## sdnea2021001

## Source details

**Canonical URL:** [sdnea2021001](https://www.imf.org/-/media/files/publications/sdn/2021/english/sdnea2021001.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/sdn/2021/english/sdnea2021001.pdf.md)
- [Structured JSON version](/-/media/files/publications/sdn/2021/english/sdnea2021001.pdf.json)

---

### Executive summary — scope and headline measures
- Corporate market power has risen in recent decades and will likely increase further as a result of the wave of bankruptcies triggered by the pandemic.
- Key aggregate empirical magnitudes (publicly listed firms in advanced economies since early 1980s):
  - “Global” markups have increased more than 30 percent since 1980.
  - Among advanced economies, markups have increased more than 35 percent.
  - Industry concentration has increased more than 30 percent since 1980.
  - Profitability (cash dividends to sales) increased more than 140 percent, rising from 1.5 percent to over 3 percent, on average.
- Geographic and sector heterogeneity:
  - Markup increase among North American (US and Canadian) firms ≈ markups among euro area firms (almost 40 percent); about two times larger than among Asian (Japanese and Korean) firms.
  - Markups increased across all broad industries; increase in health care and technology industries is more than three times larger than in industrials and consumer goods.
  - Emerging market economies show mostly constant market power measures over the sample period.
- Main themes covered: rise and persistence of market power, links to declining business dynamism and M&As, product–labor market power connections, macroeconomic policy implications, and competition policy recommendations.

### Entrenchment, concentration dynamics, and COVID-19
- Entrenchment and persistence:
  - Top decile of markup distribution more than doubled their markups; bottom half increased by just 5 percent.
  - If a firm is in the top 10 percent markups, there is now almost an 85 percent chance it remains so the following year—up almost 10 percentage points since the second half of the 1990s.
  - Declining turnover among high-markup firms suggests growing lack of competitors.
- COVID-19 effects on concentration:
  - Orbis illustration across 21 countries:
    - No–COVID-19 median concentration ratio (share of top 4 in top 20 firms’ sales within a 4-digit sector) ≈ 56 percent.
    - With COVID-19 and reassignment of bankrupt firms’ sales to leading firms, median concentration could rise by 4 percentage points, to 60 percent.
  - Pandemic-driven shift to online activity benefits large technology companies and may further strengthen concentration.

### Business dynamism and the role of M&As
- Broad-based decline in business dynamism:
  - Falling entry rate of new firms.
  - Decline in dispersion of growth rates across firms.
  - Decline in contribution of young firms (≤ five years) to aggregate output.
- Quantified links between concentration, M&As, and dynamism:
  - For countries/industries with the 25 percent largest increases in concentration, rising concentration may explain about 10 percent of the overall decline in the share of output by young firms (roughly 0.9 percentage points).
  - A one standard deviation increase in the share of M&As conducted by leading firms is associated with slightly less than half a percentage point decline in the share of output accounted for by young firms—corresponding to a 3.5 percent fall relative to the average share.
  - In the 25 percent of countries/industries with largest increases in acquisitions by leading firms, M&As may explain about 3 percent of the overall decline in the share of output by young firms and about 9 percent of the decline in growth rate dispersion.
- Firm-level effects of M&As:
  - Immediately after acquisition, acquiring firms obtain 2.6 percent higher profitability and 3.6 percent higher markups.
  - A one standard deviation increase in a firm’s share of M&As is associated with about a 25 percent increase in sales and employment growth and a 10 percent rise in R&D growth relative to average growth rates.
  - Pro-growth effects of M&As are greatly reduced if acquiring firms are market leaders.
  - A one standard deviation increase in the share of deals by a firm’s competitor reduces the firm’s sales and R&D growth by about 4 percent, relative to average growth rates; adverse effects larger when competitor is a market leader.
  - US Compustat analysis: shift in M&A activity toward largest firms associated with increases in HHI, CR4, profit share of revenue, and markups, and decreases in dispersion of firms’ growth rates.

### Industry-specific patterns and drivers
- Framing: technology, pharmaceuticals, banking (ICB-based industry definitions; industry analysis starts in 1995).
- Technology (ICB = 9; subsector ICB = 953):
  - Markups increased more than 30 percent between 1995 and 2016.
  - Concentration increased more than 10 percent globally between 1995 and 2016; sales by top 4 firms ≈ two-thirds of total sales by top 20 firms in 2016.
  - Profitability: average cash dividends to sales increased from 0.7 to 3.5 percent.
  - Melitz-Polanec decomposition: reallocation effect accounts for about 95 percent of total markup change (1995–2016); since start of decade, incumbents’ markup increases account for 15 percent of overall increase.
  - Large markup increases persist after controlling for intangible assets and overhead costs.
  - Implication: short-run growth effects ambiguous; long-run dominance may reduce incentives to innovate and slow growth.
- Pharmaceuticals (ICB = 457):
  - Since 1995: markups increased almost 40 percent; concentration rose almost 80 percent (ratio sales top 4 to top 20 from ~50 to 90 percent); profitability doubled—cash dividends > 7.7 percent of sales by 2016.
  - Decomposition: increased markups by incumbents account for >50 percent of global markup increase; market-share reallocation accounts for ~1/3; net entry minor.
  - Concerns: excessive rewards to incremental innovations, “patent thickets,” “killer acquisitions”; policy responses likely vary across jurisdictions.
  - Large markup increases persist after controlling for intangible assets and overhead costs.
- Banking:
  - Trade-offs between competition and financial stability; post-crisis forces push in opposite directions (consolidation vs. nonbank/fintech competition).
  - Lerner index appears to rise in advanced economies post-crisis, but trend largely mechanical due to falling policy rates; when corrected for policy rates and cyclical factors, no increase remains.
  - Other indicators (HHI, profitability) do not display upward trend in advanced economies.
  - In emerging market and developing economies since mid-2000s: Lerner oscillated around similar level, concentration dropped, profits/income relatively flat.
  - Methodology: Lerner estimated using data on 67 countries (27 advanced, 40 emerging) between 2000 and 2017; price = share of income to assets; marginal cost from trans-log cost function.
  - Policy takeaway: no apparent structural increase in banking market power from crisis response, but monitoring needed—prolonged low rates could trigger consolidation and reduce competition.

