Oil Prices, Monetary Conditions, and Growth Dynamics in Saudi Arabia: Evidence from an ARDL–ECM and VAR Approach

Abstract

This study examines the dynamic relationships among oil prices, monetary conditions, and nominal GDP growth in Saudi Arabia, with particular attention to short-run adjustment and long-run equilibrium patterns in an oil-dependent economy operating under a fixed exchange-rate regime. Rather than identifying structural monetary policy shocks, the study focuses on reduced-form dynamic associations between market-based monetary indicators, oil-price movements, and nominal economic activity. Using a high-frequency monthly dataset covering key macroeconomic variables, the analysis employs the Autoregressive Distributed Lag (ARDL) framework to estimate both short-run dynamics and long-run equilibrium relationships. An Error Correction Model (ECM) is used to capture the speed of adjustment toward equilibrium, while Granger causality tests assess short-term predictive linkages. The empirical results reveal that monetary indicators, particularly interest rates and money supply, exhibit lagged and non-monotonic associations with nominal GDP growth, reflecting delayed transmission under exchange-rate constraints. Oil-price movements emerge as a dominant driver, showing strong contemporaneous and lagged associations with growth, whereas inflation and exchange-rate movements display limited short-run predictive relevance. The ECM results indicate relatively rapid convergence toward long-run equilibrium, suggesting efficient adjustment dynamics. Granger causality findings further confirm the short-term predictive content of key macroeconomic variables. By integrating high-frequency data with ARDL–ECM estimation, VAR-based robustness checks, and sensitivity analysis, the study provides evidence on how oil-price movements, liquidity conditions, and interest-rate dynamics jointly shape growth fluctuations in Saudi Arabia.

1. Introduction

Monetary policy transmission refers to the process through which monetary policy actions are transmitted to the real economy, influencing variables such as inflation and economic growth (Alkhareif et al., 2017 [1]). Understanding these transmission mechanisms is essential for effective policy design, yet empirical evidence for Saudi Arabia remains relatively limited. This study contributes by analyzing the dynamic transmission and adjustment patterns of monetary policy variables in relation to Saudi Arabia’s economic growth, rather than asserting direct causal effects.

Saudi Arabia’s monetary policy framework is largely centered on liquidity management and interest-rate alignment within a fixed exchange-rate regime, where the riyal is pegged to the U.S. dollar and overseen by the Saudi Central Bank (SAMA). In such an environment, monetary policy autonomy is inherently constrained, and the effectiveness of traditional policy instruments depends heavily on external conditions and domestic liquidity dynamics. As noted by Belloumi et al. (2023) [2], maintaining monetary stability under a fixed exchange rate requires careful coordination between money supply management and external shocks.

Among the key indicators of monetary conditions in Saudi Arabia, short-term interest-rate movements—proxied by interbank rates—serve as an important channel through which monetary conditions are transmitted to credit markets. Variations in these rates are closely associated with borrowing costs, credit demand, and investment decisions, which in turn shape aggregate economic activity. Similarly, broad money supply provides a useful proxy for liquidity creation and credit availability, offering insight into the capacity of the monetary system to support private-sector activity.

Understanding the relationship between monetary policy and economic growth remains a central issue in macroeconomics. In oil-dependent economies such as Saudi Arabia, this relationship is particularly complex because domestic monetary conditions interact with fluctuations in global oil markets and fiscal dynamics. The effectiveness of monetary transmission depends not only on interest rates and liquidity but also on structural features such as the exchange-rate regime, financial market development, and the dominance of oil revenues in fiscal policy. These features can both amplify and dampen the transmission of monetary signals to real economic activity.

Over the past two decades, Saudi Arabia has implemented wide-ranging economic reforms aimed at diversification, inflation stabilization, and financial sector development in line with Vision 2030. Despite these reforms, oil revenues remain the primary driver of aggregate demand and public expenditure, rendering economic growth highly sensitive to oil-price movements. As a result, the interaction between monetary policy variables and oil-price fluctuations plays a critical role in shaping both short-run fluctuations and long-run growth trends.

Empirical evidence from oil-exporting economies provides mixed conclusions. Some studies suggest that monetary policy transmission is relatively weak when fiscal policy—largely driven by oil revenues—dominates aggregate demand (Hasanov et al., 2021 [3]). Other studies, however, find that interest rates, money supply, and inflation retain significant explanatory and predictive power for growth dynamics, particularly when long-run relationships are considered (Caporale & Helmi, 2018 [4]). Research focusing specifically on Saudi Arabia remains limited and often relies on low-frequency or bivariate approaches, which may overlook dynamic interactions and long-run equilibrium relationships among key macroeconomic variables.

This study is motivated by the need to provide a more comprehensive and high-frequency assessment of monetary policy transmission patterns in a resource-dependent economy operating under a fixed exchange-rate regime. While previous research has examined selected aspects of Saudi Arabia’s macroeconomic performance, evidence on how interest rates and money supply interact with oil prices and inflation at the monthly level remains scarce. By employing a combination of ARDL–ECM and VAR methodologies, this study captures both short-run adjustment dynamics and long-run equilibrium associations, offering a more granular perspective than existing studies.

The findings have important policy relevance. They provide insights for SAMA and fiscal authorities on the timing, persistence, and interaction of monetary and liquidity conditions with oil-price volatility. Understanding these dynamics can support more effective coordination of monetary and fiscal policies, enhance macroeconomic resilience, and reduce vulnerability to external shocks, all of which are central to the objectives of Vision 2030. By focusing on dynamic transmission rather than structural causality, this study contributes to both the academic literature on monetary policy transmission in oil-exporting economies and applied policy discussions.

Accordingly, this study addresses the following research question: How are changes in monetary policy indicators—captured by interest rates (SAIBOR) and broad money supply (M2)—dynamically associated with economic growth in Saudi Arabia, after accounting for oil prices, inflation, and exchange-rate movements? To address this question, this study employs the Autoregressive Distributed Lag (ARDL) approach (Pesaran et al., 2001 [5]) to estimate short-run dynamics and long-run relationships. Optimal lag lengths are selected using the Akaike Information Criterion (AIC), and bounds testing is used to assess cointegration. Adjustment toward long-run equilibrium is examined through an Error Correction Model (ECM), while Granger causality tests are used to evaluate short-run predictive relationships rather than causal effects.

It is important to clarify the scope of the study. The paper does not attempt to identify structural monetary policy shocks in the conventional SVAR/IRF framework. Instead, it examines the reduced-form dynamic association between monetary conditions and nominal GDP growth. Specifically, SAIBOR is used as a price-based indicator of monetary conditions, while M2 is used as a quantity-based indicator of liquidity. Oil prices are incorporated as a central external driver because, in Saudi Arabia, oil-price movements influence fiscal revenues, government spending, banking liquidity, and aggregate demand. Therefore, the empirical focus is on how these monetary and oil-related variables are dynamically associated with nominal growth, rather than on estimating the full structural monetary transmission mechanism.

