## _wp11133 - References

## Source details

**Canonical URL:** [_wp11133 - References](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11133.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11133.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11133.pdf.json)

---

### I. Introduction — key findings and motivations
- Business-cycle regularities in emerging market economies (EMEs):
  - Strongly countercyclical current account.
  - Sovereign interest rates that are highly countercyclical, very volatile, and significantly higher than the World interest rate.
  - Total consumption expenditure volatility exceeds income volatility.
- Empirical benchmark:
  - Aguiar and Gopinath (2007): consumption is 40 percent more volatile than income for emerging economies; slightly less volatile than income for developed economies.
- Paper objective:
  - Explore explanations for EME business-cycle regularities when consumption comprises consumer durables and nondurables.
  - Combine shocks to trend and shocks to cycle (Aguiar and Gopinath, 2007) with financial frictions and durable goods.
  - Calibrate model to match key business-cycle facts for Mexico.
- Core insights:
  - Introducing durable goods alters the role of trend shocks: result in Aguiar and Gopinath (2007) that trend shocks primarily drive EME cycles does not hold once durables are included.
  - Financial frictions (limited access to international borrowing, countercyclical borrowing premium) are important:
    - Endogenous borrowing limits increase consumption volatility when limits bind (De Resende, 2006).
    - Working-capital borrowing requirements and countercyclical borrowing premia amplify consumption variability via labor demand sensitivity to interest rates (Neumeyer and Perri, 2005).
    - Shocks to the volatility of the borrowing premium are significant (Fernández-Villaverde et al., 2009).
  - Interaction between financial frictions and durable goods:
    - Durable purchases yield utility over time, acting partially as postponed consumption (saving), so durable purchases respond strongly to interest rates.
    - A countercyclical interest rate during temporary income expansions encourages borrowing to finance durables, generating highly volatile durable purchases and a strongly countercyclical trade balance.

### II. Data and stylized facts
- Sample construction and grouping:
  - Countries included only if data on durable-goods spending are available.
  - Countries classified as emerging market economies if rated so by MSCI for most of the sample period.
  - Large developed economy if its GDP represented at least 2% of the world’s GDP from 2001 till 2008.
  - Three groups:
    - Small developed economies: Canada, Denmark, Finland, Netherlands, New Zealand, Spain.
    - Large developed economies: France, Italy, Japan, United Kingdom, United States.
    - Emerging market economies: Chile, Colombia, Czech Republic, Israel, Mexico, Taiwan Province of China, Turkey.
- Data sources and processing:
  - Variables (real terms, quarterly except Colombia annual): gross domestic product, total private consumption expenditure, consumption of nondurable goods, expenditure in durable goods, investment, trade balance.
  - Sources: OECD Quarterly National Accounts (developed economies and Czech Republic), Bureau of Economic Analysis (U.S.), Central Bank of Chile (Chile), DANE (Colombia), Bank of Israel and Central Bureau of Statistics (Israel), INEGI (Mexico), National Statistics, Republic of China (Taiwan) (Taiwan), Turkish Statistical Institute (Turkey).
  - All variables seasonally adjusted when needed, converted to logs, and detrended using the Hoddrick-Prescott filter.
- Stylized facts:
  - Real interest rates are countercyclical and leading in emerging markets and acyclical and lagging in developed economies.
  - Emerging market economies have higher volatilities of output and consumption compared to developed economies.
  - Consumption expenditure is more volatile than output in emerging markets while it is not (quite) as volatile as output in developed economies.
- Empirical facts on consumption, durables, and net exports:
  - Consumption of nondurables: ratio of standard deviation of consumption-to-income:
    - emerging economies: 0.9
    - developed economies: 0.72
  - Spending in durable goods: much more volatile than output in both emerging and developed economies.
  - Cross-country variation (emerging markets):
    - High end: Chile, Israel, Turkey.
    - Low end: Taiwan Province of China, Colombia, Czech Republic.
    - Colombia: total consumption = 1.02; nondurable consumption = 0.81.
    - Taiwan Province of China: total consumption = 0.80; nondurable consumption = 0.92.
    - Czech Republic: total consumption = 0.98; nondurable consumption = 0.88.
  - Trade balance correlations:
    - Developed economies: mildly countercyclical; comparable averages reported by Aguiar and Gopinath (2007) = -0.17 and Engel and Wang (2011) = -0.25.
    - Emerging economies (sample): range from procyclical (Taiwan Province of China = 0.25) to strongly countercyclical (Mexico = -0.82).
    - Sample average correlation trade balance with GDP: -0.20.
    - Larger-sample result (Aguiar and Gopinath): average correlation between net exports and output = -0.51.
- Table-level summary (selected entries preserved):
  - Table 3. Average Relative Volatility of Consumption (weighted ratios to σy):
    - Total Consumption — Large Economies 0.94, Small Economies 0.99, Emerging Economies 1.30.
    - Non-durables — Large Economies 0.68, Small Economies 0.72, Emerging Economies 0.90.
    - Durables — Large Economies 3.83, Small Economies 3.85, Emerging Economies 4.48.

