## wp1722

## Source details

**Canonical URL:** [wp1722](https://www.imf.org/-/media/files/publications/wp/wp1722.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/wp1722.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/wp1722.pdf.json)

---

### I. Scope and motivation
- Excessive corporate debt increases bankruptcy probability and can amplify liquidity constraints and aggregate economic fluctuations; empirical references:
  - Decline in employment during the global financial crisis was significantly more pronounced in highly-leveraged than in lowly-leveraged firms (Giroud and Mueller (forthcoming)).
  - Higher leverage ratios in the non-financial corporate sector are associated with a significantly higher probability of recession (Sutherland and Hoeller (2012)).
  - Buildup of private credit during expansion periods tends to make subsequent recessions more likely, deeper and longer lasting (Jordà and others (2013)).
- Most CIT systems allow interest expense deductibility but not returns to equity, creating a “debt bias.”
  - Meta-studies report a typical impact coefficient of the CIT rate on the debt-asset ratio of 0.28 (De Mooij (2011); Feld and others (2013)).
  - Example: debt bias due to a CIT rate of 25 percent would be responsible for a 7 percentage-point higher debt-to-asset ratio in an average corporation.
- Two broad policy approaches to neutralize debt bias:
  - Allowance for Corporate Equity (ACE).
  - Deny or limit interest deductibility (Thin Capitalization Rules, TCRs).
- Practical constraint: full denial of interest deductibility is not observed in practice; countries implement partial TCRs that deny interest deductibility beyond fixed levels.

### II. TCR design, data and empirical approach
- TCR dimensions:
  - Scope: related-party debt only versus all debt.
  - Test type: fixed debt-equity ratio, fixed interest-earnings ratio (“earnings stripping rules”), or arm’s-length ratio.
  - Other features: exemptions, carry-forward provisions, combinations of tests, rule strictness.
- Study distinguishing features:
  - Uses consolidated accounts for non-financial firms in 60 countries between 2005 and 2014 (excludes intracompany transactions).
  - Tests differences between related-party-only TCRs and TCRs that restrict all debt.
  - Tests for circumvention (back-to-back loans).
  - Examines heterogeneity by industry tangibility.
  - Extends analysis to firm financial distress via Altman Z-score.
- Data and sample:
  - ORBIS consolidated accounts: 369,757 observations corresponding to about 100,000 distinct firms in 60 countries, 2005–2014.
  - Z-score sample around 45,000 observations (smaller due to data availability).
  - Variables winsorized at the 0.5 percent level.
  - Macroeconomic merges: inflation and GDP growth from IMF’s World Economic Outlook; real interest rate from World Bank WDI; TCRs and statutory CIT rates from IMF Fiscal Affairs Department database.
- Empirical strategies:
  - Firm-panel model including TCR indicators (thinCap–all, thinCap–total, thinCap–related-party), statutory CIT rate (citRate), controls, year and sector fixed effects.
  - Difference-in-differences interacting industry tangibility index (Rajan and Zingales 1998).
  - Altman Z-score and Z’ formulations to measure financial distress; binary distress indicator uses Z below the 7th percentile.

