E-ISSN 2281-4612 Academic Journal of Interdisciplinary Studies Vol 12 No 2 ISSN 2281-3993 www.richtmann.org March 2023 . Research Article © 2023 Luki et al. This is an open access article licensed under the Creative Commons Attribution-NonCommercial 4.0 International License (https://creativecommons.org/licenses/by-nc/4.0/) Received: 6 December 2022 / Accepted: 10 February 2023 / Published: 5 March 2023 The Impact of Oil and Gas Extraction in Sub-Saharan Africa: Evidence from Ghana Bayuasi Nammei Luki1 Abdallah Ali-Nakyea2 Hussein Salia3 Muntari Mahama4 1Department of Management and HRM, Faculty of Business, Ghana Communication Technology University, Accra, Ghana 2University of Ghana School of Law, University of Ghana, Legon, Accra, Ghana 3Department of Accounting, School of Business, Heritage Christian University College, Amasaman, Accra, Ghana 4Brunel Business School, Brunel University London, London, United Kingdom DOI: https://doi.org/10.36941/ajis-2023-0049 Abstract The topic of how oil and gas extraction impact on economic development of nations is well researched, especially African oil-producing countries. Literature has revealed that countries endowed with natural resources tend not to grow in terms of Gross Domestic Product (GDP) compared to their counterparts with limited or no such natural resources. This phenomenon has been characterized as the paradox of plenty. The study sought to evaluate the economic impact of oil and gas extraction in Ghana. The study utilized the quantitative research approach with secondary data from World Bank Development Indicators, 2019. Johansen Co-integration Approach was also used to evaluate the extent of causality between oil production (revenues) and the various economic variables. The Johansen Co-integration test revealed that oil revenue exerts a significant effect on Ghana’s GDP (4% higher than the average GDP before the production of oil and gas). Meanwhile, the average agriculture value-added and service value-added of Ghana for the period after the production of oil is lower than the average agriculture value-added and service value-added for the period before the production of oil in commercial quantities. The study used Ghana as a case study which somehow limits its findings. Future studies may use a panel model to do a cross-country analysis of the impact of oil and gas production on the various sectors of the respective countries. Government should formulate and expand policies (such as planting for food and jobs, and agricultural mechanization) that will help grow the agricultural and service sectors of the economy using oil revenues. The study also recommends that the Public Interest and Accountability Committee (PIAC) should closely monitor and collaborate with the petroleum revenue management institutions to ensure that oil revenues allocated to the government priority sectors, especially agriculture and service, are fully implemented. This study differs from prior studies (novel) because it aimed to determine the impact of oil and gas extraction on agricultural, industrial, and service sectors before and after oil and gas production. Keywords: Agriculture, extraction, industry, impact, oil and gas, service 282 E-ISSN 2281-4612 Academic Journal of Interdisciplinary Studies Vol 12 No 2 ISSN 2281-3993 www.richtmann.org March 2023 1. Introduction In 2007, the first commercially significant discovery of crude oil was made off the coast of Cape Three Points, allowing Ghana to join the league of oil-producing nations (Oteng-Adjei, 2011). Oteng-Adjei further commented that Ghana's oil find was timely because it occurred at a time when both foreign direct investment and foreign aid to Africa were on the verge of being cut. The former President of Ghana, John Agyekum Kufuor, whose administration saw commercial quantities of crude oil discovered, was reported as remarking, "without oil, we are still doing well" (Aryeetey & Ackah, 2018). To commemorate crude oil extraction in Ghana, the third President of the Fourth Republic of Ghana flipped on the valves of the Jubilee Field on December 15, 2010 (Oteng-Adjei, 2011; Adams et al., 2019). It is, however, worth noting that the actual production of crude oil began in December 2010 at the Jubilee Field with a daily production capacity of 12,000 bpd (Aryeetey & Ackah, 2018). The reserves of the Jubilee Field were estimated at 600 mb with an upside potential of 1.8 billion barrels (Aryeetey & Ackah, 2018; Ayelazuno, 2014) making it one of the most prolific discoveries off the Gulf of Guinea in recent times (González, 2016; Oshionebo, 2018). The World Bank in 2009 estimated that Ghana could earn as much as $20 billion from the Jubilee Field Production alone between 2012 – 2030. As of January 1, 2018, the Central Intelligence Agency Factbook (CIA, 2018) estimated Ghana’s proven crude reserves at 660,000,000 barrels. Tullow Plc operates the field through its subsidiary Tullow Ghana Limited. Actual production started in 2010 with 55,000 bpd and has since increased to about 191,000 bpd as of the last quarter of 2020. It is expected that when the Jubilee Field reaches its plateau in the future, its production shall increase to 240,000 bpd (BBC News, 2010). Until the Tweneboa-Enyenra-Ntomme (TEN) and Sankofa fields began operating in 2016 and 2017 respectively, Ghana only had one field, the Jubilee Field (Aryeetey & Ackah, 2018). According to them, the TEN field is expected to produce roughly 80,000 barrels of oil and 180 million cubic feet of natural gas per day at their respective production peak. TEN has 239 million barrels of crude oil and 360 billion cubic feet of natural gas in its reserves (Aryeetey & Ackah, 2018). According to Aryeetey and Ackah (2018), the Sankofa field has a crude reserve of 204 million barrels and a natural gas reserve of 1,071 billion cubic feet. PIAC (2021) revealed that Ghana earned US$7.36 billion of its oil production as revenue. Currently, only Tullow Ghana Limited and ENI and their partners are producing crude oil in Ghana. Their combined production capacity is 55,050,391 barrels (PIAC, 2021). Aker Energy and Springfield Ghana are also expected to follow the production lead pioneered by Tullow Ghana Limited and ENI. As of the end of 2021, PIAC reports that as of the end of 2021, Ghana’s daily production stood at 88 74,891.73 barrels. This is about a 17.7% reduction compared with the 2020 production figure of 66,926,806 barrels (PIAC, 2021). Table 1 shows the oil and gas revenue earned since the inception of commercial production of crude oil in Ghana. Table 1: Ghana’s Total Oil and Gas Revenue Earned From 2011 to 2021 Year Revenue Earned (millions of US$) 2011 444.12 2012 541.62 2013 846.77 2014 978.02 2015 396.17 2016 247.18 2017 555.33 2018 977.09 2019 925.04 2020 666.39 2021 783.33 Total US$7.36 billion Source: PIAC (2021) 283 E-ISSN 2281-4612 Academic Journal of Interdisciplinary Studies Vol 12 No 2 ISSN 2281-3993 www.richtmann.org March 2023 Akinlo (2012) also assessed the contribution of Nigeria’s oil industry to its economic development from 1960 to 2009 and confirmed that the petroleum industry has a critical role to play in Nigeria’s economic development should they encourage diversification and private sector development. A study by Ingen et al. (2014) confirmed Akinlo’s (2012) assertion and added that petroleum revenue contributes almost 90% of Nigeria's export earnings. Similarly, the result from Yeboua et al. (2022) studies ranked Nigeria as Africa's top producer and exporter of crude oil. Despite the abundance of hydrocarbons in Nigeria and the enormous money generated from oil and gas production, the country is not considerably better off (Ingen et al., 2014). In other words, Nigeria has been unable to achieve significant levels of economic development by relying on oil earnings. Also, between 1974 and 2008, Asekunowo and Olaiya (2012) conducted a similar analysis of Nigeria's oil and gas industry, confirming the oil curse concept. They attributed the oil curse in Nigeria to weak institutions, excessive spending, excessive borrowing and revenue volatility. Shaw (2013) also investigated the resource curse hypothesis in Azerbaijan and found that the petroleum resource endowment is the reason for their low economic growth. This is evidenced in the quote of Friedman (NYT, 1/30/2005) when he said ‘You give me $18-a-barrel oil and I will give you political and economic reforms from Algeria to Iran’. Ross (2012) did a similar study and concluded that the oil curse results from bad democracy, weak institutions and civil war. Ross’s findings confirmed the works of Asekunowo and Olaiya (2012). Therefore, it is imperative