Do Regulatory Tariffs Curb Gas Flaring? Evidence from Nigeria

Abstract

This study examines the impact of flare tariff adjustments on gas-flaring volumes in Nigeria. Utilising a 52-year dataset, this analysis demonstrates that the effectiveness of flare tariffs in reducing gas flaring depends on the stringency of imposed charges. To isolate this effect, this study distinguishes between tariff regimes implemented before and after 2018, a pivotal year marked by the introduction of substantially higher tariffs under revised regulations. The findings indicate that the pre-2018 tariffs had no statistically significant effect on gas-flaring volumes, whereas the post-2018 tariffs led to a statistically significant reduction. Specifically, the pre-2018 tariffs were associated with a negligible reduction in flaring (0.05 percentage points), which was statistically insignificant. By contrast, the post-2018 tariff regime resulted in a 9.26 percentage-point decline in flaring volumes, significant at the 1% level. Additional factors contributing to the flaring reduction include oil production levels, oil prices, and the availability of gas infrastructure. These results highlight the critical role of sufficiently stringent tariff policies in achieving substantial reductions in global gas flaring.

1. Introduction

Despite being widely condemned for its severe environmental consequences, the flaring of associated natural gas remains a prevalent industrial practice. This inefficient and wasteful process contradicts global sustainability efforts, including the United Nations Sustainable Development Goals, which promote sustainable production and consumption to safeguard the needs of future generations [1,2]. The gas industry argues that flaring is necessary for various operational reasons, including stabilising pressure and flow from oil wells during testing, managing waste gas that cannot be captured or processed, and addressing safety or emergency concerns [3]. However, substantial scientific evidence demonstrates that gas flaring contributes to significant economic losses [4,5,6], exacerbates environmental degradation [7,8], damages human health and safety [9], and negatively impacts local communities [10,11,12].

Several countries have engaged in gas flaring for decades as it is a long-standing practice in their oil and gas industries. According to Ref. [13], 144 billion cubic metres (bcm) of gas were flared in 2021, resulting in the release of 400 million tons of carbon dioxide equivalent (CO₂-e) into the atmosphere. Notably, 75% of flaring volumes come from the top 10 flaring nations, which also account for 50% of the global oil production. These countries are Russia, Iraq, Iran, the United States, Venezuela, Algeria, Nigeria, Mexico, Libya, and China. The top seven flaring nations retained their positions for a decade, while Mexico, Libya, and China have seen increases in their flaring activities in recent years [13]. Figure 1 shows the top ten flaring nations by volume and intensity as of 2021.

The World Bank has set a goal of achieving zero routine gas flaring by 2030, yet progress towards this target has been slow [14]. The latest data show that the volume of gas flared decreased from 144 billion cubic metres (bcm) in 2021 to 139 bcm in 2022, with flaring intensity also falling from 5.2 cubic metres per barrel of oil produced in 2021 to 4.7 m³/bbl [15]. This represents a 3% reduction in global flaring volumes compared with 2021. However, this progress was short-lived, as flaring volumes rebounded in 2023, surging by 7% to reach 148 bcm—levels last observed in 2019 [14]. With only six years left in the declaration for zero routine flaring by 2030, this is a serious concern [14].

Among the top flaring nations, Nigeria has made significant contributions to global flaring reduction in recent years. By 2023, Nigeria reduced its flaring volume by 1.3 bcm, representing a 20% reduction from the 2021 level. This was linked to the slight increase in the nation’s production efficiency as represented by a reduction in the flaring intensity from 11.8 m³/bbl in 2021 to 11.1 m³/bbl in 2022 [15]. However, the largest contributor to this reduction in flaring volume was the fall by 14% in oil production during 2021–2022 [15]. The reduction resulted in Nigeria moving to the ninth position in the ranking of the top ten global flare nations from its seventh position in 2021, as the volume flared was below the volumes flared by Mexico and Libya. This study aims to investigate the determinants of gas-flaring reduction in Nigeria over the last five decades, focusing on gas flare tariff regimes. A review of the existing literature reveals that most studies examining the effects of gas-flaring policies are qualitative in nature, with quantitative research typically focusing on environmental impacts. For example, in a comparative review of policies regarding global flaring and venting regulations, the World Bank reported countries that have adopted monetary fines, flaring penalties, and sanctions as disincentives. Of twenty-one (21) countries studied, only twelve were reported to have these provisions in their regulations. The countries were Angola, Brazil, Canada (some provinces), Colombia, Gabon, Kazakhstan, Mexico, Nigeria, the Russian Federation, the United Kingdom, and some states in the US [16]. A direct literature search on the impact of the flare penalty on the volume of gas directed at the countries mentioned above returned no results, except for the work by Lade and Rudik, which explored the impact of price-based policies on gas flares using the flare gas tax [17], and Rabe et al. [18], who explored whether states have adopted market-based approaches to pricing methane flaring through severance taxes. While Lade and Rudik used quantitative methods, Rabe’s approach was qualitative. Therefore, the aim of this study is to quantitatively analyse the impact of changes in flare policy regimes on the volume of gas flared. To the best of our knowledge, this study is the first to quantitatively assess the effectiveness of a gas flare tariff in reducing the flare volume. By examining the effectiveness of market-based instruments within the context of Nigeria’s regulatory framework and energy market, the findings of our study offer valuable insights into enhancing the country’s flaring reduction efforts and supporting its transition towards more sustainable gas utilisation practices.