### III. Model structure
- Economy: small open economy two-sector neoclassical growth model with durables and nondurables.
- Goods and markets:
  - Two consumption goods: nondurables (non-tradable) and durables (tradable).
  - Markets incomplete: one financial asset, a risk-free bond paying interest in units of durable good.
- Production technologies:
  - Nondurable sector: Y_{n,t} = e^{z_{n,t}} K_{n,t}^{α_n} (Γ_t L_{n,t})^{1-α_n}.
  - Durable sector: Y_{d,t} = min{ e^{z_{d,t}} K_{d,t}^{α_d} (Γ_t L_{d,t})^{1-α_d} ; M_{n,t} / Ω } with Ω units of nondurable input per durable.
- Shocks:
  - Two temporary sectoral TFP shocks with AR(1): z_{n,t}, z_{d,t}.
  - One aggregate shock to trend: g_t, with Γ_t = Γ_{t-1} e^{g_t}.
  - Contemporaneous covariance matrix Σ_Z includes ρ_{nd} ≠ 0.
- Preferences and aggregator:
  - Utility: E_0 Σ β^t U(C_t, 1-L_t) with U(C_t,1-L_t) = (C_t^{θ}(1-L_t)^{1-θ})^{1-σ}/(1-σ).
  - Consumption aggregator: C_t = ( μ N_t^{-γ} + (1-μ) D_t^{-γ} )^{-1/γ}.
  - 1/(1+γ) is elasticity of substitution between durables and nondurables; μ is utility share of nondurables; θ utility share of consumption; σ relative risk aversion.
- Resource and accumulation equations:
  - X_{d,t} + X^k_{d,t} + X^k_{n,t} + q_t B_{t+1} = Y_{d,t} + B_t.
  - N_t = Y_{n,t} + M_{n,t}.
  - Capital and durable accumulation with depreciation δ_k, δ_d and quadratic adjustment costs Φ(·), Ψ(·). Second-hand market for durables allowed.
- Bond price specification (reduced-form borrowing premium with countercyclical term):
  - q_t = 1/(1+r^*) + χ [ exp( ( B̃_{t+1} / Γ_t ) - B̄ ) - 1 ] + η ( E_t[ Y_{t+1} / Γ_t ] - Ȳ ).
  - B̄ and Ȳ are steady-state detrended bond stock and total output.
  - η negative and much smaller for emerging economies than developed economies implies countercyclical borrowing premia.

### IV. Calibration (Mexico, quarterly)
- Income share of labor:
  - nondurables: 0.48
  - durables: 0.68
- Annual depreciation rates:
  - capital: 7.1 percent
  - durables: 15.6 percent
- Intermediate input coefficient Ω = 0.3.
- Utility and preference parameters:
  - μ = 0.881 (matches average nondurables share 91.8 percent)
  - θ = 0.413 (matches steady-state share of time devoted to work = 1/3)
  - β = 0.98
  - σ = 2
  - φ (capital adjustment cost) = 4
  - elasticity of substitution 1/(1+γ) = 0.86
- Debt and borrowing-premium parameters:
  - steady-state level of debt relative to GDP: - B̄ / Ȳ = 0.1
  - χ = -0.001
- Parameters targeted/estimated by matching business-cycle moments: variances σ_g^2, σ_n^2, σ_d^2; correlation ρ_{nd}; durable adjustment cost ψ; financial friction η.

### V. Results — No Financial Frictions (η = 0)
- Estimated parameters (column (1)):
  - σ_g = 1.8340
  - σ_n = 1.4209
  - σ_d = 1.4528
  - ψ = 0.5336
  - ρ_{nd} = 0.3631
- Fit and implications:
  - Exact matching of targeted moments (GDP, nondurable consumption, durable expenditure, net exports-GDP ratio, sectoral output correlation).
  - Random walk (permanent) component for sectoral Solow residuals substantially smaller than Aguiar and Gopinath’s estimates for Mexico.
  - Permanent shock accounts for about 32 percent of GDP volatility.
  - Model reproduces countercyclical trade balance but only about half as strong as in data.
  - Model underestimates investment volatility: simulated 3.28 vs data 6.11.
  - Model delivers strongly procyclical borrowing premium: correlation with GDP = 0.56 vs data = -0.49.
- Durability shut-down experiment:
  - Set δ_d ≈ 1 and ψ = 0, recalibrate μ and θ to keep steady-state shares/time worked.
  - Results (column (2)): corner solution with σ_d = 0; shocks to trend contribute 57 percent of GDP volatility.
  - Targeting total consumption instead of components increases importance of shocks to trend.