### III. Main empirical findings — effects on leverage and distress
- Scope matters:
  - TCRs that apply only to related-party debt have no significant impact on external borrowing of corporate groups and no effect on firm financial distress (Z-scores).
  - TCRs that target a broader corporate debt base reduce the consolidated debt ratio by about 5 percentage points, on average.
  - Wald test strongly rejects equality of total-debt versus related-party TCR effects (p-value = 0.00).
- Average TCR and tax effects (consolidated debt ratio dependent variable; Observations 262,727):
  - ThinCap–all: coefficient -1.333*** (column (1)); -1.062* (column (6) with industry-year fixed effects).
    - Interpretation: presence of some form of TCR reduces consolidated debt ratio on average by about 1.3 percentage points.
  - ThinCap–total: coefficients around -5.129***, -5.634***, -5.075***, -5.360***.
    - Interpretation: a TCR applying to total debt reduces debt ratios by about 5 percentage points on average.
  - ThinCap–related-party: coefficient -0.590*** in one specification but becomes statistically insignificant with industry-year fixed effects (e.g., -0.332, not significant).
    - Interpretation: related-party-only TCRs have much smaller effects (about one-tenth of total-debt TCRs) and less robust significance.
- Statutory CIT (citRate) effect on debt:
  - CitRate coefficients approximately 0.3 across specifications (examples: 0.301***, 0.308***, 0.313***, 0.283***, 0.293***, 0.296***).
  - Interpretation: lowering the statutory CIT rate by 10 percentage points leads to a reduction of the debt ratio by 3 percentage points.
- Industry heterogeneity — interaction with tangibility:
  - Citrate × tangibility positive and significant (e.g., 0.0064***, 0.0057***): debt bias is higher in capital-intensive industries.
  - thinCapTotal × tangibility: -0.0853***, -0.0888*** (columns (3) and (4)), indicating total-debt TCRs reduce debt ratios more in high-tangibility industries.
    - Magnitude illustration using β_sh = -0.088:
      - At 25th percentile of tangibility: ThinCap–total implied effect on debt ratio = -0.71 percentage points.
      - At 75th percentile of tangibility: ThinCap–total implied effect on debt ratio = -2.2 percentage points.
      - Difference between 75th and 25th percentile effects = -1.5 percentage points (2.2–0.71).
  - thinCapRelatedParty × tangibility: statistically insignificant (e.g., -0.00557).
- Effects on financial distress (binary indicator using Z below 7th percentile; Observations ~44,590–50,497):
  - ThinCap–total:
    - LPM: -0.050*** (Tables 5 columns (1) and (2)).
    - Logit marginal effects: -0.095*** (Tables 5 columns (3) and (4)).
    - Using Z': LPM -0.066*** to -0.067***; Logit -0.119** to -0.119***.
    - Interpretation: total-debt TCRs reduce probability of a manufacturing company facing bankruptcy risks by about 5 percentage points in LPM; larger marginal effects in logit specifications.
  - ThinCap–related-party: coefficients around -0.003 to 0.002 and generally statistically insignificant.
  - Difference-in-differences (Table 6):
    - thinCapTotal × tangibility: coefficients around -0.263*** to -0.2772*** for logit marginal effects, indicating total-debt TCRs lower likelihood of financial distress more in higher-tangibility industries.
    - thinCapAll × tangibility: -0.041***.
    - thinCapRelatedParty × tangibility: effectively zero or insignificant (e.g., -0.0006, 0.0039).
- No evidence that related-party-only TCRs induce offsetting increases in external debt via circumvention (back-to-back loans).

### IV. Key summary statistics (selected)
- debtRatio:
  - Mean 62.09
  - SD 36.18
  - Median 60.16
  - Min 4.380
  - Max 226.1
  - N 369,757
- citRate:
  - Mean 27.39
  - SD 5.217
  - Median 28
  - Min 8.500
  - Max 39
  - N 369,757
- ThinCap–all:
  - Mean 0.688
  - SD 0.463
  - Median 1
  - Min 0
  - Max 1
  - N 358,978
- ThinCap–total:
  - Mean 0.174
  - SD 0.380
  - Median 0
  - Min 0
  - Max 1
  - N 358,978
- ThinCap–related-party:
  - Mean 0.514
  - SD 0.500
  - Median 1
  - Min 0
  - Max 1
  - N 358,978
- Volatility:
  - Mean 0.0771
  - SD 0.170
  - Median 0.035
  - Min 0.000242
  - Max 1.647
  - N 264,432
- Z-score:
  - Mean 5.394
  - SD 7.026
  - Median 6.418
  - Min -26.32
  - Max 16.16
  - N 115,924
- Z’-score:
  - Mean 1.930
  - SD 3.203
  - Median 2.265
  - Min -11.80
  - Max 7.577
  - N 115,511
- Financial distress–Z:
  - Mean 0.075
  - SD 0.263
  - Median 0
  - Min 0
  - Max 1
  - N 44,590
- Financial distress–Z’:
  - Mean 0.101
  - SD 0.302
  - Median 0
  - Min 0
  - Max 1
  - N 45,283
- interestRateReal:
  - Mean 2.062
  - SD 4.046
  - Median 2.295
  - Min -42.31
  - Max 44.55
  - N 272,556
- gdpGrowth:
  - Mean 2.607
  - SD 8.135
  - Median 2.532
  - Min -51.45
  - Max 32.68
  - N 369,757
- inflationCpi:
  - Mean 2.815
  - SD 2.685
  - Median 2.321
  - Min -1.706
  - Max 34.73
  - N 369,757
- lnOperatingRev:
  - Mean 17.65
  - SD 2.247
  - Median 17.75
  - Min 11.29
  - Max 21.93
  - N 369,757
- ebitdaShareOfAssets:
  - Mean 6.067
  - SD 21.17
  - Median 8.701
  - Min -114.5
  - Max 35.42
  - N 369,757
- Descriptive note included in the source: "27.4 percent. It has a minimum value of 8.5 percent and a maximum value of 39 percent."

