## 1. Summary of Studies of RER Movements in Transition Countries

## Source details

**Canonical URL:** [1. Summary of Studies of RER Movements in Transition Countries](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp0437.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp0437.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp0437.pdf.json)

---

### I. Introduction
- Maintenance of competitiveness is an important concern in countries with currency board arrangements (CBAs) such as Bulgaria.
- Since the introduction of the CBA in 1997, the currency has appreciated by about 30 percent in real terms (Figure 1).
- The current account widened to about 8½ percent of GDP in 2003, raising concerns about competitiveness.
- Indicators suggest wage costs are rising in line with productivity and that Bulgaria continues to gain market share in partner markets.
- No prior empirical assessment had been made of the determinants of the real exchange rate or its equilibrium value for Bulgaria.
- Studies on other transition countries show much of recent real appreciation is due to changes in fundamentals or real convergence; appreciation can reflect productivity increases in traded-goods sectors and responses to terms of trade, capital inflows, net foreign asset position, or savings.
- Appreciation in response to large, potentially temporary capital inflows may adversely affect competitiveness in a fixed-rate regime.
- The paper applies a mixture of the BEER and NATREX methodologies and finds the RER has mainly been driven by fundamentals and that there is currently no major misalignment in the lev.

### II. Competitiveness, RER, and Transition Countries
- Empirical approaches reviewed:
  - Macroeconomic balance (FEER): compares current RER with a FEER consistent with full employment and current account equilibrium.
  - Behavioral equilibrium exchange rate (BEER): reduced-form econometric estimation linking RER to longer-term fundamentals and short-term variables using cointegration.
  - PEER: extracts permanent components from BEER.
  - NATREX (Stein and Allen, 1995): distinguishes medium- and long-term factors.
- Transition-specific considerations:
  - Substantial real appreciation is normal in transition due to productivity gains and market reforms.
  - Initial devaluations and hyperinflation in transition can cause early undervaluation of currencies.
  - The introduction of the CBA in Bulgaria on July 1, 1997 is a structural break: adopted after a deep financial crisis with inflation of 312 percent in 1996 and 548 percent in 1997 (Source: Bulgarian National Bank).
  - Post-CBA liberalization of most capital movements increased theoretical vulnerability to market sentiment; early inflows were mostly official finance and privatization-related FDI.
  - Structural reforms post-1997 likely boosted productivity in the open sector and influenced real appreciation.
- Prior empirical evidence for transition countries often attributes appreciation to the Balassa-Samuelson effect, productivity differentials, net foreign asset positions, and other fundamentals.

### III. Empirical Analysis of the RER in Bulgaria
- Major estimation challenge: short sample period since 1997; quarterly data from 1997:3 to 2003:1 yields 23 observations.
- Methodology: mixture of BEER and NATREX approaches; ARDL methodology of Pesaran et al (1996) applied because it can be used irrespective of whether regressors are I(1) or I(0) and is suitable for small samples.
- Five medium-term determinants used: terms of trade, productivity, world real interest rates, foreign direct investment, and gross savings.
- Cointegrated equation specified:
  - r = f(tot, productivity, savings, real_libor, fdi).                    (10)

Data (quarterly, 1997:3–2003:1):
- Real effective exchange rate (reer): CPI-based weighted average of bilateral real exchange rates; quarterly indices from Bulgarian National Bank website. An increase denotes real appreciation.
- Terms of trade (tot): numerator = Bulgarian producer price index (proxy for export prices); denominator = weighted index of partner price indices (Germany, Italy, Greece, Russia). Index heavily dependent on oil prices; an increase in oil prices worsens tot because Bulgaria is a net oil importer.
- Productivity (prod): quarterly, seasonally adjusted index = real GDP in 1995 prices divided by number of employees. Sectoral productivity unavailable; proxy implies limited capture of Balassa-Samuelson effect.
- Gross savings (savings): Gross Savings = National Disposable Income – Consumption Expenditures; NDI = GDP + Current Income from Abroad + Net Current Transfers. Gross savings in constant prices deflated by 1997-based CPI; seasonally adjusted. Initial level low after 1997 crisis; gross savings grew until third quarter of 2002, then declined.
- World real interest rate (real_libor): quarterly average three-month USD LIBOR deflated by quarterly U.S. inflation.
- Foreign direct investment (fdi): index of FDI in USD with 1997 base year from BNB balance of payments statistics; after 1997 Bulgaria attracted sizeable FDI, predominantly non-privatization transactions.

