inflation transmission mechanism of inflation in ecuador after
TRANSCRIPT
INFLATION TRANSMISSION MECHANISM IN ECUADOR AFTER DOLLARIZATIONEmilio José CalleMonetary Economics Final ProjectJohns Hopkins AAPApril 2017
CONTENTS• Background• Dollarization• Economic Concept• Hypothesis• Analysis• Conclusion
BACKGROUND• In the year 1999 Ecuador was facing high rates of
inflation that prompted the country to “dollarize” the economy
• Dollarization implies going beyond a peg or money convertibility: the whole national currency is dropped in favor of the Dollar
• This was done without coordination nor asking for authorization to the United States, it was unilateral on Ecuador’s part
ECONOMIC CONCEPTS• Under dollarization, the exchange rate is exogenous
without any possibility of intervention on the part of the dollarized country’s government
• Fixed exchange rates pins down the country’s interest rate via Uncovered Interest Parity (UIP) – forcing i to equal the foreign interest rate i*(world) – in turn, the level of i determines the level of the money supply M necessary to meet money demand.
• The country “imports” the inflation from the other country (US)
ECONOMIC CONCEPTS
HYPOTHESIS• The inflation in Ecuador should converge with the US• Oil is the main source of “fresh” money in the country
now that the country cannot print money• The inflation transmission mechanisms of Inflation in
Ecuador have to be the same as in the US
ANALYSISRegressions, ARMA, VAR
Methodology
• This study will start by dividing the monetary behavior of Ecuador into two parts: the one covering the period pre-dollarization between 1980 and 1999 (which coincides with the period after the last military dictatorship of Ecuador and the transition to democracy); and the period covered between 2000 to 2016 which covers the dollarization situation.
Data Generating Process• The data generating process of this study has two shapes. In the first one,
taking data from FRED and from the Central Bank of Ecuador it was possible to obtain month by month data on inflation from Ecuador and the United States, plus the price of oil, to form a large enough observation set to perform analysis like ARMA and correlation between the variables.
• The second set of data is annual, and covers the following additional variables from Ecuador: unemployment rate, real interest rate, terms of trade, public debt and monetary base
• Finally all variables are transformed into log-linear variables to regress the change in them rather than the nominal values
RESULTSInflation with US, Model to Explain Ecuadorian Inflation
Convergence of US and Ecuadorian Inflation
0
10
20
30
40
50
60
70
80
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998
Inflation EcuadorInflation USAPrice Oil
-20
0
20
40
60
80
100
120
2000 2002 2004 2006 2008 2010 2012 2014 2016
Inflation EcuadorInflation USAPrice Oil
Convergence of US GDP and Ecuadorian GDP Trend
1979
-01-01
1980-01
-01
1981
-01-01
1982
-01-01
1983
-01-01
1984-0
1-01
1985
-01-01
1986
-01-01
1987
-01-01
1988-01
-01
1989-01
-01
1990
-01-01
1991
-01-01
1992
-01-01
1993-01
-01
1994
-01-01
1995
-01-01
1996
-01-01
1997-01
-01
1998
-01-01
1999
-01-01
2000
-01-01
2001
-01-01
2002
-01-01
2003
-01-01
2004-0
1-01
2005
-01-01
2006
-01-01
2007
-01-01
2008
-01-01
2009
-01-01
2010
-01-01
2011-01
-01
2012
-01-01
2013
-01-01
2014
-01-01
2015-0
1-01
2016
-01-01
0.0
2000.0
4000.0
6000.0
8000.0
10000.0
12000.0
14000.0
16000.0
18000.0
20000.0
0.000000
20000000000.000000
40000000000.000000
60000000000.000000
80000000000.000000
100000000000.000000
120000000000.000000
GDP COMPARISON USA-ECUADOR
GDP USA GDP ECUADOR
Closer Look Reveals Lack of Convergence
-3
-2
-1
0
1
2
3
00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16
NEWLOGECUADORNEWLOGPOILNEWLOGUSA
Confirmed by Regressions
Model for the Ecuadorian Inflation
ΔInflationEcuador = ΔInflation USA+ΔPublic Debt + ΔGDP Ecuador + ΔMonetary Base Ecuador + ΔUnemployment + Δ Terms of Trade + Δ Real Interest Rate + Δ Price of Oil
Model for the Ecuadorian InflationDependent Variable: LN_INF_EC Method: Least Squares Date: 04/06/17 Time: 21:38 Sample: 1980 1999 Included observations: 20
Variable Coefficient Std. Error t-Statistic Prob.
