advert

JOURNAL OF RESEARCH IN NATIONAL DEVELOPMENT VOLUME 8 NO 1, JUNE, 2010


THE DYNAMICS OF GAS UTILIZATION AND THE PERFORMANCE
OF NIGERIA’S ECONOMY: AN ERROR CORRECTION ANALYSIS
Nathan Pelesai Audu
Department of Curriculum/Instruction, Niger Delta University, Wilberforce Island
E–mail: awudupel@gmail.com

Abstract: 
This study takes a look at the relationship between gas utilization and the performance of Nigeria’s economy 1970–2009 using annual data. This is necessary in view of the fact that there seems to be a cause effect relationship between the two. Nigeria is ranked as the 7th highest oil producing country in the world and produces 34% of Africa’s gas reserves. It is in recognition of this, the government via the Nigerian national petroleum corporation and its multinational joint venture partners are vigorously pursuing various gas utilization projects targeted at economically harnessing her largely wasted gas resources. It also set year 2020 as the zero flare down date and has set its aspiration for the gas sector. It is against these backdrops that this study is generally targeted at establishing the relationship between the two variables with a view to make recommendations and policies that are aimed at encouraging the growth of the Nigerian gas sector. Using the error correction analysis the empirical results shows that there exist a cause–effect relationship between gas utilization and Nigeria’s economic performance. The study also finds a strong significant relationship between foreign direct investment (FDI), broad money supply, inflation rate, fiscal deficit and crude oil production on economic growth. It was recommended that to sustain and improve Nigeria’s economic growth and development, gas utilization should be vigorously pursued and encouraged. It also recommended that for the Nigerian gas sector to be fully utilized, policies aimed at attracting FDI as well as the full liberalization  of the upstream and downstream sectors of the oil and gas industries should be vigorously pursued.
Keywords: Gas utilization, growth, error correction, stationarity, economy

 

 

Introduction
Crude oil and natural gas are some of the major natural resources endowed in Nigeria and its economy is centred on crude oil production since the early 1970s, with oil revenue collected increasing from 26% in 1970 to 62% in 1973 before reaching its peak of 82% in 1974 and by 1997 it stood at 72% (CBN, 2009). The dominance of the oil sector in the 70s, 80s and early 90s reflected the significant increases in oil prices within the period 1970 to early 1990s (Adedeji, 2001). In the recent decades, changes in Nigeria’s GDP depended mainly on her crude oil exports, therefore it has becomes increasingly clear that the reduction in the country’s dependence on crude oil via diversification is fundamental to the stability and development of her economy and it was against this backdrop that the Structural Adjustment Programme (SAP) was introduced in 1986 to reverse this mono–cultural feature and reduce the economy dependence on the oil sector (CBN, 2010). And the sure way to achieve this aim is through the utilization of the wasted natural gas because it is a vital part of natural capital and the economy exploitation of such resources is the path way to economic growth.
Diversifying the Nigerian economy has become increasingly vital to the country’s economic sustainability. Given Nigeria’s abundant gas reserves, gas utilization offers a viable option for possible diversification of her economy but available statistics shows that 70% of the gas produced in 2008 was flared, while marketed production accounted for only 18% but with increasing emphases on gas utilization and the emergence of Nigeria’s liquefied natural gas (LNG), marketed production increased to 40% and 45% of the gas produced in 2006 and 2007 (CBN, 2008). At the domestic front today, natural gas constituted about 38% of fuel consumption in Nigeria in 2007. This is more than half of the 22% of the country’s energy consumption in the mid–1990s (OPEC, 2008). IMF (2003) observed that the Nigerian government projects that foreign exchange earnings from gas will surpass that from sale of crude oil in the medium term. But this is however dependent upon the country’s ability to develop the gas sector to significantly realize its potentials. From the foregoing, it is clear that a lot needs to be done for the country to get to its desired destination in its quest for the growth of the gas sector. This paper examines the dynamic influence of economic utilization of gas flaring as well as identifies variables that could drive or constrain the growth of the gas sector and the performance of her economy.
The rest of the paper is divided into six sections. Section 1 is the introduction. Literature review, theoretical underpinning and gas production, utilization and flaring in Nigeria form the focus of section 2. In section 3, the paper highlights the model and the methodological issues. The research outcomes are discussed in section 4. Finally, section five presents the conclusion and policy recommendations of the analysis.

