5. Empirical findings

k¼1

The result of the second-generation panel unit root for the option of with and without trend are presented in Table 7. The panel unit root of CIPS using one lag order due to parsimony principle. The null hypothesis of non-stationary is fail to rejected in level in lag 1 (q = 1) for both variables option without trend (Zα) and with trend (Zτ), but rejected in first difference indicating that all series are integrated of order one or I(1) for the panel of ASEAN-5 countries.

Nonlinear Effect of Financial Development and Foreign Direct Investment in Integration… DOI: http://dx.doi.org/10.5772/intechopen.86104


#### Table 7.

yi,t ¼ θixi,t þ α<sup>0</sup>

ui,t ¼ γ<sup>0</sup>

causality model estimation based on the error-correction model as follows:

m k¼1

m k¼1

> m k¼1

m k¼1

m k¼1

m k¼1

where, EC is error-correction term comes from the FMOLS estimation, and m is the lag length. The short-run causality is determined by the statistical significance of the F-statistic associated with the corresponding right hand side variables. The presence or absence of long-run causality can be established by examining the

The result of the second-generation panel unit root for the option of with and without trend are presented in Table 7. The panel unit root of CIPS using one lag order due to parsimony principle. The null hypothesis of non-stationary is fail to rejected in level in lag 1 (q = 1) for both variables option without trend (Zα) and with trend (Zτ), but rejected in first difference indicating that all series are

λ11ikΔln FDIi,t�<sup>k</sup> þ ∑

λ13ikΔln RGDPPCi,t�<sup>k</sup> þ ∑

λ21ikΔln FinDevi,t�<sup>k</sup> þ ∑

λ31ikΔln RGDPPCi,t�<sup>k</sup> þ ∑

m k¼1

λ23ikΔ ln RGDPPCi,t�<sup>k</sup> þ ∑

λ33ikΔln FDIi,t�<sup>k</sup> þ ∑

λ41ikΔlnCPIi,t�<sup>k</sup> þ ∑

λ43ikΔln RGDPPCi,t�<sup>k</sup> þ ∑

significance of the t-statistic on the coefficient ϕ, in Eqs. (12)–(15).

integrated of order one or I(1) for the panel of ASEAN-5 countries.

structure for the errors. It can be written as:

4.4 Panel vector error-correction model

m k¼1

> m k¼1

þ ∑ m k¼1

m k¼1

þ ∑ m k¼1

m k¼1

ΔlnFDIit ¼ α1<sup>i</sup> þ ∑

ΔlnFinDevit ¼ α2<sup>i</sup> þ ∑

ΔlnRGDPPCit ¼ α3<sup>i</sup> þ ∑

ΔlnCPIit ¼ α4<sup>i</sup> þ ∑

5. Empirical findings

50

þ ∑ m k¼1

> þ ∑ m k¼1

supposed to proxy for the unobserved common factors.

i

Accounting and Finance - New Perspectives on Banking, Financial Statements and Reporting

In CS-ARDL, Eq. (9), the errors (u) is postulated a common unobserved factor

CS-ARDL is an augmented model from generic ARDL (p,q) by averaging crosssectional of the dependent and explanatory variables, as well as their lags, which are

The panel Granger causality in the framework of the panel VECM is employed to analyze the direction of the causal effect among FDI, financial development and the control variables, CPI and GDP per capita. The long-run model specified in Eq. (1) is estimated by using FMOLS to obtain the estimated residual, followed by Granger

λ12ikΔln FinDevi,t�<sup>k</sup>

λ22ikΔln FDIi,t�<sup>k</sup>

m k¼1

λ42ikΔln FinDevi,t�<sup>k</sup>

λ14ikΔlnCPIi,t�<sup>k</sup> þ ϕ1iECi,t�<sup>1</sup> þ μ1it

λ24ikΔlnCPIi,t�<sup>k</sup> þ ϕ2iECi,t�<sup>1</sup> þ μ2it

λ32ikΔln FinDevi,t�<sup>k</sup>

λ34ikΔlnCPIi,t�<sup>k</sup> þ ϕ3iECi,t�<sup>1</sup> þ μ3it

λ44ikΔln FDIi,t�<sup>k</sup> þ ϕ4iECi,t�<sup>1</sup> þ μ4it

(12)

(13)

(14)

(15)

ð Þ <sup>L</sup> <sup>Δ</sup>xit <sup>þ</sup> <sup>e</sup>ui,t (10)

iFt þ εi,t (11)

Second-generation panel unit root test of CIPS.


