Oil subsidy removal and spillovers between crude oil and stock markets: empirical study from GCC countries Open Access

Recently, the issue of using subsidies to lower oil prices has intensified, especially with the global push to reduce greenhouse gas emissions. In response, several Middle Eastern countries dismantled their subsidy systems and aligned their oil prices with international market rates. This change makes these economies more sensitive to price fluctuations, which results in higher inflation and necessitates tighter monetary policies. This move has direct implications for the region’s stock markets. We anticipate heightened interconnectedness between stock markets and oil prices following the removal of subsidies.

The Gulf Cooperation Council (GCC) countries subsidized their energy products, and all countries removed these subsidies in 2015–2016. Saudi Arabia first acted on January 29, 2016 and Kuwait implemented it last September. Subsidies, particularly significant in Saudi Arabia, were implicit: oil sold below global prices without direct government spending, making the amounts difficult to quantify. Studies have produced varying estimates using the price gap approach. For instance, Alyousef and Stevens (2011) calculated Saudi Arabia’s implicit subsidy between 2000 and 2009 at five to six billion dollars. This removal marked a significant policy shift, aligning domestic prices with global markets and reducing market distortions across the GCC region.

Subsidies were removed primarily because of market distortions and the adverse economic impact. Following the 2015 Paris Agreement, pollution reduction has become a priority, particularly in the Middle East’s heavily polluted cities. Ending subsidies raises fuel prices, reduces demand, encourages renewable energy investments, helps countries meet emission reduction targets and promotes energy diversification. This shift also generates positive spillover effects on broader economic development.

In light of the recent turbulence in the global oil and equity markets, we aim to determine the factors that cause unpredictability. We focus on the role of energy subsidies in driving both markets because they affect energy demand, output, investment and financial markets. We analyze a sample of oil-producing Middle Eastern countries, examining market interactions with subsidies and the consequences of their removal.

In response to international climate commitments, especially after the 2015 Paris Agreement, several Middle Eastern hydrocarbon-dependent economies have reformed their energy subsidies by aligning domestic fuel prices with global benchmarks. In the GCC, where subsidies are implicit, their removal marks a major shift in fiscal and energy policies. These reforms aim to improve price efficiency, reduce pollution and encourage renewable energy investment, but increase exposure to oil price volatility, which affects inflation, monetary policy and financial markets. Despite its importance, the financial market implications of subsidy removal remain underexplored (Okoroafor and Okoroafor, 2025; Rentschler et al., 2017; You-How et al., 2018). Existing studies have overlooked the temporal evolution and structural changes in oil–equity linkages that may result from such reforms.

This study investigates the impact of fuel subsidy removal on financial market interactions in the GCC by analyzing return spillovers between GCC stock markets and key international indexes – Brent crude oil, gold, the US dollar index (DXY), US Treasury bonds (T-bonds) and the CBOE crude oil volatility index (OVX) – from January 1, 2007, to November 9, 2022. We employ both the Diebold-Yilmaz (DY) spillover index and the time-varying parameter vector autoregression (TVP-VAR) connectedness framework to assess risk spillovers before and after subsidy removal. Additionally, we employ the Granger non-causality tests to evaluate the directional relationships across these periods.

Our study advances the literature by linking the macroeconomic effects of energy subsidy reforms to the oil–stock market relationship. It uniquely examines how removing oil subsidies affects the interconnectedness between the GCC stock markets and global oil prices. By comparing the periods before and after subsidy removal, we capture the dynamic market adjustments to policy changes, offering deeper insights for investors and fund managers navigating evolving policy risks.

Second, the findings have crucial implications for policymakers in hydrocarbon-dependent economies. The changes in market interconnectedness after subsidy removal show that energy reforms influence financial volatility, risk transmission and contagion across sectors and borders. This calls for integrated policies that combine energy reform with enhanced financial oversight and macroprudential regulation. Identifying the key shock transmitters and receivers enables better risk anticipation and monitoring. Moreover, the absence of major post-reform financial disruptions suggests the potential for further policies, such as carbon pricing and green finance incentives, without undermining market stability.

