1982), “Contestable Markets: an Uprising in the Theory of Industry Structure,” American Economic Review, 72(1):1-15.
- Council of Economic Advisers, 2016. “Benefits of Competition and Indicators of Market Power.” Issue Brief, Apr. 2016. https://www.whitehouse.gov/sites/default/files/page/files/20160414_cea_competition_issue_brief.pdf.
- De Loecker, J., and J. Eeckhout, 2017, “The Rise of Market Power and the Macroeconomic Implications,” NBER Working Paper No. 23687.
- De Loecker, J., P. Goldberg, A. Khandelwal and N. Pavcnik, 2016, “Prices, Markups and Trade Reform,” Econometrica, 84(2): 445–510.
- De Loecker, J., and F. Warzynski, 2012, “Markups and Firm-Level Export Status,” American Economic Review, 102(6): 2437:2471.
- Díez, F., J. Fan, and C. Villegas-Sanchez (2018), “Global Declining Competition,” forthcoming IMF Working Paper.
- Eggertsson, G. B., Robbins, J. A., and Wold, E. G. (2018), “Kaldor and Pikettys facts: The Rise of Monopoly Power in the United States.” NBER Working Paper 24287.
- Hall, Robert E., 1988, “The Relation between Price and Marginal Cost in U.S. Industry,” Journal of Political Economy, 96(5): 921–47.
- Karabarbounis, L., and B. Neiman, 2018, “Accounting for Factorless Income,” NBER Working Paper No. 24404.
- Shapiro, Carl, 2018. “Antitrust in a Time of Populism.” International Journal of Industrial Organization, forthcoming.
- Traina, J., 2018, “Is Aggregate Market Power Increasing? Production Trends Using Financial Statements” Stigler Center New Working Paper Series No 17.

### Figures (captions and key labels)
- Figure 1. United States: Evolution of Estimated Markups (Sales‐weighted mean for all publicly listed firms)
  - X-axis tick labels: 1980 1990 2000 2010 2020
  - Y-axis tick labels: 1.1 1.2 1.3 1.4 1.5 1.6
  - Source: Authors’ estimates based on Thomson Reuters WorldScope data.
- Figure 2. U.S. Firms: Markups in 2016 vs. 1980 by Industry
  - Note: Dashed line indicates 45-degree line; size of markers based on sales in 2016; color indicates ICB industry.
- Figure 3. U.S. Firms: Distribution of Markups, 1980 and 2016
  - Panel 1: All U.S. firms; Panel 2: Firms by U.S. industry
  - ICB industry mapping: 1 = Oil & Gas; 2 = Basic Materials; 3 = Industrials; 4 = Consumer Goods; 5 = Health Care; 6 = Consumer Services; 7 = Telecommunications; 8 = Utilities; 9 = Financials; 10 = Technology.
  - Density and Markup axis tick labels include: 0 .5 1 1.5 2 (Density) and 0 2 4 6 8 10 (Markup) across panels; years compared: 1980 2016.
- Figure 4. Estimates of Output Elasticity (β_i) Over Time
  - Note: U.S. data; ICB industry coverage as in Figure 3.
  - Y-axis tick labels: 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
  - Periods: 1980-1998 and 1999-2016.
- Figure 5. Estimated Markups for U.S. Firms: Baseline and Alternative Approach
  - Panel: Baseline and Controlling for SGA (Selling, General and Administrative Expenses as in equation (7))
  - X-axis tick labels: 1980 1990 2000 2010 2020
  - Y-axis tick labels include: .9 1 1.1 1.2 1.3 1.4 1.5 1.6
- Figure 6. Evolution of Estimated Markups Across Economies (Sales‐weighted mean for all publicly listed firms)
  - Note: Markup estimates for 33 advanced economies (AEs) and 41 emerging market and developing economies (EMDEs). For country groups (AE Europe, Latin America, and EMDE Asia) figure reports median. IQR denotes inter‐quartile range.
  - Y-axis tick labels: .8 1 1.2 1.4 1.6
  - Sample series labeled: United States Canada Japan AE Europe AE IQR Advanced Economies; Latin America EMDE Asia EMDE IQR Emerging Market and Developing Economies.
- Figure 7. Firms in 32 Non‑US AEs: Distribution of Markups, 1980 and 2016
  - Panels and ICB industry mapping as in Figure 3.
  - Density and Markup axis tick labels include: 0 1 2 3 (Density) and 0 2 4 6 8 10 (Markup).
- Figure 8. Investment Rate vs. Markup
  - Note: Fitted value of investment rate vs. markup based on estimates in Table 2 (column 1). Dashes indicate 90 percent confidence interval.
- Figure 9. Estimated Markup Coefficient (β̂) vs. Market Concentration
  - Note: Composite coefficient on (log) firm markup across sample range of market concentration based on Table 2 (column 2). Dashes indicate 90 percent confidence interval.
  - X-axis tick labels: -.1 -.05 0 .05 .1 (Markup coefficient)
  - Market concentration tick labels: 0 2 4 6 8 1 (plotted as 0.2.4.6.81 in source formatting).
- Figure 10. Marginal Association Between Markups and Investment Rate: Distribution for U.S. Firms
  - Note: Marginal association based on composite slope coefficient from Table 2 (column 3).
  - X-axis tick labels: 0 .1 .2 .3 .4 (Density)
  - Composite slope coefficient axis tick labels: -10 -5 0 5 10
  - Years compared: 1980 2016
- Figure 11. Investment Rate vs. Markup by Distance to Technology Frontier
  - Note: U.S. data. High/low distance indicates technology gap of i-th firm compared to TFP frontier (maximum) across all firms, by sub-sector and year, is in the top/lowest 5 percent of sample.
  - X-axis tick labels: -5 0 5 10 15 20 (Investment rate (percent))
  - Markup axis tick labels: 0 2 4 6 8

