How volatility in the oil market and uncertainty shocks affect Saudi economy: a frequency approach

Abstract

In this study, spectral Granger causality analysis is employed to investigate the spectral dynamics of uncertainty transmission and its impact on economic growth and financial development in Saudi Arabia from 1993 to 2020. We examine the relationships between crude oil volatility, geopolitical risk, global economic uncertainty, financial development (measured by market capitalisation, MCGDP, Financial Institutions Access Index, FIAIX) and economic growth. The empirical results show that negative shocks in crude oil volatility have a significant impact on financial development as measured by the MCGDP, with capital flight and reduced domestic investment playing key roles, despite the effective contribution of increased government spending to mitigate these effects. In this study, spectral Granger causality analysis is employed to investigate the spectral dynamics of uncertainty transmission and its impact on economic growth and financial development in Saudi Arabia from 1993 to 2020. We examine the relationships between crude oil volatility, geopolitical risk, global economic uncertainty, financial development (measured by market capitalisation, MCGDP, Financial Institutions Access Index, FIAIX) and economic growth. The empirical results show that negative shocks in crude oil volatility have a significant impact on financial development as measured by MCGDP, with capital flight and lower domestic investment playing a key role, despite the effective contribution of increased government spending in mitigating these effects. However, looking at the FIAIX, crude oil volatility has a different impact, with positive shocks having no significant impact on financial development. In contrast, negative shocks show long-term causal effects, which emphasises the vulnerability of the banking sector to oil price fluctuations. Geopolitical risk has a significant long-term impact on the MCGDP due to uncertainty shocks caused by regional and global geopolitical measures. For geopolitical risk, the results are mixed with a significant causality of negative shocks on the FIAIX over certain frequencies, emphasising the sensitivity of the banking sector to geopolitical tensions. In contrast, the world economic uncertainty has only a limited direct impact on the FIAIX, indicating the resilience of the Saudi financial sector to fluctuations in global uncertainty. In terms of economic growth, positive shocks have a greater impact than negative shocks due to the volatility of crude oil. However, shocks resulting from geopolitical risks, whether positive or negative, have little effect on economic growth. The industrial production index, a measure of the resilience of the Saudi economy, indicates that it is susceptible to positive shocks from fluctuations in oil prices over a range of time periods. Our study’s findings can assist Saudi Arabia’s authorities in fortifying the country’s financial and economic resilience against the spread of uncertainty. For strong financial development and sustainable economic growth, these uncertainty elements must be well monitored.

Introduction

The question of how a stable economy can be achieved in the face of uncertainty has been extensively discussed in the literature. Several authors argue that effective monitoring of uncertainty factors is essential for achieving the desired level of economic growth and financial development (Jo, 2014, Akinsola and Odhiambo, 2020). Among these uncertainty factors, fluctuations in the price of crude oil have received much attention as they directly affect a country’s prosperity, regardless of whether it is an exporter or importer of oil. For oil-exporting countries, rising oil prices can lead to higher export revenues, which can be redirected into the national production system to support economic development (Su et al., 2017; Badeeb and Lean, 2017; Su et al., 2020; Triki et al., 2023). For oil-importing countries, an improvement in oil prices can lead to a tightening of fiscal policy and a slowdown in investment in the market sector. The economic impact of oil price volatility has been a topic of great academic interest for almost four decades. Hamilton (2003) and Charles and Darné (2017) have highlighted important policy implications of crude oil price volatility for the overall economy and stock markets. In addition, Jo (2014) has found that crude oil price volatility has the greatest ability to create short-term economic uncertainty and increase risk in the energy market. Furthermore, this risk increases energy costs, which harms national budgets and reduces investment in risky assets such as crude oil commodities.

Recently, the world had to deal with various crises, including the Gulf War, the global financial crisis, Brexit, the trade war between the USA and China, the outbreak of the COVID-19 pandemic and the Russian-Ukrainian war. As a result, the debate on uncertainty in the economy has become a pressing concern, exacerbated by the decline in industrial and manufacturing activities and the economic recession (Zhang et al., 2022). Numerous scholars have claimed that the economic and financial recession is not only due to the decline in oil prices but also to other factors such as geopolitical conflicts and economic uncertainty (Lu et al., 2020; Ahmed and Sarkodie, 2021; Tissaoui et al., 2023; Zaghdoudi et al., 2024; Tissaoui et al., 2024). Many reasons are given in the literature to support the importance of these factors. Balcilar et al. (2018) found that the geopolitical risk index (GPR) is an important determinant of investment decisions because it has the potential to change economic cycles, financial markets, and economic directions. Anser et al. (2021) further argued that geopolitical risks can either mitigate or exacerbate the impact on economic activities, energy consumption, research and development, investment, and the environment. In parallel, the literature has also focused on economic policy uncertainty (EPU). Anser et al. (2021) pointed out that the International Monetary Fund Country Report No. 16/326 (2016) indicates that EPU plays a leading role in the weakness of the economy. In addition, Ahir et al. (2022) and Demir and Danisman (2021) report that the growing concern about economic uncertainty has increased worldwide since the 2008–2009 global financial crisis and is the main cause of the decline in economic performance in many economies.

In theory, many researchers have developed a comprehensive understanding of how volatility and uncertainty shocks in the oil market can affect the economy through various economic mechanisms. High price volatility in the oil market can have a significant impact on economic sectors dependent on petroleum products, including industry, transport, and agriculture. Fluctuations in the price of oil have a direct effect on production costs as energy costs and commodity prices change. Companies often have higher operating costs when oil prices rise due to volatility, and this leads to higher prices for goods and services. The presence of cost-related inflation contributes to a decline in consumer purchasing power and can limit economic growth in general (Kilian, 2008). In addition, fluctuations in oil prices have far-reaching effects on macroeconomic stability. Key economic indicators such as trade balances, GDP growth rates, and inflation can be affected by oil price fluctuations. Rising oil prices can increase inflationary pressures, especially when the economy is already operating close to its full potential. In addition, countries that import oil may experience a slowdown in economic growth as consumer spending falls and industrial costs rise. However, lower oil prices can be beneficial for oil-importing countries as they increase consumer purchasing power and reduce industrial costs (Apergis and Miller, 2009).

As far as the impact of uncertainty shocks on the economy is concerned, increased uncertainty can have a significant impact on companies’ investment decisions and consumers’ spending behaviour. Due to the unpredictability of the economy, businesses may decide to postpone investment projects or growth, and consumers may decide to save more for retirement purposes instead of purchasing durable or discretionary items (Bloom, 2009). Another reason why companies might hesitate is the lack of confidence in upcoming government measures such as trade policy changes, regulatory changes, or tax reforms. Companies may be forced to limit the scope of their activities or make important decisions that could harm the expansion of the economy when faced with uncertainty (Baker et al., 2016). In addition, increasing uncertainty may cause investors to assess risk premiums and value their investments, which could affect the overall stability of the economy and increase market volatility. Conversely, shocks to uncertainty influence asset prices and capital flows, as well as heighten market volatility (Bloom, 2009).

The main objective of this paper is to analyse the impact of uncertainty indices, especially crude oil volatility, geopolitical risk index, and global uncertainty index, on financial development and economic growth. The Kingdom of Saudi Arabia (KSA) was selected as a suitable case study for several reasons. Firstly, the Saudi economy occupies a significant position globally, as indicated by the Saudi General Authority for Statistics reports, which show that the Saudi economy recorded an annualised growth rate of 9.9% in the first quarter of 2022. This exceptional economic performance is attributed to the Kingdom’s economic policies, which have implemented various structural measures to ensure a robust recovery in both the oil and non-oil sectors. These measures include supporting liquidity and public finances in conjunction with the rise in oil prices and increased production. Secondly, the Saudi stock market has recently experienced a strong increase in value in terms of trading volume, liquidity, and investment capital. According to the World Bank collection of development indicators (2020), a positive trend has a favourable impact on financial development, as the market capitalisation to GDP ratio has increased nearly fivefold since 2018, rising from 60% to 345% in 2020. Tissaoui et al. (2021) attribute this improvement to the opening of the Saudi market to foreign investors in 2015, the listing of the leading Saudi oil company ‘Aramco’ on the Saudi stock exchange in 2019, the selection of 30 Saudi companies for inclusion in the MSCI index in 2019, and the inclusion of the Saudi stock exchange in the FTSE Russell Index for emerging markets stock exchanges in 2020.

Accordingly, this study aims to analyse the impact of different measures of uncertainty on financial development and economic growth in Saudi Arabia. Specifically, we seek to examine the impact of positive and negative shocks from sources of uncertainty on Saudi Arabia’s economic growth and financial development in the short, medium-- and long run. The motivation behind this research is that achieving economic stability and growth in Saudi Arabia depends on effective management and monitoring of risks and negative impacts associated with sources of uncertainty, especially in an uncertain and risky future. Without an adequate assessment of the risks associated with shocks from sources of uncertainty to economic and financial fundamentals, the feasibility of economic policy may be jeopardised. In addition, uncertainty and its multiple sources make laws and processes unclear and ineffective, which can lead to economic deterioration and financial disruption. This is supported by the findings of Ali et al. (2023), who showed that uncertainty causes people and organisations to postpone their interests and investments. Ultimately, the failure to assess the risks associated with shocks from sources of uncertainty to economic and financial fundamentals can lead to an increase in the time needed to deal with the negative effects, as well as increased costs to the economy.

