## _wp13234

## Source details

**Canonical URL:** [_wp13234](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp13234.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp13234.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp13234.pdf.json)

---

### Introduction and motivation
- Foreign direct investment (FDI) increased: the yearly average ratio of the world FDI net inflows to gross domestic product increased from 0.7% in the 1980’s to 2.6% in the 2000’s.
- Cross-border mergers and acquisitions (M&A) represent more than half of FDI.
- Empirical literature shows mixed productivity spillovers from FDI; an overlooked channel is corporate governance improvements and spillovers to non-target local firms.
- Prior motivating findings:
  - Cross-border transactions generate valuation gains when the acquirer’s country has stronger investor protection than the target’s country.
  - Existing studies do not provide firm-level evidence of governance improvements for rival firms.

### Research question and empirical approach
- Main question: Does a change in corporate control after a cross-border M&A lead to changes in corporate governance of non-target firms in the same country and industry?
- Focus: strategic complementarity in governance choices between target and rival firms.
- Dependent variable: firm-level governance index GOV41 measured as the percentage of 41 attributes for which firms meet minimum acceptable requirements.
- Main explanatory variable: XVAL — mean transaction value of cross-border acquisitions in the target firm’s industry (two-digit SIC) as a fraction of the industry’s market capitalization.
- Sample and exclusions:
  - Firm-level governance data 2004-2008; cross-border M&A data 2001-2008 (SDC Platinum).
  - Exclude small deals and certain deal types; only deals with transaction value of at least one million U.S. dollars.
  - Exclude firms that were targets in cross-border M&As in the current year and past three years to focus on spillover effects.
- Empirical specification includes firm fixed effects (μ), year fixed effects (λ), firm-level controls (Z), and DMA (domestic M&A dummy); all explanatory variables lagged by one year; standard errors corrected for country-level clustering.

### Theoretical model (two-stage oligopoly)
- Structure:
  - Stage 1: outside shareholders choose firm-level governance g_i.
  - Stage 2: managers choose private benefits s_i and quantity q_i, conditional on governance choices and rivals’ decisions.
- Key features and mechanisms:
  - Managers’ compensation includes fixed wage, equity share times profits, and private benefits extracted from revenues.
  - Private benefits have a quadratic cost depending on governance g_i and revenues.
  - Private benefits decrease with the manager’s equity share and corporate governance; weaker governance leads to higher production (overproduction).
  - Outside shareholders face convex costs of improving governance modeled as k g_i^φ.
- Intuition:
  - Better governance of the target reduces its overproduction, reallocating rents to rivals and increasing rivals’ marginal benefit of improving governance.
  - Convex governance costs imply larger marginal benefits are more likely to induce governance upgrades when initial governance is low.

### Testable hypotheses
- Hypothesis 1: Following a cross-border M&A from a high governance acquirer firm, non-target local firms operating in the same industry as the target firm will improve their governance.
- Hypothesis 2: The positive spillover to governance is larger in more competitive industries.
- Hypothesis 3: Following a cross-border M&A from a high governance acquirer firm, the valuation of target and non-target local firms operating in the same industry as the target firm will increase.
- Hypothesis 4: Following a cross-border M&A from a high governance acquirer firm, the productivity of target and non-target local firms operating in the same industry as the target firm will increase.

### Data overview and key variables
- GOV41: 41 attributes divided into four subcategories (board, audit, anti-takeover provisions, compensation and ownership); index measures percentage of attributes meeting minimum acceptable requirements (100% = all 41).
- M&A data (SDC Platinum, 2001-2008): selected completed deals; sample: 29,748 M&A deals, of which 9,710 are cross-border with acquirers from 99 countries.
- XVAL: annual value of cross-border deals as a fraction of market capitalization in the target’s country-industry (two-digit SIC); XVAL can exceed 100%.
- Sample composition: 2,231 firms (target and non-target) from 22 developed countries during 2004-2008; firms represent about 71% of the market capitalization of these countries.
- Key country-level GOV41 examples: Canada 67.0%, United Kingdom 53.2%, Switzerland 50.4%; lowest: Greece 33.7%, Portugal 35.2%, Belgium 35.5%.
- Key XVAL examples: United Kingdom 15.0%, Norway 12.0%, the Netherlands 7.1%; Finland and Japan: nearly zero XVAL.
- Firm-level controls and definitions reported in Appendix B; ratio controls winsorized at upper and lower 1% levels.

### Main corporate governance findings
- Aggregate governance spillovers:
  - Cross-border M&As lead to significant positive governance spillovers within the target firm’s industry.
  - Results robust to controls for firm size, growth opportunities, leverage, tangibility, ownership structure, and inclusion of firm fixed effects.
  - Instrumental variables approach (IMP = import penetration) provides evidence consistent with causality from cross-border M&A to governance improvements; first-stage F-statistic above ten.
- Coefficient magnitudes and interpretation:
  - Baseline XVAL coefficient: 0.0053, statistically significant at the 5% level (Column (2) / Table 4).
    - Interpretation: a one standard deviation increase in XVAL is associated with an increase in non-target firms’ GOV41 of 0.3 percentage points (= 0.0053 × 0.581), about 14% (= 0.3 / 2.17) of the average annual increase in GOV41.
  - Triple-interaction result (Column (5) / Table 4): XVAL 0.0138 (p-value 0.02); XVAL × IP 0.0725 (statistically significant).
    - When acquirer country investor protection is better than target country, a one standard deviation increase in XVAL is associated with a 5 percentage points (= 0.581 × (0.0138 + 0.0725)) improvement in governance of non-target firms in the same industry.
  - Instrumental variables second-stage XVAL coefficient implies: a one standard deviation increase in XVAL is associated with a 0.7 percentage points increase in GOV41 (= 0.1011 × 0.071), about 33% of the average annual increase in GOV41.
- Heterogeneity:
  - Spillovers larger when acquirer country has better investor protection than target country (XVAL × IP positive and significant).
  - Spillovers more pronounced in more competitive industries (XVAL × HHI negative and significant; effect three times larger in perfectly competitive industries: 0.0165 vs. average 0.0053 in one specification).
- Causality and robustness:
  - First-differences regressions consistent with firm fixed effects; reverse first-differences test yields statistically insignificant coefficient on change in GOV41 (inconsistent with reverse causality).
  - Robustness: restricting to majority-stake deals, alternative industry classifications (38 Fama-French), and including U.S. target firms produce similar qualitative results (some interaction significance varies).

