## 2. Monetary and Fiscal Policies

## Source details

**Canonical URL:** [2. Monetary and Fiscal Policies](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp1145.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp1145.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp1145.pdf.json)

---

### I. Introduction and paper focus
- Historical framing: initial skepticism from Modigliani and Miller (1958) vs later acceptance that financial frictions (information asymmetries, bankruptcy risk) matter for real decisions.
- Focus: supply-side credit market frictions and real effects of financial sector recapitalization policies implemented during the crisis.
- Contributions:
  - New cross-country data quantifying intensity of government resolution policies (not just binary indicators).
  - Identification: use global financial crisis as an exogenous shock to credit supply and exploit exogenous variation in firms’ dependence on external financing (Rajan and Zingales (1998) measure).
  - Emphasis on bank recapitalization as channel affecting firms’ value added via credit supply.
  - Controls for other government interventions (guarantees, asset purchases, liquidity support) and for monetary and fiscal policy actions; firm-, industry-, and country-level controls and fixed effects used to address simultaneity.
- Data scope and timing:
  - Outcome: real growth in value added at the firm level (annual growth rate in real value added in percent) for 2009.
  - Policy window: interventions announced between September 2008 and March 2009; used amounts up to end-2009.
  - Sample: listed manufacturing firms from 50 countries (advanced and emerging market economies).

### II. Policy intervention measures
- Classification of government interventions (built on Laeven and Valencia (2008)):
  - Bank recapitalization: direct capital injections (public sector recapitalizations, common and preferred stocks). Took place in roughly half of countries in sample; many countries used on average roughly half of initially committed resources. Primary variable of interest.
  - Asset purchases: measured only when conducted by the Treasury/Ministry of Finance; excludes central bank purchases of mortgage-backed securities and long-term government bonds (captured by change in money base).
    - Example: Norway announced asset purchases "in the order of 13.8 percent of GDP".
  - Guarantees and liquidity support: guarantees on bank liabilities or assets; improve access to funding; early containment phase measures.
  - Liquidity injections: proxied by change in central bank claims on financial institutions normalized by GDP (gross claims = sum of IFS lines 12E..ZF..., 12F..ZF..., and 12G..ZF...); negative results set to zero.
- Cross-country variation in macro responses:
  - Fiscal impulses range from "a 9 percent expansion in Nigeria" to "a 16 percent contraction in Latvia".
  - Most countries relaxed monetary policy; some tightened due to currency pressures.
- Distinctive sample features:
  - Quantifies scope and intensity (percent of GDP) of interventions.
  - Includes crisis and "no crisis" countries; preemptive policies included.

### III. Empirical strategy and identification
- Identification challenge: policy responses endogenous to macro environment.
- Strategy:
  - Interaction approach following Rajan and Zingales (1998): exploit within-country, across-industry variation in industry-level external finance dependence (exf) interacted with country-level interventions (G) to identify differential effects.
  - Baseline regression: dependent variable g_ik = annual growth rate in real value added (percent) of firm i in country k during 2009; primary regressor is exf_j × G_k where G_k includes fiscal policy, monetary policy, and bank recapitalization measured over September 2008–March 2009.
  - Controls: firm-level controls X_ki (size, leverage, tangibility, Tobin’s Q), industry indicators α_j, country indicators α_k, clustered standard errors.
- Measurement details:
  - External financial dependence: (capital expenditure − operating cash flow) / capital expenditure; baseline uses U.S. firms over the 1980’s (exf80); robustness uses 1980–2006 (exf8006). U.S. firms dropped from regressions.
  - Value added: EBITDA + labor expense; sample restricted to firms with fiscal year end between October and December; convert to local currency and deflate by country CPI inflation.
  - Fiscal1: change in structural balance (announced discretionary fiscal stimuli net effect on fiscal balances over 2008-10), excluding pre-crisis announcements and fiscal outlays associated with financial sector support.
  - Fiscal2: fiscal impulse following Fedelino et al. (2009) using April 2010 WEO; outcome-based, not purged of financial sector support.
  - Monetary policy: change in nominal short-term interest rates over September 2008–March 2009; alternative: change in money base over same period.
  - Bank intervention data: primary source Laeven and Valencia (2010), complemented by IMF Staff reports, central bank and ministry announcements. Recapitalization coded as committed public recapitalization amounts (announced Sep 2008–Mar 2009; used amounts up to end-2009).

### IV. Main empirical findings
- Key statistically significant results:
  - Bank recapitalization (exf × recapa) has a positive and statistically significant effect on value added growth of firms in industries more dependent on external finance.
    - Table 4 (column (4)): exf * recapa = 1.290** (0.555).
  - Discretionary fiscal policy (exf × fiscal1) has a positive and statistically significant effect:
    - Table 4 (column (4)): exf * fiscal1 = 1.357*** (0.523).
  - Monetary policy (exf × monpol) is statistically insignificant.
    - Table 4 (column (4)): exf * monpol = 0.0660 (1.015).
  - Guarantees and asset purchases individually statistically insignificant; all financial sector resolution policies jointly significant at the 10 percent level.
  - Larger firms (Size) experienced higher growth in value added:
    - Table 4 Size coefficients across columns ~1.329*** (0.322) to 1.333*** (0.291).
- Additional findings:
  - Correlation Fiscal1 and Recapa = -0.14.
  - Correlation financial dependence (1980–1989 vs 1980–2006) = 0.73.
  - Correlation financial dependence and Tobin’s Q = 0.18.
  - Table 1 selected country examples (values reproduced exactly for reference): USA: Recapitalizations pledged 5.1, used 3.1; Liquidity Support 10.4; Change in ST interest rate 1.8; Public Debt 60.9; Fiscal stimulus (fiscal1) 5.6; Fiscal stimulus (fiscal2) 7.2.

### V. Quantitative importance (economic magnitudes)
- Bank recapitalization:
  - Column 4 estimates imply a one standard deviation increase in the interaction between financial dependence and bank recapitalization increases real growth in value added by 2.3 percentage points (ceteris paribus).
  - A firm at the 75th percentile of financial dependence would have real growth in value added 1.1 percentage points higher than a firm at the 25th percentile of financial dependence if located in a country at the 75th percentile vs the 25th percentile of bank recapitalizations (ceteris paribus).
- Fiscal policy:
  - Column 4 estimates imply a one standard deviation increase in the interaction between financial dependence and fiscal expenditure increases real growth in value added by 2.5 percentage points (ceteris paribus).
  - A firm at the 75th percentile of financial dependence would have real growth in value added 1.4 percentage points higher than a firm at the 25th percentile of financial dependence if located in a country at the 75th percentile vs the 25th percentile of fiscal expenditures (ceteris paribus).
- Comparative magnitude:
  - Bank recapitalization and fiscal policy coefficients both about 1.3 (measures expressed in terms of GDP), implying a 1 percentage point of GDP increase in either would enhance growth of financially dependent firms by comparable amounts.
- Context: average firm growth in sample = -4.02 percent (Table 3).