Theoretical issues
The critical problems that plague most country’s economy in contemporary world are unemployment, economic stagnation and large trade deficits. The orthodox policy prescriptions that are used to rectify these problems have as their basis the open economy extension of the Mundell–Fleming model and the monetary approach to balance of payments (Dernburg, 2008). The orthodox perspective held that the removal of market rigidities and imperfections along with contractionary fiscal and monetary policies as well as exchange rate devaluation would bring about appropriate price movements that would raise the long–run growth rate, restore full employment and correction all trade imbalance.
The Keynesians models of open economy have provided the most significant alternative to those of the neoclassical model. The key feature of these models is the assumption that unemployment and excess capacity are structural features of the capitalist economy while in the neoclassical models, the exogenous demand growth from exports plays a central role in the accumulation of process. The staple growth, export–led growth models and the Thirlwall’s law are the contemporary extension of earlier works that is used in the heterodox literature because exports are expected to enhance economic growth (Udah, 2007). Boame, (2002) posit that there is a considerable evidence supporting the export promotion hypothesis as a development strategy. The advocates of the export–led hypothesis have argued that trade was in fact one of the major determinants of growth in South–east Asia. They argued that Hong Kong, Taiwan, Singapore and the Republic of Korea, the so called four Asian Tigers, have been successful in achieving high rate of economic growth because they consistently pursued free market and outward oriented economies (World Bank, 2009 – a and b).
There is plethora of studies that supports this hypothesis  which includes Fosu, (1996), Thornton, (1997), Shan and Sun (1998) and Udah, (2007) have demonstrated that the relationship between exports and economic growth is bidirectional. Trade theorists pinpoint a number of factors causing this link. Anyamele, (2000) observed that Nigeria continue to remain a public sector dominated economy but recent developments in economies of the industrialized economies have reinforced the importance of outward oriented growth policies. Export–led growth not import substation industrialization and trade liberalization not protection is currently the recognized aim of the economic policies in Nigeria, and most of the developing economies of the world. Therefore, it can be postulated that most developing countries’ growth has been influenced by their dependence on primary export commodities and the relevant economic theory underlying the primary–export–led growth is the staple theory.
The staple theory of trade argues that trade leads to bringing into production or cultivation of previously idle resources, creating a vent for surplus return to those resources. It is structurally similar to the vent–for–surplus theory to the extent that resources formally wasted or idle are subsequently exported. It also has some close affinity with Lewis theory of economic development with unlimited supplies of labour to be vented via trade. In this case, it is labour not natural resources to be vented. It also has some similarity with Rostow’s theory of growth to the extent that the leading sector is the staple–export sector which grows more rapidly thereby propelling the rest of the economy along the growth path. It follows that gains from trade is not once–over change in resource allocation but are also merging with gains from development. In this way international trade increases and transforms the domestic production frontier as well as the productivity of the domestic economy (Udah, 2008).
Thus, the central thesis of Staple growth theory is that when a country has comparative advantage in primary goods production, this result in the expansion of primary based export commodity which in turn induces higher rate of growth in income per capita. Also the export of primary commodities affects the rest of the economy by diminishing underdevelopment and underemployment, by inducing higher rate of domestic savings and investment, by attracting inflow of factor inputs into expanding export sector and thereby establishing linkages with other sector of the economy. These processes increase the supply responses of the domestic economy and thus the productivity of the export sector. To sum up, the theory submits that extensive growth of the primary staple leads to diversification and industrialization which results in these important benefits; improved utilization of existing resources, expanded factor endowment and linkages effects

