## _wp08191

## Source details

**Canonical URL:** [_wp08191](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2008/_wp08191.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2008/_wp08191.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2008/_wp08191.pdf.json)

---

### Introduction and objective
- Objective: develop a small macro model of the transmission mechanism of monetary policy in an economy with typical EM features and empirically implement the model on data for Brazil.
- Motivation: Brazil formally adopted inflation-targeting (IT) following a major currency crisis between late 1998 and January 1999 and the floating of the real, providing a longer time series under the new regime relative to other EMs.
- Key empirical challenge: paucity of long time series in EMs; need for parsimonious, theory-based restrictions in SVAR estimation for identification.

### Model features and augmentations
- Canonical small macro model augmented to incorporate EM-specific features:
  - External debt accumulation equation to drive the country risk premium and the exchange rate, linking to exogenous terms-of-trade fluctuations and endogenous determinants of the external trade balance such as variation in domestic absorption.
  - Inclusion of external debt accumulation imposes stock-flow dynamics that allow an invertible VAR representation, enabling estimation as a structural VAR (SVAR).
  - Bank-dependent domestic private sector to capture bank-intermediation channel (Bernanke and Blinder (1988)) and effects of shocks to banks' lending capacity (e.g., exogenous changes in reserve requirements and/or banking intermediation technology).
- Canonical model components:
  - Absorption (domestic expenditure) equation.
  - Output gap equation linking GDP, absorption and real exchange rate.
  - Phillips-curve type inflation equation including the output gap and exchange rate.
  - Taylor-type interest rate rule.
  - Exchange rate (flexible UIP) equation with a risk premium tied to net external debt.
  - Log-linearized external debt identity.
- Credit channel modeling:
  - Excess credit (ratio of nominal credit to nominal GDP relative to a “normal” level) included and ordered to allow contemporaneous effects; explanatory variables include absorption, the real interest rate (split into nominal and inflation components), and the real exchange rate.

### Econometric strategy and identification
- Empirical implementation: SVAR representation of the model; quarterly data for 1999Q1 to 2007Q2; main estimation from 1999Q2 (adoption of IT).
- Estimation choices and rationale:
  - SVAR(1) preferred after experimentation.
  - Gaps (absorption, GDP, foreign GDP) computed via Beveridge-Nelson (BN) decomposition from univariate AR(4) fits; HP-filter (1600λ) used for robustness comparisons.
  - Expectations proxied within the SVAR by regressing variables on lags and regressors rather than imposing full rational-expectations timing assumptions.
  - Real interest rate decomposed into nominal rate less target inflation and deviation of inflation from target; exchange rate allowed in the monetary policy reaction function.
- Performance: Despite short sample (1999–2007), SVAR estimates did not generate "price puzzles", "exchange rate puzzles" and other anomalies common in literature using conventional unrestricted VARs.

### Main empirical findings
- Speed of transmission:
  - Compared with the United States and other advanced countries, the transmission mechanism in Brazil works much faster, with the bulk of the effects on output and inflation taking place within a year.
  - Over 1999–2007 the transmission mechanism operates with shorter lags than in advanced economies: the bulk of effects on output and inflation take place within a year.
- Magnitude and evolution of responses:
  - The magnitude of the output gap and inflation responses to monetary policy shocks are described as smaller (relative to advanced countries).
  - Adoption of the IT framework in mid-1999 is associated with larger effects of monetary policy on inflation, consistent with a stronger expectational channel under IT and stricter central bank adherence to a low inflation mandate.
  - Inflation responses to monetary policy grew larger over time during the IT period, implying a lower “sacrifice” in output for a given disinflation effort.
- Exchange rate and external channels:
  - Exchange rate effects on inflation are important despite Brazil being relatively closed to foreign trade (ratios of exports and imports to GDP below 15%).
  - The exchange rate response to domestic policy interest rate changes is sizeable and akin to results from structural models based on uncovered interest parity.
  - Net external debt enters the exchange rate equation with a coefficient of -0.51 (statistically significant at close to 1 percent).
- Bank credit channel:
  - Two opposing channels operate:
    - Inter-temporal effect: monetary tightening (loosening) depresses (boosts) bank credit and absorption.
    - Intra-temporal effect: monetary tightening tends to appreciate the exchange rate in the short run, which can be expansionary for bank credit because of currency-mismatch balance-sheet effects and the greater bank-dependence of the non-tradable sector.
  - Net effect: contractionary but partially offset; credit shocks have limited net magnitudes.

