Understanding exchange rate pass-through in Vietnam

Figures & data

Figure 1. Consumer price index and exchange rate in Vietnam from January 2009 to May 2020. Source: International Financial Statistics (IFS).

Table 1. Data sources

Variable	Definition	Source
CPI	Consumer price index	IFS
ER	Bilateral exchange rate between Vietnam and the United States	IFS
Oil	Oil prices	Mundi index
OG	Output gap	IFS
M2	Money supply of Vietnam	SBV

Table 2A. Unit root test

Variable	ADF		PP		KPSS
	Level	1st difference	Level	1st difference	Level	1st difference
CPI	−2.52	−5.18***	−3.31**	−5.18***	1.31	0.71***
ER	−2.55	−12.01***	−2.809*	−12.01***	1.25	0.41**
M2	−2.43	−9.65***	−2.69*	−9.49***	1.46	0.45**
Oil	−1.57	−8.16***	−1.35	−7.72***	0.61***	0.32*
OG	−5.78***	−11.28***	−5.79***	−24.2***	0.05*	0.26*

Note: ***, **, and * indicate statistical significance at the 1%, 5%, and 10% levels, respectively.

Table 2B. ADF breakpoint unit root test (innovational outlier and additive outlier)

Variable	Innovational outlier		Additive outlier
	Level	1st difference	Level	1st difference
CPI	−6.71***	−7.02***	−5.32***	−7.25***
ER	−4.17	−19.37***	−4.09	−19.37***
M2	−4.14	−11.53***	−1.91	−11.64***
Oil	−4.63**	−8.60***	−4.62**	−8.93***
OG	−6.59***	−14.00***	−6.64***	−14.11***

Note: *** and ** indicate statistical significance at 1% and 5% respectively, using Vogelsang (1993) asymptotic one-sided p-values.

Table 3. Optimal lag length

Optimal lag	LR	FPE	AIC	SC	HQ
2	6	2	2	1	1

Table 4. ARDL model results. Number of observations: 135 after adjustments

Part 1: Short-run estimation ARDL (2, 1, 1, 2, 1, 0, 0)
Lag	0	1	2
$\alpha$	−0.763***
$\mathrm{\Delta }CPI$		1.333***	−0.405***
$\mathrm{\Delta }\mathrm{E}\mathrm{R}$	0.097***	−0.058**
$\mathrm{\Delta }M2$	−0.055**	0.064***
$\mathrm{\Delta }Oil$	0.016***	−0.001	−0.012***
$\mathrm{\Delta }OG$	−0.001***	0.015***
$\mathrm{\Delta }D1$	0.001
$\mathrm{\Delta }D2$	0.003*
Part 2: Long-run estimation
$\alpha$	−10.6***
ER	1.347***
M2	0.127***
Oil	0.041*
OG	0.071
D1	0.022
D2	0.042*

Note: ***, **, and * indicate statistical significance at the 1%, 5%, and 10% levels, respectively.

Table 5. NARDL model results

Part 1: Short-run estimation NARDL (2,0,0,1,2,1,0,0)
Lag	0	1	2
C	0.125*
$\mathrm{\Delta }CPI$		1.322***	−0.387***
$\mathrm{\Delta }POS$	0.093***
$\mathrm{\Delta }NEG$	0.583*
$\mathrm{\Delta }M2$	−0.051**	0.061***
$\mathrm{\Delta }Oil$	0.017***	−0.001	−0.01***
$\mathrm{\Delta }OG$	−0.01***	0.015***
$\mathrm{\Delta }D1$	−0.0006
$\mathrm{\Delta }D2$	−0.0002
Part 2: Long-run estimation
C	1.934**
POS	1.432***
NEG	9.004
M2	0.163***
Oil	0.07**
OG	0.087*
D1	−0.01
D2	−0.003

Note: ***, **, and * indicate statistical significance at the 1%, 5%, and 10% levels, respectively.

Table 6. Diagnostic statistics

F-bound Test	ECMt-1	Serial Correlation	Heteroscedasticity	CUSUM	CUSUMQ
Part 1: Diagnostic Statistics in ARDL Model
8.93***	−0.07***	3.13	11.01	Good	Good
Part 2: Diagnostic Statistics in NARDL Model
8.36***	−0.6***	3.06	12.57	Good	Good

Note: *** indicates significance at 1%.

Table 7. The existence of the threshold model for exchange rate and money supply

	Exchange rate	Money supply	Oil	Output Gap
LM-test for no threshold:	9.94	15.78	64.57	37.42
Bootstrap P-Value:	0.00	0.00	0.00	0.00

Table 8. Results of the threshold model with the threshold exchange rate and money supply variables

Variable	Order	Threshold value	Time of threshold
Exchange rate	1	9.938	03/2011
	2	10.006	12/2016
Money supply	1	14.7505	03/2011
	2	15.6105	12/2015
Oil	1	4.3051	10/2009
	2	4.4371	11/2010
Output Gap	1	0.0133	7/2009
	2	−0.0457	8/2014

Figure A1. CUSUM and CUSUMQ in ARDL model.

Figure A2. CUSUM and CUSUMQ in NARDL model. As the CUSUM and CUSUMQ lines were within the 5% significance level, the outputs signified that the parameters were stable.

Vogelsang, T. J. (1993). Essays on testing for nonstationarities and structural change in time series models. Princeton University. Ph.D. Dissertation.

 Google Scholar