that governments institute mechanisms to ensure the right institutional structures are established to efficiently manage petroleum resources. Consequently, the role of institutions in the management of petroleum resources was underscored by Sarmidi et al. (2014) in their study of the relationship between natural resource abundance and economic growth. Sarmidi and his colleagues used data from over 90 countries from 1984-2005. They confirmed that natural resource abundance affects economic growth positively after a threshold level of institutional quality. Impliedly, sovereign nations must note that the discovery or exploration of natural resources, including oil and gas, does not necessarily guarantee economic growth and development if the effort is not backed by leadership’s will to empower quality institutions to manage the resources. Relatedly, Zuo and Schieffer (2014) conducted a study in China and confirmed a negative relationship between natural resource endowment and economic development in the resource regions of China. Mavrotas et al. (2011) also researched on the relationship between natural resource endowment and economic growth among developing countries for thirty years (1970-2000) and confirmed an incongruous relationship between the natural resource endowment and economic growth among those countries. Several studies have been conducted using samples drawn from numerous African oil producing and exporting countries (Tiba, 2019; Frikha & Tiba, 2019; Ding & Field, 2005) confirming the negative relationship between petroleum resource abundance and economic development (Dogan et al., 2020). On the contrary, Alexeev and Chernyavskiy (2015) in their study showed a positive association between natural resource endowment and economic growth at the regional level in Russia. They opined that there was no negative relationship between natural resource endowment and economic growth. Mideksa (2013), Moradbeigi and Law (2017) and Dogan et al. (2020), likewise, investigated the economic impact of natural resource endowment on economic growth and found that natural resource richness boosts GDP. Ji et al. (2014) conducted a similar study in China, comparing natural resource endowment, institutional quality, and economic growth from 1990 to 2008, and found that natural resources have a positive link with economic growth due to institutional quality at the province level. Natural resources, according to Gerelmaa and Kotani (2016), had a positive relationship with economic growth between 1990 and 2010 in countries rich in natural resources. The same relationship was found to be true for Sierra Leone (Dogan et al., 2020). A study by Moshiri and Hayati (2017) of 149 countries, also showed that the negative relationship between natural resource endowment and economic growth was invalid. Their conclusions contradict many studies including Sinha and Sengupta’s (2019) research in the Asia pacific countries and Amini’s (2018) research among 284 E-ISSN 2281-4612 Academic Journal of Interdisciplinary Studies Vol 12 No 2 ISSN 2281-3993 www.richtmann.org March 2023 22 developed countries and 61 developing countries. Africa holds about 10 per cent of global proven oil and gas reserves. Africa produces 10.9% (8.4 million bpd) of global daily consumption (Garside, 2020). Natural resources, especially petroleum resources, are inputs of the natural environment and can be extracted for socio-economic development (Zalle, 2019). Oil and gas resources have the potential of generating revenues through their exports. Natural resources like oil and gas are precious assets that contribute significantly to economic growth, particularly in terms of revenue, employment, and infrastructure development (Peach & Starbuck, 2011; Adabor et al., 2021). This assertion was first established by Adam Smith and David Ricardo, that countries that are blessed with oil and gas resources would perform better than their counterparts with limited or scanty petroleum resources if those countries can leverage on the petroleum endowment by diversifying into agricultural, services and other sectors of the economy (Humphrey et al., 2007; Adabor et al., 2021). Studies have shown that 40 per cent of Ghana’s total foreign revenue is contributed by its natural resources such as gold, oil and gas, diamond, and bauxite (Adabor et al., 2021). There is, however, a paucity of empirical research on the impact of oil and gas extraction on economic growth in Ghana. The study, therefore, aims to evaluate the economic effects of oil and gas extraction in Ghana, using Johansen Co-integration Approach. This study differs from prior studies (Dah & Sulemana, 2010) because it aims to determine the impact of oil and gas extraction on agricultural, industrial, and service sectors before and after oil and gas production. Using data from the World Bank’s World Development Indicators (2019), the paper explores the existence of variations in economic growth between periods before and after oil and gas extraction in Ghana. 2. Theoretical Background/Literature In determining the influence of petroleum resource endowment on economic growth, researchers must first investigate the factors that link natural resource endowment (oil and gas) to bad economic performance (Majumder et al., 2020). Extant literature (Smith, 2004; Larsen, 2006; Brunnschweiler & Bulte, 2006; Brunnschweiler, 2007) uncovered a positive link between the discovery of petroleum resources and the economic growth of many nations. Given the preceding conclusions, one can infer that the discovery of petroleum resources in Africa (developing countries) would benefit the economies of those African countries, particularly those in sub-Saharan Africa. On the contrary, the likelihood that oil and gas discovery in Africa would be a curse instead of a blessing is a topic that has been debated among academics, policymakers and analysts within the oil and gas industry of these oil-rich countries (Ablo, 2015). The oil curse theory, also known as the paradox of plenty, has a large body of literature (Katz et al., 2004; Kopinski et al., 2013; Satti et al., 2014; Sigman et al., 2022). Extant literature also revealed that countries blessed with natural resources, especially petroleum resources in Africa, are prone to corruption, rent-seeking behaviour, petroleum revenue misallocation and utilization, bad governance and socio-economic and political crises that would subsequently weaken democratic stability and economic development (Adams et al., 2019; Majumder et al., 2020). Although the oil curse hypothesis has attracted significant academic attention, especially in the energy industry, most of these works focused on transparency and accountability as the drivers for sustainable management of petroleum resource revenues in underdeveloped settings (Van der Pleog, 2011; Collier & Goderis, 2007; Ackah-Baidoo, 2012; Corrigan, 2014; Adams et al., 2019). The pioneering studies of the resource curse hypothesis (Sachs, 1999; Sachs & Warner, 1995) are considered seminal studies that attempted to establish the link between economic growth and natural resource endowment (Frikha & Tiba, 2019). Countries blessed with oil and gas resources have some special features. Oil and gas resources generate revenue and can, thus, be an engine for industrialization (Satti et al., 2014; Stevens & Dietsche, 2008; Hasanov et al., 2022). Natural resource endowment (oil and gas) has been shown to have both direct and indirect effects on economic development in studies. According to Satti et al. 285 E-ISSN 2281-4612 Academic Journal of Interdisciplinary Studies Vol 12 No 2 ISSN 2281-3993 www.richtmann.org March 2023 (2014), the effects of natural resource endowment on economic growth are inconclusive because some empirical studies have found a positive relationship between natural resource endowment and economic growth (Brunnschweiler & Bulte, 2006; Brunnschweiler, 2007; Poncian, 2019), while others revealed a negative relationship (Bhattacharyya & Collier, 2014). Satti et. al. (2014) concluded that different investigations into the relationship between natural resource endowment and economic development presented mixed results when it came to proving the idea's veracity. This statement was backed up by Poncian (2019). Studies conducted by Majumder et al. (2020) on the sustainable management of petroleum revenues found that it was contrary to natural intuition that countries blessed with an abundance of natural resources such as oil and gas will be poor in terms of GDP. This scenario according to Frikha and Tiba (2019) imposed what they call a conceptual puzzle because the assumption is that natural resource endowment must enhance the economic growth rates of those resource-rich nations. 