## 2. Benchmark Regression

## Source details

**Canonical URL:** [2. Benchmark Regression](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2010/_wp10231.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2010/_wp10231.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2010/_wp10231.pdf.json)

---

### Major themes and objectives
- Investigate whether a country can reduce financial frictions through institutional reforms and identify which institutional reforms are most effective.
- Adapt a canonical investment model (Tobin’s Q and Abel and Eberly (1994) framework) to allow institutional differences to affect financial frictions.
- Use model-derived speed of adjustment of Q as a function of the nature of financial frictions to link institutions to movements of Q over time.

### Data and sample
- Firm-level data from the Worldscope database of Thomson Financial covering the period 1990 to 2007.
- About one million original firm-year observations (including those for which Q cannot be constructed).
- About 380,000 firm-year observations for which Q can be constructed.
- After initial cleaning: about 300,000 firm-year observations with Q.
- Benchmark-regression sample: about 75,000 firm-year observations (one reported benchmark: 74,272 observations).
- Less restricted sample (excluding Age): 147,711 observations.
- Sample countries (48): Argentina, Australia, Austria, Belgium, Brazil, Canada, Chile, China, Colombia, Czech Republic, Denmark, Finland, France, Germany, Greece, Hong Kong, Hungary, India, Indonesia, Ireland, Israel, Italy, Japan, South Korea, Luxembourg, Malaysia, Mexico, Netherlands, New Zealand, Norway, Pakistan, Peru, Philippines, Poland, Portugal, Russia, Singapore, South Africa, Spain, Sri Lanka, Sweden, Switzerland, Taiwan, Thailand, Turkey, United Kingdom, United States, and Venezuela.

### Theoretical framework and identification strategy
- Base model: Tobin’s Q framework extended to include two generic forms of financial frictions:
  - Firm-specific financial transaction costs (reduce sensitivity of investment to current firm cash flows).
  - Overall cost of capital (determined partly at the country level and partly by firm-specific required rates of return).
- Timing and setup: previous-period capital K– and beginning-of-period productivity ε determine investment I; adjustment costs φ(I, K−, K, X, W, ε) are linearly homogeneous of degree one in I and K; financial transaction costs λ(B, K−, K, X, W, ε) are linearly homogeneous in B and K and matter only when B>0.
- Investment adjustment cost specification:
  - φ(I, K) = c1(I/K) + c2(I/K)^2/2 + c3(I/K)^3/3.
- Financial transaction cost specification:
  - λ(B, K) = b1(B/K) + b2(B/K)^2/2 + b3(B/K)^3/3.
  - Interpretation: b1 and b3 lower ⇒ lower financial transaction costs; b2 captures size effect with small-firm premium when K is small relative to B.
- Equilibrium marginal condition and empirical law of motion for Tobin’s Q: model yields equations linking Q, expected Q, required rate of return r, marginal financial transaction cost terms λ1, λ2, and adjustment cost derivatives φ2 (equations presented in source as (5), (6), (7)).
- Relation to speed of adjustment of Q:
  - Speed of adjustment depends on required rate of return r and marginal financial transaction cost λ1 with opposite signs; better institutions may lower r (speeding adjustment) and reduce λ1 (slowing adjustment); net effect theoretically ambiguous.
  - Size effect λ2 implies smaller firms pay higher fees; predicted small-firm premium lower in countries with better institutions.
- Average versus marginal Q: under linear homogeneity assumptions, marginal Q equals average Q (Hayashi result), justifying use of average Q empirically.

### Estimation methodology and identification
- Core estimator: minimize the squared sum of one-period-ahead forecast errors ξ = Q − E[Q | ε] at the firm level every year (Equation (9)).
- Forecast error serially uncorrelated under model; OLS estimates unbiased and consistent; robust standard errors clustered at country-year to correct within-country correlation.
- General empirical specification expresses E[Q] as function of lagged Q and interactions of firm characteristics X and time-invariant institutional variables W with object Z = π1 − φ2 − λ1 − λ2 (Equation (8) and detailed estimating Equation (12)).
- Indicator functions used:
  - Ψijkt = 1 if Iijkt > 0, 0 otherwise.
  - Χijkt = 1 if Bijkt > 0, 0 otherwise.
- Marginal profit approximation: πijkt = (AAijkt − AAijkt−1)/AAijkt−1 using Total Assets A.
- Institutional variables are time-invariant; their level effects absorbed in country fixed effects; identification comes from interaction terms.
- Controls: industry dummies and firm age; firm size omitted as endogenous.

### Key empirical findings
- Good corporate governance (CorpGov: shareholder/anti-director rights) results:
  - Lowers the required rate of return for all firms (benchmark coefficient on lagged Q interaction with corporate governance = -0.0433).
  - Especially reduces financial transaction costs faced by small firms (reduces small-firm premium by about 3 percent of assets for a one-standard-deviation improvement in investor protection of 1.3).
  - Associated with faster convergence of Q toward firm-specific steady state (lower required rate of return speeds adjustment).
- General enforcement of financial contracts (Institution: property rights):
  - Reduces financial transaction costs for small firms (one-standard-deviation improvement of 0.8 lowers small-firm premium by about 4 percent of assets).
- Creditor rights (Creditor: strength of legal protection for lenders and borrowers):
  - Do not robustly affect either financial transaction costs or the required rate of return for the sample of listed firms.
- Product market competition (Compet):
  - More intense competition associated with a higher required rate of return (coefficient 0.0772, significant at the 10 percent level).
- Financial market development (FinMkt):
  - More developed financial markets reduce required rate of return in IV estimates (significant at the 10 percent level in some specifications).
- Speed-of-adjustment heterogeneity:
  - Differences in adjustment speed between small and large firms are smaller where shareholder protection and general institutional quality are better.
- Decomposition conclusion:
  - Institutional improvements are linked to reductions in the two modeled frictions (financial transaction costs and required rate of return) as identified via model and regression methodology.