### Governance subcategories: mechanisms and evidence
- GOV41 subcategories: Board (24 attributes), Audit (3 attributes), Anti-takeover provisions (6 attributes), Compensation and ownership (8 attributes).
- Board subcategory:
  - XVAL × IP positive and significant; spillovers on board quality occur when acquirer country investor protection is better.
  - Product market competition effect on board quality stronger in competitive industries; magnitude similar to aggregate GOV41.
  - Example magnitude: in competitive industries when acquirer country investor protection is better, one standard deviation increase in XVAL associated with a 5.4 percentage points improvement in board governance.
- Audit and anti-takeover subcategories:
  - No evidence of statistically significant spillover effects for audit and anti-takeover provisions.
- Compensation and ownership subcategory:
  - XVAL positive and significant in several specifications; XVAL × IP positive and significant.
  - Spillovers strongest in perfectly competitive industries and when acquirer country investor protection is better.
- Overall interpretation:
  - Cross-border M&A activity generates positive spillovers on internal governance mechanisms (board, compensation and ownership) rather than external mechanisms (audit, anti-takeover).

### Valuation spillovers (market-to-book, MB)
- Estimation: firm fixed effects regressions of MB with year fixed effects; GMM Arellano-Bover/Blundell-Bond used.
- Key results (Table 8):
  - XVAL coefficients positive and highly significant across specifications (e.g., Column (2): XVAL 0.4895***).
  - Economic magnitudes:
    - A one standard deviation increase in XVAL associated with a 12.5% (= 0.3951 × 0.581 / 1.842) increase in firm valuation at the median (one reported calculation).
    - In perfectly competitive industries: one standard deviation increase in XVAL associated with a 31% (= 0.581 × 0.9955 / 1.842) increase in firm valuation (at the median).
  - Interaction findings:
    - XVAL × HHI negative and significant: valuation spillovers concentrated in competitive industries.
    - XVAL × IP sometimes negative or insignificant; differences in investor protections do not appear crucial for valuation spillovers after controlling for other factors.
- Robustness: first-differences specification yields consistent MB results.

### Productivity spillovers (PRODK)
- Dependent variable: PRODK = log(output / capital) where output = total sales and capital = PPE.
- Key findings (Tables 10 and 11):
  - XVAL positive and significant in firm fixed effects regressions (e.g., Column (3): XVAL 0.1171***).
  - XVAL × HHI negative and significant: productivity spillovers concentrated in competitive industries.
  - Quantitative magnitudes:
    - Perfectly competitive industry: a one standard deviation increase in XVAL associated with a 6.8% (= 0.581 × 0.1171) increase in productivity.
    - Industries with median product market competition: corresponding increase in productivity is 3.9%.
  - Robustness: first-differences specifications consistent with fixed-effects results.
- Interpretation: governance spillovers provide an important channel explaining valuation and productivity spillovers following cross-border M&A.

### Conclusion and policy-relevant implications
- Main conclusions:
  - Cross-border M&A activity in an industry is positively related to corporate governance improvements among non-target firms in the same industry.
  - Spillovers are stronger when acquirer country investor protection is stronger than the target’s and in more competitive industries.
  - Cross-border M&A generates positive valuation and productivity externalities to industry rivals; corporate governance improvements are an important channel.
  - Effects extend beyond the target firm to industry rivals, implying the international market for corporate control can promote improved governance practices globally.

### Appendix B — variable definitions and governance attributes (summary)
- Key variable definitions include GOV41, XVAL, SIZE, SGROWTH, LEVERAGE, CASH, CAPEX, MB, ROA, R&D, PPE, FXSALES, ANALYST, CLOSE, ADR, IO_TOTAL, DMA, IP, HHI, IMP, PRODK, EMPLOYEES.
- GOV41 disaggregated into 41 firm-level attributes across Panels:
  - Panel A Board: 24 attributes (examples: 1 All directors attended 75% of board meetings; 7 Chairman and CEO positions are separated).
  - Panel B Audit: 3 attributes (examples: 25 Consulting fees < audit fees; 26 Audit committee independent).
  - Panel C Anti-takeover provisions: 6 attributes (examples: 28 Single class, common; 32 No poison pill or shareholder-approved pill).
  - Panel D Compensation & ownership: 8 attributes (examples: 34 Directors subject to stock ownership requirements; 41 Repricing prohibited).

*Source: _wp13234*

### Appendix A .............................................................................................................

### _wp13234 - Appendix A .............................................................................................................

### Appendix listings
- Appendix A ..............................................................................................................................20
- Appendix B ..............................................................................................................................24

*Source: _wp13234 - Appendix A .............................................................................................................*

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

### _wp13234 - References .............................................................................................................

### Introduction and motivation
- Foreign direct investment (FDI) has risen: the yearly average ratio of the world FDI net inflows to gross domestic product increased from 0.7% in the 1980’s to 2.6% in the 2000’s.
- Cross-border mergers and acquisitions (M&A) represent more than half of FDI.
- Empirical literature shows mixed productivity spillovers from FDI; an overlooked channel is corporate governance improvements and spillovers to non-target local firms.
- Prior findings motivating this paper:
  - Cross-border transactions generate valuation gains when the acquirer’s country has stronger investor protection than the target’s country.
  - Existing studies do not provide firm-level evidence of governance improvements for rival firms.