### VI. Robustness checks and extensions
- Weighted least squares (observations weighted by inverse of number of firm observations in each country-industry pair) leaves baseline qualitative results unchanged; differential fiscal effect = 1.72 compared to 1.36 in baseline.
- Additional firm controls (lev, ppe) included:
  - Lev (column 2 Table 5): -2.578 (2.911) (not statistically significant).
  - Ppe (column 2 Table 5): -9.351** (4.748) (statistically significant).
  - Main fiscal policy and bank recapitalization effects remain robust.
- Alternative intervention definitions:
  - Using Fiscal2 (outcome-based) yields statistically significant fiscal coefficient while bank recapitalization may lose significance in that specification.
  - Using change in monetary base normalized by GDP yields statistically insignificant monetary policy effect.
  - Using actual used amounts rather than announced amounts for recapitalization yields similar results; recapitalization remains significant with somewhat larger economic effect.
  - Emergency liquidity support shows no disproportionate effect on growth of financially dependent firms between September 2008 and March 2009.
- Systemic banking crises:
  - Interacting exf with a 2009 systemic banking crisis dummy suggests financially dependent firms are particularly hurt during systemic crises (effect not statistically significant); inclusion of crisis dummy does not alter fiscal and recapitalization impacts.
- Financial development and capital markets:
  - Financial development (private credit + stock market capitalization to GDP in 2006) has independent growth-enhancing effect for financially dependent firms driven by stock market capitalization (capital markets as a “spare wheel”).
- Working capital and short-term financing:
  - Industry liquidity needs (liq) and cash conversion cycle (ccc) indicate bank recapitalization disproportionately enhances growth of firms dependent on working capital financing; fiscal policy does not disproportionately benefit working-capital-dependent firms.
- Growth opportunities:
  - Interactions between interventions and firm Tobin’s Q (2006) are not significant, supporting a channel through financial dependence rather than changes in growth opportunities.

### VII. Interpretation and conclusions
- Supply-side financial frictions matter: evidence consistent with real effects of disruptions in credit supply alleviated by bank recapitalization.
- Mechanisms:
  - Bank recapitalization mitigates negative credit supply shocks and disproportionately benefits financially dependent firms.
  - Discretionary fiscal policy boosts domestic demand and disproportionately benefits financially dependent firms hurt most by credit supply disruptions; the approach cannot fully disentangle supply-side vs demand-side channels for fiscal policy.
  - Monetary policy and anticipated liquidity measures were less effective in producing differential benefits for financially dependent firms, plausibly because many liquidity facilities were anticipated and liquidity measures do not repair solvency-driven capital shortfalls.
- Main conclusion: statistically significant, economically important, and robust evidence that bank recapitalization policies and discretionary fiscal policy boosted growth for financially dependent firms in the sample of 50 countries. Results likely underestimate effects for smaller, more bank-dependent firms not in the listed-firm sample.

### Key statistics and preserved values (verbatim excerpts)
- Sample size (firm-level observations): 2,902 / 2,903 (reported across tables).
- Average growth in real value added (Table 3): -4.02 (Mean); sample percentiles: p25 = -24.92; p75 = 15.61; Max = 98.73; Min = -99.99.
- Recapa (Bank recapitalization, announced (percent of GDP)) mean: 2.03; Std. Dev.: 4.00; Min: 0.00; p25: 0.00; p75: 2.20; Max: 24.00.
- Liq Support mean (percent of GDP): 5.97; Std. Dev.: 10.82; Max: 57.54.
- Fiscal1 (announced discretionary fiscal policy, percent of GDP) mean: 1.02; Std. Dev.: 3.90; Min: -16.30; p25: 0.25; p75: 3.00; Max: 9.00.
- Fiscal2 (used fiscal impulse, percent of GDP) mean: 1.96; Std. Dev.: 3.19; Min: -7.20; p25: 0.70; p75: 3.20; Max: 13.90.
- Monpol (Change in short term rates, percent) mean: 1.70; Std. Dev.: 2.55; Min: -9.40; p25: 1.11; p75: 2.42; Max: 6.74.
- Correlation Fiscal1 and Recapa: -0.14.
- Correlation exf (1980–1989 vs 1980–2006): 0.73.
- Correlation exf and Tobin’s Q: 0.18.
- Economic magnitude examples:
  - One standard deviation increase in exf × recapa → +2.3 percentage points in real growth in value added (ceteris paribus).
  - One standard deviation increase in exf × fiscal expenditure → +2.5 percentage points in real growth in value added (ceteris paribus).
  - A firm at the 75th vs 25th percentile of exf in a country at 75th vs 25th percentile of recapa: +1.1 percentage points in real growth (ceteris paribus).
  - Equivalent comparison for fiscal policy: +1.4 percentage points (ceteris paribus).
- Size (log total assets) coefficient examples:
  - Table 4 Size ≈ 1.329*** (0.322) to 1.333*** (0.291) across specifications.

*Italic source attribution: IMF Working Paper chapter "2. Monetary and Fiscal Policies" (excerpt provided, PDF filename _wp1145).*

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

### _wp1145 - References .................................................................................................................................... 22

### References
- References .................................................................................................................................... 22

### Tables
- 1.  Government Interventions During the Global Financial Crisis .............................................. 25
- 2.  Summary Statistics.................................................................................................................. 27
- 3.  External Financial Dependence Across Industries in the U.S. 1980-89 and 1980-2006 ........ 28
- 4.  Government Interventions, Financial Dependence, and Real Growth in Value Added ......... 29
- 5.  Alternative Specifications ....................................................................................................... 30
- 6.  Other Country Characteristics................................................................................................. 31
- 7.  Other Industry Characteristics ................................................................................................ 32

### Figures
- 1.  Financial Policies .................................................................................................................... 34

*_wp1145 - References_*

### 2.  Monetary and Fiscal Policies .......................................................................................

### 2.  Monetary and Fiscal Policies

### I. INTRODUCTION
- Historical context:
  - "Three decades ago" skepticism prevailed that financial factors mattered for real activity, based on Modigliani and Miller (1958).
  - Acceptance grew that financial frictions (information asymmetries, bankruptcy risk) imply financial factors matter for real decisions.
- Paper’s focus and contribution:
  - Focus on supply-side credit market frictions and real effects of financial sector recapitalization policies implemented during the crisis.
  - Collects new cross-country data on resolution policies that capture the intensity of government measures across countries (moving beyond binary treatment indicators).
  - Novel identification strategy: uses the global financial crisis as a shock to the supply of credit and exploits exogenous variation in firms’ dependence on external financing via the Rajan and Zingales (1998) measure.
  - Identification emphasizes recapitalization measures that targeted the financial sector directly; argument: bank recapitalization affects firms’ value added through the supply of credit.
  - Controls for other government interventions (guarantees, asset purchases, liquidity support) and for monetary and fiscal policy actions; addresses simultaneity via country and industry fixed effects and firm-level controls.
- Data scope and timing:
  - Outcome: real growth in value added, constructed at the firm level.
  - Compare value added growth over the year 2009 in response to crisis policies announced during September 2008 to March 2009.
  - Sample: a large cross section of firms from 50 countries in both advanced and emerging market economies.
- Key summary findings (as reported in this section):
  - Recapitalization of the banking sector "turns out to significantly and disproportionately affect the growth performance of financially dependent firms".
  - Other financial sector interventions—guarantees, asset purchases, liquidity support—are "not individually significant, but all financial sector resolution policies are jointly significant".
  - "Fiscal policy was much more effective than monetary policy in mitigating the effects of the crisis on financially dependent firms".
  - Results are robust to numerous checks.

