Research Results on the Relationship Between Exchange Rate Level and FDI Capital in Vietnam and Discussion

Next, the thesis conducts a stability test of model 1. The test results show that the solutions are all within the unit circle, so the VAR model analyzing the relationship between the exchange rate level and FDI capital with lag 4 is stable (Figure 4.5).

Thus, after completing all the steps of selecting and analyzing the VAR model, the thesis has tested the stability of the model and tested the autocorrelation phenomenon of the residuals of model 1 to confirm the appropriateness of the selected model. The results show that the VAR model analyzing the relationship between the exchange rate level and FDI capital using the first-order difference data series with the optimal lag of 4 is stable and consistent with statistical rules.

Then, the thesis continues the steps of model testing and analysis including: Granger causality test, variance decomposition analysis and analysis of push-pull reaction effects between variables in model 1. Detailed results of VAR model testing coefficient, Granger causality test, variance decomposition analysis and analysis of push-pull reaction effects between variables in model 1 will be reported in section 4.2.2 of this thesis.

4.2.2. Research results on the relationship between exchange rate level and FDI capital in Vietnam and discussion

4.2.2.1. VAR test coefficient of model 1

The VAR model studying the relationship between exchange rate level and FDI capital in Vietnam with an optimal lag of 4 has the following test results:

Table 4.6. Results of testing the relationship between REER and FDI capital

Dependent variable

Independent variable	DFDI			DREER
Independent variable	Delay	Coefficient	P-value	Coefficient	P-value
DFDI	L1.	-0.8986	0.000 ***	-0.0001	0.807
	L2.	-0.6212	0.000 ***	-0.0005	0.545
	L3.	-0.4051	0.012 **	0.0005	0.520
	L4.	0.0575	0.690	0.0022	0.002 ***
DREER	L1.	10,9983	0.676	0.6017	0.000 ***
	L2.	-2.1109	0.940	-0.0527	0.697
	L3.	-7.6367	0.781	-0.3595	0.007 ***

Maybe you are interested!

L4.	76.6161	0.004 ***	0.1183	0.366
DOPEN	L1.	272,7190	0.302	1.0457	0.416
	L2.	734.3128	0.003 ***	0.0415	0.973
	L3.	861.9655	0.001 ***	2,5798	0.047 **
	L4.	394.3566	0.066 *	1,8283	0.079 *
DGROWTH	L1.	-157.7604	0.141	1,6670	0.001 ***
	L2.	21,0944	0.851	1.5679	0.004 ****
	L3.	122.8112	0.259	0.4945	0.350
	L4.	120,2879	0.351	0.1033	0.869
DDUMMY	L1.	569.5420	0.099 *	0.6986	0.678
	L2.	84,0419	0.794	-2.6964	0.085 *
	L3.	-303.7772	0.329	0.1711	0.910
	L4.	242,197	0.451	4,3302	0.006 ***
	Cons.	124,8016	0.086 *	0.5026	0.155

Note: * , ** and *** mean significant at 10%, 5% and 1% levels respectively

Source: Author's analysis

The results of testing the relationship between the level of the real multilateral exchange rate and FDI capital show that there is a bidirectional impact between the two variables. Regarding the impact of the level of the real multilateral exchange rate on FDI, the research results show that REER at lag 4 has a positive impact on FDI capital in Vietnam with a statistical significance level of 1%. The research has not found a statistically significant impact of REER at lags 1, 2 and 3 on FDI capital in Vietnam. In the opposite direction, the research shows that REER is positively impacted by FDI after 4 quarters with a statistical significance level of 1%. The research has not found a statistically significant impact of FDI at lags 1, 2 and 3 on REER of Vietnam.

The research results show that trade openness at lags 2, 3 and 4 has a positive impact on FDI capital in Vietnam. At the same time, trade openness at lags 3 and 4 has a statistically significant positive impact on Vietnam's REER.

The study has not found a statistically significant impact of economic growth on FDI inflows to Vietnam. For REER, economic growth at lag 1 is found to have a positive impact on REER at lag 1 and a negative impact on REER at lag 2. The remaining lags

The GROWTH variable in the model has not found a statistically significant impact on Vietnam's FDI capital and REER.

The global financial crisis factor at lag 1 was found to have a positive impact on FDI inflows to Vietnam. At the same time, the study found that the global financial crisis at lag 2 had a negative impact on REER and at lag 4 had a positive impact on REER.

On the other hand, the research results show that the two variables of the real multilateral exchange rate and FDI are most affected by these factors in the past. Specifically, current FDI is negatively affected by fluctuations in FDI in the past (with a lag of 1 quarter to 3 quarters). The current real multilateral exchange rate is affected by this factor in the past with opposite directions at different lags. Specifically, the current REER is positively affected by fluctuations in REER at lag 1 and negatively affected by fluctuations in REER at lag 3.