### VI. Results — Financial Frictions case (induced country risk; η < 0)
- Calibration exercise (column (4)) targeting volatilities including borrowing premium correlation:
  - Estimated η = -0.0077 (consistent with induced country risk).
  - Net exports-GDP correlation with GDP: -0.72 (much closer to sample counterpart).
  - Shocks to trend account for about 24 percent of GDP variance.
- Mechanism:
  - Countercyclical borrowing premium → interest rates fall in booms → cheaper credit → more borrowing to accumulate durables and capital → increased imports (durables tradable) → net exports fall in booms → more countercyclical trade balance.
- Impulse responses:
  - Permanent technology shock: financial friction has little effect (wealth effect dominates).
  - Temporary nondurable TFP shock: financial friction matters — borrowing premium falls only with the friction, durable spending and investment increase more, net exports/GDP drop more.
  - Temporary durable TFP shock: financial friction less important because durables are a small fraction of total output.
- Interaction patterns:
  - As financial frictions increase (η lower), correlation of net exports with GDP becomes dominated by size of friction and less sensitive to σ_g.
  - σ_g helps pin down volatilities of nondurable and durable consumption for any η.

### VII. Robustness checks
- Added orthogonal shock to borrowing premium ε_b with variance σ_b^2 (column (5)):
  - Unable to exactly match seven moments; overall quality of fit almost unchanged.
  - Estimated σ_b = 0 (no role for orthogonal shock in matching moments here).
  - Interpreted as evidence that financial frictions in emerging economies need to be correlated with income.
- Alternative borrowing-premium specification decomposing sensitivity to each shock:
  - q_t = 1/(1+r^*) + χ[ exp( B̃_{t+1}/Γ_t - B̄ ) - 1 ] + η_g (g_t - μ_g) + η_n z_{n,t} + η_d z_{d,t}.
  - Solved model for different χ: -0.0005, -0.001, -0.0015.
  - Finding: η_g decreases with more negative χ; for benchmark χ = -0.001, indebtedness effect explains positive association between shocks to trend and borrowing premium; bias on trend shock importance is second order.
  - Qualitative conclusions robust; original parsimonious specification delivers better fit.

### VIII. Key quantitative outcomes and limitations
- Selected quantitative outcomes:
  - No-friction scenario: permanent shock contribution ≈ 32 percent of GDP volatility.
  - Durability shut-down scenario: permanent shock contribution ≈ 57 percent.
  - Financial-friction scenario (η = -0.0077): permanent shock contribution ≈ 24 percent; net exports-GDP correlation ≈ -0.72.
  - Model underestimates investment volatility: simulated 3.28 vs data 6.11.
  - Orthogonal interest-rate shocks (ε_b) estimated zero in robustness exercise.
- Model difficulty:
  - Matching investment volatility remains a challenge.
  - Orthogonal interest-rate shocks do not help fit under maintained specification.

### IX. Policy and research implications / conclusions
- Financial frictions in the form of a countercyclical country risk premium are essential to:
  - Explain a strongly countercyclical trade balance in emerging markets.
  - Explain a countercyclical and very volatile borrowing premium.
- Durables accumulation interacts with financial frictions to magnify spending responses; durables matter for trade dynamics because durables are largely tradable.
- Shocks to trend are relevant but not the main source of fluctuations in emerging economies once durables and financial frictions are modeled.
- Suggested future research directions:
  - Consider exogenous shocks to the borrowing premium (orthogonal interest-rate shocks).
  - Explore other types of financial frictions and external shocks (e.g., shocks to external wealth, portfolio rebalancing, terms-of-trade shocks) and their interaction with durables and borrowing conditions.
  - Add micro-foundations to the borrowing premium mechanism.

*Source: _wp11133 - REFERENCES (IMF working paper).*

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

### _wp11133 - References

### Figures and Tables (as listed in source)
- Figures:
  - 1.Relative Volatility of Consumption and Saving Rates Across Countries
  - 2.Effect of Shocks to Trend on Key Moments
  - 3.Impulse-Response Functions for a Permanent Shock
  - 4.Impulse-Response Functions for a Shock to Nondurables
  - 5.Impulse-Response Functions for a Shock to Durables
  - 6.Effect of Financial Frictions on Key Moments
- Tables:
  - 1.Volatility of Output and Relative Volatility of Consumption, Investment and Net Exports
  - 2.Correlations with Output
  - 3.Average Relative Volatility of Consumption
  - 4.Benchmark Parameter Values
  - 5.Simulated moments and parameter estimates for Mexico
  - 6.Simulated moments and parameter estimates for an alternative specifications of the borrowing premium