### Research question and approach
- Main question: Does a change in corporate control after a cross-border M&A lead to changes in corporate governance of non-target firms in the same country and industry?
- Focus: strategic complementarity in governance choices between target and rival firms.
- Empirical strategy:
  - Dependent variable: firm-level governance index (GOV41) measured as the percentage of 41 attributes for which firms meet minimum acceptable requirements.
  - Main explanatory variable: cross-border M&A activity measured as the mean transaction value of cross-border acquisitions in the target firm’s industry (two-digit SIC) as a fraction of the industry’s market capitalization (XVAL).
  - Sample: firm-level governance data 2004-2008 and cross-border M&A data 2001-2008 (SDC Platinum), excluding small deals and certain deal types; only deals with transaction value of at least one million U.S. dollars.
  - Exclude firms that were targets in cross-border M&As in the current year and past three years to focus on spillover effects.

### Theoretical model (two-stage oligopoly model)
- Structure:
  - Stage 1: outside shareholders choose firm-level governance, g_i, given other firms’ governance choices.
  - Stage 2: managers choose private benefits s_i and quantity q_i, conditional on governance choices and rivals’ decisions.
- Key model features:
  - Managers’ compensation includes a fixed component, an equity share times profits, and private benefits extracted from revenues; private benefits have a quadratic cost depending on governance g_i and revenues.
  - Private benefits decrease with the manager’s equity share and corporate governance; weaker governance leads to higher production (overproduction).
  - Outside shareholders face convex costs of improving governance modeled as k g_i^φ where k is a scale parameter and φ captures convexity.
- Intuition:
  - Better governance of the target reduces its overproduction, reallocating rents to rivals and increasing rivals’ marginal benefit of improving governance.
  - The convex cost of improving governance implies larger marginal benefits are more likely to induce governance upgrades when initial governance is low.

### Testable hypotheses derived from the model
- Hypothesis 1: Following a cross-border M&A from a high governance acquirer firm, non-target local firms operating in the same industry as the target firm will improve their governance.
- Hypothesis 2: The positive spillover to governance is larger in more competitive industries.
- Hypothesis 3: Following a cross-border M&A from a high governance acquirer firm, the valuation of target and non-target local firms operating in the same industry as the target firm will increase.
- Hypothesis 4: Following a cross-border M&A from a high governance acquirer firm, the productivity of target and non-target local firms operating in the same industry as the target firm will increase.

### Data overview
- Governance index (GOV41):
  - 41 attributes divided into four subcategories: board, audit, anti-takeover provisions, and compensation and ownership.
  - Index measures percentage of attributes meeting minimum acceptable requirements; an index of 100% means adoption of all 41 provisions.
  - Source: RiskMetrics; firm-level governance data for 2004-2008 from Aggarwal, Erel, Ferreira, and Matos (2011).
- M&A data:
  - Source: SDC Platinum, 2001-2008.
  - Selected completed deals (public and private targets) for countries with firm-level governance data.
  - Exclusions: leveraged buyouts, spinoffs, recapitalizations, self-tenders, exchange offers, repurchases, minority stake purchases, acquisitions of remaining interest, privatizations.
  - Transaction value threshold: at least one million U.S. dollars.
  - Resulting M&A sample: 29,748 M&A deals, of which 9,710 are cross-border with acquirers from 99 countries.
  - Cross-border M&A activity variable (XVAL) = annual value of cross-border deals as a fraction of market capitalization in the target’s country-industry (two-digit SIC).
- Sample composition:
  - 2,231 (target and non-target) firms from 22 developed countries during 2004-2008.
  - Firms represent about 71% of the market capitalization of these countries.
  - Firm-level controls defined as ratios are winsorized at the upper and lower 1% levels.

### Key empirical findings
- Governance spillovers:
  - Cross-border M&As lead to significant positive governance spillovers within the target firm’s industry.
  - Results robust to controls for firm size, growth opportunities, leverage, tangibility, ownership structure, and inclusion of firm fixed effects.
  - Instrumental variables approach provides evidence consistent with causality running from cross-border M&A to governance improvements.
- Heterogeneity of effects:
  - Spillover larger when the acquirer firm comes from a country with better investor protection than the target country.
  - Spillover more pronounced in more competitive industries.
- Magnitude:
  - Improvement in governance in a non-target firm can be as high as 5 percentage points in perfectly competitive industries when the acquirer country investor protection is better than that of the target country.
  - 5 percentage points corresponds to twice the annual variation in the governance index (2.2%) or to adopting 2 (out of 41) additional governance provisions.
- Governance subcategories:
  - Cross-border M&A activity in an industry affects internal governance mechanisms (board structure and executive compensation) rather than external mechanisms (audit and anti-takeover provisions).
- Real effects:
  - Cross-border M&A activity in an industry is associated with higher valuation of non-target firms.
  - A one standard deviation increase in cross-border M&A activity is associated with a 12% increase in market-to-book relative to the sample median.
  - Significant productivity spillovers are found on target firms’ rivals following a cross-border M&A.

### Additional empirical notes
- Country-level GOV41 examples (average):
  - Canada 67.0%, United Kingdom 53.2%, Switzerland 50.4%.
  - Lowest GOV41: Greece 33.7%, Portugal 35.2%, Belgium 35.5%.
- Cross-border M&A activity (average XVAL) examples:
  - United Kingdom 15.0%, Norway 12.0%, the Netherlands 7.1%.
  - Finland and Japan: nearly zero XVAL.
- Robustness and scope:
  - Analysis focuses on non-U.S. target firms; robustness tests show consistent results when including U.S. target firms.
  - Industry aggregation uses two-digit SIC to minimize measurement error due to firm-level industry assignment variability.