2.1 Empirical literature and hypotheses Using panel data for about 95 countries from 1980-2010, Majumder et al. (2020) sought to assess the relationship between trade openness, natural resource endowment and economic growth. The study concluded that trade openness was found to be a mediating factor to ensuring that natural resource endowment inures to the benefit of oil-producing and exporting countries. Consistent with Ayelazuno (2014), nations like Switzerland, Japan, and the Asian Tigers, with no or limited natural resource abundance, recorded better economic growth rates than their counterparts such as Nigeria, Angola, and Equatorial Guinea that are endowed with natural resources. This was later confirmed by Frikha and Tiba’s (2019) conceptual puzzle scenario. The growth rates of these ‘poor’ countries in terms of per capita GDP increased about three times faster than those countries endowed with natural resources (Majumder et al., 2020). The studies of Stevens and Dietsche (2008), Frikha and Tiba (2019), and Majumder et al. (2020) attributed this paradox in resource-rich countries to a lack of diversification and weak institutional settings in oil-producing and exporting countries. Likewise, Zuo and Schieffer’s (2014) investigations on the China resource curse hypothesis showed an inverse relationship between natural resource endowment and economic development in the resource regions of the country. They attributed the resource curse in China to the crowding-out effect of education and Research and Development. Mavrotas et al. (2011) also conducted a study on the relationship between natural resource endowment and economic growth among developing countries for a thirty (30) year period (1970 – 2000). Their findings confirmed the natural resource curse hypothesis among those countries. Many other scholarly works in Africa (Tiba, 2019; Frikha & Tiba, 2019; Ding & Field, 2005) have confirmed the oil curse hypothesis (Dogan et al., 2020). However, Alexeev and Chernyavskiy (2015) found a positive association between natural resource endowment and economic growth at the regional level in Russia, indicating that there was no negative relationship between natural resource endowment and economic growth. Mideksa (2013) investigated the economic impact of natural resource endowment on economic growth and found that natural resource richness boosts GDP (Dogan et al., 2020). Ji et al. (2014) conducted a similar study in China, comparing natural resource endowment, institutional quality, and economic growth from 1990 to 2008, and found that natural resources have a positive link with economic growth due to institutional quality at the province level (Dogan et al., 2020). Natural resources, in the 1990s, according to Gerelmaa and Kotani (2016), had a positive relationship with economic growth from 1990 to 2010. The same relationship was found to be true for Sierra Leone (Dogan et al., 2020). A study by Moshiri and Hayati (2017) of 149 countries showed that the negative relationship between natural resource endowment and economic growth was invalid. Their conclusions contradict many studies, including the research of Sinha and Sengupta (2019) in the Asia-Pacific countries and that of Amini (2018) in 22 developed countries and 61 developing countries. Considering these mixed findings, and the fact that Ghana is a major oil-producing country in Africa with the government’s other priority sectors being agriculture, services and industry (Ferreira 286 E-ISSN 2281-4612 Academic Journal of Interdisciplinary Studies Vol 12 No 2 ISSN 2281-3993 www.richtmann.org March 2023 et al., 2022), it is important that the link between oil and gas production and the other sectors is uncovered. Therefore, the paper seeks to evaluate the economic effects of oil and gas extraction in Ghana with particular emphasis on Agriculture, Industry, and Service sectors using data from the World Bank’s World Development Indicators (2019). To do this, the following hypotheses were postulated. H1 the average GDP of Ghana for the period after the production of oil in commercial quantities is higher than the average GDP for the period before the production of oil in commercial quantities. H2 the average agriculture value-added of Ghana for the period after the production of oil in commercial quantities is higher than the average agriculture value-added for the period before the production of oil in commercial quantities. H3 the average service value-added of Ghana for the period after the production of oil in commercial quantities is greater than the average service value-added for the period before the production of oil in commercial quantities. H4 the average industry value-added of Ghana for the period after the production of oil in commercial quantities is greater than the average industry value-added for the period before the production of oil in commercial quantities. 3. Methodology and Data The following econometric specification (Johansen Co-integration Approach) was utilized: 3.1 Specifying the econometric Model The models for GDP, agriculture, industry, and service outputs are specified below: lnGDPt =ψ 0 +ψ1DOILt +ψ 2 lnEXRt +ψ 3 ln INFt +ψ 4 lnOILPt +ε t …………….(3.1) Where DOIL is a dummy variable that captures the period before oil discovery and the period after oil discovery, EXR is the exchange rate, INF is inflation and OILP is the oil price lnAOt =ψ 0 +ψ1DOILt +ψ lnEXR +ψ ln INF +ψ lnOILP +ε 2 t 3 t 4 t t ............(3.2) Where AO is agricultural output ln IO t = ψ 0 +ψ 1DOIL t +ψ 2 ln EXR t +ψ 3 ln INF t +ψ 4 ln OILP t + ε t ....(3.3) Where IO is the industrial output lnSOt =ψ 0 +ψ1DOILt +ψ 2 lnEXRt +ψ 3 ln INFt +ψ 4 lnOILPt +ε t ............................(3.4) Where SO is the service output 3.2 Estimation Techniques 3.2.1 Unit Root Tests The study's time series variables are checked for stationarity using a unit root test. There is a chance that a spurious relationship exists in time series that show growing or diminishing trends (Wooldridge, 2013). To ensure that there is no misleading link between the variables, the Augmented Dickey Fuller (ADF) test was utilized. The ADF test is an advanced form of the Dickey Fuller test and has been known to solve complex models. There are three basic regression models which are shown below: No constant, no trend: Δyt = γyt-1 + ut ...........................................................................(3.5) Constant, no trend: Δyt = α + γyt-1 + ut ..........................................................................(3.6) Constant and trend: Δyt = α + γyt-1 + λt + ut ..................................................................(3.7) A lagged value of the dependent variable is included in the model when the ADF test is 287 E-ISSN 2281-4612 Academic Journal of Interdisciplinary Studies Vol 12 No 2 ISSN 2281-3993 www.richtmann.org March 2023 employed. This is shown below as per the three basic regression models stated above: No constant, no trend: Δyt = γyt-1 + δsΔyt-s + ut .........................................................(3.8) Constant, no trend: Δyt = α + γyt-1 + δsΔyt-s + ut ........................................................(3.9) Constant and trend: Δyt = α + γyt-1 + λt + δsΔyt-s + ut ................................................(3.10) From the equation α is the estimated parameter, δ represents the estimated parameters of the differenced values of the lagged variables. With the ADF test, there exists a null hypothesis and an alternative hypothesis. The null hypothesis shows the existence of a unit root (H0: α = 0) in contrast to the alternative hypothesis where there is no unit root. If the test concludes on failing to reject the null hypothesis, then there is the existence of non-stationarity. On the other hand, when the test shows a rejection of the null hypothesis, then the variable is stationary or simply put, there is the existence of stationarity. 3.2.2 Cointegration Tests The Johansen co-integration test is used in this investigation. This is a vector co-integration test procedure. If the data set contains two or more time series, it has an advantage over the Engle- Granger and Phillips-Ouliaris methods in that it can estimate more than one co-integration relationship (Creswell, 2014). Instead of using OLS estimation, Johansen's technique uses maximum likelihood estimation to construct cointegrated variables. This method significantly relies on the relationship between a matrix's rank and its characteristic roots. The maximum likelihood estimation was developed by Johansen using sequential tests to determine the number of cointegrating vectors. In the sense that it relies entirely on maximum likelihood rather than relying on least squares, his method can be considered a secondary generation approach. As a result, he suggests the trace test and the maximum eigenvalue test as two different likelihood ratio tests. The estimation and test procedures are outlined below. Yt = Π1Yt−1 + Π 2Yt−2 + ...