Furthermore, it is important to outline the mechanisms through which oil price volatility, geopolitical risks and the Global Uncertainty Index can affect Saudi Arabia’s financial development and economic growth. Firstly, oil price volatility affects Saudi Arabia’s economic growth through several channels. Saudi Arabia’s income is highly dependent on oil prices as it is an oil-exporting country. Government spending is affected by changes in oil prices, particularly in the areas of social services and infrastructure. According to the IMF study No. 16/326, low oil prices led to a slowdown in GDP growth in Saudi Arabia from mid-2014 to early 2016. The country’s GDP growth rate fell from 3.6% in 2014 to 3.5% in 2015 and finally to 1.2% in 2016. In addition, investor confidence and the general business climate are affected by the volatility of oil prices. The sector’s reliance on government financing and investment may suffer from cautious investment decisions caused by the uncertainty of oil export revenues. In addition, oil price volatility can affect consumer behaviour by impacting disposable income and the pricing of products and services, thereby influencing general consumer behaviour and economic development. Oil price volatility can also affect Saudi Arabia’s financial markets. The country’s financial system, including the stock market and banking sector, is linked to the functioning of the oil industry. High volatility can lead to fluctuations in market liquidity, investor sentiment and stock market performance. In addition, changes in oil prices can affect the profitability and risk exposure of financial institutions, which in turn affects lending and overall financial stability. Secondly, the stability of exchange rates, the balance of payments and the financial market can be affected by geopolitical threats. Geopolitical tensions and regional conflicts are two examples of these risks. In addition, increased geopolitical tensions can make investors more risk-averse, which could affect foreign direct investment and portfolio investment in the Saudi Arabian financial markets.

Furthermore, our research stands out from other studies in several respects. First, while previous studies have primarily focused on the impact of oil market-induced uncertainty on Saudi Arabia’s economic growth, our study takes a broader approach by examining the reciprocal effects between different sources of uncertainty and macroeconomic indicators such as economic growth and financial development in a frequency framework over different time periods. As far as we know, there are no previous studies on the GCC and MENA countries that have examined this issue in a frequency framework from an asymmetric perspective. Secondly, our research differs from previous studies as we not only examine the impact of uncertainty on Saudi Arabia’s economic growth, but also consider the role of financial development, which plays a crucial role in influencing economic growth. Saudi Arabia has paid great attention to the development of its financial sector, especially the banking sector, the stock market and the insurance sector, in line with the strategic requirements and plans outlined in the Kingdom’s Vision 2030. Nevertheless, this development faces many challenges, the most important of which are: the inefficiency of the financial sector in relation to the growth and sustainability of sectors of the real economy, the increasing dependence on oil, the limited dependence of GDP on non-oil products, and the uncertainty related to global geopolitical conflicts and global economic uncertainty. Finally, we attempt to fill the gap in the literature on the impact of uncertainty on the economy by considering the impact of other uncertainty indices such as the GPR and the WUI index on the Saudi economy. As far as we know, there is no research in this context that has investigated the impact of positive and negative shocks of uncertainty indices on economic growth and financial development and vice versa using frequency domain analysis. To help bridge the gap between our study and previous studies, we provide many reasons to justify the use of the two uncertainty indices. First, we use the geopolitical risk index to measure the impact of global and regional wars and political tensions on the global economy, including the Saudi economy. Geopolitical concerns are a major issue and a source of uncertainty and risk, not only for the commodity and energy markets, but also for Saudi Arabia’s economic growth and financial development. These threats have the potential to affect the growth rates of the Saudi economy and derail the development process and its programs if economic policies and strategies are not implemented in a timely and effective manner. Secondly, in addition to the geopolitical risk variable, we will assess uncertainty using the World Uncertainty Index. Previous research has proposed volatility to assess uncertainty caused by financial and economic factors (Bloom, 2009, Leahy and Whited, 1995). Others, such as Baker et al. (2016), have argued in favour of economic policy uncertainty as a measure of economic uncertainty. Bilgin et al. (2021) proposed a proxy that captures both political and economic developments at the same time, in contrast to previous measures of uncertainty. It is also important to examine the impact of the World Uncertainty Index on Saudi economic growth and financial development. The remainder of this paper is organised as follows. “Literature review” explains the literary background, “Data” presents the data and methodology, and “Empirical results” discusses the main results. The robust results are discussed in “Robustness analysis”. “Discussion” contains concluding remarks and policy implications.

Literature review

Over the past three decades, both policy makers and academics have focussed on the relationship between economic performance and uncertainty shocks. Since the seminal work of Hamilton (1983), several studies have considered uncertainty caused by fluctuations in the oil market as a significant source of disruption to financial development and growth (Sreenu (2018)). However, Yıldırım et al. (2020) have shown that an increase in oil revenues has no impact on financial development in the short term, but has a positive effect in the long term. Canh and Thanh (2020) found a close and positive reciprocal relationship between the increase in oil revenues and the influence of the depth of the financial market. Indeed, a well-functioning financial system is necessary to optimally utilise the wealth of oil resources (Pradhan et al., 2016) and to provide credit to the private sector to generate economic activities (Asif et al. (2020)). When studying the Chinese market, Jiang et al. (2019) found that fluctuating oil prices have the greatest impact on financial sector output. In addition, several authors have found that the volatility of crude oil futures contracts has a significant impact on global economic and financial stability (Cheong, 2009; Lin et al., 2014; Wang et al., 2016; Gong and Lin, 2018; Wen et al., 2018). Studies on major oil-importing countries have shown that oil price volatility has a profound impact on the economy and the country needs to improve its oil futures market to ensure stable growth (Wang et al., 2012) and avoid imposing additional taxes to curb the negative impact on growth (Alkahteeb and Sultan, 2019).

Regarding the relationship between oil price volatility and economic growth, many previous studies have produced controversial results on the impact of oil prices on economic growth. They differentiate the effects depending on whether the country is an oil importer or exporter. For example, oil price increases have a positive effect on exporting countries and a negative effect on importing countries (Köse and Baimaganbetov, 2015). Maghyereh et al. (2019) reported that negative oil price shocks lead to higher economic growth, while positive shocks lead to lower economic growth. However, some studies show the importance of a negative relationship between the increase in oil prices and GDP growth. For example, Bruno and Sachs (1982) found that high oil prices had a significant effect on economic growth in the UK, especially in the industrial sector, and Hamilton (1983), who studied the US between 1949 and 1972, confirms that this effect is negative in most oil-importing economies, such as Japan (Zhang, 2008). In addition, many empirical studies have also investigated the impact of oil price fluctuations on GDP growth in developing countries compared to industrialised countries. They find a significant long-term effect for developing countries. Bacon (2006) also used a sample of 131 countries and showed that the decline in GDP growth in industrialised countries with a per capita income of more than 900 dollars would only be 0.4%, while the same effect would be 4 in developing countries with a per capita GDP of less than 300 dollars. More recently, Elhassan (2021) conducted a study on the impact of oil prices on economic growth in the KSA between 1970 and 2020. He categorised oil prices into positive and negative shocks. He found that the effects of positive shocks on growth were more significant in the short and long run than the effects of negative shocks, which were only significant in the long run. Recently, Sun et al. (2022) used micro-panel data of unlisted Chinese companies from 2010 to 2018 to investigate the relationship between oil price uncertainty and corporate credit risk. The results indicated that corporate credit risks are exacerbated by rising oil price volatility. The mediating and moderating effects showed that these risks are indirectly amplified by rising oil prices via other macroeconomic variables.

Regarding the links between the GPR and financial development, many authors have tried to answer the following questions: Does the increase in geopolitical risk, which inhibits investment and consumption, affect a country’s financial development? Would the level of financial development in turn reduce the impact of geopolitical risk? In this sense, Su et al. (2019) found that GPR has a significant impact on oil prices and then on short-term financial liquidity in KSA, which could affect the domestic market and its growth. Furthermore, Liu et al. (2019) distinguished between the GPR and its severe form of GPRS. They found that the severe form has a greater impact on oil prices and thus on financial development. As a result, to our knowledge, only one empirical study has investigated the direct relationship between GPR and financial development. The study conducted by Lu et al. (2020) examined 18 emerging markets between 1985 and 2018 and found that an increase in geopolitical risk leads to a decrease in the level of financial development as measured by the level of domestic credit. As a result, no study has examined the nature of this relationship in a country that is a major exporter of energy resources and whose GPR could increase its oil revenues and channel them into financial and then productive cycles.

Regarding the relationship between the GPR and economic growth, various studies have shown that geopolitical instability and uncertainty significantly affect economic growth (Zhang et al., 2024). Similarly, this previous research has shown that the GPR reflects the anxiety and loss of confidence of economic agents. Investors review their decisions, households review their consumption. Banks review their loan applications, financial markets panic, etc. All empirical studies agree that the rise in the GPR slows economic growth (Rigotti and Shannon, 2005; Brandt and Gao, 2019). Murdoch and Sandler (2002) empirically found that civil wars in Africa between 1960 and 1995 had a negative impact on per capita GDP growth at the national level and even in neighbouring countries. Gaibulloev and Sandler (2008), who analysed 18 European countries between 1971 and 2004, found that terrorism, especially transnational terrorism, reduces growth. Soybilgen et al. (2019), who analysed 18 emerging economies between 1986 and 2016, concluded that the impact of geopolitical risks on growth rates is negative and significant. An increase in the GPR by 10 points leads to a decline in the GDP growth rate of 0.2 to 0.4. Lanouar and Goaied (2019) also concluded that the GPR has a negative impact on exports, imports, stock and financial markets and the overall economy. Finally, Tiwari et al. (2019), while analysing its sectoral impact, found that the GPR in India adversely affects long-term tourism demand. More recently, Sharif et al. (2020) found that Covid-19 in the United States caused more damage to the GPR than to the stock market. More specifically, Adedoyin et al. (2020) examined the impact of GPR on export-led growth in Malaysia. They found that it had both positive and negative impacts, but was insignificant in the short and long term. In their study of MENA countries, Soltani et al. (2021) found that higher GPR leads to greater economic vulnerability and hinders the development and growth of certain economies.

There are fewer studies that have analysed the relationship between the WUI and economic growth. For example, Kang et al. (2020) associate global uncertainty shocks with a sharp decline in industrial production growth and inflation, with a maximum decline after 6 months, and a sharp decline in global interest rates, including the decline that would occur after a maximum of 16 months. They find that the production of China, the United States and the United Kingdom react moderately to these shocks, in contrast to the trio Russia-Brazil-South Africa, which are the most affected. Karadağ (2021) attempted to examine the effect of uncertainty in developing countries on foreign direct investment in real estate using the Gengenbach, Urbain, and Westerlund panel cointegration test and the PDOLS coefficient estimation method. It showed a positive impact of the uncertainty index on FDI in period 2008–2018. More recently, Liu and Gao (2022) examined the impact of the WUI on the US GDP growth rate. They found a stronger predictive power of the WUI for the GDP growth rate. However, most empirical studies have examined the relationship between global uncertainty and financial markets, but not the impact of global uncertainty on the financial development of economies (Chowdhury et al., 2021).