### V. Policy implications and recommendations
- To neutralize tax-induced debt bias and address macro-stability concerns, policymakers should consider TCR scope:
  - Narrow TCRs that restrict only related-party interest deductibility are unlikely to affect consolidated external leverage or firm distress.
  - Broad TCRs that cover all debt (ThinCap–total) have measurable effects, reducing consolidated debt ratios by about 5 percentage points on average and lowering bankruptcy risk, especially in high-tangibility industries.
- Industry structure matters: broader TCRs will have larger effects in industries with higher asset tangibility because firms in these sectors rely more on secured borrowing.
- Design considerations:
  - Anticipate possible circumvention (e.g., back-to-back loans) and interactions with other measures (e.g., group escape provisions using the group’s actual external leverage as a cap).
  - A uniform TCR may induce sectoral distortions; differentiation across sectors may be desirable.
- Research gaps for future work:
  - Use of country-by-country reports for multinationals to assess headquarter and subsidiary effects.
  - Effects of TCRs on investment.
  - How uniform TCRs affect firms with widely varying corporate financial structures for reasons other than taxation or tangibility.

*Source: IMF working paper — wp1722.*

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

### References

### I. Introduction — scope and motivation
- Excessive corporate debt increases bankruptcy probability and can amplify liquidity constraints and aggregate economic fluctuations; empirical findings noted:
  - Giroud and Mueller (forthcoming): decline in employment during the global financial crisis was significantly more pronounced in highly-leveraged than in lowly-leveraged firms.
  - Sutherland and Hoeller (2012): higher leverage ratios in the non-financial corporate sector are associated with a significantly higher probability of recession.
  - Jordà and others (2013): buildup of private credit during expansion periods tends to make subsequent recessions more likely, deeper and longer lasting.
- Most CIT systems allow interest expense deductibility but not returns to equity, creating a “debt bias.”
  - Meta-studies: De Mooij (2011) and Feld and others (2013) report a typical impact coefficient of the CIT rate on the debt-asset ratio of 0.28.
  - Example: debt bias due to a CIT rate of 25 percent would be responsible for a 7 percentage-point higher debt-to-asset ratio in an average corporation.
- Two broad policy approaches to neutralize debt bias:
  - Allowance for Corporate Equity (ACE) to treat equity more like debt.
  - Deny or limit interest deductibility to treat debt more like equity (partial restrictions implemented as Thin Capitalization Rules, TCRs).
- Practical constraints:
  - Full denial of interest deductibility (comprehensive business income tax) is not observed in practice due to international transaction issues and transitional problems; countries instead implement partial TCRs that deny interest deductibility beyond fixed levels.

### II. Background on Thin-Capitalization Rules (TCRs)
- As of the study, 60 countries implement some kind of TCR.
- Key dimensions of TCRs highlighted:
  - Scope: whether the rule restricts interest deductibility for only related-party debt or for all debt.
  - Test type: fixed debt-equity ratio, fixed interest-earnings ratio (“earnings stripping rules”), or arm’s-length ratio.
- Additional TCR features that matter but are harder to quantify include exemptions, carry-forward provisions, combinations of tests, and rule strictness; strictness of fixed debt-equity ratios tends not to vary much over time.
- Empirical literature context:
  - Prior studies often use unconsolidated accounts and focus on intragroup lending and debt shifting (Weichenrieder and Windischbauer 2008; Buettner and others 2012; Blouin and others 2014).
  - Debt shifting differs from debt bias: it is tax avoidance induced by cross-country CIT differences and may have limited macro-stability implications due to intra-group risk-sharing.
- This study’s distinguishing features:
  - Uses consolidated accounts for non-financial firms in 60 countries between 2005 and 2014 to capture external/group-level debt (excludes intracompany transactions).
  - Focuses on whether TCRs that restrict only related-party debt differ in impact from TCRs that restrict all debt.
  - Tests whether firms circumvent related-party-only TCRs through back-to-back loans routed via third parties.
  - Examines heterogeneous effects across industries, especially interaction with industry tangibility (higher tangibility → easier to borrow against collateral).
  - Extends analysis to firm financial distress via the Altman Z-score (Altman 1968; Altman and others 2015), which weights multiple financial ratios including debt ratio.

### III. Main empirical findings
- Effectiveness of scope of TCRs:
  - TCRs that apply only to related-party debt have no significant impact on external borrowing of corporate groups.
  - Such related-party-only TCRs also have no effect on firm financial distress as measured by Z-scores.
  - TCRs that target a broader corporate debt base are estimated to reduce the consolidated debt ratio by about 5 percentage points, on average.
- Industry heterogeneity:
  - Debt ratios are more responsive to broadly scoped TCRs in industries characterized by a high share of tangible assets, reflecting a higher propensity to borrow against collateral.
- Financial distress outcomes:
  - The paper investigates whether TCRs reduce financial distress (via substitution of equity for debt) or induce offsetting risk-increasing behaviors; overall evidence suggests scope matters for net effects on bankruptcy risk (related-party-only rules show no effect; broader rules reduce debt ratios and thus potentially reduce distress).