### II. POLICY INTERVENTION MEASURES
- Conceptual link:
  - A negative bank capital shock causes a capital crunch, disrupting the supply of credit (Peek and Rosengren, 1995; Valencia, 2008).
  - Focus on policies that replenish bank capital; but monetary and fiscal policy may also alleviate impact via aggregate demand and asset prices.
- Classification and measures (built on Laeven and Valencia (2008)):
  - Direct capital injections (public sector recapitalizations, common and preferred stocks):
    - Took place in roughly half of the countries in the sample (Figure 1 referenced).
    - Many countries pledged significant amounts, using on average roughly half of the initially committed resources.
    - This is the "key variable of interest".
  - Asset purchases:
    - Implemented by Treasuries or central banks; measure here includes only those conducted by the Treasury.
    - Excludes central bank purchases of mortgage-backed securities and long-term government bonds (effects captured by alternative monetary policy measure: change in the money base).
    - Example magnitude: Norway announced asset purchases "in the order of 13.8 percent of GDP".
  - Guarantees and liquidity support:
    - Typically deployed in early crisis containment phase (Calomiris et al., 2003).
    - Improve access to funding for banks; can be guarantees on bank liabilities or on assets.
    - Guarantees programs for credit to SMEs noted for Korea, Singapore, Malaysia, Israel (included in asset guarantee measure but small relative to other measures and not directly benefiting listed large firms in sample).
- Liquidity injections measure:
  - Figure 1 shows total liquidity injected by country over September 2008 to March 2009, measured as the change in central bank claims on financial institutions normalized by GDP.
  - Noted large injections for Luxembourg, Ireland, Iceland, Belgium relative to banking system size.
- Cross-country variation in discretionary fiscal and monetary responses (Figure 2 referenced):
  - Fiscal impulses in sample range from "a 9 percent expansion in Nigeria" to "a 16 percent contraction in Latvia".
  - Most countries relaxed monetary policy significantly; some tightened in response to currency pressures.
- Sample and timing rationale:
  - Focus on interventions between September 2008 and March 2009, which comprise the majority of actions.
  - Collapse of Lehman Brothers in September 2008 marked escalation of policy intervention.
- Distinctive contributions of the sample and approach:
  - Quantifies both scope and intensity of support rather than binary indicators.
  - Includes crisis and "no crisis" countries (including preemptive policy implementations), mitigating selection bias from analyzing only crisis countries.
  - First to quantify and analyze real effects of a comprehensive set of key policy measures taken during the recent global financial crisis across a broad cross-section of countries.

### III. EMPIRICAL STRATEGY
- Identification challenge:
  - Standard cross-country regressions suffer endogeneity since policy responses are endogenous to macroeconomic environment.
  - Approach follows Rajan and Zingales (1998): exploit within-country differences across industries (here adapted to firm-level) using interaction of firms’ external finance dependence and country-level interventions.
- Use of firm-level data:
  - Construct value-added growth measures at the firm level (annual growth rate in real value added in percent) for firm i in country k during 2009.
  - Advantages: allows firm controls, reduces reverse causality concerns (unlikely country policy responses driven systematically by individual firm prospects).
- Baseline regression specification (as presented):
  - ,12,,ikjkjkikik
    gEDGXααββ   ε=+ +   · +   +
    (1)
  - Description in words:
    - Dependent variable ,ik g is annual growth rate in real value added (in percentages) of firm i in country k during the year 2009.
    - Primary variable of interest, kj GED · , is the interaction between industry j’s dependence on external financing and a vector of country-level government interventions.
    - In the baseline specification, k G consists of: fiscal policy in country k, monetary policy in country k, and bank recapitalization in country k, all measured over the period September 2008 and March 2009.
- Controls and robustness considerations:
  - Control for other government interventions that potentially offered relief to banks: guarantees, asset purchases, liquidity support.
  - Control for monetary and fiscal policy actions that help firms and banks via lower cost of funds and stimulating domestic demand.
  - Address remaining simultaneity concerns through country and industry level fixed effects and firm-level controls.
- Methodological linkage:
  - The strategy parallels Rajan and Zingales (1998) logic but leverages firm-level observations and crisis timing as exogenous shock to credit supply.

_italicized source attribution: IMF Working Paper chapter "2.  Monetary and Fiscal Policies" (excerpt provided)_

### 2009. In the case of bank recapitalization, it corresponds to committed amounts of public

### _wp1145 - 2009. In the case of bank recapitalization, it corresponds to committed amounts of public 

### Methodology and Identification
- Baseline regression includes fiscal and monetary policy because many countries enacted bank support packages while also using macroeconomic stabilization tools.
- Regression specification includes firm-level controls X_ki, industry indicators α_j, country indicators α_k, and error term ε_ki.
- External financial dependence computed as capital expenditures minus cash flow from operations divided by capital expenditures.
- Baseline uses external financial dependence of U.S. firms over the 1980’s as in Rajan and Zingales (1998); robustness tests compute the measure over 1980-2006.
- Sample limited to manufacturing firms; standard errors adjusted for clustering at the country- and industry level.
- U.S. firms dropped from all regressions to address endogeneity concerns linked to using U.S. data for the benchmark.
- Three ways endogeneity is mitigated:
  - external finance dependence measured prior to the crisis;
  - measured using U.S. data while studying firms outside the U.S., and U.S. firms are excluded from the sample;
  - analysis at the firm level with external finance dependence measured at the industry level.