For the Saudi context, the existing literature on the impact of uncertainty on macroeconomic and financial variables is relatively scarce. A few studies have been conducted for the GCC and MENA countries. For example, Istiak and Alam (2019) examined the impact of uncertainty arising from economic uncertainty in the US on the financial markets of GCC countries. By using a linear and non-linear structural VAR model as well as an impulse response-based test, the authors found that economic policy uncertainty in the US led to a decline in the stock market index of all GCC countries. Alqahtani and Klein (2021) investigated the impact of uncertainty on the financial markets of the GCC countries. They used an ARDL model for data from May 2007 to August 2018 and found evidence of a significant impact of uncertainty on financial markets. In the case of MENA countries, Ayad et al. (2023) examined the impact of economic policy uncertainty on carbon dioxide (CO₂) emissions. The authors applied the ARDL model and its extensions to data covering the period 1970–2020. The empirical results suggest that uncertainty stemming from US economic policy played a crucial role in environmental degradation, especially in Morocco, Turkey and Iran.

Data

Data description

The data used in this study are annual values covering the period from 1993 to 2020. The data set includes five variables (Table 1), two of which relate to the Saudi economy: financial development and economic growth. The remaining three variables measure international uncertainty, specifically crude oil volatility (VTOIL), the Geopolitical Risk Index (GPR) and the World Uncertainty Index (WUI). Specifically, financial performance is represented by two measures. The first is the ratio of market capitalisation to GDP (MCGDP). The second is the index for access to financial institutions (FIAIX). Economic growth (EG), on the other hand, is measured by the annual percentage growth rate of GDP at market prices based on a constant national currency. The GPR and WUI indices are derived from. The GPR index represents the total number of newspaper articles reporting on geopolitical events and tensions, while the WUI index is compiled using the total number of words in each report for 143 countries from 1996 onwards, based on the frequency of the term “uncertainty” and its variants in the Economist Intelligence Unit’s (EIU) quarterly country reports. In addition, data on crude oil prices, represented by Brent crude oil prices, were collected from the International Energy Agency. Using this crude oil data, we estimate the following TGARCH (1.1) model proposed by Zakoian (1994) to measure crude oil volatility (VTOIL):

$${h}_{t}=\omega +{\alpha }_{1}{\varepsilon }_{t-1}+{\alpha }_{2}{{\rm{\sigma }}}_{t-1}{+\alpha }_{3}{\varepsilon }_{t-1}{I}_{t-1}^{-}$$

(1)

Where ${h}_{t}$ is the conditional standard deviation of the Brent index at time t. ${\alpha }_{1}$ and ${\alpha }_{2}$ represent the previous error term and the previous conditional variances at time t, respectively. ${\alpha }_{3}$ is the parameter for the leverage effect. There are several reasons why this period, which ranges from 1993 to 2020, was chosen. First, the variables geopolitical risk index and crude oil volatility are available for economic growth over a longer period. However, the stock market capitalisation to GDP ratio (MCGDP), which monitors financial development, is not accessible in 2021 or 2022, nor in the years prior to 1993. The indices for access to financial institutions are also available until 2021. Secondly, the World Uncertainty Index was not available until 1992. Thirdly, the year 2020 was mentioned, which is assumed to be the year in which the coronavirus outbreak will spread. For these reasons, the research period was set from 1993 to 2020.

Table 1 Description of variables.

Data analysis

Table 2 shows the summarised statistics of all variables considered in our study. Economic growth ranges from −0.041 to 0.112 with a mean and standard deviation of 0.025 and 0.038, respectively. The data show low growth rates with a mean of 0.025 and a standard deviation of 0.038. Global factors such as economic policies, technological developments and external shocks such as epidemics and financial crises have an impact on economic cycles, which are reflected, for example, over time.

Table 2 Summary statistics.

The ratio of market capitalisation to GDP is between 0.276 and 3.466 with an average and a standard deviation of 0.801 and 0.765 respectively. It shows the size of the stock market in relation to the size of the economy. The average market capitalisation to GDP ratio of 0.801 shows that changes in market capitalisation can be influenced by the position of the investor, prevailing economic conditions and regulatory changes. The market has experienced periods of exuberance and restraint during the highs and lows of the MCGDP.

In addition, the access indices of financial institutions measure the accessibility and efficiency of the banking sector. With a mean value of 0.3125, the indices range between 0.2149 and 0.4254, reflecting improvements in the accessibility and operational efficiency of the financial sector over time. Higher scores indicate improved access to financial services, while lower scores may reflect regulatory barriers or an economic downturn affecting access to banking services.

In terms of uncertainty measures, we see that the average and standard deviation of crude oil changes are 133.375 and 58.826, respectively, ranging from 44.011 to 258.491. This shows how the market is influenced by shifts in supply and demand as well as geopolitical developments. The price fluctuations in the global oil markets are shown with a standard deviation of 58.826 and an average volatility of 133.37. The peaks in volatility are shown with a standard deviation of 58.826. The spikes in volatility are associated with production disruptions, geopolitical conflicts and the unpredictability of the global economy affecting oil prices.

In addition, the GPR tracks how news outlets report on global tensions and events that affect investor confidence. The index, with an average score of 1.952 and a range of 1.643 to 2.373, indicates when there are more geopolitical risks that could affect global markets. Lower scores indicate stability in the GPR, while higher scores indicate increased uncertainty and potential market volatility.

In the end, the standard deviation was 10623.320 and the average indicator was 18441.130, indicating different levels of uncertainty. The correlation between rising WUI index values and significant financial crises and geopolitical uncertainty affecting foreign investment and business confidence explains this.

Table 3 shows the correlation coefficients between the variables used in our study. As seen, there is a low correlation between all pairs of variables, with the exception of a notable correlation between FIAIX and WUI. Specifically, we find that the correlation between financial development and economic growth is low and negative with a value of −0.23. Furthermore, the correlations between financial development, crude oil volatility and GPR are lower and negative. Furthermore, there is a positive and weaker correlation between financial development and the WUI when using the MCGDP. However, when using FIAIX as a measure of financial development, the correlation with WUI becomes significant (0.65). The volatility of these two variables with the terms of trade for oil is also low. When analysing the unit root, the Phillips-Perron unit root values reported in Table 4 show that all series in our sample are stationary at the level or first difference.

Table 3 Correlation matrix.

Table 4 Phillips-Perron Unit root test.

Methodology

From the perspective of Wiener (1956) who initiated the notion of Granger causality, Granger (1969) developed this approach based on linear modelling. Despite being one of the most used conventional time-domain causality tests, this linear process remains insufficient because it examines the interaction among the relevant variables in a statistical manner. This drawback motivates Breitung and Candelon (2006) to mention that the Granger causality can dynamically investigate the causality between variables. They develop a Granger causality test in the frequency domain to consider the spillover dynamics over different frequencies. In addition, Bozoklu and Yilanci (2013) and Tissaoui et al. (2024) indicated that the causality can be estimated at all points in the frequency distribution. This is consistent with Zaghdoudi et al. (2023) which indicates that the spectral and frequential analysis is robust to consider non-normality and seasonality in data. Thus, the principle of this spectral tool consists of decomposing the causality between two variables y and x at different frequencies to refer to the short-, medium-, and long-terms. According to Breitung and Candelon (2006), Y and X represent two-time series. Let ${V}_{t}$ = (${Y}_{t}$, ${{X}}_{t}$) be a two-dimensional vector of series achieved at t = 1….., T with a finite-order VAR-derived framework. It is illustrated as follows:

$${\theta (L)Z}_{t}={\varepsilon }_{t}$$

(2)

The model (1) can be written as follows:

$${Y}_{t}={\theta }_{11,1}{Y}_{t-1}+{\theta }_{11,2}{Y}_{t-2}+\ldots ..+{\theta }_{11,p}{Y}_{t-p}+{\theta }_{12,1}{X}_{t-1}+{\theta }_{12,2}{X}_{t-2}+,\ldots ..,+{\theta }_{12,p}{X}_{t-p}$$

$${X}_{t}={\theta }_{21,1}{X}_{t-1}+{\theta }_{21,2}{X}_{t-2}+\ldots ..+{\theta }_{21,p}{X}_{t-p}+{\theta }_{22,1}{Y}_{t-1}+{\theta }_{22,2}{Y}_{t-2}+,\ldots ..,+{\theta }_{22,p}{Y}_{t-p}$$

(3)

Also, the model (2) can be plotted in matrix representation using the lag operator (L) as follows:

$$\theta \left(L\right)\left(\begin{array}{c}{Y}_{t}\\ {X}_{t}\end{array}\right)=\left(\begin{array}{cc}{\theta }_{11} & {\theta }_{12}\\ {\theta }_{21} & {\theta }_{22}\end{array}\right)\left(\begin{array}{c}{Y}_{t}\\ {X}_{t}\end{array}\right)={\varepsilon }_{t}$$

(4)

Where $\theta (L)=1-{\theta }_{1}L-{\theta }_{2}{L}^{2}-\ldots -{\theta }_{p}{L}^{p}$ is a 2 × 2 lag polynomial, and ${\theta }_{1}-{\theta }_{2}-\ldots -{\theta }_{p}$ are 2 × 2 autoregressive coefficients matrices. The error vector ${\varepsilon }_{t}$ represents a white noise with zero mean and covariance matrix $E\left({\varepsilon }_{t},{{\varepsilon }^{,}}_{t}\right)=\sum$ where ∑ is assumed positive and symmetric. Due to these characteristics, Breitung and Candelon (2006) applied the Cholesky decomposition method to decompose the matrix ∑ as ${G}^{,}{G}={\sum }^{-1}$ where G and G’ are respectively the lower and upper triangular matrices. Thus, the moving average representation of the system is expressed as follows:

$$\left(\begin{array}{c}{Y}_{t}\\ {X}_{t}\end{array}\right)={\psi (L)\eta }_{t}=\left(\begin{array}{cc}{\psi }_{11}(L) & {\psi }_{12}(L)\\ {\psi }_{21}(L) & {\psi }_{22}(L)\end{array}\right)\left(\begin{array}{c}{\eta }_{1t}\\ {\eta }_{2t}\end{array}\right)$$