### Robustness checks and alternative specifications
- Accounting and measurement robustness: benchmark results replicate using before-tax income (Table 3a), broader investment definition including financial investments (Table 3b), and narrower external finance excluding trade credit (Table 3c); narrower external capital weakens some effects except corporate governance reduction of small-firm premium.
- One-by-one institutional regressions: estimating institutional factors individually yields similar results to benchmark; product market competition and financial market development become significant in one-by-one regressions.
- Alternative institutional proxies:
  - Spamann’s anti-director rights index reverses signs relative to La Porta et al. (1998).
  - Djankov et al. (2008) anti-self-dealing index broadly replicates benchmark except corporate governance no longer matters for required rate of return.
  - De Nicolo, Laeven, and Ueda (CGQ) broadly replicate benchmark with differences due to mean scaling.
  - Narrower creditor-rights measures sometimes associated with higher small-firm premium; survey-based bankruptcy-efficiency measures yield benchmark results (no effect of creditor rights on Q dynamics).
  - Alternative measures of general institutional quality (Rule of Law; Trust in People) reduce small-firm premium but have mixed effects on curvature.
  - Alternative competition proxies (business entry rates; costs of start-ups) show qualitative effects similar to corporate governance on small-firm premium and curvature.
- Sample selection robustness: excluding Age to raise sample from 74,272 to 147,711 yields broadly similar results, except corporate governance no longer matters for required rate of return in that specification; results unchanged when using all firms or manufacturing firms only.
- Real adjustment: good corporate governance lowers rate at which real adjustment costs increase with size (c2) but increases curvature (c3), implying small new investments cost less while big new investments cost more; overall investment effect ambiguous because intercept not identified.

### Instrumental variable estimation and measurement error tests
- Measurement error considerations: four forms of measurement errors assessed (stock price noise, accounting issues, marginal vs average Q, timing assumptions); tests for serial correlation in OLS errors fail to reject zero serial correlation; conclusion: measurement errors, if any, are very small relative to one-period-ahead forecast errors.
- IV strategy: twice-lagged Q used as instrument for lagged Q; fitted values used for interaction instruments (Wooldridge (2002, p.237)); two-stage least squares with clustering at country-year.
- IV diagnostics and results:
  - Kleibergen-Paap rk Wald F statistic = 5.14.
  - Anderson-Rubin Wald test F(71,607) = 165.17 (rejecting under-identification).
  - IV results broadly replicate OLS-fixed effects. Notable IV difference: corporate governance no longer affects required rate of return significantly in some IV specifications; IV corporate governance coefficient on required return = -0.0209 (Table 8).
- Tests for autocorrelation of fixed-effect residuals:
  - AR(1) coefficient = 0.200 with standard error 0.162 (t-statistic 1.64) when testing theoretical null of –1/(T–1) with T = 18.
  - Alternative test using all observations: AR(1) = 0.050 with standard error 0.054 (t-statistic 0.91).

### Variable definitions and institutional measures
- Marginal profit, π1: proxied by increase in earnings divided by increase in total assets. Earnings (cash-flow measure) = Net Income before Extraordinary Items and Preferred Dividends + Interest Expense on Debt + Depreciation and Amortization. Alternative before-tax measure: Operating Income + Depreciation and Amortization.
- Investment, I: benchmark = Capital Expenditure; broader definition = Capital Expenditure + Change in Cash and Short-Term Investment.
- External finance, B: benchmark = Capital Expenditure + Change in Cash and Short-Term Investment – Cash Flow from Operation – Decrease in Inventory – Decrease in Receivables – Increase in Payables; narrower definition = net increase in Total Debt + Net Proceeds from Sale/Issue of Common and Preferred Stocks.
- Q defined as (Market Capitalization + Total Asset – Total Equity) / Total Asset, measured at fiscal-year end.
- Capital stock measure: Total Asset; debt valued at par.
- Firm controls: industry dummies and firm age (Founded Date).
- Country-specific real short-term risk free rate = short-term government Treasury bill rate minus CPI inflation.
- Macroeconomic volatility: standard deviation of real GDP growth for 1995-2006.
- Institutional measures (W) used in benchmark regression:
  - CorpGov: shareholder (anti-director) rights (La Porta et al., 1998).
  - Creditor: strength of legal protection for lenders and borrowers (World Bank, 2008a).
  - Institution: property rights measure of La Porta et al. (1998).
  - Compet: measure of trade barriers (World Economic Forum, 2007).
  - FinMkt: stock market-capitalization-to-GDP for 2005 (World Bank, 2008b).

### Main interpretation and policy-relevant conclusions
- Principal empirical finding: good corporate governance (strong shareholder protection) and, less robustly, good general institutional quality are associated with lower financial transaction costs — especially by reducing the small-firm premium — and often associated with a lower required rate of return; together these imply more efficient capital allocation across firms and faster adjustment of Q to steady state.
- Creditor rights play a lesser role for listed firms in this sample; possible explanations include equity issuance options for listed firms and the importance of corporate governance for efficient use of funds regardless of finance source.
- Policy implications:
  - Improvements in corporate governance are a clear channel to reduce financial frictions and benefit firms.
  - Strengthening creditor rights may be less critical for listed firms, though reforms enhancing bankruptcy restructuring efficiency (e.g., Chapter 11–like reform) could improve ex-post contracting efficiency and be desirable.
- Caveat and research agenda: approach relies on a canonical investment theory with reduced-form financial frictions; better theoretical microfoundations linking institutions (e.g., moral hazard) to measurable financial frictions would improve empirical identification and is suggested as future research.

