## _wp0935 — References and Annexed Results

## Source details

**Canonical URL:** [_wp0935 — References and Annexed Results](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp0935.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp0935.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp0935.pdf.json)

---

### Introduction and motivation
- Ongoing financial turmoil from U.S. subprime mortgages shows asset-price dynamics amplify business cycle fluctuations by making bank lending procyclical.
- Mechanisms:
  - Rapid asset-price growth raises collateral value → stimulates credit growth.
  - Speculation and wealth effects accentuate spending booms.
  - Greater asset liquidity during euphoria reinforces lending; downturns reverse these forces.
- Micro-focused, risk-sensitive capital requirements can increase procyclicality because individual intermediaries do not internalize aggregate risk.
- Countercyclical provisioning (dynamic/statistical) advocated (example: Spain) to provision more in good times and draw reserves in bad times.
- Debate: provisioning/accounting (IAS 39) versus incomplete implementation of risk-sensitive capital regulation (Basel II).

### Empirical evidence on procyclicality of firms’ leverage
- Data: firms’ leverage measured as ratio between private credit to the corporate sector and value of equities for 16 advanced and 12 emerging European economies, January 1999–April 2008.
- Key empirical findings:
  - Firms’ leverage varies substantially over time across selected European countries.
  - Financial Conditions Indices (FCIs) constructed via country-specific VARs capture timing/transmission from financial variables to real activity.
  - Innovations to firms’ borrowing contributed to fluctuations in annual growth:
    - Austria: 40 percent
    - Sweden: 25 percent
    - United Kingdom and Greece: 10 percent
  - In emerging economies, financial conditions play a larger role:
    - Over 70 percent of growth variation in Hungary, Russia, Latvia, and Estonia attributable to financial conditions.
  - Leverage shocks are procyclical on average: increases in firms’ leverage provide greater contribution to growth during expansions.
  - Cross-country regularities:
    - Changes in firms’ borrowing more sensitive to asset-price changes where leverage co-moves closely with the business cycle.
    - Higher procyclicality in firms’ leverage associated with higher volatility in private investment.
    - Oversensitivity of credit to asset prices, greater procyclicality, and higher investment volatility characterize economies (e.g., Baltic countries) where bank lending conditions play a large role in growth fluctuations.

### Theoretical model (structure and amplification)
- Model: two-country, two-good, incomplete markets with collateral constraint à la Kiyotaki and Moore (1997).
- Key features:
  - Home: impatient, net borrower; Foreign: patient, net lender.
  - Borrowing limited by loan-to-value ratio LTV φt on equity collateral.
  - LTV φt stochastic to capture time-varying lending standards; φt assumed always binding in Home around steady state (χt>0).
  - Households: CRRA utility (σ), labor disutility parameter ψ, ν; invest in domestic equity and issue one-period non-contingent bond subject to constraint:
    - dt+1 ≤ φt pH,t qt kt+1
- Asset-pricing amplification:
  - Equity price:
    - qt = Et ∑_{j=0}^∞ (∏_{i=0}^j ˜β_{t+i}) r_{t+1+j}
  - Equity premium relative to real interest rate R_t:
    - Et(R^q_{t+1}) − R_t = −cov(β U_{c,t+1}/U_{c,t}, R^q_{t+1}) / Et[β U_{c,t+1}/U_{c,t}] + χ_t (R_t − φ_t) / Et[β U_{c,t+1}/U_{c,t}]
  - Binding collateral constraint (χt>0) creates positive wedge between actual and fundamental equity premium → higher discounting of dividends → lower equity price.
  - Procyclical φt (falls in downturns) increases wedge and further depresses equity prices → amplifies asset-price overreaction and macro volatility.
  - Higher average LTV (φ̄) narrows the wedge, reducing overreaction/volatility.

### Shocks, equilibrium, and solution
- Shocks:
  - Productivity shocks A_t (Home) and y*_t (Foreign) follow bivariate AR(1) with contemporaneous variance-covariance V.
  - Credit shocks:
    - ln(φ_t) = (1−ρφ) ln(φ̄) + ρφ ln(φ_{t−1}) + ε_{φt}, ε_{φt} ∼ iid N(0, σ^2_φ)
  - Correlation between LTV shocks and Home TFP shocks denoted ρ_{LTV,A}.
- Market clearing: Home and Foreign goods and world credit market conditions specified.
- Solution: with collateral constraint always binding in Home, model linearized and solved for policy functions.

### Calibration (benchmark and alternatives)
- Benchmark and key parameter values:
  - Preferences:
    - β = 0.96; β* = 0.99
    - σ = 1; ψ = 3; ν set to match h = 33 percent
    - η = 1.5; γ = γ* = 57.5 percent
    - ω = 0.00074; N = 1
  - Technology:
    - α = 0.30; δ = 2.5 percent per quarter; φ (adj. cost) = 10
  - Productivity shocks:
    - B11 = 0.94; σε1 = 2.2 percent
    - B22 = 0.97; σε2 = 1.14 percent
    - Corr(ε1, ε2) = 0.04
  - Loan-to-value ratio:
    - φ̄ = 40 percent; ρφ = 0.9; σφ = 1.85 percent
    - Benchmark correlation: ρ_{LTV,A} = 0.5

### Policy experiment: reducing procyclicality of lending standards
- Objective: quantify impact on cyclical properties from reducing procyclicality (introducing acyclical/countercyclical element in prudential regulation).
- Simulation design:
  - 500 replications, 150 quarters each; first 50 quarters dropped.
  - Time unit = one quarter; series filtered with Hodrick-Prescott filter; average moments computed.
  - Impulse responses to negative one-standard-deviation shocks to productivity and lending standards analyzed.
- Scenarios:
  - Baseline: Average LTV = 0.4 (φ̄ = 40 percent) with ρ_{LTV,A} = 0.5.
  - Policy: make LTV innovations independent of domestic TFP (ρ_{LTV,A} = 0).
  - Alternative leverage level: Average LTV = 0.7 examined.
- Intuition:
  - Procyclical φt amplifies equity-price overreaction to shocks and increases volatility of macro aggregates.
  - Reducing φt–cycle correlation smooths equity-price overreaction, lowering macroeconomic volatility.