Empirical literature
There is a general consensus that Sub–Saharan Africa (SSA) is lacking behind in economic growth and development. There has also been an erosion of SSA’s share of world trade, even for its traditional commodity exports, while FDI inflow into the region is at very low ebb (Mlambo, (2005). Calamitsis, (2007) observed that the problem which inhibit growth in SSA sub–region are structural policy failure, political instability and conflicts, terms of trade and external shocks, poor management of financial resources and corruption, adverse demographic and geographic condition as well as inappropriate fiscal policy. Therefore, economic growth requires increasing investment, proper management of existing and new investments, proper use of key resources and adequate mix of fiscal and monetary policies. The poor performance of most of these countries, even in the face of their natural human resources endowment partly provided a renewed impetus for the study of economic growth and development while ownership of key resources could provide a springboard for economic growth and development (Audu, 2008).
Since the emergence of the study on depletable resource economic in 1914 by Grey, several economist have been interested in the relationship between energy, exhaustible resources and macroeconomic performance and have recognized that physical and human capital, environmental and natural resource should be seen as a vital economic assets that can be called natural capital (Nyong, 2005, Adam, 2001, Iwayemi and Adenikinu, 2001, Adewuyi, 2001, Ibanga and obi, 2001 and Vincent, 1997). However, Sachs and Warner, (1995) observed that natural resource endowment gives rise to economic rent and corruption, hence the benefit of such resources do not trickle down the economy which is the central theme of resource curse hypothesis. Gylfason et al, (1999) and Ross, (2001)confirmed this in their separate studies when they held that higher levels of oil and mineral dependence economy tends to reduce a country’s rate of economic growth; while Papyrakis et al (2004) posits that the resulting sudden increase in income may lead to sloth and less need for sound economic management and institutional quality, as the boom may create a false sense of security and weaken the perceived need for investment and growth enhancing strategy. Also, as the natural resource sector expands relative to other sectors, the returns to human capital decreases as investment in education declines (Ndiyo, 2008, Deaton, 1999, Gylfalson, et al 1999 and Fosu, 2001,) while other reason for the curse of natural resources is the Dutch disease effect and the neglect in education (Bennet et al 2003 and Corden et al, 1982). From the foregoing, two reasons can be adduced why the presence of natural resources may exert negative influence on growth and development. First is the weak institutional arrangements which generate conditions that gives rise to voracity effects which prompts interest group to devote more energies trying to capture the economic rents from natural resources exports while the second is the productive structure of the economy that is related to the allocation of resources among different activities with different spillover effects on aggregate growth.
Conclusively, Manuel, (2003) observed that much of the literature on small economies focuses on the problems presented by the volatility of capital flows, however a larger problem for African economies is that their growth potential is directly affected by their ability to export and use the exports revenue to diversify production. It is against this backdrop that Fosu, (2001) and Audu (2008) in their separate study, held that it is Ricardo’s theory that motivates the need to address the macroeconomic management difficulty of developing countries that result from fluctuations in their revenue which rely heavily on a small number of products.
Gas production, utilization and flaring in Nigeria
Nigeria is a gas province with associated oil reserves but for many years, natural gas is was considered a nuisance (CBN, 2009, OPEC, 2008, Spalding, 2003, Evoh, 2002 and UNCTAD, 2009 a and b). Despite its high gas ranking, Nigeria is not among the top twenty gas producing or utilization countries. About 50% of proved gas reserved is associated gas while the rest is non–associated gas. The country has a huge amount of associated gas and the largest gas flaring nation during production of crude oil among OPEC countries. Hence, the prospects for the commercialization of natural gas via liquefaction for export and for more efficient use of gas for domestic and industrial energy generation coupled with the imperative to end gas flaring for compelling environmental reasons resulting in the emergence of a significant shift towards natural gas as the basis for Nigeria’s future hydrocarbon industry.
The continued flaring of gas is attributed to the economy’s inadequate consumption capacity for gas as manifest in the high capital cost of associated gas gathering and limited domestic demand as well as the need to produce crude oil which is the country’s main export and source of revenue. The implication of this situation, are huge waste of otherwise valuable natural resource and negative influence on the environment in the form of undesirable heat and light effects. ESMAP, (2010) observed that the factors that drive the need to reduce the volume of gas flared in Nigeria include: economic loss (about 2.5 million US dollars is lost per annum; combustion products constitute over 75% of environmental damage through the production of green house gasses and the enactment of anti–flaring laws or policies.
Dinneya, (2006) observed that for the optimal development of the gas sector, the government will have to overcome some obstacles and address several key reforms and the greatest challenge in this regard is the development of the gas sector which is needed to refurbish the power sector so that it can utilize effectively the gas output and refurbishing the power sector, requires the restructuring of the Power Holding Company of Nigeria and the privatization of some aspect of power generation and distribution. The development of a domestic market for gas and its by–products, the promotion of investment in the area of gas utilization that would lead to the substitution of imported product should be pursued vigorously as well as the need to explore the potential for exports of gas to sub–regional, regional and other markets. Finally, the elimination of gas flaring in the country is paramount because of its adverse effect to the environment. For instance, in 2009 about 3 million barrel of oil and an average of 2 million cubic meters of associated gas are produced in Nigerian daily. Of this, about 31.8% of associated gas produced is flared while about 68.2% is utilized (CBN, 2010). It is in this regard that the government has earmarked year 2020 as the date by which gas flaring will be completely eliminated and it is currently on track to achieve this target.