*Source: _wp13234 - References .............................................................................................................*

### Appendix B for detailed variable definitions.

### Appendix B for detailed variable definitions

### Measurement of key variables
- Investor protection (IP)
  - Constructed using the World Economic Forum’s country-level index of protections to minority shareholders.
  - For each cross-border M&A, calculate the difference between the acquirer’s and the target’s country index.
  - Create a dummy variable (IP) that equals one when the average (value-weighted using transaction value as weights) across cross-border M&A deals of this difference in each country-industry is positive, and zero otherwise.
  - The WEF measure ranges from 1 (not protected at all) to 7 (fully protected) and is time-varying.
- Product market competition (HHI)
  - Measured using the Herfindahl-Hirschman index (HHI) based on sales for each country and industry.
  - A low HHI indicates low concentration and potentially a more contestable and competitive industry.
- XVAL
  - XVAL calculation uses the market capitalization of public companies that operate in the same industry as the target firm. XVAL can be above 100% in industries with a significant number of private companies.
  - XVAL is defined as the value of cross-border M&As in firm i’s country-industry as a fraction of the market capitalization in year t-1.
- Other variables and controls
  - GOV41it: the corporate governance index for firm i in year t.
  - DMA: dummy equal to one if there is a domestic M&A in a firm’s industry.
  - Z: set of firm-level controls (examples reported include IO_TOTAL and CLOSE).
  - Standard errors corrected for country-level clustering.
  - All explanatory variables are lagged by one year.

### Baseline regression framework
- Baseline firm-year panel regression (equation stated in source):
  - 11141 = , itititititit GOVXVALZDMAu   + + + + + (1)
  - Where GOV41it is the dependent variable and XVALit-1 is the main explanatory variable; the regression includes firm fixed effects (μ), year fixed effects (λ), firm-level controls (Z), and DMA.
- Hypotheses tested
  - Hypothesis 1: Spillover effect larger if acquirer’s country has better investor protection than host country (interaction XVAL x IP).
  - Hypothesis 2: Spillover effect depends on product market competition in the target firm’s industry (interaction XVAL x HHI).

### Main corporate governance findings (Section A)
- Column (2) (including XVAL)
  - Estimated XVAL coefficient: 0.0053, statistically significant at the 5% level.
  - Interpretation: a one standard deviation increase in XVAL is associated with an increase in non-target firms’ governance index of 0.3 percentage points (= 0.0053 × 0.581), which represents about 14% (= 0.3 / 2.17) of the average annual increase in GOV41.
- Interaction with investor protection (Column (3))
  - XVAL coefficient: positive and significant at the 10% level.
  - XVAL x IP coefficient: positive and significant at the 5% level.
  - Interpretation: governance spillovers are concentrated in cross-border deals where the acquirer country has better investor protection than the target country; the interaction term coefficient is more than twice larger than the XVAL coefficient in column (2).
- Interaction with industry concentration (Column (4))
  - XVAL coefficient: positive and significant.
  - XVAL x HHI coefficient: negative and significant.
  - Interpretation: governance spillovers concentrated where the target firm operates in an industry with low concentration (i.e., high competition). The spillover effect is economically three times larger in perfectly competitive industries (0.0165 in column (4)) than the average effect (0.0053 in column (2)).
- Triple-interaction specification (Column (5))
  - XVAL coefficient: 0.0138 with a p-value of 0.02.
    - Indicates a significant governance spillover in perfectly competitive industries even when IP = 0.
  - XVAL x IP coefficient: 0.0725 and statistically significant.
    - Interpretation: when acquirer country investor protection is better than target country, a one standard deviation increase in XVAL is associated with a 5 percentage points (= 0.581 × (0.0138 + 0.0725)) improvement in governance of non-target firms in the same industry.
  - XVAL x HHI and the triple interaction coefficients: negative and significant, indicating the spillover effect is significantly lower in less competitive industries.
- Instrumental variables (Column (6))
  - Instrument: import penetration (IMP) = ratio of annual imports to sales in a given country-industry.
  - First-stage: positive relation between XVAL and IMP; F-statistic above ten (instruments not weak).
  - Second-stage XVAL coefficient: 0.1011 (implied by the narrative).
    - Interpretation: a one standard deviation increase in XVAL is associated with an increase in non-target firms’ governance index of 0.7 percentage points (= 0.1011 × 0.071), which represents about 33% of the average annual increase in GOV41.
- Direction of causality tests
  - First-differences regressions (Table 5): results consistent with firm fixed effects; changes in M&A activity lead to subsequent governance changes.
  - Reverse first-differences test: change in GOV41 as explanatory variable for change in XVAL yields statistically insignificant coefficient on change in GOV41 (inconsistent with reverse effect).
- Robustness checks
  - Restricting XVAL to cross-border M&As where acquirer seeks to own more than 50%: results similar to Table 4 (Panel A, Table 6).
  - Alternative industry classification (38 Fama-French industries): results similar to Table 4 except XVAL x IP interaction no longer significant though sign preserved (Panel B, Table 6).
  - Including U.S. target firms: XVAL coefficient higher than in Table 4; governance spillovers also occur when non-U.S. firm acquires U.S. firm; effect stronger when acquirer country has better investor protection than the United States; product market competition does not affect magnitude of spillover for U.S. sample (Panel C, Table 6).
- Overall conclusion from Section A
  - Cross-border M&A activity in an industry generates positive corporate governance spillovers to other firms in the same industry.
  - Spillovers are more pronounced when the foreign direct investment originates from a country with strong investor protection and when product market competition is more intense in the target firm’s industry.

### Corporate governance subcategories (Section B introduction)
- GOV41 index comprises 41 firm-level governance attributes.
- The document proceeds to examine these 41 attributes (Table B2 referenced for detailed results).