(5)

Where $\psi (L)={\theta (L)}^{-1}{G}^{-1}$. The predictive power of ${X}_{t}$ can be calculated by comparing the predictive component of the spectrum with the intrinsic component at each frequency. Referring to Geweke (1982), the measure of causality (CM) is plotted as follows:

$${{CM}}_{X\to Y\left(\omega \right)}=log \left[1+\frac{{\left|{\psi }_{12}({e}^{-i\omega })\right|}^{2}}{{\left|{\psi }_{11}({e}^{-i\omega })\right|}^{2}}\right]$$

(6)

If ${\left|{\psi }_{12}({e}^{-i\omega })\right|}^{2}=0$, $X$ does not cause $Y$ at frequency$\left(\omega \right)$. Therefore, we test H(0) which indicates no causality at frequency $\omega$ by referring to the standard F test. The frequency (ω) is equal to 2Π/cycle duration (T); for ω ∈ (0, π), where T and p represent the number of observations of a series and optimal lag order of the VAR model, respectively. Gomez-Gonzalez et al. (2015) indicated that if the values of (ω) are near zero, this is a synonym for long-term causality. However, if the values of (ω) are near to Π, this indicates a short-term causality between variables.

Empirical results

Sensitivity analysis

Before analysing the frequency domain Granger causality results, it is important to make a sensitivity analysis to check the adequacy of VAR models and to assess the impact of data variations on the analysis. To achieve this, some tests were conducted like residual autocorrelation, residual normality, and residual heteroscedasticity tests. Table 5 shows that the residual Portmanteau Tests are non-significant. This indicates that the autocorrelation of the residuals is non-evident and hence the goodness of fit of all VAR models used to make the frequency domain Granger causality results. As well, another hypothesis underlying the VAR model has been also tested. Table 5 provides evidence of a rejection of residual heteroscedasticity. All values of probability of LM ARCH tests are higher than 10%. This means that the residual variance of the regression VAR model does not change extensively. In addition to this, the normality of residuals for all VAR models was tested using the Jarque-Bera test. After performing this test, we found that the resulting P value is larger than 0.1, indicating that the residuals are normally distributed. Overall, the sensitivity tests demonstrated the goodness of fit of all VAR models utilised to generate the frequency domain Granger causality results.

Table 5 Sensitivity tests (annual analysis).

Empirical analysis

A first analysis of the different causalities consists of analysing the role of the uncertainty indices for economic growth and financial development from the perspective of the frequency domain. In all figures, the x-axis shows the frequency (ω), which is equal to 2Π/cycle length (T), while the probability is plotted on the y-axis. When interpreting the results, it is assumed that the short-term phase is less than two years, the medium-term phase is between two and five years and the long-term phase is five years or more (Prats and Sandoval, 2020). Figure 1 shows the graphical results of the spectral causality between crude oil volatility, GPR index, WUI index, economic growth and financial development. For financial development, as measured by the MCGDP, the causality between crude oil volatility and financial development (Fig. 1a) is significant overall with varying frequencies (from ω = 0.23 to ω = 2.79). This indicates a permanent dynamic spillover causality from the short-term to the long-term horizon, which leads from the uncertainty of the oil market to the financial development of the KSA. This is explained by the fact that excessive volatility in the oil market affects global policies in the Saudi financial market, especially those affecting oil-exporting countries. These countries, especially Saudi Arabia, will also have limited alternatives for financial investments due to the widespread uncertainty in the oil market. In this context, Saudi Arabia should take reflective measures that benefit financial development. However, our results show that the causality between financial development and crude oil volatility is only significant in the short term. We find that this effect is significant over frequencies from ω = 2.32 to ω = 2.79. Figure 1b also shows that there is no significant causal relationship in either direction between financial development and geopolitical risk. For the relationship between financial development and WUI (Fig. 1c), we find a non-significant causality from WUI to financial development. Conversely, we find evidence of long-term causality running from financial development to WUI across frequencies (from ω = 0.22 to ω = 1.12).

Fig. 1: Causality between the level of uncertainty indexes and the economic growth and financial development (Annual frequency).

a Oil return volatility and financial development (Measure: MCGDP). b GPR and Financial development (Measure: MCGDP). c WUI and Financial development (Measure: MCGDP). d Oil return volatility and Economic growth. e GPR and Economic growth. f WUI and Economic growth. g Oil return volatility and financial development(Measure: FIAIX). h GPR and Financial development (Measure: FIAIX). i WUI and Financial development (Measure: FIAIX). Each panel shows the relationship between different uncertainty indicators (oil return volatility, GPR, WUI) and either financial development or economic growth, with MCGDP and FIAIX as financial development measures.

Different results were found with regard to financial development, which is measured using the Financial Institutions Access Index (FIAIX). Figure 1g shows the relationship between oil volatility and the FIAIX. As observed, there is no reciprocal effect between the two variables. Comparing these results with those obtained using the MCGDP as a measure of financial development, it can be seen that financial development driven by the financial market is more sensitive to the uncertainty generated by the oil market than that generated by the banking sector. The contagion effect from the financial sector to the oil market and vice versa can explain this (Tissaoui and Azibi, 2019). As for the correlation between the geopolitical risk (GPR) and the FIAIX, Fig. 1h shows that the GPR significantly affects the FIAIX at all frequencies. However, no significant causality is observed in the opposite direction. This indicates that geopolitical risk is a significant risk factor for the performance of the Saudi banking sector. This result differs from the one obtained when analysing the MCGDP. Consequently, financial development driven by the banking sector is more sensitive to uncertainty caused by geopolitical risk than that driven by the financial market. This shows how vulnerable the Saudi banking sector is to geopolitical tensions and uncertainty. Geopolitical concerns are significantly affecting the FIAIX across all frequencies. This affects investor outlook, accessibility of financial services and regulatory stability, all factors that further deteriorate the overall health of the banking sector. When analysing the relationship between WUI and FIAIX (Fig. 1i), the results are similar to those of the MCGDP analysis. We find no significant causality between WUI and FIAIX. However, we find evidence of long-term causality between FIAIX and WUI with a frequency of ω = 0.22. The significant impact of geopolitical tensions and uncertainty on FIAIX across all frequencies shows how much they affect the Saudi banking sector. This has a negative impact on investor confidence, availability of financial services and regulatory stability. Ultimately, these factors affect the overall health of the banking sector.

Regarding the spectral causality between economic growth and uncertainty indices, Fig. 1d shows the relationship between economic growth and crude oil volatility. We show that there is no significant causality between crude oil volatility and economic growth and vice versa. Figure 1e shows that the causality from economic growth to GPR is significant at a frequency of ω = 0.22 to ω = 0.44. However, there is no significant causality in the opposite direction. For the relationship between economic growth and WUI, there is no significant causality between the two variables in either direction (Fig. 1f). As can be seen from the above results, the Granger causality test in the frequency domain shows that some causalities are significant while others are not.

Further research is needed to complete this analysis and gain a deeper understanding of the relationship between uncertainty and macroeconomic variables. We proceed by testing the effects of positive and negative shocks to the uncertainty indices on economic growth and financial development and vice versa. Figure 2a shows that there is no significant causality between the positive shock of crude oil uncertainty and financial development (MCGDP) in the KSA context. If we use FIAIX as a measure of financial development, comparable results emerge (Fig. 2g). Moreover, neither MCGDP nor FIAIX benefits significantly from the volatility of crude oil. The non-significant impact of positive shocks in crude oil price uncertainty on FIAIX and MCGDP suggests that Saudi Arabia’s financial industry and economy are somewhat resilient to temporary changes in oil prices. This resilience could be due to deliberate attempts at economic diversification and successful legislative initiatives to keep the banking sector stable in the face of unstable oil prices. However, Fig. 2 shows that the content information of financial development (MCGDP) has a significant long-run impact on the positive shock of crude oil volatility over a range of frequencies (from ω = 0.96 to ω = 1.44). The results also show that the financial development measured by the FIAIX has no impact on the uncertainty caused by the crude oil market (Fig. 2g).

Fig. 2: Causality between the positive shock of uncertainty indexes and the economic growth and financial development (Annual frequency).

a Positive crude oil volatility vs financial development (Measure: MCGDP). b Positive GPR vs financial development (Measure: MCGDP). c Positive WUI vs financial development (Measure: MCGDP). d Positive crude oil volatility vs economic growth. e Positive GPR vs economic growth. f Positive WUI vs economic growth. g Positive crude oil volatility vs financial development (Measure: FIAIX). h Positive GPR vs financial development (Measure: FIAIX). i Positive WUI vs financial development (Measure: FIAIX). Each panel presents the impact of positive shocks in uncertainty indicators (crude oil volatility, GPR, WUI) on either financial development or economic growth. MCGDP and FIAIX are used as measures of financial development.

In Fig. 3a, we found a significant long-term effect of a negative shock to crude oil volatility on financial performance over all frequencies, using MCGDP as a measure of financial performance. This shows how vulnerable the financial market is to oil price fluctuations and points to potential performance issues in the financial sector during periods of unstable oil-related economic conditions. It was found that there was no obvious impact in the other direction, suggesting that the volatility of crude oil was not much affected by the changes in financial development caused by the Saudi financial market. Due to this one-way interaction, the economic conditions caused by the fluctuations in oil prices would affect the financial market dynamics more than the other way round.