### Key quantitative statistics and estimated magnitudes
- Reported single-equation point estimates (t ratios in brackets) for the three main behavioral equations (preserved verbatim as in source):
  - tt1t1t1 ˆˆ 0.311.56 y0.12z −−− π= π +   −
  - ()()( ) 1.691.582.16− () *tt1ttt ˆˆˆ z0.62z0.69  ir0.51d −− =+−π+−
  - ()()() 4.48     2.16   2.44− () tt1t1t1t1 ˆ ˆ pc0.54pc0.08  i0.08z −−−− =−−π+
- Interpretations and elasticities:
  - A one percentage point reduction in the output gap lowers inflation by 1.5 percentage points (output gap coefficient ~1.56 in the full sample; increased to 2.18 in the 2001q2–2007q2 sub-sample).
  - A 10 percent appreciation in the real effective exchange rate leads to a 1.2 percentage point reduction in inflation, ceteris paribus.
  - A 10 percent appreciation leads to a drop in the ratio of credit to GDP close to 1 percentage point.
- Impulse response magnitudes (1999Q2–2007Q2):
  - Interest rate shock: 100 basis point rise in the yearly interest rate.
  - Response of GDP gap: cumulatively declines by 23 bps in the first year.
  - Response of annual inflation: decline peaks at just under 40 bps in the third quarter after tightening; bulk of effects occur within four quarters.
  - Credit shock: a 1 percentage point rise in excess credit (about half the standard deviation) leads to a rise in absorption and, after one period, to a rise in inflation; overall effects are not large.
- Robustness and sub-sample results:
  - Using an HP-filter (1600λ) output gap produces similar IRs; main differences: deeper fall in the output gap and a trough in inflation at 50 bps in the fourth quarter (vs. 40 bps with BN).
  - Re-estimating for 2001q2–2007q2 yields stronger short-run effects: inflation decline now peaks at 70bps (as opposed to 35bps earlier) while cumulative GDP gap drop is 26 bps in the first year (3 bps deeper than earlier estimate).

### Data, measurement choices, and estimation details
- Data sources and construction:
  - Seasonally adjusted national income account data from IMF's IFS and Brazilian Planning Ministry Research Institute (IPEA).
  - Domestic bank credit to the private sector seasonally adjusted with X11 in AREMOS.
  - Real exchange rate: IMF series computed as a weighted average of exchange rates for nearly all trading partners using CPI deflators and 2000 weights.
  - Foreign real GDP: IMF series weighted by trading partners' trade with Brazil.
  - Inflation: seasonally adjusted CPI including administered and “free market” prices (estimation results reported use “all item” CPI).
- Gap measurement:
  - Beveridge-Nelson (BN) decomposition used to extract permanent components; BN constructed from AR(p) representation of growth rates.
  - HP filter (1600λ) compared for robustness.
- Estimation specifics:
  - Quarterly data for 1999Q1 to 2007Q2; main estimation from 1999Q2.
  - SVAR(1) preferred; gaps via BN from AR(4).
  - Parameters for debt equation use average import, export and GNE ratios to GDP over 1995–2007 and average net foreign debt/GDP over the same period; target inflation set at 4.5%.

### Structural derivation and appendix highlights
- External liability equation derivation (key definitions preserved):
  - D*_t = net external liability in foreign currency.
  - M_t and X_t = import and export trade volumes.
  - i*_t = foreign (risk free) interest rate.
  - s_t = country spread over the foreign interest rate.
- Representative identities and log-linearizations preserved as in source, including the steady-state assumption c_i = ψ and the resulting simplified log-linear forms for d̂_t and related expressions.
- Note: making s_t a function of d_t and d_{t−1} stabilizes the unit-root behavior in practice; deviations around steady-state used in the SVAR.