4.2.2.2. Granger causality test

Table 4.7. Granger causality test results for model 1

Equation

Excluded	Branch 2	df	Pro> Chi 2
DFDI	DREER	8.51	4	0.075
DFDI	DOPEN	12.16	4	0.002
DFDI	DGROWTH	5.74	4	0.220
DFDI	DDUMMY	5.00	4	0.286
DFPDI	all	44.34	16	0.000
DREER	DFDI	18.45	4	0.001
DREER	DOPEN	9.30	4	0.054
DREER	DGROWTH	18.94	4	0.001
DREER	DDUMMY	11.87	4	0.018
DREER	all	50.57	16	0.000

Source: Author's analysis

The results of Granger causality test on the relationship between exchange rate level and FDI in Vietnam are presented in Table 4.7, showing a significant Granger causality effect from DREER to DFDI at the 10% significance level. In addition, DFDI is also affected by the impact of trade openness at the 1% significance level. On the contrary, the research results found a causal effect of

statistically significant effect from DFDI to DREER at 1% significance level. In addition, the DREER variable is Granger causally affected by the DGROWTH variable at 1% significance level and is affected by the DOPEN variable at 10% significance level. At the same time, the global financial crisis has a causal effect on DREER at 5% significance level.

4.2.2.3. Results of variance decomposition analysis of model 1

Table 4.8. Variance decomposition of the impact of variables on FDI capital

Delay

DFDI	DREER	DOPEN	DGROWTH	DDUMMY
0	0	0	0	0	0
1	1	0	0	0	0
2	0.9334	0.0078	0.0105	0.0239	0.0244
3	0.8820	0.0071	0.0138	0.0539	0.0431
4	0.8421	0.0221	0.0238	0.0705	0.0415
5	0.7962	0.0285	0.0284	0.0593	0.0875
6	0.7707	0.0299	0.0568	0.0579	0.0847
7	0.7466	0.0295	0.0552	0.0564	0.1123
8	0.7324	0.0320	0.0573	0.0622	0.1160

Source: Author's analysis results In general, the results of the variance decomposition analysis of the impact of variables on DFDI show that FDI capital is most affected by the FDI factor in the past, this impact tends to decrease as the lag gets further (93.3% at lag 2; 88.2% at lag 3; 84.2% at lag 4 and about 73.2% at lag 8 periods). FDI capital is affected by economic growth with an increase as the lag gets further (2.4% at lag 1; 5.4% at lag 2; 7% at lag 3 periods). Next, FDI capital is affected by trade openness with an increasing trend as the lag gets longer (1.0% at lag 2; 1.3% at lag 3; 2.3% at lag 4 and reaching nearly 5.7% at lag 8 periods). The impact of the DREER variable on the DFDI variable is quite slight in the optimal observed lags. At lags 2 and 3, the impact of the DREER variable on the DFDI variable is quite slight (about 0.7%), then increases from lag 4 (2.2%) and stabilizes in the remaining lags. The impact of the DREER variable on the DFDI variable is largest at lag 8, also reaching about 3.2%. The financial crisis alone has a slight impact on FDI capital and tends to increase slightly as the lag increases (2% at lag 2; 4% at lag 4), (see table 4.8).

Table 4.9. Variance decomposition of the impact of factors on REER

Delay

REER	FDI	Open	Growth	dummy
0	0	0	0	0	0
1	0.9997	0.0002	0	0	0
2	0.8974	0.0008	0.0006	0.0991	0.0020
3	0.8593	0.0238	0.0061	0.0986	0.0122
4	0.8451	0.0267	0.0170	0.0934	0.0177
5	0.7225	0.1220	0.0202	0.0767	0.0586
6	0.7030	0.1163	0.0203	0.0792	0.0812
7	0.6814	0.1156	0.0314	0.0791	0.0925
8	0.6545	0.1110	0.0318	0.1096	0.08929

Source: Author's analysis results Similarly, the results of variance decomposition analysis of the impact of variables on DREER in the research model of the relationship between exchange rate and FDI capital show that DREER is most affected by the DREER factor in the past, this impact tends to decrease as the lag gets further (89.7% at lag 2; 85.9% at lag 3; 84.5% at lag 4 and about 6.5% at lag 8 periods). Next, DREER is slightly affected by FDI capital with an increasing trend as the lag gets longer (2.4% at lag 3; 2.7% at lag 4) and increases sharply at lags from 4 to 8 periods (more than 11%). Economic growth has a strong impact on the real exchange rate at lags from 2 to 4 periods (more than 9%), then gradually decreases at lags from 5 to 7 periods (approximately 7%) and increases slightly at lags of 8 periods (nearly 11%). Trade openness has a very small impact on REER and tends to increase at longer lags. Similarly, financial crisis has an increased impact on DREER at longer lags (at lags of 3 periods, about 1% and at lags of 8 periods, about 9.0%, table 4.9).