Fig. 3: Causality between the negative shock of uncertainty indexes and the economic growth and the industrial production index (IP) (Annual frequency).

a Negative crude oil volatility vs financial development (Measure: MCGDP). b Negative GPR vs financial development (Measure: MCGDP). c Negative WUI vs financial development (Measure: MCGDP). d Negative crude oil volatility vs economic growth. e Negative GPR vs economic growth. f Negative WUI vs economic growth. g Negative crude oil volatility vs financial development (Measure: FIAIX). h Negative GPR vs financial development (Measure: FIAIX). i Negative WUI vs financial development (Measure: FIAIX). Each panel illustrates the effect of negative shocks in uncertainty indicators (crude oil volatility, GPR, WUI) on financial development or economic growth. MCGDP and FIAIX serve as financial development measures, and the Industrial Production Index (IP) is included to show industrial impacts.

Moreover, no reciprocal effect between the negative shock of crude oil volatility and financial development (FIAIX) was demonstrated (Fig. 3g). Since they are mutually exclusive, fluctuations in crude oil prices do not necessarily lead to proportional adjustments in the financial growth of the banking sector. This means that Saudi Arabia’s financial sector may be able to withstand unexpected changes in oil prices. Moreover, given the oil dependence of the Saudi Arabian economy, the lack of strong evidence of a link between oil price changes and banking sector growth emphasises the need for economic diversification initiatives. By encouraging the development of non-oil industries and improving macroeconomic stability, diversification can reduce the economy’s vulnerability to oil price fluctuations. We now focus on the relationship between the positive and negative shocks of GPR and financial development. Figure 2b shows that the positive shock of GPR had a remarkable impact on financial development (MCGDP) at different frequencies (from ω = 0.23 to ω = 1.86) and (from ω = 2.55 to ω = 2.79). This suggests that geopolitical tensions influence both the short and long-term movements of financial markets in the presence of positive shocks.

Similar results have been demonstrated in the measurement of financial development by the FIAIX. Figure 2h shows that the positive shock of the GPR exerts a significant impact on the FIAIX from frequencies ω = 0.46 to frequencies ω = 1.86. This is synonymous with the long-term impact of geopolitical tensions on the financial performance of the banking sector. This shows the long-term effects of geopolitical tensions on the growth of the banking sector, implying that the accessibility and expansion of the financial sector can be helped or hindered by the uncertainty caused by geopolitical concerns. Increasing geopolitical tensions generally have the potential to disrupt the banking sector and financial markets, which could lead to poorer access to financial services, higher borrowing costs and slower economic growth. Saudi Arabian policymakers need to consider better ways to mitigate the effects of geopolitical threats on the country’s financial development. This includes developing initiatives to promote stability, expanding the economy to reduce dependence on oil revenues and establishing strong risk management systems in the banking sector.

However, Fig. 3b shows a significant causality from negative GPR shocks and financial development (MCGDP) across frequencies from ω = 0.23 to ω = 1.62, suggesting a persistent and durable spillover effect where unfavourable geopolitical events negatively affect financial market development in the long term. These effects could include lower investor confidence, increased risk aversion and economic uncertainty. Conversely, Fig. 3h shows that there is no obvious causal relationship between negative GPR shocks and financial performance as measured by the FIAIX. This implies that while unfavourable geopolitical developments might affect the dynamics of the financial system, they have little impact on the accessibility and growth of the Saudi Arabian banking sector. As for the reverse direction, Fig. 3b shows that there is no significant causality from financial development (MCGDP) and negative GPR across all frequencies. This means that improvements or setbacks in financial market development have no recognisable influence on the occurrence or intensity of negative geopolitical shocks. However, Fig. 3h shows that the FIAIX influences the negative GPR in the long term, especially at frequencies from ω = 1.39 to ω = 2.32, indicating that a well-developed banking sector, as measured by the FIAIX, can help mitigate negative geopolitical shocks by strengthening Saudi Arabia’s economic resilience and stability. As the FIAIX shows, improving the accessibility and resilience of the financial sector is critical to protecting the economy from damaging geopolitical shocks. Policies aimed at improving financial sector infrastructure, expanding the availability of financial services and fostering a competitive banking environment can help promote economic stability and growth even in the face of external uncertainties.

Figure 2c shows a non-significant causal relationship between the positive WUI index and the MCGDP in both directions for the co-movement between the two variables, i.e. financial development as assessed by the MCGDP and the positive WUI index. This means that neither the expansion of the Saudi Arabian financial industry nor favourable changes in the level of global uncertainty strongly impact each other. Similarly, Fig. 2i shows that a positive WUI index has no obvious impact on the FIAIX. However, there is evidence that the FIAIX could affect the positive WUI index, especially at frequencies in the range of ω = 0.23 to ω = 0.69, implying that encouraging the expansion and accessibility of the banking sector could help Saudi Arabia mitigate the impact of global unpredictability.

In contrast to the positive index, Fig. 3c shows a significant permanent causality from the negative WUI index and the MCGDP over frequencies from ω = 0.23 to ω = 1.86. This indicates that negative global uncertainty has a long-term negative impact on the development of the financial markets in Saudi Arabia. Figure 3i, on the other hand, shows no obvious causal relationship between the negative WUI index and the FIAIX, suggesting that a decline in global uncertainty is not a direct factor in the growth of the Saudi Arabian banking sector. However, there is a strong causal relationship between the negative WUI index and the FIAIX, especially at frequencies ω = 0.24 to ω = 0.72, suggesting that a strong banking sector could help mitigate the adverse effects of global uncertainty.

As for the relationship between economic growth and the positive and negative shocks of the uncertainty indices, Fig. 2d shows that the causality from positive crude oil volatility to economic growth is evident both in the long run (from ω = 0.24 to ω = 0.96) and in the short run (from ω = 2.65 to ω = 2.89). This finding shows that the causal relationship was temporary in the short term, then no longer exists in the medium run and is significant in the long run. Conversely, we find evidence of non-significant causality from EG and the positive shock to crude oil volatility at all frequencies. For the causality between the negative shock to crude oil volatility and EG, Fig. 3d shows no significance in either direction. This result is in contrast to Maghyereh et al. (2019), who showed that negative oil price shocks lead to higher economic growth.

Furthermore, our results also show no significant effect of a positive shock to the GPR on EG (Fig. 2e). However, our results show that the causality from EG and the positive shock of the GPR is evident both in the short run (ω = 2.79) and in the long run (from ω = 0.46 to ω = 0.69). As for the causal dynamic relationship from EG to the negative shock of GPR and vice versa (Fig. 3e), no significant causality is demonstrated for either direction. Similar results are also found for the relationship between the positive shock of the WUI index and EG (Fig. 2f). However, Fig. 3f shows a non-significant causality from the negative shock of the WUI index and EG at all frequencies. So far, the causality of EG to the negative shock of the WUI index is evident both in the long term (from ω = 0.23 to ω = 1.39) and in the short term (ω = 2.79). This finding shows that the causal relationship was temporary in the short term, then no longer exists in the medium term and is significant in the long term.

Robustness analysis

This section is entirely devoted to assessing the robustness of our primary results. Both contributions are considered. First, we use the index of industrial production (IPI) instead of GDP as an indicator of the level of real economic activity. According to Nwaolisa and Chijindu (2016), industrial production increases in times of economic expansion and decreases in times of recession; therefore, a change in industrial production would indicate a change in the economy. Second, we ensure the consistency and reliability of our conclusions by conducting a high-frequency study with quarterly data. This also allows for better time series analysis and a more robust Granger inference. Moreover, we use quarterly data because we avoid the drawbacks of using annual data, which are mainly that these data do not provide many degrees of freedom for statistical estimation; moreover, annual data hide important short-term information and shifts in economic activity that could have long-term effects. Accordingly, we re-examine the causality between the uncertainty indices and the index of industrial production by applying the Granger causality test in the frequency domain to quarterly data spanning from 2015Q1 to 2022Q4 with 32 observations. The robustness analysis revealed important findings. Firstly, the sensitivity analysis confirms the robustness and suitability of the Granger causality test in the frequency domain, which is used to check the validity of VAR models and the influence of data changes on the analysis. For this purpose, certain tests were performed, such as residual autocorrelation, residual normality and residual heteroscedasticity tests. Table 6 shows that the residual Portmanteau tests are not significant. This indicates that the autocorrelation of the residuals is not evident and thus the goodness of fit of all VAR models used to generate the Granger causality results in the frequency domain. In addition, another hypothesis driving the VAR model was investigated. Table 6 supports the rejection of residual heteroscedasticity. All probability values for LM ARCH tests are greater than 10%, except for WUI _POS versus IPI, where the test value is significant. This implies that the residual variance of the regression VAR model does not vary much. In addition, the Jarque-Bera test was used to check the normality of the residuals for each VAR model. If the results of this test are above 0.1, Table 6 confirms that the residuals are regularly distributed. All VAR models used to generate the Granger causality results in the frequency domain showed satisfactory overall agreement according to the sensitivity test results. This supports the sensitivity results when we consider annual data for the analysis.

Table 6 Sensitivity tests (Quarterly analysis).

Second, the results of the spectral causality between crude oil volatility, GPR index, WUI index and IPI are shown graphically in Fig. 4. Figure 4a shows that the volatility of crude oil (VTOIL) at frequency (ω = 0.19) has a significant impact on industrial production (IPI). This suggests a discontinuous spillover causal relationship that spills over from oil market instability to KSA economic activity in the long run. However, our results show that the causal relationship between industrial production and crude oil volatility is equally strong in the long run. We show a dynamic spillover effect between the frequency ω = 0.98 and ω = 1.37. Figure 4b also shows that geopolitical risk has no significant causal impact on the IPI at all frequencies. Nevertheless, the IPI has a dynamic influence on geopolitical risk at frequencies between ω = 0.22 and ω = 1.12. In the joint movement of IPI and WUI (Fig. 4c), we see no significant causality between WUI and IPI. In contrast, we find evidence of long-term causation from IPI to WUI in the frequencies (from ω = 0.19 to ω = 0.98). As can be seen from the above data, the Granger causality test in the frequency domain shows that certain causalities are significant while others are not. Further research is required to complete this analysis and gain a better understanding of the relationship between uncertainty and industrial productivity.