### Key quantitative results — Benchmark (Average LTV = 0.4) and policy effects
- Selected simulated business-cycle moments under benchmark calibration (Table 3):
  - Volatilities:
    - σy 0.0285
    - σq 0.017
    - σx 0.0674
    - σR 0.022
    - σx/y 0.0396
    - σc 0.0256
    - σc/pH 0.0286
    - σ(c/pH)/y 0.0163
    - σca/y 0.0642
    - σd 0.033
  - Cross-correlations with Home GDP:
    - ρx,y 0.9843; ρq,y 0.9694; ρc,y 0.6995; ρd,y 0.7244
  - Cross-correlations with equity prices:
    - ρd,q 0.7721; ρca/y,q -0.3161; ρR,q 0.3361
- Policy exercise (Table 4) — change from ρ_{LTV,A} = 0.5 to ρ_{LTV,A} = 0.0 (Average LTV = 0.4):
  - Volatility differences (percent):
    - σy: 0.0285 → 0.0284 (−0.4%)
    - σx: 0.0674 → 0.065 (−3.6%)
    - σx/y: 0.0396 → 0.0372 (−6.1%)
    - σc: 0.0256 → 0.0245 (−4.3%)
    - σc/pH: 0.0286 → 0.0281 (−1.7%)
    - σ(c/pH)/y: 0.0163 → 0.0141 (−13.5%)
    - σq: 0.017 → 0.0164 (−3.5%)
    - σR: 0.022 → 0.0189 (−14.1%)
    - σca/y: 0.0642 → 0.0552 (−14.0%)
    - σd: 0.033 → 0.0284 (−13.9%)
  - Cross-correlation changes:
    - ρd,q: 0.7721 → 0.4822 (−37.5%)
    - ρR,q: 0.3361 → 0.2039 (−39.3%)
    - ρca/y,q: -0.3161 → -0.2124 (−32.8%)
    - ρd,y: 0.7244 → 0.4205 (−42.0%)
    - ρR,y: 0.1772 → 0.0664 (−62.5%)
- Policy exercise (Table 5) — Average LTV = 0.7, change from ρ_{LTV,A} = 0.5 to ρ_{LTV,A} = 0.0:
  - Volatility differences (percent):
    - σy: 0.0281 → 0.0280 (−0.4%)
    - σx: 0.0596 → 0.0570 (−4.4%)
    - σx/y: 0.034 → 0.0315 (−7.4%)
    - σc: 0.0307 → 0.0297 (−3.3%)
    - σc/pH: 0.0335 → 0.0330 (−1.5%)
    - σ(c/pH)/y: 0.0239 → 0.0213 (−10.9%)
    - σq: 0.015 → 0.0144 (−4.0%)
    - σR: 0.016 → 0.0143 (−10.6%)
    - σca/y: 0.0922 → 0.0816 (−11.5%)
    - σd: 0.0295 → 0.0261 (−11.5%)
  - Cross-correlation changes notable:
    - ρd,q: 0.7151 → 0.4055 (−43.3%)
    - ρR,q: -0.0069 → -0.1638 (2273.9%)
    - ρca/y,q: 0.0391 → 0.1674 (328.1%)

### Benchmark leverage level vs. alternative leverage (summary)
- Benchmark (ρ_{LTV,A} = 0.5):
  - Consumption volatility σc/pH slightly bigger than GDP volatility; foreign debt highly correlated with equity price: ρd,q = 0.77.
  - Current account negatively correlated with equity price: ρca/y,q = −0.32.
- Making LTV shocks independent of TFP (ρ_{LTV,A} = 0):
  - σq decreases by 3.5 percent.
  - σx decreases by 3.6 percent.
  - Debt and current account volatility reduced by as much as 14 percent.
  - σc/pH decreases by 1.7 percent.
  - σ((c/pH)/y) decreases by 13.5 percent.
- Higher average LTV (less constrained economy):
  - Lower volatility of equity prices, investment, real interest rate, and foreign debt.
  - Higher volatility of consumption in levels and as fraction of GDP.
  - Policy effects (reducing procyclicality) somewhat different:
    - Equity-price volatility: −4.0 percent.
    - Investment volatility: −4.4 percent.
    - Foreign debt volatility: −11.5 percent.
    - Consumption volatility: −1.5 percent.

### Sensitivity analysis (Table 6) — robustness across dimensions
- Dimensions tested: ψ (Frisch wage elasticity), productivity spillovers (B12, B21), consumption home bias (γ, γ*), risk aversion σ.
- Main outcomes:
  - Results robust; most relevant differences arise from preference changes.
  - Wage elasticity (ψ = 0.3) leaves qualitative gains broadly unchanged; GDP volatility falls more under the policy.
  - Productivity spillovers (B12 = B21 = 0.08) increase stabilization gains:
    - Investment: −4.3 percent (instead of −3.6 percent benchmark)
    - Consumption: −3.3 percent (instead of −1.7 percent)
    - GDP: −0.7 percent (instead of −0.4 percent)
  - Higher consumption home bias reduces gains for consumption volatility; increases real exchange rate volatility and associated gains.
  - Higher risk aversion (σ = 2):
    - Higher GDP stabilization gains; investment gains unchanged.
    - Consumption less volatile under σ = 2; policy may not reduce consumption volatility (consumption deflated by pH may be even more volatile), but consumption-to-GDP volatility gains are robust.

### VAR-based Financial Condition Indices (FCIs) — data and construction
- Data:
  - Country coverage: 16 advanced European economies and 12 emerging European economies.
  - Sample period: 1999:1–2008:4 (monthly data).
  - Country-specific VARs include: (real) GDP growth; (real) 3-month interbank rate; changes in CPI-deflated equity prices; changes in (real) private credit to corporate sector.
  - Lag order: three (Schwartz information criterion).
- FCI construction:
  - Dynamic weights computed using IRFs from each VAR to incorporate timing of transmission from financial variables to real activity.
  - Identification via Cholesky factorization; ordering: GDP growth first, changes in equity prices before credit growth, 3-month interbank rate after others.
  - FCI formula:
    - FCIt = Σ_{j=1}^n ( r_{t,j}^{t−1} + Σ_{i=2}^m r_{t,j}^{t−i} )
  - Interpretation: FCI measures total contribution to GDP growth in quarter t from shocks to financial variables over previous m quarters; dynamic weights arise from IRFs.