### Data Sources and Variable Construction
- Firm-level data: Compustat North America (US and Canada) and Global Compustat (outside the U.S.).
- External financial dependence: (capital expenditure − operating cash flow) / capital expenditure, computed over the 1980’s and 1980-2006; industry-level medians using ISIC classifications (see Table 2).
- Value added: EBITDA + labor expense; sample restricted to firms with fiscal year end between October and December.
- Real growth in value added: convert to local currency and deflate annual growth rate using CPI inflation rate in the firm’s country.
- Firm-level controls: log of total assets (size); firm leverage (lev) = total liabilities / total assets; asset tangibility (ppe) = fixed assets / total assets; Tobin’s Q = (market value of equity + book value of debt) / book value of total assets.
- Government policies computed over September 2008 to March 2009 and normalized by 2008 GDP unless indicated otherwise.
- Fiscal policy primary proxy: change in structural balance (Fiscal1) from OECD (OECD, 2009) and IMF Fiscal Monitor (cut-off 2009 Q1); measures size of announced discretionary fiscal stimuli (net effect on fiscal balances over 2008-10), excluding measures announced before the crisis and fiscal outlays associated with financial sector support packages.
- Alternative fiscal proxy (Fiscal2): fiscal impulse following Fedelino et al. (2009) using April 2010 WEO; sum of variations in cyclically adjusted primary balance (relative to potential GDP) over 2008-2010; revenue elasticity = 1; expenditure elasticity from OECD (Girouard and Andre, 2005) or set to zero when missing; Fiscal2 is actual rather than intended stimulus and not purged from financial sector support packages.
- Monetary policy proxy: change in nominal short-term interest rates over September 2008 to March 2009 (money market rates or treasury bills from IMF International Financial Statistics lines 60B..ZF... and 60C..ZF...); alternative monetary measure: change in the money base over same period from IMF IFS.
- Bank intervention data: primary source Laeven and Valencia (2010) for crisis countries; complemented for other countries using IMF Staff reports, Central Banks’, and Ministry of Finance announcements.
- Guarantees and other bank intervention measures constructed from IMF Staff reports, Central Banks’, and Ministry of Finance announcements.
- Emergency liquidity support proxied by change in gross claims of monetary authorities to financial institutions normalized by 2008 nominal GDP; negative results set to zero; gross claims = sum of IFS lines 12E..ZF..., 12F..ZF..., and 12G..ZF...
- Bank intervention coding:
  - Bank recapitalization: committed amounts of public recapitalization funds (direct capital injections in preferred/common shares and subordinated debt).
  - Asset purchases and swaps: removal of impaired or illiquid assets from banks’ balance sheets; include direct lending by Treasury/Ministry of Finance (rare); excludes central bank purchases of mortgage-backed securities and long-term government bonds (captured via monetary base change).
  - Committed amounts announced between September 2008 and March 2009; used amounts up to end-2009.
- Final sample: 50 countries chosen based on availability of firm-level data; 10 countries experienced a systemic banking crisis and 9 are borderline crisis per Laeven and Valencia (2010) (see Table 1).
- Public debt: general government gross debt as of end-2006 (% of GDP); where unavailable, central government gross debt used; from IMF WEO.
- Country controls dated 2006: openness = (exports + imports) / GDP; financial development = outstanding credit to private sector (% of GDP); stock market capitalization (% of GDP); from World Bank World Development Indicators.
- Sample exclusions/losses: lose all firms from the Republic of Korea and Japan due to missing labor expense data and fiscal year ending in March; inclusion of these firms does not alter main findings.
- Table 3: average growth in real value added for firms over sample period = -4 percent; firms at 25th percentile growth = -25 percent; firms at 75th percentile growth = 16 percent.

### Empirical Findings — Main Results
- Baseline regression results include main proxies for fiscal policy, monetary policy, and bank recapitalization; additional bank support measures (guarantees, asset purchases) added one at a time.
- Standard errors in reported regressions corrected for clustering at country and industry level (columns 4 onward).
- Key empirical findings:
  - Recapitalizing banks has a positive and statistical significant effect on value added growth of firms in industries more dependent on external finance.
  - Among financial sector interventions, only bank recapitalization is statistically significant; guarantees and asset purchases are not individually significant.
  - All bank intervention variables together are jointly significant at the 10 percent level.
  - Interaction between external financial dependence and fiscal policy is positive and statistically significant at the 1 percent level: discretionary fiscal policy gives a greater boost to growth of firms in industries more dependent on external finance.
  - Monetary policy variable is insignificant — no differential effect on returns of financially dependent firms.
  - Larger firms (by total assets) experienced higher growth in value added over the sample period.

- Interpretations and comparisons:
  - Results reject the hypothesis of a frictionless supply of credit; evidence consistent with real effects of disruptions in credit supply alleviated by bank recapitalization.
  - Findings align with prior work showing real effects of negative shocks to bank capital (Klein et al. (2002); Peek and Rosengren (1997, 2000)).
  - Fiscal policy results align with Aghion et al. (2009): countercyclical fiscal policy helps more financially dependent firms.
  - The empirical approach does not disentangle demand-side vs supply-side channels of fiscal policy (increased firm profitability reducing banks’ expected losses vs fiscal policy directly stimulating demand).
  - Insignificance of monetary policy may reflect the increasing importance of wholesale funding markets and anticipation of monetary easing prior to the sample period (e.g., Federal Funds rate target declined 5.3 percent in 2007 to 1.8 percent by September 2008).

### Quantitative Importance
- Bank recapitalization economic magnitude:
  - Column 4 estimates imply a one standard deviation increase in the interaction between financial dependence and bank recapitalization increases real growth in value added by 2.3 percentage points (ceteris paribus).
  - A firm from an industry at the 75th percentile of financial dependence would have real growth in value added 1.1 percentage points higher than a firm at the 25th percentile of financial dependence if located in a country at the 75th percentile vs the 25th percentile of bank recapitalizations (ceteris paribus).
  - Both differentials are substantial compared to the average growth rate in value added of -4.0 percent.
- Fiscal policy economic magnitude:
  - Column 4 estimates imply a one standard deviation increase in the interaction between financial dependence and fiscal expenditure increases real growth in value added by 2.5 percentage points (ceteris paribus).
  - A firm from an industry at the 75th percentile of financial dependence would have real growth in value added 1.4 percentage points higher than a firm at the 25th percentile of financial dependence if located in a country at the 75th percentile vs the 25th percentile of fiscal expenditures (ceteris paribus).
  - Both differentials are substantial compared to the average growth rate in value added of -4.0 percent.
- Comparative magnitude:
  - Bank recapitalization and fiscal policy enter with similar coefficients of about 1.3 (both measures expressed in terms of GDP), indicating that a 1 percentage point of GDP increase in either fiscal stimulus or bank recapitalizations would enhance the growth performance of financially dependent firms by comparable amounts.

*Source: _wp1145 - 2009. In the case of bank recapitalization, it corresponds to committed amounts of public*

### 1.3 percentage points. This of course is not the total impact of each corresponding measure

### 1.3 percentage points. This of course is not the total impact of each corresponding measure

### Main findings
- Bank recapitalization policies boosted the value added growth of firms that are more financially dependent.
- Discretionary fiscal policy also helped financially dependent firms grow more, with an equivalence in the magnitude of short-run effects of financial sector and fiscal policies noted in the analysis.
- Monetary policy and other bank intervention measures are statistically insignificant once fiscal policy and bank recapitalization are controlled for.
- The correlation between the main fiscal policy variable (Fiscal1) and the main bank recapitalization variable (Recapa) is -0.14.
- The correlation between measures of financial dependence computed over different periods is 0.73.
- The correlation between financial dependence and Tobin’s Q is 0.18.