To enhance robustness, the ARDL–ECM results are complemented with a Vector Autoregression (VAR) framework, which captures dynamic interdependencies without imposing long-run restrictions. The combined use of ARDL and VAR models allows for a comprehensive assessment of monetary transmission patterns and adjustment dynamics, strengthening confidence in the empirical findings.

The remainder of the paper is organized as follows. Section 2 reviews the theoretical and empirical literature on monetary policy and economic growth. Section 3 describes the data and variables. Section 4 outlines the empirical methodology. Section 5 presents and discusses the empirical results. Section 6 concludes with policy implications, limitations, and directions for future research.

2. Literature Review

The macroeconomic implications of monetary policy for economic growth have been extensively examined in the literature (Mehar, 2022 [6]). Ageli and Zaidan (2013) [7] emphasize that interest-rate and money supply channels remain central to monetary transmission, particularly in developing and emerging economies. A broad body of empirical work documents that monetary policy variables are systematically associated with economic growth through their influence on borrowing costs and liquidity conditions (Asaleye et al., 2021 [8]; Bekele, 2024 [9]). However, the strength, timing, and direction of these effects are not uniform across countries. They depend on the institutional monetary framework, the structure of the financial system, the degree of openness, and the extent to which fiscal revenues are exposed to external shocks. This is particularly important for oil-exporting economies, where monetary transmission does not operate independently from oil-price-driven fiscal and liquidity cycles.

Transmission mechanism theory explains how changes in interest rates and money supply are transmitted to output and inflation through multiple channels, including the interest-rate, credit, and balance-sheet channels. Among these, the interest-rate channel is often highlighted for its direct influence on aggregate demand and investment decisions (Bernanke & Gertler, 1995 [10]; Shapiro, 2024 [11]). However, the strength and timing of these channels depend heavily on institutional and structural characteristics. In a bank-based financial system, such as Saudi Arabia’s, monetary policy signals are transmitted mainly through bank lending rates, credit supply, liquidity conditions, and balance-sheet responses. Therefore, changes in SAIBOR and M2 may affect growth with delays rather than immediately, because firms and households adjust borrowing, investment, and spending decisions gradually.

The theoretical transmission mechanism relevant to this study can be summarized as follows. First, changes in SAIBOR affect borrowing costs and credit demand, which then influence investment and aggregate demand. Second, changes in broad money supply affect bank liquidity and credit availability, but their growth effect depends on whether liquidity is directed toward productive investment, consumption, asset markets, or precautionary balances. Third, oil-price shocks affect Saudi economic activity through fiscal revenues, public expenditure, external balances, banking-system liquidity, and private-sector confidence. Fourth, the SAR/USD peg links domestic monetary conditions to U.S. dollar monetary conditions, limiting independent monetary policy adjustment and making external monetary spillovers relevant for domestic growth dynamics. These channels suggest that monetary and oil-price effects may be lagged and non-monotonic rather than immediate and linear, while possible asymmetries between positive and negative oil-price shocks remain an important issue for future research.

Monetary policy plays a particularly complex role in open, resource-dependent economies, where domestic financial conditions interact with global commodity markets. Empirical evidence suggests that monetary transmission in such economies is often lagged, potentially asymmetric, and highly context-dependent, especially under fixed exchange-rate regimes that constrain independent policy action (Mishra et al., 2012 [12]; Guan et al., 2021 [13]). Compared with diversified economies operating under flexible exchange rates, oil-exporting economies face an additional complication: monetary variables interact with fiscal cycles because oil revenues influence government spending, deposits, banking liquidity, and aggregate demand. As a result, the estimated effect of monetary policy may partly reflect the joint operation of interest-rate, liquidity, and oil-revenue channels.

For oil-exporting economies, commodity price dependence introduces an additional layer of complexity. Oil-price shocks directly affect fiscal revenues, liquidity conditions, and investment patterns, thereby altering the transmission of conventional monetary policy tools (Alzyadat, 2024 [14]; Abid, 2025 [15]). Studies employing ARDL, VECM, and VAR approaches consistently show that oil-price movements are associated with both short-run expansions and lagged contractions in economic activity, reflecting the cyclical nature of hydrocarbon-based growth (Rehman & Ahmad, 2024 [16]; Ullah & Nobanee, 2025 [17]). These studies provide important evidence on the macroeconomic role of oil prices, but many of them treat oil prices mainly as an external control variable rather than as a central component of the monetary transmission process. This creates a gap in understanding how oil-price shocks interact with monetary indicators such as interest rates and money supply in shaping short-run growth dynamics.

Within the Gulf Cooperation Council (GCC), monetary transmission is further constrained by exchange-rate pegs to the U.S. dollar, which tie domestic interest rates to external monetary conditions (Tabash et al., 2025 [18]). Evidence from Saudi Arabia indicates that SAIBOR movements are associated with delayed and oscillatory growth responses, reflecting the interaction between domestic liquidity conditions and external monetary spillovers (Beisembina et al., 2024 [19]; Al-Nassar, 2025 [20]). Similarly, broad money supply (M2) exhibits mixed short-run associations with growth, where contemporaneous expansions may coincide with higher activity, while lagged effects may be linked to inflationary pressures or resource misallocation (Razia & Omarya, 2022 [21]). These findings suggest that the Saudi monetary transmission mechanism should not be interpreted through a simple linear interest-rate channel. Instead, it should be viewed as a dynamic process shaped by the exchange-rate peg, banking-sector liquidity, oil revenues, and fiscal-policy interactions.

Inflation dynamics also play a dual role. In the short run, moderate inflation may be associated with higher consumption and investment, whereas persistent inflation erodes purchasing power and constrains real growth over longer horizons (Blanchard, 2016 [22]). These inflationary effects interact with both monetary policy and oil-price shocks, creating a high-frequency and non-linear transmission environment (Stiglitz & Regmi, 2023 [23]). In the Saudi context, inflation may also reflect imported price pressures, domestic demand conditions, administered-price adjustments, and oil-related liquidity effects. Therefore, inflation is included not only as a macroeconomic control variable but also as a transmission channel through which monetary and oil-price shocks may affect growth.

Despite a growing literature, empirical studies focusing specifically on Saudi Arabia remain limited, with most relying on annual or quarterly data that may obscure short-run dynamics. Few studies jointly examine interest rates, money supply, oil prices, inflation, and exchange-rate movements within a unified high-frequency framework. Addressing this gap is crucial for understanding monetary transmission in the Saudi context and for informing policy design under Vision 2030. More specifically, four research gaps can be identified. First, previous studies often examine either monetary policy or oil prices separately, whereas Saudi growth dynamics are shaped by the interaction between both. Second, many studies rely on low-frequency data, which may miss short-run lag structures and adjustment patterns. Third, limited attention has been paid to how the SAR/USD peg affects monetary transmission by linking domestic interest rates to external monetary conditions. Fourth, few studies combine ARDL–ECM estimation with VAR-based robustness checks to examine both long-run adjustment and system-wide dynamic interdependence.