Fig. 4: Causality between the uncertainty indexes and the Industrial production index (IPI) (Quarterly frequency).

a Oil return volatility and IPI. b GPR and IPI. c WUI and IPI. Each panel demonstrates the influence of a specific uncertainty indicator—oil return volatility, GPR, and WUI—on industrial production, as measured by the IPI.

We analyse the effects of positive and negative shocks on the uncertainty indices of industrial production and vice versa. Figure 5a shows that the positive uncertainty shock from crude oil has a significant causal effect on the IPI with different frequencies (ω = 0.20 to ω = 0.60), (ω = 1.01 to ω = 1.41) and (ω = 2.02 to ω = 2.83). This indicates that the transmission of uncertainty from the oil market to Saudi economic activity is persistent in the short, medium and long term. However, this figure shows that the information content of the IPI has a long-run impact on the positive shock to crude oil volatility over a range of frequencies (from ω = 0.20 to ω = 1.01). In Fig. 5b, we see no significant causality between the positive shock to the GPR and the IPI in either direction. Similar results were also demonstrated for the relationship between the positive shock of the WUI and the IPI. Figure 5 shows that the spillover effect between two variables is not significant at all frequencies. Now we look at how the IPI interacts with negative shocks to the uncertainty indices. Figure 6a shows a long-term causal relationship between negative crude oil volatility and IPI at frequencies ω = 0.20 to ω = 0.40 and ω = 1.01 to ω = 1.41. Similar results are observed when examining the impact of IPI on negative crude oil volatility at the same frequency. This indicates that two variables have a long-term reciprocal influence. Regarding the causal dynamic effect of IPI on the negative shock of GPR and vice versa (Fig. 6b), no significant causal effect of the negative shock of GPR on IPI was found. Nevertheless, the long-term causal effect of IPI on the negative shock of GPR is demonstrated over the frequencies from ω = 0.20 to ω = 0.60. Finally, Fig. 6c shows that the negative shock of the WUI has a significant long-term effect on the IPI across frequencies (ω = 1.01 to ω = 2.02). However, in the long term (from ω = 0.60 to ω = 0.81), the causal impact of the IPI on the negative shock of the WUI is evident.

Fig. 5: Causality between the positive shock of uncertainty indexes and the Industrial production index (Quarterly frequency).

a Positive crude oil volatility vs IPI. b Positive GPR vs IPI. c Positive WUI vs IPI. Each panel shows the influence of positive shocks in uncertainty indicators—crude oil volatility, GPR, and WUI—on industrial production, as measured by the IPI.

Fig. 6: Causality between the negative shock of uncertainty indexes and the Industrial production index (Quarterly frequency).

a Negative crude oil volatility vs IPI. b Negative GPR vs IPI. c Negative WUI vs IPI. Each panel illustrates the impact of negative shocks in uncertainty indicators—crude oil volatility, GPR, and WUI—on industrial production, measured by the IPI.

Discussion

In this section, we will discuss the spillover dynamics across different frequencies between uncertainty indices, economic growth and financial development for the KSA in the period 1993–2020. In the first step, we consider a spectral analysis based on the Granger causality test in the frequency domain. This spectral instrument can decompose the causality between two variables at different frequencies and refer to the short, medium and long run. In the second step, we use impulse response functions to show the mechanisms by which uncertainty factors can influence financial development and economic growth. This study provides interesting insights. Looking at the causality between the uncertainty caused by crude oil volatility and financial development (MCGDP), we recognise the existence of a persistent causal relationship between the risk associated with crude oil volatility across all frequencies (Fig. 7). It is also worth noting that when uncertainty is decomposed into a negative and a positive crude oil volatility shock, financial development is persistently more susceptible to a negative shock than to a positive shock. In general, these results are consistent with the findings of Su et al. (2019). An increase in the negative crude oil volatility shock can have a negative impact on the market capitalisation of the Saudi financial market. These causality results help the Saudi government to develop appropriate fiscal policy measures to strengthen the financial market and energy companies against a negative shock from crude oil volatility. Period 8 reveals a recovery in investment as investors regain confidence as the economy stabilises. Period 10 shows that it is difficult to retain foreign investment in times of continued volatility. Secondly, despite some negative fluctuations over time, government spending remains balanced. This shows that Saudi Arabia maintains fiscal discipline in response to oil price volatility. This highlights the crucial role of government spending as a stabilising factor for the economy. Keynesian economic theory, which emphasises the counter-cyclical role of government spending in times of economic shocks, can explain this result. According to Baunsgaard and Symansky (2009), when oil prices fluctuate, the Saudi government uses fiscal policy measures to offset private sector cuts. VTOIL influences investor behaviour, while the government uses spending measures as a stabilising instrument. The government maintains economic stability by slightly adjusting its spending so that the MCGDP does not experience major disruptions. Third, domestic investment (LOCINVEST) shows an early decline reflecting uncertainty caused by oil price volatility (period 2). This result is consistent with Bloom (2009), who has shown that uncertainty caused by oil price volatility leads to a delay in investment. In oil-rich countries such as Saudi Arabia, oil price volatility has contributed to a contraction in private sector investment, resulting in a slowdown in long-term economic growth and financial development. However, when the economy stabilises in the seventh and eighth periods, investment rises again and becomes positive. This is due to the importance of domestic variables, such as government spending and domestic investment, in maintaining financial growth. Conversely, during foreign investment recovery periods (period 8), there were positive improvements in MCGDP, indicating that foreign investment plays a more important role in promoting financial development. Regarding the causality between financial development and crude oil volatility, we find a transitory effect of financial development on crude oil volatility. The results also show that financial development, as measured by market capitalisation in Saudi Arabia, is a source of a positive shock to crude oil volatility in the long run. This implies that the Saudi government should ensure the stability of the financial sector in the long run by promoting financial inclusion and developing the financing environment for energy companies to manage their production and export of crude oil and thus participate in managing crude oil volatility in the future. Comparing the measures of financial development using the FIAIX and the MCGDP leads to quite different results. There is no obvious reciprocal effect compared to the financial market as measured by the MCGDP, indicating that bank-driven financial development is less susceptible to changes in the oil price. The positive shock of crude oil volatility has a non-significant effect on the FIAIX, suggesting that the Saudi economy and financial sector are resilient to short-term fluctuations in oil prices. Furthermore, no correlation was found between the FIAIX and the negative shock of crude oil volatility, suggesting that different financial market groups have different sensitivities. Regarding the relationship between financial development measured by the MCGDP and geopolitical risk, no causality was found from the GPR to financial development. Furthermore, it is worth noting that when considering a decomposition of uncertainty into a negative and a positive GPR shock, two important results were demonstrated. First, the positive shock of geopolitical tensions has an impact on financial development in the short and long run. Second, the negative GPR shock also affects financial development, but only in the long run. This emphasises that Saudi Arabia’s financial performance depends on the uncertainty of regional and global geopolitical developments and policies, especially in the long term. This prompts the Saudi government to develop financial measures to protect financial development by increasing market capitalisation through strengthening the role of funds in financing the national economy, attracting foreign investors to the Saudi financial market, diversifying the investment products and mechanisms available in the financial market, creating the regulatory environment for the financial market and increasing the level of transparency and disclosure in the financial market. The Saudi government can provide additional funding to companies in times of high geopolitical risk and directly support investments to boost economic performance. Indeed, additional financing for companies can reduce the negative impact of geopolitical risks. It can also help to accelerate financial and economic development. The banking sector is highly vulnerable to geopolitical tensions, as shown by the strong causality from GPR and FIAIX at all frequencies. It is clear from this that more efforts and robust laws are needed to reduce these risks. However, when we look at the positive shock to the GPR, we show that positive GPR shocks have a significant impact on the FIAIX across all frequencies, emphasising the impact of geopolitical tensions on banking sector dynamics. Differential effects on different financial market segments are evident in that the negative shock to the GPR does not show strong causality for the FIAIX.

Fig. 7: Impulse response of VTOIL on MCGDP, FORINVEST, LOCINVEST and GEXP over time.

Each panel shows the dynamic response of the respective variable to a positive or negative shock in VTOIL.

It is also important to show the mechanisms by which the geopolitical risk index can influence financial development. Figure 8 shows important results using the impulse response function. Firstly, FORINVEST shows significant volatility in response to geopolitical risk. In the early periods, FORINVEST increases (period 2), indicating that initial GPR shocks may lead to capital outflows as investors seek safer markets. Nevertheless, FORINVEST falls rapidly (periods 3 and 4) while geopolitical risk persists, indicating continued capital flight. This pattern is consistent with the flight to safety theory (Baur and Lucey, 2010), which states that in times of uncertainty, investors reallocate their capital to less risky markets. Bekaert and Harvey (2000) have shown that investors change their perception of risk as information becomes clearer over time. This suggests that the adjustment of investors’ decisions to the geopolitical environment may be responsible for the final returns of FORINVEST in subsequent periods, as was observed in periods 8 and 9. However, the negative change in period 10 highlights the long-term risks faced by foreign investors with GPR. A second finding is that government spending (GEXP) has gradually increased over time, indicating the Saudi government’s strategy to offset the negative impact of geopolitical risks through fiscal measures. The government is actively working to stabilise the economy by pumping money into the system, be it through social programmes, infrastructure projects or economic stimulus, which is evident in the increase in government spending. This is especially true in times when the MCGDP is declining (periods 4–7). The fact that GEXP remains positive over time suggests that government measures are a permanent factor that reduces the impact of exogenous shocks such as the GPR. The countercyclical fiscal policy hypothesis (e.g. Blanchard and Perotti (2002) and Ramey (2011)) show that governments increase spending to stabilise the economy in times of external shocks, confirming this approach to fiscal policy. Third, home bias theory (French and Poterba, 1991) argues that domestic investors are more inclined to hold on to their investments in local markets despite foreign threats (periods 2–4). This could be the reason why domestic investment initially react favourably to the GPR. This means that domestic investors continue to invest despite the withdrawal of capital by international investors, perhaps due to internal economic incentives or government support. However, as long as geopolitical uncertainties persist, investment gradually declines and eventually turns negative (period 9 and period 10). This shows that local investors are less confident due to the permanent uncertainty in the world, which weakens confidence and reduces long-term investment intentions. Despite the unfavourable FDI inflows, the MCGDP often performs well in the first periods. This is due to growing government spending and ongoing domestic investment. The rise in MCGDP in periods 2 and 3 is in line with increased government spending and investment and shows that the economy is able to temporarily protect itself from external shocks through internal forces. MCGDP declines in periods 4–7 as geopolitical uncertainties persist, indicating the limits of internal and government-led stabilisation initiatives. Despite the fact that FORINVEST picks up in period 8, MCGDP only partially recovers. This suggests that geopolitical issues continue to exist as a challenge. The recovery in MCGDP in period 10 is due to government spending and the brief positive boost from FORINVEST in preceding periods, but the recovery is uncertain due to ongoing negative domestic investment.