Third, this study links subsidy reforms to GCC stock market behavior, which is crucial for economic diversification. The findings show varied country responses due to differences in economic structures, fiscal policies and market maturity. This insight helps policymakers align diversification strategies with energy reforms and evolving markets, highlighting the key role of financial market development as both an indicator and a facilitator of successful economic transformation.

Finally, this study enhances our understanding of the GCC progress toward sustainable development and the United Nations Sustainable Development Goals (UN SDGs). The fact that subsidy removal has not destabilized financial markets suggests that GCC countries can implement carbon pricing and environmental policies. These initiatives are vital for achieving climate goals and green growth, aligning hydrocarbon-dependent economies with global sustainability and guiding coordinated energy and financial reforms for long-term objectives.

Our empirical findings reveal the varying effects of abolishing subsidies on the oil–stock market relationship in GCC countries. In the UAE, the bidirectional causality between oil and stock prices persisted before and after subsidy removal. However, in Saudi Arabia, bidirectional causality was observed only before removal. Oil price volatility has a unidirectional effect on stock markets in all cases. Despite subsidy elimination, the market connection does not strengthen, as the results become less significant for certain countries. The connectedness analysis indicates distinct patterns in the oil–stock market relationship before and after subsidy removal. The stock markets, interest rates, and oil volatility of the UAE and Qatar serve as primary channels for transmitting returns.

The remainder of this paper is organized as follows: Section 2 presents the literature review. Section 3 describes the methodology, and Section 4 provides an overview of the data and descriptive statistics. Section 5 presents the empirical results. Finally, Section 6 concludes the paper.

Degiannakis et al. (2018) identified several mechanisms by which oil prices influence stock markets. The stock valuation channel affects company cash flows differently depending on the country’s involvement in the oil industry. The monetary channel affects inflation and interest rates, which, in turn, influence stock prices. The output effect describes how higher oil prices boost production in oil-exporting countries, thus positively affecting stock markets. Additionally, a fiscal channel exists through which oil revenues fund infrastructure investments that raise stock prices. Conversely, the uncertainty channel suggests that elevated oil prices increase economic uncertainty, reduce investment and depress stock prices.

Empirical research on oil price–stock market linkages often uses time-series models that focus on risk. Basher and Sadorsky (2006) employed a multifactor model in emerging economies, while Elyasiani et al. (2011) confirmed significant oil–stock price links at the industrial level. Degiannakis et al. (2018) show that this relationship varies over time and depends on the nature of the shocks. The extant literature shows mixed results: Ferson and Harvey (1994) report a negative correlation, Joo and Park (2021) find that oil volatility positively influences returns when equity returns are high, and Huang et al. (1996) find no such correlation.

Several studies highlight that the oil–stock relationship differs by country type (exporters versus importers) and exhibits asymmetries. Park and Ratti (2008) find that increased oil volatility lowers stock returns in Europe but not in the USA, with no evidence of asymmetric responses. Gupta (2016) observes that oil-producing countries’ stock returns are more sensitive to oil price changes than those of non-producers. Similarly, Atif et al. (2022) show that oil-exporting countries are more affected by oil price fluctuations, with COVID-19 inducing negative global impacts. Yadav et al. (2024) document mixed connectedness between Saudi crude oil and GCC equity markets, noting Omani markets as shock transmitters and Bahraini markets as receivers with stronger long-run linkages.

Alhassan et al. (2019) focus on firms reliant on oil in subsidizing countries and find a stronger oil–stock return correlation that weakens during volatile periods. A firm’s oil dependence and subsidy status significantly influence this relationship. Studies conducted before and after subsidy removal provide mixed conclusions. Li and Sun (2018) argue that China’s subsidy phase-out was driven by low global prices and policy reforms; however, subsidy removal alone may increase CO₂ emissions by shifting consumption to coal. Okoroafor and Okoroafor (2025) highlight the negative effects of subsidy removal on households and inflation, but the positive impacts on government revenue and emissions. Rentschler et al. (2017) showed that firms can offset subsidy-driven cost increases through efficiency and fuel switching. You-How et al. (2018) examined Malaysia’s 2014 subsidy cut and found intensified information spillovers from Brent crude oil to the stock market, indicating increased investor sensitivity to post-reform oil price signals.