Four research gaps can therefore be identified. First, previous studies often examine monetary policy and oil prices separately, although Saudi Arabia’s growth dynamics are shaped by their interaction. Second, much of the existing literature relies on annual or quarterly data, which may obscure short-run lag structures and adjustment dynamics. Third, limited attention has been paid to how the SAR/USD peg affects monetary transmission by linking domestic interest rates to external U.S. dollar conditions. Fourth, few studies combine ARDL–ECM estimation with VAR-based robustness checks to examine both long-run adjustment and system-wide dynamic interdependence. This study addresses these gaps by using monthly data and an integrated framework linking SAIBOR, M2, oil prices, inflation, exchange-rate conditions, and nominal GDP growth.

In summary, existing studies suggest that monetary policy transmission in Saudi Arabia is shaped by oil-price volatility, liquidity management, and exchange-rate rigidity, with growth responses to monetary variables tending to be lagged, non-linear, and context-dependent. These insights motivate the present study’s use of monthly high-frequency data and robust multivariate econometric techniques, specifically ARDL–ECM and VAR models, to capture both short-run adjustment dynamics and long-run equilibrium relationships in Saudi Arabia’s economic growth. By linking monetary transmission theory with the institutional realities of an oil-dependent economy operating under a fixed exchange-rate regime, this study strengthens the academic dialog between the monetary policy literature, the oil-shock literature, and the GCC macroeconomic literature.

3. Data

The dependent variable in this study is the nominal GDP growth rate (NGDPG), which serves as a high-frequency proxy for short-run economic activity in Saudi Arabia. NGDPG is defined as the monthly percentage change in nominal GDP at current prices, capturing fluctuations in aggregate demand that are particularly relevant in a resource-dependent economy. Since official GDP data published by the General Authority for Statistics (GASTAT) are available only at the quarterly frequency, quarterly GDP series are temporally disaggregated to a monthly frequency using the Chow–Lin (1971) [24] interpolation method. This interpolation was performed by the author; the monthly GDP series was not taken directly from an external monthly GDP database.

The use of temporal disaggregation is justified by the empirical design of the study. The remaining variables used in the analysis, including SAIBOR, M2, CPI, Brent oil prices, and the exchange rate, are available at monthly frequency. Maintaining the analysis at quarterly frequency would considerably reduce the number of observations and weaken the model’s ability to capture short-run monetary transmission, lagged adjustment, and oil-price effects. Therefore, constructing a monthly GDP proxy is necessary to align the dependent variable with the frequency of the monetary and external indicators.

The Chow–Lin method is appropriate because it uses related high-frequency information to distribute a low-frequency series across higher-frequency observations while preserving consistency with the official quarterly data. In this study, monthly Brent crude oil prices are used as the high-frequency indicator guiding the temporal disaggregation. This choice is economically justified because Saudi Arabia’s nominal economic activity is strongly influenced by oil-market conditions through oil revenues, fiscal spending, liquidity creation, and aggregate demand. As a result, Brent oil prices provide relevant intra-quarter information for constructing a monthly proxy of Saudi nominal GDP.

The monthly growth rate is computed as the first difference of the natural logarithm of the interpolated GDP series multiplied by 100, a transformation that mitigates heteroskedasticity, captures proportional changes, and allows a direct interpretation of short-run growth dynamics.

It is important to clarify that the dependent variable (NGDPG) is defined as a monthly growth rate derived from log-differenced interpolated GDP. Consequently, the estimated ARDL long-run relationships should be interpreted as equilibrium associations among growth dynamics rather than level-based steady-state relationships. This distinction is consistent with the use of high-frequency growth proxies in macroeconomic time-series analysis. The use of nominal GDP growth is consistent with the objective of analyzing high-frequency nominal economic activity rather than estimating real welfare or productivity effects. Since monetary conditions, liquidity, oil revenues, and price-level movements directly affect nominal spending and income, NGDPG is an appropriate dependent variable for capturing short-run macroeconomic fluctuations. Inflation is included as a control variable to separate general price-level dynamics from movements in nominal economic activity.

The inclusion of the nominal exchange-rate variable is also justified by Saudi Arabia’s fixed exchange-rate regime. Although the SAR/USD peg is highly stable, the exchange-rate arrangement is central to the monetary transmission mechanism because it links domestic interest-rate movements to external U.S. dollar monetary conditions and constrains independent monetary policy adjustment. Therefore, EXCR is included not as a discretionary exchange-rate policy variable, but as a control for open-economy monetary spillovers, peg-related external conditions, and residual monthly variations associated with settlement timing and market microstructure. Including EXCR helps isolate the effects of domestic monetary indicators and oil prices on economic growth.

Monetary policy conditions are proxied by two complementary indicators reflecting both price-based and quantity-based transmission channels. The first indicator is the Saudi Interbank Offered Rate (SAIBOR) at the 3-month maturity, expressed as an annual percentage rate. SAIBOR represents the benchmark cost of unsecured interbank lending in Saudi riyals and is closely aligned with the Saudi Central Bank’s (SAMA) policy stance, particularly under the fixed exchange-rate regime. Monthly averages of daily SAIBOR are employed to smooth short-term volatility and better reflect prevailing monetary conditions, with data sourced from SAMA’s Monthly Statistics reports.

The second monetary indicator is broad money supply (M2), defined as the sum of currency in circulation, demand deposits (M1), and time and savings deposits. These variables capture liquidity creation and credit availability within the banking system, which are central to the monetary transmission process. M2 is expressed in millions of Saudi riyals at end-of-month values and obtained from SAMA’s monetary aggregates database. To ensure comparability across variables and address scale and variance instability, M2 is transformed into natural logarithms, and where appropriate, short-run dynamics are analyzed using first differences of the logged series.

SAIBOR and M2 are included simultaneously because they capture different dimensions of monetary conditions. SAIBOR represents the price-based channel, reflecting the cost of short-term funds and the interest-rate environment faced by banks and borrowers. M2 represents the quantity-based channel, reflecting liquidity creation and the availability of monetary balances in the banking system. Their joint inclusion allows the model to distinguish between interest-rate transmission and liquidity transmission.

To control additional macroeconomic forces influencing growth and monetary transmission, the model incorporates three key control variables.

The first is the Consumer Price Index (CPI), which measures headline inflation and is compiled monthly by GASTAT with a base year of 2015 = 100. CPI is included in natural logarithmic form, with monthly inflation rates calculated as first differences of the log-transformed series, enabling a consistent treatment of price dynamics alongside other variables.

The second control variable is the Brent crude oil price (OIL), expressed in U.S. dollars per barrel and measured as the monthly average of daily spot prices (FOB), sourced from the U.S. Energy Information Administration (EIA). Given Saudi Arabia’s strong dependence on hydrocarbon revenues, oil prices constitute a fundamental external driver of fiscal capacity, liquidity conditions, and aggregate demand, and their inclusion is essential for isolating the transmission of monetary variables.