By employing multivariate linear regression models, we study the effect of oil production, oil price, gas price, the availability of gas infrastructure, and flare tariffs on the volume of gas flared. Nigeria enacted a new regulation (Flare Gas (Prevention of Waste and Pollution) Regulations) in 2018, which increased the flare tariff by 38 times relative to the tariff in operation before that time. The results show that the new tariff after 2018 is much more effective in reducing the volume of gas flared.

Research on the determinants of gas flaring has identified a wide range of factors that contribute to the persistence of this practice. Some studies suggest that the insufficient pricing of gas, either domestically or close to the flare point, is a key inhibitor [19,20,21], whereas others point to insufficient punitive governance, policy, and regulation as the cause of the persistence of gas flaring [12,22,23,24]. Even when regulations exist, effective enforcement by responsible bodies is lacking [25]. Other reasons, such as the lack of sufficient gas infrastructure [26], lack of mature domestic gas markets, the sparse nature of gas-flaring points, and insufficient monitoring by regulatory bodies, have also been noted [27,28,29]. During the early years of Nigeria’s oil industry development in the 1960s and the 1970s, gas was not a popular energy source; thus, little attention was given to building a gas infrastructure [30]. Uncoordinated capacities were built by international oil companies in their early years, mostly to support their operations. The first major gas pipeline was commissioned in 1978 by an NNPC–Shell joint venture to deliver gas to the then Nigerian Electric Power Authority (NEPA) turbine station at Sapele, followed by two other power stations with a combined capacity of 0.27 billion cubic feet per day [31]. With the creation of the Nigerian Gas Company (NGC) in 1988, additional capacities were steadily added to the infrastructure, to the current 2.9 bscf/d as of 2019, which was relatively small for the economic development of Nigeria [26].

Various global attempts have been made to identify strategies for reducing and ultimately eliminating gas flaring, such as the completion of the largest flare gas-to-power project in the Middle East commissioned by Aggreko in Southeast Kurdistan [32], Hoerbiger’s eleven flare gas projects in Ecuador [33], and Nigeria’s gas flare commercialisation programme [34]. These policies, governing structures, and regulations have been applied in different contexts and locations, and most recommendations focus on legalising the prohibition of gas flaring and promoting market-based initiatives for flaring reduction. Nigeria, for instance, has gone through various efforts to reduce gas flaring since 1969, when the Petroleum (Drilling and Production) Regulation mandated that operators submit gas utilisation proposals for new fields coming on stream. This was closely followed by the Petroleum (Amendment) Act of 1973, which empowers the government to take gas at a flare site without payment to the operator [35]. By 1979, the Associated Gas Reinjection Act had declared flaring illegal starting January 1984 [36]. The same act introduced the first penalty regime for flaring gas with an effective date of 1985 [5]. Since then, the flare penalty has been reviewed three times in 1992, 1998, and, most recently, in 2018, by the signing of the Flare Gas (Prevention of Waste and Pollution) Regulations, 2018. Other notable policies targeted at gas flaring include the 1989 NLNG Act, the Associated Gas Framework Agreement of 1992, Finance Decree 18 of 1998, and Decree 30 of 1999, which extends all incentives relating to associated gas to non-associated gas, the National Domestic Gas Supply and Pricing Policy/Regulations that mandated the allocation of gas reserves for domestic use as well as providing a framework for establishing minimum gas prices.

Following Nigeria’s ratification of the Paris Agreement in 2016, in which the country committed to reducing and ultimately eliminating gas flaring as part of its efforts to curb carbon and methane emissions, the Flare Gas (Prevention of Waste and Pollution) Regulations, 2018 were enacted. This regulation established a nationwide framework for gas flare elimination by facilitating the auctioning of flare sites for project developers interested in gas monetisation. Additionally, the regulation increased flare tariffs to an average of USD 2.50 per thousand cubic feet [37]. Furthermore, the passage of the Petroleum Industry Act (PIA) 2021 by the National Assembly, followed by its ratification by the President and the Minister of Petroleum Resources, introduced a revised tariff for unauthorised flaring, venting, or wastage of natural gas, set at USD 3.50 per thousand standard cubic feet (Mscf) [38]. Figure 2 illustrates the regulatory measures aimed at addressing gas flaring in Nigeria.