Despite growing interest, research on the effects of fossil fuel subsidy removal on oil stock spillovers and investor behavior remains limited. To fill this gap, this study examines the interactions between the GCC stock markets and oil prices before and after subsidy reforms, capturing shifts in spillovers, risk transmission and investor responses. We expect subsidy removal to initially disrupt market interactions owing to higher energy costs and uncertainty. However, aligning domestic prices with global benchmarks may enhance market efficiency and transparency and improve the influence of oil price signals on financial markets. These findings provide insights into energy policy, sovereign wealth fund strategies and macroprudential regulation design, which are vital to the resilience of GCC economies during energy transitions.

Based on this literature, we hypothesize the following hypothesis:

The TVP-VAR model, based on Diebold and Yilmaz’s framework, allows coefficients to vary freely over time without imposing structural assumptions, thus avoiding misspecification risks common in STAR models. Antonakakis et al.’s (2020) specification is widely used for its robustness and capacity to capture evolving cross-market interactions through stochastic parameter evolution, making it well suited for analyzing dynamic relationships across assets and markets under changing economic conditions.

Using stochastic volatility Kalman Filter estimation ^[1], the TVP-VAR model can be expressed as follows:

where $Yt$ is an $N×1$ vector of conditional volatilities, $Yt−1$ is an $Np×1$ lagged conditional vector, $βt$ is an $N×Np$ matrix of time-varying coefficients and $ϵ$ is an $N×1$ vector of error disturbance terms with an $N×N$ matrix of time-varying variance-covariance matrix, $St$⁠. In Eq. (2), $βt−1$ is determined by the lagged values of $βt$ and an $N×Np$ error matrix with an $Np×Np$ variance-covariance matrix.

To estimate the generalized forecast error variance decomposition (GFEVD) of Pesaran and Shin (1998), the VAR model is transformed into a vector moving average (VMA) representation following the Wold representation theorem:

where $βt$ is the transpose of the VAR coefficients at time $t$⁠, and $At=[A1,t,A2,t,...,Ap,t]′$⁠. The generalized impulse response function (GIRF) is derived by comparing the differences between a $J$-step ahead forecast with and without a shock in variable $i$ and can be computed as the following equation:

For $J$-step, is the forecast horizon, $δj,t$ is the selection vector, where only the $j$-th position is set to one. The GFEVD, which measures the variance attributable to a given variable relative to the others, is normalized as follows:

Such that $∑j=1Nϕ̅ij,tN(J)=1$ and $∑i,j=1Nϕ̅ij,tN(J)=N$⁠. The total connectedness index (TCI) can be estimated as

The TO directional connectedness index is given by

The FROM directional connectedness index is as follows:

Finally, the NET directional connectedness index is calculated as follows:

where a positive (negative) value of NET $Ci,tg$ suggests that variable $i$ is the net shock transmitter (receiver). Additionally, the pairwise connectedness index (PCI), which quantifies the interconnectedness between variables $i$ and $j$⁠, is determined as follows:

In this analysis, connectedness measures were estimated using a one-lag order (BIC) and 10-forecast horizon. To illustrate the time-varying connectedness index, a rolling window approach with a 200-day estimation is applied.

This study employs the time-varying parameter vector autoregression (TVP-VAR) connectedness approach developed by Antonakakis et al. (2020) to examine dynamic spillover effects across markets. This method surpasses constant-parameter and rolling-window VAR models (Pham et al., 2023) by avoiding sensitivity to window size, thus reducing estimation bias and data loss (Younis et al., 2024). The TVP-VAR allows parameters to evolve over time, rapidly adjusting to structural shocks and major events, capturing time-varying relationships with high accuracy. This is crucial for our sample period from January 1, 2007, to November 9, 2022, encompassing the Global Financial Crisis, the 2014–2016 oil glut, the 2017 Gulf diplomatic crisis and the COVID-19 pandemic – times when static or rolling-window models often underperform. As Antonakakis et al. (2020) demonstrated, TVP-VAR is robust to outliers and structural breaks, improving spillover measurement reliability and providing a solid foundation for connectedness analysis.