*Source: Appendix B for detailed variable definitions from the provided IMF content unit.*

### Appendix shows the list of attributes. The attributes cover four broad subcategories: (i) Board

### _wp13234 - Appendix shows the list of attributes. The attributes cover four broad subcategories: (i) Board

### Governance subcategories and regression evidence
- Attributes cover four broad subcategories: (i) Board (24 attributes), (ii) Audit (3 attributes), (iii) Anti-takeover provisions (6 attributes), and (iv) Compensation and ownership (8 attributes).
- Empirical strategy: estimate corporate governance spillover effects generated by cross-border M&A separately for each governance subcategory.
- Board subcategory (Panel A, Table 7):
  - XVAL is insignificant in column (1).
  - In column (2) the XVAL x IP coefficient is positive and significant, indicating positive spillovers on board quality of non-target firms when the acquirer country investor protection is better than that of the target country.
  - Column (3) shows product market competition effect is statistically insignificant.
  - Column (4), including XVAL x IP, XVAL x HHI, and XVAL x HHI x IP, shows the magnitude of the spillover to board attributes is strongest when the acquirer country investor protection is better than the target’s and in competitive industries.
  - The magnitude of the spillover effects is similar to those in Table 4 for the aggregate variable GOV41.
  - Example magnitude: the board subcategory spillover effect in more competitive industries when the acquirer country investor protection is better than the target country investor protection (XVAL x IP coefficient in column (4)), a one standard deviation increase in XVAL is associated with a 5.4 percentage points improvement in governance of non-target firms that operate in the same industry as the target firm.
- Audit and anti-takeover subcategories (Panels B and C, Table 7):
  - No evidence of statistically significant spillover effects of cross-border M&A activity for these two subcategories.
- Compensation and ownership subcategory (Panel D, Table 7):
  - XVAL is positive and significant in column (1), consistent with a positive spillover effect.
  - In column (2) XVAL is positive and significant at the 10% level, indicating significant spillovers even when the acquirer country has weaker investor protection than the target country.
  - The XVAL x IP interaction is positive and significant, indicating the effect is significantly stronger when the acquirer country has stronger investor protection than the target country.
  - Column (3) shows significant spillovers regardless of product market competition intensity (positive and significant XVAL).
  - Column (4) (including interactions) shows the magnitude of the spillover to compensation and ownership attributes is strongest in perfectly competitive industries and when the acquirer country investor protection is better than the target country investor protection.
- Overall interpretation:
  - Cross-border M&A activity in an industry generates positive spillover effects on board, compensation, and ownership of non-target firms in the same industry as the target.
  - Effects are not important for audit and anti-takeover provisions.
  - Cross-border M&A mainly affects internal governance mechanisms rather than external governance mechanisms.
  - Rationale: external governance mechanisms such as anti-takeover provisions may not be as important outside of the U.S. for lack of an active market for corporate control.

### Valuation results (Section C)
- Hypothesis tested: acquisitions by foreign rivals lead to corporate governance improvements and valuation gains to shareholders of non-target firms in the same industry (Hypothesis 3).
- Dependent variable: market-to-book ratio (MB); explanatory variable of interest: XVAL.
- Estimation: firm fixed-effects regressions of MB with year fixed effects; standard errors clustered by country; GMM using Arellano-Bover/Blundell-Bond estimator with forward orthogonal deviations.
- Key findings (Table 8):
  - Column (2): XVAL coefficient is positive and significant, indicating cross-border M&A activity generates positive valuation spillovers to non-target firms.
  - Economic magnitude: a one standard deviation increase in XVAL is associated with a 12.5% (= 0.3951  0.581 / 1.842) increase in firm valuation at the median of the data.
  - Column (3): positive difference in investor protection between acquirer and target country does not increase the non-target firm valuation spillover effect; XVAL x IP coefficient is negative (inconsistent with theory) but its significance disappears when adding more controls in column (5).
  - Column (4): including HHI and XVAL x HHI shows XVAL positive and significant and XVAL x HHI negative and significant — valuation spillover effect concentrated in competitive industries.
  - Perfectly competitive industries: one standard deviation increase in XVAL is associated with a 31% (= 0.581  0.9955 / 1.842) increase in firm valuation (at the median).
- Robustness: first-differences specification (Table 9) yields consistent results; some interaction effects stronger in column (5) of Table 9 relative to column (5) of Table 8.
- Short summary: cross-border M&A activity generates substantial positive valuation externalities to other firms in the same industry; product market competition is an important condition; differences in investor protections do not seem crucial for valuation spillovers.

### Productivity results (Section D)
- Hypothesis tested: cross-border M&A activity generates productivity spillovers to non-target firms in the same industry (Hypothesis 4).
- Dependent variable: capital productivity PRODK = log(output / capital); output = total sales; capital = property, plant and equipment (PPE).
- Controls: log number of workers (EMPLOYEES) and same control set as Table 8; all explanatory variables lagged by one year.
- Estimation: firm fixed effects regressions of PRODK (Table 10) and robustness with first-differences (Table 11).
- Key findings:
  - Columns (2) and (3): XVAL included; IP and XVAL x IP coefficients insignificant in both cases.
  - Columns (4) and (5): including HHI and interactions shows XVAL positive and significant; XVAL x HHI negative and significant — productivity spillovers concentrated in competitive industries.
  - Quantitative magnitudes:
    - Perfectly competitive industry: a one standard deviation increase in XVAL is associated with a 6.8% (= 0.581  0.1171) increase in productivity.
    - Industries with median product market competition: corresponding increase in productivity is 3.9%.
- Robustness: first-differences specifications (Table 11) are consistent with firm fixed effects.
- Interpretation: evidence supports significant positive firm valuation and productivity spillover effects following cross-border M&A; corporate governance spillovers are an important channel to explain valuation and productivity spillovers.

### Conclusion (Section V)
- Main findings:
  - Positive relation between cross-border M&A activity in an industry and corporate governance of non-target firms in the same industry.
  - Cross-border M&A is especially effective in improving corporate governance in more competitive industries and when acquirer country investor protection is stronger than the target’s.
  - Cross-border M&A leads to increases in firm valuation and productivity of non-target firms, indicating FDI affects corporate governance and real firm outcomes.
  - The paper establishes that effects spread beyond the target firm to the target’s industry rivals, generating positive externalities.
  - Market forces, namely the international market for corporate control, can promote good corporate governance practices worldwide.