#### Table 8.

Second-generation panel cointegration results.

Since all variables are integrated of order one, the panel cointegration test is employed to measure the existence of long-run relationship in Eq. (1). The results of second-generation of panel cointegration presented in Table 8. The panel cointegration test shows that P<sup>α</sup> and P<sup>τ</sup> test statistics reject the null hypothesis of no cointegration at 1, 5 and 10% significance level for both models using DCPS and PCDM, for the specification without trend. The P<sup>α</sup> and P<sup>τ</sup> test statistics have the highest power and are the most robust to cross-sectional correlation [21]. Thus, the evidence from the second-generation panel cointegration test supports the presence of a cointegrating relationship among FDI, financial development, price and economic development in ASEAN-5 countries.

Due to the existence of cointegration among variables in the region, the FMOLS estimator is used to estimate the long-run equilibrium. Table 9 reports that Model 1–3 estimates the linear and nonlinear relationship between financial development and FDI in the long-run, by using DCPS, LL and PCDM as a proxy for financial development, respectively. Long-run covariance estimates pre-whitening with lag 1, where the automatic bandwidth selection is based on Newey-West fixed bandwidth and Bartlett kernel. In the linear specification, the relationship between financial development and FDI are not significant in all models. However, in contrast, there is significant of nonlinear relationship between financial development and FDI. The nonlinear relationship between these variables is anti-Kuznets or U-shape curve, where α<sup>1</sup> the coefficient of financial development and α<sup>2</sup> the financial development square coefficient (Eq. (2)) is negative and positive,


DCPS is used as a proxy in Model 1, while PCDM in Model 2.

\*\*\*Significant at 1% level. \*

denotes significant at 10% level.

#### Table 9.

Panel FMOLS long-run estimation (dependent variable: FDI).

respectively. The U-shape curve indicating that the financial development exceeded the threshold level, its incremental will attract more FDI inflows.

The result show that the negative effect of low level of financial development below the threshold level, in general at 70% of GDP that portrays the financial illness in host country. Initially, the negative relationship associated with the underdeveloped financial sector that may discourage the investor's decision to invest to the host country for those investors who are preferring on resource-seeking and market-seeking. Low level in credit market will reduce the purchasing power of parity among the citizens, and as a result, the innovative products produced by foreign firms may become unmarketable or over-supplied in the host country.

In the other hand, the high financial development reflects high financial strength that might attracts the inflows of FDI that related to assist them to set-up new business in host country and survival in their day-to-day business. When the level of financial development that above the 70% of GDP threshold point, it influences the positive impact on FDI inflows. Specifically, based on the quadratic specification of Model 1 in Table 7, the financial development threshold point is 73% of GDP (7.98/(0.93(2))), and Model 3 is 94.48% of GDP (8.46/(0.93(2))). The result showed that the DCPS surpassed the threshold point at median value to accelerate inflows of FDI, while PCDM should beyond the 75% quantile. The nonlinear relationship between financial development and FDI in this study is however differed from those of previous studies which examined the linear relationship (see [25–27]). The U-shape curve commensurate with the argument that well-developed financial market benefited FDI in host country [28].

threshold point at median level, it will attract the FDI inflows to ASEAN-5 countries. Specifically, by considering the cross-sectional dependency among ASEAN-5 countries the liquid liabilities threshold point is 70% of GDP (29.34/(3.46(2))) in

DCPS is used as a proxy in Model 1, while LL and PCDM in Model 2 and Model 3, respectively.