### Labor-market power and worker outcomes
- Measures: employer concentration in local labor markets, institutional indicators, firm-specific labor supply elasticity (labor markdown) extending De Loecker and Warzynski (2012) using Orbis.
- Cross-country firm-level findings (2000–2015):
  - Firms’ labor markdown did not rise across the board, but a small fraction of already powerful firms slightly increased labor market power.
  - Heterogeneity by industry: some increase in manufacturing; decline in finance and insurance.
  - Larger firms tend to have greater labor market power: may pay higher wages in levels but pay less relative to workers’ marginal productivity.
  - Evidence that a fraction of powerful product-market firms can suppress wages; powerful product-market firms gained more labor market power than others since early 2000s.
- Policy implications:
  - Firms’ labor market power can be relevant for competition authorities in some cases, including M&As.
  - Authorities should vigorously enforce prohibitions against “no-poaching” agreements.

### Macroeconomic policy transmission — monetary and fiscal effects
- Monetary policy transmission:
  - Strong market power dampens monetary transmission because high-markup firms’ demand is less sensitive to relative price changes; profits largely economic rents, so less incentive to adjust inputs when costs change.
  - High-markup firms’ financial buffers shelter them from shifts in external funding conditions.
  - GIMF simulations: US employment response to a 100-basis-point decline in nominal policy rate is weaker with higher markups observed today versus early 1980s markups.
  - Orbis firm-level local projections: low-markup firms’ output responds more to central bank policy rate changes; high-markup firms’ output responds less; young low-markup firms respond more than young and old high-markup firms.
  - Monetary policy shocks computed as forecast errors of policy rates not explained by forecast errors for growth and inflation; analyses control for country-sector-time and firm fixed effects and firm characteristics.
- Fiscal policy transmission:
  - Lower market competition weakens transmission of targeted fiscal stimulus; prices/markups increase in imperfectly competitive sectors, limiting output response.
  - China 2009–10 stimulus (RMB 4 trillion; more than 10 percent of 2008 GDP): in cities with more competitive construction sectors, private investment in construction responded more to local government stimulus; instrumental variable evidence supports causal interpretation.
  - Analysis used data for 334 Chinese municipalities grouped into 50 bins, controlling for province fixed effects. Construction competitiveness measured as log of total number of construction and real estate firms per 1,000 people.
  - Policy implication: promoting competitiveness strengthens fiscal stimulus effectiveness and should inform stimulus design by accounting for supply-side heterogeneity.

### Competition-policy adjustments and enforcement recommendations
- Overarching guidance:
  - Adjustments tailored to national circumstances and digital-economy challenges; strengthen competition policy without abandoning fundamentals (for example, consumer welfare criterion) but with broader interpretation to include innovation incentives and dynamism.
- M&A enforcement:
  - Risk of underenforcement; M&As—especially by leading firms—can harm competitors and business dynamism.
  - Recommendations:
    - Reconsider jurisdictional thresholds (revenue, deal value, market share) to capture acquisitions of nascent competitors; consider deal-price thresholds and notification of series of small acquisitions by same acquirer.
    - Bring test cases to clarify legislation application.
    - Conduct ex post analyses of past M&A decisions to improve enforcement; accept some false positives ex post to avoid too many false negatives.
    - Assess corporate power in input markets, notably labor markets; enforce prohibitions on “no-poaching” agreements.
  - Notable factual examples preserved:
    - February 2020: US FTC issued special orders to several large technology firms requiring information about previous acquisitions over the past 10 years not reported under the Hart-Scott-Rodino Act.
    - December 2020: FTC and group of 48 Attorney Generals filed lawsuits alleging Facebook engaged in anticompetitive practices relating in part to its acquisitions of Instagram and WhatsApp.
    - Example legislative change: new German legislation allowing competition authority to request that specific large businesses notify it of much smaller acquisitions in certain sectors.
- Abuse of dominant position and market investigations:
  - Active enforcement of abuse restrictions can mitigate entrenchment.
  - Market investigations/sector inquiries useful for identifying anticompetitive behavior and imposing remedies; resource-intensive but appropriate for markets transformed by technology or critical infrastructure.
  - Example: UK Competition and Markets Authority’s in-depth market investigations can change firm conduct or require divestments.
  - Resourcing needs: more and deeper market/M&A investigations, new sector expertise, ex post evaluation, stronger domestic and international cooperation; evidence suggests some authorities’ budgets have not kept pace (example: combined FTC and DOJ Antitrust Division budget roughly halved as share of US GDP during 1980–2018).
- Digital-economy-specific responses:
  - Institutional tools:
    - Build dedicated digital market units; use interim (conservatory) measures prior to final decisions.
    - Reconsider revenue-based thresholds to capture nascent competitors.
  - Regulatory and enforcement mix:
    - Interoperability requirements, data portability, flexible regulation (codes of conduct, targeted rules).
    - Examples/precedents: number portability; Open Banking; Australia’s consumer data right; EU regulation on platform-to-business relations; Australia and UK discussions on divestments.
  - Caution on structural remedies:
    - Regulating digital firms like utilities or “breaking them up” poses major challenges due to rapid evolution and complexity of digital services; utility-style regulation could be costly and complex.
  - Data access and portability:
    - Data access is a source of market power amplified by processing power and algorithms.
    - Policy tools: data portability, data interoperability, open standards where appropriate; design must protect privacy and firms’ incentives, keep obligations simple to avoid excessive compliance costs that could bar entry.
- International cooperation and norms:
  - Need stronger cross-border cooperation and information sharing; promote international best-practice guidelines; strengthen role of international institutions (OECD, UNCTAD, International Competition Network).
  - Recommendation: coordinate to avoid fragmentation; support international dialogue (example: Canada hosting annual Digital Enforcement Summit).
- Risks and safeguards:
  - Avoid using competition policy for industrial objectives; international coordination reduces risk of protectionist misuse.
  - Maintain strong domestic competition authorities with comparable objectives to reduce strategic misuse risk.