2. Literature Review

2.1. Theoretical Evidence

A survey of the existing literature suggests that a change in policy is vital for combatting gas flaring both globally and locally. At the global level, Ref. [39] discussed measures to put in place to ensure the elimination of gas flaring. The enablers to such measures include the development of specifically appropriate legal and regulatory frameworks by governments, the reformation of gas markets, the elimination of subsidies for competing alternative fuels, and the involvement of the private sector in the development of gas infrastructure. They also recommend amendments to the royalty and tax systems that discourage gas utilisation by operators. Writing on the effects that flaring had on the Niger Delta from 1969 to 2001, Ref. [30] opined that the regulations and incentives put in place by the Nigerian government to abate gas flaring were not enough to discourage the practice of perpetrating oil industry operators. As a result, the study recommended an upward review of the tax/penalties on gas flared and the amendments of property rights that would foster sustainable energy utilisation as well as community participation. Ref. [40], on the other hand, compares Nigeria’s legal and institutional frameworks on gas flaring to those of Canada, the UK, and Saudi Arabia. They recommend amendments to the legal framework governing the sector responsible for flaring as well as the institutions for law enforcement. This opinion was echoed by the author of [41], who showed that flaring in the Niger Delta region of Nigeria persisted due to the failure of government and government institutions responsible for regulating the industry to raise their expectation as well as multinational companies operating in the region to operate responsibly. He argued that these issues are linked to corruption and the inept attitudes of MNCs. Addressing problems and prospects of gas flaring in Nigeria, Ref. [27] reviewed the applicable laws governing the oil and gas industry, especially those on gas flaring and their flaws, to ensure the elimination of flares. Their recommendation also borders on the enactment of more stringent laws that outlaw flaring altogether rather than paying fines.

Other flare policy studies by the authors of [42,43] argued that the passage of the PIB and strict improvement of its monitoring for implementation can be game changers in the sustainable development of gas in Nigeria and the elimination of gas flaring. Ref. [24] reveals hindrances that prevent the success of such laws and policies and recommends measures to overcome them. Ref. [23] reviewed the literature on the effect of gas flaring on ecosystems and suggested that stricter measures be adopted by the government to end the wasteful process.

Discussing Nigeria’s recent gas flare commercialisation programme (NGFCP) and other policies in the same area, Ref. [44] recommended that, with proper implementation and relaxation of some strict conditions imposed by the government to participate in the flaring programme, the NGFCP could change the narrative of addressing gas flaring in the Nigerian Delta region, while Ref. [29] argues that imposing a flaring ban as a law or regulation does not work. He believed that flaring occurred when the following two conditions were met: First, the country has saturated oil reservoirs with rich solution gas and gas caps, making reinjection an enhanced oil recovery method for maintaining reservoir pressure as non-viable. The second condition is when the domestic market for natural gas is not developed or when the pricing of the product is not profitable enough to warrant infrastructure investment. The lack of gas transport and processing infrastructure can also promote flaring, as the gas, even after capture, cannot be further processed and transported to the market. Ref. [45] compares the gas-flaring legal framework of Nigeria to that of Russia, the US, and Norway. The analysis identified the weak enforcement of existing laws as enablers of gas flaring in Nigeria. A recommendation for the stringent enforcement of the PIA 2021 was made, and the adoption of other laws as identified in the higher-income countries was compared with Nigeria in the study.

From the cited theories published within the last decade, a narrative suggesting more stringent laws, effective monitoring, and the supervision of such laws, creating an enabling environment for the promotion of gas project development, and enabling the creation of sustainable gas markets (especially local markets) led to the recommended actions. As clear as these recommendations can be, though, empirical studies supporting them are sparse. All these can be lumped into policy. Although it may be challenging to represent some of the policies in an empirical study, others can be appropriately represented by indicators. The following section reviews the most recent empirical studies on the issue of gas flaring, covering some of these policies.

2.2. Empirical Evidence

A large number of empirical studies have focused on the effect of gas flaring on the environment or economy. See, for example, [2,46,47,48]. Hassan [46], for instance, studied the vulnerability of Nigeria’s GDP to environmental pollution caused by gas flaring using the autoregressive distributed lag (ARDL) model and Granger causality to run the regression. The study found that gas utilisation policies and transparency that were introduced in the oil sector reduced the level of environmental pollution through flaring and increased gas utilisation projects.

Following Hassan’s work, Okoro et al. [47] developed an econometric model to investigate the determinants of gas flaring with a focus on some identified contributing variables such as gas price, crude oil production, utilisation, and GDP growth. Flaring was found to be persistent, as there was an increase in flaring activities by about 0.37% to 0.38% compared to flaring in the recent past. His research shows that gas flaring in Nigeria is largely determined by the consumption and pricing of gas, as well as the past activities of oil and gas companies that sustain the practice. Introducing policies to address the associated gas flaring and increasing private sector participation in both the upstream and downstream sectors are recommended. In a similar study of the effect of gas flaring on the GDP, Diugwu et al. [48] discovered that gas flaring has a significant negative effect on the GDP; that is, as flaring increases, GDP falls because of lost revenue that could have been accrued to the government by utilisation. On the other hand, the amount of gas utilised in the country was found to have an insignificant effect on GDP. Hassan et al. [46] used a time series model for the period of 1965–2009 to measure the effect of oil and gas production, investment in gas utilisation, and the export price of gas on gas flared. The study found that the size and environmental philosophy of the industry have a strong positive impact on gas flaring-related carbon dioxide emissions.

A recent study using the ARDL error correction model by Okoye et al. [2] tested the relationship between oil rent, fossil fuel production, and gas flaring on the economy of Nigeria. The result of the estimation found a significant long-run positive contribution of oil rent and fossil fuel production to the economy, while gas flaring was found to depress economic performance. Another study by the author of [22] along this line compared the flaring activities of seven (7) major oil exploration companies in Nigeria and how they affect the economy. The results clearly indicate a negative impact on the economy. The writer recommended imposing stricter fines on flaring as a mitigating factor and designing a programme that will ensure a gradual reduction in flaring over time.

