While the literature is flooded with analyses on the risk connectedness among oil and stock markets, the literature has failed to provide conclusive evidence of the direction of risk transmission and whether economic policy uncertainty impacts the level of risk spillover among both assets. Based on different data sets from different countries and regions, a bulk of early research studies mostly focused on the link between oil price changes and economic uncertainty (Hamilton, 1983; You et al. 2017; Fang and You, 2014; Antonakakis et al. 2014; Inchaupse et al. 2015, among others), while another group of studies exhibited evidence of a relationship between economic uncertainty and stock price changes (Kang et al. 2017; Kang and Ritta, 2013, 2015; Arouri et al. 2014; Mensi et al. 2014, among others). These studies have ignored the effect of economic uncertainty on the risk spillover among both assets.
Most recently, however, after the presence of the GFC and the significant recent volatility of oil prices, a new research direction on risk transmission among both assets has been noticed in the literature. Although an immense amount of research studies examined the risk transmission (or spillover) between oil and stock prices, they failed to provide convincing evidence and did not involve the effect of economic policy uncertainty on the risk spillover, particularly during the COVID-19 pandemic period. For instance, Arouri (2011) argued that the European sector stock markets and oil price risks are strongly interrelated, and the correlation varies greatly over time and across sectors. Using the generalized VAR-GARCH approach, intense volatility spillovers among world oil prices and the Gulf Cooperation Council (GCC) stock markets were reported by Arouri et al. (2011a). Using a similar methodology, Arouri et al. (2011b) found a strong volatility spillover among oil and stock markets in Europe and the United States at the sectoral level. Arouri et al. (2012) found that volatility spillovers between oil prices and sector stock returns are strong at both the aggregate and sector levels of European stock markets. This is a very important finding for portfolio managers.
Taking the 2014 oil price crisis into account, Awartania et al. (2016) argued that oil price risk is the main driver of stock market prices and that the risk transmission from oil to equity markets rose during the crisis. In a similar vein, Maghyereh et al. (2016) found that risk spillovers between oil and eleven major equity markets around the world were largely dominated by spillovers from the oil market to equity markets and not the other way around, due to the start of global recovery after the global financial crisis (GFC) in 2007–2008. However, in an international study, Zhang (2017) argued that oil shocks had a limited contribution to the world financial system, meaning that oil shocks rarely contribute to the stock markets of the six major stock markets around the world. Using the multivariate ARMA-GARCH approach and wavelet multiresolution analysis, Boubaker and Raza (2017) found that risk spillovers among oil prices and the BRICS stock markets were very strong. In summary, the above-referenced studies have provided conflicting results and inconclusive evidence of the risk transformation among the oil-stock nexus.
Turning to the impact of economic policy uncertainty, the literature remains scarce to some extent. Although few studies have examined the impact of economic policy uncertainty on the risk connectedness among the oil-stock nexus (Kang and Ratti, 2013; You et al. 2017), this topic is still rather understudied and remains an appealing research venue. Starting with the study of Kang and Ratti (2013), an increase in oil-market demand led to a significant increase in economic policy uncertainty but reduced real stock returns. In a similar vein, Kang and Ratti (2015) reported that an increase in economic policy uncertainty in China led to a less negative effect on real oil prices and real stock market returns, concluding that volatility in oil prices causes a significant rise in China's economic policy uncertainty but reduces the real stock market returns.
Another study by You et al. (2017) indicates that stock returns were asymmetrically responsive and highly correlated to the effects of oil price shocks and China's economic policy uncertainty. Based on the DCC-MIDAS model, Fang et al. (2018) also reported a significant positive influence of EPU on the long-run oil-stock correlation. However, when the EPU category-specific indices were used, they found a strong positive impact on the correlation except for the monetary policy uncertainty and national security uncertainty reported.Another study by You et al. (2017) indicates that stock returns were asymmetrically responsive and highly correlated to the effects of oil price shocks and China's economic policy uncertainty. Based on the DCC-MIDAS model, Fang et al. (2018) also reported a significant positive influence of EPU on the long-run oil-stock correlation. However, when the EPU category-specific indices were used, they found a strong positive impact on the correlation except for the monetary policy uncertainty and national security uncertainty reported. By exploring the inherent dynamics and the casual interrelationships between various types of financial uncertainty indices and oil markets, Uddin et al. (2018) found strong heterogeneity in the revealed interrelationships among the indices over time and across scales.