### Tables — key captions and reported numeric results (exact values preserved)
- Table 1. Relation Between Markup and Other Measures of Market Power
  - Note: Firm markup relation estimated with firm and time-fixed effects; sector-level aggregated markup relation estimated with country- and time-fixed effects; standard errors clustered by sector. Significance: *** p<0.01, ** p<0.05, * p<0.1.
  - USA vs Other AEs coefficients:
    - Dividend/sales: 1.923*** (0.233) for USA; 1.309*** (0.056) for Other AEs.
    - Market cap/sales: 0.019*** (0.001) for USA; 0.016*** (0.001) for Other AEs.
    - Concentration: 0.654*** (0.037) for USA; 0.135*** (0.014) for Other AEs.

- Table 2. Investment Rate Equation Estimates
  - Note: U.S. data. Markup denotes log of markup of i-th firm. Additional controls included (Tobin’s Q and sales-to-lagged assets ratio in previous year). Heteroskedasticity-robust standard errors. *** p<0.01, ** p<0.05, * p<0.1.
  - Columns (1), (2), (3):
    - Markup: 0.061*** (0.007); 0.072*** (0.012); 0.107*** (0.012)
    - Markup × markup: -0.045*** (0.005); -0.043*** (0.005)
    - Concentration: 0.025 (0.016); 0.022 (0.015)
    - Markup × concentration: -0.152*** (0.031); -0.131*** (0.031)
    - Firm FE: Yes; Time FE: Yes
    - Number of observations: 57,371; 52,319; 52,319
    - R²: 0.228; 0.221; 0.22

- Table 3. R&D Rate Equation Estimates
  - Note: U.S. data. Additional controls included. Heteroskedasticity-robust standard errors. Significance as above.
  - Columns (1), (2), (3):
    - Markup: 0.027*** (0.004); 0.036*** (0.006); 0.043*** (0.007)
    - Markup × markup: -0.009** (0.003); -0.008** (0.004)
    - Concentration: 0.014* (0.007); 0.013* (0.007)
    - Markup × concentration: -0.048*** (0.016); -0.045*** (0.016)
    - Firm FE: Yes; Time FE: Yes
    - Number of observations: 59,470; 54,627; 54,627
    - R²: 0.055; 0.055; 0.055