3. Materials and Methods

3.1. Data Collection and Preparation

Historical annual time series data of all variables contained in the models covering 1970–2021 were collected from different sources. Data on gas-flared volumes (billion standard cubic feet (BSCF)) were collected from the Nigerian National Petroleum Corporation (NNPC) Annual Statistical Bulletin (ASB) [49]. The gas-flaring volume is the dependent variable and is labelled as the total gas flared (TGF). Data on Nigerian oil production (thousand barrels per day) [50] and historical oil prices (USD/bbl) were obtained from the database of the Organization of Petroleum Exporting Countries (OPEC) [51] and are denoted as the total oil produced (TOP) and oil price (OPR), respectively. Nigeria, a member of the OPEC, is obligated to submit monthly production and supply statements (PSSs) as well as complete the OPEC questionnaire that gathers data on member countries’ economic indicators on an annual basis. To capture the impact of insufficient gas infrastructure, historical gas pipeline capacities (Mscf/km) were extracted from [31,52], whereas recent updates as of 2018 were sourced from [26]. Historical gas flare tariff data were obtained from the Department of Petroleum Resources (now Nigerian Upstream Petroleum Regulatory Commission (NUPRC), the regulator responsible for administering the tariffs and other regulatory functions in the Nigerian oil and gas industry [37]. The gas flare tariff is the main variable of interest and is denoted as the adjusted flare tariff (AFT), which accounts for the conversion of the tariff value from Nigerian Naira to US Dollars using the exchange rate obtained from the World Bank [53]. Finally, gas prices (GPR) were obtained from the US Energy Information Administration (EIA) website in the form of LNG import prices [54]. All monetary values (tariffs, oil price, and gas price) were adjusted for inflation using the Nigerian Consumer Price Index for the 2010 base year.

3.2. Data Visualisation

Figure 3 provides a snapshot of the total volume of flared gas and crude oil produced during the period under consideration. A sharp and steady increase in the volume flared between 1970 and 1974 corresponds to a surge in oil production following the end of the civil war that raged in the country during the period 1967–1970 [55]. In contrast, the sharp decline in the volume of gas flared witnessed in the 1980s marked the global oil glut and the military coup that overthrew Nigeria’s democratically elected government [56]. This political and economic instability has led to a significant drop in crude production and associated gas flaring. In the early 2000s, a steady decline in gas flaring was witnessed, marking the commissioning of the Nigerian Liquefied Natural Gas (NLNG) company, which has a combined nameplate capacity of 8.85 million tons per year of liquefied natural gas and natural gas liquids supplied to the global natural gas market [57]. Other monetisation projects targeting the power sector, particularly the construction of seven integrated power projects, have significantly improved gas utilisation in Nigeria [58].

Figure 4 depicts the historical evolution of crude oil prices using the OPEC reference basket, which weighs the average prices of oil. The almost steady rise in prices from 1970 to 1980 corresponded to the Yom Kippur War as well as the Iranian revolution. The low prices witnessed from 1990 to around 2000 coincided with the introduction of netback pricing as well as the invasion of Kuwait by Iraq. Other peaks around 2008 were triggered by the invasion of Iraq by the US, while the highest peak around 2012 was attributed to the Arab Spring [59]. Similar justifications for price fluctuations issued by the U.S. Energy Information Administration (EIA) were changes in global economic growth expectations, followed by concerns over supply disruptions from producing nations such as Syria, Yemen, and Sudan, with a potential cut of approximately one million barrels per day from the global oil market. The last factor was the sanction on Iranian oil imports by the EU and the US, aimed at pressuring the Persian nation to abandon its nuclear programme [60]. Global oil price is employed in all models, as 60% of the oil produced in Nigeria is exported, while only 40% is allocated for domestic refining and utilisation (combined refining capacity of 445,000 barrels/day).

Figure 5 presents the historical trend of flare tariff rates starting in the 1970s, when no tariffs existed. The first tariff regime was introduced in 1984 at a rate of N0.02 per thousand standard cubic feet (USD 0.02/mscf) and remained in place until 1992. This was followed by an upward review of N0.5/Mscf (USD 0.003/mscf) during 1992–1998. By 1999, a new regime had kicked in with an average tariff rate of N10/Mscf (USD 0.142/mscf) until 2018 [37]. Before the recent review, the 1999 rate dipped to USD 0.028 because of currency exchange depreciation. The current tariff that became effective in 2019 saw a rise in the rate to an average of USD 1.066/mscf following the passing of the Flare Gas (Prevention of Waste and Pollution) Regulations, 2018 [61].