The Model and methodological issues
The model
The model is designed to explain the factors that affect gas production in Nigeria and the relationship between gas utilization and the Nigerian economy using the augmented Cobb–Douglas production function. The novelty of the augmented Cobb–Douglas production function is that it allows in addition to the traditional inputs of production, the inclusion of non–traditional inputs to capture their influence to growth. This model is used by various scholars, among others, Udah, (2007), Papyrakis et al (2004) and Fosu, (1990). Following the adoption of the augmented Cobb–Douglas production function, the general models to be estimated for Nigeria are as defined:

 

RGDP = δ0 + δ1GU + δ2MS + δ3FD + δ4INF + δ5FDI + δ6COP + µ1 …….………… (1)
            δ1 > 0; δ2 > 0; δ3 > 0; δ4 > 0; δ5 > 0; δ6 > 0
Where RGDP = Real gross domestic product, FDI = Foreign direct investment, COP = Crude oil production, GU = Gas utilization, MS = Broad money supply, FD = Fiscal deficit and INF = Inflation rate while µ1 is the stochastic error term.

However, since the data are time series, we explore their long–run properties. The stationarity of the series is tested using the Augmented Dickey–Fuller (ADF) test statistics to investigate the presence of unit root under the alternative hypothesis that the series is stationary around a fixed –term trend. ADF test are performed using the ordinary least square technique to estimate the following equation:

Where t is a time trend; the null hypothesis of non–stationarity is rejected if δ1 is less than zero and statistically significant. But the basic model can be reformulated with the error correction representation as:

Where Z is the residual term from the static regression of Yt on Xt. The optimal lag length is determined using the Akaike information criterion.

The methodological issues
One of the objectives of this study is to investigate the long–run dynamic relationship among the selected variables. The system is represented in equations (1), (2) and (3). To do this we explore the cointegration theory and the error correction mechanism. Given data instability in Nigeria occasioned by policy instability, political cum economic disruption etc, it becomes increasingly useful to test the time series properly of the variables for meaningful economic results. It is clear that OLS regression estimates with non–stationary time series data often produced unacceptable results, even though the overall results may suggest a high degree of fit as measured by the coefficient of multiple correlation, R2 or adjusted coefficient of R2, high auto–correlated residuals and statistical significance as measured by the usual t–statistics (Gujarati, 2004). However, many economic variables have a strong tendency to trend over time, such that the levels of these variables can be characterize as non–stationary since they do not have constant mean over time (Udah, 2007).
This study therefore adopts the cointegration and error correction methodology to estimate equations (1), (2) and (3). This selection is based on the premise that if the variables are non–stationary, the desirable properties of consistency, efficiency and unbiasedness will be lost if the OLS technique is used to estimate the equations, which could lead to spurious results and inference, hence, inaccurate predictions. Cointegration and error correction is used because it adds riches, flexibility and versatility to the econometric modeling and integrates short–run dynamics with long–run equilibrium, hence accurate prediction can be more confidently made on the economic relationship between the variables. Also, apart from the examination of the long–run cointegration of the variables of interest, we will explore the short–run dynamics by performing Granger causality tests for cointegrating systems. Such as exercise will provide an understanding of the interactions among the variables in the systems and this will shed more light on the directions of the causality.

The research outcomes.
The ADF test shows that the null hypothesis of non–stationary for all the variables would be accepted. Therefore, differencing the non–stationary series once resulted in their stationary (that is they are I(1)). Therefore, to avoid spurious regressions, we conducted a cointegration test on the I(1) variables using the Johansen cointegration test.

Table 1: Augmented Dickey–Fuller tests for the presence of unit root


Variables

Levels

1st Difference

Lag

Decision

RGDP

–2.894820

–5.290760

2

I(1)

FDI

–2.324115

–4.268796

2

I(1)

COP

–1.916276

–3.901410

2

I(1)

GU

–2.304381

–4.457726

2

I(1)

MS

–2.741056

–4.602062

2

I(1)

FD

–2.951711

–6.076740

2

I(0)

INF

–2.608350

–4.457942

2

I(1)

Critical values
5%

 

–2.951125

 

–2.943427

 

 

       Source: Author’s computation

 

Table 2: Cointegration test


Sample(adjusted): 1973 2009

Series: RGDP GU FDI COP MS FD INF

Lags interval (in first differences): 1 to 2

Unrestricted Cointegration Rank Test

 

Hypothesized
No. of CE(s)

Eigenvalue

Trace
Statistic

5 Percent
Critical Value

1 Percent
Critical Value

 

None **

 0.732672

 88.10873

 68.52

 76.07

 

At most 1

 0.343376

 39.29539

 47.21

 54.46

 

At most 2

 0.301472

 23.73159

 29.68

 35.65

 

At most 3

 0.182624

 10.45675

 15.41

 20.04

 

At most 4

 0.077768

 2.995471

  3.76

  6.65

 

 *(**) denotes rejection of the hypothesis at the 5% (1%) level.

