Regression Analysis Of Mauritius Nigeria And South Africa Economics Essay

The intent of the empirical probe is to gauge the effects of FDI on economic growing and place factors that enhance or hinder the effects of FDI on growing. In peculiar, as discussed in chapter 2, this subdivision examines whether FDI interacts with the stock of human capital, the development of substructure, and other determiners of establishments and the investing clime measured among other things.All arrested developments are based on a panel information of 29 old ages from1980 to 2008. Section 4.1 will be carried out to analyze the dependance of economic growing on the chief factors of each of the 3 African states and subdivision 4.

2 will affect the arrested development utilizing the Hausman trial. The FE and the RE are besides tested which will be elaborated afterwards depending on the consequences of the Hausman trial. The undermentioned theoretical account is being used for the intent of appraisal:LNRGDPPCAP = I?0 + I?1 ( FDI ) + I?2 ( OPEN ) + I?3 ( HUMCAP ) + I?4 ( GOVSIZE ) +I?5 ( GOVINV ) + I?6 ( INFL ) + I?7 ( INFRAS ) + I?8 ( POP ) + Iµtwhere,GDPPCAP = existent gross domestic merchandise per capita ( in log signifier )FDI = foreign direct investing defined as ( FDI/GDP*100 )OPEN = openness of the economic system as a per centum of GDPHUMCAP = the degree of human capital ( gross secondary school registration ratio )GOVSIZE = authorities ingestion as a ratio of GDPGOVINV = GFCF as a per centum of GDPINFL = the rate of rising pricesINFRAS = substructure development as a per centum of GDPPOP = one-year population growing rate

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now

4.1 Country wise Analysis of Datas

This subdivision will include the arrested development of the 3 African states: Mauritius, Nigeria and South Africa utilizing the OLS appraisal accompanied with other relevant specification trials. Likewise the informations will be interpreted and will be compared among the states.


1.1 Mauritius Regression Analysis

Table 4.1: Mauritius Regression Results

Dependent Variable: LNReal GDP Per Capita

Appraisal from 1980 to 2008

VariablesCoefficientStd ErrorT-RatioFDI/GDP-0.00803560.008085-0.99INF-0.00934730.0018375-5.




244187721.92R2 0.9808F-Test ( 8, 20 ) 0.0000*Durbin-Watson Statisticss 1.56*Significant at 5 % degreeThe above tabulated consequence is obtained after taking into history the consecutive correlativity job. No hint of multicollinearity and autocorrelation could be found which is 2.

18 and 1.56 severally. The consequence shows that rising prices, openness to merchandise every bit good as human capital significantly explains growing in Mauritius.

FDI is negatively correlated with growing as in the early yearss Mauritius were non able to pull FDI and as such it had an inauspicious consequence on the economic growing. In other words, it means a 1 % lessening in FDI would give a 0.008 addition in Real GDP per capita. Government investing and substructure appears to be negatively correlated with growing at 5 % significance degree while authorities ingestion and population are insignificantly positive linked with growing at same significance degree.


2 Nigeria Regression Analysis

Table 4.2: Nigeria Regression Results

Dependent Variable: LNReal GDP Per Capita

Appraisal from 1980 to 2008

VariablesCoefficientStd ErrorT-RatioFDI/GDP0.00011630.





3119125.32R2 0.6833F-Test ( 8, 20 ) 0.0011*Durbin-Watson Statisticss 1.

11*Significant at 10 % degreeFrom the above consequence, it can be seen that none of the explanatory variables except openness to trade are significantly correlated with growing. The degree of FDI attracted by Nigeria is mean ( Asiedu, 2003 ) compared with the resource base and possible demand. The consequences of surveies carried out in association between FDI and economic growing in Nigeria are non common in their entries. A closer scrutiny of these old surveies reveals that witting attempt was non made to take attention of the fact that more than 60 % of the FDI inflows into Nigeria is made into the extractive ( oil ) industry. The clang of universe oil monetary values in 1980 caused a monolithic divestment from the state and the low degree of inflow obtained until 1986. There is non consecutive correlativity in the theoretical account. However the VIF which is 7.90 is instead high but it instead indicates no multicollinearity.