### Tables and empirical estimates (selected)
- Table 1: AR(1) estimates for demeaned LTVs (country / observations / ̂ρ / Std.err. / ̂σφ):
  - Croatia 46 0.92 0.052 0.055
  - Estonia 49 0.89 0.0458 0.076
  - Lithuania 34 0.91 0.083 0.046
  - Russia 37 0.65 0.157 0.062
  - Slovenia 52 0.84 0.0922 0.046
  - Turkey 53 0.8 0.110 0.084
- Table 2: Benchmark calibration (selected entries):
  - β Home discount factor 0.96; β* Foreign discount factor 0.99
  - σ Coefficient of risk aversion 1; ψ = 3
  - B11 = 0.94; σε1 = 0.022; B22 = 0.97; σε2 = 0.0114
  - φ̄ Average LTV ratio 0.4; ρφ = 0.9; σφ = 0.0185; ρLT V,A = 0.5
- Table 3: Business Cycle Moments (selected values reproduced above)
- Tables 4–6: Policy exercise and sensitivity results (selected values reproduced above)

### Conclusions and policy implications
- Empirical-consistent findings:
  - Firms’ borrowing more sensitive to asset-price changes where leverage co-moves closely with the business cycle.
  - Higher procyclicality in firms’ leverage associated with higher volatility in private investment.
  - Economies where bank lending conditions dominate growth fluctuations (e.g., Baltic countries) exhibit oversensitivity of credit to asset prices and greater macro volatility.
- Model implications:
  - Equities serve both as collateral in international lending and as capital value for production, creating amplification channels.
  - Less financially constrained markets (higher stochastic LTV) alter trade-offs between investment and consumption volatility.
  - Stabilizing lending standards (reducing procyclicality) yields sizeable reductions in volatility of equity prices, investment, consumption, and external imbalances.
  - Cross-country technology spillovers increase stabilization gains.
  - Results robust to variations in wage elasticity, home-bias in consumption, and risk aversion.

*Source: _wp0935 — IMF Working Paper (References, tables, figures and annexed results as presented in the source PDF)._*

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

### _wp0935 - References  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

### Introduction
- Ongoing financial turmoil originating in U.S. subprime mortgages shows asset price dynamics amplify business cycle fluctuations by making bank lending procyclical.
- Mechanisms described:
  - Rapid asset-price growth raises collateral value → stimulates credit growth.
  - Speculation and wealth effects accentuate spending booms.
  - Greater asset liquidity during euphoria reinforces lending.
  - Downturn: asset-price deflation, debtor distress, creditor liquidation → reduced asset liquidity and sharper downturn.
- Micro-focused, risk-sensitive capital requirements can increase procyclicality because individual intermediaries do not internalize aggregate risk.
- Countercyclical provisioning (dynamic/statistical) has been advocated (example: Spain) to provision more in good times and draw reserves in bad times to reduce procyclicality.
- Debate exists whether the source is provisioning/accounting (IAS 39) or incomplete implementation of risk-sensitive capital regulation (Basel II).

### Empirical evidence on procyclicality of firms’ leverage
- Data: firms’ leverage measured as ratio between private credit to the corporate sector and value of equities for 16 advanced and 12 emerging European economies, January 1999–April 2008.
- Key empirical findings:
  - Firms’ leverage varies substantially over time across selected European countries.
  - Financial Conditions Indices (FCIs) constructed via country-specific VARs capture timing and transmission from financial variables to real activity.
  - Innovations to firms’ borrowing contributed substantially to fluctuations in annual growth in selected countries:
    - Austria: 40 percent
    - Sweden: 25 percent
    - United Kingdom and Greece: 10 percent
  - In emerging economies, financial conditions play a larger role:
    - Over 70 percent of growth variation in Hungary, Russia, Latvia, and Estonia is attributable to financial conditions.
  - An increase in firms’ leverage provides a greater contribution to growth in periods of economic expansion (leverage shocks are procyclical on average).
  - Cross-country regularities:
    - Changes in firms’ borrowing are more sensitive to asset price changes in economies where leverage co-moves closely with the business cycle.
    - Higher degree of procyclicality in firms’ leverage is associated with higher volatility in private investment.
    - Oversensitivity of credit to asset prices, greater procyclicality, and higher investment volatility characterize economies (e.g., Baltic countries) where bank lending conditions play a large role in growth fluctuations.

### Theoretical model (two-country, two-good, incomplete markets with collateral constraint)
- Purpose: quantify gains in macroeconomic volatility from policies that mitigate procyclicality in lending standards.
- Key model features:
  - Two countries: Home (impatient, net borrower) and Foreign (patient, net lender).
  - Financial friction à la Kiyotaki and Moore (1997): borrowing limited to fraction of value of collateral (loan-to-value ratio, LTV φt); collateral is equity.
  - LTV φt is stochastic to capture time-varying lending standards and calibrated to match empirical cyclicality.
  - Households: infinitely lived, utility specified with CRRA consumption and labor disutility (σ, ψ, ν), invest in domestic equity, issue one-period non-contingent bond, face borrowing constraint:
    - dt+1 ≤ φt pH,t qt kt+1
  - Constraint assumed always binding in Home economy around steady state (χt>0); model solved by linearization around steady state with binding collateral constraint.
- Asset pricing and amplification mechanisms:
  - Equity price qt satisfies:
    - qt = Et ∑_{j=0}^∞ (∏_{i=0}^j ˜β_{t+i}) r_{t+1+j}
  - Equity premium relative to real interest rate R_t:
    - Et(R^q_{t+1}) − R_t = −cov(β U_{c,t+1}/U_{c,t}, R^q_{t+1}) / Et[β U_{c,t+1}/U_{c,t}] + χ_t (R_t − φ_t) / Et[β U_{c,t+1}/U_{c,t}]
  - Binding collateral constraint (χt>0) creates a positive wedge between actual and fundamental equity premium → higher discounting of dividends → lower equity price.
  - Procyclical lending standards (φt falls in downturns) increase the equity-premium wedge and further depress equity prices → amplifies asset-price overreaction and macro volatility.
  - Higher average LTV (φ̄) narrows gap between actual and fundamental risk premium, reducing overreaction/volatility.