### Figures and selected descriptive statistics (captions and explicit figures preserved)
- Figure 1. Brazil: Monetary and Price Indicators — Panels: Inflation and the Real/US$ Exchange Rate; Nominal and Real Interest Rates (percent); Real Effective Exchange Rate and Relative Non-Tradable Prices (Indices 2000=100).
- Figure 2. Brazil: Output Indicators — Real GDP Growth Rates (percent a year). Explicit sample statistics:
  - 1980q1-1994q2: Growth  = 1.7%
  - σ_Δy = 4.72%
  - 1994q3-2007q2: Growth  = 2.9% (t=1.51)
  - σ_Δy = 2.66%
- Figure 3. Brazil: External Indicators — Panels: External Terms of Trade and Trade Balance/GDP (TOT and (X-M)/GDP, right axis); Net External Debt/GDP and EMBI spreads (bps).
- Figure 4. Brazil: Financial Indicators — Panels: Bank Spreads and Money Market Interest Rate (percent); Bank Credit and the Real Exchange Rate (REER: 2000=100; Credit: 3-Q MA of annualized % change).
- Figures 6–11. Impulse-Response Results (1999q2:2007q2 and sub-samples) — include:
  - Figure 6: Impulse-Responses to 100 bp Monetary Tightening, 1999q2:2007q2 (in percent).
  - Figure 7: Impulse-Responses to 1% Credit Growth Shock, 1999q2:2007q2 (in percent).
  - Figure 8: Impulse-Responses to 100 bp Monetary Tightening with HP Gap Measures, 1999:2-2007Q2 (in percent).
  - Figure 9: Impulse-Responses to 100 bp Monetary Tightening, 2001q2-2007q2 (in percent).
  - Figure 10: Impulse-Responses to 1% Credit Growth Shock, 2001q2-2007q2 (in percent).
  - Figure 11: Recursive Coefficient Estimates of Output Gap in Inflation Equation (time series from 2002Q2 to 2007Q2).
- Impulse-response axes and scales preserved in the figures as presented (e.g., credit growth panels with ranges such as 0 to 1.2; external debt/GDP panels with explicit ranges).

### Robustness, limitations, and policy implications
- Robustness and limitations:
  - Short sample (34 quarterly observations) implies many parameters relative to data; individual t-ratios less informative though equation signs and most magnitudes conform to theoretical priors; single-equation R2s range from 0.45 (inflation) to 0.90.
  - Absorption shock interpretation ambiguous (transitory demand vs. supply); GDP shock conditional on absorption.
  - Bank credit statistics may understate broader credit effects where informal credit sectors exist; micro-level evidence would help isolate the “pure credit” channel.
  - Inclusion of stock external debt variable important for invertibility and avoiding non-invertible MA representations.
- Policy implications and conclusions:
  - Shorter transmission lags in Brazil imply the bulk of monetary policy effects on output and inflation occur within a year, facilitating policy fine-tuning and faster reversal of policy overshooting or undershooting.
  - Exchange rate movements significantly amplify disinflation via pass-through and an interest-rate sensitive exchange rate (consistent with UIP plus an endogenous country risk premium), even in a relatively closed economy.
  - The bank credit channel matters: monetary tightening reduces credit via the inter-temporal channel but can temporarily boost credit via intra-temporal exchange rate appreciation and balance-sheet effects—net effect is contractionary but partially offset.
  - Over the IT period inflation became more responsive to monetary policy and to the output gap, implying lower output costs (lower sacrifice ratios) for a given disinflation path and highlighting the payoff of credible and consistently firm monetary policy under IT.

*Source: _wp08191 - References*

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

### _wp08191 - References

### Introduction and objective
- The paper addresses the paucity of empirical research on the monetary transmission mechanism in emerging-market (EM) economies relative to advanced countries, noting problems such as shorter time series.
- Objective: develop a small macro model of the transmission mechanism of monetary policy in an economy with typical EM features and empirically implement the model on data for Brazil.
- Motivation: Brazil formally adopted inflation-targeting (IT) following a major currency crisis between late 1998 and January 1999 and the floating of the real, providing a longer time series under the new regime relative to other EMs.

### Model features and augmentations
- Canonical small macro model augmented to incorporate EM-specific features:
  - External debt accumulation equation to drive the country risk premium and the exchange rate, linking to exogenous terms-of-trade fluctuations and endogenous determinants of the external trade balance such as variation in domestic absorption.
  - Inclusion of external debt accumulation imposes stock-flow dynamics that allow an invertible VAR representation, enabling estimation as a structural VAR (SVAR).
  - Bank-dependent domestic private sector to capture bank-intermediation channel (Bernanke and Blinder (1988)) and effects of shocks to banks' lending capacity (e.g., exogenous changes in reserve requirements and/or banking intermediation technology).
- Rationale: these augmentations permit tracing the effects of monetary policy shocks through both external aggregates and the bank-intermediation channel, which is argued to be particularly relevant in EMs.

