## _wp07176 - References

## Source details

**Canonical URL:** [_wp07176 - References](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp07176.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp07176.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp07176.pdf.json)

---

### Introduction and research question
- Latin America experienced average growth of over 5 percent from 2004 to 2006, described as the most vigorous three-year expansion since the late 1970s.
- Expansion occurred with low and falling inflation, rapidly growing imports and public spending, declining public debt, and record primary fiscal balances and external current account surpluses in 2006.
- Central research questions:
  - Can Latin America’s current growth be expected to continue if external conditions deteriorate?
  - What impact would external shocks—both real and financial—have on Latin America’s growth performance?

### Methodology: mean-adjusted Bayesian VAR with informative steady-state priors
- Model and estimation:
  - Bayesian Vector-Autoregressive (BVAR) with “informative priors” on steady state values following Villani (2005).
  - Key notation and priors preserved from source: Σ prior ∝ (Σ)−(n+p+1)/2, Gvec ~ N(G, θGΩ), ψ ~ N(ψ, θψΩ).
  - Mean-adjusted and standard forms represented as in source: (1) tt LημxG+=() and (2) tt LηψxG=()().
  - Gibbs sampler with the number of draws set to 10,000.
- Empirical implementation and data:
  - External factors proxied by world GDP growth, a trade-weighted commodity price index relevant for Latin America, U.S. three-month treasury bill rates, and the U.S. high-yield corporate bond spread.
  - Alternative specification replaces world growth with U.S. growth and U.S. inflation.
  - Latin American growth measure: weighted index for Argentina, Brazil, Chile, Colombia, Mexico and Peru (LA6), covering almost 90 percent of Latin American output.
  - Additional variable: Latin America subcomponent of JP Morgan’s emerging market bond index (EMBI).
  - Model forms (notation preserved): equations (3) and (4) as in source.
  - Data frequency and sample: quarterly, from 1994Q2 to 2006Q4.
  - Variable treatment: world (or US) GDP, commodity prices, and LA6 GDP modeled in first differences; remaining variables modeled in levels (ADF and KPSS unit root testing referenced).
  - Priors on dynamic coefficients: modified “Minnesota priors” with prior means on first own lag for variables in first differences set to zero, and for variables in levels set to 0.9.
- Steady-state priors and rationale:
  - World and U.S. GDP growth priors based on IMF’s World Economic Outlook medium-term projections.
  - U.S. T-Bill rate and CPI inflation priors combine a 2 percent inflation target and an equilibrium real interest rate of ~2 percent.
  - LA6 growth prior centered on 4.25 percent based on econometric studies of reforms and long-run growth.
  - U.S. high-yield bond and EMBI spreads assigned relatively wide distributions reflecting historical behavior.
  - Commodity prices assumed as random walk with small drift; steady-state growth rate centered on 1 percent.

### Key empirical findings: impulse responses and variance decompositions
- External factors explain more than half of the variance of the growth rate of an aggregate Latin American output index at standard medium-term horizons (depending on the model, between 50 and 60 percent).
  - Financing shocks are the most important external component, explaining over half of the contribution of external shocks.
- One-standard-deviation shock magnitudes used in the model:
  - world growth: about 0.28 percentage point
  - U.S. T-bill rates: 27 basis points
  - U.S. high-yield bond spread: 90 basis points
  - commodity prices: 5 percent
  - Latin EMBI: 115 basis points
- Impulse-response “rules of thumb” and specific effects (model averages):
  - Overall impact of a world or U.S. growth shock on Latin America is roughly one-for-one over time.
  - A 0.3 percent world growth shock leads to an increase in four-quarter Latin American growth by about 0.4 percentage point after four quarters.
  - U.S. growth shocks (standard deviation ≈ 0.5 percentage points) result in an increase of four-quarter Latin American growth by about 0.6 percentage point.
  - A standard deviation (90-basis-point) rise in the U.S. high-yield bond spread leads to a decline of four-quarter growth in Latin America by about 0.9 percentage point after three quarters.
  - A standard deviation commodity shock (change of almost 5 percent in a quarter) leads to a change in four-quarter Latin American growth of about 1/3 percentage point after two quarters.
  - A 115-basis point rise in the Latin EMBI is associated with a drop in four-quarter growth by 0.5 percent after four quarters.
- Dynamics of identified shocks (Cholesky ordering as in equations (3) and (4)):
  - World growth shock raises U.S. short-term rates over two to three quarters and lowers high-yield spreads; when U.S. inflation is included it also raises inflation—interpreted as aggregate demand shocks.
  - U.S. interest rate hike reduces world/U.S. growth after four to six quarters and immediately increases the high-yield bond spread.
  - High-yield bond spread shock causes a dip in world growth, gradual easing of U.S. interest rates, and a sharp immediate jump in the Latin EMBI of about the same magnitude as the high-yield shock.
- Variance decomposition (medium term, 10–20 quarter horizon):
  - Model using world GDP growth:
    - Approximately 12 percent by world growth shocks.
    - 6 percent by commodity prices.
    - 34 percent by U.S. financial conditions (combined U.S. short-term interest rates and U.S. high-yield bond spread).
  - Alternative model (with U.S. growth and inflation): external influence rises to about 57 percent:
    - 16 percent corresponds to U.S. growth.
    - 5 percent to commodity prices.
    - U.S. financial conditions account for 27 percent of the variance (rising to 36 percent if the contribution of U.S. inflation is included in this category).