*Source: sdnea2021001 - EXECUTIVE SUMMARY*

### EXECUTIVE SUMMARY____________________________________________________________________________________ 5

### EXECUTIVE SUMMARY

### Introduction
- Corporate market power has risen in recent decades and will likely increase further as a result of the wave of bankruptcies triggered by the pandemic.
- The note provides fresh evidence on the rise in market power, differences across industries, links to declining business dynamism and M&As, connections between product- and labor-market power, implications for competition policy frameworks, and implications for macroeconomic policies.

### Key empirical findings on market power
- Corporate market power has increased significantly among publicly listed firms in advanced economies since the early 1980s.
- Measures and magnitudes:
  - “Global” markups have increased more than 30 percent, on average, since 1980.
  - Among advanced economies, markups have increased more than 35 percent.
  - Industry concentration has increased more than 30 percent since 1980.
  - Profitability, measured as the ratio of cash dividends to sales, increased more than 140 percent, rising from 1.5 percent to over 3 percent, on average.
- Geographic and industry heterogeneity:
  - The increase in markups among North American (US and Canadian) firms is about the same as that of euro area firms (almost 40 percent) but two times larger than the increase among Asian (Japanese and Korean) firms.
  - Markups have increased across all broad industries but are substantially larger in some sectors; the increase among firms in the health care and technology industries is more than three times larger than among firms in the industrials and consumer goods industries.
- Emerging market economies show mostly constant market power measures over the sample period.

### Concentration, persistence, and business dynamism
- The rise in market power is concentrated among a small group of firms whose high price markups are increasingly persistent.
- Concomitant broad-based decline in business dynamism:
  - Falling share of economic activity accounted for by young firms.
  - Decline in the dispersion of growth rates across firms.
- While rising market power is not the dominant driver of declining business dynamism, it has been a contributing factor.

### Role of mergers and acquisitions (M&As)
- The rise in M&As, especially by dominant firms, partly underlies the connection between rising market power and declining business dynamism.
- M&As can have mixed effects:
  - Potential benefits: exit strategy for entrepreneurs, economies of scale and scope, possible consumer benefits.
  - Potential harms: strengthen ability to charge higher prices, weaken incentives to innovate, discourage competitors from innovating.
- Industry- and firm-level evidence:
  - M&As by dominant firms are associated with lower business dynamism at the industry level.
  - Acquiring firms increase market power following transactions.
  - Competitors’ growth and research and development take a hit after rivals’ M&As.

### Industry-specific patterns
- Technology:
  - The overall increase in the tech sector reflects mostly the growing market share of high-markup (and usually more productive) firms.
- Pharmaceuticals:
  - Increase driven primarily by rising market power within incumbent firms.
  - Can reward major breakthroughs (example: recent Covid-19 vaccines, whose development also benefitted from government subsidies) but can be harmful when rewarding excessively incremental innovations.
- Banking:
  - Market power does not appear to have increased significantly in the banking industry; evidence is mixed.

### Effects on macroeconomic policy effectiveness
- Rising market power has made monetary and fiscal policies less effective:
  - Macroeconomic policies have a smaller impact on high-markup than on low-markup firms because high-markup firms have fewer incentives to adjust output when input costs change and are more immune to shifts in external financing conditions.
- Rising market power, while making economies more stable, reduces the transmission of policy stimulus to the real economy.

### Labor market power and worker outcomes
- The note finds some evidence that large firms can suppress wages, possibly because increased employer concentration restricts workers’ alternatives.
- Implications:
  - Firms’ labor market power can be relevant for competition authorities in some cases, including M&As.
  - Authorities should vigorously enforce prohibitions against “no-poaching” agreements.

### Competition policy implications and recommended responses
- Competition authorities should be increasingly vigilant when enforcing merger control:
  - Consider reconsidering thresholds used to determine jurisdiction to ensure authorities can review all relevant cases, including acquisitions by dominant firms of currently small but potentially large future competitors.
  - Conduct ex post analysis of past M&A control decisions to help adjust enforcement.
- Beyond M&A control:
  - Greater use of market investigations with appropriate remedies.
  - Active enforcement of prohibitions against abuse of dominant positions.
  - Authorities should assess firms’ power in input markets, including labor markets.
- Digital economy-specific measures:
  - Authorities need to be more responsive, including greater use of interim (conservatory) measures imposed before a final decision.
  - Develop specific expertise (for example, by building digital economy units).
  - Regulating digital firms like public utilities would be difficult given the fast-evolving nature of digital businesses and challenges of price regulation in a digital context.
- International cooperation:
  - Strengthen international cooperation among competition authorities via enhanced information exchange and expansion of international best-practice guidelines.
  - Cross-border cases and digital businesses necessitate stronger cooperation to avoid higher costs and uncertainty for firms and global spillovers from major authorities’ decisions.
- Resource and framework considerations:
  - Some competition authorities may need additional resources to implement these changes.
  - Changes may require significant adjustments to existing competition policy frameworks, although not an overhaul of the consumer welfare criterion.