### Econometric strategy and identification
- Empirical implementation: SVAR representation of the model to identify key features and speeds of the monetary policy transmission mechanism in Brazil.
- Challenges and approach:
  - Small sample of observations in EMs makes sensible theoretical restrictions important for dimensionality reduction.
  - The unrestricted version of the model has a large number of parameters; imposing parsimony is necessary to aid interpretation.
  - Alternatives like imposing a factor structure are possible, but the paper follows a strategy of applying theory-based restrictions to reach a balance between fit and parsimony.
- Performance: Despite the relatively short time span available for estimation (1999–2007), the SVAR estimates did not generate "price puzzles", "exchange rate puzzles" and other anomalies common in literature using conventional unrestricted VARs.

### Main results and empirical findings
- Speed of transmission:
  - Compared with the United States and other advanced countries, the transmission mechanism in Brazil works much faster, with the bulk of the effects on output and inflation taking place within a year.
  - This faster transmission is consistent with structural features of the Brazilian economy such as shorter maturity of domestic credit and relatively high exchange rate pass-through.
- Magnitude of responses:
  - While the transmission operates faster, the magnitude of the output gap and inflation responses to monetary policy shocks are described as smaller (relative to advanced countries), as indicated in the text.

### Relevance and policy implications
- The framework allows evaluation of central-bank responses under an IT framework that has been maintained in Brazil despite major external shocks and political gyrations.
- The model provides a tool to trace how high real interest rates—often employed by the central bank in response to inflation—propagate through output, inflation, external balances, and banking intermediation.
- The inclusion of banking intermediation and external debt dynamics highlights policy channels (reserve requirements, banking technology, external-debt accumulation) through which monetary policy and shocks to financial intermediation affect the macroeconomy.

_Source: _wp08191 - References

### introduction of the IT framework in mid-1999, the effects on inflation are estimated to be

### _wp08191 - introduction of the IT framework in mid-1999, the effects on inflation are estimated to be

### Major empirical findings
- Adoption of the IT framework in mid-1999 is associated with larger effects of monetary policy on inflation, consistent with a stronger expectational channel under IT and stricter central bank adherence to a low inflation mandate.
- Exchange rate effects on inflation are important despite Brazil being relatively closed to foreign trade (ratios of exports and imports to GDP below 15%).
- The exchange rate response to domestic policy interest rate changes is sizeable and akin to results from structural models based on uncovered interest parity.
- The bank credit channel plays a role in monetary transmission, though net magnitudes are not very large; two opposing channels operate:
  - Inter-temporal effect: monetary tightening (loosening) depresses (boosts) bank credit and absorption.
  - Intra-temporal effect: monetary tightening tends to appreciate the exchange rate in the short run, which can be expansionary for bank credit because of currency-mismatch balance-sheet effects and the greater bank-dependence of the non-tradable sector.
- Over 1999–2007 the transmission mechanism operates with shorter lags than in advanced economies: the bulk of effects on output and inflation take place within a year.
- Inflation responses to monetary policy grew larger over time during the IT period, implying a lower “sacrifice” in output for a given disinflation effort.

### Key quantitative statistics and estimated magnitudes (preserved as in source)
- Single-equation point estimates (t ratios in brackets) reported for the three main behavioral equations:
  - tt1t1t1 ˆˆ 0.311.56 y0.12z −−− π= π +   −
  - ()()( ) 1.691.582.16− () *tt1ttt ˆˆˆ z0.62z0.69  ir0.51d −− =+−π+−
  - ()()() 4.48     2.16   2.44− () tt1t1t1t1 ˆ ˆ pc0.54pc0.08  i0.08z −−−− =−−π+
- Interpretation of coefficients and elasticities:
  - A one percentage point reduction in the output gap lowers inflation by 1.5 percentage points (output gap coefficient ~1.56 in the full sample; increased to 2.18 in the 2001q2–2007q2 sub-sample).
  - A 10 percent appreciation in the real effective exchange rate leads to a 1.2 percentage point reduction in inflation, ceteris paribus.
  - A 10 percent appreciation leads to a drop in the ratio of credit to GDP close to 1 percentage point.
  - Net external debt enters the exchange rate equation with a coefficient of -0.51 (statistically significant at close to 1 percent).
- Impulse response magnitudes (1999Q2–2007Q2):
  - Interest rate shock: 100 basis point rise in the yearly interest rate.
  - Response of GDP gap: cumulatively declines by 23 bps in the first year.
  - Response of annual inflation: decline peaks at just under 40 bps in the third quarter after tightening; bulk of effects occur within four quarters.
  - Credit shock: a 1 percentage point rise in excess credit (about half the standard deviation) leads to a rise in absorption and, after one period, to a rise in inflation; overall effects are not large.
- Robustness checks and alternative measures:
  - Using an HP-filter (1600λ) output gap produces similar IRs; main differences: deeper fall in the output gap and a trough in inflation at 50 bps in the fourth quarter (vs. 40 bps with BN).
  - Re-estimating for 2001q2–2007q2 (two years after IT onset) yields stronger short-run effects: inflation decline now peaks at 70bps (as opposed to 35bps earlier) while cumulative GDP gap drop is 26 bps in the first year (3 bps deeper than earlier estimate).