### Key exact numeric highlights and statistics
- Data period: 1990 to 2007.
- Countries: 48.
- Original firm-year observations: about one million.
- Firm-year observations with Q after deletions: about 300,000.
- Benchmark regression sample: about 75,000 firm-year observations (one reported benchmark: 74,272 observations).
- Less restricted sample (excluding Age): 147,711 observations.
- Benchmark coefficient (corporate governance on required return) = -0.0433.
- One-standard-deviation improvement in anti-director rights index = increase of 1.3; implied Q reduction next period = 0.16 for the average firm with Q = 3.
- Small-firm premium reductions: corporate governance one-standard-deviation (1.3) lowers premium by about 3 percent of assets; general institutional quality one-standard-deviation (0.8) lowers premium by about 4 percent of assets.
- Product market competition coefficient on required return = 0.0772 (significant at 10 percent).
- Table 6 reported: observations = 147,711; R-squared = 0.435.
- Instrumental variable estimation (Table 8): corporate governance coefficient on required return = -0.0209; Kleibergen-Paap rk Wald F statistic = 5.14; Anderson-Rubin Wald test F(71,607) = 165.17.
- Tests for autocorrelation of fixed-effect residuals: AR(1) = 0.200 (standard error 0.162, t-statistic 1.64) with theoretical null –1/(T–1), T = 18; alternative AR(1) = 0.050 (standard error 0.054, t-statistic 0.91).
- Macroeconomic volatility measure: standard deviation of real GDP growth for 1995-2006.

*Excerpted from the IMF working paper PDF _wp10231 (Worldscope firm-level data, 1990–2007).*

### 2. Benchmark Regression...............................................................................32

### 2. Benchmark Regression

### Major themes and objectives
- Investigate whether a country can reduce financial frictions through institutional reforms and identify which institutional reforms are most effective.
- Adapt a canonical investment model (Tobin’s Q and Abel and Eberly (1994) framework) to allow institutional differences to affect financial frictions.
- Use model-derived speed of adjustment of Q as a function of the nature of financial frictions to link institutions to movements of Q over time.

### Data and sample
- About 75,000 firm-year observations.
- 48 major advanced and emerging market economies.
- Period covered: 1990-2007.

### Theoretical framework and identification strategy
- Base model: Tobin’s Q framework extended to include two generic forms of financial frictions:
  - Firm-specific financial transaction costs (reduce sensitivity of investment to current firm cash flows).
  - Overall cost of capital (determined partly at the country level and partly by firm-specific required rates of return).
- Institutional differences are modeled to generate these financial frictions in specific ways, allowing derivation of a theoretical relation between institutions and the stochastic dynamics of Q.
- Estimation strategy: develop a simple regression methodology using differences between predicted values and realized values of Q to estimate structural parameters linking institutions to each form of friction.
- Estimators obtained by minimizing the squared sum of the one-period ahead forecast errors (the difference between theory-based predictions and realized values of Q at the firm level every year).
- Instrumental variable estimations are also conducted to address potential endogeneity; results reported as robust to IV estimation.

### Key empirical findings
- Good corporate governance:
  - Lowers the required rate of return for all firms.
  - Especially reduces the financial transaction costs faced by small firms.
  - Interpretation: lowers informational and agency problems, thereby reducing financial frictions.
- Better general enforcement of financial contracts (e.g., adherence to the rule of law and well-established property rights):
  - Reduces financial transaction costs for small firms.
- Stronger creditor rights:
  - Do not robustly affect either financial transaction costs or the required rate of return for the sample of listed firms.
  - Interpretation offered: collateral constraints appear not important for listed firms, or measured creditor rights do not capture relevant aspects of collateral constraints.
- Decomposition:
  - Institutional improvements are linked to reductions in the two modeled frictions (financial transaction costs and required rate of return) in ways identified via the model and regression methodology.

### Econometric and measurement considerations
- Addresses identification problems traditional to investment regressions by grounding estimation in a structural model and minimizing one-period ahead forecast errors.
- Concerns about Q being a “noisy” variable (inefficient stock markets, limited arbitrage, poor accounting) and model specification errors leading to measurement error are explicitly considered.
- Empirical tests confirm measurement errors are not significantly large relative to one-period ahead forecast errors in the methodology.
- Instrumental variable estimations are performed and show robustness of the main results.

### Contribution to literature
- Combines a structural Q-based investment model with cross-country institutional variation to identify which institutional differences give rise to firm-level financial frictions.
- Adds to literature on finance and macroeconomic fluctuations, firm-level credit constraints, and the role of institutions in economic and financial development by:
  - Explicitly decomposing institutional effects onto two forms of financial frictions.
  - Using large international firm-level data to exploit slowly changing cross-country institutional variation.

*Source: _wp10231 - 2. Benchmark Regression (excerpt) — IMF PDF chapter content provided*

### Section IV describes the data set used for this study; Section V presents the estimation

### _wp10231 - Section IV describes the data set used for this study; Section V presents the estimation

### Theoretical model: setup and key assumptions
- Model adapts Hayashi (1982), Abel and Blanchard (1986), Abel and Eberly (1994), Gomes (2001), and Hennessy, Levy, and Whited (2007).
- Timing: based on previous-period capital K– and beginning-of-period productivity ε, investment I is chosen, adjustment costs paid, new capital K formed immediately; production occurs using K and ε.
- Within-period “working capital” finance (credit lines or trade credit) is assumed to involve no financial transaction costs. External finance B is costly and determined at end-of-period when gross profits (return to capital) π are realized.
- Adjustment costs of investment: φ(I, K−, K, X, W, ε) are linearly homogeneous of degree one in I and K; X denotes time-varying predictable firm characteristics; W denotes time-invariant institutional quality.
- Financial transaction costs: λ(B, K−, K, X, W, ε) are linearly homogeneous in B and K and matter only when B>0.
- Production and labor assumptions: competitive labor market, constant-returns-to-scale production f, competitive wage o with marginal condition of employment, and competitive product markets.
- Productivity shocks ε are serially correlated with distribution P(ε+ | ε); firm observes ε but not aggregate/industry common components when choosing investment.
- Capital law of motion: K = (1 − δ)K− + I. (Equation (1))

### Investment and financing cost functions (parameterization)
- Investment adjustment cost functional form used:
  - φ(I, K) = c1(I/K) + c2(I/K)^2/2 + c3(I/K)^3/3 (presented as homogeneous convex specification with coefficients c2, c3). (Equation (10))
- Financial transaction cost functional form used:
  - λ(B, K) = b1(B/K) + b2(B/K)^2/2 + b3(B/K)^3/3 (generalized version of Gomes (2001)). (Equation (11))
- Interpretations:
  - b1 and b3 lower ⇒ lower financial transaction costs.
  - b2 captures size effect: financial transaction costs decrease with firm size; small firms pay a premium when K is small relative to B.