### I. INTRODUCTION — key findings and motivations
- Business-cycle regularities in emerging market economies (EMEs):
  - Strongly countercyclical current account.
  - Sovereign interest rates that are highly countercyclical, very volatile, and significantly higher than the World interest rate.
  - Total consumption expenditure volatility exceeds income volatility.
- Empirical benchmark cited:
  - Aguiar and Gopinath (2007) report consumption is 40 percent more volatile than income for emerging economies, while slightly less volatile than income for developed economies (the "excess volatility of consumption" puzzle).
- Paper objective:
  - Explore explanations for EME business-cycle regularities when consumption comprises consumer durables and nondurables.
  - Build a model combining shocks to trend and shocks to cycle (Aguiar and Gopinath, 2007) with financial frictions and durable goods.
  - Calibrate the model to match key business-cycle facts for Mexico.
- Core insights:
  - Introducing durable goods alters the role of trend shocks: the result in Aguiar and Gopinath (2007) that trend shocks primarily drive EME cycles does not hold once durables are included.
  - Financial frictions (limited access to international borrowing, countercyclical borrowing premium) are important:
    - Endogenous borrowing limits increase consumption volatility when limits bind (De Resende, 2006).
    - Working-capital borrowing requirements and countercyclical borrowing premia amplify consumption variability via labor demand sensitivity to interest rates (Neumeyer and Perri, 2005).
    - Shocks to the volatility of the borrowing premium are significant (Fernández-Villaverde et al., 2009).
  - Interaction between financial frictions and durable goods is especially important:
    - Durable purchases yield utility over time, acting partially as postponed consumption (saving), so durable purchases respond strongly to interest rates.
    - A countercyclical interest rate during temporary income expansions encourages borrowing to finance durables, generating highly volatile durable purchases and a strongly countercyclical trade balance.
- Comparative literature positioning:
  - Extends Aguiar and Gopinath (2007) by including durables.
  - Highlights differences with Neumeyer and Perri (2005) and García-Cicco, Pancrazi, and Uribe (2010) by emphasizing financial frictions in the presence of durables.
  - Notes other relevant studies (De Gregorio, Guidotti, and Vegh, 1998; Engel and Wang, 2011) but underscores that they do not jointly study financial frictions and durables for EMEs.

### Main quantitative and conceptual results summarized
- Decomposing consumption into durables and nondurables:
  - The excess volatility of consumption puzzle is "explained away" in the sense that nondurable consumption is not more volatile than income in either emerging or developed economies.
  - Empirical ratios:
    - Ratio of the standard deviation of nondurable consumption relative to that of income is 0.9 for a sample of emerging economies.
    - The same ratio is 0.72 for a sample of developed economies.
- Role of shocks:
  - With durables and nondurables, shocks to trend appear less important as the main driving force for EME cycles than reported by Aguiar and Gopinath (2007).
  - A financial friction in the form of a countercyclical borrowing premium (the "induced country risk" hypothesis of Neumeyer and Perri, 2005) substantially improves the model's fit to key business-cycle moments for Mexico and reduces the importance attributed to trend shocks.
- Policy-relevant conclusion:
  - Accounting for EMEs' main business-cycle characteristics requires explicit modeling of financial frictions alongside durable goods; relying solely on properties of technology shocks (trend shocks) is insufficient.

### II. DATA AND STYLIZED FACTS
- Sample construction and grouping:
  - Countries included only if data on durable-goods spending are available.
  - Countries classified as emerging market economies if rated so by MSCI for most of the sample period.
  - Developed economies split by size: an economy is "large" if its GDP represented at least 2% of the world’s GDP from 2001 till 2008.
  - Three groups:
    - Small developed economies: Canada, Denmark, Finland, Netherlands, New Zealand, Spain.
    - Large developed economies: France, Italy, Japan, United Kingdom, United States.
    - Emerging market economies: Chile, Colombia, Czech Republic, Israel, Mexico, Taiwan Province of China, Turkey.
  - Note on classification: Israel included as emerging for sample period despite later upgrade; Czech Republic included as emerging despite IMF World Economic Outlook upgrade after sample period.
- Data sources and processing:
  - Variables (real terms, quarterly except Colombia annual): gross domestic product, total private consumption expenditure, consumption of nondurable goods, expenditure in durable goods, investment, trade balance.
  - Sources: OECD Quarterly National Accounts (developed economies and Czech Republic), Bureau of Economic Analysis (U.S.), Central Bank of Chile (Chile), DANE (Colombia), Bank of Israel and Central Bureau of Statistics (Israel), INEGI (Mexico), National Statistics, Republic of China (Taiwan) (Taiwan), Turkish Statistical Institute (Turkey).
  - All variables seasonally adjusted when needed, converted to logs, and detrended using the Hoddrick-Prescott filter.
- Stylized facts for EMEs and small open developed economies (from literature: Aguiar and Gopinath, 2007; Neumeyer and Perri, 2005; Uribe and Yue, 2006; García-Cicco, Pancrazi, and Uribe, 2010):
  1. Real interest rates are countercyclical and leading in emerging markets and acyclical and lagging in developed economies;
  2. Emerging market economies have higher volatilities of output and consumption compared to developed economies;
  3. Consumption expenditure is more volatile than output in emerging markets while it is not (quite) as volatile as output in developed economies.