+ Πρ X t−ρ + ut .........................................(3.11) where Yt is an n X 1 vector of variables that are integrated of order one, that is, I(1), ut is an nX1 vector of innovations while Π1 through Πρ are m X m coefficient matrices. Reparametrizing equation 3.10, that is, subtracting Yt-1 on both sides, leads to ΔYt = Γ1ΔYt−1 + Γ2ΔYt−2 + ...+ Γρ−1Yt−ρ+1 − ΓρYt− put ............................(3.12) Γ1 = Π1 − I Γ = Π − Γ Γ =Π −Γ Γ = I −Π1 −Π 2 − ....−Π2 2 1 3 3 2 ρ The impact matrix is a matrix that determines the extent to which the system is cointegrated. The trace test and the highest eigenvalue are two likelihood ratio tests suggested by Johansen. 3.2.3 The Trace Examination The trace test compares the alternative hypothesis of n cointegrating vectors to the null hypothesis of r cointegrating vectors. The test statistic is calculated as follows: n( jtrace = −T ln(1− λˆi ) i=r+1 ………………………………………………………………………...(3.13) 3.2.4 The Maximum Eigenvalue On the other hand, the maximum eigenvalue test compares the null hypothesis of r cointegrating vectors to the alternative hypothesis of (r+1) cointegrating vectors. Its test statistic is calculated as follows: jmax = −T (1− λˆr+1) ……………………………………………………………………….…....(3.14) where T is the sample size and λ̂i is the ith largest canonical correlation. 288 E-ISSN 2281-4612 Academic Journal of Interdisciplinary Studies Vol 12 No 2 ISSN 2281-3993 www.richtmann.org March 2023 3.2.5 Fully Modified OLS Estimator To evaluate the effect of oil and gas extraction in Ghana, the study uses a fully modified OLS (FMOLS) model. Because the model variables are cointegrated, a long-run FMOLS estimate of the GDP model can be used to display them. Phillip and Hansen proposed the Fully Modified Ordinary Least Square method (FMOLS) first (1990). To reduce the long-run link between the cointegrating equation and the innovations, the method uses a semi-parametric correction. In other words, FMOLS modifies least squares to explicate serial correlation. 4. Empirical Results 4.1 Stationary test results This test was carried out to ensure that there was no integration of order greater than 1(2). This assures that the final product is free of erroneous results. The presence of a unit root was checked, as well as the sequence of integration, using the Augmented Dickey-fuller (ADF) test. Table 2 summarizes the findings of the ADF test. The following econometric specification (Johansen Cointegration Approach) was utilized: 4.2 Specifying the econometric Model The models for GDP, agriculture, industry, and service outputs are specified below: lnGDPt =ψ 0 +ψ1DOILt +ψ 2 lnEXRt +ψ 3 ln INFt +ψ 4 lnOILPt +ε t …………….(3.1) Where DOIL is a dummy variable that captures the period before oil discovery and the period after oil discovery, EXR is exchange rate, INF is inflation and OILP is oil price lnAOt =ψ 0 +ψ1DOILt +ψ 2 lnEXRt +ψ 3 ln INFt +ψ 4 lnOILPt +ε t ...................(3.2) Where AO is agricultural output ln IOt =ψ 0 +ψ1DOILt +ψ 2 lnEXRt +ψ 3 ln INFt +ψ 4 lnOILPt +ε t ........................(3.3) Where IO is industrial output lnSOt =ψ 0 +ψ1DOILt +ψ 2 lnEXRt +ψ 3 ln INFt +ψ 4 lnOILPt +ε t .........................(3.4) Where SO is service output 4.3 Estimation Techniques 4.3.1 Unit Root Tests he study’s time series variables are checked for stationarity using a unit root test. There is a chance that a spurious relationship exists in time series that show growing or diminishing trends (Wooldridge, 2013). To ensure that there is no misleading link between the variables, the Augmented Dickey Fuller (ADF) test was utilized. The ADF test is an advanced form of the Dickey Fuller tests and has been known to solve complex models. There are three basic regression models which are shown below: No constant, no trend: Δyt = γyt-1 + ut .............................................................................(3.5) Constant, no trend: Δyt = α + γyt-1 + ut ...........................................................................(3.6) Constant and trend: Δyt = α + γyt-1 + λt + ut ...................................................................(3.7) A lagged value of the dependent variable is included in the model when the ADF test is employed. This is shown below as per the three basic regression models stated above: No constant, no trend: Δyt = γyt-1 + δsΔyt-s + ut .............................................................(3.8) 289 E-ISSN 2281-4612 Academic Journal of Interdisciplinary Studies Vol 12 No 2 ISSN 2281-3993 www.richtmann.org March 2023 Constant, no trend: Δyt = α + γyt-1 + δsΔyt-s + ut ............................................................(3.9) Constant and trend: Δyt = α + γyt-1 + λt + δsΔyt-s + ut ....................................................(3.10) From the equation α is the estimated parameter, δ represents the estimated parameters of the differenced values of the lagged variables. With the ADF test, there exists a null hypothesis and an alternative hypothesis. The null hypothesis shows the existence of a unit root (H0: α = 0) in contrast to the alternative hypothesis where there is no unit root. If the test concludes on failing to reject the null hypothesis, then there is the existence of non-stationarity. On the other hand, when the test shows a rejection of the null hypothesis, then the variable is stationary or simply put, there is the existence of stationarity. 4.3.2 Co-integration Tests The Johansen co-integration test is used in this investigation. This is a vector co-integration test procedure. If the data set contains two or more time series, it has an advantage over the Engle- Granger and Phillips-Ouliaris methods in that it can estimate more than one co-integration relationship (Creswell, 2014). Instead of using OLS estimation, Johansen’s technique uses maximum likelihood estimation to construct cointegrated variables. This method significantly relies on the relationship between a matrix’s rank and its characteristic roots. The maximum likelihood estimation was developed by Johansen using sequential tests to determine the number of cointegrating vectors. In the sense that it relies entirely on maximum likelihood rather than relying on least squares, his method can be considered a secondary generation approach. As a result, he suggests the trace test and the maximum eigenvalue test as two different likelihood ratio tests. The estimation and test procedures are outlined below. Y = Π Y + Π Y + ...+ Π X + u t 1 t−1 2 t−2 ρ t−ρ t .......................................................................(3.11) where Yt is an n X 1 vector of variables that are integrated of order one, that is, I(1), ut is an nX1 vector of innovations while Π1 through Πρ are m X m coefficient matrices. Reparametrizing equation 3.10, that is, subtracting Yt-1 on both sides, leads to ΔYt = Γ1ΔYt−1 + Γ2ΔYt−2 + ...+ Γρ−1Yt−ρ+1 − ΓρYt− put ................................................(3.12) Γ1 = Π1 − I Γ2 = Π 2 − Γ1 Γ =Π −Γ Γ = I −Π −Π3 3 2 1 2 − ....−Πρ The impact matrix is a matrix that determines the extent to which the system is cointegrated. The trace test and the highest eigenvalue are two likelihood ratio tests suggested by Johansen. 4.3.3 The Trace Examination The trace test compares the alternative hypothesis of n cointegrating vectors to the null hypothesis of r cointegrating vectors. The test statistic is calculated as follows: n( jtrace = −T ln(1− λˆi ) i=r+1 …………………………………………………………..(3.13) 4.3.4 The Maximum Eigenvalue On the other hand, the maximum eigenvalue test compares the null hypothesis of r cointegrating vectors to the alternative hypothesis of (r+1) cointegrating vectors. Its test statistic is calculated as follows: jmax = −T (1− λˆr+1 ) …………………………………………………………………….(3.14) where T is the sample size, and λ̂i is the ith largest canonical correlation. 290 E-ISSN 2281-4612 Academic Journal of Interdisciplinary Studies Vol 12 No 2 ISSN 2281-3993 www.richtmann.org March 2023 4.3.5 Fully Modified OLS Estimator To evaluate the effect of oil and gas extraction in Ghana, the study uses a fully modified OLS (FMOLS) model. Because the model variables are cointegrated, a long-run FMOLS estimate of the GDP model can be used to display them. Phillip and Hansen proposed the Fully Modified Ordinary Least Square method (FMOLS) first (1990). To reduce the long-run link between the cointegrating equation and the innovations, the method uses a semi-parametric correction. In other words, FMOLS modifies least squares to explicate serial correlation. 