### Robustness checks and extensions (VI)
- A. Unbalanced number of firms
  - Weighted least squares (observations weighted by inverse of number of firm observations in each country-industry pair) leaves baseline qualitative results unchanged.
  - Differential effect of fiscal policy on value added growth of financially dependent firms: 1.72 compared to 1.36 in baseline.
- B. Additional firm controls
  - Include firm leverage (Lev = liabilities / total assets) and tangibility (ppe = fixed assets / total assets).
  - High leverage negatively affects growth performance (not statistically significant).
  - High tangibility negatively and significantly affects growth.
  - Main findings for fiscal policy and bank recapitalizations remain unaltered.
- C. Alternative definitions of intervention variables
  - Fiscal2 (fiscal impulse estimated from outcomes, includes outlays targeting financial sector) enters with a statistically significant coefficient while bank recapitalization does not when Fiscal2 is used.
  - Using expansion of monetary base normalized by GDP yields statistically insignificant effect for monetary policy on firm growth.
  - Using actual amounts used for bank recapitalization and asset purchases (rather than announced amounts) produces similar results; bank recapitalization remains statistically significant with a somewhat more pronounced economic effect.
  - Emergency liquidity support to banks shows no disproportionate effect on growth performance of financially dependent firms between September 2008 and March 2009.
    - Rationale: liquidity support was widely anticipated; many liquidity facilities had been expanded before September 2008 (examples cited include TAF, emergency liquidity to broker-dealers, U.S. dollar swap lines).
    - Liquidity injections do not repair capital shortfalls when banks face solvency rather than liquidity problems.
- D. Systemic banking crises
  - Interacting financial dependence with a 2009 systemic banking crisis dummy (Laeven and Valencia (2010) definition) shows financially dependent firms are particularly hurt during systemic crises, though the effect is not statistically significant.
  - Impact of fiscal policy and bank recapitalization policies is unaltered by inclusion of systemic crisis indicator.
- E. Financial development
  - Financial development measured as private credit + stock market capitalization to GDP in 2006 has an independent growth-enhancing effect for financially dependent firms.
  - This effect is entirely driven by capital market development (stock market capitalization to GDP), not banking sector development (private credit to GDP).
  - Interpretation: capital markets can act as a “spare wheel” during banking distress.
- F. Openness and integration
  - Interaction between trade openness (trade/GDP) and financial dependence does not alter main findings on fiscal policy and bank recapitalization.
- G. Fiscal space and size of government
  - Using total public debt as of 2006 (percentage points of GDP) does not change main findings.
- H. Additional industry characteristics
  - Using Rajan and Zingales (1998) measure of dependence on equity financing yields qualitatively similar results to main measure of financial dependence.
  - Financial dependence computed over 1980-2006 (Compustat) also yields qualitatively similar results.
- I. Short-term financing needs
  - Industry liquidity needs (liq = median inventories/sales, 2000–2006) and cash conversion cycle (ccc = inventories/sales + receivables/sales − payables/cost of goods sold, 2000–2006) used to capture working capital dependence.
  - Bank recapitalization measures disproportionately enhance growth of firms dependent on working capital financing.
  - Fiscal policy does not disproportionately benefit working-capital-dependent firms.
  - Theoretical rationale: fiscal policy primarily expands market size for long-term projects and thus affects long-term investment rather than short-term liquidity needs.
- J. Growth opportunities
  - Interactions between intervention policies and firm Tobin’s Q (computed at year-end 2006 as market value of equity + book value of liabilities divided by book value of total assets) do not enter significantly and do not alter findings.
  - This supports the channel operating through financial dependence rather than through changes in growth opportunities.

### Interpretation and mechanisms
- Supply-side financial frictions matter: the paper’s evidence emphasizes the importance of supply-side frictions in the credit market in affecting real activity.
- Fiscal policy effect: by boosting domestic demand, fiscal stimulus disproportionately benefits firms that are financially dependent because they are hurt most by disruptions in credit supply.
- Bank recapitalization effect: by mitigating negative credit supply shocks, recapitalizations disproportionately benefit financially dependent firms facing financial constraints.
- Monetary policy and anticipated liquidity measures were less effective:
  - Likely because liquidity measures were largely anticipated and less targeted to solvency problems.
  - Central bank liquidity facilities had expanded prior to the policy measurement window, reducing marginal real effects.

### Key statistics and values preserved verbatim
- Correlation Fiscal1 and Recapa: -0.14
- Differential fiscal effect on value added growth (weighted WLS): 1.72 compared to 1.36
- Correlation financial dependence (1980–1989 vs 1980–2006): 0.73
- Correlation financial dependence and Tobin’s Q: 0.18
- Policy measurement period for interventions covered: September 2008 to March 2009
- Financial development measure year: 2006

### Conclusions (VII)
- Statistically significant, economically important, and robust evidence that bank recapitalization policies and discretionary fiscal policy boosted growth for financially dependent firms in the sample of 50 countries.
- Results are not driven by correlation between fiscal policy and bank recapitalization.
- Findings likely underestimate effects because the sample focuses on large and listed firms that have better non-bank financing alternatives than smaller and bank-dependent firms.
- Overall, results provide new insights on the relative effectiveness of government interventions during financial crises and on the role of credit market frictions in explaining growth differences.

*Source: _wp1145 - 1.3 percentage points. This of course is not the total impact of each corresponding measure*