### IV. Policy implications and interpretation
- To neutralize tax-induced debt bias and address macro-stability concerns, policymakers should consider the scope of TCRs:
  - Narrow TCRs that restrict only related-party interest deductibility are unlikely to affect consolidated external leverage or firm distress.
  - Broad TCRs that cover all debt have measurable effects, reducing consolidated debt ratios by about 5 percentage points on average.
- Industry structure matters: broader TCRs will have larger effects in industries with higher asset tangibility because firms in these sectors rely more on secured borrowing.
- When designing TCRs, consider possible circumvention (e.g., back-to-back loans) and interactions with other measures (e.g., group escape provisions that use the group’s actual external leverage as a cap).

### V. Data and empirical approach (high level)
- Data: panel of non-financial firms in 60 countries, 2005–2014, using consolidated accounts to capture external/group-level debt.
- Empirical strategies:
  - Cross-country variation in TCR introduction and scope.
  - Difference-in-differences framework interacting industry tangibility (in the spirit of Rajan and Zingales 2005).
  - Use of Altman Z-score as a comprehensive indicator of financial distress.

*wp1722 - References*

### Introduction

### Introduction

### International Thin Capitalization Rules and Related Parameters
- The source lists fixed debt-equity and EBITDFA percentage rules by country, distinguishing rules for related-party debt and for total debt. Examples include:
  - Related-party fixed debt-equity or EBITDFA rules: Argentina 1999 2:1; Belarus 2013 1:1; Brazil 2011 2:1; Canada 1972 1.5:1; Chile 2012 3:1; China 2008 2:1; Czech Republic 2007 4:1; Ecuador 2007 3:1; Egypt 2005 4:1; El Salvador 2012 3:1; France 2007 1.5:1; Ghana 2000 2:1; Gibraltar 2010 5:1; Kenya 2006 3:1; Korea, Republic of 1997 2:1; Lithuania 2002 4:1; Macedonia 3:1; Mexico 2005 3:1; Mongolia 2005 3:1; Mozambique 2008 2:1; Namibia 2012 3:1; Oman 2012 2:1; Peru 2007 3:1; Poland 1999 1:1; Rwanda 2008 4:1; Slovenia 2005 4:1; Sri Lanka 2006 3:1; Taiwan 2011 3:1; Turkey 2006 3:1; Uganda 2013 1.5:1; United States 1989 1.5:1; Venezuela 2007 1:1; Yemen 2010 7:3.
  - Fixed debt-equity rule for total debt: Albania 2000 4:1; Australia 1997 1.5:1; Bulgaria 2006 3:1; Colombia 2013 3:1; Croatia 2005 4:1; Denmark 1998 4:1; Dominican Republic 2013 3:1; Georgia 2018 (planned) 3:1; Hungary 2000 3:1; Indonesia 2016 4:1; Japan 1992 3:1; Latvia 2003 4:1; New Zealand 1995 1.6:1; Papua New Guinea 2013 2:1; Romania 2006 3:1; Serbia 2001 4:1; Zimbabwe 2011 3:1.
  - Arm's-length rule examples: Kazakhstan 2008; South Africa 1995; United Kingdom 1999.
  - Interest-stripping rule for total debt: Germany 1994 30%; Greece 2010 40%; Italy 2003 30%; Portugal 1996 30%; Spain 1996 30%.
  - Interest-stripping rule for related-party debt: Finland 2013 25%; Norway 2014 30%; Slovakia 2015 25%.
- Footnotes in the source note:
  - "The TCR in the year of introduction may differ from the one applied in 2016. For example, Germany introduced a TCR in 1994 in the form of a safe-harbor ratio. The interest stripping rule was introduced in 2008."
  - "Has both interest-EBITDA and equity-debt ratios."

### Alternative Policy Options Discussed in the Literature
- Formula apportionment of worldwide interest expenses across affiliates based on location of assets, external debt, gross profits, or employment.
- Net financing deduction approach: deny interest deduction if borrowing finances an equity injection in foreign affiliates (Desai and Dharmapala 2015).
- Taxation of intracompany interest at the rate of the country from which it is borrowed to eliminate incentives for debt shifting (IMF 2016a).
- Group-wide test or worldwide cap: restrict interest deductibility of an affiliate according to its share in the group’s worldwide activity or deny deductions exceeding the worldwide third-party interest expense of the group.