### Structural model, identification, and estimation approach
- Model features and motivations:
  - Canonical small open-economy gap model with: an absorption (domestic expenditure) equation, an output gap equation linking GDP, absorption and real exchange rate, a Phillips-curve type inflation equation including the output gap and exchange rate, a Taylor-type interest rate rule, an exchange rate (flexible UIP) equation with a risk premium tied to net external debt, and a log-linearized external debt identity.
  - Augmentations relative to prior literature: separation of absorption and output effects, inclusion of an external liability (debt) equation to ensure invertibility, addition of private bank credit growth (excess credit), and inclusion of exogenous foreign variables (real US money market rate and terms of trade) to capture external spread and terms-of-trade effects.
- Expectations and SVAR representation:
  - Expectations are proxied within the SVAR by regressing variables on their lags and appropriate regressors rather than imposing full rational-expectations timing assumptions, given short sample and limited inflation-expectations series.
  - Real interest rate decomposed into nominal rate less target inflation and deviation of inflation from target to allow expected inflation to deviate from target and permit short-run nominal effects beyond real-rate channels.
  - Exchange rate allowed in the monetary policy reaction function to capture expectational effects and “leaning against the wind” considerations.
- Credit channel modeling:
  - Excess credit (ratio of nominal credit to nominal GDP relative to a “normal” level) is included before other system variables to allow for contemporaneous effects; explanatory variables in the excess credit equation include absorption, the real interest rate (split into nominal and inflation components), and the real exchange rate.
- Estimation details:
  - Quarterly data for 1999Q1 to 2007Q2; main estimation from 1999Q2 (adoption of IT).
  - SVAR(1) preferred after experimentation; gaps (absorption, GDP, foreign GDP) computed via Beveridge-Nelson (BN) decomposition from univariate AR(4) fits.
  - Parameters for debt equation use average import, export and GNE ratios to GDP over 1995–2007 and average net foreign debt/GDP over the same period; target inflation set at 4.5%.

### Data and measurement choices
- Data sources and construction:
  - Seasonally adjusted national income account data from IMF's IFS and Brazilian Planning Ministry Research Institute (IPEA).
  - Domestic bank credit to the private sector from the same sources, seasonally adjusted with X11 in AREMOS.
  - Real exchange rate: IMF series computed as a weighted average of exchange rates for nearly all trading partners using CPI deflators and 2000 weights.
  - Foreign real GDP: IMF series weighted by trading partners' trade with Brazil.
  - Inflation: seasonally adjusted CPI including administered and “free market” prices (estimation results reported use “all item” CPI).
- Gap measurement:
  - The Beveridge-Nelson (BN) decomposition is used to extract permanent components given its I(1) assumption and consistency with the gap-model framework; BN constructed from AR(p) representation of growth rates.
  - The HP filter (1600λ) is compared; BN from AR(4) resembles HP (1600λ) though HP is smoother; estimation limits to post-1994Q3 because of structural changes in moments of output growth.

### Robustness, limitations, and sensitivity
- Model and sample limitations:
  - Short sample (34 quarterly observations) implies many parameters relative to data; individual t-ratios less informative than usual, though equation signs and most magnitudes conform to theoretical priors and yield single-equation R2s ranging from 0.45 (inflation) to 0.90.
  - The absorption shock is hard to interpret (could be transitory demand or supply); the GDP shock is conditional on absorption and can capture shifts in domestic goods preferences.
  - Bank credit statistics may understate broader credit effects where informal credit sectors exist; disaggregated firm- or sector-level evidence would be valuable for pinning down the “pure credit” channel.
- Econometric identification:
  - Inclusion of the stock external debt variable is important for invertibility and to avoid non-invertible MA representations that would arise from compressing the system to flows-only variables.