LN_GDP 1.803100 1.898116 0.949942 0.3625LN_INF_USA 0.740578 0.912610 0.811495 0.4343
LN_MB -1.107301 0.627307 -1.765166 0.1052LN_OIL -0.428684 1.116093 -0.384093 0.7082LN_PD 0.166753 0.749092 0.222607 0.8279LN_RIR 0.191633 0.163953 1.168833 0.2672
LN_TOFT 0.117223 0.252176 0.464844 0.6511LN_U 0.738546 0.897529 0.822866 0.4281
C -35.63916 46.36398 -0.768682 0.4583
R-squared 0.744245 Mean dependent var 3.494548Adjusted R-squared 0.558242 S.D. dependent var 0.476429S.E. of regression 0.316658 Akaike info criterion 0.840174Sum squared resid 1.102995 Schwarz criterion 1.288254Log likelihood 0.598256 Hannan-Quinn criter. 0.927644F-statistic 4.001241 Durbin-Watson stat 1.596951Prob(F-statistic) 0.018492
Dependent Variable: LN_INF_EC Method: Least Squares Date: 04/06/17 Time: 21:24 Sample (adjusted): 2000 2015 Included observations: 16 after adjustments
Variable Coefficient Std. Error t-Statistic Prob.
LN_MB -11.66994 7.299544 -1.598722 0.1539LN_PD 0.258321 0.146329 1.765350 0.1209LN_PIB 9.066064 6.998477 1.295434 0.2363
LN_POIL -0.769176 0.882327 -0.871759 0.4122LN_RIR -0.172561 0.326041 -0.529263 0.6130
LN_TOFT 0.942149 0.551581 1.708089 0.1314LN_UM -0.096181 1.033973 -0.093021 0.9285LN_USA -0.328007 0.355385 -0.922962 0.3867
C -120.4652 101.8660 -1.182585 0.2756
R-squared 0.912883 Mean dependent var 1.777165Adjusted R-squared 0.813322 S.D. dependent var 1.036236S.E. of regression 0.447719 Akaike info criterion 1.529020Sum squared resid 1.403167 Schwarz criterion 1.963602Log likelihood -3.232163 Hannan-Quinn criter. 1.551275F-statistic 9.169016 Durbin-Watson stat 1.938770Prob(F-statistic) 0.004243
Monetary Base was Statistically significant
Public Debt and Terms of Trade have become closer to being Statistically significant
Model for the Ecuadorian Inflation
Low Predictability before Dollarization High Predictability after Dollarization
-.4
-.2
.0
.2
.4
.6
2.4
2.8
3.2
3.6
4.0
4.4
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998
Residual Actual Fitted
-.8
-.4
.0
.4
.8
0
1
2
3
4
5
00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15
Residual Actual Fitted
Model for the Ecuadorian Inflation
Low Persistence and Autoregression before dollarization
Higher Persistence and Autoregression afterdollarization