The third control variable is the nominal exchange rate (EXCR), expressed as SAR/USD (approximately 0.2665), representing the amount of U.S. dollars per Saudi riyal. This corresponds to the long-standing peg of approximately 3.75 SAR per USD maintained by SAMA. Although the exchange rate regime is highly stable, small month-to-month variations arise due to settlement timing and market microstructure effects, and including this variable allows us to account for open-economy spillovers and residual exchange-rate-related transmission effects. Exchange-rate data are obtained from the IMF’s International Financial Statistics via the Federal Reserve Bank of St. Louis (FRED).

Although fiscal policy is central to Saudi Arabia’s macroeconomic adjustment process, monthly fiscal expenditure and fiscal-balance data are not consistently available over the full sample period at the same frequency as the monetary and financial variables. Therefore, fiscal effects are indirectly captured through oil prices, liquidity conditions, and nominal activity, while the absence of an explicit fiscal block is acknowledged as a limitation and a priority for future research.

The analysis covers the period from January 2010 to March 2025, yielding 183 monthly observations. This sample length is consistent with standard time-series applications in macroeconomics, particularly for single-country studies employing high-frequency data and providing sufficient degrees of freedom to capture dynamic adjustment processes and lagged transmission effects. Table 1 reports descriptive statistics for all variables used in empirical analysis.

Table 1. Descriptive statistics.

Variable	Obs.	Mean	Std. Dev.	Min	Max
NGDPG	183	0.5323	1.8489	−5.0587	4.5895
SAIBOR	183	2.1317	1.7968	0.6000	6.3207
M2	183	14.2859	0.2948	13.6406	14.8412
CPI	183	4.6222	0.0814	4.4245	4.7646
OIL	183	4.3098	0.3271	3.2715	4.8354
EXCR	183	0.2665	0.0002	0.2655	0.2667

Table 1 presents the descriptive statistics for all variables over the sample period.

The nominal GDP growth rate (NGDPG) averages 0.53% per month, with a standard deviation of 1.85 percentage points, ranging from a contraction of −5.06% to an expansion of 4.59%. The 3-month Saudi Interbank Offered Rate (SAIBOR) has a mean level of 2.13% per annum, exhibiting substantial variation (standard deviation 1.80), with rates as low as 0.60% and peaking at 6.32% during the tightening phase. Broad money supply (M2), measured in natural logarithms of millions of Saudi riyals, averages 14.29, with relatively low dispersion (0.29) and a range between 13.64 and 14.84. The Consumer Price Index (CPI), also in logs, shows a stable pattern with a mean of 4.62 and a narrow spread (0.08), reflecting limited monthly inflation volatility. Brent crude oil prices (OIL), in logarithmic form, average 4.31 (USD per barrel), with a standard deviation of 0.33 and extremes from 3.27 to 4.84, capturing both sharp downturns and peaks in the oil market. The nominal exchange rate (EXCR), in log form, remains remarkably stable around its pegged value, averaging 0.2665 with a minimal range between 0.2655 and 0.2667, consistent with Saudi Arabia’s fixed SAR/USD regime.

To ensure the stationarity of our time-series data, we perform the Augmented Dickey–Fuller (ADF) unit root test at the 5% significance level. The results (Table 2) indicate that the nominal GDP growth rate (NGDPG) and the nominal exchange rate (EXCR) are stationary in levels, with ADF statistics of −5.625 and −5.282, respectively, both exceeding the critical value of −2.885. In contrast, the remaining variables, 3-month Saudi Interbank Offered Rate (SAIBOR), broad money supply (M2), Consumer Price Index (CPI), and Brent crude oil price (OIL), are non-stationary in levels, as their ADF statistics are higher than the critical value, indicating the presence of a unit root. However, after first difference, all variables become stationary, with first-differenced series (DNGDPG, DSAIBOR, DM2, DCPI, DOIL, DEXCR) showing ADF statistics well below the 5% critical value and p-values of 0.0000. These results confirm that all series are integrated of order one, I(1), except for NGDPG and EXCR, which are I(0), providing a basis for appropriate modeling of short- and long-run dynamics in the monetary policy transmission analysis.

Table 2. Augmented Dickey–Fuller (ADF) unit root test results for stationarity.

Variable	ADF Statistic	5% Critical Value	p-Value	Variable in First Difference	ADF Statistic	5% Critical Value	p-Value
NGDPG	−5.6250	−2.8850	0.0000	DNGDPG	−9.5060	−2.8850	0.0000
SAIBOR	−0.5690	−2.8850	0.8778	DSAIBOR	−7.3850	−2.8850	0.0000
M2	−1.4780	−2.8850	0.5441	DM2	−8.8960	−2.8850	0.0000
CPI	−1.8760	−2.8850	0.3433	DCPI	−9.8220	−2.8850	0.0000
OIL	−2.3440	−2.8850	0.1583	DOIL	−9.5160	−2.8850	0.0000
EXCR	−5.2820	−2.8850	0.0000	DEXCR	−10.9370	−2.8850	0.0000

Potential multicollinearity among the explanatory variables was assessed using the variance inflation factor (VIF) diagnostic, with the results reported in Table 3.

Table 3. Variance inflation factor (VIF) diagnostic results for multicollinearity.

Variable	VIF	1/VIF
DOIL	1.1200	0.8940
EXCR	1.0800	0.9260
DSAIBOR	1.0400	0.9590
DM2	1.0000	0.9960
DCPI	1.0000	0.9980
Mean	1.0500

Table 3. Variance inflation factor (VIF) diagnostic results for multicollinearity.

Variable	VIF	1/VIF
DOIL	1.1200	0.8940
EXCR	1.0800	0.9260
DSAIBOR	1.0400	0.9590
DM2	1.0000	0.9960
DCPI	1.0000	0.9980
Mean	1.0500

Variance inflation factor (VIF) analysis indicates no evidence of problematic multicollinearity among the explanatory variables. All VIF scores are well below the common threshold of 10 (and even the more conservative threshold of 5), with a mean VIF of 1.05, suggesting that linear relationships among regressors are minimal and will not bias coefficient estimates through collinearity effects.

4. Methodology

This study examines the dynamic relationship between monetary policy indicators and economic growth in Saudi Arabia using monthly time-series data. The dependent variable is nominal GDP growth (NGDPG), while the explanatory variables include the 3-month Saudi Interbank Offered Rate (SAIBOR), broad money supply (M2), consumer prices (CPI), Brent oil prices (OIL), and the nominal exchange rate (EXCR). The empirical strategy focuses on dynamic associations, long-run equilibrium relationships, and adjustment mechanisms rather than strict structural causality.

4.1. ARDL Model

The Autoregressive Distributed Lag (ARDL) approach is used because it is suitable for variables integrated of order I(0) and I(1), provided that none is I(2). It also performs well in moderate samples and allows the estimation of both short-run and long-run relationships. The general ARDL model is specified as follows:

$$NGDP{G}_t={\alpha }_0+{\displaystyle \sum _{i=1}^p}{\alpha }_{i}NGDP{G}_{t-i}+{\displaystyle \sum _{j=0}^q}{\beta }_j{X}_{t-j}+{\epsilon }_t$$

(1)

where $X_t$ represents SAIBOR, M2, CPI, OIL, and EXCR; $p$ and $q$ denote lag lengths selected using information criteria; and ${\epsilon }_t$ is the error term. The ARDL coefficients are interpreted as reduced-form dynamic transmission effects rather than contemporaneous causal impacts.