3.3. Model Specification

To analyse the effect of gas-flaring tariffs on the volume of gas flared, the flare volume was expressed as a function of key determinants of flaring identified by Refs. [2,47]. These determinants include oil production, oil and gas prices, and flare tariffs. Gas pipeline capacity was also included in the model to measure the impact of gas infrastructure provision on flaring. The first linear econometric model is specified as shown in Equation (1) below. TOP is the total oil production (million barrels), OPR denotes the oil price (USD/barrel), and GPR refers to the price of LNG imported to the United States (US), used as a proxy for gas prices in Nigeria because of the lack of domestic price data. GPC is the gas pipeline capacity in thousand standard cubic feet per kilometre (Mscf/km), whereas AFT represents the adjusted gas-flaring tariff (USD/Mscf). Subscript t indicates the period from 1970 to 2021 in this case.

$${TGF}_t={{\phi }_0+{\phi }_{1}TOP}_t+{\phi }_2{OPR}_t+{\phi }_3{GPR}_t+{\phi }_4{GPC}_t+{\phi }_5{AFT}_t+{\epsilon }_t$$

(1)

The ${\phi }_i(i=0,1\dots )$ represent the magnitude and direction of the estimated coefficients including the intercept. The error term is denoted as ε_t. The a priori expectation is that the volume of gas flared is positively associated with the volume of oil production. Similarly, oil prices are expected to have a positive impact on the volume of gas flared, as higher oil prices incentivise producers to increase oil production, thereby leading to higher associated gas production and flaring. As favourable gas prices may motivate producers to invest in gas utilisation technologies, higher gas prices are expected to reduce the volume of gas flared. Similarly, the availability of gas infrastructure is expected to lower flaring, as gas can be easily channelled to the market. As flare tariffs serve as a disincentive for gas flaring, it is expected that the higher the flare tariff, the lower the volume of gas flared. To check the robustness of the model, we estimate the semi-log form of the equation. The re-specified model takes the following form:

$${logTGF}_t={{\phi }_0+{\phi }_{1}TOP}_t+{\phi }_2{OPR}_t+{\phi }_3{GPR}_t+{\phi }_4{GPC}_t{+\phi }_5{AFT}_t+{\epsilon }_t$$

(2)

The third model we estimate is the log-log form as follows:

$${logTGF}_t={{\phi }_0+{\phi }_{1}logTOP}_t+{\phi }_2{logOPR}_t+{\phi }_3{logGPR}_t+{\phi }_{4}log{GPC}_t+{\phi }_5{logAFT}_t+{\epsilon }_t$$

(3)

The AFT in Equations (1)–(3) is a continuous variable that represents the cost of flaring. Therefore, the corresponding coefficient ${\phi }_1$ indicates the average marginal change in gas flared associated with one incremental change in flare tariff.

Figure 5 highlights a significant increase in the flaring tariff over the years, with recent tariff levels nearly 50 times higher than previous rates. Given this substantial change, the average marginal effect may not capture its impact adequately. In fact, during the period considered, there were five distinct regimes covering 1970–1984, 1985–1992, 1993–1998, 1999–2018, and 2019. To account for the shift in tariff regimes, the continuous tariff variable AFT is divided into two dummy variables: Pre2019AFT and Post2019AFT. Since tariff variations before 2018 were relatively minor, all pre-2018 rates were grouped together under the Pre2019AFT, which takes a value of one for the years 1970–2018 and zero otherwise. The variable Post2019AFT is equal to one for 2018 onwards and zero otherwise, capturing the implementation of the Flare Gas (Prevention of Waste and Pollution) Regulations in 2018 and the significantly higher flaring tariff that followed. Splitting the continuous tariff variable into two dummy variables to assess the effects of different tariff regimes was inspired by the author of [62], who applied a similar approach to examine whether Chinese national oil companies paid a premium for acquiring foreign assets compared with their counterparts across two distinct periods. Three other models, like Equations (1)–(3), were specified with two dummy variables. However, only the log-log model is presented in Equation (4) as follows:

$$\begin{array}{c}{logTGF}_t={{\phi }_0+{\phi }_{1}logTOP}_t+{\phi }_2{logOPR}_t+{\phi }_3{logGPR}_t+{\phi }_4{GPC}_t+{\phi }_4{Pre2019AFT}_t\\ +{\phi }_5{Post2019AFT}_t+{\epsilon }_t\end{array}$$

(4)

When the flare tariff in the pre-2019 period was relatively low compared to oil prices in the international market, producers preferred to flare and pay the penalty rather than investing in monetising associated gas. However, following the introduction of the 2018 Regulation, the increased flare penalty is expected to have a significant impact on reducing gas flaring. As the Nigerian government opined, the penalty was designed to “bite but not kill” the operators [63].

4. Results and Discussions

4.1. Summary Statistics

Table 1. Descriptive statistics of variables.

	Total Gas Flared (Bscf)	Total Oil Produced (Mbbls/d)	Oil Price (USD/bbl)	Gas Price (USD/Mscf)	Gas Pipeline Capacity (Mscf/km)	Adjusted Flare Tariff (USD/Mscf)
Statistics	TGF	TOP	OPR	GPR	GPC	AFT
Mean	637.199	1780.385	48.773	8.152	1.630	0.217
Median	655.385	1819.350	46.095	6.830	2.040	0.005
Maximum	953.000	2366.000	104.210	16.900	2.900	3.417
Minimum	187.820	1084.500	10.120	2.650	0.000	0.000
Kurtosis	−1.170	−0.628	−0.540	0.640	−1.589	12.951
Skewness	−0.300	−0.295	0.656	1.193	−0.327	3.739
Range	765.180	1281.500	94.090	14.250	2.900	3.417
Std. Dev	231.011	307.093	26.346	3.681	1.158	0.772
Obs	52	52	52	33	52	52

Source: Authors (generated with collected data).

presents the descriptive statistics of the dependent and independent variables used in this study. The total gas flared serves as the dependent variable (TGF), while the independent variables consist of oil production (TOP) in thousands of barrels per day (Mbbls/d), oil price (OPR) in United States dollars per barrel (USD/barrel), gas price (GPR) in United States dollars per thousand standard cubic feet (USD/Mscf), gas pipeline capacity in thousands of standard cubic feet per kilometre (Mscf/km), and the adjusted flare tariff (AFT) in United States dollars per thousand standard cubic feet (USD/Mscf). All monetary variables are in real terms.