 Trace test indicates 1 cointegrating equation(s) at both 5% and 1% levels.


Cointegration test
Johansen cointegration test suggests that there is one cointegrating equation at both 1% and 5% significance levels. In all, the integration result implies that there exists a long –run relationship between gas utilization and economic growth.


Table 3: The long–run regression results


Dependent Variable: LOG(RGDP)

Method: Least Squares

Date: 03/03/10   Time: 11:35

Sample: 1970 2009

Included observations: 40

Variable

Coefficient

Std. Error

t-Statistic

Prob.

C

9.060058

5.047142

1.795087

0.0818

LOG(GU)

0.030398

0.179392

0.169451

0.8665

LOG(FDI)

0.742310

0.225285

3.294980

0.0024

LOG(COP)

0.819319

0.381972

2.144970

0.0394

LOG(MS)

0.264694

0.245674

1.077419

0.2891

FD

-0.656604

1.787812

-0.367267

0.7158

INF

0.002123

0.004048

0.524631

0.6033

R-squared

0.902104

    Mean dependent var

12.63027

Adjusted R-squared

0.884305

    S.D. dependent var

2.315258

S.E. of regression

0.397974

    Akaike info criterion

1.152767

Sum squared resid

5.226645

    Schwarz criterion

1.448321

Log likelihood

-16.05534

    F-statistic

214.4900

Durbin-Watson stat

0.793859

    Prob(F-statistic)

0.000000

Table 4: Unit root test for the residual from the long–run model


Null Hypothesis: ECM has a unit root

Exogenous: Constant

Lag Length: 2 (Fixed)

 

 

 

t-Statistic

  Prob.*

Augmented Dickey-Fuller test statistic

-5.671496

 0.0000


The long–run model
Having checked for cointegration between contemporaneous economic growth and gas sector variables in levels, we obtained the long–run results of gas sector liberalization–economic nexus by estimating the general model (equation 1) and testing for stationarity of the residuals. Once the residuals were found to be stationary, we tested for the goodness of fit (R2) using the F–test. All these are as reported in tables 3 and 4.
The high F–statistic and R–squared indicate the joint significance of the explanatory variable and the high degree to which variations in the rates of economic growth are explained by variations in gas sector liberalization indicators. The Durbin–Watson statistic indicates a strong case of serial correlation. Indeed the long–run model shows that though all the explanatory variables are significant in explaining the Nigerian growth process, some of them came up with the wrong signs.
To capture the short–run dynamics, we used the residual from the cointegration regression as the error correction term (equation 3). We start with an over–parameterized model which we then tested for down until we arrived at our preferred parsimonious model. Here, variables with low t–statistic were dropped.
The over–parameterized model of the study is as shown in table 5. The model shows that the explanatory variables are quite able to explain variations in the Nigerian growth process over the entire period of study. Indeed, both the over–parameterized and the parsimonious models showed that the explanatory variables accounted for 98% and 89% variation in the Nigerian growth process, respectively.
The results of the reduced short–run dynamic growth model are presented in table 6. As expected, the error correction term (ECMt–1) is of the expected negative sign and significant in the growth function. This result substantiates the finding of cointegration among the variables reported earlier, but more importantly, it suggests that one cannot overlook the cointegrating relationship among variables in the model; otherwise, this could introduce misspecification in the underlying dynamic structure. The absolute value of the coefficient of the error–correction term indicates that about 65.4% of the disequilibrium in the growth model is been offset by short–run adjustment within a year. In this case, the full adjustments are achieved, and take twelve months to complete the cycles. Thus, to maintain a long–run equilibrium, it is important to reduce the existing disequilibrium over time.
In addition to the disequilibrium effect, the results in table 6 show that growth is been influenced by changes in the first lags of growth and fiscal deficit. This means that growth in Nigeria and fiscal deficit depended on their past levels. The same position holds for the current gas utilization, second lag of foreign direct investment as well as third lags of crude oil production, broad money supply and inflation rate, which means that a unit rise in these variables will raise growth in Nigeria by 59, 94, 45, 56 and 0.4 percent respectively. The first lag of growth has the correct sign and is significant at 5%. The parameter estimate for current gas utilization is correctly signed and significant at 5 percent. In the model even though contemporaneous, the first lag of fiscal deficit do not have the correct sign but was significant at 5%. Furthermore, the third lags of crude oil production, broad money supply and inflation rate were correctly signed and significant at 5 percent respectively. However, the fourth lags of crude oil production and inflation rate were insignificant. As such, they are not good determinants of growth process in Nigeria within the period of study.