### Forecasting performance and conditional forecasts
- Forecast generation: 10,000 Gibbs sampling iterations; central forecast is the median forecast across the forecast density at each horizon.
- Out-of-sample performance:
  - Mean-adjusted BVAR generally outperforms the classical VAR (relative RMSE almost always smaller than unity).
  - Relative to the traditional BVAR, the mean-adjusted BVAR produces smaller RMSEs at all horizons for three out of six variables; for world growth, LA6 growth, and commodities performance depends on horizon.
- Comparison with IMF WEO forecasts (annual growth rates, two-year horizon):
  - One year horizon RMSE: WEO = 1.1 percentage points; model = 1.4 percentage points.
  - Two year horizon RMSE: WEO = 2.6; model = 2.1.
- Baseline projections for LA6 growth (model with world growth, conditional and unconditional virtually same):
  - 4.6 for 2007, and 4.2 for 2008.
  - Spring 2007 WEO projections: 4.8 and 4.3 percent, respectively.
  - Model based on U.S. growth only: 3.8 percent for 2007 and 3.5 for 2008.

### Conditional scenarios and quantitative impacts
- Baseline conditional forecast: world growth and commodity paths from April 2007 WEO; flat U.S. high yield bond spread and Latin EMBI for two quarters after end of sample. Baseline LA6 annual averages: 4.6 for 2007 and 4.2 for 2008.
- Moderate external downturns:
  - If slowdown originates in U.S., impact on Latin America larger than via world growth: around 0.5 lower average growth relative to baseline in 2007 and about 0.7 lower growth in 2008 (model including U.S. growth and inflation).
- Disruptive unwinding of global imbalances (example scenario: 400 basis point jump in U.S. high-yield spread and U.S. growth declines to 1 percent for two years):
  - Model featuring world growth: Latin American four-quarter growth would fall to less than 1 percent by early 2008; average annual growth about 2¾ percent in 2007 and 1½ percent in 2008.
  - Model featuring U.S. growth: four-quarter growth turns sharply negative after about 4 quarters; average annual growth about one percent in 2007 and about zero in 2008.
- Declines in nonfuel commodity prices (holding baseline world growth):
  - A 20 percent fall to early 2007 levels: predicted Latin American growth about 3½–3¾ percent on average in 2007 and 2008.
  - A 50 percent decline: four-quarter growth would fall to around ¾ percent by mid-2008; average growth would decline to 3 percent in 2007 and about 1¼ percent in 2008.
  - Large commodity shocks also affect financing conditions via EMBI spreads, trade balances and fiscal primary surpluses.
- Emerging market financing shock (400-basis-point shock to Latin EMBI and 200-basis-point increase in U.S. high-yield spread, without industrial country growth impact):
  - Model predicts growth to drop to just under 3 percent in 2007 and to about 2¾ percent in 2008.
  - Authors note this likely overestimates impact given improved macro fundamentals relative to 1994–2006 sample period.