4.2. Empirical Results

The empirical results from the estimating model (1)–(3) are reported in

Table 2. Estimation results for the level, semi-log, and log-log models.

	(1)	(2)	(3)
Variables	Level Form Model	Semi-Log Model	Log-Log Model
Oil produced	0.775 ***	0.001 ***	2.160 ***
	(0.109)	(0.000)	(0.410)
Oil price	−3.334 ***	−0.004 ***	−0.290 ***
	(1.071)	(0.002)	(0.092)
Gas price	−24.087	−0.043	0.175
	(15.319)	(0.027)	(0.146)
Gas pipeline cap.	−83.281 ***	−0.193 **	−0.133
	(40.843)	(0.072)	(0.110)
Flare tariff	−206.869 ***	−0.620 ***	−0.103 ***
	(77.345)	(0.135)	(0.023)
Constant	−294.497 **	4.723 ***	−9.224 ***
	(143.882)	(0.252)	(2.983)
Observations	52	52	52
Adj. R2	0.799	0.83	0.83
p-value of F-stat.	0.000	0.000	0.000

Note: *** represents statistical significance at less than 1%, and ** represents 1%, levels, respectively. Robust standard errors are in parentheses. Adj. R² represents the adjusted R-squared values. All price items are in real terms adjusted to 2010 prices using the historical CPI sourced from the World Bank data bank. Local currencies in the form of flare tariffs are also converted to their USD equivalent using the prevailing exchange rate data. Source: Authors’ computations.

. Column (1) of Table 2 presents the results of the level form (Equation (1)). Thereafter, the results of the semi-logged model (Equation (2)) are presented in Column (2) of Table 2. The log-log specification of the model are presented in column (3) of Table 2 below.

The baseline model estimated using Equation (1) contains a full sample of variables in their level form. The variables include total gas flare volume, total oil produced, oil price, gas price, gas pipeline capacity, and flare tariff adjusted for inflation and currency exchange. As seen from the value of the adjusted R², 80% of the variations in the volume of gas flared are explained by the model at the adopted 5% significance level. Four of the five explanatory variables were found to be statistically significant at the 1% level, except for gas price. This might be because the US LNG import price used as a proxy for the price of gas is not a true representation of the domestic price of the commodity in local markets. Additional reasons may be related to various missing data points that could generate a result with few degrees of freedom to validate the overall outcome. The total oil produced has a positive relationship with the volume of gas flared. This is logical because the more crude oil is produced, the more associated gas is produced as well.

The prevailing relationship between the volume of gas flared and gas price is also an inverse one but statistically insignificant in the model. The size of the gas price coefficient obtained shows that a 1-USD/mscf increase in the price of gas would cause a 24.09 billion scf reduction in the amount of gas flared. This is expected, as favourable gas pricing can motivate producers to find alternative uses for the associated gas as against flaring it. This contradicts the findings of [47], who find that gas prices exert a positive and significant impact on gas flaring. The coefficient of the flare tariff indicated that a 1-USD/mscf increase in the penalty charged by the regulators would result in a 207 billion scf reduction in the volume of flared gas. Thus, the model indicated that there is a statistically significant inverse correlation between the amount charged for the penalty and the volume of gas flared.

Both the semi-log and log-log transforms indicate that 83% of the variations in the dependent variable were explained by the independent variables, with both models statistically significant at the 1% level. The semi-log results show that a 1 Mbbl/d increase in the volume of oil produced leads to a corresponding increase of 0.62% in the volume of gas flared. Therefore, the relationship between oil production and gas flared is directly proportional. On the other hand, the inverse relationship between gas flared and the price of oil shows that a 1-USD/bbl increase in the price of oil would yield a reduction of 0.004% in the volume of gas flared. Similarly, a 1-USD/mscf increase in the price of gas results in a 0.04% reduction in the volume of gas flared. Finally, a 1-USD/mscf increase in the chargeable flare tariff would result in a 0.62% reduction in the volume of gas flared. All variables are statistically significant, except for gas price. The insignificance of the gas price to the volume of gas flared can be attributed to the relatively high skewness of the gas price data, which is partially eased after the log-log transformation in Column (3) of Table 2.

The log-transformed specification presented in column (3) of Table 2 provides the best fit for the data based on post-diagnostic tests of normality and constant variance. The adjusted R² value suggests that the model explains 83% of the variations in the volume of the gas flared. The signs and significance levels of the estimates remain largely consistent with those in column (2). Specifically, oil price and flare tariffs are inversely related to gas flaring, while oil production and gas prices are positively associated with the dependent variable. A detailed breakdown of the results shows that a 1% increase in the amount of oil produced leads to a 2.16% increase in the volume of the gas flared. However, a 1% increase in the price of oil leads to a 0.29% decrease in the volume of gas flared. This negative relationship between oil price and the volume of gas flared can be attributed to Nigeria’s membership in the OPEC. As a member of the OPEC, Nigeria must adhere to production quotas, which may have restricted its ability to expand its oil production during the study period. Gas prices show a positive relationship with the volume of gas flared, in that a 1% increase in the gas price leads to a 0.18% increase in the volume of gas flared. The key variable of interest, the flare tariff, exhibits an inverse relationship with the volume of the gas flared. The results show that a 1% increase in the tariff rate leads to a 0.10% reduction in the volume of flared gas. All variables in the log-transformed model are statistically significant at the 1% level, except for gas price.