Table 5. Short–run overparameterized growth model


Dependent Variable: LOG(RGDP)

Method: Least Squares

Date: 03/03/10   Time: 22:38

Sample(adjusted): 1975 2009

Included observations: 35 after adjusting endpoints

Variable

Coefficient

Std. Error

t-Statistic

Prob. 

C

-38.51516

49.47963

-0.778404

0.5789

∆(LOG(GU))

0.899120

0.372593

2.413142

0.0324

∆(LOG(GU(-1)))

-5.103283

27.34331

-0.186637

0.8825

∆(LOG(GU(-2)))

3.923282

1.436975

2.730237

0.0260

∆(LOG(GU(-3)))

-0.671624

5.255682

-0.127790

0.9191

∆(LOG(GU(-4)))

0.906149

6.512824

0.139133

0.9120

∆(LOG(FDI))

-0.062884

0.621998

-0.101101

0.9359

∆(LOG(FDI(-1)))

0.182445

0.694368

0.262750

0.8364

∆(LOG(FDI(-2)))

1.282146

0.440750

2.909010

0.0047

∆(LOG(FDI(-3)))

-1.039077

1.348420

-0.770589

0.5820

∆(LOG(FDI(-4)))

0.728846

1.292352

0.563968

0.6731

∆(LOG(COP))

1.048231

2.510426

0.417551

0.7482

∆(LOG(COP(-1)))

0.470608

3.048808

0.154358

0.9025

∆(LOG(COP(-2)))

-0.840868

3.094293

-0.271748

0.8311

∆(LOG(COP(-3)))

4.407670

1.662132

2.651817

0.0459

∆(LOG(COP(-4)))

-2.170452

0.782107

-2.775134

0.0277

∆(LOG(MS))

-1.414262

1.780247

-0.794419

0.5726

∆(LOG(MS(-1)))

-0.189465

2.225153

-0.085147

0.9459

∆(LOG(MS(-2)))

-1.267315

1.358915

-0.932593

0.5222

∆(LOG(MS(-3)))

2.637484

2.805728

3.273417

0.0007

∆(LOG(MS(-4)))

-0.661370

1.653421

-0.400001

0.7578

∆FD

-4.910847

7.625436

-0.644009

0.6358

∆FD(-1)

-7.285979

2.873664

-2.535432

0.0411

∆FD(-2)

-2.691866

4.313324

-0.624082

0.6448

∆FD(-3)

-10.12764

13.26092

-0.763721

0.5848

∆FD(-4)

-5.573549

6.155737

-0.905424

0.5316

∆INF

-0.004821

0.017894

-0.269398

0.8325

∆INF(-1)

0.007622

0.030482

0.250049

0.8440

∆INF(-2)

-0.030153

0.035009

-0.861313

0.5473

∆INF(-3)

0.013574

0.004960

2.716974

0.0390

∆INF(-4)

-0.009465

0.003189

-2.968015

0.0035

ECM(-1)

3.629113

27.98483

0.129681

0.9179

R-squared

0.999698

    Mean dependent var

13.07842

Adjusted R-squared

0.989724

    S.D. dependent var

2.109654

S.E. of regression

0.213852

    Akaike info criterion

-1.859553

Sum squared resid

0.045733

    Schwarz criterion

-0.348644

Log likelihood

66.54218

    F-statistic

100.2370

Durbin-Watson stat

3.332455

    Prob(F-statistic)

0.078957

Note: (1)∆ before any variable stands for first difference (2) Log for natural logarithms

Table 6. Short–run parsimonious growth model


Dependent Variable: LOG(RGDP)

Method: Least Squares

Date: 03/03/10   Time: 22:47

Sample(adjusted): 1975 2009

Included observations: 35 after adjusting endpoints

Variable

Coefficient

Std. Error

t-Statistic

Prob. 