### Methodology and Empirical Specification
- Three main empirical specifications are used to identify the impact of Thin Capitalization Rules (TCRs) on corporate debt ratios.

- Firm-panel model (equation (1)):
  - Consolidated total debt-asset ratio variable: ݐܾ݁݀
௜௦௖௧
.
  - TCR indicator: ܴܥܶ
௖௧
 takes the value 1 if a country adopts a thin capitalization rule in year t, and 0 otherwise.
  - Statutory CIT rate: ݔܽܶ
௖௧
.
  - Controls vector: ࢄ
࢚ࢉ࢙࢏
 with coefficients ࢼ.
  - Includes year fixed effects (ߣ
௧
) and sector fixed effects (ߤ
௦
); variants include industry-year fixed effects (θ_st).
  - Main coefficients of interest:
    - ܽ
ଶ
: expected positive effect on consolidated debt ratio (depicting debt bias).
    - ܽ
ଵ
௝
: j denotes TCR type — thinCap-total (applies to total debt or broad debt) and thinCap-related-party (applies only to related-party debt). Both TCR variables are included; thinCap-all denotes no distinction.
    - A negative ߙ
ଵ
௝
 indicates a TCR lowers the consolidated debt ratio.
    - Expected impact of related-party TCR on consolidated debt is a priori unclear (ambiguity in ܽ
ଵ
௧௛௜௡஼௔௣_௧௢௧௔௟ and related expressions in source); possibility of substitution via back-to-back loans is noted.

- Difference-in-difference specification (equation (2)):
  - Interaction between an industry-specific tangibility index (݃݊ܽܶ
௦
) and country-level tax variables.
  - Equation (2) displayed as:
    ݐܾ݁݀
௜௦௖௧
ߚൌ
଴
ߚ൅
ଵ
ሺ
݃݊ܽܶ
௦
ݔܽܶൈ
௖௧
ሻ
ߚ൅
ଶ
ሺ
݃݊ܽܶ
௦
ܴܥܶൈ	
௖௧
ሻ
ߠ൅
௦
߰൅
௖௧
݁൅
௜௦௖௧
.
  - ݃݊ܽܶ
௦
 is the industry median share of tangible assets in total assets at NACE revision 2 using U.S. data (Rajan and Zingales 1998).
  - Coefficients ߚ
ଵ
 and ߚ
ଶ
 capture differential effects of taxes and TCRs by industry tangibility.
    - Positive ߚ
ଵ
 implies high-tangibility industries are more responsive to taxes.
    - Negative ߚ
ଶ
 implies debt ratios of high-tangibility industries are more negatively affected by TCRs.
  - Explored for both thinCap-total and thinCap-related-party.

- Bankruptcy risk and Z-score analysis:
  - Altman Z-score (equation (3)) used to measure financial distress:
    Z = 3.25 + 6.56 X1 + 3.26 X2 + 6.72 X3 + 1.05 X4,
    where:
      - X1 = working capital / total assets
      - X2 = retained earnings / total assets
      - X3 = earnings before interest and taxes / total assets
      - X4 = book value of equity / book value of total liabilities
    - Higher Z indicates lower risk of failure.
  - Alternative Z’ (equation (4)) following Bernanke and Campbell (1988) and Altman (1968):
    Z’ = 1.2 X1 + 1.4 X2 + 3.3 X3 + 0.6 X4 + 1.0 X5,
    where X5 = sales / total assets.
    - Low values of Z’ indicate financial distress and higher likelihoods of bankruptcy.
  - Binary financial distress indicator:
    - Firms with Z value lower than the 7th percentile of the sample Z are set to 1 (in financial distress), others 0.
    - This cutoff corresponds to the literature’s "zone of financial turmoil" criterion.
  - Volatility of corporate earnings measure (equation (5)):
    - Constructed as the deviation of firm EBITDA/Assets from country average in a given year and the standard deviation of this measure over a three-year rolling window.
    - Equation displayed in source as:
      ଌݕݐ݈݅݅ݐ݈ܽ݋ܸ
ට
ଵ
்
∑
ܧቀ
௜௦௖௧
െ
ଵ
்
∑
ܧ
௜௦௖௧
்
ଵ
ቁ
ଶ
்
ଵ
,
      with ܧ
௜௦௖௧
 defined in the source.

### Data
- Firm-level data source: ORBIS database of the Bureau van Dijk.
  - Contains balance sheet items, consolidated vs unconsolidated accounts, and economic sector.
  - Sample: consolidated accounts for 369,757 observations corresponding to about 100,000 distinct firms in 60 countries, covering the period 2005 to 2014.
  - Analysis excludes financial sector firms where TCRs often do not apply.
  - Variables winsorized at the 0.5 percent level to control for outliers.
  - Z-score sample is smaller due to data availability; ultimate sample for Z-score has around 45,000 observations.
- Macroeconomic merges:
  - Inflation and GDP growth from IMF’s World Economic Outlook database.
  - Real interest rate data from World Bank World Development Indicators.
  - TCRs and statutory CIT rates from the IMF’s Fiscal Affairs Department database.