### Equilibrium marginal condition and law of motion for Tobin’s Q
- Firm value derivative V1 = Q (marginal Q). Optimality condition (see Appendix):
  - (Equation (5)) 11 2 2 )(1() r Q E Q Q δ λ π φ λ − − + − + − − = + . [printed in source as equation (5) — terms λ1, λ2 are partial derivatives of λ w.r.t. B and K; φ2 is ∂φ/∂K]
- Rearranged empirical formula used for testing:
  - (Equation (6)) [ ] 11 2 2 (1(1)) E Q r Q δ λ π φ λ − + − + + = + + + . (as presented in source)
- Notes:
  - When external finance B ≤ 0, marginal financial transaction cost terms λ1 and λ2 vanish from (6).
  - When investment I ≤ 0, investment adjustment cost φ2 drops from (6).
  - Value function has one state variable K plus predetermined X and time-invariant W.

### Relation to speed of adjustment of Q
- Rewriting (6) yields:
  - (Equation (7)) [ ] 1221 )(1(1) ( ) ) ( E Q r Q Q Q λ π δ φ λ δ − − − − − + + = − − + + . (as presented)
- Implications:
  - Speed of adjustment of Q depends on required rate of return r and marginal financial transaction cost λ1 with opposite signs.
  - Better institutions may lower r (speeding adjustment) but may also reduce λ1 (slowing adjustment); net effect on adjustment speed is theoretically ambiguous.
  - Size effect λ2 implies smaller firms pay higher fees; small-firm premium predicted to be lower in countries with better institutions.

### Average versus marginal Q
- Under linear homogeneity assumptions for φ and λ, marginal Q equals average Q (Hayashi result). This justifies using average Q in empirical work given the model’s homogeneity assumptions.

### Estimation methodology
- Core strategy: minimize one-period-ahead forecast errors by matching expected end-of-period Q to realized Q.
- Forecast error definition: ξ = Q − E[Q | ε]. (Equation (9))
  - One-period-ahead forecast error is serially uncorrelated even if productivity shocks are serially correlated; implies OLS estimates are unbiased and consistent (robust standard errors clustered at country-year used to correct for within-country correlation).
- General empirical specification (compact form):
  - Equation (8) expresses expected Q as function of lagged Q, interactions of X and W with Z (where Z = π1 − φ2 − λ1 − λ2, an object capturing marginal profits net of marginal costs), and other interaction terms; interaction terms identify effects of institutions on financial transaction cost function and required rate of return.
- Full estimation equation structure:
  - Equation (12) presents the detailed estimating equation with indicator functions for I>0 and B>0, interaction terms between Total Assets A (firm size), profits π, investment I, external finance B, firm characteristics X, and country institutional variables W.
  - Indicator functions:
    - Ψijkt = 1 if Iijkt > 0, 0 otherwise.
    - Χijkt = 1 if Bijkt > 0, 0 otherwise.
- Marginal return approximation used:
  - πijkt = (AAijkt − AAijkt−1)/AAijkt−1 (presented in source as approximation using Total Assets A).

### Identification and controls
- Institutional variables are time-invariant; their level effects are absorbed in country fixed effects; identification of institutional effects arises from interaction terms.
- Only external finance is interacted with profits (coefficient b1) because investment is not assumed to be influenced by current profits, while financing is.
- Controls include industry dummies and firm age in X to capture differences in required rate of return across industries and vintages.

### Data description (lead-in)
- All variable definitions, data sources and some sample statistics are provided in Table (referenced in source). Section IV contains the data set description; Section V presents estimation results; Section VI examines measurement error issues; Section VII concludes.

* _wp10231 - Section IV describes the data set used for this study; Section V presents the estimation_

### 1. We use firm level data from the Worldscope database of Thomson Financial. The data

### _wp10231 - 1. We use firm level data from the Worldscope database of Thomson Financial. The data

### Data, sample construction, and key sample sizes
- Worldscope firm-level data cover the period 1990 to 2007 for 48 countries and contain about 380,000 firm-year observations for which Q can be constructed.  
- The number of original firm-year observations (including those for which Q cannot be constructed) is about one million.  
- Sequential deletions and filters produce:  
  - about 300,000 firm-year observations with Q after initial cleaning;  
  - benchmark-regression sample of about 75,000 firm-year observations (sample shrinkage driven by unavailability of lagged Q and other variables, and by firm age availability).  
  - alternative, less restricted regressions use sample sizes of 74,272 and 147,711 observations.  
- The 48 sample countries are: Argentina, Australia, Austria, Belgium, Brazil, Canada, Chile, China, Colombia, Czech Republic, Denmark, Finland, France, Germany, Greece, Hong Kong, Hungary, India, Indonesia, Ireland, Israel, Italy, Japan, South Korea, Luxembourg, Malaysia, Mexico, Netherlands, New Zealand, Norway, Pakistan, Peru, Philippines, Poland, Portugal, Russia, Singapore, South Africa, Spain, Sri Lanka, Sweden, Switzerland, Taiwan, Thailand, Turkey, United Kingdom, United States, and Venezuela.