Fig. 8: Impulse response of GPR on MCGDP, FORINVEST, LOCINVEST and GEXP over time.

Each panel depicts the dynamic response of the respective variable to positive or negative shocks in GPR.

As far as the relationship between financial development and WUI is concerned, we show no causality between WUI and financial development. Thus, the decomposition of the WUI into positive and negative shocks shows different results. Firstly, the negative shock to the WUI influences financial development at the long-run level. This indicates that the negative shock of uncertainty in globally influential economies affects Saudi Arabia’s financial development in the long run. This leads to a slowdown in the investment decision-making process and reduces consumer appetite in the long run. Secondly, there is no evidence of a significant impact of the positive shock of the WUI on financial development. As for the causality between financial development and WUI, we find a permanent effect of financial development on WUI. This implies that Saudi policy makers should ensure the stability of financial development to contribute to the reduction of WUI in the long run by promoting financial inclusion and developing the financing environment for listed companies.

In addition, the impulse response function revealed crucial findings. Figure 9 shows how financial development in Saudi Arabia responds to fluctuations in the World Uncertainty Index (WUI) and its indirect effects through foreign direct investment (FORINVEST), domestic investment (LOCINVEST) and government spending (GEXP). This chart shows that the MCGDP is very volatile when uncertainty levels fluctuate. When uncertainty subsides, financial development can accelerate in the short run (as in period 3). However, prolonged uncertainty leads to significant declines, as can be shown in periods 5, 6 and 10. This reveals how financial development is influenced by longer periods of global instability. FORINVEST and LOCINVEST are directly related to changes in MCGDP. MCGDP often increases when both domestic and foreign investments are positive, as in periods 3 and 8, but financial development decreases when both are negative (periods 5, 6 and 9). This suggests that maintaining strong investment inflows is necessary to sustain the growth of the MCGDP. The global risk aversion hypothesis (Bekaert and Harvey, 2000) explains this by the fact that increased uncertainty leads investors to withdraw from riskier investments, which affects capital flows to developing countries. At the same time, GEXP becomes important. When government spending is higher, MCGDP generally grows or remains stable. However, as shown in periods 8 and 9, prolonged periods of uncertainty can gradually reduce the effectiveness of government initiatives, highlighting the limits of fiscal policy’s ability to promote financial development in times of prolonged global uncertainty.

Fig. 9: Impulse response of WUI on MCGDP, FORINVEST, LOCINVEST and GEXP over time.

Each panel illustrates the dynamic response of these variables to positive or negative shocks in WUI.

However, looking at the FIAIX as a measure of financial development, no significant causal relationship was found between the Global Uncertainty Index and the Global Uncertainty Index, indicating that global uncertainty does not have a direct impact on the development of the banking sector in Saudi Arabia. In addition, long-run causality was observed between the FIAIX and the WUI at certain frequencies (ω = 0.22), indicating a possible role of banking sector flexibility in reducing the impact of global uncertainty. Although, the decomposition of the WUI into positive and negative shocks shows important results. As well, when we consider the positive shock, no impact of WUI is evident on the FIAIX, meaning that the effect of WUI on the Saudi Arabia’s banking sector is limited. No significant causality was found for the FIAIX when taking into consideration the negative shock, either. As for the relationship between crude oil volatility and economic growth, no effect of crude oil volatility on economic growth was found, as demonstrated in the study by AlNemer et al. (2023). When we analyse uncertainty in both negative and positive crude oil volatility shocks, economic growth is more susceptible to a positive shock. However, it is not affected by a negative shock. Accordingly, it is possible to accept the hypothesis that economic growth in the KSA is affected by the shock of positive crude oil volatility from the energy sector during the period under study. Furthermore, this can be explained by the reaction of real GDP to fluctuations in international crude oil prices.

It is also important to show the mechanisms by which the volatility of crude oil can influence economic growth. Figure 10 shows important results using the impulse response function. First, foreign direct investment outflows have a significant impact on economic growth. Economic growth seems to recover in times when FORINVEST is positive or comparatively stable (periods 3, 4). However, significant outflows (periods 2, 6, 8) are directly correlated with slower growth rates, suggesting that economic growth in times of constant oil prices is highly dependent on attracting and retaining foreign capital. Secondly, government spending helps to stabilise economic growth. Compared to other indicators, the GEXP falls less sharply in times of increased oil price volatility. This suggests that the Saudi government is cushioning the negative impact of oil price shocks on the economy through its fiscal policy, as evidenced by the recovery in economic growth in periods 6 and 10 despite the volatility. Thirdly, sustainable economic growth requires domestic investment. In periods of positive or stable domestic investment, as in period 6, the decline in economic growth is less pronounced. However, periods (e.g. periods 2, 7, 8) in which domestic investment declines have a negative impact on economic growth. This exhibits how salient it is to keep domestic investment constant despite fluctuating global oil prices. All in all, VTOIL has a significant negative impact on Saudi Arabia’s economic growth (EG), especially regarding the country’s domestic and foreign direct investment outflows (FORINVEST and LOCINVEST). These results confirm the results of Arezki and van der Ploeg (2007). They found that oil price fluctuations not only reduce the inflow of foreign investment, but also reduce domestic investment, which leads to lower economic growth. Saudi government spending helps to mitigate the negative effects of oil price volatility, but the stable declines point to the challenges that arise during extended periods of instability. Saudi Arabia’s dependence on oil prices makes it vulnerable to conditions in the global market, emphasising the need for diversification to reduce this dependence and promote more stable growth even during times of high oil prices. Furthermore, this relationship can be understood through the resource curse theory (Sachs and Warner, 2001), which states that economies that are highly dependent on natural resources are more vulnerable to external shocks such as the possibility of rising oil prices. The negative effect of crude oil prices on the Saudi Arabian EG draws attention to the need for economic diversification, as the country’s dependence on oil prices exposes it to global market conditions.

Fig. 10

Impulse response of crude oil volatility on EG, FORINVEST, LOCINVEST and GEXP over time.

These results could prompt the Saudi authorities to further expand the economy by rising investment in key non-oil producing sectors of the economy and expanding small and medium enterprises to protect themselves from the shocks of oil price volatility. All this will lead to a balance between non-oil revenues and spending needs, so that a large part of oil revenues will not be diverted. As for the relationship between the uncertainty created by the GPR and economic growth, the results show no impact of the GPR on economic growth. Regardless of the positive or negative impact of the GPR, there is no impact on the economic growth of the KSA.

Nevertheless, when the impulse response function was incorporated, the findings were different. Figure 11 shows the direct and indirect effects of the GPR on the Saudi Arabian economy, broken down by government spending, domestic investment and foreign direct investment. Firstly, there are notable outflows of foreign direct investment in periods 3–5 with a simultaneous decline in economic growth. This suggests that foreign investors are pulling back in the face of increasing geopolitical risks, which is slowing economic growth. Conversely, the positive inflows of foreign investment in periods 8–10 point to a recovery in investor confidence and a slight acceleration in GDP growth (period 9). This shows the importance of foreign investment in economic growth and recovery. Secondly, government spending is continuously increasing in reaction to geopolitical risks, which is stabilising economic expansion. In periods when geopolitical risks are high (periods 4–6), the increase in government spending helps maintain positive economic growth even as FDI outflows rise and domestic investment begins to slow. Even if government spending remains high in the following decades, its ability to mitigate the negative effects of GPR diminishes slightly as continued geopolitical instability restricts both international and local investment flows. Third, despite the geopolitical risks, domestic investment increases in the initial period, perhaps due to the fiscal measures adopted by the government. This can favour positive economic growth in periods 2–6. However, the decline in domestic investment in recent years indicates a deterioration in investor confidence, which also reduces economic growth in these times. Government interventions initially ensure a favourable development of domestic investment. However, the impact of ongoing geopolitical uncertainty on the EG is obviously felt when it negatively affects domestic and international investment (periods 7–10). This points to the need for effective investment to maintain long-term growth. Overall, the GPR affects economic growth in Saudi Arabia in both the short and long term. The EG is anticipated to be fuelled by robust government spending and strong domestic investment. However, continued geopolitical uncertainty ultimately led to a decline in foreign and domestic investment and an overall slowdown in economic development. The results are consistent with the risk aversion hypothesis. Bekaert et al. (2005) hypothesise that increased capital flight triggered by geopolitical risks reduces investment in productive sectors and thus slows economic growth. When geopolitical problems calm down, FORINVEST and LOCINVEST are the driving forces for economic recovery; in contrast, GEXP plays an essential role in maintaining financial stability in difficult times. Blonigen and Piger (2014) note that investor confidence in foreign direct investment needs to be restored to support long-term growth, as it remains vulnerable to political and economic uncertainties. Saudi Arabia needs to diversify its economy and reduce its dependence on foreign investment and government spending in order to achieve more economic growth in the face of uncertainty.

Fig. 11

Impulse response of GPR on EG, FORINVEST, LOCINVEST and GEXP over time.

Regarding the effect of economic growth on geopolitical risk, we provide evidence of a permanent effect of economic growth on geopolitical risk. As for the causal relationship between the uncertainty generated by the WUI and economic growth, we show no effect of the index on economic growth. Regardless of the positive or negative impact of the WUI, no significant impact on the economic growth of the KSA is found. Regarding the effect of economic growth on the WUI, we show a permanent impact of economic growth on the WUI. This means that policy makers in Saudi Arabia should ensure the stability of economic growth to reduce geopolitical and global uncertainties and risks in the long run.