- Table 4. Technology Gap, Markups, Investment, and R&D
  - Note: Technology gap indicates i-th firm’s distance to the TFP frontier (maximum) across all firms in the subsector by year; inverse measure of “neck-and-neckness.” Heteroskedasticity-robust standard errors. Significance as above.
  - Dependent variable columns: Investment and R&D
    - Markup: 0.104*** (0.010) for Investment; 0.030*** (0.005) for R&D
    - Markup × markup: -0.078*** (0.009) for Investment; -0.013*** (0.005) for R&D
    - Markup × Technology distance: -0.035*** (0.011) for Investment; -0.001 (0.005) for R&D
    - Markup × markup × Technology distance: 0.028*** (0.010) for Investment; 0.003 (0.005) for R&D
    - Firm FE: Yes; Time FE: Yes
    - Number of observations: 49,921 (Investment); 51,095 (R&D)
    - R²: 0.237 (Investment); 0.059 (R&D)

- Table 5. Labor Share Equation Estimates
  - Note: U.S. data. Heteroskedasticity-robust standard errors. Significance as above.
  - Columns (1), (2), (3):
    - Markup: -0.095*** (0.026); 0.128*** (0.038); 0.100** (0.044)
    - Markup × markup: 0.009 (0.019); 0.029 (0.019)
    - Concentration: 0.182*** (0.053); 0.182*** (0.053)
    - Markup × concentration: -0.654*** (0.104); -0.657*** (0.104)
    - Firm FE: Yes; Time FE: Yes
    - Number of observations: 87,129; 80,888; 80,888
    - R²: 0.035; 0.036; 0.03

- Table 6. Firm-level Equation Estimates for the United States and Other AEs
  - Note: Markup denotes log of markup of i-th firm. Heteroskedasticity-robust standard errors. Significance as above.
  - Reported coefficients across models (Investment, R&D, Labor share) for USA and non‑US AE:
    - Markup: 0.107*** (0.012) USA; 0.091*** (0.011) non‑US AE (Investment)
    - Markup: 0.043*** (0.007) USA; 1.028*** (0.122) non‑US AE (R&D)
    - Markup: 0.100** (0.044) USA; 0.065 (0.045) non‑US AE (Labor share)
    - Markup × markup: -0.043*** (0.005) USA; -0.053*** (0.013) non‑US AE; other reported interactions per table including standard errors.
    - Concentration and Markup × concentration coefficients and standard errors reported per table.
    - Firm FE: Yes across specifications; Time FE and Time × country FE included as indicated.
    - Observations and R² values: Observations 52,319 (USA, Investment) / 72,616 (non‑US AE); 54,627 / 70,661 (R&D); 80,888 / 64,799 (Labor share). R² reported: 0.22, 0.155, 0.055, 0.127, 0.036, 0.124 (as formatted in source).

