For the purpose of this study secondary date has been used. All the data has been acquired from different publications of State Bank of Pakistan.
3.2 Sample Size
Since the data of total investment inflows to the textile sector is not available, we will use foreign direct investment in textile sector as a proxy for investment in textile sector. For this purpose monthly data of foreign direct investment in textile sector, consumer price index and interest rate from July 2003 to August 2010 has been employed.
3.3 Research Model developed
In our model there are two dependent and two independent variables. We will employ a statistical technique in order to evaluate the affects of independent variables on dependent variable.
3.4 Statistical Technique
We will employ a multiple linear regression as used by Bell (2007). Multiple linear regression is a technique for determining the linear relationship between one dependent variable and two or more independent variables. The equation of multiple linear regression of our model can be presented as below:
FDI = α + β1 (CPI) + β2 (IR) +
Where as,
FDI = foreign direct investment in textile sector.
α = the intercept of the equation.
β1 = the coefficient of consumer price index.
CPI = consumer price index as a measure of inflation rate.
β2 = the coefficient of interest rate.
IR = Weighted average lending rate of all banks.
= the error term.
Transformation of Variables
Foreign direct investment (FDI) is dependent variable in our model. Transformation is applied by taking a log of each value. After taking a log our new variable is ln_FDI
Inflation (CPI) is independent variable in our model. In order to fix the normality issue we have taken a difference of inflation. It is calculated by subtracting the current value by its previous value. Since our data is monthly, we have subtracted each month's value from its previous month's value. The new values represent our new variable which is indicated by diff1_inflation.
Lending rate is also independent variable in our model. It is transformed first by taking inverse of each value and then subtracting each month's value from its previous month's value. New values represent our new variable which is indicated by inv_diff_interest_rate in our model.
CHAPTER 4: RESULTS
Following are the results.
Table No. 1
Model Summary
R
R Square
Adjusted R Square
Std. Error of the Estimate
Durbin-Watson
Model
1
.274
.075
.052
.31003
1.705
Predictors: (Constant), inv_diff1_intrest_rate, diff1_inflation
Dependent Variable: ln_FDI
R is the correlation between observed and predicted values of dependent variable. If the value of R is close to 1 it means that the correlation is very high between observed and predicted values. Above table shows the value of R which is 0.274. It reveals that the correlation between observed and predicted values is very weak. R Square represents the proportion of variation in dependent variable explained by a model. In our model value of R Square is 0.075 which is very low. When we have two or more predictors in a model we consider Adjusted R Square's value instead of R Square. In our model Adjusted R Square is 0.052 which shows that model is not best fitted.
Table No. 2
ANOVA
Sum of Squares
df
Mean Square
F
Model
1
Regression
0.624
2
0.312
3.247
Residual
7.689
80
0.096
Total
8.314
82
a. Predictors: (Constant), inv_diff1_intrest_rate, diff1_inflation
b. Dependent Variable: ln_FDI
The sig-value of our model is 0.044. The value shows that regression model we developed is significant.
Table No.3
Coefficients
Unstandardized Coefficients
Standardized Coefficients
t
Co linearity Statistics
B
Std.Error
Beta
Tolerance
VIF
Model
1
(Constant)
1.102
0.288
3.547
diff1_inflation
-.043
0.027
-.171
-1.582
0.986
1.014
inv_diff1_intrest_rate
0.809
0.373
0.235
2.170
0.986
1.014
a. Dependent Variable: ln_FDI
Table No.3 shows that one variable is significant which is inv_diff1_intrest_rate. Variable(s) in order to be significant need to have a sig value of less than 0.05. Inflation and interest rate have sig-values of 0.118 and 0.33 respectively. We will rerun the regression to see if interest rate alone is significant in our model or not.
Table No. 4
Model Summary
R
R Square
Adjusted R Square
Std. Error of the Estimate
Durbin-Watson
Model
1
0.215
0.046
0.034
0.31289
1.653
Predictors: (Constant), inv_diff1_intrest_rate
Dependent Variable: ln_FDI
In the above table value of adjusted R square is 0.034. It shows that relationship between independent and dependent variable is very weak.
Table No.5
ANOVA
Sum of Squares
df
Mean Square
F
Model
1
Regression
.384
1
.384
3.918
Residual
7.930
81
.098
Total
8.314
82
Predictors: (Constant), inv_diff1_intrest_rate
Dependent Variable: ln_FDI
The sig-value of our model is 0.051. The value shows that regression model we developed is insignificant.
Table No.6
Coefficients
Unstandardized Coefficients
Standardized Coefficients
t
Co linearity Statistics
B
Std.Error
Beta
Tolerance
VIF
Model
1
(Constant)
.960
.288
3.337
inv_diff1_intrest_rate
.739
.374
.215
1.980
1.000
1.000
a. Dependent Variable: ln_FDI
Interest rate alone is insignificant in our model since its sig-value is 0.051. After employing the regression twice we found no empirical evidence that foreign direct investment in textile sector in affected by inflation rate and interest rate.
4.2 Hypothesis Assessment Summary
Table No.7
Hypothesis
Sig Value
Empirical Findings
H1: Negative relationship exists between foreign direct investment and interest rate.
0.051
Hypothesis not accepted.
H2: Positive relationship exists between foreign direct investment and inflation
0.118
Hypothesis not accepted
CHAPTER 5: DISCUSSION, CONCLUSION,
IMPLICATIONS AND FUTURE RESEARCH
5.1 Conclusion:
The study intended to examine the affects of inflation and interest on investment in the textile sector of Pakistan. The investigation has been conducted using linear regression technique. Since the data of total investment inflows to the textile sector is not available, foreign direct investment in textile sector has been used as a proxy for investment. After performing analysis, the results brought to the conclusion that interest rate and inflation are not significant predictor of foreign direct investment. The hypothesis that Interest rate is inversely related to FDI in textile sector is not accepted and it's concluded that there is no significant relationship between interest rate and foreign direct investment. Similarly the hypothesis that Inflation is positively related to FDI in textile sector is not accepted either and concluded that inflation has no significant relationship with foreign direct investment. These findings reveal that investment in textile sector is not affected by interest rate and inflations and they have no significant relationship with investment. It is possible that there would be some other variables which can affect investment in textile sector.
5.2 Discussions
We used FDI in textile sector as a proxy for investment in textile sector due to lack of data. Government should work to design policies in a way that it reduces the overall input cost of business so that our industries become competitive globally.
5.3 Implications and Recommendations
In a country like Pakistan where high rate of inflation exists, interest rates are of key importance to curd the inflation. Normally interest rates are increased to curb inflation but it also has a downside, it increases the cost of borrowing. As our results suggest that FDI in textile sector is not affected by interest rate and inflation, in a way it is favorable for the textile sector.
5.4 Future Research
The present study should be of significant interest to both researchers and policymakers in the arena of economic development. Certainly, the present findings are Pakistan-specific, and further work is needed to establish whether it may be generalized results for the global economy. Future research could be conducted to find the variables that effect the overall investment in textile sector since in our findings both predictors are found to be insignificant.