This study considers daily closing price data for seven GCC equity markets (Abu Dhabi, Bahrain, Dubai, Kuwait, Oman, Qatar and Saudi Arabia), Brent oil, gold, DXY, T-bonds and the OVX, spanning January 1, 2007, to November 9, 2022, from DataStream. The sample enables the analysis of differing oil dependency impacts on subsidy removal. For example, Saudi Arabia, a major producer, may react differently to Oman, a minor producer. Fuel subsidies were removed in 2015–2016 in Saudi Arabia (January 29, 2015), the United Arab Emirates (August 1, 2015), Oman (January 15, 2016), Bahrain (January 12, 2016), Qatar (May 1, 2016) and Kuwait (September 1, 2016).

The timeframe for this study is deliberately selected to capture the major global and regional events influencing the financial and commodity markets, including the 2008 Global Financial Crisis, the 2014–2016 oil price collapse, the coordinated removal of fuel subsidies across the GCC countries in 2015–2016, and the 2020 COVID-19 pandemic. These episodes mark distinct phases of economic turbulence and policy change, enabling an analysis of evolving market dynamics and interdependence. This period allows a rigorous assessment of the short- and long-term effects of energy subsidy reforms in oil-exporting economies and differential market reactions across varying oil dependence, ensuring robust, contextually relevant and empirically meaningful results.

We compute the continuously compounded daily returns by calculating the difference between the logarithms of the percentage change in two successive prices, defined as $rt=ln(Pt/Pt−1)×100$⁠. Figure A1 presents the dynamics of the stock, Brent, gold, DXY and T-bond markets along with the OVX. Markets showed synchronized movements, with sharp declines during crises, such as the 2008 Global Financial Crisis, the oil surplus of 2014–2016 and the COVID-19 pandemic in 2020. Oil prices fell during these events, affecting oil-dependent GCC economies, where reduced oil revenues weakened stock markets. Saudi Arabia’s market nearly doubled from 2009 to 2020, rising sharply until 2022, whereas those of Bahrain and Oman declined. Gold and the dollar surged during crises, showing safe haven roles. OVX peaked in 2020, reflecting extreme oil price volatility.

Panel B illustrates the significant volatility in dynamic returns during crises, indicating the markets’ heightened sensitivity to external shocks and economic news. These fluctuations reflect unexpected price movements, prompting investors to use derivatives, including options and futures, as hedging tools to mitigate the adverse effects of sudden market changes.

Table A1 summarizes the returns for the seven GCC stock markets, Brent, gold, DXY, T-bonds and OVX. The mean returns are positive, except for Abu Dhabi and Qatar; Bahrain has the highest (0.276) and Abu Dhabi has the lowest (−0.118). Dubai showed the highest risk (13.6) and Oman the lowest (5.716). Brent and T-bond have positive average returns, whereas gold and DXY are negative. The skewness is negative, except for DXY, indicating more frequent positive returns. The GCC equities, oil, and T-bonds exhibit fat tails. The Jarque–Bera tests reject the normality. Elliott et al.’s (1996) unit root test confirms stationarity. The Q(20) and Q2(20) tests indicate serial autocorrelations in returns and squared residuals, respectively.

Figure A2 presents a heat map of the pairwise correlations between various assets, illustrating the relationships between asset returns. The DXY shows negative correlations with all markets, especially oil and gold, because a stronger dollar increases costs for foreign buyers, reduces demand, and lowers prices. GCC stock markets exhibit high correlations, reflecting close economic interdependence. Brent crude oil is positively correlated with GCC stocks because rising oil prices enhance earnings and growth in oil-exporting countries, and boost stock prices. Conversely, OVX is negatively correlated with the GCC stocks, indicating that unpredictable oil prices undermine economic stability. Gold shows low correlations, except for the US dollar, suggesting portfolio diversification benefits.