### Variable definitions and measurement choices
- Marginal profit, π1, proxied by increase in earnings divided by increase in total assets. Earnings (cash-flow measure) = Net Income before Extraordinary Items and Preferred Dividends + Interest Expense on Debt + Depreciation and Amortization. Alternative before-tax measure used: Operating Income + Depreciation and Amortization.
- Investment, I: benchmark = Capital Expenditure. Broader definition used in robustness checks = Capital Expenditure + Change in Cash and Short-Term Investment.
- External finance, B: benchmark = Capital Expenditure + Change in Cash and Short-Term Investment – Cash Flow from Operation – Decrease in Inventory – Decrease in Receivables – Increase in Payables (adds change in cash to Rajan and Zingales (1998) definition). Narrower external finance concept (robustness) = net increase in Total Debt + Net Proceeds from Sale/Issue of Common and Preferred Stocks (excludes trade credit).
- Q defined as (Market Capitalization + Total Asset – Total Equity) / Total Asset. Q is measured at fiscal-year end.
- Capital stock measure: Total Asset used as broad measure of capital. Debt valued at par.
- Firm controls: industry dummies and firm age (Founded Date). Firm size omitted as endogenous.
- Country-specific real short-term risk free rate = short-term government Treasury bill rate minus CPI inflation.
- Macroeconomic variables: CPI inflation and macroeconomic volatility measured as standard deviation of real GDP growth for 1995-2006. CPI and real growth rates from World Development Indicators. Short-term Treasury bill rates mainly from IMF’s International Financial Statistics.

### Institutional measures (W) used in benchmark regression
- Corporate governance (CorpGov): shareholder (anti-director) rights (La Porta et al., 1998) — benchmark corporate governance measure.  
- Creditor rights (Creditor): strength of legal protection for lenders and borrowers (World Bank, 2008a).  
- General institutional quality (Institution): property rights measure of La Porta et al. (1998).  
- Product market competition (Compet): measure of trade barriers (World Economic Forum, 2007).  
- Financial market development (FinMkt): stock market-capitalization-to-GDP for 2005 (World Bank, 2008b).  
- Interaction specification: interaction terms of lagged Q with X, W, industry dummies, Age, RiskFreeRate, Inflation, Macro, CorpGov, Creditor, Institution, Compet, FinMkt as shown in Equation (13) and level effects in Equation (14).

### Benchmark estimation results — main findings (fixed effects, Table 2 and discussion)
- Required rate of return: better corporate governance (shareholder protection) is associated with a significantly lower required rate of return; estimated coefficient on lagged Q interaction with corporate governance = -0.0433.  
  - Interpretation provided by authors: a one-standard-deviation improvement (increase of 1.3) in the anti-director rights index would lower Q in the next period by 0.16 for the average firm (with a Q of 3).
- Product market competition: more intense competition associated with a higher required rate of return (coefficient 0.0772), significant only at the 10 percent level.
- Firm age: higher firm age associated with a higher required rate of return but effect is negligible — Q increases by 0.01 for a firm one year older with “average characteristics.”
- Financial transaction costs (columns 2–4 of Table 2):  
  - Corporate governance increases the intercept of the marginal financial transaction cost curve (small effect; significant at 10 percent).  
  - Crucially, better corporate governance reduces the extra costs that small firms have to pay (the small-firm premium): a one-standard-deviation improvement in investor protection (1.3) lowers this premium by about 3 percent of assets.  
  - Better general institutional quality (one-standard-deviation improvement, 0.8) lowers the small-firm premium by about 4 percent of assets.  
  - Creditor rights increase the small-firm premium in some specifications, but the statistical significance is low and not robust.  
  - Other institutional factors generally do not show robust effects on firm-level financial transaction costs.
- Speed of convergence of Q: Q moves more rapidly toward firm-specific steady state in countries with better corporate governance (due to lower required rate of return). Differences in adjustment speed between small and large firms are smaller where shareholder protection and general institutional quality are better.

### Robustness checks and alternative specifications
- Accounting and measurement robustness: Tables 3a–3c replicate benchmark results when using: before-tax income (Table 3a), broader investment definition including financial investments (Table 3b), and narrower external finance excluding trade credit (Table 3c). Using narrower external capital makes effects less significant except the reduction in the small-firm premium associated with better corporate governance.
- One-by-one institutional regressions (Table 4): estimating each institutional factor without others yields virtually same results as benchmark, except product market competition and financial market development become significant in one-by-one regressions — indicating correlations among institutional variables but no systematic bias.
- Alternative institutional proxies (Table 5):  
  - Spamann’s version of anti-director rights reverses signs (Spamann argues index differs from La Porta et al., 1998).  
  - Djankov et al. (2008) anti-self-dealing index broadly replicates benchmark results except corporate governance no longer matters for required rate of return.  
  - De Nicolo, Laeven, and Ueda (CGQ) broadly replicate benchmark results except effect on required rate of return differs (mean scales differ).  
  - Narrower creditor-rights measures (Djankov, McLiesh, and Shleifer, 2007) associated with a higher small-firm premium in some specifications; survey-based bankruptcy-efficiency measures (Djankov, Hart, McLiesh, and Shleifer, 2008) yield benchmark results (no effect of creditor rights on Q dynamics).  
  - Alternative measures of general institutional quality (Rule of Law; Trust in People) reduce the small-firm premium but have mixed/fragile effects on curvature of financial transaction cost function.  
  - Alternative competition proxies (business entry rates; costs of start-ups) show effects qualitatively similar to corporate governance on small-firm premium and curvature.
- Sample selection robustness: excluding Age (to increase sample size from 74,272 to 147,711) yields broadly similar results, except corporate governance no longer matters for required rate of return in that specification (Table 6). Results unchanged when using all firms or manufacturing firms only.
- Real adjustment (Table 7): institutional factors affect real investment adjustment costs. Good corporate governance lowers rate at which costs increase with size (c2), but increases curvature (c3) so small new investments cost less while big new investments cost more. Overall effect on investment (relative to assets) is ambiguous because intercept not identified.

