Estimation Relationship Between Nuclear Energy Consumption and Economic Growth in Pakistan Using ARDL Approach

Background/objectives: The use of nuclear energy plays very imperative role in economic growth. Method/findings: The present study employs ARDL approach to inspect alternative and nuclear power (% of total power use) and GDP per capita (constant 2010 US$) (economic growth) in Pakistan 1971–2014. Findings: The ARDL approach suggests that the dependent variable and regressor have long-term equilibrium but they are insignificant. The outcomes also propose that, both the long-term and short term, nuclear power utilization has a non-negative impact on economic growth. As a final point, the Granger causality analysis shows a unidirectional causality between alternative and nuclear power (% of total power use) and GDP per capita (constant 2010 US$). Application: Furthermore, ARDL test in this study suggests that nuclear energy is an effective solution to boost economic growth in Pakistan and gives useful information for policymakers.


Introduction
A few years ago, nuclear mishap in Fukushima, Japan was a gigantic strike to the status and belief in nuclear power creation, which make a number of countries to re-evaluate the mechanism of generation of nuclear power. The current financial upheaval might be limited in the developing or the developed countries if as the positive environmental effects of the nuclear power establishment are substantial. Thus, tremendous literature is available regarding the relationship flunked by use of nuclear power and economic growth. Abundant studies have been conducted in the past two decades to find an association flunked by these variables. The general findings of the studies show that the use of nuclear power and economic growth have a strong relationship. 1 For instance, Schurr 2 has perceived a nonnegative liaison between economic growth and use of nuclear power. Some countries are using nuclear power at large scale; others are introducing nuclear power first time. This is one of the main topics on Nuclear Power in the 21st century. This is well known that Pakistan is the slowest producer and consumer of nuclear power. Pakistan is not a member has not top position in the list of expanding countries in the area of nuclear power. Currently, seven reactors are under construction in Russia, six in India and three in the Republic of Korea and two in Pakistan. However, the United States, France, Japan, and China are utilizing more nuclear power than all other countries in the world.
Power is a main pillar for the development of any nation. It supplies power to industrial sector by which goods and services are produced in the economy and in return it generates economic growth and brings opulence in society. In Refs. [4][5][6][7] , authors emphasize an optimistic effect nuclear power on economic growth in several nations.
Power is a main input factor in production procedure, nevertheless, some researchers argued that it play a vital role along with production factors like capital, land, and technology in the production process 8 in line with Cheng and Lai 9 who also argue use of power to boost the efficiency of other production factors. The endorsement of International Power Agency 10 demonstrates that the consumption of the power achievement, power-related emanations of CO 2, and the demand and consumption of oil will increase more than twofold by 2050. Therefore, without any sufficient amendment in government policies, consumers continuously utilize power more and more, thus dependency on fossil fuels would increase by which climate change and it would affect economic growth.
At this point, many countries have switched from the use of fossil fuels to hygienic surrogate power resources such as hydropower; nuclear power; wind power; and natural gas that will bail out them face escalating consumers demand and tumble the dependency on imported oil, nevertheless, Pakistan uses less nuclear energy than all these countries. The use of nuclear power brings price stabilization and reduces emission of greenhouse gases; consequently, it will increase supply of secure power. Thus, nuclear power is considered a carbon free source of power and this is the best solutions to secure power in the future and to prevent global warming. 11 It can be examined that both variables EG (economic growth) and NEC (nuclear energy consumption) have a non-negative tendency through all over the study period; nonetheless, the nuclear utilization raises in onwards period. 12,13 The query arises hereby whether the uses of nuclear power affect the growth and development positively in shape of "by-products" 4 .
Actually, Pakistan has gifted with huge number of natural resources of power like wood, water, coal, gas, sunshine, oil, and wind, but unfortunately, these resources are not correctly consumed even though most of them stayed unused for decades. Subsequently, due to poor investments in power sector, thus power deficit is natural phenomenon for Pakistan. The economic growth and development in Pakistan faces enormous hurdle due to insufficient of power services. As a result, Pakistan is making try to promote nuclear power sector by which increase provision of power services to the Pakistani people. Thus, this will increase the per capita power consumption and growth of the country substantially. The consumption of energy will increase in South Asian countries because Pakistan, India, and Bangladesh consume more energy due to excess pressure of population. 14 Nevertheless, issues of use of nuclear energy have not been examined appropriately in Pakistan. In Pakistan, the nature has gifted a large number of resources, but these resources are not used properly. Therefore, in energy infrastructure, Pakistan faces serious energy shortfalls, because there is insufficient investment in energy infrastructure. Thus, Pakistan is making investment in the nuclear sector to enhance delivery of energy facilities to people in Pakistan.