### Shocks and equilibrium
- Shocks:
  - Productivity shocks A_t (Home) and y*_t (Foreign) follow a bivariate AR(1):
    - ln(A_t) and ln(y*_t) dynamics per equation (30) with contemporaneous variance-covariance V.
  - Credit shocks: ln(φ_t) = (1−ρφ) ln(φ̄) + ρφ ln(φ_{t−1}) + ε_{φt}, ε_{φt} ∼ iid N(0, σ^2_φ).
- Correlation between LTV shocks and Home TFP shocks (ρ_{LTV,A}) used to model different cyclical patterns of lending standards.
- Market clearing and solution:
  - Home and Foreign goods and world credit market clearing conditions specified.
  - With collateral constraint always binding in Home, model is linearized and solved for policy functions.

### Calibration (benchmark and alternatives)
- Benchmark calibration uses standard literature values and post-1999 Estonia (Home proxy) and Sweden (Foreign proxy) where pertinent.
- Preference and technology parameters:
  - β* = 0.99, β = 0.96
  - σ = 1 (benchmark), ψ inverse Frisch wage elasticity = 3 (elasticity 1/3)
  - ν set to match average time worked = 33 percent
  - η = 1.5; γ = γ* = 57.5 percent
  - ω (portfolio adjustment cost) = 0.00074
  - N = 1 (symmetric two-country in benchmark)
  - α = 0.30; δ = 2.5 percent per quarter; capital adjustment cost parameter φ = 10
- Productivity shock parameters:
  - Home AR parameter B11 = 0.94
  - Foreign AR parameter B22 = 0.97
  - Spillovers B12 = B21 = 0 in benchmark
  - σ_{ε1} = 2.2 percent (Home TFP shocks) → Home GDP volatility ≈ 2.8 percent
  - σ_{ε2} = 1.14 percent (Foreign)
  - Corr(ε1, ε2) = 0.04
- Loan-to-value ratio parameters:
  - Unconditional mean φ̄ = 40 percent (benchmark)
  - φ process parameters: ρφ = 0.9, σφ = 1.85 percent
  - Empirical fit: AR(1) estimated on demeaned LTVs from emerging Europe yields ρφ within estimated range; σφ is conservative relative to data.
  - Benchmark correlation between LTV shocks and Home TFP shocks: ρ_{LTV,A} = 0.5 (in line with Baltic empirical evidence).

### Policy experiment: altering cyclical pattern of lending standards
- Objective: quantify impact on cyclical properties of macro variables from reducing procyclicality in lending standards (interpreted as introducing a countercyclical element in prudential regulation).
- Benchmark behavior: lending standards procyclical—credit grows more than the cyclical increase in collateral value in upturns (credit booms) and contracts more in downturns (credit squeezes).
- Simulation design:
  - 500 replications, each 150 quarters long; first 50 quarters dropped to avoid starting-value bias.
  - Time unit = one quarter.
  - Series filtered with Hodrick-Prescott filter; average moments computed across simulations.
  - Impulse responses to negative one-standard-deviation shocks to productivity and lending standards analyzed (Figures referenced in source).
- Scenarios and comparative experiments:
  - Alternative average LTV levels examined (example averages include LTV = 0.4 and LTV = 0.7 in tables and figures).
  - Policy interpreted as exogenous drop in correlation of LTV ratio with cycle (i.e., reducing ρ_{LTV,A}).
  - Sensitivity analysis performed for different model specifications (e.g., alternative σ, γ, spillovers).
- Intuition from simulations:
  - Procyclical φt amplifies equity-price overreaction to shocks and increases volatility of key macro aggregates.
  - Reducing φt–cycle correlation (making lending standards less procyclical) is expected to lower macroeconomic volatility; quantitative gains are explored via simulation (summary results appear in Tables 4–6 and sensitivity analyses in Table 6).

*Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp0935.pdf*

### 0.3 to 0.7) and different degrees of correlation between lending standards and productivity

### 0.3 to 0.7) and different degrees of correlation between lending standards and productivity

### A. Benchmark leverage level
- Calibration assumption: correlation between innovations to the LTV ratio and productivity, ρLT V,A = 0.5 (procyclical credit conditions).
- Key statistical patterns under the Benchmark calibration:
  - Volatility of consumption σc/pH (deflated with the GDP deflator pH) is slightly bigger than volatility of GDP (relative volatility of consumption above one).
  - Foreign debt is very volatile and highly correlated with the price of equity in the Home economy: ρd,q = 0.77 (hence highly procyclical).
  - Current account is negatively correlated with the price of equity: ρca/y,q = −0.32 and highly volatile.
- Policy experiment: make innovations to the LTV ratio independent of domestic TFP (ρLT V,A = 0).
  - Effects (relative to benchmark procyclicality):
    - σq (equity-price volatility) decreases by 3.5 percent.
    - σx (investment volatility) decreases by 3.6 percent.
    - Volatility of debt and current account reduced by as much as 14 percent.
    - σc/pH decreases by 1.7 percent.
    - σ((c/pH)/y) decreases by 13.5 percent (consumption volatility as share of domestic GDP).
    - GDP volatility is reduced mildly.
    - Correlation of equity prices with foreign debt and with the real interest rate decreases.
    - ρca/y,q decreases by about one third.
- Mechanism:
  - A financially constrained economy generates a wedge between the actual and the “fundamental” equity premium (endogenous).
  - Procyclical LTV magnifies the wedge and increases volatility of the equity premium and equity price.
  - Reducing procyclicality smooths the “overreaction” of equity prices to fundamentals, lowering macroeconomic volatility.
- Note on investment reporting:
  - Reported results for investment (x) correspond to gross investment (kt+1 − (1−δ)kt).
  - Reduction of volatility of each variable is expressed as a percentage difference with respect to the volatility under the benchmark cyclicality of lending standards.