5. Empirical Results 5.1 Stationary test results This test was carried out to ensure that there was no integration of order greater than 1(2). This assures that the final product is free of erroneous results. The presence of a unit root was checked, as well as the sequence of integration, using the Augmented Dickey-fuller (ADF) test. Table 2 summarizes the findings of the ADF test. Table 2: Augmented Dickey Fuller (ADF) Test Level First Difference Variables Intercept Trend Intercept Trend LNAGRIC -1.2872 -4.0244** -2.7526* -3.2427* LNEXR 0.0411 -2.5036 -2.9592** -3.2179* LNGDP -1.2643 -1.8965 -3.2037** -3.3074* LNINDUSTRY -1.4367 -2.4496 -2.7045* -3.7992** LNINF -1.4913 -2.1729 -4.5944*** -4.5244*** LNOILPRICE -1.9764 -1.3932 -6.6241*** -6.8887*** LNSERVICE -2.3114 -1.9888 -2.7290* -3.7827** Note: that the symbols ***, ** and * denote statistical significance at the 1%, 5% and 10% levels respectively. Source: Authors’ calculations based on World Development Indicators data from the World Bank (2019) It is shown in Table 2 that agriculture value-added is non-stationary at the level without trend and stationary with the trend but became stationary after the first difference, thus integrated of order one, I (1). Both exchange rate and GDP contain unit root at the level but the unit root was removed after the first difference, thus, both variables are I (1). The results further revealed that industry value- added, inflation, oil price and service value-added are all non-stationary at the level but became stationary after the first difference, thus, integrated of order zero, I (1). Given that the variables are I (1) series, the Johansen Co-integration is the appropriate approach for the estimation in this study. 5.2 Co-integration Results This section discusses the Johansen co-integration results on the effect of oil revenue on GDP and sector outputs in Ghana. The results are displayed in Table 3. The Johansen co-integration result based on trace statistics indicates that there exists 1 co-integrating equation for the GDP model at the 5 per cent significance level. In other words, there is co- integration at rank 1. Thus, the trace test supports the hypothesis that there exists a long-run relationship between GDP and oil revenue, exchange rate, inflation, and oil price. The co-integration test based on the Max-Eigen test statistic indicates that there exists 1 co-integrating equation at the 5 291 E-ISSN 2281-4612 Academic Journal of Interdisciplinary Studies Vol 12 No 2 ISSN 2281-3993 www.richtmann.org March 2023 per cent level of significance. That is, there is co-integration at rank 1. Thus, the Max-Eigen statistics support the hypothesis that there exists a long-run relationship between GDP and oil revenue, exchange rate, inflation, and oil price. Table 3: Results of Johansen’s Test for Co-integration among Common Components GDP Model Agric Sector Model Industry Sector Model Service Sector Model H Trace Max-Eigen Trace Max-Eigen Trace Max-Eigen Trace Max-Eigen 0 Statistic Value Statistic Value Statistic Value Statistic Value None 70.4984** 35.5344** 82.3156** 38.4197** 97.6517** 47.6469** 88.7173** 39.4150** At most 1 34.9640 16.7356 43.8959 23.6429 50.0049** 18.7374 49.3024** 22.3968 At most 2 18.2284 12.5947 20.2530 13.7710 31.2674** 17.0892 26.9056 13.6238 At most 3 5.6338 3.6984 6.4819 4.9759 14.1783* 8.3488 13.28181 7.4872 At most 4 1.9354 1.9354 1.5060 1.5060 5.8295** 5.8295** 5.7946 5.7946 Note: that the symbols ***, ** and * denote statistical significance at the 1%, 5% and 10% levels respectively Source: Authors’ calculations based on World Development Indicators data from the World Bank (2019) The Johansen co-integration result based on trace statistics indicates that there exists 1 co-integrating equation for the agriculture value-added model at the 5 per cent significance level. That is, there is co-integration at rank 1. Thus, the trace test supports the hypothesis that there exists a long-run relationship between agriculture value-added and oil revenue, exchange rate, inflation, and oil price. The co-integration test based on the Max-Eigen test statistic indicates that there exists 1 co- integrating equation at the 5 per cent level of significance. In other words, there is co-integration at rank 1. Thus, the Max-Eigen statistics support the hypothesis that there exists a long-run relationship between agriculture value-added and oil revenue, exchange rate, inflation, and oil price. The Johansen co-integration result based on trace statistics indicates that there exist 3 co- integrating equations for the industry value-added model at the 5 per cent significance level. That is, there is co-integration at ranks 1, 2 and 3. Thus, the trace test supports the hypothesis that there exists a long-run relationship between industry value-added and oil revenue, exchange rate, inflation, and oil price. The co-integrating test based on the Max-Eigen test statistic indicates that there exists 1 co-integrating equation at the 5 per cent level of significance. In other words, there is co-integration at rank 1. Thus, the Max-Eigen statistics support the hypothesis that there exists a long-run relationship between industry value-added and oil revenue, exchange rate, inflation, and oil price. Finally, the Johansen co-integration result based on trace statistics indicates that there exist 2 co-integrating equations for the service value-added model at the 5 per cent significance level. That is, there is co-integration at ranks 1 and 2. Thus, the trace test supports the hypothesis that there exists a long-run relationship between service value-added and oil revenue, exchange rate, inflation, and oil price. The co-integrating test based on the Max-Eigen test statistic indicates that there exists 1 co-integrating equation at the 5 per cent level of significance. In other words, there is co-integration at rank 1. Thus, the Max-Eigen statistics support the hypothesis that there exists a long-run relationship between service value-added and oil revenue, exchange rate, inflation, and oil price. 5.3 Long-run Estimations The results show that oil revenue has a significant effect on GDP, agricultural value-added, industry value-added and service value-added at the 10 per cent and 1 per cent levels of significance respectively. Focusing on the GDP model, the positive sign of the coefficient indicates that the average GDP of Ghana for the period after the production of oil in commercial quantities is higher than the average GDP for the period before the production of oil in commercial quantities. This shows that oil and gas production (revenues) have a positive impact on Ghana's economic growth 292 E-ISSN 2281-4612 Academic Journal of Interdisciplinary Studies Vol 12 No 2 ISSN 2281-3993 www.richtmann.org March 2023 (GDP). The magnitude of the coefficient shows that the average GDP for the period after the production of oil in commercial quantities is 4% higher than the average GDP before the production of oil in commercial quantities. The findings are incongruous with the works of Moradbeigi and Law (2017) and Cockx and Franken (2016). This, however, supports the oil and gas bless hypothesis that the discovery of oil stimulates a nation's revenue thereby improving the level of infrastructural development and consequently increasing GDP (Adabor et al., 2021). Hence, the study accepted the hypothesis (H1) that there is a significant difference in the GDP performance of Ghana before and after oil and gas production. The comments are also in line with empirical studies of Luki (2016), and Oteng-Adjei (2011) that oil and gas exploration and extraction have a positive effect on the economies (GDP) of oil-producing and exporting countries. Focusing on the agriculture value-added and service value models, the negative sign of the coefficients indicates that the average agriculture value-added and service value-added of Ghana for the period after the production of oil in commercial quantities are lower than the average agriculture value-added and service value-added for the period before production of oil in commercial quantities. The magnitude of the coefficients shows that the average agriculture value-added and service value for the period after the production of oil in commercial quantities is 12% and 31% lower than the average agriculture value-added and service value-added for the period before the production of oil in commercial quantities respectively. Hence, the study rejects