Table 1 presents the return connectedness estimates using the DY method, which reveal an overall spillover of 44.6%. Each GCC equity market is primarily driven by its own shocks, with Bahrain being the most impacted and Dubai the least impacted. Thus, 44.6% of the forecast error variance is due to spillovers from other markets, indicating moderate interdependence, whereas 55.4% is due to market-specific shocks. This result underscores the importance of accounting for both external and domestic factors in empirical models of return transmission, and highlights the potential for portfolio diversification across markets with lower spillover exposure. Abu Dhabi and Dubai are the main transmitters of returns, with Abu Dhabi transmitting 21.6% to Dubai and receiving 22%. Qatar mainly transmitted to Kuwait (9.8%), whereas Abu Dhabi and Dubai significantly influenced Oman. These results align with those of Hanif et al. (2025), who identified the UAE and Saudi Arabia as the dominant players affecting GCC equity dynamics. The GCC markets show limited influence from non-equity assets; however, Brent and OVX are key contributors to GCC returns, highlighting the importance of oil, consistent with Arouri et al. (2011) and Ziadat and McMillan (2022). OVX is the largest net return transmitter, followed by T-Bond, while gold and Saudi Arabia’s stock index are net recipients. Gold’s negative spillover (−11.1%) confirms its role as a safe haven and diversifier (Mensi et al., 2015).

We analyze return spillovers using the TVP-VAR method with the Bayesian information criterion to select the optimal lag (Table A2). The total connectedness index is 42.7% and all markets are significantly influenced by their own shocks. Abu Dhabi, Dubai and Qatar are the main transmitters and receivers. OVX is the largest net transmitter, followed by Qatar; gold is the largest net receiver, followed by DXY. Under the TVP-VAR model the UAE and Kuwait shift from net transmitters in the DY framework to net receivers. This confirms H1, showing that oil–stock spillovers vary by country and are influenced by domestic energy pricing policies and oil revenue dependence. Countries with substantial energy subsidies exhibit dampened spillovers, whereas those with market-based pricing and higher oil reliance exhibit greater sensitivity. This highlights the critical role of subsidy reform in enhancing market efficiency and transparency but also signals increased volatility, necessitating strong regulatory oversight. Policies focusing on economic diversification and financial resilience are essential for sustainable development during fluctuations in oil prices.

The TVP-VAR connectedness results align with the Granger causality findings, showing strengthened market linkages after subsidy removal and a shift toward market-driven oil prices. This finding supports You-How et al. (2018), who found intensified Brent oil price spillovers to Malaysia’s stock market after subsidy cuts. Abu Dhabi’s strong ties with Oman and Qatar before subsidy removal became more complex afterward. Qatar’s influence remained significant, Oman’s influence declined, and Dubai and Kuwait’s links strengthened. Abu Dhabi maintained strong ties with gold, OVX and interest rates. After subsidy elimination, Bahrain and Dubai showed increased market interconnectedness, whereas Qatar and Saudi Arabia’s connectedness remained stable. Oman displayed higher connectivity with regional markets before the subsidy removal, with oil prices influencing its markets equally before and after the reform. These differences reflect the market structure, reform strategies and fiscal capacity of each country. Bahrain and Dubai combined subsidy cuts with rapid non-oil diversification, leading to stronger regional spillover. With their hydrocarbon-focused exports and large sovereign buffers, Qatar and Saudi Arabia have maintained stable connectedness. Oman’s smaller trade open market showed greater pre-reform co-movement, while oil price impacts persisted.

Overall, oil prices have minimal effects on most markets except for Oman and Saudi Arabia. In Saudi Arabia, oil prices influenced markets before but not after subsidy removal. The impact of the volatility index increased after the removal, indicating a market shift toward oil price volatility and related derivative products.

Figure A3 shows the dynamic total connectedness from 2007 to 2022 using the TVP-VAR and DY12 methods, both moving in tandem. The DY spillover index ranged from 35% to 60% during the global financial crisis and the COVID-19 pandemic. Alshater et al. (2024) found similar heightened connectedness during the COVID-19 pandemic among GCC equity markets and various regional and global financial markets. Return spillovers react to crises and subsidy removal, rising from 36% to 45% during the 2015–2016 GCC fuel subsidy removal, confirming the subsidy’s impact on GCC markets (You-How et al., 2018). Subsidy removal affects oil prices, company costs and stock volatility, which spill over to markets and influence investors’ behavior.