### Heterogeneity across countries
- Contribution of external factors to 20-quarter horizon variance (model with world growth):
  - Around 40 percent for Brazil, Mexico and Peru.
  - Around 50 percent for Chile and Colombia.
  - Over 60 percent for Argentina.
- Alternative model (U.S. growth and inflation): influence of external factors approximately 50 percent for Mexico and Brazil; just over 40 percent for Peru; similar to world-growth model for Argentina, Chile, Colombia.
- Country-specific impulse-response highlights:
  - U.S. growth shocks larger and tighter for Mexico (and to a lesser extent Argentina) than world growth shocks; Mexican GDP response to a 0.5 point U.S. growth shock: 0.3 on impact rising to almost 1.0.
  - Argentina shows the greatest contractionary response to U.S. high-yield bond spread and Latin EMBI (reflecting 1998–2001 episodes); Brazil, Colombia, Mexico, Peru show contractionary responses but more muted.
  - Chile shows Latin EMBI shocks associated with an expansionary response.
  - Commodity shocks elicit largest output responses in Argentina, followed by Peru and Chile; Chile’s response peaks after about 5 quarters, while Argentina and Peru react more immediately.

### Policy-relevant implications and recommendations
- Macro context:
  - Improved macroeconomic policy frameworks in Latin America help support growth, but the external environment has been unusually favorable (high world growth, ample private financing, historically low emerging market risk premia, and high commodity prices), contributing materially to the recovery.
- Key monitoring priorities:
  - Monitor external financing conditions closely because financing shocks are the dominant external source of growth volatility in the model.
- Robustness and limits:
  - Latin America is fairly robust to moderate external shocks (e.g., moderate U.S. slowdown, 20 percent commodity price reduction).
  - Extreme combined shocks (world growth slowdown plus much higher financing premia) would have materially larger adverse effects, though considered unlikely based on the model and information at the time.
  - Results reflect average behavior over 1994–2006; structural changes since then (notably reduction in currency mismatches) may have reduced vulnerability to external financing shocks, so model may overstate current vulnerability.
- Policy recommendations to reduce regional vulnerabilities:
  - Lower public debt.
  - Make budgets more flexible.
  - Strengthen financial systems.
  - Diversify export structures.
  - Reduce fiscal dependence on commodity revenues.

*Source: _wp07176 (IMF working paper content summarized from the referenced chapter).*

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

### _wp07176 - References.......................................................................................................................

### Introduction and research question
- Latin America experienced average growth of over 5 percent from 2004 to 2006, described as the most vigorous three-year expansion since the late 1970s.
- The expansion occurred with low and falling inflation, rapidly growing imports and public spending, declining public debt, and record primary fiscal balances and external current account surpluses in 2006.
- Main question: Can Latin America’s current growth be expected to continue if external conditions deteriorate? What impact would external shocks—both real and financial—have on Latin America’s growth performance?

### Methodology: mean-adjusted Bayesian VAR with informative steady-state priors
- Model: Bayesian Vector-Autoregressive (BVAR) with “informative priors” on steady state values, following Villani (2005).
- Key equations and parameterization:
  - Standard form: (1) tt LημxG+=()
  - Mean-adjusted form: (2) tt LηψxG=()()
  - Priors: Σ prior ∝ (Σ)−(n+p+1)/2, Gvec ~ N(G, θGΩ), ψ ~ N(ψ, θψΩ) (notation preserved as in source).
  - Numerical evaluation: Gibbs sampler with the number of draws set to 10,000.
- Empirical implementation:
  - External factors proxied by world GDP growth, a trade-weighted commodity price index relevant for Latin America, U.S. three-month treasury bill rates, and the U.S. high-yield corporate bond spread.
  - Alternative specification replaces world growth with U.S. growth and U.S. inflation.
  - Latin American growth measure: weighted index for Argentina, Brazil, Chile, Colombia, Mexico and Peru (LA6), covering almost 90 percent of Latin American output.
  - Additional included variable: Latin America subcomponent of JP Morgan’s emerging market bond index (EMBI).
  - Model forms:
    - equation (3): () worldUS ttt tttt yiHYycEMBI ′ = ΔΔ Δx
    - equation (4): () USUSUS tttt tttt yiHYycEMBIπ ′ = ΔΔ Δx
  - Data frequency and sample: quarterly, from 1994Q2 to 2006Q4.
  - Variable treatment: world (or US) GDP, commodity prices, and LA6 GDP modeled in first differences; remaining variables modeled in levels (unit root testing referenced: ADF and KPSS).
  - Priors on dynamic coefficients: slightly modified “Minnesota priors” with prior means on first own lag for variables in first differences set to zero, and for variables in levels set to 0.9.
  - Steady-state priors rationale:
    - World and U.S. GDP growth priors based on IMF’s World Economic Outlook medium-term projections.
    - U.S. T-Bill rate and CPI inflation priors follow conventional choices combining a 2 percent inflation target and an equilibrium real interest rate of ~2 percent.
    - LA6 growth prior centered on 4.25 percent based on econometric studies of reforms and long-run growth.
    - U.S. high-yield bond and EMBI spreads assigned relatively wide distributions reflecting historical behavior.
    - Commodity prices assumed as random walk with small drift; steady-state growth rate centered on 1 percent.