4.3. Results of the Tariff Regime Change

To effectively evaluate the impact of changes in the gas-flaring tariff regime, tariff data were divided into two distinct periods. The first period, spanning 1970 to 2018, corresponds to the regime in which the gas-flaring tariff remained below USD 1/MScf. The second period began with the enforcement of the Flare Gas (Prevention of Waste and Pollution) Regulations, 2018, signed into law by the President and Minister of Petroleum Resources. This regulation introduced a revised tariff regime, effective from 2019, with an average flare penalty of USD 1.50/MScf.

Equation (4) is estimated using two dummy variables to represent these tariff regimes: Pre2019AFT represents the flare tariff regime for the period prior to the implementation of the 2018 regulation, from the introduction of the flaring tariff in the early 1980s to mid-2018. Post2019AFT represents the tariff regime from 2019 when the revised tariff regime became operational. The results of this study are presented in

Table 3. Estimation results before and after the 2018 gas flare regulation.

	(1)	(2)	(3)
Variables	Level Form Model	Semi-Log Model	Log-Log Model
Oil produced	0.771 ***	0.001 ***	2.541 ***
	(0.103)	(0.000)	(0.397)
Oil price	−3.070 ***	−0.003 ***	−0.282 ***
	(0.972)	(0.000)	(0.084)
Gas price	−19.801	−0.038	−0.074
	(17.007)	(0.031)	(0.159)
Gas pipeline cap.	−75.572 *	−0.202 ***	−0.230 ***
	(38.774)	(0.070)	(0.106)
Pre2019AFT	−22.306	−0.053 *	−0.045
	(16.043)	(0.029)	(0.029)
Post2019AFT	−3471.499 *	−8.632 ***	−9.263 ***
	(1728.176)	(3.128)	3.299
Constant	−428.383 ***	4.328 ***	−11.479 ***
	(139.381)	(0.252)	(2.824)
Observations	52	52	52
Adj. R2	0.829	0.849	0.864
p-value	0.00	0.00	0.00

Note: Other variables are included as in the previous models but not explained. Interpretation centres on before and after the passage of the 2018 gas flare regulations. *** represents statistical significance at less than 1%, and * represents 5% levels, respectively. Robust standard errors are in parentheses; Adj. R² represents the adjusted R-squared values. All price items are in real terms adjusted to 2010 prices using the historical CPI sourced from the World Bank data bank. Local currencies in the form of flare tariffs are also converted to their USD equivalent using the prevailing exchange rate data. Source: Authors’ computations.

.

As with before the creation of the dummy variables, three models were specified: the base, the semi-log, and the log-log models. Among the three models, the log-log model fit the data best, as shown by the highest R-squared value of 86%. Since the results of the first two models were similar to those prior to the separation of the flare tariffs, we report on the best-fit model only to save space. Column 3 of Table 3 presents this. Similarly, because both the dependent and independent variables are measured logarithmically, the estimated coefficients can be effectively interpreted as elasticities. The results indicate that a 1% change in the oil produced would lead to a 2.54 percentage change in the volume of gas flared. The relationship is statistically significant at less than 1% level with a corresponding p-value of 4.48 × 10⁻⁷. This is expected, as the majority of flaring volumes in Nigeria come from the associated gas. Therefore, in fields with no installed solution for processing the associated gas, an increase in oil production would lead to an increase in flaring.

The estimated coefficient of oil prices was negative and statistically significant (p = 0.002). This was not expected because higher oil prices would cause operators to ramp up production (subject to the availability of capacity), thereby increasing flaring. A possible explanation for this violation is that Nigeria is a member of the OPEC with an allocated production quota per day. Thus, even if there is an increase in the price of oil, the country cannot ramp up production with the intention of exporting, which could be interpreted as cheating by other members of the cartel. Likewise, the estimated coefficient of gas price is negative but statistically insignificant. As mentioned during the initial interpretation, this may be because the proxy used for gas price is not a good fit for the country. On the other hand, a 1% increase in gas pipeline capacity is seen to decrease gas flaring by 0.23 percentage points, and this decrease is statistically significant at the 1% level (p-value of 0.04). As additional gas infrastructure would aid the transportation of gas to desired markets, the exhibited relationship was expected, meaning that we have evidence to support our expectation that additional gas infrastructure would cause a drop in the volume of gas flared.

Furthermore, the results show that, while Pre2019AFT is statistically insignificant with a very small estimated coefficient of 0.05, Post2019AFT is statistically significant at the 1% level. The magnitude of the Post2019AFT coefficient of 9.26 (p-value of 0.009) indicates that a 1% increase in the flare tariff is associated with a −9.26% decrease in the volume of gas flared after the introduction of the new regulation. Therefore, we have evidence that, although previous tariffs have no significant effect on the volume of gas flared, current tariffs have a negative and statistically significant effect. These findings imply that higher tariffs significantly discourage wasteful flaring, without necessarily hurting oil production.