4.1.3 South Africa Regression Analysis

Table 4.3: South Africa Regression Results

Dependent Variable: LNReal GDP Per Capita

Appraisal from 1980 to 2008

VariablesCoefficientStd ErrorT-RatioFDI/GDP-0.





04859310.0144383.37*Changeless6.9172610.198551934.84R2 0.9530F-Test ( 8, 20 ) 0.0000*Durbin-Watson Statisticss 1.

87*Significant at 5 % degreeFrom these consequences, it can be concluded that there is no autocorrelation where d-value=1.87. The VIF is 6.17 which is less than 10 and it indicates no multicollinearity. The coefficient of FDI/GDP is consistent with the theoretical literature, which postulates that FDI plays a important function in advancing economic growing in South Africa. The empirical consequences show that a 1 % addition in FDI will increase growing by about 0.00074 % in South Africa. As expected the coefficient of rising prices is negative.

Government ingestion, openness to merchandise, human capital and population is positively correlated with growing whereas authorities investing and infrastructural development is undistinguished with the existent GDP per capita.

4.1.4 Comparison between all 3 African States

This subdivision attempts to analyze farther the informations sample by comparing the 3 African states with one another.

The impact of FDI in Mauritius over the period of 1980-2008 showed a negative relationship with growing as 25 old ages ago most FDI were in low-skill low-value added sectors. Since the late-1990s, Mauritius has been seeking to develop competitory advantages in higher value added sectors and therefore is looking to pull FDI in the higher value added activities of the industrial and service sectors ( fiscal sector, EPZ and the Information Technology sector ) . In the twelvemonth 1990s, the FDI influxs were limited to a really minimal degree. Both the EPZ and touristry saw the influxs of FDI lessening well, being close to nil in 2001. The Banking sector witnessed an increasing investing in the late ninetiess, with a extremum in 1997 of Rs.1, 122 million due to a individual investing by a South-African bank in the State Bank of Mauritius, but fell back to no FDI inflows in the early 2000s.

FDI in Nigeria has a positive but non important relationship with economic growing. With a GDP of US $ 43.4 billion and a population of 132.8 million in 2002, Nigeria was the 4th largest economic system in Africa and has an internal market that has no challenger within the continent. The state has failed to unleash its FDI potency mostly for self-inflicted grounds and it has made small advancement in pulling FDI despite its huge homo and natural resources.

Nigeria merely carefully and late, in the mid 1980s, embarked on a reform way but this was characterized by frequent break by political dazes and policy reversals. The FDI environment in Nigeria has improved, at least comparative the state of affairs in the pre 1980s, although it is still less suiting sometimes hostile and unequal to pull high quality, efficiency-seeking FDI. Overall, Nigeria has made small advancement in pulling FDI, when related to the size of its economic system ; it becomes evident that the sums of FDI into Nigeria are instead littleOver the last one-fourth of a century, South Africa has attracted really small foreign investing, due to the political environment like the infliction of trade and fiscal countenances in the mid-1980s, the subsequent fiscal crisis, the tightening of capital controls. Accumulative FDI influxs in 1980-93 amounted to merely over $ 0.3 billion.

After 1993, FDI increased, with two major events ruling this period: the partial sale of authorities portions in Telkom in 1997 and the coup d’etat of De Beers by Anglo American in 2001. Overall, nevertheless, FDI has stayed at comparatively low degrees averaging about 1.5 per centum of GDP during 1994-2002. In footings of sector distribution, the FDI influxs have been comparatively diversified. Contrary to what one would anticipate, the function of natural resources is less of import despite South Africa ‘s big mineral militias. Non-mining activities have drawn more than two-thirds of the FDI influxs, proposing that the chief purpose of foreign investing in South Africa has been to capture domestic and regional markets. Given South Africa ‘s low degrees of domestic economy and investing, higher FDI influxs are critical to spur growing.