### Appendix tables and sample coverage
- Table A1. List of Countries in Sample
  - Note: Table reports list of countries indicated by ISO three-letter codes, and number of firm-year markup estimates (N).
  - Examples (exact country entries and N preserved as in source formatting):
    - AUS 23,874; AUT 1,882; BEL 2,796; CAN 29,142; CHE 4,870; CZE 558; DEU 15,482; DNK 3,422; ESP 3,554; EST 189; FIN 2,941; FRA 16,784; GBR 35,989; GRC 4,785; HKG 13,968; IRL 1,357; ISL 196; ISR 5,418; ITA 5,513; JPN 80,690; KOR 24,103; LTU 335; LUX 509; NLD 4,334; NOR 3,477; NZL 2,047; PRT 1,405; SGP 9,887; SVK 257; SVN 510; SWE 7,924; TWN 24,427; USA 133,231.
    - Emerging market and developing economies examples: ARE 605; ARG 1,475; BGR 2,648; BHR 229; BRA 5,215; CHN 36,000; COL 861; EGY 1,878; HRV 925; HUN 701; IDN 6,433; IND 27,350; JOR 1,657; KAZ 298; KEN 386; KWT 1,403; LBN 48; LKA 2,174; MAR 728; MEX 2,443; MUS 338; MYS 16,398; NGA 723; OMN 969; PAK 3,457; PER 1,463; PHL 3,323; POL 5,509; QAT 321; ROU 1,505; RUS 5,842; SAU 1,236; SRB 902; THA 9,364; TUR 4,697; UKR 912; VEN 355; VNM 6,385; ZAF 5,537; ZWE 184.
- Table A2. Summary Statistics for Selected Variables
  - Note: Summary statistics for full sample of 74 economies and number of observations (N) available for each variable (table content summarized in source).
- Table A3. U.S. Firm-level Investment Equation Estimates: by U.S. Industry
  - Note: Markup denotes log of markup; results for 10 ICB industries; additional controls and fixed effects as in Table 2 (column 1). Significance: *** p<0.01, ** p<0.05, * p<0.1.
  - Selected firm-level percentiles and sector coefficients preserved:
    - Variable percentiles (25th 50th 75th N):
      - Sales-to-COGS ratio: 1.18 1.39 1.85 717,958
      - Investment rate (percent): 2.81 7.09 17.22 294,418
      - R&D rate (percent): 0.25 7.18 50.72 281,970
      - Employment (number of employees): 156 675 2,700 626,304
    - Sector markup and markup × markup reported for sectors:
      - Oil & Gas 0.094***   -0.060**
      - Basic Materials 0.055** -0.044*
      - Industrials 0.111*** -0.048***
      - Consumer Goods 0.084*** -0.050**
      - Health Care -0.006 -0.002
      - Consumer Services 0.053** -0.037**
      - Telecommunications 0.015 -0.007
      - Utilities 0.045 -0.012
      - Financials 0.081** -0.058
      - Technology 0.068*** -0.054***
    - Firm FE: Yes; Time FE: Yes; Number of observations: 62,726; R²: 0.25
- Table A4. U.S. Firm-level Equation Estimates: OLS and Instrumental Variables
  - Note: Instrument for markup is median markup for other firms in same ICB sub-sector (excluding i-th firm). Additional controls included. Heteroskedasticity-robust standard errors. Significance as above.
  - Columns (1)-(4) (Dependent variables: Investment, Investment, R&D, R&D; Estimation procedure OLS, IV, OLS, IV):
    - Markup: 0.061*** (0.007) OLS Investment; 0.202*** (0.060) IV Investment; 0.027*** (0.004) OLS R&D; 0.045 (0.027) IV R&D.
    - Markup × markup: -0.045*** (0.005) OLS Investment; -0.228*** (0.054) IV Investment; -0.009** (0.003) OLS R&D; -0.055** (0.027) IV R&D.
    - Firm FE: Yes; Time FE: Yes
    - Number of observations: 57,371; 57,292; 59,470; 59,393
    - R²: 0.22; 80.181; 0.055; 0.019 (as formatted in source).
- Table A5. U.S. Firm-level Equation Estimates: Additional Fixed Effects
  - Note: Columns (2) and (4) include sector × time fixed effects. Additional controls included as in previous tables.
  - Columns (1)-(4) (Investment, Investment, R&D, R&D):
    - Markup: 0.061*** (0.007) across columns reported.
    - Markup × markup: -0.045*** (0.005); -0.041*** (0.005); -0.009** (0.003); -0.008** (0.003)
    - Firm FE: Yes; Time FE and Time × sector FE included as indicated.
    - Number of observations: 57,371; 57,371; 59,470; 59,470
    - R²: 0.22; 80.255; 0.055; 0.070 (as formatted in source).
- Table A6. U.S. Firm-level Investment Rate Equation Estimates: Market Concentration Based on Mean-adjusted HHI Index
  - Note: Column (1) baseline (as in Table 1, column 2). Column (2) HHI index standardized. Column (3) alternative HHI index defined in source text (formula preserved in source). Markup denotes log of markup. Additional controls included. Heteroskedasticity-robust standard errors. Significance as above.
  - Columns (1), (2), (3):
    - Markup: 0.072*** (0.012); 0.021*** (0.004); 0.015*** (0.004)
    - Concentration: 0.025 (0.016); 0.003 (0.002); -0.007 (0.007)
    - Markup × concentration: -0.152*** (0.031); -0.021*** (0.004); -0.023*** (0.007)
    - Firm FE: Yes; Time FE: Yes for columns as indicated.
    - Number of observations: 52,319; 52,319; 57,371
    - R²: 0.221; 0.221; 0.227

*Source: wp18137 - REFERENCES (pdf content provided).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2018/wp18137.pdf_