To interpret this outcome in the context of environmental economics, we assess the implied marginal abatement cost of reducing greenhouse gas (GHG) emissions resulting from an increase in the flare tariff. Using Nigeria’s 2022 gas-flaring baseline of roughly 7.4 billion cubic metres (bcm), equivalent to 261.2 billion standard cubic feet (scf) [13], a 1% hike in the flare tariff is projected to decrease flaring by about 24.2 billion scf annually. With a standard emission factor of 54.9 kg of CO₂-e per thousand scf of flared gas, this reduction translates to avoiding approximately 1.33 million tonnes of CO₂ equivalent (CO₂-e) emissions each year. At a prevailing tariff rate of USD 2.00/mscf [61], a 1% increase signifies an additional cost of USD 0.02/mscf. By multiplying this cost by the volume of gas reduced, the total annual cost associated with the avoided flaring amounted to approximately USD 484 million. Dividing this cost by the emissions avoided results in an implied marginal abatement cost of around USD 364 per tonne of CO₂-e avoided. Although this figure may seem high compared to global carbon pricing benchmarks, it reflects the cost of behavioural change in a setting characterised by outdated infrastructure, limited gas capture technologies, and historically weak enforcement. This finding underscores the potential effectiveness of economic instruments such as flare tariffs in reducing emissions in the oil and gas sector. However, it also highlights the necessity for complementary policies, including investments in gas utilisation infrastructure and enforcement mechanisms, to reduce costs and enhance compliance over time.

This implies that claims by authors such as [64], Refs. [12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45] who conducted qualitative studies and concluded that the pre-2019 tariff rates were ineffective in curbing gas flaring, are now backed by this quantitative study. Thus, the newly introduced tariff is not only effective but also the largest contributor to gas-flaring reduction based on the determinants considered in the final model. Our result is also in line with the findings of [17], who show that taxing flared gas at levels near federal land royalty rates significantly reduces flaring by 99% at 46% of the cost of carbon in North Dakota. The results also upheld the recommendation by [30], who advocated for an upward revision of the flare penalty in Nigeria but disputed the claims of [27], who said that paying fines cannot reduce flaring. Although the implementation of appropriate penalties is not the only factor driving reductions in gas flaring, the findings of this research contribute to quantifying the significance of setting appropriate tariffs that could bite but not kill the operators [63]. Therefore, finding an appropriate tariffing rate coupled with other incentives could help Nigeria put out its flares for good.

5. Conclusions/Policy Implication

This study investigated the impact of policy change on the volume of gas flared in Nigeria, with an emphasis on flare penalty regimes. By collating time-series data on the determinants of gas flaring and specifying econometric models, we were able to determine whether setting an appropriate tariff has a significant effect on the volume of gas flared in Nigeria. The regression results indicate that, while tariffs set prior to 2018 have a statistically insignificant effect on the volume of gas flared, those introduced after 2018 through the Gas Flare Regulation of 2018 have a significant impact on curbing flaring. The insignificance of applicable tariff regimes prior to 2018 did not serve as a sufficient disincentive for producers to find alternative uses of the gas being flared. This is because the insignificant effect of the tariffs on the operator’s economics compared to the magnitude of investments needed to install flare capture technologies to utilise gas at the flare is negligible. As a result, producers found it more economically beneficial to flare and pay rather than capture resources and utilise them. With the substantial increase in payable tariffs that became effective in 2019 owing to the passing of the Gas Flare (Prevention of Waste and Pollution) Regulations, 2018, the tariff was found to be statistically significant in reducing gas flaring in Nigeria. In addition to being statistically insignificant (with a p-value of 0.14), the initial regime covering the periods prior to 2018 contributed to a meagre 0.05% reduction in the volume of gas being flared. On the other hand, the regime after 2018 led to a 9.62% reduction in the volume of gas flared, and the coefficient was found to be statistically significant at the 1% level (p-value of 0.009). Hence, this study proves that finding an appropriate tariff that would motivate producers to find alternative uses for gas flares while maintaining their production operations is essential to the elimination of gas flares altogether, particularly in the Nigerian context.

Thus, the government’s decision to significantly increase tariff rates while advocating for other market-driven solutions could help the country end more than six decades of activity. As mentioned in this body of work, additional measures should be put in place to work together with the increased tariff rates to promote the utilisation of gas at the flare. It is interesting to note that Nigeria is walking through the talk by approaching this menace from multiple fronts. The policy change that inspired this research also introduced the Nigerian Gas Flare Commercialisation Programme (NGFCP), which aims to auction flare sites for third-party investors to commercialise gas resources. Subsequent papers in this research aim to explore the viability of the commercialisation programmes being promoted, as well as the prospect of the implementation framework in ending flaring in Nigeria.

To attain flare elimination targets faster, these combined initiatives must be followed by the well-targeted monitoring and supervision of the entire programme to ensure its full implementation and curve any potential diversion from laid-out processes and procedures. How effectively the nation enforces these regulations and the sustainability of flare reduction could be an area for further research when more data become available.