∆(LOG(RGDP(-1)))

0.842740

0.095124

8.859394

0.0000

C

8.552860

4.079108

2.096747

0.0467

∆(LOG(GU))

0.599116

0.176355

3.397219

0.0024

∆(LOG(FDI(-2)))

0.459242

0.163337

2.811623

0.0097

∆(LOG(COP(-3)))

0.948873

0.363350

2.611461

0.0353

∆(LOG(COP(-4)))

-0.440683

0.340947

-1.292524

0.2085

∆(LOG(MS(-3)))

0.560026

0.164056

3.413630

0.0023

∆FD(-1)

-3.183967

1.088653

-2.924684

0.0074

∆INF(-3)

0.004189

0.001561

2.683536

0.0132

∆INF(-4)

0.000179

0.002805

0.063749

0.9497

ECM(-1)

-0.653550

0.201337

-3.246046

0.0034

R-squared

0.932238

    Mean dependent var

13.07842

Adjusted R-squared

0.889004

    S.D. dependent var

2.109654

S.E. of regression

0.221227

    Akaike info criterion

0.072019

Sum squared resid

1.174589

    Schwarz criterion

0.560843

Log likelihood

9.739670

    F-statistic

306.7907

Durbin-Watson stat

2.416817

    Prob(F-statistic)

0.000000

Note: (1)∆ before any variable stands for first difference (2) Log for natural logarithms

The coefficient of determination (adjusted R2) at 0.89 percent used to measure the goodness–of– fit of the estimated model, indicates that the model is reasonable fit in prediction (i.e. the model explains about 89% of the behaviour of growth in Nigeria) while the F–statistics value of 306.8 shows that the entire model is statistical significant. At 2.4, the Durbin–Watson statistics suggests the absence of autocorrelation. Furthermore, the central issue for the empirical analysis is stability of the parameters of the growth function or equation, which we reported in table 6. It is now a standard practice to incorporate short–run dynamics in testing for stability of the long–run parameters of the growth equation. To this end, this paper adopted Bahmani–Oskooee and Shin (2002), as well as applying the cumulative sum of recursive residuals (CUSUM) to the residual of the parsimonious model. The forecasting evaluation tests carried out equally showed that the model has a superior forecasting performance. This is revealed in the root mean squared error of 0.22 and all other forecasting indicators showed a similar picture. All these are as shown below.