Automatic ARIMA ForecastingSelected dependent variable: LN_INF_ECDate: 04/05/17 Time: 21:03Sample: 1980M01 1999M12Included observations: 240Forecast length: 0
Number of estimated ARMA models: 25Number of non-converged estimations: 0Selected ARMA model: (2,1)(0,0)AIC value: 0.594444013121
Automatic ARIMA ForecastingSelected dependent variable: D(LN_INF_EQ)Date: 04/05/17 Time: 21:08Sample: 2000M01 2016M12Included observations: 203Forecast length: 0
Number of estimated ARMA models: 25Number of non-converged estimations: 0Selected ARMA model: (3,2)(0,0)AIC value: 0.374247145334
VAR Analysis to find the current Transmission Mechanism 1980-1999
-4.0E-13
0.0E+00
4.0E-13
8.0E-13
1.2E-12
1 2 3 4 5 6 7 8 9 10
Response of LN_INF_EC to LN_INF_EC
-4.0E-13
0.0E+00
4.0E-13
8.0E-13
1.2E-12
1 2 3 4 5 6 7 8 9 10
Response of LN_INF_EC to LN_MB
-4.0E-13
0.0E+00
4.0E-13
8.0E-13
1.2E-12
1 2 3 4 5 6 7 8 9 10
Response of LN_INF_EC to LN_PD
-4.0E-13
0.0E+00
4.0E-13
8.0E-13
1.2E-12
1 2 3 4 5 6 7 8 9 10
Response of LN_INF_EC to LN_TOFT
-5.0E-13
0.0E+00
5.0E-13
1E-12
1.5E-12
2.0E-12
1 2 3 4 5 6 7 8 9 10
Response of LN_MB to LN_INF_EC
-5.0E-13
0.0E+00
5.0E-13
1E-12
1.5E-12
2.0E-12
1 2 3 4 5 6 7 8 9 10
Response of LN_MB to LN_MB
-5.0E-13
0.0E+00
5.0E-13
1E-12
1.5E-12
2.0E-12
1 2 3 4 5 6 7 8 9 10
Response of LN_MB to LN_PD
-5.0E-13
0.0E+00
5.0E-13
1E-12
1.5E-12
2.0E-12
1 2 3 4 5 6 7 8 9 10
Response of LN_MB to LN_TOFT
-4.0E-13
0.0E+00
4.0E-13
8.0E-13
1.2E-12
1 2 3 4 5 6 7 8 9 10
Response of LN_PD to LN_INF_EC
-4.0E-13
0.0E+00
4.0E-13
8.0E-13
1.2E-12
1 2 3 4 5 6 7 8 9 10
Response of LN_PD to LN_MB
-4.0E-13
0.0E+00
4.0E-13
8.0E-13
1.2E-12
1 2 3 4 5 6 7 8 9 10
Response of LN_PD to LN_PD
-4.0E-13
0.0E+00
4.0E-13
8.0E-13
1.2E-12
1 2 3 4 5 6 7 8 9 10
Response of LN_PD to LN_TOFT
-4.0E-12
0.0E+00
4.0E-12
8.0E-12
1.2E-11
1 2 3 4 5 6 7 8 9 10
Response of LN_TOFT to LN_INF_EC
-4.0E-12
0.0E+00
4.0E-12
8.0E-12
1.2E-11
1 2 3 4 5 6 7 8 9 10
Response of LN_TOFT to LN_MB
-4.0E-12
0.0E+00
4.0E-12
8.0E-12
1.2E-11
1 2 3 4 5 6 7 8 9 10
Response of LN_T OFT to LN_PD
-4.0E-12
0.0E+00
4.0E-12
8.0E-12
1.2E-11
1 2 3 4 5 6 7 8 9 10
Response of LN_TOFT to LN_TOFT
Response to Generalized One S.D. Innov ations ± 2 S.E.