Time series properties:
- All series except savings are I(1) and stationary in first differences according to ADF; savings is I(0) by ADF, but Phillips-Perron does not reject unit root for savings. Short sample makes integration determination controversial.

Estimation approach:
- ARDL two-step:
  1. Test existence of long-term relationship via F-statistic in ECM form.
  2. Estimate long-run coefficients and ECM via OLS on ARDL.

Estimated dynamic ARDL specification (notation: l = logarithm, d = first difference, numbers in parentheses are lags, SQ2 seasonal variable):
- RER = 12.27 –0.24*lRER(-1) -0.98*lTOT +0.01*lSAVINGS – 0.32*lPRODUCTIVITY –0.08*REAL_LIBOR +0.003*lFDI +1.36*dlTOT +1.46*dlTOT(-1) +1.15*dlTOT(-2) +0.37*dlTOT(-3) +0.10*dREAL_LIBOR –0.04*dlSAVINGS(-2) +0.56*dlPRODUCTIVITY +0.01*dREAL_LIBOR(-2) –0.08*SQ2

Estimated long-run relationship (coefficients obtained by dividing log-coefficients by coefficient of lRER(-1) = –0.245):
- RER = -49.99 + 3.99ltot – 0.06lsavings + 1.3lproductivity + 0.34real_libor – 0.01lfdi
  - t-values (in brackets): (-27.96) (22.36) (-2.89) (19.74) (30.99) (-1.78)

Interpretation of coefficients:
- Coefficient signs as expected.
- Terms of trade, productivity, and real LIBOR significant at the 1 percent level.
- Gross savings significant at the 10 percent level.
- Foreign direct investment significant at the 5 percent level.
- Coefficient on lRER(-1) = –0.245 implies RER reaches equilibrium after approximately 4 periods (i.e., a year) following a shock.
- Tests of serial correlation and normality favorable; heteroskedasticity could not be tested due to short sample.
- Cautions: results should be interpreted cautiously given short sample and data limitations.

### C. Misalignment of Real Exchange Rate
- Definitions:
  - Short-run current misalignment (speculative): deviation of actual RER from actual equilibrium RER estimated using actual values of fundamentals.
  - Medium-run total misalignment (cyclical and speculative): deviation of actual RER from sustainable equilibrium RER estimated using sustainable values of fundamentals (obtained via Hodrick-Prescott filter).
- Calculations for 2003:Q1:
  - Current misalignment estimated at -0.14 percent.
  - Total misalignment estimated at 1.77 percent.
- Conclusion: no significant deviation between actual RER and estimated short- and medium-term equilibrium levels for the sample period.
- Additional summary of misalignment dynamics:
  - Undervaluation of the real exchange rate from 1998 to the end of 2001.
  - Since 2002 there has been some overvaluation but it has been decreasing.
  - During the period of the study, the total misalignment is in the range from –6.2 percent to 7.1 percent and its trend is approaching zero.
  - The long-run tendency is a convergence of the actual to the equilibrium level of the real effective exchange rate.
  - The use of total productivity data instead of sectoral may have biased the results towards finding overvaluation, which would strengthen the result of no overvaluation.
  - Given the small sample and data problems, the results should be considered tentative and subject to some uncertainty.