*Source: sdnea2021001 - EXECUTIVE SUMMARY*

### 7.       The drivers and macroeconomic

### 7.       The drivers and macroeconomic implications of rising market power

### A. Cross-industry framing
- Three industries analyzed: technology, pharmaceuticals, and banking.  
- Different perceived drivers by industry:
  - Technology: winner-takes-all dynamics driven by network effects.
  - Pharmaceuticals: strong intellectual property rights, most prominently drug patents.
  - Banking: concentration—including by policy design due to financial stability considerations.
- Notes on industry definitions and data:
  - Industry analysis uses the Industry Classification Benchmark (ICB) and starts in 1995 due to sample-size considerations.
  - Financial industry includes bank and nonbank institutions; traditional market power measures such as markups may not be appropriate for banks, especially in the context of very low policy interest rates.

### B. Market Power in Technology: Rising through Market Share Reallocation
Findings
- Markups in the technology industry increased more than 30 percent between 1995 and 2016.
- Concentration increased more than 10 percent globally between 1995 and 2016, with sales by the top 4 firms accounting for about two-thirds of total sales by the top 20 firms in 2016.
- Profitability rose: average cash dividends, as a share of sales, increased from 0.7 to 3.5 percent.
- Decomposition of the markup increase (Melitz-Polanec decomposition) indicates:
  - The reallocation effect (high-markup firms gaining market share without increasing markups) accounts for about 95 percent of the total markup change between 1995 and 2016.
  - Since the start of this decade, incumbent firms increased their markups more substantially, accounting for 15 percent of the overall increase.
Implications
- The surge of dominant, high-markup firms cements industry leadership and raises concentration.
- Growth effects are ambiguous:
  - Short-run: market-share gains by more productive firms could be growth-enhancing.
  - Longer-run: dominance may reduce incentives to innovate, discourage competitors from innovating, reduce business dynamism, and slow overall economic growth.

Notes
- Technology industry defined as ICB = 9 (“Technology”) and subsector ICB = 953 (“Software & Computer Services”).
- The large markup increases persist even after controlling for intangible assets and overhead costs.

### C. Market Power in Pharmaceuticals: Rising within Incumbent Firms
Findings
- Pharmaceuticals is characterized by intellectual property rights (patents) that create monopoly rents to incentivize innovation and cover development costs; government subsidies also played a role in Covid-19 vaccine development.
- Since 1995:
  - Markups have increased almost 40 percent.
  - Concentration has risen almost 80 percent, with the ratio of sales by the top 4 firms to the top 20 firms climbing from roughly 50 to 90 percent.
  - Profitability doubled, with cash dividends accounting for more than 7.7 percent of sales by 2016.
- Decomposition of the markup increase shows:
  - Increased markups charged by incumbent firms account for more than half of the global increase in markups in the industry.
  - Market share reallocation toward high-markup firms accounts for only about a third.
  - Net entry of firms plays a comparatively minor role.
Implications and policy considerations
- The industry’s need for large operating profits to reward major innovations must be balanced against concerns that excessive rewards for minor incremental innovations can hurt consumers and stifle innovation.
- Specific concerns include:
  - Excessive rewards to minor incremental innovations.
  - “Patent thickets”: overly complicated patent setups requiring agreements with many parties to use a technology.
  - “Killer acquisitions”: acquisitions of patents or firms that are not commercialized to eliminate potential competition.
- Further work is needed on whether, and if so what, government intervention is warranted; policy responses are likely to vary across jurisdictions given cross-country variation in intellectual property rights regimes.

Notes
- Pharmaceuticals industry analysis uses ICB = 457 (“Pharmaceuticals & Biotechnology”).
- Large markup increases documented persist after controlling for intangible assets and overhead costs.