_Italic line: Source: _wp11133 - References (from the provided PDF content)._

### 4.  Net exports are more volatile and more countercyclical in emerging markets when com-

### 4.  Net exports are more volatile and more countercyclical in emerging markets when com-

### Empirical facts on consumption, durables, and net exports
- Consumption of nondurables: ratio of standard deviation of consumption-to-income
  - emerging economies: 0.9
  - developed economies: 0.72
- Spending in durable goods: much more volatile than output in both emerging and
  developed economies.
- Overall: consumption spending in both durables and nondurables is relatively more
  volatile in emerging markets than in developed economies.
- Cross-country variation (emerging markets):
  - High end: Chile, Israel, Turkey.
  - Low end: Taiwan Province of China, Colombia, Czech Republic.
  - Colombia: total consumption = 1.02; nondurable consumption = 0.81 (annual data
    frequency explanation).
  - Taiwan Province of China: total consumption = 0.80; nondurable consumption = 0.92.
  - Czech Republic: total consumption = 0.98; nondurable consumption = 0.88.
- Relationship between relative volatility of consumption and average saving rate for
  1960-1995: negative relationship (World Bank’s World Saving Database).
- Trade balance correlations:
  - Developed economies: mildly countercyclical; comparable averages reported by
    Aguiar and Gopinath (2007) = -0.17 and Engel and Wang (2011) = -0.25.
  - Emerging economies (sample): range from procyclical (Taiwan Province of China = 0.25)
    to strongly countercyclical (Mexico = -0.82).
  - Sample average correlation trade balance with GDP: -0.20.
  - Larger-sample result (Aguiar and Gopinath): average correlation between net exports and
    output = -0.51.

### Model structure (small open economy two-sector neoclassical growth model)
- Two consumption goods: nondurables (non-tradable) and durables (tradable).
- Markets incomplete: one financial asset, a risk-free bond paying interest in units of durable
  good.
- Production:
  - Nondurable sector: Cobb-Douglas technology Y_{n,t} = e^{z_{n,t}} K_{n,t}^{α_n} (Γ_t L_{n,t})^{1-α_n}.
  - Durable sector: Leontief intermediate input with value-added Cobb-Douglas:
    Y_{d,t} = min{ e^{z_{d,t}} K_{d,t}^{α_d} (Γ_t L_{d,t})^{1-α_d} ; M_{n,t} / Ω }.
  - M_{n,t} is nondurable intermediate input; Ω units needed per durable.
- Shocks:
  - Two temporary sectoral TFP shocks with AR(1): z_{n,t}, z_{d,t}.
  - One aggregate shock to trend: g_t, with Γ_t = Γ_{t-1} e^{g_t}.
  - Contemporaneous covariance matrix Σ_Z includes ρ_{nd} ≠ 0.
- Preferences and consumption aggregator:
  - Utility: E_0 Σ β^t U(C_t, 1-L_t) with U(C_t,1-L_t) = (C_t^{θ}(1-L_t)^{1-θ})^{1-σ}/(1-σ).
  - Consumption aggregator: C_t = ( μ N_t^{-γ} + (1-μ) D_t^{-γ} )^{-1/γ}.
  - 1/(1+γ) is elasticity of substitution between durables and nondurables; μ is utility
    share of nondurables; θ utility share of consumption; σ relative risk aversion.
- Resource constraints and laws of motion:
  - X_{d,t} + X^k_{d,t} + X^k_{n,t} + q_t B_{t+1} = Y_{d,t} + B_t.
  - N_t = Y_{n,t} + M_{n,t}.
  - Capital and durable accumulation with depreciation δ_k, δ_d and quadratic adjustment
    costs Φ(·), Ψ(·). Second-hand market for durables allowed.
- Bond price q_t specification (reduced form borrowing premium with countercyclical term):
  q_t = 1/(1+r^*) + χ [ exp( ( B̃_{t+1} / Γ_t ) - B̄ ) - 1 ] + η ( E_t[ Y_{t+1} / Γ_t ] - Ȳ ).
  - B̄ and Ȳ are steady-state detrended bond stock and total output.
  - η negative and much smaller for emerging economies than developed economies implies
    countercyclical borrowing premia.

### Calibration (Mexican economy, quarterly frequency)
- Income share of labor:
  - nondurables: 0.48
  - durables: 0.68
- Annual depreciation rates:
  - capital: 7.1 percent
  - durables: 15.6 percent
- Intermediate input coefficient Ω = 0.3.
- Utility and preference parameters:
  - μ = 0.881 (matches average nondurables share 91.8 percent)
  - θ = 0.413 (matches steady-state share of time devoted to work = 1/3)
  - β = 0.98
  - σ = 2
  - φ (capital adjustment cost) = 4
  - elasticity of substitution 1/(1+γ) = 0.86
- Debt and borrowing-premium parameters:
  - steady-state level of debt relative to GDP: - B̄ / Ȳ = 0.1
  - χ = -0.001
- Parameters estimated/calibrated by matching business-cycle moments: variances σ_g^2, σ_n^2, σ_d^2, correlation ρ_{nd}, durable adjustment cost ψ, financial friction η.