4.2. Error Correction Model (ECM)

After testing for cointegration, the ARDL model is reparameterized into an Error Correction Model (ECM) to capture short-run adjustment toward long-run equilibrium:

$$\Delta NGDP{G}_t={\gamma }_0+\sum {\gamma }_i\Delta NGDP{G}_{t-i}+\sum {\delta }_j\Delta X_{t-j}+\varphi EC{M}_{t-1}+u_t$$

(2)

where $EC{M}_{t-1}$ is the lagged error correction term and $\varphi$ measures the speed of adjustment. A negative and statistically significant $\varphi$ indicates convergence toward long-run equilibrium.

4.3. VAR Robustness Check

To verify the robustness of the ARDL–ECM findings, a Vector Autoregression (VAR) model is also estimated. The VAR framework treats all variables as jointly endogenous and captures dynamic interdependencies without imposing long-run restrictions:

$$Y_t=c+A_1{Y}_{t-1}+\cdots +A_p{Y}_{t-p}+e_t$$

(3)

where $Y_t$ includes NGDPG, SAIBOR, M2, CPI, OIL, and EXCR; $A_i$ are coefficient matrices; and $e_t$ is the vector of innovations. Model adequacy is assessed using residual autocorrelation, heteroskedasticity, and stability diagnostics.

4.4. Methodological Scope and Sensitivity Checks

The ARDL–ECM and VAR approaches are used to identify dynamic transmission patterns, predictive linkages, and equilibrium adjustment mechanisms. Therefore, the empirical findings should be interpreted as reduced-form associations rather than definitive causal effects. This approach is appropriate for analyzing monetary transmission in a resource-dependent economy operating under a fixed exchange-rate regime.

To further assess the robustness of the empirical design, an additional sensitivity check was conducted by re-estimating the baseline ARDL specification after excluding the nominal exchange-rate variable (EXCR). This check is particularly relevant because the Saudi riyal is strongly anchored to the U.S. dollar, which may raise concerns that the inclusion of a nearly fixed exchange-rate variable could mechanically affect the results. The coefficients and significance patterns from the ARDL model without EXCR were then compared with those of the baseline specification.

The empirical strategy is designed to identify reduced-form dynamic associations, predictive relationships, and equilibrium adjustment mechanisms. It does not identify structural monetary policy shocks in the conventional SVAR or impulse-response framework. Therefore, the term “transmission” is used in a limited sense, referring to the dynamic association between market-based monetary indicators, oil-price movements, and nominal GDP growth. Future research could extend this framework using structural VAR, sign-restricted VAR, or local projections to estimate impulse responses to explicitly identified monetary policy and oil-price shocks.

5. Results

The optimal lag structure for the ARDL model was selected using the Akaike Information Criterion (AIC), which is appropriate for capturing short-run dynamics in monthly macroeconomic time series. This procedure ensures that the specification adequately reflects the lagged transmission patterns of monetary conditions, liquidity, inflation, oil prices, and exchange-rate movements in relation to Saudi Arabia’s economic growth. Based on AIC, the selected specification is ARDL(4,4,2,4,3,3). This lag structure is consistent with the view that monetary and external shocks propagate gradually, generating delayed and potentially oscillatory responses in high-frequency macroeconomic data, while also preserving sufficient degrees of freedom for reliable inference. Table 4 reports the estimated short-run ARDL coefficients.

Table 4. ARDL model estimation results.

Economic Growth
Variable	Lag	Coefficient	Std. Error	p-value	95% Confidence Interval
NGDPG	L1	0.8680	0.0729	0.0000	[0.7237, 1.0116]
NGDPG	L2	−0.0400	0.0930	0.6700	[−0.2269, 0.1462]
NGDPG	L3	−0.4270	0.0949	0.0000	[−0.6136, −0.2394]
NGDPG	L4	0.4240	0.0751	0.0000	[0.2758, 0.5727]
Monetary Policy
Variable	Lag	Coefficient	Std. Error	p-value	95% Confidence Interval
DSAIBOR	Current	−1.0170	0.5556	0.0690	[−2.1129, 0.0796]
DSAIBOR	L1	1.3640	0.6317	0.0320	[0.1190, 2.6088]
DSAIBOR	L2	−1.8400	0.6410	0.0050	[−3.1074, −0.5720]
DSAIBOR	L3	1.1760	0.6322	0.0650	[−0.0741, 2.4261]
DSAIBOR	L4	−0.8960	0.5643	0.1130	[−2.0088, 0.2161]
Money Supply
Variable	Lag	Coefficient	Std. Error	p-value	95% Confidence Interval
DM2	Current	8.2430	6.1059	0.1770	[−3.7774, 20.2642]
DM2	L1	−15.8910	6.1378	0.0100	[−28.0065, −3.7755]
DM2	L2	−9.4730	6.2762	0.1320	[−21.8300, 2.8836]
Inflation
Variable	Lag	Coefficient	Std. Error	p-value	95% Confidence Interval
DCPI	Current	4.7790	15.2490	0.7580	[−25.7639, 35.3209]
DCPI	L1	51.7900	15.0626	0.0010	[22.0772, 81.5021]
DCPI	L2	2.4812	13.7748	0.8570	[−24.7295, 29.6918]
DCPI	L3	9.6181	13.5136	0.4810	[−17.2674, 36.5036]
DCPI	L4	−29.6360	13.6464	0.0310	[−56.5969, −2.6747]
Oil Prices
Variable	Lag	Coefficient	Std. Error	p-value	95% Confidence Interval
DOIL	Current	3.8540	1.0250	0.0000	[1.8308, 5.8776]
DOIL	L1	1.0262	1.0581	0.3330	[−1.0615, 3.1139]
DOIL	L2	−2.3620	1.0922	0.0320	[−4.5217, −0.2017]
DOIL	L3	−2.0590	1.0618	0.0540	[−4.1503, 0.0328]
Exchange Rate
Variable	Lag	Coefficient	Std. Error	p-value	95% Confidence Interval
EXCR	Current	−16.8686	839.5838	0.9810	[−1435.1760, 1401.4390]
EXCR	L1	301.1289	795.0764	0.7080	[−1284.6310, 1886.8890]
EXCR	L2	−2248.2790	780.3090	0.0050	[−3792.1180, −704.4395]
EXCR	L3	1164.1600	718.9160	0.1070	[−254.3807, 2582.7010]
Constant	--	213.3582	206.9290	0.3020	[−194.0389, 620.7554]
Model Statistics		R-squared	0.7543	F-statistic	18.6700
		Adjusted R-squared	0.7139	p-value	0.0001

5.1. Short-Run ARDL Estimates and Interpretation

The lagged coefficients on nominal GDP growth (NGDPG) indicate strong persistence and cyclical adjustment. In particular, NGDPG exhibits a positive and highly significant first lag (0.868, p < 0.001), suggesting that monthly growth is strongly state-dependent, while the third lag is negative and significant (−0.427, p < 0.001), indicating short-run cyclical correction consistent with momentum followed by partial reversal. The fourth lag is again positive and significant (0.424, p < 0.001), reinforcing the presence of dynamic oscillation typical of macroeconomic adjustment processes. This pattern suggests that Saudi economic activity responds gradually over several months, which is consistent with the delayed effects of fiscal spending, investment cycles, and liquidity conditions.