### Instrumental variable estimation and measurement error tests
- Measurement error assessment: four forms of measurement errors considered (stock price noise, accounting issues, marginal vs average Q, timing assumptions). Tests for serial correlation in OLS errors fail to reject zero serial correlation; conclusion: measurement errors, if any, are very small relative to one-period-ahead forecast errors.
- IV strategy: twice-lagged Q used as instrument for lagged Q; fitted values used for interaction instruments as per Wooldridge (2002, p.237). Standard two-stage least squares with clustering at country-year. Tests reported: Kleibergen-Paap rk Wald F statistic = 5.14; Anderson-Rubin Wald test F(71,607)=165.17 (rejecting under-identification).
- IV results (Table 8): broadly replicate OLS-fixed effects results. Notable IV differences: corporate governance no longer affects required rate of return significantly but reduces curvature of financial frictions significantly; small-firm premium lower where corporate governance and general institutional quality are better; creditor rights do not have robust effects; more developed financial markets reduce required rate of return (significant at 10 percent). IV estimates reported include coefficient for corporate governance on required return = -0.0209 (Table 8).

### Main interpretation and policy-relevant conclusions
- Principal empirical finding: good corporate governance (strong shareholder protection) and, less robustly, good general institutional quality are associated with lower financial transaction costs, especially by reducing the small-firm premium; good corporate governance often associated with a lower required rate of return. Together, these institutional features imply more efficient capital allocation across firms and faster adjustment of Q to steady state.
- Creditor rights (a narrower contractual enforcement concept) play a lesser role in this sample of listed firms. Possible explanations offered: listed firms can in theory issue equity, so marginal cost of external finance determined by equity; good corporate governance necessary for efficient use of funds regardless of finance source; credit constraints may be binding but weaker creditor rights do not strongly influence constraints for listed firms.
- Policy implications emphasized by authors: improvements in corporate governance are a clear channel to reduce financial frictions and benefit firms. Strengthening creditor rights may be less critical for listed firms, though bankruptcy restructuring schemes (e.g., Chapter 11–like reform) could still enhance ex-post contracting efficiency and be a desirable policy.
- Caveats and research agenda: approach relies on a canonical investment theory with somewhat reduced-form financial frictions; better theoretical microfoundations linking institutions (e.g., moral hazard) to measurable financial frictions would improve empirical identification — suggested as an agenda for future research.

### Key exact numeric highlights and statistics (preserved)
- Data period: 1990 to 2007.  
- Countries: 48.  
- Original firm-year observations: about one million.  
- Firm-year observations with Q after deletions: about 300,000.  
- Benchmark regression sample: about 75,000 firm-year observations (one reported benchmark: 74,272 observations).  
- Less restricted sample (excluding Age): 147,711 observations.  
- Benchmark coefficient: corporate governance (lagged Q interaction on required return) = -0.0433.  
- Standard-deviation interpretation: one-standard-deviation improvement in anti-director rights index = increase of 1.3; implied Q reduction next period = 0.16 for average firm with Q = 3.  
- Small-firm premium reductions: corporate governance one-standard-deviation (1.3) lowers premium by about 3 percent of assets; general institutional quality one-standard-deviation (0.8) lowers premium by about 4 percent of assets.  
- Table 2 product market competition coefficient on required return = 0.0772 (significant at 10 percent).  
- Table 6 observations reported: 147,711; R-squared = 0.435.  
- Instrumental variable estimation (Table 8): corporate governance coefficient on required return = -0.0209; Kleibergen-Paap rk Wald F statistic = 5.14; Anderson-Rubin Wald test F(71,607)=165.17.  
- Tests for autocorrelation of fixed-effect residuals: AR(1) coefficient = 0.200 with standard error 0.162 (t-statistic 1.64) when testing theoretical null of –1/(T–1) with T=18; alternative test using all observations AR(1) = 0.050 with standard error 0.054 (t-statistic 0.91).  
- Macroeconomic volatility measure used: standard deviation of real GDP growth for period 1995-2006.  
- CPI and real growth sources: World Development Indicators; Treasury bill rates: IMF’s International Financial Statistics.

*Italic source attribution: Excerpted from the IMF working paper PDF _wp10231 (Worldscope firm-level data, 1990–2007) — content used exactly as supplied.*

### REFERENCES

### _wp10231 - REFERENCES

### Investment theory and corporate finance
- Abel, Andrew B., and Olivier J. Blanchard, 1986, “The Present Value of Profits and Cyclical Movements in Investment,” Econometrica, Vol. 54, No. 2, pp. 249-73.
- Abel, Andrew B., Avinash Dixit, Janice C. Eberly, and Robert Pindyck, 1996, “Options, the Value of Capital, and Investment,” Quarterly Journal of Economics, Vol. 111, No. 3, pp. 753-77.
- Abel, Andrew B, and Janice C. Eberly, 1994, “A Unified Model of Investment under Uncertainty,” American Economic Review, Vol. 84, No. 5, pp. 1369-384.
- Abel, Andrew B, and Janice C. Eberly, 2008, “How Q and Cash Flow Affect Investment without Frictions: An Analytic Explanation,” unpublished manuscript, Wharton School and Kellogg School.
- Barnet, Steven A., and Plutarchos Sakellaris, 1999, “A New Look at Firm Market Value, Investment, and Adjustment Costs,” The Review of Economics and Statistics, Vol. 81, No. 2, pp. 250-60.
- Blanchard, Olivier, Changyong Rhee, and Lawrence Summers, 1993, “The Stock Market, Profit, and Investment,” The Quarterly Journal of Economics, Vol. 108, No. 1, pp. 115-36.
- Gomes, Joao F., 2001, “Financing Investment,” American Economic Review, Vol. 91, No. 5, pp. 1263-285.
- Hayashi, Fumio, 1982, “Tobin's Marginal q and Average q: A Neoclassical Interpretation,” Econometrica, Vol. 50, No. 1, pp. 213-24.
- Hennessy, Christopher A., Amnon Levy, and Toni M. Whited, 2007, “Testing Q Theory with Financing Frictions,” Journal of Financial Economics, Vol. 83, No. 3, pp. 691-717.
- Mussa, Michael, 1977, “External and Internal Adjustment Costs and the Theory of Aggregate and Firm Investment,” Economica, Vol. 44, No. 174, pp. 163-78.
- Myers, Stewart, and Nicholas Majluf, 1984, “"Corporate Financing and Investment Decisions When Firms Have Information That Investors Do Not Have," Journal of Financial Economics, Vol. 13, pp. 187–221.
- Tobin, James, 1969 “A General Equilibrium Approach to Monetary Theory,” Journal of Money Credit and Banking, Vol. 1, No. 1, pp 15-29.

