Economical determinants of domestic

3.Theoretical and empirical consideration
A number of models have been employed in the literature to explain the determinants
of investment, among these models are the Neoclassical investment model and the accelerator
investment model. The Neoclassical model has been criticized for its shortcomings in
estimating investment function for developing countries. These criticisms are related to the
lack of readily available measures of capital stock and/or returns to capital (Blejer and Khan,
1984). In that regard, the study will consider the accelerator investment model for the same
reason.
Fry (1998) established a flexible accelerator model and developed by Agrawal (2000).
Fry developed an investment model in terms of the ratio of investment to GDP based on the
flexible accelerator model. The accelerator model has desired capital stock k* proportion to
real output, y:
k * =αy
(1)
This can be expressed in terms of desired ratio of net investment to output
(I /Y) *:
(I /Y)* =αγ
(2)
Where Iis gross domestic investment in current prices, Y denotes GDP in current
prices and γ is the growth rate of real GDP. A partial adjustment mechanism allows the actual

---

European Scientific Journal
April edition vol. 8, No.7
ISSN: 1857 – 7881 (Print)
e -
ISSN 1857- 7431
6
investment rate to adjust to the difference between the desired investment rate and the
investment rate in previous period:
Δ (I /Y) = λ[(I/Y) * -(I/Y)
t-1
]
orI/Y = λ(I/Y) * +(1- λ)(I/Y)
t-1
(4)
Where λ denotes the coefficient of adjustment. The flexible accelerator model allows
economic conditions to influence the adjustment coefficient λ . Specifically:
λ β
0
+ β
1
Z
1
+ β
2
Z
2
+ β
3
Z
3
+....
(I/y) *-(I/Y)
t-1
(5)
Where Z
i
are the variables (include an intercept term for constant depreciation rate)
that affect λ rate, and β
i
are their respective coefficients.
Ghura and Goodwin (2000) also employed the following empirical framework for the
analysis of the determinants of domestic investment using panel data from (31) developing
countries:
Y1 =α + βX
i
+ e
i
(6)
Where y
i
is the ratio of domestic investment to GDP, X
i
are the observable variables
representing factors affecting domestic investment, α and β are parameters to be estimated,
and e
i
is a random error term with a mean of zero.
In this line of research, most researchers have included all or a subset of the following
variables (among others) as the exogenous variables in the domestic investment equation:
FDI, financial intermediation, exports, human capital, and domestic credit availability. See
for example: Ghura and Goodwin (2000), Fry (1998), and Agrawal (2000). These studies
implicitly assumed the existence of an underlying equilibrium relationship between domestic
investment and a given set of explanatory variables. Our estimation technique differs from
these earlier studies in the way that handles the non-stationarity feature of the data.
Theoretically, most literature pointed out that all these variables contribute positively to the
growth of domestic investment in developing countries (see among others: Lucus, 1998;
Romer, 1990; Borensztein, et al., 1998; Levin and Beck, 2000; Gura and Goodwin, 2000;
Madsen, 2002). Specifically, the model used is:
GDI =β
0
+β
1
Gr+β
2
FDI+β
3
FI+β
4
H+β
5
Cr+β
6
X+t+e
(7)
Where:
GDI Denotes domestic investment (net of FDI)
Gr Denotes the growth rate of real GDP,
FDI Denotes foreign direct investment as a ratio of GDP,
X Denotes the exports of goods and services as a ratio of GDP,

---

European Scientific Journal
April edition vol. 8, No.7
ISSN: 1857 – 7881 (Print)
e -
ISSN 1857- 7431
7
FI Denotes financial intermediation as calculated by M2 as a ratio of GDP,
H Denotes human capital proxied by secondary school enrolment ratio,
Cr Denotes domestic credit availability as a ratio of GDP,
T Denotes trend.
ε
Denotes error term