### Key empirical findings (impulse responses, variance decompositions, and forecasting)
- External shocks explain more than half of the variance of the growth rate of an aggregate Latin American output index at standard medium-term horizons (depending on the model, the number varies between 50 and 60 percent).
  - Of the external shocks, financing shocks are the most important, explaining over half of the contribution of external shocks.
- Impulse response “rules of thumb”:
  - The overall impact of a world or U.S. growth shock on Latin America is roughly one-for-one over time.
  - One-standard-deviation shock magnitudes (as used in the model):
    - world growth: about 0.28 percentage point
    - U.S. T-bill rates: 27 basis points
    - U.S. high-yield bond spread: 90 basis points
    - commodity prices: 5 percent
    - Latin EMBI: 115 basis points
  - Estimated impacts on Latin American growth from one-standard-deviation shocks (model averages reported):
    - commodity prices and Latin EMBI (changes of about 5 percent and 115 basis points, respectively, within one quarter): change in Latin American growth of around 0.4 to 0.5 percentage point.
    - U.S. high-yield bond spread (standard deviation 90 basis points): estimated effect of around 0.9 percentage point on Latin American growth.
- Dynamics of identified shocks (Cholesky ordering as in equations (3) and (4)):
  - A world growth shock raises U.S. short-term interest rates over two to three quarters and lowers high-yield bond spreads; in the version including U.S. inflation it also raises inflation—interpreted as aggregate demand shocks.
  - A U.S. interest rate hike reduces world/U.S. growth after four to six quarters and immediately increases the high-yield bond spread.
  - A high-yield bond spread shock causes a dip in world growth, gradual easing of U.S. interest rates, and a sharp immediate jump in the Latin EMBI of about the same magnitude as the high-yield shock.
- Forecasting performance:
  - The mean-adjusted BVAR model outperforms both a classical VAR and a conventional BVAR in out-of-sample forecasting at most horizons.
- Conditional forecast exercises and scenarios:
  - Conditional forecasts suggest Latin American growth would be fairly resilient to plausible external risks consistent with IMF’s April 2007 World Economic Outlook, including:
    - a moderate slowing of U.S. and world growth, and
    - a decline in commodity export prices of around 20 percent.
  - Even with such moderate shocks, Latin America’s external environment would remain relatively favorable.
  - A significant slowing of Latin American growth would require a stronger set of shocks—such as a combined world growth slowdown and much higher financing premia—which appear unlikely based on the model and current information.
- Caveat on interpretation:
  - Results reflect average behavior over the 1994–2006 sample period.
  - Structural changes in many Latin American economies since then—most notably a large reduction in currency mismatches—may have reduced vulnerability to external financing shocks; consequently, the model may overstate current vulnerability to external shocks.