### Appendix A — model intuition and testable hypotheses
- Model setup:
  - Industry oligopoly with N identical firms; two-stage game: (1) outside shareholders choose firm-level governance g; (2) managers choose private benefits s_i and quantity q_i.
  - Manager’s compensation includes fixed wage, equity share, and private benefits; private benefits cost modeled as quadratic and depend on governance.
  - Demand specification with negative unit elasticity (convenient analytically).
- Key model results:
  - Managers choose higher private benefits when manager equity share is lower and governance is weaker.
  - Firms produce more when governance is weaker (overproduction externality).
  - Outside shareholders choose governance balancing profits net of wages and governance improvement costs.
  - Higher costs of adopting stronger governance (higher parameter ϕ) → governance decreases and private benefits increase.
  - More ex ante competition (higher parameter θ) decreases ability to extract private benefits and thus decreases incentives to improve governance.
- Spillover effects from a cross-border M&A:
  - Model scenario: target firm “f” is acquired by an acquirer that imposes better governance (lower g_f); local firms labeled “l”.
  - Result 1: Better governance imposed on the target firm (lower g_f) leads to lower private benefits in remaining local firms (lower s_l).
  - Testable Hypothesis 1: Following a cross-border M&A from a high governance acquirer firm, non-target local firms in the same industry are expected to improve their governance.
  - Result 2: Higher industry competition increases the responsiveness of governance in the industry to the cross-border M&A.
  - Testable Hypothesis 2: Following a cross-border M&A from a high governance acquirer firm, the positive spillover to governance of non-target local firms is expected to be higher in industries with high competition.

*Source: Appendix and Sections C–V of the provided content unit.*

### APPENDIX B

### _wp13234 - APPENDIX B

### Variable definitions
- GOV41: Firm-level index of corporate governance based on 41 attributes on board, audit, anti-takeover provisions, and compensation and ownership structure (RiskMetrics). See Table B2 for details.
- XVAL: Value of cross-border M&As (SDC Platinum) divided by market capitalization (Worldscope item 08001) in a country-industry (two-digit SIC level).
- SIZE: Log of total assets in thousands of U.S. dollars (Worldscope item 02999).
- SGROWTH: Two-year  geometric average of annual growth rate in sales in U.S.dollars (Worldscope item 01001).
- LEVERAGE: Total  debt (Worldscope item 03255) divided by total assets (Worldscope item 02999).
- CASH: Cash and short-term investments (Worldscope item 02001) divided by total assets (Worldscope item 02999).
- CAPEX: Capital expenditures (Worldscope item 04601) divided by total assets (Worldscope item 02999).
- MB: Market value of equity (Worldscope item 08001) divided by book value of equity (Worldscope item 03501).
- ROA: Ratio of net income before extraordinary items (Worldscope item 01551) plus interest expenses (Worldscope  item 01151) to total assets (Worldscope item 02999).
- R&D: Research and development expenditures (Worldscope item 01201) divided by total assets (Worldscope item 02999).
- PPE: Property, plant, and equipment (Worldscope item 02501) divided by total assets (Worldscope item 02999).
- FXSALES: International sales (Worldscope item 07101) as a proportion of sales (Worldscope item 01001).
- ANALYST: Number of analysts following a firm (IBES).
- CLOSE: Number of shares held by insiders (shareholders who hold five percent or more of the outstanding shares, such as officers, directors, and immediate families, other corporations or individuals), as a fraction of the number of shares outstanding (Worldscope item 08021).
- ADR: Dummy that equals one if a firm is cross-listed on a U.S. exchange through a level 2-3 ADR or ordinary listing, and zero otherwise (major depositary institutions and U.S. stock exchanges).
- IO_TOTAL: Holdings byinstitutional investors as a fraction of market capitalization (FactSet/LionShares).
- DMA: Dummy that takes the value one if there is at least one domestic M&A deal in a country-industry.
- IP: Dummy variable that takes the value of one if the value-weighted average (weights are value of transaction) difference between acquirer and target’s country-level strenght of investor protection (World Economic Forum) in a given country-industry (two-digit SIC) is positive, and zero otherwise.
- HHI: Herfindahl-Hirschman index of the market shares in a country-industry using sales (Worldscope item 01001).
- IMP: Import penetration, defined as imports (United Nations Comtrade) over sales (Worldscope item 01001) in a country-industry (two-digit SIC).
- PRODK: Log of the ratio of sales (Worldscope item 01001) to property, plant, and equipment (Worldscope item 02501).
- EMPLOYEES: Log of total number of employees (Worldscope item 07011).

### Firm-Level Governance Attributes (Table B2)
Panel A: Board
- 1 All directors attended 75% of board meetings or had a valid excuse
- 2 CEO serves on the boards of two or fewer public companies
- 3 Board is controlled by more than 50% independent outside directors
- 4 Board size is at greater than five but less than sixteen
- 5 CEO is not listed as having a related-party transaction
- 6 Compensation committee composed solely of independent outsiders
- 7 Chairman and CEO positions are separated, or there is a lead director
- 8 Nominating committee composed solely of independent outsiders
- 9 Governance committee exists and met in the past year
- 10 Shareholders vote on directors selected to fill vacancies
- 11 Governance guidelines are publicly disclosed
- 12 Annually elected board (no staggered board)
- 13 Policy exists on outside directorships (four or fewer boards is the limit)
- 14 Shareholders have cumulative voting rights
- 15 Shareholder approval is required to increase/decrease board size
- 16 Majority vote requirement to amend charter/bylaws (not supermajority)
- 17 Board has the express authority to hire its own advisers
- 18 Performance of the board is reviewed regularly
- 19 Board-approved succession plan in place for the CEO
- 20 Outside directors meet without CEO and disclose number of times met
- 21 Directors are required to submit resignation upon a change in job
- 22 Board cannot amend bylaws without shareholder approval or can do so only under limited circumstances
- 23 Does not ignore shareholder proposal
- 24 Qualifies for proxy contest defenses combination points

Panel B: Audit
- 25 Consulting fees paid to auditors are less than audit fees paid to auditors
- 26 Audit committee composed solely of independent outsiders
- 27 Auditors ratified at most recent annual meeting

Panel C: Anti-takeover provisions
- 28 Single class, common
- 29 Majority vote requirement to approve mergers (not supermajority)
- 30 Shareholders may call special meetings
- 31 Shareholders may act by written consent
- 32 Company either has no poison pill or a pill that is shareholder approved.
- 33 Company is not authorized to issue blank check preferred