### References

### _wp1145 - References

### References cited
- Aghion, Philippe, David Hemous, and Enisse Kharroubi, 2009, “Credit constraints, Cyclical Fiscal Policy and Industry Growth,” NBER working paper 15119.
- Ashcraft, Adam, 2005, “Are Banks Really Special? New Evidence from the FDIC-Induced Failure of Healthy Banks,” American Economic Review Vol. 95, pp. 1712-730.
- Bernanke, Ben, and Cara Lown, 1991, “The Credit Crunch,” Brookings Papers on Economic Activity, 1991(2), pp. 205–39.
- Blanchard, Olivier, and Roberto Perotti, 2002, “An Empirical Characterization of the Dynamic Effects of Changes in Government Spending and Taxes on Output,” The Quarterly Journal of Economics, Vol. 117, No. 4, pp. 1329–368.
- Calomiris, Calomiris, Daniela Klingebiel, and Luc Laeven, 2003, “Financial Crisis Policies and Resolution Mechanisms: A Taxonomy from Cross-Country Experience,” in: P. Honohan and L. Laeven, eds., Systemic Financial Distress: Containment and Resolution, Chapter 2, Cambridge: Cambridge University Press.
- Claessens, Stijn, Daniela Klingebiel, and Luc Laeven, 2005, “Crisis Resolution, Policies, and Institutions: Empirical Evidence,” in Patrick Honohan and Luc Laeven (eds.), Systemic Financial Distress: Containment and Resolution, Chapter 6, pp. 169–96, Cambridge: Cambridge University Press.
- Dell'Ariccia, Giovanni, Enrica Detragiache, and Raghuram Rajan, 2008, “The Real Effects of Banking Crises,” Journal of Financial Intermediation, Vol. 17, pp. 89–112.
- Detragiache, Enrica, and Giang Ho, 2009, “Responding to Banking Crises: Lessons from Cross-Country Evidence,” IMF Working Paper 10/18. Washington: International Monetary Fund.
- Farhi, Emmanuel and Jean Tirole 2010, “Collective Moral Hazard, Maturity Mismatch and Systemic Bailouts,” American Economic Review, forthcoming.
- Fatás, Antonio, and Ilian Mihov, 2001, “Government Size and Automatic Stabilizers,” Journal of International Economics, Vol. 55, pp. 3–28.
- Fedelino, Annalisa, Mark Horton, and Anna Ivanova, 2009, “Computing Cyclically-Adjusted Balances and Automatic Stabilizers,” IMF Technical Notes and Manuals 09/05.
- Gertler, Mark, and Simon Gilchrist, 1994, “Monetary Policy, Business Cycles, and the Behavior of Small Manufacturing Firms,” Quarterly Journal of Economics, No. 59, pp. 309–40.
- Giannetti, Mariassunta, and Simonov, Andrei, 2009, “On the Real Effects of Bank Bailouts: Micro-Evidence from Japan," CEPR Discussion Papers 7441.
- Girouard, Nathalie, and Christophe André, 2005, “Measuring Cyclically-adjusted Budget Balances for OECD Countries,” OECD Economics Department Working Papers 434.
- Hancock, Diana, and James Wilcox, 1994, “Bank Capital and the Credit Crunch: The Roles of Risk-Weighted and Unweighted Capital Regulations,” Journal of the American Real Estate and Urban Economics Association, Vol. 22, No. I, pp. 59–94.
- International Monetary Fund, 2009, “The State of Public Finances: Outlook and Medium-Term Policies After the 2008 Crisis,” IMF fiscal Monitor Companion Paper.
- Kashyap, Anil, Jeremy Stein, and David Wilcox, 1993, “Monetary Policy and Credit Conditions: Evidence from the Composition of External Finance,” American Economic Review, Vol. 83, No. 1, pp. 78–98.
- Kashyap, Anil, Owen Lamont, and Jeremy Stein, 1994, “Credit Conditions and the Cyclical Behavior of Inventories,” Quarterly Journal of Economics, Vol. 109, pp. 565–92.
- Kashyap, Anil and Jeremy Stein. 2000, “What Do A Million Observations on Banks Say About the Transmission of Monetary Policy?” American Economic Review, Vol. 90, No. 3, pp. 407–28.
- Klein, Michael, Joe Peek, and Eric Rosengren, 2002, “Troubled Banks, Impaired Foreign Direct Investment: The Role of Relative Access to Credit,” American Economic Review, Vol. 92, pp. 664–82.
- Klingebiel, Daniela, Randall Kroszner, and Luc Laeven, 2001, “Stock Market Responses to Bank Restructuring Policies during the East Asian Crisis,” World Bank Policy Research Working Paper No. 2571.
- Kroszner, Randall, Luc Laeven, and Daniela Klingebiel, 2007, “Banking Crises, Financial Dependence, and Growth, Journal of Financial Economics, Vol. 84, pp. 187–228.
- Laeven, Luc and Fabián Valencia, 2008, “Systemic Banking Crises: a New Database,” IMF Working Paper No. 08/224. Washington: International Monetary Fund.
- Laeven, Luc and Fabián Valencia, 2010, “Resolution of Banking Crises: The Good, the Bad, and the Ugly,” IMF Working Paper No. 10/146. Washington: International Monetary Fund.
- Levine, Ross, 2005, “Finance and Growth: Theory and Evidence,” in P. Aghion and S. Durlauf, eds., Handbook of Economic Growth, Elsevier Science, The Netherlands.
- Modigliani, Franco, and Miller, Merton, 1958, “The Cost of Capital, Corporate Finance and the Theory of Investment,” American Economic Review, Vol. 48, pp. 261–97.
- Mountford, Andrew, and Harald Uhlig, 2008, "What are the Effects of Fiscal Policy Shocks?" NBER Working Paper No. 14551.
- OECD, 2009, “Fiscal Packages Across OECD Countries: Overview and Country Details,” OECD Report.
- Peek, Joe, and Eric Rosengren, 1995, “The Capital Crunch: Neither a Borrower nor a Lender Be,” Journal of Money, Credit and Banking, Vol. 27, pp. 625–38.
- Peek, Joe, and Eric Rosengren, 1997, “The International Transmission of Financial Shocks: The Case of Japan,” American Economic Review, Vol. 87, pp. 495–505.
- Peek, Joe, and Eric Rosengren, 2000, “Collateral Damage: Effects of the Japanese Banking Crisis on Real Activity in the United States,” American Economic Review, Vol. 90, pp. 30–45.
- Peek, Joe, Eric Rosengren, and Geoffrey Tootell, 2003, “Identifying the Macroeconomic Effect of Loan Supply Shocks,” Journal of Money, Credit and Banking, Vol. 35, No. 6, pp. 931–46.
- Perotti, Roberto, 2005, "Estimating the Effects of Fiscal Policy in OECD Countries," CEPR Discussion paper No. 4842.
- Philippon, Thomas and Philipp Schnabl, 2010, “Efficient Bank Recapitalization” New York University working paper.
- Raddatz, Claudio, 2006, “Liquidity Needs and Vulnerability to Financial Underdevelopment,” Journal of Financial Economics, Vol. 80, pp. 677–722.
- Rajan, Raghuram, and Luigi Zingales, 1998, “Financial Dependence and Growth,” American Economic Review, Vol. 88, pp. 393–410.
- Reinhart, Carmen, and Kenneth Rogoff, 2009, “This Time Is Different: Eight Centuries of Financial Folly,” Princeton University Press.
- Romer, Christina and David Romer, 1990, “New Evidence on the Monetary Transmission Mechanism,” Brookings Papers on Economic Activity, Vol. 1, pp. 149–214.
- Valencia, Fabián, 2008, “Banks’ Precautionary Capital and Credit Crunches,” IMF Working Paper No. 08/248, Washington: International Monetary Fund.

### Table 1 — Government interventions during the Global Financial Crisis (selected country figures)
- Column definitions (as given in the source):
  - Recapitalizations: Pledged amount (Percent of GDP); Used amount (Percent of GDP).
  - Guarantees: Asset guarantees (Percent of GDP); Bank Creditor's guarantees (Percent of GDP).
  - Asset Purchases and Lending by Treasury: Pledged amount (Percent of GDP); Used amount (Percent of GDP).
  - Liquidity Support (Percent of GDP).
  - Change in ST interest rate (percent, 08/08-03/09).
  - Change in Money base or Reserve Money (% of GDP).
  - Public Debt (Percent of 2006 GDP).
  - Fiscal stimulus (fiscal1) Percent of GDP.
  - Fiscal stimulus (fiscal2) Percent of GDP.
- Selected country entries (values reproduced exactly):
  - Argentina: 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, -2.2, 0.6, 76.5, 1.5, 2.8
  - Australia: 0.0, 0.0, 0.0, 97.0, 0.0, 0.0, 1.4, 3.6, -0.2, 10.1, 4.6, 5.2
  - Iceland: 24.0, 13.0, 0.0, 108.4, 0.0, 0.0, 24.7, -0.4, -2.3, 30.1, -9.4, 5.6
  - Ireland: 7.6, 7.6, 0.0, 200.0, 0.0, 0.0, 42.2, 2.4, 0.1, 24.8, -4.4, 5.8
  - Luxembourg: 7.7, 7.7, 0.0, 12.0, 0.0, 0.0, 57.5, 2.4, 0.4, 6.6, 3.6, 4.4
  - Netherlands: 6.5, 6.5, 10.8, 34.8, 6.1, 6.1, 9.3, 2.4, 0.0, 47.4, 1.5, 4.0
  - UK: 8.2, 5.0, 14.5, 40.0, 0.0, 0.0, 7.9, 4.7, 0.0, 43.2, 1.4, 5.2
  - USA: 5.1, 3.1, 0.0, 10.4, 2.3, 1.6, 3.5, 1.8, 0.0, 60.9, 5.6, 7.2