### Policy implications and conclusions
- Shorter transmission lags in Brazil imply that the bulk of monetary policy effects on output and inflation occur within a year, facilitating policy fine-tuning and faster reversal of policy overshooting or undershooting.
- Exchange rate movements significantly amplify disinflation via pass-through and an interest-rate sensitive exchange rate (consistent with UIP plus an endogenous country risk premium), even in a relatively closed economy.
- The bank credit channel matters: monetary tightening reduces credit via the inter-temporal channel but can temporarily boost credit via intra-temporal exchange rate appreciation and balance-sheet effects—net effect is contractionary but partially offset.
- Over the IT period inflation became more responsive to monetary policy and to the output gap, implying lower output costs (lower sacrifice ratios) for a given disinflation path and highlighting the payoff of credible and consistently firm monetary policy under IT.

*Source: _wp08191 - introduction of the IT framework in mid-1999, the effects on inflation are estimated to be*

### REFERENCES

### _wp08191 - REFERENCES

### References (selected citations)
- Anderson, H.M., C.N. Low and R.D. Snyder, 2006, “Single Source of Error State-Space Approach to the Beveridge-Nelson Decomposition,” Economics Letters, Vol. 91, pp.104–109.
- Angeloni, I, A. Kashyap, B. Mojon and D. Terlizzese, 2003, “The Output Composition Puzzle: A Difference in the Monetary Transmission in the Euro Area and the United States,” Journal of Monetary, Credit and Banking, Vol. 35, No. 6, pp.1265–1305.
- Berg, A., P. Karam, and D. Laxton, 2006, “A Practical Model Based Approach to Monetary Analysis – Overview”, IMF Working Paper 06/80.
- Bernanke, B.S. and A. Blinder, 1988, “Credit, Money, and Aggregate Demand,” American Economic Review, Vol. 73, pp. 257–76.
- Bevilaqua, A. S., M. Mesquita and A. Minella, 2007, “Brazil: Taming Inflation Expectations,” Central Bank of Brazil, Working Paper No. 129.
- Bogdanski, J., A. A. Tombini, and S. R. C. Werlang, 2000, “Implementing Inflation Targeting in Brazil,” Central Bank of Brazil Working Paper No. 1.
- Bonomo, M., B. Martins, and R. Pinto, 2004, “Debt Composition and Exchange Rate Balance Sheet Effects in Brazil: A Firm Level Analysis,” Working Paper No. 535, Fundacao Getulio Vargas, Rio de Janeiro.
- Calvo, G., A. Izquierdo, and L. Fernando Mejia, 2004, “On the Empirics of Sudden Stops: The relevance of Balance Sheet Effects,” NBER Working Paper No. 10520.
- Calvo, G. and C.M. Reinhart, 2002, “Fear of Floating,” Quarterly Journal of Economics, No. 117, pp. 379–408.
- Catão, L.A.V., and M.E. Terrones, 2005, “Fiscal Deficits and Inflation”, Journal of Monetary Economics, Vol. 52, pp. 529–54.
- Cespedes, L. F. and C. Soto, 2005, “Credibility and Inflation Targeting in an Emerging Market: Lessons from the Chilean Experience,” International Finance No. 8, pp 545–75.
- Christiano, L., M. Eichenbaum, and C. Evans, 1998, “Monetary Policy Shocks: What Have We Learned and to What End?” Handbook of Macroeconomics.
- Corbo, V., O. Lanbderretche, and K. Schmidt-Hebbel, 2001, “Assessing Inflation Targeting After a Decade of World Experience,” International Journal of Finance and Economics Vol. 6, pp. 343–68.
- Dungey, M. and A.R. Pagan, 2000, “A Structural VAR Model of the Australian Economy,” Economic Record, Vol. 76, pp. 321–42.
- De Gregorio, J., A. Tokman, and R. Valdes, 2005, “Flexible Exchange Rate with Inflation Targeting in Chile: Experience and Issues,” Inter-American Development Bank, Working Paper No. 540.
- Edwards, S. and C. Végh, 1997, “Banks and Macroeconomic Disturbances Under Predetermined Exchange Rates,” Journal of Monetary Economics, Vol. No. 40, pp. 239–78.
- Eichenbaum, M., 1994, “Comment”, in “Monetary Policy and Bank Lending,” in Mankiw, N.G. (ed.), Monetary Policy, Chicago.
- Favero, C. and F. Giavazzi, 2004, “Inflation Targetting and Debt: Lessons from Brazil,” Centre for Economic Policy Research, Discussion Paper No. 4376, London.
- Fraga, A., I., Goldfajn, and A. Minella, 2005, “Inflation Targeting in Emerging Market Economies,” in Gertler and Rogoff (eds.), NBER Macroeconomics, Annual 2003, Cambridge.
- Fukacs, M., and A.R. Pagan, 2006, “Limited Information Estimation and Evaluation of DSGE Models,” NCER Working Paper No. 6.
- Goodfriend, M., 2005, “Inflation Targeting in the US?” in B. Bernanke and M. Woodford (eds), The Inflation Targeting Debate, Chicago.
- Goodfriend, M. and B. MaCallum, 2007, “Banking and Interest Rates in Monetary Policy Analysis: A Quantitative Exploration,” Journal of Monetary Economics, Vol. No. 54, pp. 1480–1507.
- Harvey, A.C. and A. Jaeger, 1993, “De-Trending, Stylized Facts and the Business Cycle,” Journal of Applied Econometrics, Vol. No. 8, pp. 231–47.
- Kamenik, O., H. Kiem, V. Klyuev, and D. Laxton, 2008, “Why is Canada’s Price Level so Predictable?” IMF Working Paper No. 08/25 (January 2008), available at www.imf.org.
- Kashyap, A. and J. Stein, 1994, “Monetary Policy and Bank Lending,” in Mankiw, N.G. (ed.), Monetary Policy, Chicago.
- Laxton, D., K. Shoom, and R. Tetlow, 1992, “Should the Change in the Gap Appear in the Phillips Curve? Some Consequences of Mis-measuring Potential Output,” Working Paper No. 92-1, Bank of Canada.
- Loayza, N. and R. Soto, 2002, “Inflation Targeting: Design, Performance, Challenges,” Central Bank of Chile.
- Masson, P., M. Savastano, and S. Sharma, 1998, “The Scope for Inflation Targeting in Developing Countries,” IMF Working Paper No. 97/130 (Washington, DC: International Monetary Fund).
- Minella, A., P. Springer, I. Goldfajn, and M. Muinhos, 2003, “Inflation Targeting in Brazil: Constructing Credibility Under Exchange Rate Volatility,” Working Paper No. 77, Central Bank of Brazil.
- Mishkin, F., 2004, “Can Inflation Targeting Work in Emerging Markets?,” in Reinhart, C., A.Velasco, and C. Végh, A Festschrift for Guillermo Calvo.
- Mishkin, F. and K. Schmidt-Hebbel, 2007, Monetary Policy Under Inflation Targeting, Chile.
- Minella, A., 2003, “Monetary Policy and Inflation in Brazil (1975–2000): A VAR Estimation,” Revista Brasileira de Economia, No. 57, pp. 605–35.
- Morley, J., 2007, “Two Interpretations of the Beveridge-Nelson Decomposition,” Mimeo, Washington University.
- Schimitt-Grohe, S. and M. Uribe, 2003, “Closing Small Open Economy Models,” Journal of International Economics, Vol. 61, pp. 163–85.
- Svensson, L. O., 2000, “Open Economy Inflation Targeting,” Journal of International Economics, Vol. 50, pp. 155–83.
- Tombini, A. and S. A. Lago Alves, 2006, “The Recent Brazilian Disinflation Process and Costs,” Central Bank of Brazil Working Paper 109.
- Tornell, A. and F. Westermann, 2005, “Boom-Bust Cycles and Financial Liberalization,” MIT Press.