The negative rising prices in Mauritius and Nigeria suggests that the macroeconomic policies in topographic point encourage growing. It is expected that ability to command rising prices should cut down investing hazards and enhance FDI and growing. The positive relationship of the rising prices coefficient in South Africa implies that the development within the macro economic system is such that it encourages FDI influxs bespeaking that the assorted policy enterprises aimed at promoting investors is giving the expected consequences.In all the 3 states, the impact of authorities investing is positive but undistinguished. Though this is perplexing, given the strong theoretical statement for a positive important consequence of authorities investings on economic growing, the theoretical analysis in the literature reappraisal indicated that non all types of authorities investing have a strong direct impact on economic growing. This means that authorities investing has an indirect impact on GDP through other variables, such as public ingestion and private investing.The positive relationship of the Government size variable in the South African instance is implicative of the fact that the nature of authorities ingestion is such as to promote FDI influxs in footings of proviso of substructure and enabling environment. Both Mauritius ‘s and Nigeria ‘s instances indicate a positive relationship, and although non statistically important, these findings may propose a demand side consequence.

The non significance of the substructure variable in all the 3 states indicates there is the demand for productive attending to be given to proviso of needful substructure, particularly power coevals and distribution and H2O supply, to heighten economic growing. The ground that substructure has a negative consequence on the economic growing may be due to the fact that substructure is oversupplied compared to the economic graduated table.Openness in Mauritius and every bit in South Africa has a positive and important relationship with economic growing. This is expected and is consistent with old consequences such as those of Asiedu ( 2001 ) .

This consequence stresses the importance of divergences in export and import monetary values on per capita GDP growing. These divergences are a major beginning of economic instability in developing states, where the majority of export net incomes is from primary trade goods. However, the consequences show that Openness in Nigeria is negatively related to FDI influx which agrees with Anyanwu ( 1998 ) , who gave the ground for such observation as the inauspicious structural accommodation programme policy steps in topographic point that led to the capital flight experience.Human capital measured by secondary registration in Mauritius and South Africa is statistically important. This confirms the anticipations of the endogenous growing theory on the importance of human capital for economic growing as the application of advanced engineering embodied in FDI requires a sufficient degree of human capital. The degree of available human capital in Nigeria is low and there is demand for more accent on preparation to heighten its possible to lend to economic growing.

Even if the estimated coefficient of population growing rate in Nigeria is non important, negative coefficients indicate that higher population growing rates tend to keep economic growing. The population growing rate in Mauritius and South Africa appears to be important connoting that the demand for population is to make economic systems of graduated table that addition efficiency. Modern growing theory emphasizes the functions of capital formation and new engineering to accomplish higher carrying capacity for an economic system.

4.2 Panel Data Analysis & A ; Results Appraisals

Table 4.4: Hausman Test for Fixed versus Random Effectss

Dependent Variable: LNReal GDP Per Capita

3 States used for appraisal from 1980 to 2008


Prob & gt ; Chi2

Model to utilize

179.370.0000FE ModelAs per tabular array 4.

4, the process indicates that the single effects are supposed to be fixed. A P value greater than 0.05 favors RE theoretical account and in this peculiar instance the value is in existent fact less than 0.05. Thus the Fixed effects theoretical account is more efficient relation to the Random effects theoretical account under H0.


2.1 Specification Trials

The RESET trial provides strong grounds that the consequences indicate in this paper do non hold theoretical account misspecifications. The Breusch and Pagan test expresses that there is no heteroskedasticity job in the theoretical account.