### Tables and figures referenced (content preserved from source)
- Tables:
  - Table 1. Steady State Prior and Posterior Distributions (prior and posterior probability intervals shown for: World growth, U.S. growth, U.S. inflation, U.S. T-Bill rate, U.S. HY spread, Commodity prices, LA6 growth, Latin EMBI spread).
    - Examples of prior/posterior intervals shown in source (probability intervals refer to a normal distribution):
      - World growth prior (3.75, 4.75) / posterior (3.4, 4.1)
      - U.S. growth prior (2.0, 4.0) / posterior (2.8, 3.9)
      - U.S. inflation prior (1.0, 3.0) / posterior (2.0, 2.8)
      - U.S. T-Bill rate prior (3.0, 5.0) / posterior (3.5, 5.0)
      - U.S. HY spread prior (3.0, 6.0) / posterior (3.8, 5.9)
      - Commodity prices prior (-2.0, 4.0) / posterior (-1.4, 4.4)
      - LA6 growth prior (3.5, 5.0) / posterior (3.4, 4.8)
      - Latin EMBI spread prior (2.0, 5.0) / posterior (2.1, 4.9)
- Figures:
  - Figure 1: Data plots for World Growth, U.S. T-bill, U.S. HY Bond Spread, LA6 GDP Growth, Commodities Index, Latin EMBI (growth rates given as percentage changes with respect to the same quarter in the preceding year).
  - Figure 2: Impulse Response Functions from Mean-Adjusted Bayesian VAR, 1994Q2–2006Q4 (horizons one to twenty quarters; colored bands are 50 and 90 percent confidence bands).
  - Additional figures and tables listed in the source include variance decompositions, forecasting performance comparisons, unconditional vs. conditional forecasts, scenarios of external slowdown and combined shocks, commodity index, effects of declines in commodity prices, and effects of emerging market financing shocks.

### Policy-relevant implications (as presented)
- Improved macroeconomic policy frameworks in Latin America help support growth, but external environment has been unusually favorable—high world growth, ample private financing, historically low emerging market risk premia, and high commodity prices—contributing materially to the recovery.
- Monitoring external financing conditions is crucial because financing shocks are the dominant external source of growth volatility in the model.
- While moderate external shocks (e.g., moderate world/U.S. slowdown or a ~20 percent decline in commodity export prices) are unlikely to derail growth prospects, policymakers should be mindful of the possibility of larger combined shocks (world growth slowdown plus much higher financing premia) which would have materially larger adverse effects.

*Source: _wp07176 - References (PDF), estimations and results summarized from the original chapter content.*

### 0.3 percent world growth shock leads to an increase in (four-quarter) Latin American

### _wp07176 - 0.3 percent world growth shock leads to an increase in (four-quarter) Latin American

### Key impulse response findings
- A 0.3 percent world growth shock leads to an increase in (four-quarter) Latin American growth by about 0.4 percentage point after four quarters.  
- Latin American growth reaction to U.S. interest rates is muted: the sign is as expected after about three to four quarters (a hike reduces growth), but the effect is small and barely statistically significant.  
- A standard deviation (90-basis-point) rise in the U.S. high yield bond spread leads to a decline of four-quarter growth in Latin America by about 0.9 percentage point after three quarters.  
  - The U.S. high yield bond spread also appears to have strong effects on the Latin EMBI and on world growth (or more strongly on U.S. growth), providing additional transmission channels.  
- A standard deviation commodity shock—here a change of almost 5 percent in a quarter—leads to a change in four-quarter Latin American growth of about ⅓ percentage point after two quarters.  
- A 115-basis point rise in the Latin EMBI is associated with a drop in four-quarter growth by 0.5 percent after four quarters.  
- Alternative model (includes U.S. growth and inflation instead of world growth):  
  - Similar effects for commodities and EMBI shocks.  
  - Reaction to a shock to the U.S. high yield bond spread is slightly more muted: LA6 growth decreases by about 0.8 percentage point after three quarters.  
- U.S. growth shocks (standard deviation of around 0.5 percentage points) lead to a faster and slightly larger reaction than world growth shocks: a standard deviation U.S. growth shock results in an increase of four-quarter Latin American growth by about 0.6 points.  
  - Since the U.S. growth shock leads to a total increase in U.S. growth of about 0.55 percentage point after three quarters, the overall average reaction of Latin American growth to U.S. growth is about one-for-one.

### Variance decomposition (medium term, 10–20 quarter horizon)
- More than half of the medium-term variance of Latin American GDP growth is explained by external factors. Breakdown (model using world GDP growth):  
  - Approximately 12 percent by world growth shocks.  
  - 6 percent by commodity prices.  
  - 34 percent by U.S. financial conditions (combined influence of U.S. short-term interest rates and the U.S. high yield bond spread).  
- Alternative model (with U.S. growth and inflation): external influence rises to about 57 percent:  
  - 16 percent corresponds to U.S. growth.  
  - 5 percent to commodity prices.  
  - U.S. financial conditions account for 27 percent of the variance (rising to 36 percent if the contribution of U.S. inflation is included in this category).