Figure A4 presents the net directional spillovers for each market. The DY method shows the UAE stock markets as net transmitters, while TVP-VAR indicates them as net recipients, especially during the September 2008 Global Financial Crisis and the March 2020 COVID-19 outbreak. This difference arises because TVP-VAR allows continuously evolving parameters to capture abrupt dynamic changes. Both methods show that Qatar is a consistent net contributor. Other GCC markets are typically net receivers, except Kuwait, which became a contributor after the January 2016 subsidy cuts. Brent crude oil was a net contributor before the January 2015–January 2016 subsidy removal but became a net receiver afterward. Gold and the DXY remain net recipients, whereas T-bonds and the OVX are persistent net transmitters. These findings align with those of Alqahtani et al. (2019) and Hussain and Rehman (2023), who document significant, heterogeneous and divergent effects of oil price uncertainty on GCC stock market spillovers.

Our findings reveal notable market interconnectedness. Figure A5 illustrates the pairwise directional network using the DY (Panel A) and TVP-VAR (Panel B) models for the entire sample. Panel A underscores the pivotal role of oil volatility in shaping oil price returns, indicating that volatility is a key driver rather than a mere outcome of price movements. This finding suggests that uncertainty directly influences returns through shifts in risk premiums and market sentiments. The T-bond rate impacts the DXY and gold, as rising US interest rates attract foreign investment in dollar assets, strengthening the DXY and discouraging gold purchases (Wang and Chueh, 2013). Among the GCC markets, Abu Dhabi, Dubai and Qatar emerged as a major spillover contributors, whereas Saudi Arabia, Oman and Bahrain were primary recipients. This asymmetry suggests a regional hierarchy in market influence, with the former exerting a stronger cross-market impact, potentially due to higher liquidity, greater integration with global financial markets and more dynamic economic structures. Panel B shows that the UAE markets have become the primary recipients of spillovers from Qatar, which remains the leading transmitter.

We examine return spillovers among the GCC countries before and after fuel subsidy removal (Figures A6–A12). Subsidies reduced production costs, stabilized inflation and insulated markets from global shocks but weakened integration with international finance. Before its removal, Qatar, Oman and Dubai contributed the most to Abu Dhabi, with T-bonds and Bahrain acting as receivers (Figure A6). After removal, Qatar and OVX became key contributors, while gold, T-bonds and DXY were primary recipients, reflecting flight to quality during uncertainty. OVX’s increased role shows rising oil risk premiums. Gold, T-bonds and the DXY acted as safe havens, while Bahrain’s limited post-reform connectedness highlights its diversification potential.

Figure A7 indicates that the Bahrain market was the primary transmitter to the T-bond and receiver from Kuwait prior to the removal of subsidies (Panel A). Following its removal, DXY became the primary recipient of this market (Panel B). Prior to the subsidy cuts, Qatar was the primary contributor to Dubai (Figure A8). After subsidy cuts, the OVX emerged as the primary transmitter in Dubai. In the Kuwaiti market (Figure A9), a similar pattern can be observed, with a weak connection with oil volatility prior to the subsidy cuts, but a significant return spillover from OVX afterward. As depicted in Panel A of Figure A10, Qatar (DXY) was the primary contributor (recipient) to Oman before and after fuel subsidies were removed. Oman was the primary contributor to Abu Dhabi and the main recipient of Qatar. However, Brent acted as a net transmitter of the Omani market’s return spillovers both before and after the removal of fuel subsidies. The Qatari stock market is a net contributor to all other markets (Figure A11), except OVX, which functions as a net transmitter to the Qatari stock market before and after subsidy cuts. While Brent was the primary return spillover contributor to Saudi Arabia before the removal of fuel subsidies (Figure A12), OVX became the strongest transmitter of returns to the market after subsidy cuts. These findings demonstrate the impact of fuel subsidy removal on the connectedness between the GCC markets and other markets. These findings confirm H2, indicating significant shifts in spillovers between the oil and GCC stock markets following fuel subsidy reforms, reflecting evolving investor behavior, market sensitivity and risk dynamics across the region.