Background of the Problems
The aforementioned studies on alternative and nuclear power (% of total power use) and GDP per capita (constant 2010 US$) (economic growth) show these has not been studied properly but it is considered very important for the benefit and welfare for the masses. The researchers found limited studies nuclear energy and economic growth. Consequently, more researches are needed in this field for significant improvement in Pakistan. Nevertheless, the alternative and null hypotheses are given as below: H 0 . There is no cointegration between use of nuclear energy consumption and economic growth. H 1 . There is cointegration between use of nuclear energy consumption and economic growth.
The null hypothesis of no long term liaison is H 0 :¥ = ¥ 1 = 0 and it was checked under the direction of alternative hypothesis. The null hypothesis is ¥=¥ 1 which shows that there is no long-term liaison between variables, while in the alternative hypothesis is ¥≠ ¥ 1 ≠ 0 which shows that there is long-run relationship between variables.
The remaining part of this research paper is organized as follows. The effect of nuclear power utilization on economic growth is briefed and discussed in the second section. Empirical approach and model specification is described in section 3. Section 4 demonstrates mechanism of data collection. The findings and discussion are depicted in sector 5. The finishing comments are presented in the last sector.

Gap in Literature
According to researchers' views, the study on alternative and nuclear power (% of total power use) and GDP per capita (constant 2010 US$) (economic growth) is required in a wider context in Pakistan. However, extensive literature on the alternative and nuclear power (% of total power use) and GDP per capita (constant 2010 US$) (economic growth) eminence showed that theoretical and empirical study regarding on alternative and nuclear power (% of total power use) and GDP per capita (constant 2010 US$) (economic growth) prestige is limited. In addition, researchers strongly recommended that there is excessive need to use of nuclear energy and economic growth in scenario of present era for the welfare of society.
In nutshell, the main purpose this research is to determine the liaison between GDP per capita (constant 2010 US$) and use of nuclear power measured as alternative and nuclear power (% of total power use) in the Pakistan for the period from 1971 to 2014. GDP per capita (constant 2010 US$) has been taken as a proxy for economic growth and alternative and nuclear power (% of total power use) for use of nuclear energy.
A detailed discussion of the principles and the mathematical foundations regarding alternative and nuclear power (% of total power use) and GDP per capita (constant 2010 US$) (economic growth) are presented in the subsequent sections along with the methodology by which relationship alternative and nuclear power (% of total power use) and GDP per capita (constant 2010 US$) (economic growth) are explored.

Materials and Methods
The mechanism of Cobb-Douglas Production Function is represented as: Here y is Domestic Gross Productive per capita (constant 2010 US$), nec is alternative and nuclear power (% of total power use). The equation may be predicted in a linear form such as below.
Log of proxy for economic growth is GDP per capita (constant 2010 US$), α is the intercept and β 1 is the coefficient of nec where nec is the log of nuclear power and μ is the error term. Expectation: α > 0, β 1 > 0.

Model Specification
The ARDL methodology for cointegrating test was established by Pesaran, Shin, and Smith (2001) which is valid regardless of whether the regressor variables are I(0), I(1), or mutually cointegrated.
There are the basic steps of ARDL model is as follows It may be extended as The range of summation in various term in Equation (2) are represented from 1 to q1, 0 to q2, 0 to q3 and 0 to q4, respectively.
The appropriate values for maximum lags, p, q1, q2, q3 and q4 are selected by using the information criterion to select for using bound test again as below Perform an "F-test" for acceptation or rejection of null hypothesis and alternative null hypothesis.
In each case, the lower bounds is based on the assumption that all the variables are I(0), and the upper bound is based on the assumption that all the variables are I (1).
as well as the usual ECM: ADF test is used to check stationarity of the variables at I(0) or I (1). The bound test with an ARDL methodology is used to check the validity of long-run associations of variables.
The ARDL approach is usage to conclude the relationship between alternative and nuclear power (% of total power use) and GDP per capita (constant 2010 US$) (economic growth. The ARDL econometric technique familiarized by Pesaran et al. 16 and adopted. 17 This technique is several time better over the cointegration technique represented by the authors. 18,19 This approach is adopted small sample and it has not imposed condition the variables have the same order of integration. Furthermore, dependent and independent variables are explained by their past. In the next step, computed value of the F-statistics are compare with tabulated critical bounds values, which were depicted by Granger. 20 The study will find a long-run relationship between variables. Thus, the null hypothesis of a long-term alliance is refused. Furthermore, thevalue of F-statistic surpassed over critical bound values, despite orders of integration of the variables is I (0) or I (1), it means that there is long-term liaison between variables. In the same way, if the computed values of F-statistics are below the critical value, the null hypothesis is not rejected and it shows that there is no long-term relationship between dependent variable and regressor. Furthermore, if the F-statistic values fall between these two bounds critical values then findings are undecided.
The mechanism of ARDL model is estimated as below:   Thus, Equation (6) As a final point, for reliable findings a stability test is used to check the fitness of model. The cumulative sum of CUSUM and CUSUMSQ tests depicts stability of the model. The both tests are employed at the 5% significance level and they indicate the null hypothesis is not rejected because the graphs of the CUSUM and CUSUMSQ statistics remain inside the boundary lines. Thus, it shows that model is fit for analysis and findings of analysis are confirmed.