### D. Market Power in Banking: Mixed Signals
Context and trade-offs
- The case for unrestricted competition in banking is ambiguous:
  - Benefits of competition: lower cost of services, improved credit allocation.
  - Risks of competition: erosion of banks’ franchise value, increased risk-taking incentives, threats to financial stability, weakened incentives to invest in relationship lending, and reduced access to finance for firms reliant on relationship lending (e.g., SMEs).
  - Trade-off between competition and financial stability can become acute during financial crises.
Post-crisis dynamics and plausible offsets
- Post-global financial crisis effects could push in opposite directions:
  - Consolidation (stronger banks taking over failing banks) and increased fixed compliance costs from enhanced regulation could reduce competition.
  - Offsetting forces include greater participation of less-regulated nonbank financial institutions and the rise of fintech providers as competitors to banks.
Empirical indicators and interpretation
- A simple Lerner index (price minus marginal cost, expressed as a share of price) appears to show a marked increase in market power in advanced economies, especially after the global financial crisis.
- However, this upward trend in the Lerner index is largely a mechanical result of falling policy rates:
  - Since the global financial crisis, bank interest expenses have declined along with policy rates.
  - Interest income on loans has also declined, leaving net interest margins (NIMs) fairly flat.
  - Regression analysis shows that lower policy rates are significantly correlated with a higher Lerner index, particularly after the crisis.
- When the Lerner index is corrected for the effects of policy rates and cyclical factors, it no longer shows an increase; this holds for both the United States and European countries.
- Other indicators—concentration (HHI) and profitability measures—do not display an upward trend in advanced economies.
- Among emerging market and developing economies since the mid-2000s:
  - The Lerner index has oscillated around a similar level.
  - Market concentration has dropped.
  - Profits and income have remained relatively flat.
Methodological notes
- Lerner index estimation uses data on 67 countries (27 advanced and 40 emerging market and developing economies) between 2000 and 2017. Price captured by share of income to assets; marginal cost estimated from a trans-log cost function including deposits, wages, and other expenses.
- When interest rates are close to or below zero, the Lerner index becomes less informative because the ratio between interest paid and interest earned mechanically moves toward 0 and pushes the index toward 1.
Policy takeaways
- The policy response to the global financial crisis does not seem to have resulted in a structural increase in market power in banking.
- Policymakers should continue to monitor evolving competition in banking because:
  - Whether the (adjusted) Lerner index remains stable depends on how long rates stay at the effective lower bound and how banks respond.
  - Prolonged low rates increase pressure on bank profits and could trigger consolidation, potentially reducing competition.
  - Future markups will depend on the extent to which banks can overcome structural challenges such as the rise of fintech, which could weaken banks’ market power.

*International Monetary Fund*

### 19.      A small group of high-markup firms have increasingly entrenched market power. Firms

### 19.      A small group of high-markup firms have increasingly entrenched market power. Firms

### Market power entrenchment: markups and persistence
- Firms at the top decile of the markup distribution more than doubled their markups, while firms at the bottom half of the distribution increased theirs by just 5 percent (Figure 8, panel 1).
- The likelihood of persistently remaining a high-markup firm has increased since 2000:
  - If a firm has one of the top 10 percent highest markups, there is now almost an 85 percent chance this will still be the case the following year—up almost 10 percentage points since the second half of the 1990s (Figure 8, panel 2).
- Declining firm turnover among the group of high-markup firms suggests a growing lack of competitors for dominant firms.

### COVID-19 effects on concentration and market power
- The pandemic is expected to strengthen concentration and market power through sectoral bankruptcies that disproportionately affect small and medium-sized enterprises.
- Illustration using Orbis data across 21 countries:
  - No–COVID-19 scenario median concentration ratio (share of the top 4 firms in total sales of the top 20 firms within a 4-digit sector) would be about 56 percent.
  - With COVID-19 and reassignment of bankrupt firms’ sales to leading firms within their sector, this median concentration could rise by 4 percentage points, to 60 percent (Figure 9).
- The pandemic-driven shift to online activities is benefiting large technology companies and could further strengthen market concentration in their industries.

### Declining business dynamism associated with rising market power
- Cross-country Orbis analysis covering 28 economies since 2000 documents declines in:
  - Entry rate of new firms (Figure 10, panel 1).
  - Dispersion of growth rates across firms (Figure 10, panel 2).
  - Contribution of young firms (five years old or younger) to aggregate output (Figure 10, panel 3).
- Detailed empirical analysis finds a strong association between signs of falling business dynamism and rising market concentration at the country-industry level.
- For countries and industries with the 25 percent largest increases in concentration, rising concentration might explain about 10 percent of the overall decline in the share of output by young firms (roughly 0.9 percentage points).

### Role of M&As, especially by dominant firms
- M&As can generate efficiency gains and exits for entrepreneurs but can also reduce competition and weaken incentives to innovate.
- Country-industry analysis (Orbis merged with Zephyr M&A data) finds that business dynamism declines when leading firms conduct relatively more M&A deals (Figure 11):
  - A one standard deviation increase in the share of M&As conducted by leading firms is associated with slightly less than half a percentage point decline in the share of output accounted for by young firms—corresponding to a 3.5 percent fall relative to the average share of output accounted for by young firms.
  - Focusing on countries and industries with the 25 percent largest increases in acquisitions by leading firms, these estimates may explain about 3 percent of the overall decline in the share of output by young firms and about 9 percent of the decline in growth rate dispersion.
- These findings suggest a chilling, albeit moderate, effect of some M&As conducted by dominant firms on overall business dynamism and point to the need for competition policies to take a broad, dynamic perspective when reviewing deals.

### Effects of M&As on acquiring firms and competitors
- Firm-level analyses (Worldscope merged with SDC) show consistent findings:
  - Right after acquisition, the typical acquiring firm obtains 2.6 percent higher profitability and 3.6 percent higher markups (Figure 12).
  - A one standard deviation increase in a firm’s share of M&As is associated with about a 25 percent increase in sales and employment growth and a 10 percent rise in R&D growth relative to the respective average growth rates.
  - These pro-growth effects of M&As are greatly reduced if the acquiring firms are market leaders (in terms of sales within a given country-industry).
- Effects on competitors:
  - A one standard deviation increase in the share of deals by a firm’s competitor reduces the firm’s sales and R&D growth by about 4 percent, relative to the average sales and R&D growth rates (Figure 13).
  - The adverse effect on a firm’s growth is larger when the competitor is a market leader.
- US-specific analysis using Compustat confirms cross-country results:
  - A shift in M&A activities toward the largest firms in a sector is associated with increases in the Herfindahl-Hirschman Index (HHI), the share of the largest four firms (CR4), profit share of revenue, and markups, and with decreases in the dispersion of firms’ growth rates (Figure 14).