### Financial frictions, credit cycles, and intermediation
- Bernanke, Ben, and Mark Gertler, 1989, “Agency Costs, Net Worth, and Business Fluctuations,” American Economic Review, Vol. 79, No. 1, pp. 14-31.
- Chari, V. V., Patrick Kehoe, and Ellen McGrattan, 2006, "Business Cycle Accounting," Econometrica, Vol. 75, Vol. 3, pp. 781-836.
- Christiano, Laurence, Roberto Motto, and Massimo Rostagno, 2010, "Financial Factors in Business Cycles," ECB Working Paper, No. 1192.
- Gertler, Mark, and Nobuhiro Kiyotaki, 2010, "Financial Intermediation in Credit Policy in Business Cycle Analysis," Handbook of Monetary Economics, manuscript.
- Kiyotaki, Nobuhiro, and Moore, John 1997, "Credit Cycles," Journal of Political Economy, Vol. 105, No 2, pp. 211–48.
- Khan, Aubhik, and Julia K. Thomas, 2008, “Idiosyncratic Shocks and the Role of Nonconvexities in Plant and Aggregate Investment Dynamics,” Econometrica, Vol. 76, No. 2, pp. 395-436.
- Lorenzoni, Guido, and Karl Walentin, 2007, "Financial Frictions, Investment, and Tobin's q," NBER Working Paper, No. 13092.
- Philippon, Thomas, 2009, “The Bond Market’s Q,” Quarterly Journal of Economics, Vol. 124, No. 3, pp. 1011–056.
- Townsend, Robert M., 1979 "Optimal Contracts and Competitive Markets with Costly State Verification," Journal of Economic Theory, Vol. 21, No. 2, pp. 265-93.
- Paulson, Anna L., and Robert M. Townsend, 2005, “Financial constraints and entrepreneurship: Evidence from the Thai financial crisis,” Federal Reserve Bank of Chicago Economic Perspectives, Vol. 3, pp. 34-48.
- Paulson, Anna L., and Robert M. Townsend, 2008, “Entrepreneurial Heterogeneity: Insights from the Thai Financial Crisis,” unpublished manuscript, FRB Chicago and MIT.
- Acharya, Viral V., Jean M. Imbs, and Jason Sturgess, 2010, “Finance and Efficiency: Do Bank Branching Regulations Matter?” Review of Financial Studies, forthcoming.
- Karaivanov, Alexander, Jesus Saurina, Robert M. Townsend, and Sonia Ruano, 2010, “No Bank, One Bank, Several Banks: Does It Matter for Investment?” a paper presented at the 10th Econometric Society World Congress in Shanghai.

### Financial development, market structure, and growth
- Allen, Franklin, and Douglas Gale, 2000, Comparing Financial Systems, Cambridge, MIT Press.
- Beck, Thorsten, Ross Levine, and Norman Loayza, 2000, “Finance and the Sources of Growth,” Journal of Financial Economics, Vol. 58, No. 1-2, pp. 261-300.
- Greenwood, Jeremy, and Boyan Jovanovic 1990, “Financial Development, Growth, and the Distribution of Income,” Journal of Political Economy, Vol. 98, No. 5, pp. 1076–107.
- Rajan, Raghuram G., and Luigi Zingales, 1998, “Financial Dependence and Growth,” American Economic Review, Vol. 88, No. 3, pp. 559-86.
- Levine, Ross, 2005, “Finance and Growth: Theory and Evidence,” in Philippe Aghion and Steven Durlauf, eds., Handbook of Economic Growth. The Netherlands: Elsevier Science.
- Wurgler, Jeffery, 2000, “Financial Markets and the Allocation of Capital,” Journal of Financial Economics, Vol. 58, No. 1-2, pp. 187-214.
- Acemoglu, Daron, and Fabrizio Zilibotti, 1997, "Was Prometheus Unbound by Chance? Risk, Diversification, and Growth," Journal of Political Economy, Vol. 105, No. 4, pp. 709-51.
- Greenwood, Robin, 2005, “Aggregate Corporate Liquidity and Stock Returns,” Harvard Business School Working Paper, No. 05-014.