In add-on, since all specifications are found to hold a consecutive correlativity job which is based on the Lagrange Multiplier trial, the corrected trial to be used alternatively is the Fixed Effect arrested development with AR ( 1 ) perturbations. Based upon these trials for econometric issues, such as theoretical account specification, autocorrelation, heteroskedasticity and Multicollinearity, it is implied that the decisions provided in this subdivision are dependable.

4.2.2 Testing for Model Multicollinearity

Many variables may travel together in systematic ways. Such variables are known as collinear variables. The job of collinearity would take to incorrect decisions which are called collinearity or Multicollinearity.

A simple trial for multicollinearity is to look at the correlativity coefficients among explanatory variables must non transcend 0.85 and another manner to observe multicollinearity is through the discrepancy rising prices factor ( vif ) . The average VIF is 6.54 and it can be claimed that there is no multicollinearity happening in the theoretical account.

4.2.3 Testing for Model Autocorrelation: Lagrange Multiplier Test

When utilizing panel informations, there is a demand to see the being of autocorrelation.

If the job of autocorrelation is neglected, the consequence may be overestimated or underestimated and it could act upon the truth of interval appraisal and hypothesis trial. The LM trial consequences point out that its matching p-value of 0.0290 is good below the significance degree of 0.05. Therefore, after rectifying for autocorrelation in the theoretical account, the p-value gives 0.2677 which suggests that the theoretical account is non car correlated.


4 Testing for Model Heteroskedasticity: Breusch-Pagan Trial

In statistics, the Breusch-Pagan trial is used to prove for heteroskedasticity in a additive arrested development theoretical account. Breusch and Pagan have devised a Lagrange multiplier trial of the hypothesis that I?i2 = I?2 degree Fahrenheit ( I±0 + I± zi ) , where zi is a vector of independent variables. This trial can be applied to a assortment of theoretical accounts. Under normalcy, this modified statistics will hold the same asymptotic distribution that it provides a more powerful trial. The chance of the chi-square theoretical account ( p = 0.

4121 ) is greater than the significance degree of 5 % . Therefore it can be stated that there is no heteroskedasticity in the theoretical account.


5 Testing for Model Misspecification: The RESET Test

The RESET trial ( Regression Specification Error Test ) is designed to observe omitted variables and wrong functional signifier. It examines whether the functional signifier of theoretical accounts suffered from misspecification mistakes and omitted variable prejudice. Overall, the general doctrine of the trial is: if we can significantly better the theoretical account by unnaturally including powers of the anticipations of the theoretical account, so the original theoretical account must hold been unequal.

The RESET trial consequences indicate that its matching p-value of 0.505 is good above the conventional significance degree of 0.05. There is undistinguished grounds to propose that the theoretical accounts are unequal. The deficiency of significance of the RESET statistic suggests that the theoretical accounts are appropriate.

4.3 Interpretation of FE Results

The equation for the relationship between FDI and Economic Growth is as follows:LNRGDPPC = 5.865 – 0.

00253FDI – 0.000326INF + 0.000203GFCF + 0.0193GCONS – 0.129INFRAS + 0.

0018HUMAN + 0.0267OPEN + 0.183POP + Iµt

Table 4.5: FE Regression Consequences

Dependent Variable: LNReal GDP Per Capita

3 States used for appraisal from 1980 to 2008

VariablesCoefficientStd MistakesT- RatioFDI/GDP-0.00253030.0019199-1.32INF-0.00032550.





03Changeless5.864860.249937723.47Note: Real GDP per capita is expressed in log signifier ; t statistics: important at 5 %Within R2 = 0.7824 Between R2 = 0.

9881 Overall R2 = 0.8663The FE arrested development tests the hypothesis to which FDI impacts negatively on growing while it was expected to be positively important. The consequences in Table 4.2 confirm this hypothesis that FDI does non hold an overall positive consequence on economic growing, after commanding for rising prices, authorities ingestion, authorities investing, population growing, human capital, openness to merchandise, and substructure.