Cross-sectional dependence ARDL (CS-ARDL) estimation (dependent variable: FDI).

FinDev 0.61 0.32 0.11 10.36 3.42 9.20 FinDev<sup>2</sup> ——— 1.34 0.54 1.27 RGDPPC 7.11\*\*\* 5.39\*\*\* 5.86\*\*\* 10.02 5.73 4.89 CPI 2.69 2.85 2.98\* 2.16 1.88 1.76

Nonlinear Effect of Financial Development and Foreign Direct Investment in Integration…

FDI 0.96\*\*\* 1.00\*\*\* 0.96\*\*\* 1.04\*\*\* 1.12\*\*\* 1.16\*\*\* FinDev 0.05 0.07 0.53 1.33 29.34\*\*\* 4.16 FinDev<sup>2</sup> ——— 0.21 3.46\*\*\* 0.38 RGDPPC 4.66\*\* 2.37 3.93\*\* 8.90 4.92 6.55 CPI 1.85 2.03 1.07 2.00 0.66 0.73\*\* R2 0.64 0.65 0.65 0.74 0.72 0.76 CD statistic 3.91 3.86 3.60 3.55 3.91 3.80 p-value 0.00 0.00 0.00 0.00 0.00 0.00 RMSE 0.37 0.36 0.36 0.33 0.34 0.31 Threshold value (% of GDP) — 69.66 —

Linear model Quadratic model Model 1 Model 2 Model 3 Model 1 Model 2 Model 3

Although the more financial development can attract the FDI inflows, but the quality of financial reporting is important channel of its information for investors, since the ASEAN-5 countries are committed in complying IFRS rules. The information in financial reporting provided by firms and financial institution in host countries as a canal on presenting it financial position that leads better decision for foreign investors. The foreign investors may unable to make decision of lacking information on financial condition in host countries. A superior financial reporting system lowers the cost of capital and improves capital allocation efficiency [29]. The quality of financial reporting would lead to transparent and clear information that reducing asymmetric information between foreign investors and financial institu-

In further investigation, the causality between variables are tested by using Granger causality test based on VECM model as shown in Table 11. The lag length is based on Akaike information criterion. All models shown the negatively significant of error-correction term for the FDI equation, that suggesting the existence of longrun relationship when the FDI is dependent variable. Similarly, there are exists the long-run causality when financial development and CPI as a dependent variable for all models. The causality between financial development and FDI inflows is however occurred only in the long-run. As shown in Table 11, there is a unidirectional causal effect running from real GDP per capita to DCPS and PCDM, CPI to FDI, CPI

long-run.

53

Table 10.

Short-run estimation

DOI: http://dx.doi.org/10.5772/intechopen.86104

Long-run estimation

\*\*\*denotes significant at 1% level. \*\*denotes significant at 5% level.

tions in ASEAN-5 countries.

The result from FMOLS is however not considering the cross-sectional dependency among ASEAN-5 countries. The long-run coefficients are further estimated by using CS-ARDL for robustness check as shown in Table 10. Similar with Table 9 previously, the relationship between financial development on FDI inflows are not significant by using linear specification for all models in short-run and long-run estimations as shown in Table 10. The relationship between financial development and FDI inflows is however absence in the short-run estimation. But the relationship is existed in quadratic model specification, but result shows that only Model 2 has significant U-shape relationship between LL and FDI inflows in the long-run. The relationship is significant only in the long-run due to time lag effect in materializing the benefit of FDI inflows influenced by financial development. The Ushape relationship indicates by the negative coefficient for α^1, and positive coefficient for α^<sup>2</sup> (refer to Eq. (2)), which both coefficients are significant at 1% level. It means, when the level of liquid liabilities as a percentage of GDP beyond the


Nonlinear Effect of Financial Development and Foreign Direct Investment in Integration… DOI: http://dx.doi.org/10.5772/intechopen.86104

DCPS is used as a proxy in Model 1, while LL and PCDM in Model 2 and Model 3, respectively.

\*\*\*denotes significant at 1% level. \*\*denotes significant at 5% level.