### Market power in labor markets
- Competition law and policy often cover buyer/monopsony power; labor markets have received renewed attention as an input market where firms may exercise monopsony power.
- Methods to measure firm labor market power include employer concentration in local labor markets, institutional/regulatory indicators (unions, collective bargaining, job protection), and a firm-specific labor supply elasticity approach (labor markdown), extending De Loecker and Warzynski (2012) using Orbis (see Online Annex III).
- Cross-country firm-level findings on labor market power (2000–2015):
  - While firms’ labor markdown does not seem to have risen across the board, a small fraction of already powerful firms slightly increased their labor market power between 2000 and 2015 (Figure 15, panel 1).
  - Labor market power trends are heterogeneous across industries: some evidence of an increase in manufacturing but a decline in finance and insurance.
  - Larger firms tend to have greater labor market power (Figure 15, panel 2); larger firms may pay higher wages in levels but pay less relative to workers’ marginal productivity.
  - There is evidence that a fraction of powerful firms in product markets are also able to suppress wages, and that powerful firms in product markets have gained more labor market power than other firms since the early 2000s.

*Source: IMF staff calculations based on Orbis, Zephyr, Worldscope, SDC, Compustat and analyses presented in the source content.*

### 29.      Adjustments to competition policy frameworks appear to be needed, tailored to

### sdnea2021001 - 29.      Adjustments to competition policy frameworks appear to be needed, tailored to

### Need for adjustments to competition policy frameworks
- Corporate market power has been rising, with adverse consequences for market dynamism and growth.
- Rising market power has weakened the effectiveness of monetary and fiscal policies (Boxes 1 and 2).
- Adjustments should be tailored to national circumstances and to specific challenges raised by the surge of the digital economy.
- Strengthening competition policy does not require abandoning fundamentals (for example, the consumer welfare criterion) but does require significant adjustments and a broad interpretation of existing welfare benchmarks (for example, maintaining high innovation incentives and strong industry dynamism).

### Existing competition laws and institutional differences
- More than 130 countries have adopted competition laws.
- Jurisdictional procedure differences:
  - European Union and Japan: competition authorities can make first-instance decisions to prohibit an M&A or impose a fine; these can be appealed to a court or tribunal.
  - United States and Canada: competition authorities investigate but must get a court or tribunal to prohibit an M&A or anticompetitive conduct.
- Substantive law differences (examples):
  - United States prohibits “monopolization.”
  - European Union prohibits “abuse of a dominant position” (may cover conduct like excessive pricing that is not captured under U.S. monopolization).
- Competition policy should be complemented by regulatory interventions, but regulators’ actions should be designed to avoid inadvertently strengthening established firms’ market power.

### Exceptional exemptions during extraordinary events (COVID-19 and similar)
- Authorities should remain flexible to allow temporary, exceptional exemptions to competition rules in extraordinary events.
- Examples of permissible exceptional measures:
  - Easing state-aid restrictions (European Union).
  - Tax exemptions for affected businesses, direct government grants, state guarantees for loans, safeguards for banks channeling aid, short-term export credit.
  - Guidance to firms on legitimate coordination to allocate scarce critical supplies (example: grocery sector in the United Kingdom).
- Design of such measures:
  - Should be exceptional (temporary) and aim to maintain some degree of competition.
  - Competition authorities can mitigate anticompetitive effects by:
    - Closely monitoring pricing of benefiting firms.
    - Working with consumer protection agencies to prevent abuse of dominant positions.
    - Providing specific guidance on legitimate coordination.

### A. Addressing rising market power across the economy — M&A enforcement
- Risks and challenges:
  - Growing risk of underenforcement in merger control.
  - M&As—especially involving leading firms—can harm competitors and weaken business dynamism, with risks for innovation and long-term growth.
  - Difficult-to-factor effects include impacts on product quality, nonprice parameters, and incentives for tacit collusion.
- Recommendations:
  - Bring test cases to clarify application of legislation where appropriate.
  - Ensure jurisdiction to review all M&As that may impact future competition, including some seemingly smaller deals.
- Jurisdictional thresholds:
  - Typical thresholds: revenue, value of acquisition, or market share.
  - Concern: revenue thresholds can be too high and miss relevant deals (Wollman 2019).
  - Revenue-based thresholds can miss high-value, low-current-revenue targets (examples: early-stage pharmaceutical firms with substantial intellectual property; digital businesses that have not yet monetized consumer engagement or data).
  - Responses taken by some countries:
    - Germany and Austria introduced new thresholds based on deal price.
  - Consider notifying competition authorities of series of small acquisitions by the same acquirer with significant cumulative impact.
  - Filings impose costs; thresholds should be proportional to risk of harm.
- Ex post evaluation:
  - Competition authorities would benefit from ex post analyses of previous M&A control decisions to identify ways to improve future enforcement and detect systematic errors (Lear 2019; Neumann and Sanderson 2007; Shapiro 2018).
  - Accepting some false positives ex post may be necessary to avoid too many false negatives.
- Labor-market considerations:
  - Authorities should assess corporate power in input markets, notably labor markets.
  - Enforcement should include vigorous action against “no-poaching” agreements.

- Notable factual examples and dates preserved from the source:
  - February 2020: US Federal Trade Commission (FTC) issued special orders to several large technology firms requiring information about previous acquisitions over the past 10 years not reported under the Hart-Scott-Rodino Act.
  - December 2020: FTC and a group of 48 Attorney Generals filed lawsuits alleging Facebook engaged in anticompetitive practices relating in part to its acquisitions of Instagram and WhatsApp.
  - Example legislative change: new German legislation allowing the competition authority to request that specific large businesses notify it of much smaller acquisitions in certain sectors.

