Economical determinants of domestic

5.3 The ECM Estimation
The short-run dynamics of domestic investment function in Jordan was also estimated
using the ARDL approach to co-integration of Peseran et al. (2006). The results of ECM
estimation based on the ARDL technique for domestic investment model along with
diagnostic tests are reported in Table (4).
The diagnostic statistics in Table (4) indicated that the equation was well specified.
None of the statistics shown in the table were significant at the 5% significance level. The
model fulfilled the conditions of non-autocorrelated, homoskedasticity and normal
distribution; i.e. the χ
2
tests showed no evidence on residual serial correlation, while the
Ramsey’s RESET tests showed no functional form of misspecification. Furthermore, the χ
2
tests did not indicate any evidence of normality problem or heteroscedasticity of residual.
Indeed the Adjusted R
2
is 0.836 suggesting that the error correction models (ECM) fitted the
data reasonably well.
As shown in Table 4, the estimated values of the lagged error-correction term (ECM
-1
)
based on the ARDL method is -0.5702 and statistically significant, which suggest that the
ECM tends to cause domestic investment to converge monotonically to its long-run
equilibrium path in relation to changes in the exogenous “forcing variables”. Again the
statistically significant and the correct sign of ECM
-1
coefficients confirm the presence of
long-run equilibrium between domestic investment and its determinants.
The empirical results also showed that the short-run movement in most of the
variables of domestic investment equation had the correct signs and were statistically
significant, suggesting the existence of long term relationship between the model’s variables.
The coefficients of exports (0.3517) and the growth rate of GDP (0.255) carry positive signs
and are noticeably larger than most of the other variables in ECM. This result indicates that if
the growth rates of both real GDP and export are sustainable during the next years, as in the
period 2006-2009, the growth rates of domestic investment will be better than during the pre-
2000’s period. Since the growth rate of both GDP and exports were very small during the
decades of the 1980’s and1990’s, compared to those of the 2000’s, this could be the reason
why the growth rate of domestic investment in Jordan was weak and ultimately slowed the
economic growth during the pre-2006s period.

---

European Scientific Journal
April edition vol. 8, No.7
ISSN: 1857 – 7881 (Print)
e -
ISSN 1857- 7431
12
The empirical results suggest that the inflow of FDI have a “crowd in” effect on
domestic investment, and that there was complementary relationship between FDI and
domestic investment in Jordan. Another important point is that FDI had a smaller impact on
domestic investment stimulation. For example, this result suggests that a 1 million JD increase
in FDI inflow to Jordan could result in an increase in domestic investment by only 90
thousand JD. Also, the results showed that the availability of domestic credit had a short-run
effect on domestic investment while financial intermediation, and human capital showed no
clear effect on domestic investment in the short-term.
These results are consistent with the previous findings in that the growth of real GDP
stimulates domestic investment (see: De Long and Summers, 1992; Chaudhari and Wilson,
2000; Ghura and Goodwin, 2000; Podrecca and Carmeci, 2001). Booth (1999) argued that
rapid growth leads to high rates of investment and vice versa. Numerous studies including
Carkovic and Levine (2002), Marchant, et. al., (2002), Agrawal (2000) and Graham and
Krugman (1991) found that the increase in the FDI inflows were associated with a many-fold
increase in investment by national investors. For example, Borenszteinet. al. (1998)
investigated the effect of FDI on domestic investment. His results were supportive of “a
crowding in” effect, and he found that a 1 US$ increase in the net inflow of FDI is associated
with an increase in total investment in the host economy by more than 1 US$. Borensztein
interpreted his finding by the advanced technology and management skills carried by FDI
transfer to domestic investment. Our study confirmed that the expansion of the export of
goods and services inspired domestic investment (see: Jansen, 1995; Cuvers, 1996).
Table 4. Error Correction Models based on the ARDL approach:
Short-Run Estimations for domestic investment Model
RegressorsDependent Variable Domestic Investment
SBC Selected model
Coefficients T-ratio [P-value]
ECT
-0.5702
-4.0464[0.001]***
dGr
0.2551 4.3776[0.000]***
dX 0.3517 2.9941[0.008]**
dFI 0.2847
1.2162[0.241]
dFDI 0.0981 3.6062[0.002]**
dH 0.3026 0.8986 [0.381]
dCR
0.0199 2.1523 [.042]**

---

European Scientific Journal
April edition vol. 8, No.7
ISSN: 1857 – 7881 (Print)
e -
ISSN 1857- 7431
13
C
-1.3260
-2.0971 [0.051]**
R
2
0.8365
Diagnostic Tests:
[p-value]