### Methodology and Model Structure
- Behavioral model framework:
  - r_t = β1 Z_1t + β2 Z_2t + τ T_t + ε = r^E + γ T + ε
  - r^E = α Z_1 + β Z_2
  - r^F = α Z^F_1 + β Z^F_2
- Grouping of fundamentals:
  - Z1: persistent fundamentals (terms of trade (tot), sectoral or aggregate productivity (prod), net domestic savings (sav), balance of public finance (fisc), net foreign assets (nfa)); improvements lead to appreciation.
  - Z2: fundamentals affecting RER over the medium term (may coincide with business cycle); includes real interest rate differential (RID) and risk premium (σ_t = (λ_t + k)).
- Risk premium modeled as a positive function of relative supply of domestic and foreign debt: σ_t = (debt_t/debt*_t); increased domestic debt supply relative to foreign debt supply increases risk premium and requires depreciation.
- BEER specification:
  - BEER = r^E(tot, prod, fisc, sav, nfa, i-i*, debt/debt*)

BEER and NATREX considerations:
- BEER limitations:
  - Does not identify which fundamentals are sustainable; sustainable levels should be calibrated (e.g., Hodrick-Prescott filter).
  - Reflects both long- and medium-term factors; excluding medium-term factors yields a measure close to NATREX.
- NATREX description:
  - Fluctuating medium- to long-term equilibrium exchange rate determined by fundamentals in absence of speculative capital flows and changes in monetary reserves.
  - Determinants include savings, productivity, foreign debt, exogenous terms of trade, and world real interest rates.
  - NATREX is not necessarily welfare-optimal and is non-stationary because its fundamentals are not stationary.
  - Medium-term intracyclical equilibrium condition: I – S + CA = 0 (I = intended investment, S = intended savings, CA = intended current account).

### Model Shortcomings and Caveats
- Main problems:
  - Identification and choice of short-term, medium-term, and long-term fundamentals.
  - Calibration of sustainable levels of fundamentals is difficult.
  - Calibration of external balance is difficult because uncovered interest parity may not hold under external financing constraints.
  - Cointegration analysis requires long samples; short time series may reflect only short- and medium-term relationships.
  - The correction term from cointegration may not capture a clean medium-term reaction if the state of activity changes quickly after exogenous shocks.
- Results are tentative due to the small sample and data problems.

### Estimation Results (selected diagnostics and coefficients)
- Estimation context:
  - Dependent Variable: LREER
  - Method: Least Squares
  - Date: 08/22/03 Time: 16:45
  - Sample(adjusted): 1998:3 2003:1
  - Included observations: 19 after adjusting endpoints
- Selected estimated coefficients (Variable — Coefficient — Std. Err. — t-Statistic — Prob.):
  - C — 12.27347 — 0.438909 — 27.96357 — 0.0001
  - LREER(-1) — -0.245496 — 0.049092 — -5.000715 — 0.0154
  - LTOT — -0.979524 — 0.043801 — -22.36329 — 0.0002
  - LSAVINGS — 0.013974 — 0.004839 — 2.887876 — 0.0631
  - LPRODUCTIVITY — -0.319561 — 0.016189 — -19.73952 — 0.0003
  - REAL_LIBOR01 — -0.084057 — 0.002712 — -30.9938 — 0.0001
  - LFDI — 0.003484 — 0.001957 — 1.780599 — 0.1730
  - DLTOT — 1.361993 — 0.053704 — 25.36124 — 0.0001
  - DLTOT(-1) — 1.460687 — 0.058488 — 24.97424 — 0.0001
  - DLTOT(-2) — 1.153782 — 0.04667 — 24.72193 — 0.0001
  - DLTOT(-3) — 0.365241 — 0.015666 — 3.31426 — 0.0002
  - DREAL_LIBOR01 — 0.099495 — 0.003583 — 27.7711 — 0.0001
  - DLSAVINGS(-2) — -0.036536 — 0.003485 — -10.48348 — 0.0019
  - DLPRODUCTIVITY — 0.560812 — 0.047237 — 11.87228 — 0.0013
  - DREAL_LIBOR01(-2) — 0.014337 — 0.002207 — 6.497198 — 0.0074
  - SQ2 — -0.077514 — 0.003659 — -21.18219 — 0.0002
- Model fit and diagnostics:
  - R-squared 0.999692
  - Adjusted R-squared 0.998152
  - S.E. of regression 0.002560
  - Sum squared resid 1.97E-05
  - Log likelihood 103.9611
  - F-statistic 649.0900
  - Prob(F-statistic) 0.000088
  - Durbin-Watson stat 1.602860
- Residual diagnostics (selected):
  - Normality (Jarque-Bera) — Jarque-Bera 1.383482 — Probability 0.500703
  - ARCH Test — F-statistic 0.983238 — Probability 0.398504; Obs*R-squared 2.093767 — Probability 0.351030
  - Breusch-Godfrey Serial Correlation LM Test — F-statistic 0.197464 — Probability 0.846690; Obs*R-squared 5.379221 — Probability 0.067907