### B. Alternative leverage levels
- Scenario: less financially constrained setting (higher average LTV ratio). Simulate both benchmark (procyclical) and active policy scenarios.
- Under procyclical lending standards with higher leverage:
  - Lower volatility of equity prices, investment, real interest rate, and foreign debt.
  - Higher volatility of consumption, both in levels and as fraction of GDP.
- Intuition:
  - Higher leverage implies a given collateral value shock creates a larger wealth effect on household budgets → larger changes in spending (consumption and equity accumulation).
  - Higher leverage reduces average size of the wedge between actual and “fundamental” equity price → lower “overreaction” of equity prices to fundamentals → lower incentive to adjust equity accumulation → investment volatility falls while consumption volatility rises.
  - Following a positive productivity shock, the trade-off tilts more toward consumption in a higher-leverage setting.
- Correlations and external balances under higher leverage:
  - Real interest rate is more correlated with GDP and less correlated with equity prices.
  - Trade balance becomes positively correlated with equity prices.
  - Current account deficit becomes essentially independent from equity-price evolution.
- Policy experiment effects in higher-leverage economy (relative reductions after introducing acyclical credit conditions):
  - Equity-price volatility: −4.0 percent.
  - Investment volatility: −4.4 percent.
  - Foreign debt volatility: −11.5 percent.
  - Consumption volatility: −1.5 percent.
  - Correlations of equity prices with debt and real interest rate reduced; real interest rate correlation becomes even negative.
  - Equity prices and current account deficit co-move in opposite directions: rise in equity prices associated with an improvement in the external deficit.
- Comparative message:
  - When the economy is more financially constrained, there is slightly more to gain in stabilizing consumption and external imbalances (debt, trade balance, current account deficit).
  - The opposite holds for stabilization of equity prices and investment.
  - Volatility gains are estimated around two different steady states when considering the two leverage scenarios.

### VI. Sensitivity analysis
- Dimensions tested: Frisch wage elasticity of labor supply (1/ψ), productivity spillovers across countries (B12 and B21), degree of home bias (γ and γ*), risk aversion coefficient (σ).
- Overall: simulation results are extremely robust; most relevant differences arise from preference changes.
- Wage elasticity of labor supply:
  - Benchmark uses ψ consistent with micro-data; alternative tested ψ = 0.3 (instead of ψ = 3).
  - Volatility gains from smoothing procyclicality broadly unchanged qualitatively and quantitatively.
  - Reduction in investment volatility slightly higher; consumption volatility reduction slightly lower (ratios to GDP essentially unchanged).
  - Larger fall in GDP volatility implied by the policy under ψ = 0.3.
- Productivity spillovers:
  - Alternative assumed B12 = B21 = 0.08 (similar to Backus, Kehoe and Kydland (1992); within 0-0.1 used in Heathcote and Perri (2002)).
  - Adjust B11 and B22 proportionally to keep largest eigenvalue equal to benchmark.
  - Implementing the policy with spillovers yields higher stabilization gains:
    - Investment: −4.3 percent (instead of −3.6 percent).
    - Consumption: −3.3 percent (instead of −1.7 percent).
    - GDP: −0.7 percent (instead of −0.4 percent).
- Consumption home bias:
  - Higher home bias in consumption → policy has similar gains for investment volatility but lower gains for consumption volatility.
  - Real exchange rate volatility higher with higher home bias (approximately four times higher than benchmark); therefore volatility gains from policy on the real exchange rate would be higher.
- Preferences / risk aversion:
  - Benchmark uses logarithmic utility (standard for always-binding collateral constraint models).
  - Alternative tested σ = 2.
  - With σ = 2:
    - Higher gain in GDP stabilization; gains in investment volatility unchanged.
    - Consumption is much less volatile than with log utility under any degree of leverage cyclicality.
    - No reduction in consumption volatility from the policy; consumption deflated by pH may be even more volatile.
    - Gains in consumption-to-GDP volatility are unchanged relative to benchmark and hence robust to changes in σ.

### VII. Conclusions (key findings and policy implications)
- Empirical-consistent findings:
  - Changes in firms’ borrowing are more sensitive to asset-price changes in economies where firms’ leverage co-moves more closely with the business cycle.
  - Higher procyclicality in firms’ leverage is associated with higher volatility in private investment.
  - Oversensitivity of credit availability to asset-price changes, greater procyclicality in firms’ leverage, and higher investment volatility likely characterize economies where growth fluctuations are more vulnerable to bank lending conditions (example: Baltic countries).
- Stylized two-country model highlights:
  - Equities play a double role: collateral in international lending and capital value for production.
  - Lending conditions and discount factors need not be symmetric across countries to generate cross-country differences in investment and current account dynamics.
  - One economy modeled as credit constrained and “impatient” (preferences tilted toward current consumption) with borrowing constrained by collateral value tied to equity prices.
  - Less financially constrained markets represented by higher borrowing limit (stochastic LTV process).
  - Model consistent with empirical observation that investment and current-account dynamics are more responsive to financial conditions in tighter-borrowing economies.
- Policy experiment summary:
  - Smoothing the cyclical pattern of lending standards in a financially-constrained economy leads to sizeable reductions in volatility of:
    - Equity prices
    - Investment
    - Consumption
    - External imbalances: current account deficit, trade balance, net foreign position
  - Cross-country technology spillovers increase stabilization gains.
  - Results robust to variations in wage elasticity, home-bias in consumption, and, to a degree, risk aversion.