Regarding monetary policy, changes in the 3-month SAIBOR (DSAIBOR) show non-monotonic and lag-dependent associations with growth, consistent with delayed pass-through under a fixed exchange-rate regime. The contemporaneous change is marginally negative (−1.017, p = 0.069), suggesting a weak short-run contractionary association, while the first lag is positive and statistically significant (1.364, p = 0.032) and the second lag is negative and significant (−1.840, p = 0.005). This alternating pattern implies that interest-rate shocks are transmitted to real activity with delays and potential overshooting, rather than through a single immediate effect. The third lag is again marginally positive (1.176, p = 0.065) and the fourth lag is negative but not statistically significant (p = 0.113). Economically, the alternating SAIBOR coefficients indicate that interest-rate transmission in Saudi Arabia is not immediate. Under the SAR/USD peg, short-term domestic rates are closely linked to external monetary conditions, while the effect on bank lending, credit demand, and investment decisions appears with a delay.

Money supply dynamics (DM2) exhibit a short-run pattern consistent with liquidity expansion followed by delayed adjustment. The contemporaneous coefficient is positive but insignificant (8.243, p = 0.177), whereas the first lag is negative and significant (−15.891, p = 0.010). This result suggests that prior liquidity expansions are associated with lower subsequent monthly growth, which may reflect delayed inflationary pressures, portfolio shifts, or short-run misallocation effects. The second lag remains negative but statistically insignificant (p = 0.132). This finding implies that liquidity growth does not automatically translate into immediate productive activity. In a bank-based financial system, the growth effect of money supply depends on whether additional liquidity is directed toward productive credit, consumption, precautionary balances, or asset-market activity.

Inflation dynamics (DCPI) show heterogeneous lagged associations. Only the first lag is positive and highly significant (51.790, p = 0.001), while the fourth lag is negative and statistically significant (−29.636, p = 0.031). This sign reversal is consistent with the interpretation that inflation may coincide with demand expansion in the very short run, but persistent or lagged inflation is associated with a subsequent deterioration in real activity through purchasing-power and cost channels. The remaining lags are statistically insignificant. Thus, inflation appears to operate as a short-run demand signal but becomes a constraint when its effects persist over time.

Oil prices (DOIL) show strong and economically meaningful short-run associations with growth, which is expected in an oil-dependent economy. The contemporaneous coefficient is positive and highly significant (3.854, p < 0.001), while the second lag is negative and significant (−2.362, p = 0.032), and the third lag is marginal (p = 0.054). This pattern is consistent with an immediate revenue-and-demand impulse followed by short-run normalization or adjustment effects, reflecting the cyclical nature of oil-driven fiscal and liquidity conditions. This result is particularly important for Saudi Arabia because oil-price increases can rapidly support fiscal revenues, liquidity, and confidence, while the negative lagged coefficients suggest partial adjustment after the initial oil-price impulse.

The exchange rate (EXCR), despite the SAR/USD peg, shows a significant negative coefficient at the second lag (−2248.279, p = 0.005), while other lags are not significant. Given the near-fixed exchange-rate regime, this result should be interpreted cautiously as capturing residual monthly variations arising from settlement timing, market microstructure, and external financial conditions rather than discretionary exchange-rate policy movements. The exchange-rate variable is therefore interpreted as a control for external monetary and peg-related conditions rather than as an active exchange-rate policy instrument.

The model exhibits strong explanatory power with R-squared = 0.754 and adjusted R-squared = 0.714, and the regressors are jointly significant (F-statistic = 18.67, p < 0.001). These results indicate that growth dynamics are strongly linked to lagged monetary and external conditions; however, the estimates are interpreted as reduced-form dynamic associations rather than structural causal effects. In summary, the ARDL results show that Saudi Arabia’s short-run growth dynamics are mainly shaped by delayed monetary transmission, liquidity adjustment, and oil-price movements.

Overall, the ARDL results suggest that Saudi Arabia’s short-run growth dynamics are shaped by delayed monetary adjustment, liquidity conditions, and oil-price movements. The results should be interpreted as reduced-form dynamic associations rather than structural monetary policy effects.

5.2. Diagnostic Tests

Table 5 reports standard diagnostic checks confirming the adequacy of the ARDL specification.

Table 5. Model diagnostic results.

Test	Name	Statistic	p-Value
Autocorrelation	Breusch–Godfrey LM test	0.0110	0.9184
Heteroskedasticity	White’s test	178.0000	0.4647
Heteroskedasticity	Breusch–Pagan/Cook–Weisberg test	2.6200	0.1056
Normality	Cameron & Trivedi IM-test—Skewness	11.2300	0.9918
Normality	Cameron & Trivedi IM-test—Kurtosis	2.4800	0.1155
Overall IM-test	Cameron & Trivedi IM-test—Total	191.7100	0.7047

The Breusch–Godfrey LM test indicates no evidence of residual autocorrelation (p = 0.9184). Heteroskedasticity tests (White and Breusch–Pagan/Cook–Weisberg) do not reject homoskedasticity (p = 0.4647 and p = 0.1056, respectively). Normality diagnostics based on the Cameron & Trivedi IM-tests indicate that residual skewness and kurtosis are consistent with normality (p = 0.9918 and p = 0.1155), with the overall IM-test also supporting well-behaved residuals (p = 0.7047). Collectively, these results support the reliability of inference from the estimated ARDL model.

5.3. Bounds Test for Cointegration

The ARDL bounds test was conducted to evaluate the existence of a long-run relationship between NGDPG and its determinants (SAIBOR, M2, CPI, OIL, EXCR). The results in Table 6 indicate an F-statistic of 4.88 (p = 0.0003), leading to a strong rejection of the null hypothesis that the lagged level terms are jointly zero. This provides evidence consistent with a stable long-run relationship among the variables and supports the use of the ECM representation.

Table 6. ARDL bounds test results for cointegration.

Test	H0	Statistic	p-Value
Bounds Test	All lagged variables jointly equal zero	4.8800	0.0003

5.4. Error Correction Model (ECM) Results

The ECM estimates in Table 7 indicate that the error correction term is negative and highly significant (−0.3633, p < 0.001), implying that approximately 36% of deviations from the long-run equilibrium are corrected within one month. This magnitude indicates relatively rapid adjustment dynamics, consistent with high-frequency macroeconomic equilibrium correction in a system shaped by strong fiscal capacity and liquidity management. Economically, this adjustment speed suggests that shocks to Saudi growth dynamics are partially absorbed within a short period, reflecting the role of fiscal stabilization, liquidity management, and the banking sector in restoring equilibrium.