### Results — No Financial Frictions case (η = 0)
- Estimated parameters (column (1) results):
  - σ_g = 1.8340
  - σ_n = 1.4209
  - σ_d = 1.4528
  - ψ = 0.5336
  - ρ_{nd} = 0.3631
- Fit and implications:
  - Exact matching of targeted moments (GDP, nondurable consumption, durable expenditure,
    net exports-GDP ratio, and sectoral output correlation).
  - Random walk (permanent) component for sectoral Solow residuals substantially smaller than
    Aguiar and Gopinath’s estimates for Mexico.
  - Permanent shock accounts for about 32 percent of GDP volatility.
  - Model reproduces countercyclical trade balance but only about half as strong as in data.
  - Model underestimates investment volatility: simulated 3.28 vs data 6.11.
  - Model delivers strongly procyclical borrowing premium: correlation with GDP = 0.56 vs
    data = -0.49.
- Durability shut-down experiment:
  - Set δ_d ≈ 1 and ψ = 0, recalibrate μ and θ to keep steady-state shares/time worked.
  - Results (column (2)): corner solution with σ_d = 0; shocks to trend contribute 57 percent of GDP volatility.
  - Targeting total consumption instead of components increases importance of shocks to trend.

### Results — Financial Frictions case (induced country risk; η < 0)
- Calibration exercise (column (4)) targeting volatilities including borrowing premium correlation:
  - Estimated η = -0.0077 (consistent with induced country risk).
  - Net exports-GDP correlation with GDP: -0.72 (much closer to sample counterpart).
  - Shocks to trend account for about 24 percent of GDP variance.
- Mechanism:
  - Countercyclical borrowing premium → interest rates fall in booms → cheaper credit → more
    borrowing to accumulate durables and capital → increased imports (durables tradable) →
    net exports fall in booms → more countercyclical trade balance.
- Impulse responses:
  - Permanent technology shock: financial friction has little effect (wealth effect dominates).
  - Temporary nondurable TFP shock: financial friction matters — borrowing premium falls only with
    the friction, durable spending and investment increase more, net exports/GDP drop more.
  - Temporary durable TFP shock: financial friction less important because durables are a small
    fraction of total output.
- Interaction plots (Figure 6 summary):
  - As financial frictions increase (η lower), correlation of net exports with GDP becomes
    dominated by size of friction and less sensitive to σ_g.
  - σ_g helps pin down volatilities of nondurable and durable consumption for any η.

### Robustness checks
- Added orthogonal shock to borrowing premium ε_b with variance σ_b^2 (column (5)):
  - Unable to exactly match seven moments; overall quality of fit almost unchanged.
  - Estimated σ_b = 0 (no role for orthogonal shock in matching moments here).
  - Interpreted as evidence that financial frictions in emerging economies need to be correlated with income.
- Alternative borrowing-premium specification decomposing sensitivity to each shock:
  q_t = 1/(1+r^*) + χ[ exp( B̃_{t+1}/Γ_t - B̄ ) - 1 ] + η_g (g_t - μ_g) + η_n z_{n,t} + η_d z_{d,t}.
  - Solved model for different χ: -0.0005, -0.001, -0.0015.
  - Finding: η_g decreases with more negative χ; for benchmark χ = -0.001, indebtedness effect explains
    positive association between shocks to trend and borrowing premium; bias on trend shock importance
    is second order.
  - Qualitative conclusions robust; original parsimonious specification delivers better fit.

### Key quantitative outcomes and model limitations
- No-friction scenario: permanent shock contribution ≈ 32 percent of GDP volatility.
- Durability shut-down scenario: permanent shock contribution ≈ 57 percent.
- Financial-friction scenario (η = -0.0077): permanent shock contribution ≈ 24 percent; net exports-GDP
  correlation ≈ -0.72.
- Model difficulty: matching investment volatility (simulated 3.28 vs data 6.11).
- Orthogonal interest-rate shocks (ε_b) estimated zero in robustness exercise.

### Policy and research implications / conclusions
- Financial frictions in the form of a countercyclical country risk premium are essential to:
  - Explain a strongly countercyclical trade balance in emerging markets.
  - Explain a countercyclical and very volatile borrowing premium.
- Durables accumulation interacts with financial frictions to magnify spending responses; durables
  matter for trade dynamics because durables are largely tradable.
- Shocks to trend are relevant but not the main source of fluctuations in emerging economies.
- Future research directions suggested:
  - Consider exogenous shocks to the borrowing premium (orthogonal interest-rate shocks).
  - Explore other types of financial frictions and external shocks (e.g., shocks to external wealth,
    portfolio rebalancing, terms-of-trade shocks) and their interaction with durables and borrowing
    conditions.
  - Add micro-foundations to the borrowing premium mechanism.