### Out-of-sample forecasting performance
- Forecast generation: 10,000 Gibbs sampling iterations; central forecast is the median forecast across the forecast density at each horizon.  
- Forecast comparison (mean-adjusted BVAR vs traditional BVAR and classical VAR):  
  - The mean-adjusted BVAR generally outperforms the classical VAR (relative RMSE almost always smaller than unity).  
  - Compared to the traditional BVAR, the mean-adjusted BVAR produces smaller RMSEs at all horizons for three out of six variables; for world growth, LA6 growth, and commodities the relative performance depends on horizon (short end similar, traditional BVAR somewhat better at medium horizons 4–8 quarters, mean-adjusted BVAR better at longer horizons).  
- Comparison with IMF WEO forecasts (annual growth rates, two-year horizon):  
  - One year horizon RMSE: WEO = 1.1 percentage points; model = 1.4 percentage points (WEO better at one year).  
  - Two year horizon RMSE: WEO = 2.6; model = 2.1 (model better at two years).  
- Baseline projections for LA6 growth (model with world growth, conditional and unconditional virtually same):  
  - 4.6 for 2007, and 4.2 for 2008.  
  - Spring 2007 WEO projections: 4.8 and 4.3 percent, respectively.  
  - Model based on U.S. growth only: 3.8 percent for 2007 and 3.5 for 2008.

### Conditional forecasts and scenario analysis (selected scenarios and quantitative impacts)
- Baseline conditional forecast: world growth path and commodity path from April 2007 WEO; flat U.S. high yield bond spread and Latin EMBI for two quarters after end of sample. Baseline LA6 annual averages: 4.6 for 2007 and 4.2 for 2008.  
- Moderate downturn in external growth:  
  - If slowdown originates in U.S., impact on Latin America larger than via world growth: around 0.5 lower average growth relative to baseline in 2007 and about 0.7 lower growth in 2008 (model including U.S. growth and inflation).  
- Disruptive unwinding of global imbalances (captured by a 400 basis point jump in the U.S. high yield bond spread and much lower external growth; assumed U.S. growth declines to 1 percent for two years):  
  - Model featuring world growth: Latin American four-quarter growth would fall to less than 1 percent by early 2008; average annual growth about 2¾ percent in 2007 and 1½ percent in 2008.  
  - Model featuring U.S. growth: four-quarter growth turns sharply negative after about 4 quarters; average annual growth about one percent in 2007 and about zero in 2008.  
- Declines in nonfuel commodity prices (holding baseline world growth):  
  - A 20 percent fall to early 2007 levels: predicted Latin American growth about 3½–3¾ percent on average in 2007 and 2008.  
  - A 50 percent decline: four-quarter growth would fall to around ¾ percent by mid-2008; average growth would decline to 3 percent in 2007 and about 1¼ percent in 2008.  
  - Large commodity shocks also impact financing conditions via EMBI spreads, affect trade balances and fiscal primary surpluses.  
- Emerging market financing shock (400-basis-point shock to Latin EMBI and 200-basis-point increase in U.S. high-yield spread, without industrial country growth impact):  
  - Model predicts growth to drop to just under 3 percent in 2007 and to about 2¾ percent in 2008.  
  - Authors note this likely overestimates impact given improved macro fundamentals relative to 1994–2006 sample period.

### Heterogeneity across individual countries (country-level findings)
- Contribution of external factors to 20-quarter horizon variance (model with world growth): ranges from around 40 percent for Brazil, Mexico and Peru; around 50 percent for Chile and Colombia; over 60 percent for Argentina.  
- In alternative model (U.S. growth and inflation): influence of external factors approximately 50 percent for Mexico and Brazil; touch over 40 percent for Peru; similar to world-growth model for Argentina, Chile, Colombia.  
- Notable country-specific impulse responses:  
  - U.S. growth shocks larger and tighter for Mexico (and to a lesser extent Argentina) than world growth shocks; Mexican GDP response to a 0.5 point U.S. growth shock: 0.3 on impact rising to almost 1.0.  
  - Argentina shows the greatest contractionary response to U.S. high yield bond spread and Latin EMBI (reflecting 1998–2001 episodes); Brazil, Colombia, Mexico, Peru show contractionary responses but more muted. Chile shows Latin EMBI shocks associated with an expansionary response.  
  - Commodities shocks solicit largest output responses in Argentina, followed by Peru and Chile; Chile’s response peaks after about 5 quarters, while Argentina and Peru react more immediately.