### Appendix — VAR-based Financial Condition Indices (FCIs)
- Data and estimation:
  - Country coverage: 16 advanced European economies and 12 emerging European economies.
  - Sample period: 1999:1–2008:4 (monthly data).
  - Country-specific VARs include four endogenous variables:
    1. (real) GDP growth
    2. (real) 3-month interbank rate (CPI)
    3. changes in equity prices (deflated using the CPI)
    4. changes in (real) private credit to the corporate sector
  - With exception of the interest rate, all variables are in logs. A constant is included as exogenous.
  - Lag order: three (Schwartz information criterion).
- FCI construction:
  - Dynamic weight structure computed using impulse response functions (IRFs) from each VAR to incorporate timing of transmission from financial markets to real activity.
  - Identification of contemporaneous shocks via Cholesky factorization; variables ordered by relative sluggishness.
    - GDP growth ordered first (assumed relatively sluggish).
    - Changes in equity prices precede credit growth; the 3-month interbank rate ordered after other variables.
  - FCI formula:
    - FCIt = Σ_{j=1}^n ( r_{t,j}^{t−1} + Σ_{i=2}^m r_{t,j}^{t−i} )
    - r’s denote IRFs of GDP growth with respect to each other variable; j indexes variables; i indexes time periods.
    - First term: response of GDP growth in quarter t to a financial shock occurring in previous quarter.
    - Second summation: marginal current impact of shocks that occurred before the previous quarter measured via their effect on GDP growth in the previous period.
  - Interpretation:
    - FCI measures total contribution to GDP growth in a quarter from shocks to financial variables over the previous m quarters.
    - Dynamic weights arise naturally from IRFs: the weight on a variable i periods into the future equals the response of economic activity at t+i to a shock at t.

*Source: _wp0935 - 0.3 to 0.7) and different degrees of correlation between lending standards and productivity*

### REFERENCES

### _wp0935 - REFERENCES

### References
- Aiyagari, S. Rao and Mark Gertler, 1999, “Overreaction of Asset Prices in General Equilibrium” Review of Economic Dynamics, Vol. 2(1),pp. 3-35.
- Arvai, Zsofia, Karl Driessen, Daniel Hardy, and Inci tker-Robe, 2008, “Regional Financial Interlinkages and Contagion Channels” Chapter 2, Box 5, inRegional Economic Outlook: Europe (Washington: International Monetary Fund, April), pp. 28-30.
- Asea, Patrick K. and Brock Blomberg, 1998, “Lending Cycles”Journal of Econometrics, Vol. 83, pp. 89-128.
- Ayuso, Juan, Perez, Daniel and Jesus Saurina, 2004, “Are capital buffers pro-cyclical?: Evidence from Spanish panel data”Journal of Financial Intermediation, Vol. 13(2), pp. 249-64.
- Backus, David K., Patrick J. Kehoe and Finn E. Kydland, 1992, “International Real Business Cycles” Journal of Political Economy, Vol. 100, pp. 745-75.
- Borio, Claudio, Furfine, Craig and Philip Lowe, 2001, “Procyclicality of the financial system and financial stability: issues and policy options” inMarrying the macro- and micro-prudential dimensions of financial stability, BIS Papers No. 1, March, pp. 1-57.
- Callegari, Giovanni, 2007, “Financial Frictions and Household Debt: a New Perspective on Twin Deficits” unpublished.
- Calza, Alessandro, Monacelli, Tommaso, and Livio Stracca, 2007, “Mortgage Markets, Collateral Constraints, and Monetary Policy: Do Institutional Factors Matter?” CEPR Discussion Papers 6231 (London: Centre for Economic Policy Research).
- Caruana, Jaime and Aditya Narain, 2008, “Banking on More Capital”Finance & Development (Washington: International Monetary Fund), Vol. 45(2), pp. 24-28.
- Dell’Ariccia, Giovanni and Marquez, Robert, 2006, “Lending Booms and Lending Standards” emph Journal of Finance, Vol. 61(5), pp. 2511-46.
- Galí, Jordi, Gertler, Mark and J. David López-Salido, 2007, “Markups, Gaps, and the Welfare Costs of Business Fluctuations”The Review of Economics and Statistics, Vol. 89(1), pp. 44-59.
- Ghironi, Fabio, Iscan, Talan B., and Alessandro Rebucci, 2005, “Net Foreign Asset Positions and Consumption Dynamics in the International Economy” IMF Working Papers 05/82 (Washington: International Monetary Fund).
- Goodhart, Charles, and Avinash Persaud, 2008, “How to Avoid the Next Crash”Financial Times, January 30th.
- Guajardo, Jaime, 2008, “Financial Frictions and Business Cycles in Middle-Income Countries” IMF Working Papers 08/20 (Washington: International Monetary Fund).
- Iacoviello, Matteo, 2005, “House Prices, Borrowing Constraints, and Monetary Policy in the Business Cycle”American Economic Association, vol. 95(3), pages 739-64.
- Iacoviello, Matteo and Raoul Minetti, 2006, “Financial Liberalisation and the Sensitivity of House Prices to Monetary Policy: Theory and Evidence” Boston College Working Papers in Economics No. 538.
- Iacoviello, Matteo and Raoul Minetti, 2006, “International Business Cycles with Domestic and Foreign Lenders”Journal of Monetary Economics, Vol. 53(8), pp. 2267-82.
- Iacoviello, Matteo and Stefano Neri, 2008, “Housing Market Spillovers: Evidence from an Estimated DSGE Model” Boston College Working Papers in Economics No. 659.
- Jiménez, Gabriel and Jesús Saurina, 2006, “Credit Cycles, Credit Risk, and Prudential Regulation” International Journal of Central Banking, Vol. 2(2), pp. 65-98.
- Jokipii, Terhi and Milne, Alistair, 2006, “The Cyclical Behaviour of European Bank Capital Buffers” Bank of Finland Discussion Paper No. 17/2006.
- Kiyotaki, Nobuhiro and John Moore, 1997, “Credit Cycles ”Journal of Political Economy, Vol. 105, pp. 211-48.
- Mendoza, Enrique G., 2006, “Endogenous Sudden Stops In A Business Cycle Model With Collateral Constraints: A Fisherian Deflation Of Tobin’s Q” NBER Working Papers 12564.
- Mendoza, Enrique G. and Katherine A. Smith, 2006, “Quantitative implications of a debt-deflation theory of Sudden Stops and asset prices”Journal of International Economics, Elsevier, Vol. 70(1), pp. 82-114.
- Mendoza, Enrique G. and Marco E. Terrones, 2008, “An Anatomy Of Credit Booms: Evidence From Macro Aggregates And Micro Data” NBER Working Papers 14049.
- Heathcote, Jonathan and Fabrizio Perri, 2002, “Financial Autarky and International Business Cycles.”Journal of Monetary Economics, Elsevier, vol. 49(3), pp. 601-27, April.
- Repullo, Rafael, and Javier Suarez, 2008, “The Procyclical Effects of Basel II.” CEPR Discussion Paper No. 6862 (London: Centre for Economic Policy Research).
- Saurina, Jesús, and Carlos Trucharte, 2007, “An Assessment of Basel II Procyclicality in Mortgage Portfolios.”Journal of Financial Services Research, Vol. 32 (October), pp. 81-101.
- Schmitt-Grohé, Stephanie, and Martin Uribe, 2003, “Closing Small Open Economy Models.” Journal of International Economics, Vol. 61(1), pp. 163-85.
- Swiston, Andrew, 2008, “A U.S. Financial Conditions Index: Putting Credit Where Credit Is Due.” IMF Working Paper 08/161 (Washington: International Monetary Fund).
- Taylor, Ashley, and Charles Goodhart, 2006, “Procyclicality and Volatility in the Financial System: The Implementation of Basel II and IAS 39” inProcyclicality of Financial Systems in Asia, ed. by Stefn Gerlach and Paul Gruenwald (Houndsmill, Basingstoke: Palgrave Macmillan).
- Uribe, Martin, and Vivian Z. Yue 2006. “Country Spreads and Emerging Countries: Who Drives Whom?”Journal of International Economics, Vol. 69(1), pp. 6-36.