Looking at the literature during the COVID-19 crisis, it is noticed that although evidence of the spillover effect among oil and stock markets was reported, the literature does not heavily focus on the impact of economic policy uncertainty on the risk connectedness among both markets and rarely uses the real economic uncertainty (particularly the business condition index). For instance, Sharif et al. (2020) argued that the effect of the COVID-19 pandemic on the geopolitical risk was substantially higher than on the US economic uncertainty, and the COVID-19 risk was perceived differently over the short and long runs. Using the MWC method, Ali et al. (2022) found strong co-movement between the oil futures market and five stock markets (the United States (US), Canada, China, Russia, and Venezuela) before and during the COVID-19 pandemic at lower frequencies. They argued that, at the beginning of the COVID-19 period, strong co-movements were found at high frequency, but positive co-movements were found at low frequency during the overall COVID-19 period.
Using the TVP-VAR framework, Bahloul and Khemakhem (2021) showed evidence of spillover risk transmission among oil and Islamic stock markets after the COVID-19 pandemic. Another analysis performed by Belhassine and Karamti (2021) showed that risk spillovers between the oil prices and the stock markets of rich oil importing and exporting countries are strongly interdependent in the long run between oil and stock indices; oil-exporting countries showed a higher correlation in the very long run than importing countries. Furthermore, Cui et al. (2021) stated that the total risk spillovers among oil and stock markets in oil-importing and oil-exporting countries were mostly transmitted in the long run, in which the oil market receives much more risk spillovers from the stock markets in the US, EU, Canada, and Russia. Furthermore, they argued that the major global crises, such as the GFC, the oil price collapse in 2014, and the COVID-19 pandemic, had greatly contributed to the increase in the level of risk spillover.
Another strand of literature has substantially used the MWC and PWC methods to examine the level of coherencies (time and frequency dependence) on risk connectedness. For instance, Xiang et al. (2021) found a positive pass-through from oil price volatility to inflation in China in the short run, but in the medium and long run, the pass-through persists over time. Choi (2022) found evidence of interdependence between the global geopolitical risks (GPR) and the volatility of stock markets indices of the three East-Asian countries (China, South Korea, and Japan) in the short run, and concluded that the connectedness between GPR and volatility a time dependence behavior. When filtering out for the Korean GPR, they found more co-movement between the global GPR, rather than the Korean GPR, and volatility of the Korean and Japanese stock market indices.
Aguiar-Conraria et al. (2018a) found lead-lag relationships between the US federal funds rate (as a monetary policy proxy) and inflation and the output gap that differ over time and cyclical frequencies. In light of the relationship between economic policy uncertainty and stock price, Ko and Lee (2015) argue that economic policy uncertainty is negatively associated with stock price, but that association changes over time, exhibiting low- to high-frequency cycles. Furthermore, the timing of frequency changes overlaps when there is a strong correlation between US policy uncertainty and other countries’ policy uncertainty. In introducing a new direction to the relationship between EPU and stock prices, Das and Kumar (2018) showed that the combined effect of the international (US EPU) and the domestic EPU is more significant for stock prices in developed markets (European countries and Japan), while stock prices in emerging markets are more sensitive to the domestic EPU.
Wu, Zhu, Xu and Yang (2020) pointed out that crude oil is a major driver of co-movement between global stock markets in the median and long term. When comparing the oil-importing and oil-exporting countries separately, they reported that the crude oil has relative lesser impact on the co-movement in oil-importing or in oil-exporting countries, indicating that its co-movement is caused by other factors. Albulescu and Mutascu (2021) found a strong co-movement among fuel prices in France, Germany, and Italy at medium and small frequencies, claiming that this co-movement is largely driven by the international oil prices in the medium and long run. When using the PWC and controlling for the effect of crude oil prices, the co-movements started to deteriorate due to the presence of changes in European fuel taxes and the lack of a real cost competition relying on fuel price dynamics. The co-movements of gasoline prices are stronger than those reported for diesel prices.
Altogether, although the existing studies have used the PWC and MWC, they have reported mixed and inconclusive evidence of the presence of coherency in the dynamic spillover among financial assets. Given that the existing literature has mostly focused on economic uncertainty and price changes and returns, there is a lack of literature examining the effect of real economic uncertainty on the risk connectedness in the oil-equity nexus. Furthermore, one can also notice that they have generally ignored the effect of the COVID-19 crisis on the level of co-movements. Furthermore, although the literature has extensively used the economic policy uncertainty proposed by Baker et al. (2016) with their category-specific indices, they rarely used the real economic uncertainty indices, in particular, the business conditions index. In terms of the methodological approaches, it is clear that the literature has focused on volatilities measures such as DCC-GARCH, the structural VAR model, the time-varying copula-GARCH-based CoVaR approach, the multivariate ARMA-GARCH approach, and wavelet multiresolution analysis, Quantile-VAR, and VAR-GARCH, but they have largely ignored the use of multiple and partial wavelet coherencies in examining the role of the real economic uncertainty on the risk transmission among oil and stock markets.