*Source: _wp11133 - 4.  Net exports are more volatile and more countercyclical in emerging markets when com- (IMF working paper).*

### REFERENCES

### _wp11133 - REFERENCES

### References
- Aguiar, Mark, and Gita Gopinath, 2007, “Emerging Market Business Cycles: The Cycle is the Trend,” Journal of Political Economy, Vol. 115, No. 1, pp. 69–102.
- Baxter, Marianne, 1996, “Are Consumer Durables Important for Business Cycles?” Review of Economics and Statistics, Vol. 78, No. 1, pp. 147–155.
- Boz, Emine, Christian Daude, and Ceyhun B. Durdu, 2008, “Emerging Market Business Cycles Revisited: Learning About the Trend,” International Finance Discussion Paper 927, Federal Reserve Board.
- Caballero, Ricardo J., 1993, “Durable Goods: An Explanation for Their Slow Adjustment,” Journal of Political Economy, Vol. 101, No. 2, pp. 351–384.
- De Gregorio, Jose, Pablo E. Guidotti, and Carlos A. Vegh, 1998, “Inflation Stabilisation and the Consumption of Durable Goods,” Economic Journal, Vol. 108, No. 446, pp. 105–131.
- De Resende, Carlos, 2006, “Endogenous Borrowing Constraints and Consumption Volatility in a Small Open Economy,” Working Paper 06-37, Bank of Canada.
- Eaton, Jonathan, and Mark Gersovitz, 1981, “Debt with Potential Repudiation: Theoretical and Empirical Analysis,” Review of Economic Studies, Vol. 48, No. 2, pp. 289–309.
- Engel, Charles, and Jian Wang, 2011, “International Trade in Durable Goods: Understanding Volatility, Cyclicality, and Elasticities,” Journal of International Economics, Vol. 83, No. 1, pp. 37–52.
- Fernández-Villaverde, Jesús, Pablo A. Guerrón-Quintana, Juan Rubio-Ramírez, and Martín Uribe, 2009, “Risk Matters: The Real Effects of Volatility Shocks,” Working Paper 14875, National Bureau of Economic Research.
- Galí, Jordi, 1993, “Variability of Durable and Nondurable Consumption: Evidence for Six O.E.C.D. Countries,” Review of Economics and Statistics, Vol. 75, No. 3, pp. 418–428.
- García-Cicco, Javier, Roberto Pancrazi, and Martín Uribe, 2010, “Real Business Cycles in Emerging Countries?” American Economic Review, Vol. 100, No. 5, pp. 2510–31.
- Gomes, João F., Leonid Kogan, and Motohiro Yogo, 2009, “Durability of Output and Expected Stock Returns,” Journal of Political Economy, Vol. 117, No. 5, pp. 941–986.
- Gruss, Bertrand, and Karel Mertens, 2009, “Regime Switching Interest Rates and Fluctuations in Emerging Markets,” Economics Working Paper ECO2009/22, European University Institute.
- Hornstein, Andreas, and Jack Praschnik, 1997, “Intermediate Inputs and Sectoral Comovement in the Business Cycle,” Journal of Monetary Economics, Vol. 40, No. 3, pp. 573–595.
- Kehoe, Patrick J., and Timothy J. Kehoe, 1994, “A Primer on Static Applied General Equilibrium Models,” Federal Reserve Bank of Minneapolis Quarterly Review, Vol. 18, No. 1, pp. 2–16.
- Kouparitsas, Michael, 1998, “Dynamic Trade Liberalization Analysis: Steady State, Transitional and Inter-Industry Effects,” Working Paper WP-98-15, Federal Reserve Bank of Chicago.
- Mendoza, Enrique G., 1991, “Real Business Cycles in a Small Open Economy,” American Economic Review, Vol. 81, No. 4, pp. 797–818.
- Neumeyer, Pablo A., and Fabrizio Perri, 2005, “Business Cycles in Emerging Economies: The Role of Interest Rates,” Journal of Monetary Economics, Vol. 52, No. 2, pp. 345–380.
- Ogaki, Masao, and Carmen M. Reinhart, 1998, “Measuring Intertemporal Substitution: The Role of Durable Goods,” Journal of Political Economy, Vol. 106, No. 5, pp. 1078–1098.
- Restrepo-Echevarria, Paulina, 2008, “Business Cycles in Developing vs. Developed Countries: The Importance of the Informal Economy,” Working paper, UCLA.
- Rosenzweig, Mark R., and Kenneth I. Wolpin, 1993, “Credit Market Constraints, Consumption Smoothing, and the Accumulation of Durable Production Assets in Low-Income Countries: Investments in Bullocks in India,” Journal of Political Economy, Vol. 101, No. 2, pp. 223–244.
- Schmitt-Grohé, Stephanie, and Martín Uribe, 2003, “Closing Small Open Economy Models,” Journal of International Economics, Vol. 61, No. 1, pp. 163–185.
- Uribe, Martín, and Vivian Z. Yue, 2006, “Country Spreads and Emerging Countries: Who Drives Whom?” Journal of International Economics, Vol. 69, No. 1, pp. 6–36.