### Key Summary Statistics (Table 2)
- debtRatio:
  - Mean 62.09
  - SD 36.18
  - Median 60.16
  - Min 4.380
  - Max 226.1
  - N 369,757
- citRate:
  - Mean 27.39
  - SD 5.217
  - Median 28
  - Min 8.500
  - Max 39
  - N 369,757
- ThinCap–all:
  - Mean 0.688
  - SD 0.463
  - Median 1
  - Min 0
  - Max 1
  - N 358,978
- ThinCap–total:
  - Mean 0.174
  - SD 0.380
  - Median 0
  - Min 0
  - Max 1
  - N 358,978
- ThinCap–related-party:
  - Mean 0.514
  - SD 0.500
  - Median 1
  - Min 0
  - Max 1
  - N 358,978
- Volatility:
  - Mean 0.0771
  - SD 0.170
  - Median 0.035
  - Min 0.000242
  - Max 1.647
  - N 264,432
- Z-score:
  - Mean 5.394
  - SD 7.026
  - Median 6.418
  - Min -26.32
  - Max 16.16
  - N 115,924
- Z’-score:
  - Mean 1.930
  - SD 3.203
  - Median 2.265
  - Min -11.80
  - Max 7.577
  - N 115,511
- Financial distress–Z:
  - Mean 0.075
  - SD 0.263
  - Median 0
  - Min 0
  - Max 1
  - N 44,590
- Financial distress–Z’:
  - Mean 0.101
  - SD 0.302
  - Median 0
  - Min 0
  - Max 1
  - N 45,283
- interestRateReal:
  - Mean 2.062
  - SD 4.046
  - Median 2.295
  - Min -42.31
  - Max 44.55
  - N 272,556
- gdpGrowth:
  - Mean 2.607
  - SD 8.135
  - Median 2.532
  - Min -51.45
  - Max 32.68
  - N 369,757
- inflationCpi:
  - Mean 2.815
  - SD 2.685
  - Median 2.321
  - Min -1.706
  - Max 34.73
  - N 369,757
- lnOperatingRev:
  - Mean 17.65
  - SD 2.247
  - Median 17.75
  - Min 11.29
  - Max 21.93
  - N 369,757
- ebitdaShareOfAssets:
  - Mean 6.067
  - SD 21.17
  - Median 8.701
  - Min -114.5
  - Max 35.42
  - N 369,757

*Source: IMF working paper — Introduction (wp1722).*

### 27.4 percent. It has a minimum value of 8.5 percent and a maximum value of 39 percent

### wp1722 - 27.4 percent. It has a minimum value of 8.5 percent and a maximum value of 39 percent

### Summary of study and context
- Focus: Effectiveness of thin capitalization rules (TCRs) limiting interest deductibility on corporate debt ratios and corporate stability using ORBIS data for 2005–14.
- TCR variables:
  - ThinCap–all: whether a country embraces a TCR.
  - ThinCap–total: equal to one if the TCR rule applies to total debt, zero otherwise.
  - ThinCap–related-party: equal to one if the TCR applies only to related-party debt, zero otherwise.
- Key descriptive: "27.4 percent. It has a minimum value of 8.5 percent and a maximum value of 39 percent" (as reported in the source).

### Main empirical findings on debt ratios
- Statutory corporate income tax (CitRate) effect:
  - CitRate coefficients approximately 0.3 across specifications (e.g., 0.301***, 0.308***, 0.313***, 0.283***, 0.293***, 0.296***).
  - Interpretation: lowering the statutory CIT rate by 10 percentage points leads to a reduction of the debt ratio by 3 percentage points.
- TCR average effects (Table 3, consolidated debt ratio dependent variable; Observations 262,727):
  - ThinCap–all: coefficient -1.333*** (column (1)); -1.062* (column (6) with industry-year fixed effects).
    - Interpretation: presence of some form of TCR reduces consolidated debt ratio on average by about 1.3 percentage points.
  - ThinCap–total: coefficients around -5.129***, -5.634***, -5.075***, -5.360***.
    - Interpretation: a TCR applying to total debt reduces debt ratios by about 5 percentage points on average.
  - ThinCap–related-party: coefficient -0.590*** in one specification but becomes statistically insignificant with industry-year fixed effects (e.g., -0.332, not significant).
    - Interpretation: related-party-only TCRs have much smaller effects (about one-tenth of total-debt TCRs) and less robust significance.
  - A Wald test strongly rejects equality of total-debt versus related-party TCR effects (p-value = 0.00).