### Table 2 — External financial dependence across U.S. industries (selected rows)
- Column definitions: ISIC code; Industrial sector; exf80; exf8006.
- Selected ISIC entries (values reproduced exactly):
  - 314 Tobacco: -0.45; -2.80
  - 361 Pottery: -0.15; -0.22
  - 323 Leather: -0.14; -1.30
  - 3211 Spinning: -0.09; -0.11
  - 324 Footwear: -0.08; -0.78
  - 372 Nonferrous metal: 0.01; -0.01
  - 322 Apparel: 0.03; -0.08
  - 353 Petroleum refineries: 0.04; -0.11
  - 369 Nonmetal products: 0.06; -0.01
  - 3522 Drugs: 1.49; 2.10
  - 3825 Office and computing: 1.06; 1.05
  - 3832 Radio: 1.04; 1.14
  - 385 Professional goods: 0.96; 1.13

### Table 3 — Summary statistics (selected variables; values reproduced exactly)
- Firm-level variables (Obs, Mean, Std. Dev., Min, p25, p75, Max):
  - Growth (Growth in real value added): 2902; -4.02; 34.98; -99.99; -24.92; 15.61; 98.73
  - Size (Firms’ total assets): 2902; 5.01; 2.63; -9.13; 3.90; 6.45; 12.62
  - Lev (Firms’ leverage (liabilities/assets)): 2902; 0.49; 0.25; 0.02; 0.32; 0.62; 4.28
  - Ppe (Firms’ asset tangibility (fixed assets/total assets)): 2902; 0.31; 0.19; 0.00; 0.15; 0.44; 0.99
- Industry-country level variables (Obs, Mean, Std. Dev., Min, p25, p75, Max):
  - exf * fiscal1: 771; 0.50; 1.83; -24.31; 0.02; 0.77; 9.55
  - exf * monpol: 771; 0.74; 1.36; -14.02; 0.15; 0.99; 10.05
  - exf * bg: 771; 17.35; 44.00; -52.15; 0.00; 15.42; 440.29
  - exf * tcap: 754; 81.28; 115.31; -149.74; 14.07; 103.44; 910.94
  - exf * trade: 768; 41.46; 71.64; -95.30; 6.88; 49.40; 681.10
  - exf * pdebt: 771; 17.81; 24.30; -40.04; 3.04; 22.14; 158.85
- Country-level variables (Obs, Mean, Std. Dev., Min, p25, p75, Max):
  - Fiscal1 (Discretionary fiscal policy, announced (percent of GDP)): 50; 1.02; 3.90; -16.30; 0.25; 3.00; 9.00
  - Fiscal2 (Discretionary fiscal policy, used (percent of GDP)): 50; 1.96; 3.19; -7.20; 0.70; 3.20; 13.90
  - Monpol (Change in short term rates, percent): 50; 1.70; 2.55; -9.40; 1.11; 2.42; 6.74
  - Monbase (Change in money base, percent of GDP): 50; 0.07; 1.53; -4.99; 0.00; 0.14; 5.83
  - Bg (Guarantees of bank liabilities (percent of GDP)): 50; 36.63; 64.58; 0.00; 0.00; 46.80; 295.20
  - Recapa (Bank recapitalization, announced (percent of GDP)): 50; 2.03; 4.00; 0.00; 0.00; 2.20; 24.00
  - Liq Support (Liquidity support (percent of GDP)): 50; 5.97; 10.82; 0.00; 0.29; 7.87; 57.54
  - Public Debt (Public Sector debt as of 2006 (Percent of GDP)): 50; 48.02; 26.54; 1.67; 29.53; 65.92; 106.50

### Table 4 — Government interventions, financial dependence, and real growth in value added (key regression coefficients)
- Dependent variable: firm's real growth in value added during 2009 (percent).
- Model results (coefficients with cluster-robust standard errors in parentheses; significance stars as in source):
  - Size:
    - (1): 1.333*** (0.291)
    - (2): 1.331*** (0.292)
    - (3): 1.329*** (0.292)
    - (4): 1.329*** (0.322)
  - exf * fiscal1:
    - (1): 1.280** (0.516)
    - (2): 1.291** (0.513)
    - (3): 1.357*** (0.516)
    - (4): 1.357*** (0.523)
  - exf * monpol:
    - (1): -0.0737 (0.937)
    - (2): -0.0522 (0.942)
    - (3): 0.0660 (0.956)
    - (4): 0.0660 (1.015)
  - exf * recapa:
    - (1): 1.336** (0.603)
    - (2): 1.348** (0.601)
    - (3): 1.290** (0.604)
    - (4): 1.290** (0.555)
  - exf * bg:
    - (1): 0.00539 (0.0190)
    - (2): 0.00841 (0.0195)
    - (3): 0.00841 (0.0155)
  - exf * asseta:
    - (3): 0.509 (0.613)
    - (4): 0.509 (0.562)
- Model fit and sample:
  - Observations: 2,903 (for all columns)
  - R-squared: column (1) 0.140; (2) 0.140; (3) 0.141; (4) 0.141
  - Clustering: column (4) specifies Country-Industry clustering; earlier columns report "No" (as presented).

### Table 5 — Alternative specifications (selected coefficients)
- Dependent variable: firm's real growth in value added during 2009 (percent). Various additional controls and specifications across columns (1)-(6).
- Selected coefficient estimates (coefficient (standard error)):
  - Size:
    - (1): 1.014* (0.534)
    - (2): 1.342*** (0.323)
    - (3): 1.338*** (0.322)
    - (4): 1.335*** (0.323)
    - (5): 1.331*** (0.322)
    - (6): 1.322*** (0.323)
  - Lev (column 2): -2.578 (2.911)
  - Ppe (column 2): -9.351** (4.748)
  - exf * fiscal1:
    - (1): 1.710** (0.811)
    - (2): 1.352** (0.528)
    - (3): 1.315** (0.535)
    - (4): 1.456*** (0.537)
    - (5): 1.322** (0.524)
  - exf * monpol:
    - (1): -1.442 (1.021)
    - (2): -0.0311 (1.006)
    - (3): 0.0334 (0.966)
    - (4): -0.0802 (1.074)
    - (5): 0.0878 (1.021)
  - exf * recapa:
    - (1): 1.199* (0.689)
    - (2): 1.279** (0.555)
    - (3): 0.896 (0.559)
    - (4): 1.320** (0.540)
    - (6): 1.670** (0.683)
  - exf * bg (across columns): values range from 0.00656 to 0.0140 with standard errors reported (e.g., (0.0180), (0.0157), (0.0162), (0.0153), (0.0155), (0.0156))
  - exf * recapu (column 4/5/6 variants): 1.978** (0.817) reported for one alternative (as presented)
- Observations and fit:
  - Observations: 2,902 for all columns
  - R-squared: column (1) 0.111; columns (2)-(6) 0.141

### Table 6 — Other country characteristics (context for regressions; variables defined)
- Variables and definitions (as presented):
  - Size: natural logarithm of the firm’s total assets expressed in millions of U.S. dollars.
  - Exf: external financial dependence during 1980’s.
  - Fiscal policy (fiscal1): change in the fiscal structural balance as announced for 2008-2010 normalized by GDP in 2008.
  - Monetary policy (monpol): absolute decrease in nominal interest rates between Sep 2008 and Mar 2009.
  - Recapitalization (recapa): bank recapitalizations announced.
  - Guarantees (bg): guarantees on bank assets plus guarantees on bank liabilities.
  - Asset purchases (asseta): asset purchases announced.
  - Crisis: dummy for systemic or borderline systemic banking crisis.
  - Tcap: total financial sector capitalization computed as sum of private credit to GDP and stock market capitalization to GDP in 2006.
  - Privo: private credit to GDP (definition appears in source continuation).
- (Table is introductory to additional regressions; specific coefficient values and further rows appear in the subsequent pages of the source.)