### Appendix: Derivation of External Liability Equation (key steps and definitions)
- Definitions:
  - D*_t = net external liability in foreign currency.
  - M_t and X_t = import and export trade volumes.
  - i*_t = foreign (risk free) interest rate.
  - s_t = country spread over the foreign interest rate.
- Starting (unnumbered) identity (as presented):
  - ( ) **** * t t t t 1 m, t t x, t t D 1 is D p M p X − = + + + −
- Dividing through by the nominal foreign currency value of potential GDP (E_t Q_t P_t):
  - ( ) * * t t t t t 1 t t t D 1 T T d 1 is D M X E Q P Q Z Q Z − == + + + −
  - where Z_t = p_t^m E_t / Q_t is the real exchange rate and T*_x,t / p^m,t = T_t* is the terms of trade (as given).
- Intermediate log-linearized form (symbols preserved as in source):
  - ( ) ()() * t t t t t t 1 t t t 1 T T d 1 is 1 e 1 d M X Q Z Q Z − = + + − Δ − ψ + −
- Introducing c,t and ψ definitions:
  - c,t: where * c,t t t i s = + (i.e., c_t = i*_t + s_t).
  - ψ_t = π_t + q, where q is the growth rate in potential GDP.
- Log-linearization around equilibrium (c_i, e, Δ = 0, ψ) yields:
  - ( ) ()() t c t 1 t 1 c,t c t t ˆ ˆ d 1 i 1 d ˆ d 1 l 1 i 1 e 1 i ˆ ˆ d − − ⎡⎤ = + − ψ + − ψ ++ − ψ Δ − + π ⎢⎥ ⎣⎦
  - plus terms with m_tˆ, x_tˆ, ω_m, ω_x (equilibrium import and export shares).
- Assumption for steady state:
  - For analytical convenience and well-defined steady state, assume c_i = ψ, which results in the simplified form (as presented):
  - ( ) ()() t t 1 t 1 c,t c t m t t ˆ ˆ d d d 1 l 1 i m z − − ⎡⎤ = + − ψ − + ψ + ω − ⎢⎥ ⎣⎦
  - and remaining terms with x_tˆ and ω_x.
- Stationarity and deviations:
  - The text notes the equation still has a unit root, but if s_t is made a function of d_t and d_{t−1} it should stabilize at some level d.
  - Defining d̂_t as the deviation around steady-state and ignoring interaction terms with d̂_{t−1} yields:
    - ( ) ()() t c,t c t m t t ˆ ˆ ˆ ˆ d d 1 l 1 i m z ⎡⎤ Δ = − ψ − + π + ω − ⎢⎥ ⎣⎦
    - plus terms with x_tˆ and ω_x.
- From the log-linearized national income identity:
  - y_t = n_t + x_t + m_t relations yield the expression (as shown):
    - ( ) ()() t c,t c t n t t x m t x t ˆ ˆ ˆ ˆ d d 1 i 1 iny z tt ⎡⎤ Δ = − ψ − + π + ω − + ω − ω − ω ⎢⎥ ⎣⎦