### Tables and Key Numeric Results
- Table 1: Results from Estimating an AR(1) Processes to Demeaned LTVs.
  - Equation: ln(φt) = ρφ ln(φt−1) + εφt, εφt ∼ iid N(0, σ2φ)
  - Country / Observations / ̂ρ / Std.err. / ̂σφ
    - Croatia 46 0.92 0.052 0.055
    - Estonia 49 0.89 0.0458 0.076
    - Lithuania 34 0.91 0.083 0.046
    - Russia 37 0.65 0.157 0.062
    - Slovenia 52 0.84 0.0922 0.046
    - Turkey 53 0.8 0.110 0.084

- Table 2: Benchmark Calibration.
  - Preferences
    - β Home discount factor 0.96
    - β* Foreign discount factor 0.99
    - σ Coefficient of risk aversion 1
    - ψ Inverse of (Frisch) wage elasticity of labor supply 3
    - ν Disutility of labor parameter Set to match h=33% in SS
    - γ, γ* Bias in consumption towards domestically produced goods 0.575
    - η Elasticity of substitution between domestic and foreign goods 1.5
  - Technology and Productivity
    - α Capital share in Production 0.3
    - δ Capital depreciation rate (quarterly) 0.025
    - φ Capital adjustment cost parameter 10
    - B11 Home TFP autoregressive parameter 0.94
    - σε1 Standard deviation of Home TFP shocks 0.022
    - B22 Foreign endowment autoregressive parameter 0.97
    - σε2 Standard deviation of Foreign endowment shocks 0.0114
    - B12 Spillover parameter, Foreign to Home 0
    - B21 Spillover parameter, Home to Foreign B12
    - ρ ε1,ε2 Cross-correlation A-y* shocks 0.07
  - Credit standards
    - ̄φ Average LTV ratio 0.4
    - ρφ LTV autoregressive parameter 0.9
    - σφ Standard deviation of lending standards shocks 0.0185
    - ρLT V,A Cross-correlation LTV-A shocks 0.5
  - Other
    - ω Portfolio adjustment cost parameter 0.00074
    - N Relative size of Foreign relative to Home economy 1

- Table 3: Business Cycle Moments from Simulated Series under Benchmark Calibration.
  - Volatilities
    - σy 0.0285
    - σq 0.017
    - σx 0.0674
    - σR 0.022
    - σx/y 0.0396
    - σζ 0.0026
    - σc 0.0256
    - σnx/y 0.0144
    - σc/pH 0.0286
    - σca/y 0.0642
    - σ(c/pH)/y 0.0163
    - σd 0.033
  - Cross-correlations with Home GDP
    - ρx,y 0.9843
    - ρq,y 0.9694
    - ρx/y,y 0.9538
    - ρR,y 0.1772
    - ρc,y 0.6995
    - ρd,y 0.7244
    - ρc/pH ,y 0.8369
    - ρnx/y,y -0.0939
    - ρ(c/pH)/y ,y -0.2834
    - ρca/y,y -0.1575
    - ρh,y 0.2834
    - ρy,y−1 0.7212
  - Cross-correlations with equity prices
    - ρx,q 0.9963
    - ρR,q 0.3361
    - ρc,q 0.7823
    - ρnx/y,q -0.2455
    - ρd,q 0.7721
    - ρca/y,q -0.3161

- Table 4: Policy Exercise Results (Average LTV = 0.4). (Columns: ρLT V,A = 0.5 ; ρLT V,A = 0% ; difference)
  - Volatilities
    - σy 0.0285 / 0.0284 / -0.4%
    - σx 0.0674 / 0.065 / -3.6%
    - σx/y 0.0396 / 0.0372 / -6.1%
    - σc 0.0256 / 0.0245 / -4.3%
    - σc/pH 0.0286 / 0.0281 / -1.7%
    - σ(c/pH)/y 0.0163 / 0.0141 / -13.5%
    - σq 0.017 / 0.0164 / -3.5%
    - σR 0.022 / 0.0189 / -14.1%
    - σnx/y 0.0144 / 0.0125 / -13.2%
    - σca/y 0.0642 / 0.0552 / -14.0%
    - σd 0.033 / 0.0284 / -13.9%
  - Cross-correlations with Home GDP
    - ρx,y 0.9843 / 0.9865 / 0.2%
    - ρc,y 0.6995 / 0.7574 / 8.3%
    - ρq,y 0.9694 / 0.973 / 0.4%
    - ρR,y 0.1772 / 0.0664 / -62.5%
    - ρnx/y,y -0.0939 / -0.1395 / 48.6%
    - ρca/y,y -0.1575 / -0.0765 / -51.4%
    - ρd,y 0.7244 / 0.4205 / -42.0%
  - Cross-correlations with equity prices
    - ρd,q 0.7721 / 0.4822 / -37.5%
    - ρR,q 0.3361 / 0.2039 / -39.3%
    - ρnx/y,q -0.2455 / -0.2605 / 6.1%
    - ρca/y,q -0.3161 / -0.2124 / -32.8%