### Figures (captions)
- Figure 1. Relative Volatility of Consumption and Saving Rates Across Countries
- Figure 2. Effect of Shocks to Trend on Key Moments
- Figure 3. Impulse-Response Functions for a Permanent Shock
  - Note: The dotted line refers to the trend of the variable under consideration.
- Figure 4. Impulse-Response Functions for a Shock to Nondurables
- Figure 5. Impulse-Response Functions for a Shock to Durables
- Figure 6. Effect of Financial Frictions on Key Moments

### Tables and selected numeric highlights
- Table 1. Volatility of Output and Relative Volatility of Consumption, Investment and Net Exports
  - Macroeconomic volatility measured by standard deviation (σ) of GDP (y), total consumption expenditure (c), consumption of nondurable goods (cn), expenditure in durable goods (cd), investment (i), and net exports as a fraction of GDP (nx).
  - Example country entries (σy, σc, σcn, σcd, σi, σnx):
    - Chile 96:I-08:I — 2.38, 3.39, 1.43, 2.86, 1.20, 1.05, 4.64, 1.28, 2.78 (table rows preserve column ordering as in source).
    - United States 47:I-08:IV — 1.65, 1.26, 0.76, 0.79, 0.48, 5.13, 3.11, 4.77, 0.36.
  - All data is quarterly except for Colombia, for which it is annual. All variables are in logs (except nx) and detrended using the HP filter.

- Table 2. Correlations with Output
  - Variables include ρy,c, ρy,cn, ρy,cd, ρy,i, ρy,nx, ρy,y−1, ρ∆y,∆y−1.
  - Example entries:
    - Mexico 93:I-07:IV — ρy,c 0.94, ρy,cn 0.92, ρy,cd 0.92, ρy,i 0.92, ρy,nx -0.82, ρy,y−1 0.81, ρ∆y,∆y−1 0.25.
    - United States 47:I-08:IV — ρy,c 0.76, ρy,cn 0.78, ρy,cd 0.59, ρy,i 0.83, ρy,nx -0.28, ρy,y−1 0.84, ρ∆y,∆y−1 0.34.

- Table 3. Average Relative Volatility of Consumption
  - Weighted Average of Ratio to σy:
    - Total Consumption — Large Economies 0.94, Small Economies 0.99, Emerging Economies 1.30.
    - Non-durables — Large Economies 0.68, Small Economies 0.72, Emerging Economies 0.90.
    - Durables — Large Economies 3.83, Small Economies 3.85, Emerging Economies 4.48.

- Table 4. Benchmark Parameter Values
  - Time discount factor β 0.98
  - Consumption utility share θ 0.413
  - Exponent in the consumption aggregator function γ 0.1628
  - Risk aversion σ 2
  - Utility share of nondurables μ 0.881
  - Capital income share in nondurables sector αn 0.52
  - Capital income share in durables sector αd 0.32
  - Depreciation rate of capital stock δk 0.01775
  - Depreciation rate of stock of durables δd 0.039
  - Amount of nondurable good needed to produce one unit of durable good Ω 0.3
  - Aggregate productivity’s long-run mean growth rate μg 1.006
  - Autocorrelation of shock to trend ρg 0.01
  - Autocorrelation of shock to nondurable goods ρn 0.95
  - Autocorrelation of shock to durable goods ρd 0.95
  - Capital adjustment cost φ 4
  - Debt coefficient on interest rate premium χ -0.001
  - Steady-state normalized debt B̄/Ȳ -0.1

- Table 5. Simulated moments and parameter estimates for Mexico
  - Simulated standard deviations and correlations for GDP (y), ∆y, nondurables production (yn), durables production (yd), consumption measures, investment (i), durable expenditure (cd), borrowing premium (bp), net exports-output ratio (nx).
  - Example simulated values (Data and selected columns):
    - σy Data 2.35; column (1) 2.35; column (5) 2.3500; column (6) 2.4056.
    - σ∆y Data 1.50; column (1) 1.76; column (6) 1.8012.
    - σcd Data 4.20; column (1) 4.20; column (3) 0.9083; column (6) 4.2399.
    - ρc,y Data 0.92; column (1) 0.9237; column (6) 0.9374.
    - σg 1.8340 (Data) and simulated values such as 2.4776, 1.9107, 1.5795, 1.5410 across columns.
    - Quality of Fit entries: Data 0.8363; column (1) 0.7160; column (4) 0.8969; column (5) 0.8939.
  - Parameters and estimates listed in the table include ψ 0.5336, ρnd 0.3631, η000 -0.0077 -0.0059, δd 0.039, wn 0.2779, wd 0.4266, Share of εg in GDP variance decomposition (%) 0.3300 and simulated counterparts.

- Table 6. Simulated moments and parameter estimates for alternative specifications of the borrowing premium
  - Columns for χ = −.0005, χ = −.001, χ = −.0015.
  - Selected simulated outcomes:
    - σy 2.3198, 2.3436, 2.3630.
    - σ∆y 1.7386, 1.7501, 1.7588.
    - σcd 4.1305, 4.1851, 4.2304.
    - ρnx,y -0.5402, -0.5918, -0.5975.
    - σg 1.4106, 1.5963, 1.7643.
    - Share of εg in GDP variance decomposition 0.0976, 0.1317, 0.1667.

*Content unit: _wp11133 - REFERENCES (PDF).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11133.pdf_