These findings reveal growing financial interconnectedness in the Gulf, with GCC stock markets becoming increasingly sensitive to energy pricing reforms. Post-reform, global risk factors such as oil price volatility and safe haven assets have become major spillover transmitters, highlighting greater integration with global finance. This shift requires improved risk management, market transparency, infrastructure and capital market development to enhance resilience.

To ensure the robustness and reliability of our findings, we conducted a comprehensive robustness check by performing a sensitivity analysis and estimating the Dynamic Equicorrelation GARCH (DECO-GARCH) model following Engle and Kelly (2012). Table A3 reports the estimate results of the DECO-GARCH model ^[2]. This analysis examines whether the Total Connectedness Index (TCI) is affected by variations in key modeling parameters, specifically the forecast horizon and rolling window size used in the estimation process. Figure 1 demonstrates that TCI remains consistently stable across different specifications, regardless of whether short- or long-term horizons or narrow or broad windows are applied. This stability suggests that methodological artifacts do not drive the observed connectedness dynamics. The robustness of TCI to these alternative specifications strengthens the credibility of our empirical findings and affirms the reliability of the interconnectedness patterns documented in the main analysis.

Regarding the relationship between stock prices and oil volatility, Table A4 presents the Granger causality test of oil volatility on all GCC stock prices with and without subsidies.

This study explores the impact of fuel subsidy removal on the nexus between the oil and GCC stock markets using a Granger causality test and the TVP-VAR spillover index method to measure the linkage before and after subsidy reduction.

Our empirical analysis reveals that subsidy removal impacts oil–stock market interactions differently across GCC countries and varies by causality direction. The UAE sustained a bidirectional relationship between oil prices and stock markets, whereas Saudi Arabia exhibited two-way causality only before subsidy elimination. This difference arises because the UAE’s diversified economy uses monthly benchmark-linked fuel pricing that quickly passes global oil price changes to domestic markets, while maintaining strong bidirectional links despite subsidy removal. In contrast, Saudi Arabia’s oil-dependent economy implemented a one-time administrative subsidy reform in 2015, with few adjustments afterward, thereby reducing market sensitivity to oil price swings. Its fiscal buffers further weakened the oil-equity links, ending the two-way post-reform causality. Unidirectional causality exists between oil volatility and other GCC stock markets. Subsidy removal does not significantly strengthen market connections to oil prices or volatility because domestic subsidies minimally affect broader economies. The connectedness analysis aligns with the causality findings, with oil volatility, the UAE and Qatari stock markets and interest rates acting as the main net transmitters. In conclusion, factors beyond subsidies, such as the number of oil-dependent firms, influence oil-stock market interactions, enhance market responsiveness and improve asset allocation.

These results have important implications for financial regulations, investment strategies and energy policies in hydrocarbon-dependent economies. Regulators should enhance their financial surveillance and incorporate oil price volatility into stress tests, particularly when equity markets transmit regional shocks. Identifying key net transmitters, such as the UAE and Qatar, highlights the need for proactive oversight to limit the contagion. Sovereign wealth funds should be recalibrated by increasing allocations to low-carbon, less-oil-correlated assets, such as green bonds, ESG equities and climate-resilient infrastructure. Dynamic portfolio rebalancing, which reflects energy market conditions, can improve capital preservation, diversification and alignment with climate transition goals. Our findings suggest that the GCC governments can pursue carbon pricing and green finance reforms without causing market instability. Subsidy removal creates fiscal space for ambitious environmental policies, such as carbon taxes, enabling public resources to shift toward sustainable development and climate mitigation, marking a crucial step in the region’s green transition.

Future research should disaggregate stock markets by sector, especially in energy-intensive and export-oriented industries, to capture nuanced subsidy reform effects. Incorporating investor sentiment and herd behavior can improve causality and connectedness. Using high-frequency data can reveal the short-term market responses. Additionally, exploring the interactions among subsidy policies, environmental uncertainty and geopolitical risks, considering energy transitions and climate commitments, would deepen the understanding of financial risk dynamics in the region.

The supplementary material for this article can be found online.