VAR Analysis to find the current Transmission Mechanism 2000-2016
-2.0E-12
-1E-12
0.0E+00
1E-12
2.0E-12
1 2 3 4 5 6 7 8 9 10
Response of LN_INF_EC to LN_INF_EC
-2.0E-12
-1E-12
0.0E+00
1E-12
2.0E-12
1 2 3 4 5 6 7 8 9 10
Response of LN_INF_EC to LN_MB
-2.0E-12
-1E-12
0.0E+00
1E-12
2.0E-12
1 2 3 4 5 6 7 8 9 10
Response of LN_INF_EC to LN_PD
-2.0E-12
-1E-12
0.0E+00
1E-12
2.0E-12
1 2 3 4 5 6 7 8 9 10
Response of LN_INF_EC to LN_TOFT
-4.0E-11
-2.0E-11
0.0E+00
2.0E-11
4.0E-11
1 2 3 4 5 6 7 8 9 10
Response of LN_MB to LN_INF_EC
-4.0E-11
-2.0E-11
0.0E+00
2.0E-11
4.0E-11
1 2 3 4 5 6 7 8 9 10
Response of LN_MB to LN_MB
-4.0E-11
-2.0E-11
0.0E+00
2.0E-11
4.0E-11
1 2 3 4 5 6 7 8 9 10
Response of LN_MB to LN_PD
-4.0E-11
-2.0E-11
0.0E+00
2.0E-11
4.0E-11
1 2 3 4 5 6 7 8 9 10
Response of LN_MB to LN_TOFT
-6.0E-12
-4.0E-12
-2.0E-12
0.0E+00
2.0E-12
4.0E-12
6.0E-12
1 2 3 4 5 6 7 8 9 10
Response of LN_PD to LN_INF_EC
-6.0E-12
-4.0E-12
-2.0E-12
0.0E+00
2.0E-12
4.0E-12
6.0E-12
1 2 3 4 5 6 7 8 9 10
Response of LN_PD to LN_MB
-6.0E-12
-4.0E-12
-2.0E-12
0.0E+00
2.0E-12
4.0E-12
6.0E-12
1 2 3 4 5 6 7 8 9 10
Response of LN_PD to LN_PD
-6.0E-12
-4.0E-12
-2.0E-12
0.0E+00
2.0E-12
4.0E-12
6.0E-12
1 2 3 4 5 6 7 8 9 10
Response of LN_PD to LN_TOFT
-2.0E-10
-1.0E-10
0.0E+00
1.0E-10
2.0E-10
1 2 3 4 5 6 7 8 9 10
Response of LN_TOFT to LN_INF_EC
-2.0E-10
-1.0E-10
0.0E+00
1.0E-10
2.0E-10
1 2 3 4 5 6 7 8 9 10
Response of LN_TOFT to LN_MB
-2.0E-10
-1.0E-10
0.0E+00
1.0E-10
2.0E-10
1 2 3 4 5 6 7 8 9 10
Response of LN_TOFT to LN_PD
-2.0E-10
-1.0E-10
0.0E+00
1.0E-10
2.0E-10
1 2 3 4 5 6 7 8 9 10
Response of LN_TOFT to LN_TOFT
Response to Generalized One S.D. Innov ations ± 2 S.E.
ConclusionsAnswers to Hypothesis
Ho1: The inflation in Ecuador should converge with US
• The data shows this has not happened and there is no tendency that indicates that it will.
• Although the US inflation becomes much more significant in explaining Ecuadorian inflation after dollarization, this could have more to do with Ecuador being forced into having OECD-country like inflation instead of the previous scenarios with high volatility, than with both inflations converging
• Ecuador’s inflation has dropped but it most likely approaches a worldwide rate of inflation rather than a US only one
• Hypothesis Rejected
Ho2: Oil as main source of liquidity• The data shows this is not true, contrary to the consideration of Ecuador being
a Oil-based economy• The Prior to this hypothesis is that Ecuador was oil-neutral: as Ecuador does
not produce enough oil derivatives to satisfy internal demand, a higher price of crude oil only neutralizes higher prices of imports
• The results prove this prior right: oil is NOT statistically significant to Ecuador’s inflation either before or after dollarization
• Hypothesis rejected
Ho3: The inflation transmission mechanisms of Inflation in Ecuador have to be the same as in the US
• As the US and Ecuadorian inflations have not and are not converging, then this hypothesis has to be false
• But then what are the mechanisms explaining Ecuador’s inflation?• From the VAR analysis and the Impulse functions, the answer is shown in the
next slide• Hypothesis rejected
Inflation Transmission Mechanism in Ecuador after dollarization
FINAL CONCLUSION• Ecuador has switched from Seignorage to Borrowing for its liquidity needs• As so, when Monetary Base increases now it actually lowers inflation as it’s an
endogeneous growth without borrowing nor paying interests (through transfers and remittances for example)
• As there are no exchange rates, Terms of Trade (relative prices between Ecuador and the world) are the only, exogeneous variable that determine import prices and their influence on inflation (price levels) in Ecuador
• Both Interest Rates and Terms of Trade are exogeneous in this model