Furthermore, the coefficient of FDI is -0.00253, which shows that a unit alteration in foreign direct investing consequences in a positive alteration in gross domestic merchandise by a factor of about 0.03 per centum. It is unusual for FDI to be negatively correlated with growing and this is due to the fact that human capital has a negative impact on FDI.Inflation displays a negative and undistinguished correlativity with economic growing, as expected. This consequence is in conformity with other surveies ( as De Gregorio, 1992 ) .

The penetration for this judgement is simple in saying that a high rising prices rate jeopardizes fight and hence exports, and may be a symptom of the being of deformations in the markets, deficiency of financial subject or hapless macroeconomic stableness ( as argued by Fischer, 1993 ) .The GFCF is considered as a booster of growing and the consequences shows that it is statistically undistinguished at 5 per centum. The coefficient of GFCF is 0.000203 which implies that a one unit addition in GFCF leads to a rise of 0.0002 units in GDP. Thus GFCF remains an indispensable constituent of the outgo on GDP and the 3 African states should heighten investing in edifices, workss, conveyance & A ; equipment in order to supply foreign investors the necessary clime for puting in host states.Government size has a positive and important relationship with growing, proposing that authorities outgo encourages economic growing via investings ; in other words, there is a “ crowding in ” consequence of authorities outgo.

This is contrary to the entry of Akinlo ( 2004 ) but in line with the findings of Adelegan ( 2000 ) . The positive coefficient of the authorities size indicates that the nature of authorities ingestion is such as to promote FDI influxs in footings of proviso of substructure and enabling environment.The substructure variable has a positive and important relationship with growing similar to the instance of Easterly and Rebelo ( 1993 ) . Technological substructure is one of the many countries in which foreign direct investing is meant to profit a state. With the aid of foreign direct investing being made in a state, the authorities can build, every bit good as, better the bing technological tools at their disposal.

This in bend besides plays a really important function in the economic development of a state as this technological promotion assists a state in upgrading its industries and therefore helps them to confront the challenges of the modern-day planetary economic system.The coefficient of Secondary Enrolment Rate is positive with a value of 0.0267.

This is in conformity with a priori outlook resembling the theory of Borensztein et Al. ( 1998 ) . The t-statistic value of 12.91 implies statistical significance. This indicates the being of a positive relationship between registration in secondary schools and economic growing such that a one per centum addition in secondary registration leads to a important and more than proportionate addition in GDP and finally, economic growing.

Openness has a positive and important relationship with economic growing. This is expected and is consistent with old consequences such as those of Asiedu ( 2001 ) . Li and Liu ( 2004 ) and Flexner ( 2000 ) besides report a positive relationship between trade and economic growing in China and Bolivia, severally. This consequence stresses the importance of fluctuations in export and import monetary values on per capita GDP growing. These fluctuations are a major beginning of economic instability in less developed states, particularly in Africa, where the majority of export net incomes is from primary trade goods.Population has a positive and important relationship with economic growing. The coefficient of population is 0.

183 which means a one per centum addition in population would give a 0.2 addition in GDP. Empirical grounds on the effects of population on growing reflects the contradiction bing in the literature by Kelley and Schmidt ( 1994 ) whose consequences show that the least developed states have the strongest positive relation between population growing and economic growing,R2 measures the goodness of tantrum of the arrested development equation.

If in a theoretical account, the R2 is closer to 1, there will be greater ability of that theoretical account to foretell a tendency. A value of R2=1 would connote that the theoretical account provides perfect anticipations. The within R2 shows that over 78 per centum of the fluctuation in GDP is explained by the independent variable for all the 3 African states included in the analysis and once more for each state, a between R2 of 0.98 indicates that a 98 per centum fluctuation in GDP is explained by the independent variable for the given state. The overall R2 provinces that the regressors are explicating about 87 percent fluctuation in the dependant variable which is good since most of the variables are important.


I'm Ruth!

Would you like to get a custom essay? How about receiving a customized one?

Check it out