Mechanism of Granger Causality Test
The estimation of cointegration shows a causal relationship flunked by dependent variable and regressors, but the path of causality is not clear. The path of causality among variables can be investigated by employing the Granger Causality test. 21,22 If the cointegration flunked by variables exists at non-stationary level, then, Granger causality test among the variables is verified through system of the following Equation (7)

Data Collection
In order to find the relationship flunked by use of nuclear power and economic growth, the data for economic growth and nuclear power utilization have been taken from world indicator for the period 1971-2014. Economic growth is proxy for GDP per capita (constant 2010 US$) and use of nuclear power is proxy for alternative and nuclear power (% of total power use). The use of GDP per capita is more suitable than gross national product (GNP) because goods and services which are fabricated within the nation not with the use of nuclear power. The GDP per capita is measured in USA currency as a dollar. The use of nuclear power is measured in unit of Terawatthours (TWh).

Results and Discussions
The ARDL model is employed if the variables are I (1) or

ARDL Cointegration Approach
Before employing ARDL bound test, the choice of the optimal lag order of variables is most important because it affects the statistical value of the F-test. The Lag order selection criteria are used to find the optimal lag. The findings are presented in Table 2, which remarks the optimal lag order is 1.
With the help of optimal lag order, the F-test is used to investigate the cointegrating liaison between the variables. The findings points out that the F-statistics values do not surpassed the lower critical and the upper critical bound at the 5% and 10% levels. The outcomes are depicted in Table 3. In Ref., 19 author suggested that the null hypothesis of no long-run liaison is accepted and the alternative null hypothesis of long-term liaison is rejected. The findings of ARDL bound test verified that alternative and nuclear power (% of total power use) and GDP per capita (constant 2010 US$) (economic growth) have not a long-run relationship because the calculated F-statistics values do not exceed the upper and lower critical bounds, it infers that there was no a long-term liaison between dependent variable and regressors. It is verified that dependent variable and regressor have not a long-run relationship between each other, which is represented in Table 4. The coefficient of nuclear power is nonnegative, but it is significant which suggest that there is a positive influence of the use of nuclear power on economic growth in Pakistan, but there is not a long-run relationship because the conditions of Wald test are not justified.
The long-run elasticity nuclear power for economic growth is 0.88, which indicate that 0.88% boost up in economic growth but it is insignificant even at 5% level    if a 1% increase in use of nuclear power. But, the use of nuclear power does not appear to be an imperative factor on economic growth in the long term. The findings suggest that there is a nonnegative liaison between dependent variable and regressor in Pakistan's economy. The findings of this study confirm the result of Granger 20 , Yoo and Kwak 21 and Elliott et al. 22 The estimated shortterm coefficient of Ecm(t−1) is −0.02 in error-correction model which is negative and statistically insignificant. This error-correction divulges the deviation from the long-term balanced level of economic growth in one year and speed of the adjustment is represented by 2% in the next year, but it is insignificant.

The VECM
The findings of the GCT are presented in Table 5. The findings show nuclear energy consumption and economic growth have a unidirectional liaison in a short-run. These findings imply that the economic growth and use of nuclear power have unidirectional causality.

The Instability Test Results for Longterm
The parameters create unreliability in the model, thus, CUSUM and USUMSQ technique are practiced to determine the stability of the parameters. Therefore, results of the CUSUM and USUMSQ predict that all coefficients remain stable and credible in the errorcorrection model because the graph of both tests in Figures 1 and 2 remain within the boundaries lines and this shows the stability of parameters in this model over the study period. So, the selected model is fit for analysis, thus policy decision-maker can use these results for the welfare of society.

Graph of cumulative sum of recursive residuals
For short-run period. This study also uses CUSUM and USUMSQ to find the steadiness of the parameters in the

Recursive Coefficient Test for Stability
The recursive coefficient test for stability depicts the model is fit for analysis, because graph of all variables lays between the boundaries lines as shown in Figure 5.

Findings/Conclusion
This research observes the relationship between nuclear power and economic growth in Pakistan by using ARDL method and Granger Causality technique. The conclusions suggested that nuclear energy and economic growth have not a long-run relationship between each other and similarly in a short-run. This study discovers that use of nuclear power causes a significant increase in economic growth in Pakistan but it is not in a long term. The estimated short-run coefficient of Ecm (t−1) is −0.02 in error-correction model which is negative and statistically insignificant. This error-correction reveals the deviance from the long-term balanced level of economic growth in one year and speed of the adjustment is represented by 2% over the next year but it is insignificant. The GC test shows nuclear power and economic growths have unidirectional causal relationship. Thus, these findings have imperative suggestion for policy making. The findings consist with the results of specifically; the findings show that if the nation will use more unit of nuclear power in coming decades, it will achieve economic stability and sustainable growth in future.