### Interaction with industry tangibility (Tables 4 and 6)
- Tangibility measure: industry-specific index capturing industry needs for external financing (Rajan and Zingales (1998)).
- Taxes × tangibility:
  - Citrate × tangibility positive and significant (e.g., 0.0064***, 0.0057***), indicating debt bias is higher in capital-intensive industries.
- TCR × tangibility (difference-in-difference, Table 4; Observations up to 369,757):
  - thinCapAll × tangibility: -0.0261* (column (1)).
  - thinCapTotal × tangibility: -0.0853***, -0.0888*** (columns (3) and (4)).
  - thinCapRelatedParty × tangibility: statistically insignificant (e.g., -0.00557).
  - Magnitude illustration (using estimated β_sh = -0.088 in column 4):
    - At 25th percentile of tangibility: ThinCap–total implied effect on debt ratio = -0.71 percentage points.
    - At 75th percentile of tangibility: ThinCap–total implied effect on debt ratio = -2.2 percentage points.
    - Difference between 75th and 25th percentile effects (absolute value) = -1.5 percentage points (2.2–0.71).
- Table 6 (diff-in-diff on financial distress; logit marginal effects, Observations up to 50,497):
  - thinCapAll × tangibility: -0.041***.
  - thinCapTotal × tangibility: coefficients around -0.263*** to -0.2772***, indicating total-debt TCRs lower likelihood of financial distress more in high-tangibility industries.
  - thinCapRelatedParty × tangibility: effectively zero or insignificant (e.g., -0.0006, 0.0039).

### Effects on financial distress and corporate stability (Tables 5 and 6)
- Financial distress indicator: coded 1 if firm Z-score is below the 7th percentile; specifications use Z (equation 3) and Z' (equation 4).
- Table 5 (LPM and logit; Observations ~44,590–45,239):
  - ThinCap–total:
    - LPM: -0.050*** (columns (1) and (2)).
    - Logit marginal effects: -0.095*** (columns (3) and (4)).
    - Using Z': LPM -0.066*** to -0.067***; Logit -0.119** to -0.119***.
    - Interpretation: total-debt TCRs reduce probability of a manufacturing company facing bankruptcy risks by about 5 percentage points in LPM; larger marginal effects in logit specifications.
  - ThinCap–related-party: coefficients around -0.003 to 0.002 and generally statistically insignificant.
  - Other controls:
    - gdpGrowth negative and significant (e.g., -0.002***).
    - Volatility positive and significant (e.g., 0.504***, 0.510***).
    - interestRateReal positive and significant in some specifications (e.g., 0.002***).
- Table 6 (logit diff-in-diff on financial distress):
  - Confirms total-debt TCRs reduce likelihood of financial distress more in higher-tangibility industries.
  - Controls: ebitdaShareOfAssets negative and significant (e.g., -0.004*** to -0.0054***); operatingRevenueRatio negative and significant; Volatility positive and significant.

### Robustness and interpretation
- Results robust across specifications, fixed effects choices, and estimators (LPM vs logit).
- Main pattern: TCRs that apply to total debt have a robust, large negative effect on corporate debt ratios and reduce financial distress; TCRs that apply only to related-party debt show little or no effect.
- No evidence found that related-party TCRs indirectly increase external debt through back-to-back loans.

### Policy implications and conclusions
- If a country adopts a TCR to address stability concerns from debt bias, it should apply it to all debt (ThinCap–total), not only related-party debt.
- Two-thirds of observed TCRs worldwide target related-party debt only; such design choices limit effectiveness in addressing debt bias and macro-financial stability.
- A uniform TCR may induce sectoral distortions because firms differ in tangibility and non-tax financial decisions; differentiation across sectors may be desirable.
- Areas for further research highlighted:
  - Impact of TCRs using country-by-country reports for multinationals to assess headquarter and subsidiary effects.
  - Effects of TCRs on investment.
  - How uniform TCRs affect firms with widely varying corporate financial structures for reasons other than taxation or tangibility.