Panel D: Compensation and ownership
- 34 Directors are subject to stock ownership requirements
- 35 Executives are subject to stock ownership guidelines
- 36 No interlocks among compensation committee members
- 37 Directors receive all or a portion of their fees in stock
- 38 All stock-incentive plans adopted with shareholder approval
- 39 Options grants align with company performance and reasonable burn rate
- 40 Officers’ and directors’ stock ownership is at least 1% but not over 30% of total shares outstanding
- 41 Repricing prohibited

*Source: _wp13234 - APPENDIX B*

### REFERENCES

### _wp13234 - REFERENCES

### References cited
- Acharya, V., and P. Volpin, 2010, “Corporate Governance Externalities,” Review of Finance 14, pp. 1-33.
- Aggarwal, R., I. Erel, R. Stulz, and R. Williamson, 2009, “Differences in Governance Practices between U.S. and Foreign Firms: Measurement, Causes, and Consequences,” Review of Financial Studies 22, pp. 3131-3169.
- Aggarwal, R., I. Erel, M. Ferreira, P. Matos, 2011, “Does Governance Travel Around the World? Evidence from institutional investors,” Journal of Financial Economics 100, pp. 154-181.
- Aitken, B., and A. Harrison, 1999, “Do Domestic Firms Benefit from Direct Foreign Investment? Evidence from Venezuela,” American Economic Review 89, p. 605-618.
- Albuquerque, R., and N. Wang, 2008, “Agency Conflicts, Investment, and Asset Pricing,” Journal of Finance 63, pp. 1-40.
- (Additional citations continue as listed in the source content, including works by Bloomberg, Bris & Cabolis, Cheng, Dicks, Djankov et al., Doidge et al., Fernandes, Giroud & Mueller, Gompers et al., Haddad & Harrison, Harford, Haskel et al., Javorcik, La Porta et al., Levit & Malenko, Martynova & Renneboog, OECD, Philippon, Rossi & Volpin, Servaes & Tamayo, Shleifer & Wolfenzon, Syverson.)

### Table 1 — Corporate Governance and Cross-Border M&A by Target Country: key statistics
- Sample: Worldscope firms with GOV41 data in 2004-2008.
- Total: Number of firms 2,223; Number of observations 9,817; GOV41 mean 0.460; GOV41 standard deviation 0.108; XVAL mean 0.050; XVAL standard deviation 0.581.
- Country examples (Number of firms, Number of observations, GOV41 mean, GOV41 sd, XVAL mean, XVAL sd):
  - Australia: 119, 495, 0.472, 0.064, 0.036, 0.134
  - Canada: 188, 797, 0.670, 0.073, 0.060, 0.405
  - Japan: 599, 2,845, 0.377, 0.032, 0.001, 0.013
  - United Kingdom: 520, 2,088, 0.532, 0.075, 0.150, 1.201

### Table 2 — Corporate Governance and Cross-Border M&A by Industry: key statistics
- Sample: Worldscope non-target firms with GOV41 in 2004-2008.
- Panel A (Top ten industries by XVAL) — examples:
  - Transportation services: observations 68, GOV41 mean 0.452, GOV41 sd 0.076, XVAL mean 2.058, XVAL sd 6.304
  - Construction-special trade contractors: 13, GOV41 mean 0.454, GOV41 sd 0.076, XVAL mean 1.095, XVAL sd 2.640
  - Primary metal industries: 216, GOV41 mean 0.434, GOV41 sd 0.089, XVAL mean 0.113, XVAL sd 0.682
- Panel B (Bottom ten industries by XVAL) — examples:
  - Petroleum refining & related industries: observations 78, GOV41 mean 0.485, GOV41 sd 0.145, XVAL mean 0.000, XVAL sd 0.001
  - Food Stores: 149, GOV41 mean 0.461, GOV41 sd 0.117, XVAL mean 0.003, XVAL sd 0.010

### Table 3 — Summary statistics (sample: Worldscope firms with GOV41 in 2004-2008)
- Selected variables (Mean, Median, Standard deviation, Minimum, Maximum, Number of observations):
  - GOV41: 0.460, 0.439, 0.108, 0.195, 0.878, 9,817
  - XVAL: 0.050, 0.001, 0.581, 0.000, 28.672, 9,816
  - SIZE: 14.598, 14.579, 2.237, 5.347, 22.052, 9,221
  - MB: 2.501, 1.842, 2.444, -1.870, 15.969, 9,418
  - ROA: 0.047, 0.054, 0.126, -1.766, 0.325, 9,095
  - R&D: 0.014, 0.000, 0.044, 0.000, 0.629, 9,817
  - IO_TOTAL: 0.214, 0.147, 0.233, 0.000, 1.000, 9,817
  - IMP: 1.172, 0.000, 3.852, 0.000, 27.226, 10,069
  - PRODK: 1.367, 1.350, 1.415, -2.581, 5.038, 10,264

### Table 4 — Cross-Border M&A and Non-Target Corporate Governance: Firm Fixed Effects (selected coefficient estimates)
- Dependent variable: GOV41 of non-target firms; explanatory: XVAL (lagged one year); sample: Worldscope firms 2004-2008; robust p-values adjusted for country-level clustering in brackets; *, **, *** denote 10%, 5%, 1% significance.
- Coefficients on XVAL across columns:
  - Column (1): XVAL 0.0053** (0.031)
  - Column (2): XVAL 0.0026* (0.095)
  - Column (3): XVAL 0.0165** (0.035)
  - Column (4): XVAL 0.0138** (0.021)
  - Column (5): XVAL 0.1011** (0.046) — Column (6) reports IV with IMP as instrument (observations 7,293; R-squared not reported)
- Notable interaction terms:
  - XVAL × IP: 0.0123** (0.011) in one specification; 0.0725*** (0.009) in another
  - XVAL × HHI: -0.0240** (0.049); -0.0253** (0.048)
  - XVAL × HHI × IP: -0.1001** (0.028)
- Control variable examples:
  - SIZE: -0.0056* (0.097) to -0.0045 (0.225)
  - ROA: 0.0200*** (0.010) to 0.0157** (0.011)
  - CLOSE: -0.0142*** (0.007) to -0.0141*** (0.001)
  - IO_TOTAL: 0.0253** (0.024) to 0.0176** (0.014)
- Number of observations ranges: 7,293–7,398; R-squared ~0.875–0.876.