### Policy Implications and Recommendations
- Widening current account deficit appears driven by demand pressures (increased capital inflows, growing bank credit) rather than competitiveness deterioration.
- Policy recommendation: implement demand management policies, with emphasis on a cautious fiscal stance.

*Source: _wp0437*

### 1. Summary of Studies of RER Movements in Transition Countries....................................   7

### 1. Summary of Studies of RER Movements in Transition Countries

### Figures
- Figure 1. Real Effective Exchange Rate of Lev ................................................................................   4
- Figure 2. Actual and Total Misalignment of Real Bulgarian Exchange Rate (ARDL Method) ....... 13

### Appendices
- I. Explanation of Model and Its Shortcomings ............................................................. 14
- II. Estimation Results..................................................................................................... 18

*Source: _wp0437 - 1. Summary of Studies of RER Movements in Transition Countries....................................   7*

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

### References

### I. INTRODUCTION
- Maintenance of competitiveness is an important concern in countries with currency board arrangements (CBAs) such as Bulgaria.
- Since the introduction of the CBA in 1997, the currency has appreciated by about 30 percent in real terms (Figure 1).
- The current account widened to about 8½ percent of GDP in 2003, raising concerns about competitiveness.
- Indicators suggest wage costs are rising in line with productivity and that Bulgaria continues to gain market share in partner markets (see IMF Country Report No. 03/206).
- No prior empirical assessment had been made of the determinants of the real exchange rate or its equilibrium value for Bulgaria.
- Studies on other transition countries show much of recent real appreciation is due to changes in fundamentals or real convergence; appreciation can reflect productivity increases in traded-goods sectors and responses to terms of trade, capital inflows, net foreign asset position, or savings.
- Appreciation in response to large, potentially temporary capital inflows may adversely affect competitiveness in a fixed-rate regime.
- The paper applies a mixture of the BEER and NATREX methodologies and finds the RER has mainly been driven by fundamentals and that there is currently no major misalignment in the lev.

### II. COMPETITIVENESS, RER, AND TRANSITION COUNTRIES
- Empirical approaches reviewed:
  - Macroeconomic balance (FEER): compares current RER with a FEER consistent with full employment and current account equilibrium.
  - Behavioral equilibrium exchange rate (BEER): reduced-form econometric estimation linking RER to longer-term fundamentals and short-term variables using cointegration.
  - PEER: extracts permanent components from BEER.
  - NATREX (Stein and Allen, 1995): distinguishes medium- and long-term factors.
- Transition-specific considerations:
  - Substantial real appreciation is normal in transition due to productivity gains and market reforms.
  - Initial devaluations and hyperinflation in transition can cause early undervaluation of currencies.
  - The introduction of the CBA in Bulgaria on July 1, 1997 is a structural break: adopted after a deep financial crisis with inflation of 312 percent in 1996 and 548 percent in 1997 (Source: Bulgarian National Bank).
  - Post-CBA liberalization of most capital movements increased theoretical vulnerability to market sentiment; early inflows were mostly official finance and privatization-related FDI.
  - Structural reforms post-1997 likely boosted productivity in the open sector and influenced real appreciation.
- Prior empirical evidence for transition countries often attributes appreciation to the Balassa-Samuelson effect, productivity differentials, net foreign asset positions, and other fundamentals (see Table 1 summary of studies).