- Table 5: Policy Exercise Results (Average LTV = 0.7). (Columns: ρLT V,A = 0.5 ; ρLT V,A = 0% ; difference)
  - Volatilities
    - σy 0.0281 / 0.028 / -0.4%
    - σx 0.0596 / 0.057 / -4.4%
    - σx/y 0.034 / 0.0315 / -7.4%
    - σc 0.0307 / 0.0297 / -3.3%
    - σc/pH 0.0335 / 0.033 / -1.5%
    - σ(c/pH)/y 0.0239 / 0.0213 / -10.9%
    - σq 0.015 / 0.0144 / -4.0%
    - σR 0.016 / 0.0143 / -10.6%
    - σnx/y 0.0156 / 0.014 / -10.3%
    - σca/y 0.0922 / 0.0816 / -11.5%
    - σd 0.0295 / 0.0261 / -11.5%
  - Cross-correlations with Home GDP
    - ρx,y 0.9501 / 0.9523 / 0.2%
    - ρc,y 0.5789 / 0.6492 / 12.1%
    - ρq,y 0.9444 / 0.9439 / -0.1%
    - ρR,y 0.2163 / 0.0381 / -82.4%
    - ρnx/y,y -0.088 / -0.1638 / 85.2%
    - ρca/y,y -0.1933 / -0.0489 / -74.7%
    - ρd,y 0.697 / 0.3398 / -51.2%
  - Cross-correlations with equity prices
    - ρd,q 0.7151 / 0.4055 / -43.3%
    - ρR,q -0.0069 / -0.1638 / 2273.9%
    - ρnx/y,q 0.1792 / 0.1102 / -38.5%
    - ρca/y,q 0.0391 / 0.1674 / 328.1%

- Table 6: Sensitivity Analysis. (Sensitivity Analysis - LTV=0.4)
  - Scenarios shown: Benchmark ; ψ=0.3 ; B12 = B21 = 0.08 ; γ = γ* = 0.75 ; σ = 2
  - Selected volatilities and percent differences versus benchmark:
    - σy: 0.0285 (-0.4%); 0.0307 (-1.6%); 0.0272 (-0.7%); 0.0286 (-1.0%); 0.0264 (-1.9%)
    - σx: 0.0674 (-3.6%); 0.0711 (-4.1%); 0.0634 (-4.3%); 0.0649 (-3.5%); 0.0715 (-3.8%)
    - σx/y: 0.0396 (-6.1%); 0.0411 (-6.1%); 0.0368 (-6.8%); 0.0367 (-5.7%); 0.0458 (-4.8%)
    - σc: 0.0256 (-4.3%); 0.024 (-3.3%); 0.0248 (-5.2%); 0.0269 (-2.2%); 0.0184 0.0%
    - σ(c/pH)/y: 0.0163 (-13.5%); 0.0161 (-13.7%); 0.0159 (-11.3%); 0.0134 (-14.9%); 0.0107 (-14.0%)
    - σR: 0.022 (-14.1%); 0.0198 (-13.6%); 0.0219 (-13.7%); 0.023 (-15.7%); 0.0247 (-14.6%)
    - ρx,y: 0.9843 0.2%; 0.9876 0.2%; 0.9864 0.1%; 0.9926 0.0%; 0.9828 0.2%
    - ρc,y: 0.6995 8.3%; 0.7064 8.2%; 0.7042 4.6%; 0.7988 6.4%; 0.8018 5.4%
    - ρR,y: 0.1772 -62.5%; 0.0404 -238.6%; 0.2457 -56.3%; 0.2302 -43.0%; 0.1823 -50.7%
    - ρnx/y,y: -0.0939 48.6%; 0.0424 -148.8%; -0.0801 -3.4%; -0.1286 5.6%; -0.0717 161.5%
    - ρca/y,y: -0.1575 -51.4%; -0.019 -348.9%; -0.2095 -54.7%; -0.2133 -34.0%; -0.1688 -41.2%
    - ρd,y: 0.7244 -42.0%; 0.7183 -39.8%; 0.7222 -44.6%; 0.7785 -33.9%; 0.7609 -35.4%
    - ρd,q: 0.7721 -37.5%; 0.7651 -36.2%; 0.7561 -38.8%; 0.7921 -31.3%; 0.7958 -32.4%
    - ρR,q: 0.3361 -39.3%; 0.1794 -66.6%; 0.3973 -38.0%; 0.2266 -43.5%; 0.3324 -33.7%
    - ρnx/y,q: -0.2455 6.1%; -0.0879 36.1%; -0.2277 -13.5%; -0.1097 68.3%; -0.2258 37.4%
    - ρca/y,q: -0.3161 -32.8%; -0.1574 -57.6%; -0.3618 -36.0%; -0.2075 -35.4%; -0.321 -28.0%

### Figures (listed in source)
- Figure 1: Time Variation in Loan-To-Value Ratios.
- Figure 2: Share of Output Variation Explained by Credit and Asset Price Shocks.
- Figure 3: Degree of Cyclicality in Credit Innovations.
- Figure 4: Procyclicality in Credit Innovations and Sensitivity of Credit to Asset Price Shocks.
- Figure 5: Procyclicality in Credit Innovations and Macroeconomic Volatility.
- Figure 6: Increasing Reliance of Emerging Europe on Foreign Funding.
- Figure 7: Concentration of Emerging Europe Exposure to Western Europe.
- Figure 8: IRFs to a Negative Productivity Shock under Alternative Leverage Levels.
- Figure 9: IRFs to a Negative Shock to Lending Standards under Alternative Leverage Levels.
- Figure 10: Sensitivity of Volatility to Different Degrees of Cyclicality in Lending Standards Under Alternative Leverage Levels.

*Content of this unit: REFERENCES, tables, and figures as presented in the source PDF.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp0935.pdf_