*Italic source attribution: Extracted from the PDF filename "_wp1145 - References".*

### 2006. Mcap is stock market capitalization to GDP in 2006. Pdebt is government debt to GDP in 2006. All

### _wp1145 - 2006. Mcap is stock market capitalization to GDP in 2006. Pdebt is government debt to GDP in 2006. All

### Major regression results (Table with VARIANTS (1)–(5))
- Dependent variable: the firm’s real growth in value added during 2009 (in percentages).
- Size (natural logarithm of firm’s total assets expressed in millions of U.S. dollars):
  - (1) Size = 1.329*** (0.322)
  - (2) Size = 1.339*** (0.323)
  - (3) Size = 1.331*** (0.323)
  - (4) Size = 1.322*** (0.322)
  - (5) Size = 1.329*** (0.322)
- exf * fiscal1 (external financial dependence × fiscal policy):
  - (1) 1.356** (0.527)
  - (2) 1.299** (0.526)
  - (3) 1.403*** (0.530)
  - (4) 1.277*** (0.478)
  - (5) 1.365*** (0.517)
- exf * monpol (external financial dependence × monetary policy):
  - (1) 0.0669 (1.017)
  - (2) -0.190 (1.014)
  - (3) -0.0965 (0.984)
  - (4) 0.234 (1.077)
  - (5) 0.0750 (1.013)
- exf * recapa (external financial dependence × recapitalization):
  - (1) 1.296* (0.664)
  - (2) 1.197** (0.565)
  - (3) 1.488** (0.585)
  - (4) 1.394** (0.574)
  - (5) 1.290** (0.554)
- exf * bg (external financial dependence × guarantees):
  - (1) 0.00836 (0.0162)
  - (2) -0.0329 (0.0256)
  - (3) -0.0315 (0.0251)
  - (4) -0.00786 (0.0271)
  - (5) 0.00742 (0.0201)
- exf * asseta (external financial dependence × asset purchases):
  - (1) 0.510 (0.571)
  - (2) 0.108 (0.607)
  - (3) 0.191 (0.585)
  - (4) 0.444 (0.562)
  - (5) 0.513 (0.563)
- exf * cri:
  - (2) -0.0519 (3.734)
- exf * tcap:
  - (3) 0.0294** (0.0149)
- exf * mcap:
  - (4) 0.0350* (0.0179)
- exf * trade:
  - (5) 0.0143 (0.0223)
- exf * pdebt:
  - (5) -0.00593 (0.0700)
- Constant terms:
  - (1) -3.673 (2.510)
  - (2) -4.317* (2.535)
  - (3) -4.082 (2.509)
  - (4) -3.944 (2.487)
  - (5) -3.633 (2.575)
- Sample size and fit:
  - Observations: (1) 2,902; (2) 2,902; (3) 2,902; (4) 2,899; (5) 2,902
  - R-squared: (1) 0.141; (2) 0.141; (3) 0.141; (4) 0.140; (5) 0.140

### Additional industry-characteristic regressions (Table 7)
- Size (repeat) in alternative specifications:
  - (1) 1.334*** (0.321)
  - (2) 1.353*** (0.323)
  - (3) 1.315*** (0.320)
  - (4) 1.319*** (0.320)
  - (5) 1.251*** (0.333)
- eqf80 * fiscal1 (industry dependence on equity financing during 1980’s × fiscal policy):
  - (1) 2.886*** (0.881)
- eqf80 * recapa:
  - (1) 2.023** (0.938)
- exf8006 * fiscal1 (external financial dependence during 1980-2006 × fiscal policy):
  - (3) 0.991** (0.457)
- exf8006 * recapa:
  - (3) 0.924** (0.470)
- liqn00 * fiscal1 (liquidity needs 2000-06 × fiscal policy):
  - (4) -10.45 (7.153)
- liqn00 * recapa:
  - (4) 13.79** (6.254)
- ccc00 * recapa:
  - (5) 7.598** (3.065)
- exf * fiscal1 (again in these specs):
  - (5) 1.435** (0.592)
- exf * recapa:
  - (5) 1.093** (0.517)
- tobinq * recapa:
  - (5) 0.477** (0.214)
- tobinq * bg:
  - (5) 0.0140* (0.00723)
- tobinq * asseta:
  - (5) -0.287 (0.183)
- Constants (Table 7):
  - (1) -2.932 (2.466)
  - (2) -2.774 (2.523)
  - (3) -5.612 (4.576)
  - (4) -8.042** (3.678)
  - (5) -6.420** (3.028)
- Sample size and fit (Table 7):
  - Observations: (1) 2,902; (2) 2,902; (3) 2,902; (4) 2,902; (5) 2,478
  - R-squared: (1) 0.141; (2) 0.140; (3) 0.141; (4) 0.140; (5) 0.156

### Variable definitions, data scope, and estimation details
- Dependent variable: firm’s real growth in value added during 2009 (in percentages).
- Size: natural logarithm of firm’s total assets expressed in millions of U.S. dollars.
- Exf: external financial dependence during 1980’s.
- Eqf80: industry-level dependence on equity financing during 1980’s.
- Exf8006: external financial dependence during 1980-2006.
- Eqf8006: dependence on equity financing during 1980-2006.
- Liqn: liquidity needs, computed as median ratio of inventories to sales during 2000-06.
- Ccc: working capital needs, computed using cash conversion cycle method as ratio of inventories to sales plus ratio of receivables to sales minus accounts payables to cost of goods sold, during 2000-06.
- Sales growth: industry sales growth during 2000-06.
- Fiscal policy (fiscal1): change in the fiscal structural balance as announced for the period 2008-2010 normalized by GDP in 2008.
- Monetary policy (monpol): absolute decrease in nominal interest rates between Sep 2008 and Mar 2009.
- Recapitalization (recapa): announced bank recapitalizations.
- Guarantees (bg): guarantees on bank assets plus guarantees on bank liabilities.
- Asset purchases (asseta): asset purchases.
- Estimation: All regressions include country and industry fixed effects (not included). Standard errors are corrected for clustering at the country-industry pair level. ***, **, and * denote significance at the 1%, 5%, and 10% level, respectively.

*Source: WEO, Official Websites, IMF IFS, and authors’ calculations.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp1145.pdf_