the hypotheses that: H2 the average agriculture value-added of Ghana for the period after the production of oil in commercial quantities is higher than the average agriculture value-added for the period before the production of oil in commercial quantities. H3 the average service value-added of Ghana for the period after the production of oil in commercial quantities is greater than the average service value-added for the period before the production of oil in commercial quantities. However, the positive sign of the coefficient for industry value-added indicates that the average industry value-added of Ghana for the period after the production of oil in commercial quantities is higher than the average industry value-added for the period before the production of oil in commercial quantities. The magnitude of the coefficient shows that the average industry value-added for the period after the production of oil in commercial quantities is 54 per cent higher than the average industry value added before the production of oil in commercial quantities. As a result, the study supports the hypothesis that H4 the average industry value-added of Ghana for the period after the production of oil in commercial quantities is greater than the average industry value-added for the period before the production of oil in commercial quantities. The results are displayed in Table 4. Table 4: Fully-Modified Estimation of GDP, Agric, Industry and Service GDP Model Agric Sector Model Industry Sector Model Service Sector Model Variables Coefficient Coefficient Coefficient Coefficient DOILREV 0.0411* -0.1158*** 0.5449*** -0.3151*** (0.0244) (0.0407) (0.0818) (0.0926) LNEXR 0.4308*** -0.2212** -0.0929 0.2580*** (0.0173) (0.0289) (0.0580) (0.0657) LNINF -0.0757*** 0.1441*** 0.0025 -0.0869 (0.0204) (0.0339) (0.0681) (0.0771) LNOIL PRICE 0.1453*** -0.0879*** -0.1867*** 0.2950*** (0.0149) (0.0249) (0.0500) (0.0566) CONSTANT 24.6413*** 3.4118*** 3.7916*** 2.7331*** (0.0998) (0.1663) (0.3341) (0.3783) Note: that the symbols ***, ** and * denote statistical significance at the 1%, 5% and 10% Source: Authors’ calculations based on World Development Indicators data from the World Bank (2019) 293 E-ISSN 2281-4612 Academic Journal of Interdisciplinary Studies Vol 12 No 2 ISSN 2281-3993 www.richtmann.org March 2023 6. Summary The literature presents a mixed verification of the effect of oil and gas extraction on economic development. While some studies found a positive relationship between natural resource abundance and economic growth (Dong et al., 2019), others found a negative correlation (Frikha & Tiba, 2019; Majumder et al., 2020). The study's goal was to look at the effect of oil and gas extraction on the GDP performance of Ghana. To do this, the study specifically focused on the effect of oil and gas extraction on the Agricultural, Services and Industry sectors of Ghana. In terms of the GDP model, the positive sign of the coefficient shows that Ghana's average GDP for the period following commercial oil and gas production is higher than the average GDP for the period before commercial oil and gas production. This demonstrates that revenue from oil and gas production has a favourable impact on Ghana's economic growth (GDP). The magnitude of the coefficient indicates that the average GDP for the time following commercial oil production is 4% higher than the average GDP before commercial oil production. The finding is incongruous with the works of Moradbeigi and Law (2017) and Cockx and Franken (2016). This result, however, supports the oil and gas blessing hypothesis that the discovery of oil and gas stimulates a nation's revenue thereby improving the level of infrastructural development and consequently, increasing GDP (Adabor, et al., 2021). Hence the study accepted the hypothesis (H5) that there is a significant difference in the GDP performance of Ghana before and after oil and gas production. The findings are also in line with the result of other empirical studies (Luki, 2016; Oteng-Adjei, 2011) that oil and gas exploration and extraction have an effect on the economies (GDP) of oil-producing and exporting countries. 7. Conclusion, Recommendation and Policy Implications Even though oil and gas extraction has had a positive impact on the economy (GDP) of Ghana, its impact on the agricultural and service sectors is negative as evidenced in Table 4. However, the study found evidence that the average industry value-added of Ghana for the period after the production of oil in commercial quantities is greater than the average industry value-added for the period before the production of oil in commercial quantities (H4). The study also alludes to the empirical findings that those countries with abundant natural resources, such as petroleum, have an economic edge over those with little or no such resources. Thus, discovering a natural asset such as crude oil could be a blessing or a curse, depending on how the cash generated from it is managed. Furthermore, the study's conclusions can be applied to other African oil-producing and exporting countries that have similar institutional setups. Governments, policymakers, and other oil and gas- producing and exporting countries, particularly in sub-Saharan Africa, will benefit from the study's conclusions. For policymakers, the findings of the study provide insights into Ghana's diversification and industrialization drive. i. When looking at the agriculture and service value-added models, the negative sign of the coefficient indicates that Ghana's average agriculture and service value-added for the period after commercial oil production is lower than the average agriculture and service value- added for the period before commercial oil production. The magnitude of the coefficient shows that the average agriculture and service value-added for the period after the production of oil in commercial quantities is 12% lower and 31% lower than the average agriculture and service value-added for the period before the production of oil in commercial quantities respectively. Government should therefore formulate and expand policies (notably, planting for food and jobs, and agricultural mechanization) to help expand the agricultural and service base of the economy using oil revenues. ii. Though previous studies have identified factual evidence of the oil curse, the study revealed a positive link between oil and gas extraction (oil revenues) and economic growth (GDP). Nonetheless, oil and gas extraction has had no major impact on the agricultural and service 294 E-ISSN 2281-4612 Academic Journal of Interdisciplinary Studies Vol 12 No 2 ISSN 2281-3993 www.richtmann.org March 2023 sectors, providing credence for the oil curse hypothesis (Dutch Disease). The study, therefore, recommends that the Public Interest and Accountability Committee (PIAC) should closely monitor and collaborate with the other petroleum revenue management institutions to ensure oil revenues are allocated to the government priority sectors, especially agriculture and service, are fully implemented. iii. Future research may look at specific government interventions or investments in the various sectors of the economy. This will give a better indication as to what specific investment decisions have to be made. References Ablo, A. D. (2015). Local content and participation in Ghana's oil and gas industry: can enterprise development make a difference? Extractive Industries and Societies, 2, 320–327. http://dx.doi. org/10.1016/j.exis.2 015.02.003. Ackah-Baidoo, A. (2012). Enclave development and “offshore corporate social responsibility” implications for oil- rich sub-Saharan Africa. Resources Policy, 37, 152–159. http://dx.doi.org/10.1016/j.resourpol.2011.12.010. Adabor, O, Buabeng, E. & Choumbou, R. F.D. (2021). Asymmetrical effects of oil and gas resource rent on economic growth: Empirical evidence from Ghana. Cogent Economics & Finance, 9(1), retrieved on November 16, 2022 from: DOI: 10.1080/23322039.2021.1971355 Adams, D., Ullaha, S., Akhtara, P., Adams, K., & Saidi, S. (2019). The role of country-level institutional factors in escaping the natural resource curse: Insights from Ghana. Resources Policy, 61, 433–440. Akinlo, H. (2012.). How important is oil in Nigeria's economic growth? Journal of Sustainable Development, 5(4), 165-179. Alexeev, M., & Chernyavskiy, A. (2015). Taxation of natural resources and economic growth in Russia’s regions. Economic Systems, 39 (2), 317–338. Amini, A. (2018). Studying the Effect of Abundance of Natural Resources on Economic Growth. European Journal of Sustainable Development, 7(1), 201. https://doi.org/10.14207/ejsd.2018.v7n1p201 Aryeetey, E. & Ackah, I. (2018). The boom, the bust, and the dynamics of oil resource management in Ghana, WIDER Working Paper, No. 2018/89, ISBN 978-92-9256-531-2, The United Nations University World Institute for Development Economics Research (UNU-WIDER), Helsinki. Asekunowo, V. O., & Olaiya, S. A. (2012). Crude oil revenue and economic development in Nigeria (1974 2008). OPEC Energy Revenue 36 (2), 138–169. https://doi.org/10.1111/j.1753-0237.2011.00205.x Ayelazuno, J. (2014). Oil wealth and the well-being of the subaltern classes in Sub-Saharan Africa: A critical analysis of the resource curse in Ghana. Resources Policy. 