Conclusion and policy recommendations.
The paper attempts to explore the link between gas utilization and the performance of the Nigerian economy. Selected indicators of gas utilization and growth were used. We examined the relationships between the variables by analyzing their long–run relationship and short–run dynamics. The econometric results from the error correction model show that gas utilization leads to economic growth in Nigeria. Overall, the unexpected signs of some of the explanatory variables could be attributed to a myriad of other factors, such as policy inconsistency and policy mortality, infrastructure failure as well as high risk and insecurity in South–South region (Niger Delta) could have their own effects on gas production and utilization and consequently on the Nigerian growth process.
From the foregoing, it is clear that the recent amnesty granted to militants in the Niger Delta region by President Yar’Adua as well as the reforms in the gas sector are steps in the right direction. This means that more revenue from the sale of gas can be channeled into productive use to enhance growth. Also, there is the need for the sustenance of the amnesty programme to the latter, as well as the adoption of further reforms in the gas sector that will lead to the development of more conducive environment for the operators. Although, the last few years have witnessed reduction in gas flaring, this would have to be pursued to a logical end in order to boost the confidence of the players in the private sector that Nigeria indeed is pursuing private sector led growth policies. Therefore, it must be noted that the variables considered in this study are not the only variables that can promote growth; there is the need to address other factors that continue to hinder growth in Nigeria generally and the gas sector.
Based on the above empirical evidence, it is therefore suggested that the present reforms in the gas sector be sustained given the fact that the Nigerian gas sector has a significant role to play in channeling resources for investment and productive processes in other sectors of the economy. Moreover, government should put in play appropriate macroeconomic policies that will boost productive activities and encourage investment in the sector, especially foreign direct investment as well as the use of gas for power generation as this will reduce drastically the cost of manufacturing and production and thereby leading to industrial growth and development. Such policies include tax incentive, security of lives and properties, maintenance of the stability of the naira, reduction in the cost of borrowing, research and local content, transparency as well as the strengthening of the Nigerian capital market.
References
Adam J. A. (2001): Implications of oil resource availability and exhaustibility. Nigeria Economic Society Annual Conference.
Adedeji, O. (2001): The size and sustainability of Nigerian current account deficits. IMF Working Paper 01/87. June.
Adewuyi, A. (2001): The implication of crude oil exploitation and export on the environment and level of economic growth and development in Nigeria. Nigeria Economic Society Annual Conference
Anyamele, D. O. (2000): Export–led growth in a public sector dominated economy: A macroeconomic model of Nigeria. Unpublished Ph.D. dissertation Department of Business and Economics, University of Maryland, Maryland, USA.
Audu, N. P. (2008): Globalization and Nigeria’s agricultural exports. Ikogho 5(3) 37 – 50
Bahmani–Oskooee, M. and Shin, S. (2002): Stability of the demand for money Korea. International Economic Journal, 16(2). 85 – 95.
Bennett, S. and Ossowski, R. (2003): ‘What goes up….’. Finance and Development. Washington: IMF. March.
Boame, A. K. (2002): Primary export –led growth: The evidence from Ghana. Journal of Economic Development 23(1). June. 34 – 44
Calamitsis, E. (2007): The need for stronger domestic policies and international support. Journal of Finance and Development 38 (4). December. 10 – 15
Central Bank of Nigeria (2008): Annual Report and Statement of Account
Central Bank of Nigeria (2010): Bullion
Central Bank of Nigeria (2009): Statistical Bulletin
Corden, W. M. and Neary, S. P. (1982): Booming sector deindustrialization in a small open economy. Economic Journal 92. December. 87 – 101.
Deaton, A. (1999): Commodity prices and growth in Africa. Journal of Economic Perspective 13. 23 – 40.
Dernburg, T. F. (2008): Global Macroeconomics. New York. Harper Collins 5ed.
Dinneya, G. (2006): Political Economy of democratization in Nigeria. Lagos. Concept Publication
ESMAP (2010): Strategic plan for Nigeria. Joint UNDP/World bank Energy Sector Management Assistant Programme. January
Evoh, C. (2002): Gas flaring, oil companies and politics in Nigeria. The Guardian. 18th February.
Fosu, A. K. (1990): Exports and economic growth: The African case. World Economy 18. 831 – 835
Fosu, A. K. (1996): Primary exports and economic growth in developing countries. World Economy 19. 465 – 475
Fosu, A. K. (2001): The global setting and African economic growth. Journal of African Economies 10. 282 – 310
Glyfalson, T., Herbertson, T. T. and Zoega, G. (1999): A mixed blessing: Natural resource and economic growth. Macroeconomic Dynamics 3.  June. 204 – 225.
Gujarati, D. N. (2004): Basic Econometric. India. Tata McGraw–Hill
Ibanga, I and Obi, P. B. (2001): Sustainability and depletion of resource use in Nigeria: The case of crude oil. Nigeria Economic Society Annual Conference
IMF, (2003): International Financial Statistics
Iwayemi, A. and Adenikuni, A. (2001): Energy–Environment: Economy linkage in Nigeria. Nigeria Economic Society Annual Conference
Manuel, T. A. (2003): Africa and the Washington consensus: Finding the right path. Finance and Development 4. September
Mlambo, K. (2005): Reviving FDI in Southern Africa: Constraints and Policies. Tunis, Blackwell Publishing Company.
Ndiyo, A. N. (2008): Poverty for sustainable development: A Communist based approach. Calabar. University of Calabar Press
Nyong, M. O. (2005): International Economics: Theory, policy and application. Calabar. Wusen Publications
OPEC, (2008): Annual Statistical Bulletin
Papyrakis, E. and Reyer, G. (2004): The resource curse hypothesis and its transmission channels. Journal of Comparative Economics 32.  181 – 193
Ross, M. (2001): Does oil hinder democracy? World Politics 53. 325 – 361.
Sachs, J. D. and Warner, A. M. (1995): Natural resource abundance and economic growth. Harvard. Institute for International Development, Development Discussion Paper 517. October
Shan, J. and Sun, F. (1998): On the export–led growth hypothesis: The econometriv evidence from China. Applied Economics 30. 1055 – 1065
Spalding, M. C. (2003): The Economics of Gas development in Saudi Arabia. World Energy Council.
Thornton, J. (1997): Exports and Economic growth: Evidence from 19th century Europe. Economic Letters 55. 235 – 240
Udah, E. B. (2007): Export–growth hypothesis: An econometric analysis of the Nigerian case. Ikogho 4(4) 70 – 84
UNCTAD, (2009a): Trade and Development Report. New York and Geneva: United Nations
UNCTAD, (2009b): World Investment Report. New York and Geneva: United Nations
Vincent, J. R. (1997): Resource depletion and economic sustainability in Malaysia. Environmental and Development Economics 2. 19 – 37
World Bank (2009): World Development indicators, Washington, DC.