Table 7. ECM estimation results.

Variable	Lag	Coefficient	Std. Err.	p-Value	95% Confidence Interval
ECM	ECM_lag	−0.3633	0.0596	0.0000	[−0.4810, −0.2456]
Monetary Policy (SAIBOR)	D1	−0.2718	0.5538	0.6240	[−1.3648, 0.8211]
Money Supply (M2)	D1	8.9580	4.5566	0.0510	[−0.0354, 17.9514]
Inflation (CPI)	D1	17.9456	11.3930	0.1170	[−4.5407, 40.4319]
Oil Prices (OIL)	D1	2.6687	0.7935	0.0010	[1.1026, 4.2348]
Exchange Rate (EXCR)	D1	272.7199	674.5345	0.6860	[−1058.603, 1604.043]
Constant	-	0.0016	0.0912	0.9860	[−0.1784, 0.1816]
Model Statistics		R-squared	0.2056	F-statistic	7.5100
		Adjusted R-squared	0.1782	p-value	0.0000

In the short run, the first difference of money supply is positive and marginally significant (8.9580, p = 0.051), and oil prices are positive and highly significant (2.6687, p = 0.001). This highlights the dominant role of oil-driven liquidity and revenue impulses in short-run monthly growth fluctuations. By contrast, short-run changes in SAIBOR, CPI, and EXCR are statistically insignificant, implying that their immediate month-to-month predictive relevance is weaker once equilibrium correction is accounted for. This result suggests that oil prices and liquidity conditions affect growth more rapidly than interest-rate and exchange-rate channels, whose effects are more likely to appear through delayed financial transmission.

5.5. Short-Run Granger Causality

A short-run Granger causality test was conducted to evaluate whether the differenced explanatory variables jointly predict changes in nominal GDP growth. As shown in Table 8, the null hypothesis that all differenced regressors jointly do not Granger-cause D.NGDPG is rejected (F = 3.50, p = 0.0048). This finding should be interpreted as evidence of short-run predictive content rather than structural causality, consistent with the reduced-form nature of the test.

Table 8. Short-run Granger causality test results.

Test	H0	Statistic	p-Value
Short-Run Granger Causality	All differenced variables jointly do not Granger-cause D.NGDPG	3.5000	0.0048

5.6. Robustness Checks: VAR Estimates and Diagnostics

To assess robustness, a VAR model was estimated using the same set of variables (Table 9). The VAR results broadly corroborate the ARDL–ECM patterns; notably: (i) strong persistence and cyclical adjustment in NGDPG; (ii) lag-dependent SAIBOR associations (with significant negative effects at L2 and L4); (iii) a significant negative association of lagged money supply (DM2 at L1); and (iv) statistically significant inflation and oil-price dynamics at specific lags. The replication of key signs and significance across ARDL and VAR strengthens confidence that the observed relationships reflect stable transmission patterns rather than model-specific artifacts. This consistency supports the view that the estimated dynamics are not specific to the ARDL specification but reflect broader macroeconomic interactions among growth, monetary conditions, liquidity, inflation, oil prices, and exchange-rate-related factors.

Table 9. VAR model estimation results.

Economic Growth
Variable	Lag	Coefficient	Std. Error	p-value	95% Confidence Interval
NGDPG	L1	0.8915	0.0718	0.0000	[0.7509, 1.0322]
NGDPG	L2	0.0247	0.0903	0.7840	[−0.1522, 0.2016]
NGDPG	L3	−0.3760	0.0906	0.0000	[−0.5536, −0.1985]
NGDPG	L4	0.4400	0.0747	0.0000	[0.2936, 0.5864]
Monetary Policy
Variable	Lag	Coefficient	Std. Error	p-value	95% Confidence Interval
DSAIBOR	L1	0.6950	0.5362	0.1950	[−0.3560, 1.7460]
DSAIBOR	L2	−1.9281	0.6150	0.0020	[−3.1336, −0.7227]
DSAIBOR	L3	1.1746	0.6199	0.0580	[−0.0405, 2.3896]
DSAIBOR	L4	−1.1090	0.5313	0.0370	[−2.1503, −0.0678]
Money Supply
Variable	Lag	Coefficient	Std. Error	p-value	95% Confidence Interval
DM2	L1	−14.4743	5.9606	0.0150	[−26.1569, −2.7917]
DM2	L2	−7.4355	6.1912	0.2300	[−19.5701, 4.6991]
DM2	L3	−5.4165	6.1517	0.3790	[−17.4735, 6.6406]
DM2	L4	−1.7810	6.2583	0.7760	[−14.0470, 10.4850]
Inflation
Variable	Lag	Coefficient	Std. Error	p-value	95% Confidence Interval
DCPI	L1	46.7991	14.5806	0.0010	[18.2217, 75.3764]
DCPI	L2	2.9022	14.0455	0.8360	[−24.6265, 30.4309]
DCPI	L3	3.3531	13.1279	0.7980	[−22.3771, 29.0832]
DCPI	L4	−23.8062	13.2118	0.0720	[−49.7009, 2.0884]
Oil Prices
Variable	Lag	Coefficient	Std. Error	p-value	95% Confidence Interval
DOIL	L1	0.6527	1.0314	0.5270	[−1.3687, 2.6741]
DOIL	L2	−3.5428	1.0307	0.0010	[−5.5631, −1.5226]
DOIL	L3	−2.6843	1.0383	0.0100	[−4.7193, −0.6492]
DOIL	L4	−1.0364	1.0149	0.3070	[−3.0255, 0.9528]
Exchange Rate
Variable	Lag	Coefficient	Std. Error	p-value	95% Confidence Interval
EXCR	L1	428.6360	686.4553	0.5320	[−916.7917, 1774.0640]
EXCR	L2	−2418.9760	772.9533	0.0020	[−3933.9370, −904.0156]
EXCR	L3	592.0973	797.1902	0.4580	[−970.3669, 2154.5610]
EXCR	L4	443.5605	735.3677	0.5460	[−997.7337, 1884.8550]
_cons	--	254.6354	194.1925	0.1900	[−125.9749, 635.2457]
Model Statistics		R-squared	0.7274
		chi squared	475.0385
		p-value	0.0000

VAR diagnostic tests (Table 10) support model validity. The LM test indicates no autocorrelation at conventional levels (p = 0.6442 for lag 1; p = 0.0851 for lag 2). White’s test suggests no heteroskedasticity (p = 0.1435). The eigenvalue stability condition is satisfied with a maximum modulus of 0.8384, confirming that the VAR system is dynamically stable. These diagnostics validate the reliability of the VAR-based robustness analysis.

Table 10. VAR model validity diagnostic results.