### Conclusions and policy implications
- The mean-adjusted BVAR model outperforms a classical VAR and a conventional BVAR as a forecasting tool, and outperforms WEO forecasts at the 2 year/next year horizon (but not at the 1 year/same year horizon).  
- Latin America is fairly robust to moderate external shocks (e.g., moderate U.S. slowdown, 20 percent commodity price reduction).  
- Extreme scenarios (e.g., sudden unwinding of global imbalances with much lower U.S./world growth and much higher premia) would have very large impacts, though these scenarios are considered unlikely at the time of analysis.  
- Financial shocks account for a large share of external-factor influence (more than 60 percent of the contribution of external factors to the variance of growth in the model’s estimates), and financial channels magnify real shocks such as commodity price declines. Commodity prices remain an important determinant of short-term fluctuations.  
- Policy recommendations to reduce regional vulnerabilities:  
  - Lower public debt.  
  - Make budgets more flexible.  
  - Strengthen financial systems.  
  - Diversify export structures.  
  - Reduce fiscal dependence on commodity revenues.

*Source: _wp07176 - 0.3 percent world growth shock leads to an increase in (four-quarter) Latin American (IMF working paper content provided).*

### REFERENCES

### REFERENCES

### Forecasting, Judgment, and Policy Rules
- Adolfson, Malin, Michael K. Andersson, Jesper Lindé, Mattias Villani, and Anders Vredin 2005, “Modern Forecasting Models in Action: Improving Macro Economic Analyses at Central Banks,” Working Paper No. 190, Sveriges Riksbank.
- Clarida, Richard, Jordi Gali, and Mark Gertler, 1998, “Monetary Policy Rules in Practice. Some International Evidence,” European Economic Review Vol. 42, pp. 1033–1067.
- Croushore, Dean and Tom Stark, 2002, “Forecasting with a Real-Time Data Set for Macroeconomists,” Journal of Macroeconomics Vol. 24, pp. 473–484.
- Lawrence, M. J., R. H. Edmundson, and M. J. O’Connor, 1986, “The Accuracy of Combining Judgmental and Statistical Forecasts,” Management Science, Vol. 32, pp. 1521–1532.
- McNees, Stephen K., 1990, “The Role of Judgment in Macroeconomic Forecasting Accuracy,” International Journal of Forecasting, Vol. 6, pp. 287-299.
- Österholm, Pär, 2006, “Incorporating Judgement in Fan Charts,” Finance and Economics Discussion Series 2006–39, Board of Governors of the Federal Reserve System.
- Svensson, Lars E. O., 2005, “Monetary Policy with Judgement: Forecast Targeting,” International Journal of Central Banking, Vol. 1, pp. 1–54.
- Taylor, John B., 1993, “Discretion versus Policy Rules in Practice,” Carnegie-Rochester Conference Series on Public Policy, Vol. 39, pp.195–214.
- Leeper, Eric M. and Tao Zha, 2003, “Modest Policy Interventions,” Journal of Monetary Economics, Vol. 50, 1673–1700.
- Sims, Chris A., 1982, “Policy Analysis with Econometric Models,” Brookings Papers on Economic Activity, 1982:1, pp. 107–164.