### III. EMPIRICAL ANALYSIS OF THE RER IN BULGARIA
- Major estimation challenge: short sample period since 1997; quarterly data from 1997:3 to 2003:1 yields 23 observations.
- Methodology: mixture of BEER and NATREX approaches; ARDL methodology of Pesaran et al (1996) applied because it can be used irrespective of whether regressors are I(1) or I(0) and is suitable for small samples.
- Five medium-term determinants used: terms of trade, productivity, world real interest rates, foreign direct investment, and gross savings.
- Cointegrated equation specified as:
  - r = f(tot, productivity, savings, real_libor, fdi).                    (10)

Data (quarterly, 1997:3–2003:1):
- Real effective exchange rate (reer): CPI-based weighted average of bilateral real exchange rates; quarterly indices from Bulgarian National Bank website. An increase denotes real appreciation.
- Terms of trade (tot): numerator = Bulgarian producer price index (proxy for export prices); denominator = weighted index of partner price indices (Germany, Italy, Greece, Russia). Structure makes index heavily dependent on oil prices; an increase in oil prices worsens tot because Bulgaria is a net oil importer.
- Productivity (prod): quarterly, seasonally adjusted index = real GDP in 1995 prices divided by number of employees (National Statistics Institute data). Sectoral productivity unavailable; proxy implies limited capture of Balassa-Samuelson effect.
- Gross savings (savings): computed from nominal GDP with Gross Savings = National Disposable Income – Consumption Expenditures; NDI = GDP + Current Income from Abroad + Net Current Transfers. Gross savings in constant prices deflated by 1997-based CPI; seasonally adjusted. Initial level low after 1997 crisis; gross savings grew until third quarter of 2002, then declined.
- World real interest rate (real_libor): quarterly average three-month USD LIBOR deflated by quarterly U.S. inflation; data from Bloomberg and http://www.stls.frb.org/.
- Foreign direct investment (fdi): index of FDI in USD with 1997 base year from BNB balance of payments statistics; after 1997 Bulgaria attracted sizeable FDI, predominantly non-privatization transactions.

Time series properties:
- All series except savings are I(1) and stationary in first differences according to ADF; savings is I(0) by ADF, but Phillips-Perron does not reject unit root for savings. Short sample makes integration determination controversial.

Estimation of cointegrated equation:
- ARDL two-step: (1) test existence of long-term relationship via F-statistic in ECM form; (2) estimate long-run coefficients and ECM via OLS on ARDL.
- Estimated dynamic ARDL specification (with notation: l = logarithm, d = first difference, numbers in parentheses are lags, SQ2 seasonal variable):
  - RER = 12.27 –0.24*lRER(-1) -0.98*lTOT +0.01*lSAVINGS – 0.32*lPRODUCTIVITY –0.08*REAL_LIBOR +0.003*lFDI +1.36*dlTOT +1.46*dlTOT(-1) +1.15*dlTOT(-2) +0.37*dlTOT(-3) +0.10*dREAL_LIBOR –0.04*dlSAVINGS(-2) +0.56*dlPRODUCTIVITY +0.01*dREAL_LIBOR(-2) –0.08*SQ2
- Estimated long-run relationship (coefficients obtained by dividing log-coefficients by coefficient of lRER(-1) = –0.245):
  - RER = -49.99 + 3.99ltot – 0.06lsavings + 1.3lproductivity + 0.34real_libor – 0.01lfdi
    - t-values (in brackets): (-27.96) (22.36) (-2.89) (19.74) (30.99) (-1.78)
- Interpretation:
  - Coefficients signs as expected.
  - Terms of trade, productivity, and real LIBOR significant at the 1 percent level.
  - Gross savings significant at the 10 percent level.
  - Foreign direct investment significant at the 5 percent level.
  - Coefficient on lRER(-1) = –0.245 implies RER reaches equilibrium after approximately 4 periods (i.e., a year) following a shock.
  - Tests of serial correlation and normality favorable; heteroskedasticity could not be tested due to short sample.
- Cautions: results should be interpreted cautiously given short sample and data limitations (see Appendix II for discussion).