40, 66-73. https://doi.org/10.1016/j.r esourpol.2013.06.009 Bhattacharyya, S., & Hodler, R. (2014). Do natural resource revenues hinder financial development? The role of political institutions. World Development. 57, 101–113. Brunnschweiler, C. N. (2007). Cursing the Blessings? Natural Resource Abundance, Institutions, and Economic Growth. World Development, 36(3), 399–419. doi: 10.1016/j.worlddev.2007.03.004 Brunnschweiler, C. N., & Bulte, E. H. (2006). The Resource curse revisited and revised: A tale of paradoxes and red herrings, working paper 06/61, 2006 December at www.ideas.repec.org/p/eth/wpsuif/06- 61.html (last visited 11/07/2019) CIA (2018). at https://www.cia.gov/library/publications/resources/the-world-factbook/fields/264rank.html (last visited 22/05/2020) Cockx, L., & Francken, N, (2016). Natural resources: A curse on education spending? Energy Policy, 92, 394–408. https://doi.org/10.1016/j.enpol.2016.02.027 Collier, P., & Goderis, B. (2007). Commodity Prices, Growth and the Natural Resource Curse: Reconciling a Conundrum. CSAE WPS 15. Corrigan, C. C. (2014). Breaking the resource curse: transparency in the natural resource sector and the extractive industries transparency initiative. Resources Policy, 40, 17–30. http://dx.doi.org/10.1016/j.resourpol.2 013.10.003. Creswell, J. W. (2014). Research Design: Qualitative, Quantitative and Mixed Methods Approaches. (4th ed.). Thousand Oaks, CA: Sage. Dah, F. K., & Sulemana, K. M. (2010). The contribution of oil to the economic development of Ghana: The role of foreign direct investments (FDI) and government policies. Master thesis in International Business 15 ECT, Department of Economics and Informatics, University West Spring term 2010 295 E-ISSN 2281-4612 Academic Journal of Interdisciplinary Studies Vol 12 No 2 ISSN 2281-3993 www.richtmann.org March 2023 Ding, N., & Field, B. C. (2005). Natural resource abundance and economic growths. Land Economies, 81(4), 496–502. Dogan, E., Altinoz, B., & Tzeremes, P. (2020). The analysis of Financial Resource Curse’ hypothesis for developed countries: Evidence from asymmetric effects with quantile regression. Resources Policy, 68, 101773. Dong, B., Zhang, Y., & Song, H. (2019). Corruption as a natural resource curse: Evidence from the Chine. China Economic Review, 57, 101314. https://doi.org/10.1016/j.chieco.2019.101314 Ferreira, V., Almazán-Gómez, M.Á., Nechifor, V. & Ferreira, V (2022). The role of the agricultural sector in Ghanaian development: a multiregional SAM-based analysis. Economic Structures, 11(6), 1-27. https://doi.org/10.1186/s40008-022-00265-9 Frikha, M., & Tiba, S. (2019). The controversy of the resource curse and the environment in the SDGs background: The African context. Resources Policy, 62, 437-452. Garside, M. (2020). Volume of global oil reserves from 1990 to 2018 at https://www.statista.com/stati stics/236657/global-crude-oil-reserves-since-1990/ (last accessed May 22, 2020). Gerelmaa, L., & Kotani, K. (2016). Further investigation of natural resources and economic growth: do natural resources depress economic growth? Resources Policy, 50, 312–321 González, Y. S. (2016). The Gulf of Guinea: the future African Persian Gulf? Brazilian Journal of African Studies, 1(1), 85-105. Hasanov, F., Aliyev, R., Tashkin, D., & Suleymanov, E. (2022). Oil Rents and Non-Oil Economic Growth in CIS Oil Exporters. The Role of Financial Development. Available at SSRN: https://ssrn.com/abstract=416149 2 or http://dx.doi.org/10.2139/ssrn.4161492 Humphrey, M., Sachs, J. D., & Stiglitz, J. E. (2007). Escaping the Resource Curse. New York USA, Colombia University Press. Ingen, C., Wait, R., & Kleynhans, E. (2014). Fiscal policy and revenue management in resource-rich African countries: A comparative study of Norway and Nigeria, South African Journal of International Affairs, 21(3), 367-390. DOI: 10.1080/10220461.2014.968200 Ji, K., Magnus, J. R., & Wang, W. (2014). Natural resources, institutional quality, and economic growth in China. Environmental Resources. Economics, 57(3), 323–343. https://doi.org/10.1007/s10 640-013-9673-8 Katz, M, Bartsch, U., Malothra, H., & Cuc, M. (2004). Lifting the Oil Curse: Improving Petroleum Revenue Management in Sub-Saharan Africa. Washington DC, International Monetary Fund Kopiński, D., Polus, A. & Tycholiz, W. (2013). Resource curse or resource disease? Oil in Ghana. African Affairs, 112(449) 583–601. http://dx.doi.org/10.1093/afraf/adt056. Larsen, E. R. (2006). Escaping the resource curse and the Dutch disease? American Journal of Economics and Sociology, 65(3), 605–640. http://dx.doi.org/10.1111/j.1536-7150.2006.00476.x. Luki, B. N. (2016). Sustainable Management of Petroleum Resources: An Examination of the Ghana Heritage Fund (GHF) of the Petroleum Revenue Management Act, 2010 (Act 815). Journal of Resources Development and Management, 19, 64-73. Majumder, M. K., Raghavan, M., & Vespignani, J. (2020). Oil curse, economic growth and trade openness.Energy Economics, 91, 104896. https://doi.org/10.1016/j.eneco.2020.104896 Mavrotas, G., Murshed, S. M., & Torres, S. (2011). Natural resource dependence and economic performance in the 1970–2000 period. Revenue Development Economies, 15(1), 124–138. Mideksa, T. K. (2013). The economic impact of natural resources. Journal Environmental Economics Management, 65(2), 277–289. Moradbeigi, M., & Law, S. H. (2017). The role of financial development in the oil-growth nexus. Resources Policy, 53(16), 164 -172. https://ezproxy.ug.edu.gh:2057/10.1016/j.resourpol.2017.06.016 Moshiri, S. & Hayati, S. (2017). Natural resources, institutions quality, and economic growth; A cross-country analysis. Iran Economy Review, 21(3), 661–693. Oteng-Adjei, J. (Minister of Energy, Ghana) 'meet-the-press series' Accra, Thursday, 06/10/2011 Oshionebo, E. (2018). Sovereign Wealth Funds in Developing Countries: A Case Study of the Ghana Petroleum Funds. Journal of Energy & Natural Resources Law, 36(1), 33–59. Peach, J. and CM Starbuck (2011). Oil and Gas Production and Economic Growth in New Mexico. Journal of Economic Issues, 45(2), 511-526. DOI: 10.2753/ JEI0021-3624450228 PIAC (2020). Public Interest and Accountability Committee Reports 2011-2020 at https://www.piacghana.org/port al/5/25/piac-reports (last accessed December 3, 2022). Poncian, J. (2019). Extractive resource ownership and the subnational resource curse: Insights from Tanzania. Extractive Industries and Society 6(2) 332-342. https://doi.org/10.1016/j.exis.2018.08.013 296 E-ISSN 2281-4612 Academic Journal of Interdisciplinary Studies Vol 12 No 2 ISSN 2281-3993 www.richtmann.org March 2023 Ross, M. L. (2012). The oil curse: How Petroleum wealth shapes the Development of Nations. New Jersey USA, Princeton University Press. Sachs, D. J., & Warner, A. M. (1995). Natural resource abundance and economic growth. National Bureau of Economic Research (NBER) working paper no. 5398. Sachs, J. A. (1999). The big push, natural resource booms and growth. Journal of Development Economics,59, 43– 76. Sarmidi, T. L., Law, S. H., & Jafari, Y (2014). Resource curse: new evidence on the role of institutions. International Journal of Economics, 28(28), 191-206. Satti, L. S., Farooq, A., & Loganathan, N., (2014). Empirical evidence on the resource curse hypothesis in oil abundant economy. Economic Modelling, 42, 421-429. Shaw, D. L. (2013). Good Governance in the Post – Soviet South: Testing Theories of the Resource Curse in Azerbaijan. Journal of Politics & International Studies, Vol. 9, 520-561. Sigman, R., Harris, A. S., Meyer-Sahling, J-H., Mikkelsen, K. S. & Schuster, C. (2022): Do Bureaucrats Contribute to the Resource Curse? Evidence from a Survey Experiment in New Oil States. The Journal of Development Studies. 