### Latin America, External Shocks, and Regional Outlooks
- Calvo, Guillermo and Ernesto Talvi, 2007, “Current Account Surplus in Latin America: Recipe Against Capital Market Crises?” May, Available via the internet: www.rgemonitor.com/latam-blog
- Canova, Fabio, 2005, “The Transmission of US Shocks to Latin America,” Journal of Applied Econometrics Vol. 20, pp. 229–25
- Cuevas, Alfredo C., Miguel Messmacher, Alejandro M. Werner, 2003, “Sincronizacion Macroeconomica entre Mexico y sus Socios Comerciales del TLCAN,” Banco de México Working Paper, No. 2003-01.
- Gaytan González, A. and J. R. Gonzalez-Garcia, 2006, “Structural Changes in the Transmission Mechanism of Monetary Policy in Mexico: A Non-Linear VAR Approach,” Working Paper 2006-06, Banco de México.
- Izquierdo, Alejandro, Randall Romero, and Ernesto Talvi, 2007, “Business Cycles in Latin America: the Role of External Factors,” forthcoming IADB Working Paper (Washington: Inter-American Development Bank).
- Loayza, Norman, Paulo Fajnzylber, and Cesar Calderón, 2004, “Economic Growth in Latin America and the Caribbean: Stylized Facts, Explanations and Forecasts,” Working Papers of Central Bank of Chile 265.
- Roache, Shaun, 2007, “Economic Linkages Across the Western Hemisphere: A Survey,” (unpublished draft; Washington: International Monetary Fund).
- Talvi, Ernesto, 2006, “Risks for Latin America Given Main Scenarios for the World Economy,” presented at the XXIV Meetings of the Latin American Network of Central Banks and Finance Ministries, Inter-American Development Bank, Washington, DC, October.
- Talvi, Ernesto, 2007, “Beyond Perceptions: Macroeconomics in LAC,” presented at the XXV Meetings of the Latin American Network of Central Banks and Finance Ministries, Inter-American Development Bank, Washington, DC, May.
- Zettelmeyer, Jeromin, 2006, “Growth and Reforms in Latin America: A Survey of Facts and Arguments,” IMF Working Paper 06/210 (Washington: International Monetary Fund).

### Trade Agreements and Regional Integration
- Kose, M. Ayhan, and Alessandro Rebucci, 2005, “How Might CAFTA Change Macroeconomic Fluctuations in Central America? Lessons from NAFTA,” Journal of Asian Economics Vol. 16, pp. 77–104.
- Cuevas, Alfredo C., Miguel Messmacher, Alejandro M. Werner, 2003, “Sincronizacion Macroeconomica entre Mexico y sus Socios Comerciales del TLCAN,” Banco de México Working Paper, No. 2003-01.

### Bayesian Methods, VARs, and Multivariate Modeling
- Litterman, Robert B.,1986, “Forecasting with Bayesian Vector Autoregressions–Five Years of Experience,” Journal of Business and Economic Statistics, Vol.5, pp. 25–38.
- Villani, Mattias, 2005, “Inference in Vector Autoregressive Models with an Informative Prior on the Steady State,” Working Paper No. 181, Sveriges Riksbank.
- Villani, Mattias, and Anders Warne, 2003, “Monetary Policy Analysis in a Small Open Economy Using Bayesian Cointegrated Structural VARs,” Working Paper No. 296, European Central Bank.
- Tierny, Luke, 1994, “Markov Chains for Exploring Posterior Distributions,” Annals of Statistics, Vol. 22, pp. 1701–1762.
- Waggoner, Daniel F. and Tao Zha, 1999, “Conditional Forecasts in Dynamic Multivariate Models,” Review of Economics and Statistics, Vol.81, pp. 639–651.

### Econometric Tests and Time-Series Methodology
- Said, Said E. and David A. Dickey, 1984, “Testing for Unit Roots in Autoregressive-Moving Average Models of Unknown Order,” Biometrika, Vol 71, pp. 599–607.
- Kwiatkowski, Denis, Peter C. B. Phillips, Peter J. Schmidt, and Yongcheol Shin, 1992, “Testing the Null Hypothesis of Stationarity Against the Alternative of a Unit Root: How Sure Are We That Time Series Have a Unit Root?,” Journal of Econometrics, Vol. 54, pp. 159–178.
- Orphanides, Athanasios and Simon van Norden, 2002, “The Unreliability of Output-Gap Estimates in Real Time,” Review of Economics and Statistics, Vol. 84, pp.569–583.
- Leeper, Eric M. and Tao Zha, 2003, “Modest Policy Interventions,” Journal of Monetary Economics, Vol. 50, 1673–1700.

### International Institutions and Global Outlooks
- International Monetary Fund, 2006a, Regional Economic Outlook: Western Hemisphere, (Washington).
- International Monetary Fund, 2006b, World Economic Outlook, September (Washington).
- International Monetary Fund, 2007, World Economic Outlook, April (Washington).
- World Bank, 2006, Global Economic Prospects 2007: Managing the Next Wave of Globalization (Washington).

*Source: _wp07176 - REFERENCES (IMF PDF).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp07176.pdf_