### Figures and Key Statistics (captions and explicit figures preserved)
- Figure 1. Brazil: Monetary and Price Indicators
  - Panels include: Inflation and the Real/US$ Exchange Rate; Nominal and Real Interest Rates (percent); Real Effective Exchange Rate and Relative Non-Tradable Prices (Indices 2000=100).
- Figure 2. Brazil: Output Indicators
  - Real GDP Growth Rates (percent a year) time series shown.
  - Explicit sample statistics:
    - 1980q1-1994q2: Growth  = 1.7%
    - σ_Δy = 4.72%
    - 1994q3-2007q2: Growth  = 2.9% (t=1.51)
    - σ_Δy = 2.66%
  - Estimates of the Output Gap (Beveridge-Nelson filter) plotted for GDP gap and GNE gap.
- Figure 3. Brazil: External Indicators
  - Panels: External Terms of Trade and Trade Balance/GDP (TOT and (X-M)/GDP, right axis); Net External Debt/GDP and EMBI spreads (bps).
- Figure 4. Brazil: Financial Indicators
  - Panels: Bank Spreads and Money Market Interest Rate (percent); Bank Credit and the Real Exchange Rate (REER: 2000=100; Credit: 3-Q MA of annualized % change).
- Figure 5. BN- and HP-filter Gaps
  - Time series comparison of HP and BN gap measures from 1998Q2 to 2006Q3.
- Figures 6–11. Impulse-Response Results (1999q2:2007q2 and sub-samples)
  - Figure 6: Impulse-Responses to 100 bp Monetary Tightening, 1999q2:2007q2 (in percent). Panels include GDP and GNE gaps; Actual Yearly Inflation - Target Inflation; log of REER; Credit Growth; Trade Balance "Gap"; External Debt/GDP.
  - Figure 7: Impulse-Responses to 1% Credit Growth Shock, 1999q2:2007q2 (in percent). Panels as above.
  - Figure 8: Impulse-Responses to 100 bp Monetary Tightening with HP Gap Measures, 1999:2-2007Q2 (in percent).
  - Figure 9: Impulse-Responses to 100 bp Monetary Tightening, 2001q2-2007q2 (in percent).
  - Figure 10: Impulse-Responses to 1% Credit Growth Shock, 2001q2-2007q2 (in percent).
  - Figure 11: Recursive Coefficient Estimates of Output Gap in Inflation Equation (time series from 2002Q2 to 2007Q2).
- Impulse-response axes and scales are preserved in the figures as presented (e.g., credit growth panels with ranges such as 0 to 1.2 in some figures; external debt/GDP panels with ranges shown explicitly).

*Content and equations reproduced from _wp08191 - REFERENCES.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2008/_wp08191.pdf_