When analysing the impact of WUI on growth, it is important to show the possible ways in which WUI could influence economic growth. Figure 12 shows the empirical results that emerge from the impulse response function. It analyses the relationship between WUI and government spending (GEXP), investment (LOCINVEST) and foreign direct investment flows (FORINVEST) and the long-run effects of these variables on economic growth. Firstly, the EG tends to decline during periods of high FDI inflows (e.g. periods 2 and 5). This relationship emphasises how important FDI is for the EC. The economy experiences adverse effects when foreign investors withdraw from large investments due to increased uncertainty. But as in periods 4, 7 and 10, the inflow of foreign capital has often been a driver of economic growth, which is favourable for EC, especially for FORINVEST returns. Secondly, in uncertain times, GEXP continuously provides stability. When both domestic and international investments decline, the increase in government spending serves as a hedge against the negative impact on economic growth. For instance, when EG falls in period 5, government spending mitigates a more severe recession. In addition, the continuous rise in GEXP shows the Saudi government’s commitment to using fiscal policy as a tool to maintain economic stability. On the other hand, prolonged reliance on public spending without a comparable increase in private sector investment may eventually pose problems with fiscal sustainability.

Fig. 12

Impulse response of WUI on EG, FORINVEST, LOCINVEST and GEXP over time.

Reinhart and Rogoff (2010) display that relying only on government spending is problematic for long-term fiscal sustainability. A cautious equilibrium must be realised between public and private investment to boost sustainable growth. Third, domestic investment positively plays a part in economic growth in early phases of immense uncertainty. This indicates a domestic investors’ confidence in overcoming global challenges. This indicates that domestic investors, especially with business-friendly government policies, are more resilient to shocks coming from outside the world. This is consistent with the research of Aggarwal et al. (2011), which shows that domestic investors are often able to withstand shocks from around the world, especially when supportive government policies are in place. Our findings demonstrate this resilience by illustrating that domestic investment contributed to economic expansion in times of huge uncertainty. This is further evidence in favour of the theory that local confidence plays a key role in mitigating the damaging effects of global uncertainty. When EG falls in periods 5 and 8, it coincides with a fall in the LOCINVEST index. This emphasises the importance of domestic investment to economic growth. Strong domestic investment is crucial to offset the volatility of foreign investment, as shown by the correlation between higher domestic investment in periods 6 and 10 and higher economic growth. The WUI has an impact on economic growth, as it influences both domestic investment and foreign direct investment. Government spending is also essential for maintaining economic stability in turbulent times. However, if persistent financial problems arise, their long-term sustainability may be jeopardised.

In addition, many important results can be emphasised by the robustness analysis. In contrast to economic growth, which had no significant impact on uncertainty, the study found that using quarterly industrial production as a measure of economic activity has a significant impact on measures of uncertainty. This can be explained by the fact that manufacturing activity in Saudi Arabia is largely based on mining, quarrying and, most importantly, oil production. This diversity in the mix of industrial production, which is mainly based on oil production, influences oil prices and their fluctuations in global markets, especially since Saudi Arabia is known as the largest oil producer in the world. However, the results show that the impact of various uncertainty metrics on Saudi industrial production is non-existent. With the exception of crude oil volatility, the robustness analysis revealed a discontinuous impact in the long run. This can be explained by the efforts to increase economic stability in Saudi Arabia through the introduction of the Kingdom’s Vision 2030, which focuses on economic diversification and the promotion of non-oil industries such as manufacturing. This economic diversity has helped to reduce the impact of the Industrial Production Index on changes in oil production volumes as well as the impact of shocks resulting from various sources of uncertainty. Dividing the uncertainty indicators into positive and negative shocks and examining their impact on industrial production has yielded more profound and important results. First, the results show that the effect of the positive shock of crude oil volatility on industrial production is persistent from short to long-term horizons. This suggests that an increase in crude oil volatility in the international oil market affects Saudi economic activity as measured by industrial production. These results are consistent with Köse and Baimaganbetov (2015). This shows that the Saudi economy is vulnerable to the effects of positive shocks from oil volatility. Therefore, policymakers in Saudi Arabia need to promote economic diversification by supporting the non-oil private sector, retail, tourism, construction and especially investments in alternative energy sources such as renewable energy. Therefore, we encourage the Saudi authorities to utilise alternative renewable energy sources such as solar energy, especially in times of high oil prices. I also advise Saudi Arabia to quadruple the research projects and demonstrations of renewable energy technologies to tackle the problem of full correlation with oil. As a result, oil resources must be replaced by environmentally friendly alternatives. More new energies are urgently needed to achieve thermal power and independence from oil as the main resource. Solar and wind energy can help cover a large part of Saudi Arabia’s energy needs. Secondly, the findings suggest that the decline in crude oil volatility has a lasting effect on industrial production, which means that reducing turbulence in the oil market can affect economic activity. Saudi policy can capitalise on this by boosting industrial oil production.

Conclusion and policy implications

The interaction between uncertainty and economic indicators has changed in recent decades. Volatility in the oil market as well as geopolitical and economic events have emerged as the main causes of this uncertainty. Many economies have tried to overcome these risks by implementing appropriate measures and plans. This has also prompted many economic researchers to investigate the relationship between uncertainty, financial development and economic growth using various methods and data. In this study, we revisit the debate on this topic by examining the impact of different sources of uncertainty (crude oil volatility, geopolitical risk index and world uncertainty index) on financial development as well as economic growth in Saudi Arabia. By applying Granger causality in the frequency domain in conjunction with the impulse response function (IRF), we were able to gain important insights. We show that negative shocks in crude oil volatility significantly influence financial development as represented by the MCGDP. This is due to capital flight and the decline in domestic investment, despite the positive effect of government spending as a stabiliser of financial development. On the other hand, if we consider the FIAIX as a measure of financial development, the results show that the volatility of crude oil has different effects. When the shocks are positive, no significant contribution to financial development was found. However, when we consider the negative shocks, longer-term causal effects are evident and consequently a vulnerability of the financial sector to changes in oil prices was obtained. In addition, geopolitical risk has a significant impact on financial development as measured by the MCGDP, especially in the long term. It leads to capital outflows from abroad, although increased government spending and the initial resilience of domestic investment play an important role in mitigating the negative effect of capital outflows. However, continued instability blocks foreign and domestic investment, which in turn slows the acceleration of financial development. In terms of geopolitical risk, there are contradictory results. On the other hand, negative shocks have a significant causal relationship for the FIAIX at certain frequencies, which indicates the high susceptibility of the banking sector to geopolitical instability. Global economic uncertainty, on the other hand, has no direct effect on the FIAIX, suggesting that the Saudi financial sector is resilient to shifts in global uncertainty. Positive shocks to crude oil volatility have a much greater effect on economic growth than negative shocks. This means that crude oil volatility has a negative impact on the Saudi Arabian EG by affecting capital outflows and domestic investment, although government spending plays an important role in mitigating these effects. Similarly, neither positive nor negative shocks from geopolitical risks affect economic growth when Granger causality is used in the frequency domain. However, the impulse response function shows that GPR limits Saudi Arabia’s economic growth through a negative effect on domestic investment and capital outflows. This situation has been exacerbated despite the stabilising role of government spending in times of ongoing uncertainty. The robustness assessment of the industrial production index shows that the Saudi economy is vulnerable to positive shocks from oil price fluctuations over a range of time horizons. These results show the vulnerability of the Saudi economy to uncertainty factors across quarterly data frequencies and GDP indicators. The findings of this study have significant policy implications. The study emphasises the need for Saudi Arabia to diversify its economy beyond its dependence on the monopoly of oil revenues. Promoting non-oil industries, including manufacturing, tourism and renewable energy, could make the economy less vulnerable to oil price fluctuations. This kind of diversity could ultimately contribute to the growth of a strong and stable economy. Secondly, there is a need to strengthen the resilience of the financial sector in Saudi Arabia. According to the study, fluctuations in geopolitical tensions and oil prices could affect financial systems, particularly in indicators of access to financial institutions and market capitalisation. Plans that prioritise the expansion of financial access, the development of regulatory oversight and the improvement of risk management systems in financial institutions reduce these risks. Third, our study shows the important role that geopolitical issues play for the financial sector and economic growth in Saudi Arabia. To minimise the negative impact on the economy, policymakers should prioritise supporting economic stability by improving investor protection and transparency, which can boost investor confidence and attract foreign capital. Fourth, Saudi Arabia’s economic growth should continue despite international uncertainty through effective long-term economic planning. To this end, measures must be taken to strengthen institutional capacity, promote entrepreneurship and innovation, and diversify the economy. Investment in infrastructure and human capital development can promote long-term economic growth and reduce dependence on unstable sectors. Fifth, the study emphasises the importance of improving financial inclusion in Saudi Arabia. Improving access to financial services for underserved populations, including SMEs and rural communities, can boost economic growth, create employment opportunities and reduce income inequality. Implementing policies that improve financial literacy and expand microfinance programmes can enable greater economic empowerment for both people and small businesses. In addition, implementing sustainable energy practises is more important than ever and this approach would certainly mitigate the financial risks associated with oil price volatility. On the other hand, this strategy will improve Saudi Arabia’s position in terms of achieving the Sustainable Development Goals. Although this paper provides several important findings and policy implications, it suffers from some limitations. The research has not considered global economic policy uncertainty as a variable that can affect economic growth and financial development. Moreover, we can apply our model through a cross-country study in the region. Similarly, we can further detail our model by adding a regional factor and study the asymmetric analysis of the impact of crude oil change on the sustainable development of KSA. It would also have been better to use monthly data in the statistical analysis as this would be more useful for obtaining more accurate results. In addition, the research question did not consider comparative analysis with other MENA economies. If these limitations are considered, the results of this study can be further investigated and improved. In the future, new interpretable frequency methods and more uncertainty variables can be used.