4.2.2.4. Results of analysis of the push-pull reaction effects between variables in model 1

Impact of factors on current FDI capital (Figure 4.6)

The push response of the multilateral real exchange rate to FDI in the model studying the relationship between the exchange rate level and FDI capital shows that, when there is a shock in REER, FDI capital is stable in the first two quarters, has a slight decline in the third quarter and increases sharply in the fourth quarter. This impact gradually eases in the following quarters and stabilizes after 8 quarters. In addition, the push response shows that FDI shocks in the past (with a lag of 1 quarter and 2 quarters) will impact the decline in FDI in the present. After 2 quarters, the impacts

This effect gradually weakens and FDI gradually recovers. When there is a shock to trade openness, FDI increases slightly in the first quarter and decreases slightly in the second quarter, then continues to decrease significantly in the third quarter and stabilizes in the fourth quarter. These effects increase sharply in the fifth quarter and decrease sharply in the sixth quarter, then gradually weaken and stabilize. The push response of economic growth to FDI shows that, when there is a shock in economic growth, FDI capital into Vietnam decreases in the first quarter and increases again in the second quarter, then gradually stabilizes in the third and fourth quarters. From the fourth quarter, these effects begin another cycle of fluctuations. When there is a shock to the global financial crisis, FDI increases in the first quarter, then decreases in the second quarter and gradually recovers in the third quarter and peaks in the fourth quarter. After the fourth quarter, the shocks of the global financial crisis continue to cause fluctuations in FDI capital flows into Vietnam (Figure 4.6).

Impact of factors in model 1 on current REER (figure 4.7)

The push response of FDI to the multilateral real exchange rate in the model studying the relationship between the exchange rate level and FDI capital shows that when there is a shock in FDI capital, REER is stable in the first quarter, has a slight decline in the second quarter and increases sharply in the third quarter, peaks in the fourth quarter, and then declines in the fifth quarter. After the fifth quarter, these impacts gradually ease and stabilize after 8 quarters. In addition, the push response shows that REER shocks in the past (with a lag of 1 to 3 quarters) will impact the current REER decline. After that, these impacts weaken and REER gradually recovers.

When there is a shock to trade openness, REER increases slightly in the first quarter and decreases slightly in the second quarter, stabilizes in the third quarter and decreases sharply in the fourth quarter. In the fifth quarter, REER recovers and increases sharply in the sixth quarter, then REER shows a decrease after the sixth quarter. In addition, the push response of economic growth to REER shows that, when there is a shock in economic growth, Vietnam's REER increases in the first quarter and decreases in the second quarter, fluctuating slightly from the third quarter to the fifth quarter. From the sixth quarter, REER fluctuates more strongly than before the shock of economic growth. On the other hand, when there is a shock of the global financial crisis, REER is stable in the first quarter, then decreases in the second quarter, stabilizes in the third quarter and increases sharply in the fourth quarter. After the fourth quarter, the shocks of the global financial crisis continue to cause fluctuations in Vietnam's REER (Figure 4.7).

95% CI impulse response function (irf)

irf1, dreer, dfdi

100

-50

-100

step

Graphs by irfname, impulse variable, and response variable

The push-pull response of the multilateral real exchange rate to FDI

irf1, dopen, dfdi

1000

500

-500

-1000

95% CI

step

impulse response function (irf)

Graphs by irfname, impulse variable, and response variable

irf1, dfdi, dfdi

-.5

-1

step

95% CI

impulse response function (irf)

Graphs by irfname, impulse variable, and response variable

The push reaction of past FDI to current FDI

The push-pull response of trade openness to FDI

irf1, dgrowth, dfdi

500

-500

95% CI

step

impulse response function (irf)

Graphs by irfname, impulse variable, and response variable

irf1, ddummy, dfdi

2000

1000

-1000

-2000

95% CI

step

impulse response function (irf)

Graphs by irfname, impulse variable, and response variable

The push response of economic growth to FDI The push response of economic crisis to FDI

Figure 4.6. Push response of variables to current FDI

95% CI impulse response function (irf)

irf1, dfdi, dreer

.004

.002

-.002

step

Graphs by irfname, impulse variable, and response variable

The push response of FDI to the multilateral real exchange rate

irf1, dreer, dreer

-.5

step

95% CI

impulse response function (irf)

Graphs by irfname, impulse variable, and response variable

irf1, dopen, dreer

-2

-4

step

95% CI

impulse response function (irf)

Graphs by irfname, impulse variable, and response variable

Push-pull response of past REER to current REER Push-pull response of trade openness to REER

irf1, dgrowth, dreer

-2

step

95% CI

impulse response function (irf)

Graphs by irfname, impulse variable, and response variable

irf1, ddummy, dreer

-5

step

95% CI

impulse response function (irf)

Graphs by irfname, impulse variable, and response variable

The push reaction of economic growth to the REER The push reaction of economic crisis to the REER

Figure 4.7. Impulse response of variables in model 1 to current REER

Comment