1-17. DOI: 10.1080/00220388.2021.2013468 Sinha, A., & Sengupta, T. (2019). Impact of natural resource rents on human development: What is the role of globalization in Asia Pacific countries? Resources Policy, 63(C), 1-1. Smith, B. (2004). Oil wealth and regime survival in the developing world, 1960–1999. American Journal of Political Science, 48(2), 232-246. Steven, P, & Dietsche, E. (2008). The Resource Curse. An analysis of causes, experience and possible ways forward. Energy Policy, 36(1), 56-65. Tiba, S. (2019). Modelling the nexus between resources abundance and economic growth: an overview from the PSTR model. Resources Policy, 64, 101503. https://doi.org/10.1016/j.resourpo l.2019.101503 Van der Ploeg, F. (2011). Natural resources: curse or blessing? Journal of Economic Literature. 49, 366–420. http://dx.doi.org/10.1257/jel.49.2.366. Yeboua, K. Cilliers, J. & Roux, A. (2022). Nigeria in 2050: major player in the global economy or poverty capital?” Institute for Security Studies, March 15, 2022, 12-15, https://tinyurl.com/242hh4ya. Zalle, O. (2019). Natural resources and economic growth in Africa: the role of institutional quality and human capital. Resources Policy, 62, 616–624. https://doi.org/10.1016/j.resourpol.2018.11.009 Zuo, N., & Schiefer, J. (2014). Are Resources a Curse? An Investigation of Chinese Provinces. Selected Paper prepared for presentation at the Southern Agricultural Economics Association (SAEA) Annual Meeting, Dallas, Texas, 1-4 February 2014. Appendix 1: Secondary Data (World Bank Development Indicators, 2019) Year OIL GDP Exchange rate Inflation Doilrev Service Industry Agric 2001 23.12 54848740700 0.716305158 32.9054089 0 29.15890698 25.21597974 35.24180012 2002 24.36 57316933800 0.792417084 14.81624006 0 29.21411322 25.27567435 35.14837706 2003 28.1 60297414400 0.866764327 26.67494973 0 29.06676018 25.2099151 36.5454059 2004 36.05 63674069600 0.899494854 12.62457406 0 28.67829438 24.71675627 37.95242609 2005 50.59 67430842200 0.905209486 15.11818572 0 28.91462953 25.12543594 37.45301293 2006 61 71746357100 0.915106799 10.91516997 0 46.45903144 19.80338501 28.94894567 2007 69.04 74865041500 0.932619195 10.73272807 0 47.1685501 19.49280657 27.29411448 2008 94.1 81715042300 1.052275 16.52214331 0 46.17389681 19.39956455 29.40809676 2009 60.86 85673716900 1.404966667 19.25071443 0 47.93620711 18.51409213 30.99338421 2010 77.38 92441693700 1.429983333 10.70756812 0 48.18116083 18.01494281 28.03873796 2011 107.46 1.05427E+11 1.520625 8.726836831 1 45.8448786 23.86366009 23.66370497 2012 109.45 1.15224E+11 1.824866667 7.12635006 1 47.58295025 27.13653359 22.13115469 2013 105.87 1.2365E+11 1.98135 11.66619231 1 40.58855099 34.08982225 20.36984251 2014 96.29 1.27233E+11 2.896575 15.48961603 1 39.09721473 35.19175722 20.74452957 2015 49.49 1.30004E+11 3.714641667 17.1499695 1 42.14244854 32.13687984 20.78519837 2016 40.68 1.34486E+11 3.909816667 17.45463471 1 45.25817355 28.34427253 21.48137806 2017 52.51 1.45438E+11 4.350533333 12.37192155 1 44.63646636 30.37133117 20.08068434 2018 68.09 1.54548E+11 4.585325 7.808765166 1 45.00482887 31.39229929 18.68726875 2019 63.05 1.6456E+11 5.217366667 7.175922847 1 45.88186902 31.5979152 17.60748628 2020 50.23 1.65993E+11 5.595708333 9.95344087 42.63263497 34.6854824 18.23772918 year Agric EXR GDP Industry Service Oil price Inflation Doilrev 2001Q1 35.24180012 54848740700 25.21598 29.15891 25.72 41.95 0 2001Q2 35.21844436 0.735333 55465788975 25.2309 29.17271 26.72 36.84 0 2001Q3 35.19508859 0.754361 56082837250 25.24583 29.18651 26.06 28.31 0 2001Q4 35.17173283 0.773389 56699885525 25.26075 29.20031 19.33 21.29 0 2002Q1 35.14837706 0.792417 57316933800 25.27567 29.21411 24.03 16.00 0 2002Q2 35.49763427 0.811004 58062053950 25.25923 29.17727 24.51 13.70 0 2002Q3 35.84689148 0.829591 58807174100 25.24279 29.14044 28.32 12.90 0 297 E-ISSN 2281-4612 Academic Journal of Interdisciplinary Studies Vol 12 No 2 ISSN 2281-3993 www.richtmann.org March 2023 year Agric EXR GDP Industry Service Oil price Inflation Doilrev 2002Q4 36.19614869 0.848178 59552294250 25.22635 29.1036 27.45 15.20 0 2003Q1 36.5454059 0.866764 60297414400 25.20992 29.06676 29.45 29.84 0 2003Q2 36.89716095 0.874947 61141578200 25.08663 28.96964 27.16 32.94 0 2003Q3 37.248916 0.88313 61985742000 24.96334 28.87253 26.78 29.79 0 2003Q4 37.60067105 0.891312 62829905800 24.84005 28.77541 29.58 31.27 0 2004Q1 37.95242609 0.899495 63674069600 24.71676 28.67829 32.73 15.62 0 2004Q2 37.8275728 0.900924 64613262750 24.81893 28.73738 35.52 18.00 0 2004Q3 37.70271951 0.902352 65552455900 24.9211 28.79646 42.82 19.64 0 2004Q4 37.57786622 0.903781 66491649050 25.02327 28.85555 40.64 16.44 0 2005Q1 37.45301293 0.905209 67430842200 25.12544 28.91463 53.28 17.79 0 2005Q2 35.32699611 0.907684 68509720925 23.79492 33.30073 57.61 14.05 0 2005Q3 33.2009793 0.910158 69588599650 22.46441 37.68683 63.72 14.34 0 2005Q4 31.07496248 0.912632 70667478375 21.1339 42.07293 55.60 13.91 0 2006Q1 28.94894567 0.915107 71746357100 19.80339 46.45903 63.06 11.28 0 2006Q2 28.53523787 0.919485 72526028200 19.72574 46.63641 69.74 11.39 0 2006Q3 28.12153008 0.923863 73305699300 19.6481 46.81379 63.49 11.67 0 2006Q4 27.70782228 0.928241 74085370400 19.57045 46.99117 62.54 10.92 0 2007Q1 27.29411448 0.932619 74865041500 19.49281 47.16855 62.36 10.19 0 2007Q2 27.82261005 0.962533 76577541700 19.4695 46.91989 70.55 10.69 0 2007Q3 28.35110562 0.992447 78290041900 19.44619 46.67122 77.00 10.19 0 2007Q4 28.87960119 1.022361 80002542100 19.42288 46.42256 91.51 12.75 0 2008Q1 29.40809676 1.052275 81715042300 19.39956 46.1739 102.98 13.79 0 2008Q2 29.80441862 1.140448 82704710950 19.1782 46.61447 133.47 18.41 0 2008Q3 30.20074048 1.228621 83694379600 18.95683 47.05505 100.75 17.89 0 2008Q4 30.59706235 1.316794 84684048250 18.73546 47.49563 43.29 18.13 0 2009Q1 30.99338421 1.404967 85673716900 18.51409 47.93621 47.32 20.53 0 2009Q2 30.25472265 1.411221 87365711100 18.3893 47.99745 69.34 20.74 0 2009Q3 29.51606109 1.417475 89057705300 18.26452 48.05868 68.19 18.37 0 2009Q4 28.77739953 1.423729 90749699500 18.13973 48.11992 75.24 15.98 0 2010Q1 28.03873796 1.429983 92441693700 18.01494 48.18116 79.90 13.30 0 2010Q2 26.94497971 1.452644 95688043450 19.47712 47.59709 75.66 9.50 0 2010Q3 25.85122147 1.475304 98934393200 20.9393 47.01302 78.21 9.40 0 2010Q4 24.75746322 1.497965 1.02181E+11 22.40148 46.42895 92.34 8.60 0 2011Q1 23.66370497 1.520625 1.05427E+11 23.86366 45.84488 114.62 9.13 1 2011Q2 23.2805674 1.596685 1.07876E+11 24.68188 46.2794 113.91 8.59 1 2011Q3 22.89742983 1.672746 1.10326E+11 25.5001 46.71391 109.96 8.40 1 2011Q4 22.51429226 1.748806 1.12775E+11 26.31832 47.14843 107.72 8.58 1 2012Q1 22.13115469 1.824867 1.15224E+11 27.13653 47.58295 124.62 8.78 1 2012Q2 21.69082665 1.863988 1.17331E+11 28.87486 45.83435 95.89 9.44 1 2012Q3 21.2504986 1.903108 1.19437E+11 30.61318 44.08575 113.04 9.43 1 2012Q4 20.81017056 1.942229 1.21544E+11 32.3515 42.33715 109.19 8.84 1 2013Q1 20.36984251 1.98135 1.2365E+11 34.08982 40.58855 109.53 10.78 1 2013Q2 20.46351428 2.210156 1.24546E+11 34.36531 40.21572 103.30 11.63 1 2013Q3 20.55718604 2.438963 1.25441E+11 34.64079 39.84288 111.21 11.95 1 2013Q4 20.65085781 2.667769 1.26337E+11 34.91627 39.47005 110.60 13.50 1 2014Q1 20.74452957 2.896575 1.27233E+11 35.19176 39.09721 107.68 14.52 1 2014Q2 20.75469677 3.101092 1.27926E+11 34.42804 39.85852 111.97 14.99 1 2014Q3 20.76486397 3.305608 1.28618E+11 33.66432 40.61983 98.56 16.47 1 2014Q4 20.77503117 3.510125 1.29311E+11 32.9006 41.38114 62.36 16.99 1 2015Q1 20.78519837 3.714642 1.30004E+11 32.13688 42.14245 57.01 16.64 1 2015Q2 20.95924329 3.763435 1.31125E+11 31.18873 42.92138 63.75 17.08 1 2015Q3 21.13328822 3.812229 1.32245E+11 30.24058 43.70031 48.57 17.36 1 2015Q4 21.30733314 3.861023 1.33366E+11 29.29242 44.47924 38.92 17.70 1 2016Q1 21.48137806 3.909817 1.34486E+11 28.34427 45.25817 39.80 19.20 1 2016Q2 21.13120463 4.019996 1.37224E+11 28.85104 45.10275 49.89 18.40 1 2016Q3 20.7810312 4.130175 1.39962E+11 29.3578 44.94732 47.23 17.20 1 2016Q4 20.43085777 4.240354 1.427E+11 29.86457 44.79189 54.93 15.40 1 2017Q1 20.08068434 4.350533 1.45438E+11 30.37133 44.63647 52.53 12.80 1 2017Q2 19.73233044 4.409231 1.47716E+11 30.62657 44.72856 47.54 12.10 1 2017Q3 19.38397654 4.467929 1.49993E+11 30.88182 44.82065 55.23 12.20 1 2017Q4 19.03562265 4.526627 1.5227E+11 31.13706 44.91274 64.27 11.80 1 2018Q1 18.68726875 4.585325 1.54548E+11 31.3923 45.00483 66.68 10.36 1 2018Q2 18.41732313 4.743335 1.57051E+11 31.4437 45.22409 75.94 9.97 1 2018Q3 18.14737752 4.901346 1.59554E+11 31.49511 45.44335 79.09 9.79 1 2018Q4 17.8774319 5.059356 1.62057E+11 31.54651 45.66261 57.67 9.43 1 2019Q1 17.60748628 5.217367 1.6456E+11 31.59792 45.88187 67.05 9.28 1 2019Q2 17.76504701 5.311952 1.64918E+11 32.36981 45.06956 63.05 9.12 1 2019Q3 17.92260773 5.406538 1.65276E+11 33.1417 44.25725 63.05 9.12 1 2019Q4 18.08016846 5.501123 1.65634E+11 33.91359 43.44494 63.05 9.12 1 2020Q1 18.23772918 5.595708 1.65993E+11 34.68548 42.63263 33.73 7.80 1 298