### A. Addressing rising market power across the economy — Abuse of dominant positions and market investigations
- Enforcement of restrictions on abuse of a dominant position can mitigate effects of corporate market power and prevent entrenchment.
- Market investigations and sector inquiries:
  - Useful to analyze market functioning, identify anticompetitive behavior, and impose remedies or recommend legislative changes.
  - Example: United Kingdom Competition and Markets Authority can conduct in-depth market investigations that can lead to changes in firm conduct or requirements to divest parts of business.
  - Appropriate for markets transformed by technology and critical infrastructure (transport, energy, telecommunications) or markets not operating efficiently.
  - Market investigations can be resource-intensive.
- Resourcing:
  - Additional resources may be needed for more and deeper market and M&A investigations, new sectoral expertise, ex post evaluation, and stronger cooperation domestically and internationally.
  - Evidence suggests budgets of some competition authorities have not kept pace with economy size or number of M&As.
  - Example: during 1980–2018, combined budget of the Federal Trade Commission and the Department of Justice Antitrust Division was roughly halved as a share of US GDP; it fell even more (in real terms) compared with sharply higher total number of deals reported under the Hart-Scott-Rodino Antitrust Improvements Act.

### B. Responding to the rise of the digital economy — Challenges
- Digital-economy-specific pressures on regulators:
  - Need for quicker decisions in more complex and fast-moving environments.
  - Features: multiple-sided markets, zero-monetary-price services, ecosystem building, role of data, network effects, economies of scale and scope, winner-takes-all dynamics.
  - Data combined with algorithms and artificial intelligence increases risk of hard-to-detect anticompetitive collusion.
  - Difficulty assessing firm activity impact in innovative and fast-moving markets where future developments are hard to predict.
- Large online platforms:
  - Operate in multiple-sided markets with strong network effects and economies of scale that can lead to convergence toward a small number of platforms with large market shares.
  - Risks: abuse of dominant position harming users on one or both sides and potential entrants.
  - Policy focus: ensure multihoming and switching are viable; competitive dynamics vary across platforms—case-by-case approach needed.

### B. Responding to the rise of the digital economy — Enforcement and regulation mix
- Use a mixture of competition enforcement and regulatory intervention, drawing on experiences from utilities and telecommunications.
- Potential regulatory tools:
  - Interoperability requirements to allow entrants access to platform networks.
  - Measures to allow consumers to retain data when switching between platforms (data portability), analogous to Open Banking.
  - Flexible regulation forms: codes of conduct or targeted regulation to rebalance platform-participant relationships.
- Examples and precedents:
  - Telecommunications: incumbent network access requirements; number portability to retain mobile phone number when switching providers.
  - Australia and United Kingdom have considered or recommended divestments in some digital cases (Australian Competition and Consumer Commission 2019; Competition and Markets Authority 2019).
  - Codes of conduct: recommended by Australian Competition and Consumer Commission’s Digital Platforms Inquiry and the Furman Review.
  - European Union introduced a regulation on platform-to-business relations with new rules for fair and transparent processes for traders using digital platforms.

### B. Responding to the rise of the digital economy — Structural remedies and fragmentation
- Idea of regulating big tech as utilities or “breaking them up”:
  - Has gained attention but raises major challenges.
  - Vertical separation (network ownership vs. network use) used historically in utilities/telecommunications; potential analogies exist for platforms given high fixed costs and low marginal costs.
  - Divestments have been considered in Australia and the United Kingdom.
  - Caution: digital businesses provide complex and rapidly evolving services, making utility-style regulation and structural remedies potentially costly and complex.

### B. Responding to the rise of the digital economy — Data access, portability, and interoperability
- Access to data is a specific source of corporate market power amplified by increased processing power and algorithms.
- Enforcement actions:
  - Active enforcement of laws against abuse of dominance to prevent data-based market power from harming users.
- Policy tools:
  - Data portability: enable consumers to transfer their data to improve switching and competition, with limited privacy risks; precedent: Open Banking.
  - Data interoperability: enable software, devices, or systems to exchange data and integrate services to allow new entrants and encourage innovation.
  - Government may coordinate establishment of open standards in some cases.
- Design considerations:
  - Protect privacy and firms’ incentives to collect data and innovate.
  - Keep data protection obligations as simple as possible to avoid excessive compliance costs that could disproportionately harm smaller firms and act as barriers to entry in data-intensive markets.

*Source: sdnea2021001 - 29.      Adjustments to competition policy frameworks appear to be needed, tailored to*

### 46.      For competition authorities to move faster in rapidly evolving markets, it is also

### 46.      For competition authorities to move faster in rapidly evolving markets, it is also

### A. Knowledge, institutional tools, and merger review of nascent competitors
- Findings and concerns:
  - Competition authorities need extensive knowledge of the digital economy—including data science, key technologies, and market dynamics—to avert irreversible damage to competition.
  - In cases with substantial network effects and high returns to scale, foreclosure of new entrants by dominant incumbents can be difficult to remedy without timely intervention.
  - Authorities must consider the impact on future competition and innovation of acquisitions of businesses that have the potential to become competitors of the acquirer, even if they are not strong competitors at present (for example because they have not yet monetized their customer base or data).
- Institutional options and tools:
  - Set up dedicated digital market units to build required expertise.
  - Use interim measures—conservatory measures imposed on firms before the competition authority reaches a final decision regarding any specific conduct—to rapidly address potential harm.
  - Reconsider revenue-based thresholds to capture acquisitions of nascent competitors (relevance varies across jurisdictions).
- Examples and precedents cited:
  - The GDPR makes a general right for all personal data in the European Union.
  - Australia has introduced a consumer data right to enable consumers to transfer their data to trusted parties in banking and will extend this to other sectors.
  - The United Kingdom will launch a Digital Markets Unit, as called for in the Furman Review.
  - In October 2019, the European Commission imposed interim measures on Broadcom for TV and modem chipsets, prohibiting Broadcom from applying provisions the Commission prima facie considers infringement of EU competition rules until the Commission concludes its assessment on the substance of the case.