### C. MISALIGNMENT OF REAL EXCHANGE RATE
- Two types of deviations defined:
  - Short-run current misalignment (speculative): deviation of actual RER from actual equilibrium RER estimated using actual values of fundamentals.
  - Medium-run total misalignment (cyclical and speculative): deviation of actual RER from sustainable equilibrium RER estimated using sustainable values of fundamentals (obtained via Hodrick-Prescott filter).
- Calculations for 2003:Q1:
  - Current misalignment estimated at -0.14 percent.
  - Total misalignment estimated at 1.77 percent.
- Conclusion: no significant deviation between actual RER and estimated short- and medium-term equilibrium levels for the sample period.

*Source: _wp0437 - References (IMF working paper content as provided).*

### 2003. This suggests that there is no short-run or significant long-run misalignment,

### _wp0437 - 2003. This suggests that there is no short-run or significant long-run misalignment,

### Key findings on real exchange rate (RER) misalignment
- According to the calculations, there is an undervaluation of the real exchange rate from 1998 to the end of 2001.
- Since 2002 there has been some overvaluation but it has been decreasing.
- During the period of the study, the total misalignment is in the range from –6.2 percent to 7.1 percent and its trend is approaching zero.
- The long-run tendency is a convergence of the actual to the equilibrium level of the real effective exchange rate.
- The use of total productivity data instead of sectoral may have biased the results towards finding overvaluation, which would strengthen the result of no overvaluation.
- Given the small sample and data problems, the results should be considered tentative and subject to some uncertainty.
- The widening current account deficit is not attributed to problems with competitiveness; continued growth of exports in recent years (despite sluggish demand in main partner markets in 2003) supports this.
- Rapid growth of imports is likely to reflect demand pressures from increased capital inflows and growing bank credit, implying a need for demand management policies, especially a cautious fiscal stance.

### Methodology and model structure
- The study uses behavioral models of exchange rates (especially the BEER and the NATREX) and constructs an econometric behavioral model to analyze medium-term and long-term dynamics of the real exchange rate.
- Reduced-form representation:
  - r_t = β1 Z_1t + β2 Z_2t + τ T_t + ε = r^E + γ T + ε
  - r^E = α Z_1 + β Z_2
  - r^F = α Z^F_1 + β Z^F_2
- Grouping of fundamentals:
  - Z1: fundamental factors with persistent effects (terms of trade (tot), sectoral or aggregate productivity (prod), net domestic savings (sav), balance of public finance (fisc), net foreign assets (nfa)); improvements lead to appreciation.
  - Z2: fundamentals affecting RER over the medium term (may coincide with business cycle); includes real interest rate differential (RID) and risk premium (σ_t = (λ_t + k)).
- Risk premium modeled as a positive function of relative supply of domestic and foreign debt: σ_t = (debt_t/debt*_t); increased domestic debt supply relative to foreign debt supply increases risk premium and requires depreciation.
- BEER specification:
  - BEER = r^E(tot, prod, fisc, sav, nfa, i-i*, debt/debt*)
- BEER limitations:
  - Does not identify which fundamentals are sustainable; sustainable levels should be calibrated (e.g., Hodrick-Prescott filter).
  - Reflects both long- and medium-term factors; excluding medium-term factors yields a measure close to NATREX.
- NATREX description:
  - Fluctuating medium- to long-term equilibrium exchange rate determined by fundamentals in absence of speculative capital flows and changes in monetary reserves.
  - Determinants include savings, productivity, foreign debt, exogenous terms of trade, and world real interest rates.
  - NATREX is not necessarily welfare-optimal and is non-stationary because its fundamentals are not stationary.
  - Medium-term intracyclical equilibrium condition: I – S + CA = 0 (I = intended investment, S = intended savings, CA = intended current account).