### Table 5 — First Differences regressions (GOV41)
- Dependent variable: changes in GOV41; regressions include country, industry and year fixed effects.
- ∆XVAL coefficients:
  - Column (1): ∆XVAL 0.0047*** (0.008)
  - Column (2): ∆XVAL 0.0026* (0.062)
  - Column (3): ∆XVAL 0.0143* (0.061)
  - Column (4): ∆XVAL 0.0148** (0.010)
- Interaction ∆(XVAL × IP): 0.0117** (0.047); 0.0579* (0.069) in another.
- ∆SIZE: -0.0087*** (0.001) across specifications.
- ∆ROA: 0.0168** (0.024) to 0.0178** (0.029).
- Number of observations: 5,473–5,529; R-squared ~0.209–0.212.

### Table 6 — Robustness (Majority stakes, 38 Fama-French industries, Sample including U.S. firms)
- Panel A (Majority stakes): XVAL coefficients e.g., 0.0053** (0.032); XVAL × IP 0.0116** (0.013); XVAL × HHI -0.0236* (0.060); XVAL × HHI × IP -0.0958** (0.033). Observations up to 7,398; R-squared 0.876.
- Panel B (38 Fama-French industries): XVAL 0.0135*** (0.002); XVAL × HHI -0.1086*** (0.009). Observations ~7,368–7,397; R-squared 0.875–0.876.
- Panel C (Including U.S. firms): XVAL 0.0073*** (0.001); XVAL × IP 0.0138*** (0.000); observations up to 24,749; R-squared 0.920–0.921.

### Table 7 — GOV41 Subcategories (Board, Audit, Anti-takeover, Compensation & Ownership)
- Panel A (Board): XVAL × IP 0.0155** (0.028); XVAL × HHI -0.0517*** (0.009); XVAL × HHI × IP -0.0827* (0.054). Number of observations 7,384–7,398; R-squared ~0.892–0.893.
- Panel B (Audit): IP -0.0297** (0.033) in one specification; HHI × IP 0.0548** (0.032). R-squared ~0.681–0.682.
- Panel C (Anti-takeover provisions): no statistically significant XVAL coefficients reported; R-squared ~0.785–0.786.
- Panel D (Compensation and ownership): XVAL 0.0170** (0.034) to 0.0287** (0.026); XVAL × IP 0.0305** (0.016); XVAL × HHI × IP -0.2468** (0.025). R-squared ~0.736–0.737.

### Table 8 — Non-Target Firm Valuation (MB): Firm Fixed Effects (GMM Arellano-Bover/Blundell-Bond)
- Dependent variable: MB; sample: Worldscope firms with GOV41 2004-2008.
- XVAL coefficients:
  - Column (1): XVAL 0.3951*** (0.000)
  - Column (2): XVAL 0.4895*** (0.000)
  - Column (3): XVAL 0.9955*** (0.000)
  - Column (4): XVAL 0.8633*** (0.005)
- Interaction XVAL × IP: -0.5557*** (0.002) in one specification.
- HHI: 0.3003** (0.027); XVAL × HHI -1.3503*** (0.000) in one specification.
- Control examples:
  - SIZE: -0.0720** (0.030) to -0.0538** (0.026)
  - LEVERAGE: 0.7959*** (0.002) to 0.6583*** (0.003)
  - CASH: 0.9899** (0.010) to 0.9154*** (0.008)
  - R&D: 2.0026** (0.011) to 1.8109*** (0.004)
- Number of observations: 6,843–6,863.

### Table 9 — MB: First Differences
- ∆XVAL coefficients:
  - Column (1): ∆XVAL 0.3851*** (0.003)
  - Column (2): ∆XVAL 0.4726*** (0.001)
  - Column (3): ∆XVAL 0.8951*** (0.000)
  - Column (4): ∆XVAL 0.8754*** (0.000)
- ∆(XVAL × IP) -0.5946*** (0.002) in one specification.
- ∆ HHI: -0.9062** (0.025); ∆(XVAL × HHI) -1.1295*** (0.001).
- ∆CASH: 1.0643** (0.048) to 1.0919** (0.041).
- Number of observations: 5,343–5,378; R-squared ~0.131–0.137.

### Table 10 — Productivity of Capital (PRODK): Firm Fixed Effects
- Dependent variable: PRODK = log(sales/capital); sample: Worldscope firms 2004-2008.
- XVAL coefficients:
  - Column (3): XVAL 0.1171*** (0.001)
  - Column (4): XVAL 0.1100*** (0.006)
- XVAL × HHI: -0.2567*** (0.001); -0.2474*** (0.002) in specifications (3) and (4).
- IP × HHI: -0.0647* (0.079) in one specification.
- Control examples:
  - SGROWTH 0.1565** (0.020) across specifications.
  - CASH 0.1389* (0.072) to 0.1453* (0.063).
  - FXSALES 0.0844* (0.065) to 0.0817* (0.069).
- Number of observations: 6,935–6,949; R-squared 0.962.

### Table 11 — PRODK: First Differences
- ∆XVAL coefficients:
  - ∆XVAL 0.0924** (0.014) and 0.1044** (0.011) in specifications (3) and (4).
- ∆(XVAL × HHI) -0.1893*** (0.009); -0.2212*** (0.005) in specifications (3) and (4).
- ∆MB -0.0017** (0.016) to -0.0017** (various p-values).
- ∆ROA -0.2465** (0.048) to -0.2416* (0.050).
- ∆R&D -0.7336* (0.061) across specifications.
- Observations: 5,060–5,071; R-squared ~0.038–0.039.

*Source: _wp13234 - REFERENCES (source PDF content provided).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp13234.pdf_