### B. Strengthening international cooperation
- Rationale:
  - As global markets become increasingly interconnected, competition authorities need to work together to avoid international fragmentation.
  - In the absence of a global competition authority, cooperation between national or regional competition authorities is needed for cross-border issues.
  - Without international cooperation, firms would face 130 different competition authorities with potentially divergent approaches and interests, raising firm costs and uncertainty.
  - A decision by a major authority to block or condition a multinational M&A will typically spill over to markets and regulators in other jurisdictions.
  - Data access issues have implications for trade policy, competition policy, and data protection and will require greater cooperation.
- Policy recommendations:
  - Competition authorities should coordinate more, including by sharing more information with one another on cross-border cases.
  - Develop and promote international best-practice guidelines to encourage homogeneous and consistent approaches by competition authorities.
  - Strengthen the role of international institutions (for example, Organisation for Economic Co-operation and Development; United Nations Conference on Trade and Development; International Competition Network) to facilitate development and expansion of best practices.
  - Support other efforts to improve international dialogue (example: Canada announced it will host an annual Digital Enforcement Summit).

### C. Risks of using competition policy for industrial objectives
- Findings:
  - International coordination and convergence to best practices will alleviate the risk of domestic competition policies being used to favor domestic producers over foreign competitors.
  - Domestic consumers would suffer higher prices, lower quality, and less choice if competition policy were used strategically to achieve industrial policy objectives (such as protecting large domestic firms).
  - A national industrial policy is more likely to succeed if accompanied by strong competition policy to maintain dynamic and efficient markets.
- Recommendation:
  - Maintain strong domestic competition authorities with comparable objectives to reduce the risk of protectionist or strategic misuse of competition policy.

### Box 1 — The role of market power for monetary policy transmission
- Model and empirical findings:
  - Strong market power dampens monetary policy transmission because customers’ demand for a high-markup firm does not vary much when its relative price changes, so output responds little to changes in input costs driven by monetary policy.
  - Profits of high-markup firms consist largely of economic rents, making them less sensitive to input cost changes and reducing incentives to adjust inputs when costs change.
  - High-markup firms’ financial buffers help shelter them from shifts in external funding conditions, enabling continued financing when cost of credit rises.
- Quantitative illustrations:
  - Simulated responses of US employment to a 100-basis-point decline in the nominal policy rate are shown under two scenarios in the IMF’s Global Integrated Monetary and Fiscal (GIMF) model: one with average markups at early 1980s levels and one with higher markups observed today. Higher markups render monetary policy less effective at stabilizing US business cycles.
  - Cross-country firm-level analysis (Orbis data) using the local projection method shows that, compared with the response of the average firm in their country and industry:
    - Low-markup firms’ output responds more to changes in central banks’ policy rates.
    - High-markup firms’ output responds less to policy rate changes.
  - The analysis finds high-markup firms have financial buffers that mitigate the impact of monetary policy actions.
  - Interactions with firm age: young low-markup firms respond more to monetary policy actions than young and old high-markup firms.
- Methodological notes:
  - Monetary policy shocks are computed as forecast errors of policy rates that cannot be explained by forecast errors for growth and inflation.
  - Analysis controls for country-sector-time and firm fixed effects and a wide range of firm-level characteristics other than markups.

### Box 2 — The role of market power for fiscal policy transmission
- Key finding:
  - Lower market competition weakens the transmission of targeted fiscal stimulus; sectoral output in the targeted sector responds more in more competitive product markets.
- Mechanism:
  - In a targeted stimulus (for example, infrastructure), demand for the targeted sector increases. With perfect competition, markups/prices do not respond and output increases substantially. With imperfect competition, markups/prices increase in response to higher demand, limiting output response.
- Empirical evidence (China 2009–10 stimulus):
  - The 2009–10 Chinese government stimulus amounted to RMB 4 trillion (more than 10 percent of 2008 GDP), mostly carried out by local governments and targeted to infrastructure and housing projects.
  - At the city level, in cities with a more competitive construction sector (proxied by the number of firms per 10,000 people), private investment in the construction sector responded more to local government stimulus spending.
  - An instrumental variable approach exploiting local entry barriers in the construction sector supports a causal interpretation: in response to the stimulus, private investment in construction increased more, whereas land prices increased less in cities with greater market competitiveness.
- Quantitative details:
  - Analysis used data for 334 Chinese municipalities, grouped into 50 bins and controlling for province fixed effects.
  - Construction sector competitiveness is measured as the log of the total number of construction and real estate firms per 1,000 people in each city.
  - Private real estate investment is constructed using the total investment of all private real estate and construction firms from 2008 to 2013.
  - Public infrastructure spending is the total investment of local government financing vehicles between 2008 and 2013.
- Policy implication:
  - Promoting competitiveness strengthens fiscal stimulus effectiveness and should inform stimulus design by accounting for supply-side heterogeneity across sectors and locations.

*IMF staff calculations and analysis as presented in the source document.*

---


_Source: https://www.imf.org/-/media/files/publications/sdn/2021/english/sdnea2021001.pdf_