### Model shortcomings and caveats
- Main problems: identification and choice of short-term, medium-term, and long-term fundamentals; calibration of sustainable levels of fundamentals.
- Calibration of external balance is difficult because uncovered interest parity may not hold under external financing constraints.
- Cointegration analysis requires long samples; short time series may reflect only short- and medium-term relationships.
- The correction term from cointegration may not capture a clean medium-term reaction if the state of activity changes quickly after exogenous shocks.
- Results are tentative due to the small sample and data problems.

### Estimation results (selected diagnostics and coefficients)
- Estimation context:
  - Dependent Variable: LREER
  - Method: Least Squares
  - Date: 08/22/03 Time: 16:45
  - Sample(adjusted): 1998:3 2003:1
  - Included observations: 19 after adjusting endpoints
- Selected estimated coefficients (Variable — Coefficient — Std. Err. — t-Statistic — Prob.):
  - C — 12.27347 — 0.438909 — 27.96357 — 0.0001
  - LREER(-1) — -0.245496 — 0.049092 — -5.000715 — 0.0154
  - LTOT — -0.979524 — 0.043801 — -22.36329 — 0.0002
  - LSAVINGS — 0.013974 — 0.004839 — 2.887876 — 0.0631
  - LPRODUCTIVITY — -0.319561 — 0.016189 — -19.73952 — 0.0003
  - REAL_LIBOR01 — -0.084057 — 0.002712 — -30.9938 — 0.0001
  - LFDI — 0.003484 — 0.001957 — 1.780599 — 0.1730
  - DLTOT — 1.361993 — 0.053704 — 25.36124 — 0.0001
  - DLTOT(-1) — 1.460687 — 0.058488 — 24.97424 — 0.0001
  - DLTOT(-2) — 1.153782 — 0.04667 — 24.72193 — 0.0001
  - DLTOT(-3) — 0.365241 — 0.015666 — 3.31426 — 0.0002
  - DREAL_LIBOR01 — 0.099495 — 0.003583 — 27.7711 — 0.0001
  - DLSAVINGS(-2) — -0.036536 — 0.003485 — -10.48348 — 0.0019
  - DLPRODUCTIVITY — 0.560812 — 0.047237 — 11.87228 — 0.0013
  - DREAL_LIBOR01(-2) — 0.014337 — 0.002207 — 6.497198 — 0.0074
  - SQ2 — -0.077514 — 0.003659 — -21.18219 — 0.0002
- Model fit and diagnostics:
  - R-squared 0.999692
  - Adjusted R-squared 0.998152
  - S.E. of regression 0.002560
  - Sum squared resid 1.97E-05
  - Log likelihood 103.9611
  - F-statistic 649.0900
  - Prob(F-statistic) 0.000088
  - Durbin-Watson stat 1.602860
- Residual diagnostics (selected):
  - Normality (Jarque-Bera) — Jarque-Bera 1.383482 — Probability 0.500703
  - ARCH Test — F-statistic 0.983238 — Probability 0.398504; Obs*R-squared 2.093767 — Probability 0.351030
  - Breusch-Godfrey Serial Correlation LM Test — F-statistic 0.197464 — Probability 0.846690; Obs*R-squared 5.379221 — Probability 0.067907

### Policy implications and recommendations
- Widening current account deficit appears driven by demand pressures (increased capital inflows, growing bank credit) rather than competitiveness deterioration.
- Policy recommendation: implement demand management policies, with emphasis on a cautious fiscal stance.

*Source: _wp0437 - 2003. This suggests that there is no short-run or significant long-run misalignment,*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp0437.pdf_