*Source: wp1722 - 27.4 percent. It has a minimum value of 8.5 percent and a maximum value of 39 percent (PDF chapter/section).*

### REFERENCES

### REFERENCES

### Cited literature

- Acemoglu, D., V. Carvalho, A. Ozdaglar, and A. Tahbaz-Salehi, 2012, “The Network Origins of Aggregate Fluctuations,” Econometrica, Vol. 80, pp. 1977–2016.
- Altman, E.I., 1968, “Financial Ratios, Discriminant Analysis and the Prediction of Corporate Bankruptcy,” Journal of Finance, Vol. 23, pp. 589–609.
- ———, M. Iwanicz-Drozdowska, E. Laitinen, and A. Suvas, 2014, Distressed Firms and Bankruptcy Prediction in an International Context: A Review of Empirical Analysis of Altman’s Z-score Model (New York: New York University Stern Business School).
- Bernanke, B., and J.Y. Campbell, 1988, “Is There a Corporate Debt Crisis?” Brookings Papers on Economic Activity No. 1 (Washington, Brookings Institute).
- Bianchi, J., 2011, “Overborrowing and Systemic Externalities in the Business Cycle,” American Economic Review, Vol. 101, pp. 3400–26.
- Blouin, J., H. Huizinga, L. Laeven, and N. Gaëtan, 2014, “Thin Capitalization Rules and Multinational Firm Capital Structure,” IMF Working Paper No. 14/12 (Washington: International Monetary Fund).
- Buettner, T., M. Overesch, U. Schreiber, and G. Wamser, 2012, “The Impact of Thin-Capitalization Rules on the Capital Structure of Multinational Firms,” Journal of Public Economics, Vol. 96, pp. 930–38.
- De Mooij, R., 2011, “The Tax Elasticity of Corporate Debt: A Synthesis of Size and Variations,” IMF Working Paper No. 11/95 (Washington: International Monetary Fund).
- Desai, M., and D. Dharmapala, 2015, “Interest Deductions in a Multijurisdictional World,” National Tax Journal, Vol. 68, No. 3, pp. 653–80.
- Feld, L. P., J. H. Heckemeyer, and M. Overesch, 2013, “Capital Structure Choice and Company Taxation—A Meta Study,” Journal of Banking and Finance, Vol. 37, pp. 2850–66.
- Friedman, B. M., 1986, “Increasing Indebtedness and Financial Stability in the United States,” NBER Working Paper No. 2072 (Cambridge: National Bureau for Economic Research).
- Giroud, X., and H. Mueller, “Firm Leverage and Unemployment during the Great Recession,” Quarterly Journal of Economics (forthcoming).
- Haufler, A., and M. Runkel, 2012, “Firms’ Financial Choices and Thin Capitalization Rules under Corporate Tax Competition,” European Economic Review, Vol. 56, 1087–103.
- Hebous, S., and M. Ruf, 2015, “Evaluating the Effects of ACE Systems on Multinational Debt Financing and Investment,” CESifo Working Paper No. 5360 (Munich: CESifo).
- Heckemeyer, J., and R. De Mooij, 2017, “Taxation and Corporate Debt: Are Banks any Different?” National Tax Journal, Vol. 70, No. 1.
- Huizinga, H., L. Laeven, and G. Nicodeme, 2008, “Capital Structure and International Debt Shifting, Journal of Financial Economics, Vol. 88, pp. 80–118.
- International Monetary Fund, 2016a, “Tax Policy, Leverage and Macroeconomic Stability,” IMF Policy Paper FO/Dis/16/151 (Washington).
- ———, 2016b, Fiscal Monitor: Debt: Use it Wisely, October (Washington).
- John, K., L. Litov, and B. Yeung, 2008, “Corporate Governance and Risk-taking,” Journal of Finance, Vol. LXIII, pp. 1679–728.
- Jordà, Ò, M. Schularick, and A. M. Taylor, 2013, “When Credit Bites Back,” Journal of Money, Credit and Banking, Vol. 45, pp. 3–28.
- Merlo, V, N. Riedel, and G. Wamser, 2015, The Impact of Thin Capitalization Rules on the Location of Multinational Firms' Foreign Affiliates, Mimeo.
- Organisation for Economic Co-operation and Development (OECD), 2015, “Base Erosion and Profit Shifting, Action 4: Limiting Base Erosion Involving Interest Deductions and Other Financial Payments,” Final Report (Paris).
- Rajan, R., and L. Zingales, 1998, “Financial Dependence and Growth,” American Economic Review, Vol. 88, pp. 559–86.
- Sutherland, D., and P. Hoeller, 2012, “Debt and Macroeconomic Stability: An Overview of the Literature and Some Empirics,” OECD Economics Department Working Papers No. 1006 (Paris: Organisation for Economic Co-operation and Development).
- Weichenrieder, A, and H. Windischbauer, 2008, “Thin-Capitalization Rules and Company Responses—Experience from German Legislation,” CESifo Working Paper No. 2456 (Munich: CESifo).

*wp1722 - REFERENCES*

---


_Source: https://www.imf.org/-/media/files/publications/wp/wp1722.pdf_