### Corporate governance, law, and institutions
- De Nicolo, Gianni, Luc Laeven, and Kenichi Ueda, 2008, “Corporate Governance Quality: Trends and Real Effects,” Journal of Financial Intermediation, Vol. 17, No. 2, pp. 198–228.
- Djankov, Simeon, Oliver Hart, Caralee McLiesh, and Andrei Shleifer, 2008, “Debt Enforcement around the World,” Journal of Political Economy, Vol. 116, No. 6, pp. 1105-49.
- Djankov, Simeon, Caralee McLiesh, Andrei Shleifer, 2007, “Private Credit in 129 Countries,” Journal of Financial Economics, Vol. 84, No. 2, pp. 299–329.
- Djankov, Simeon, La Porta, Rafael, Florencio Lopez-de-Silanes, and Andrei Shleifer, 2008, “The Law and Economics of Self-Dealing,” Journal of Financial Economics, Vol. 88, No. 3, pp. 430–65.
- Doidge, Craig G., Andrew Karolyi, and Rene M. Stulz, 2007, “Why Do Countries Matter So Much for Corporate Governance?” Journal of Financial Economics, Vol. 86, No. 1, pp. 1–39.
- La Porta, Rafael, Florencio Lopez-de-Silanes, and Andrei Shleifer, 2008, “The Economic Consequences of Legal Origins,” Journal of Economic Literature, Vol. 46, No. 2, pp. 285–332.
- La Porta, Rafael, Florencio Lopez-de-Silanes, and Robert Vishny, 1998, “Law and Finance,” Journal of Political Economy, Vol. 106, No. 6, pp. 1113–155.
- Morck, Randall, Daniel Wolfenzon, and Bernard Yeung, 2005, "Corporate Governance, Economic Entrenchment, and Growth," Journal of Economic Literature, Vol. 43, No. 3, pp. 655-720.
- Spamann, Holger, 2010, “The Antidirector Rights Index Revisited,” Review of Financial Studies, Vol. 23, No. 2, pp. 467-86.
- Dittmar, Amy, Jan Mahrt-Smith, and Henri Servaes, 2003, “International Corporate Governance and Corporate Cash Holdings,” Journal of Financial and Quantitative Analysis, Vol. 38, No. 1, pp.111-33.
- Tosi, Henry L., Steve Werner, Jefferey P. Katz, and Luis R. Gomez-Mejia, 2000, "How Much Does Performance Matter? A Meta-analysis of CEO Pay Studies," Journal of Management, Vol. 26, pp. 301-39.

### Misallocation, firm heterogeneity, and entrepreneurship
- Hsieh, Chang-Tai, and Peter Klenow 2009, “Misallocation and Manufacturing TFP in China and India,” Quarterly Journal of Economics, Vol. 124, No. 4, pp. 1403-1448.
- Cagetti, Marco, and Mariacristina De Nardi, 2006, "Entrepreneurship, Frictions, and Wealth," Journal of Political Economy, Vol. 114, No. 5, pp. 835-70.
- Banerjee, Abhijit V., and Andrew F. Newman (1993), “Occupational Choice and the Process of Development,” Journal of Political Economy, Vol. 101, No. 2, pp. 274-98.
- Parente, Stephen, and Edward Prescott, 2000, Barriers to Riches, MIT Press, Cambridge.
- Paulson, Anna L., and Robert M. Townsend, 2005, “Financial constraints and entrepreneurship: Evidence from the Thai financial crisis,” Federal Reserve Bank of Chicago Economic Perspectives, Vol. 3, pp. 34-48.
- Paulson, Anna L., and Robert M. Townsend, 2008, “Entrepreneurial Heterogeneity: Insights from the Thai Financial Crisis,” unpublished manuscript, FRB Chicago and MIT.
- Klapper, Leora, and Inessa Love 2010, “The Impact of the Financial Crisis on New Firm Creation,” unpublished manuscript, World Bank.

### Measurement, econometrics, and methodology
- Alvarez, Fernando, and Nancy L. Stokey, 1998, "Dynamic Programming with Homogeneous Functions," Journal of Economic Theory, Vol.82, No.1, pp.167-89.
- Arellano, Manuel ,and Stephen Bond, 1991, "Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations," Review of Economic Studies, Vol. 58(2), pp. 277-97.
- Baum, Christopher F., Mark E. Schaffer, and Steven Stillman, 2007, “Enhanced Routines for Instrumental Variables/Generalized Method of Moments Estimation and Testing,” The Stata Journal, Vol. 7, pp. 465-506.
- Duffie, Darrell, 2010, "Presidential Address: Asset Price Dynamics with Slow-Moving Capital,” Journal of Finance, Vol. 65, pp. 1237–267.
- Erickson, Timothy, and Toni M. Whited, 2000, "Measurement Error and the Relationship between Investment and Q," Journal of Political Economy, Vol. 108, No. 5, pp. 1027-057.
- Almeida, Heitor, Murillo Campello, and Antonio F. Galvao Jr., 2010, “Measurement Errors in Investment Equations,” Review of Finance, forthcoming.
- Wooldridge, Jeffrey M., 2002, Econometric Analysis of Cross Section and Panel Data. MIT Press, Cambridge.
- Almeida, Heitor, Murillo Campello, and Michael S. Weisbach, 2010, “Corporate Financial and Investment Policies when Future Financing is Not Frictionless,” Journal of Corporate Finance, forthcoming.
- Almeida, Heitor, and Murillo Campello, 2010, “Financing Frictions and the Substitution between Internal and External Funds,” Journal of Financial and Quantitative Analysis, forthcoming.
- Hennessy, Christopher A., Amnon Levy, and Toni M. Whited, 2007, “Testing Q Theory with Financing Frictions,” Journal of Financial Economics, Vol. 83, No. 3, pp. 691-717.
- Barnet, Steven A., and Plutarchos Sakellaris, 1999, “A New Look at Firm Market Value, Investment, and Adjustment Costs,” The Review of Economics and Statistics, Vol. 81, No. 2, pp. 250-60.

### Data sources, indexes, and reports
- International Monetary Fund, 2009, International Financial Statistics, Washington, D.C.
- Kaufmann, Daniel, Aart Kraay, and Massimo Mastruzzi, 2003, “Governance Matters III: Updated Governance Indicators for 1996–02,” Working paper, Washington, D.C. World Bank.
- World Economic Forum, 2007, The Global Competitiveness Report 2006–2007, Oxford University Press, New York.
- World Bank, 2008a, Doing Business, Washington, D.C.
- World Bank, 2008b, World Development Indicators, Washington, D.C.
- World Values Survey 1990-1993, available at www.worldvaluessurvey.org

*Source: _wp10231 - REFERENCES*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2010/_wp10231.pdf_