Test	Lag	Statistic	p-Value
VAR Autocorrelation (LM test)	1	32.3225	0.6442
	2	42.7924	0.0851
White’s Heteroskedasticity Test	–	1.6200	0.1435
Eigenvalue Stability Check	–	Max modulus = 0.8384	–

Although the VAR framework does not explicitly identify cointegrating vectors, it provides system-wide dynamic validation that complements the single-equation ARDL results. The VAR results therefore reinforce the main conclusion that Saudi growth responses to monetary and external variables are delayed and non-monotonic.

5.7. Additional Sensitivity Check

As an additional sensitivity check, the ARDL model was re-estimated after excluding the nominal exchange-rate variable (EXCR), given the high stability of the SAR/USD peg. This test evaluates whether the baseline findings are mechanically driven by the inclusion of a nearly fixed exchange-rate variable (Table 11).

Table 11. Sensitivity check: ARDL model without the exchange-rate variable.

Economic Growth
Variable	Lag	Coefficient	Std. Error	p-value	95% Confidence Interval
NGDPG	L1	0.8378	0.0727	0.0000	[0.6941, 0.9814]
NGDPG	L2	−0.0123	0.0920	0.8940	[−0.1941, 0.1695]
NGDPG	L3	−0.4722	0.0923	0.0000	[−0.6545, −0.2898]
NGDPG	L4	0.4369	0.0727	0.0000	[0.2933, 0.5804]
Monetary Policy
Variable	Lag	Coefficient	Std. Error	p-value	95% Confidence Interval
DSAIBOR	Current	−0.9004	0.5304	0.0920	[−1.9481, 0.1473]
DSAIBOR	L1	1.6300	0.6000	0.0070	[0.4449, 2.8151]
DSAIBOR	L2	−2.2953	0.6134	0.0000	[−3.5069, −1.0836]
DSAIBOR	L3	1.5905	0.6213	0.0110	[0.3633, 2.8176]
DSAIBOR	L4	−0.9129	0.5532	0.1010	[−2.0056, 0.1797]
Money Supply
Variable	Lag	Coefficient	Std. Error	p-value	95% Confidence Interval
DM2	Current	8.1245	6.1542	0.1890	[−4.0317, 20.2807]
DM2	L1	−14.6211	6.1851	0.0190	[−26.8384, −2.4038]
DM2	L2	−9.2124	6.2804	0.1440	[−21.6180, 3.1932]
Inflation
Variable	Lag	Coefficient	Std. Error	p-value	95% Confidence Interval
DCPI	Current	11.1033	15.4318	0.4730	[−19.3789, 41.5855]
DCPI	L1	49.6307	15.0384	0.0010	[19.9255, 79.3358]
DCPI	L2	3.8669	13.9380	0.7820	[−23.6647, 31.3985]
DCPI	L3	11.1327	13.7826	0.4200	[−16.0919, 38.3573]
DCPI	L4	−31.4248	13.8633	0.0250	[−58.8088, −4.0408]
Oil Prices
Variable	Lag	Coefficient	Std. Error	p-value	95% Confidence Interval
DOIL	Current	4.0478	0.9788	0.0000	[2.1144, 5.9812]
DOIL	L1	1.4927	0.9928	0.1350	[−0.4684, 3.4537]
DOIL	L2	−3.0550	1.0326	0.0040	[−5.0947, −1.0153]
DOIL	L3	−2.4114	1.0507	0.0230	[−4.4868, −0.3361]
Constant	—	0.1548	0.1202	0.2000	[−0.0826, 0.3922]
Model Statistics		R-squared	0.7391	F-statistic	21.0400
		Adjusted R-squared	0.7040	p-value	0.0000

Table 11. Sensitivity check: ARDL model without the exchange-rate variable.

Economic Growth
Variable	Lag	Coefficient	Std. Error	p-value	95% Confidence Interval
NGDPG	L1	0.8378	0.0727	0.0000	[0.6941, 0.9814]
NGDPG	L2	−0.0123	0.0920	0.8940	[−0.1941, 0.1695]
NGDPG	L3	−0.4722	0.0923	0.0000	[−0.6545, −0.2898]
NGDPG	L4	0.4369	0.0727	0.0000	[0.2933, 0.5804]
Monetary Policy
Variable	Lag	Coefficient	Std. Error	p-value	95% Confidence Interval
DSAIBOR	Current	−0.9004	0.5304	0.0920	[−1.9481, 0.1473]
DSAIBOR	L1	1.6300	0.6000	0.0070	[0.4449, 2.8151]
DSAIBOR	L2	−2.2953	0.6134	0.0000	[−3.5069, −1.0836]
DSAIBOR	L3	1.5905	0.6213	0.0110	[0.3633, 2.8176]
DSAIBOR	L4	−0.9129	0.5532	0.1010	[−2.0056, 0.1797]
Money Supply
Variable	Lag	Coefficient	Std. Error	p-value	95% Confidence Interval
DM2	Current	8.1245	6.1542	0.1890	[−4.0317, 20.2807]
DM2	L1	−14.6211	6.1851	0.0190	[−26.8384, −2.4038]
DM2	L2	−9.2124	6.2804	0.1440	[−21.6180, 3.1932]
Inflation
Variable	Lag	Coefficient	Std. Error	p-value	95% Confidence Interval
DCPI	Current	11.1033	15.4318	0.4730	[−19.3789, 41.5855]
DCPI	L1	49.6307	15.0384	0.0010	[19.9255, 79.3358]
DCPI	L2	3.8669	13.9380	0.7820	[−23.6647, 31.3985]
DCPI	L3	11.1327	13.7826	0.4200	[−16.0919, 38.3573]
DCPI	L4	−31.4248	13.8633	0.0250	[−58.8088, −4.0408]
Oil Prices
Variable	Lag	Coefficient	Std. Error	p-value	95% Confidence Interval
DOIL	Current	4.0478	0.9788	0.0000	[2.1144, 5.9812]
DOIL	L1	1.4927	0.9928	0.1350	[−0.4684, 3.4537]
DOIL	L2	−3.0550	1.0326	0.0040	[−5.0947, −1.0153]
DOIL	L3	−2.4114	1.0507	0.0230	[−4.4868, −0.3361]
Constant	—	0.1548	0.1202	0.2000	[−0.0826, 0.3922]
Model Statistics		R-squared	0.7391	F-statistic	21.0400
		Adjusted R-squared	0.7040	p-value	0.0000

The alternative ARDL specification remains highly significant overall, with F = 21.04, p < 0.001, and an adjusted (R²) of 0.704. Although the adjusted explanatory power is slightly lower than in the baseline model including EXCR, the main coefficient patterns remain qualitatively unchanged. In particular, lagged NGDPG remains strongly significant, SAIBOR continues to exhibit an alternating lagged transmission pattern, M2 remains negative and significant at the first lag, CPI remains positive at the first lag and negative at the fourth lag, and oil prices retain strong contemporaneous and lagged significance. These results confirm that the core conclusions are not driven by the inclusion of EXCR. Therefore, EXCR is retained in the baseline specification because it reflects Saudi Arabia’s fixed exchange-rate monetary framework and controls for external monetary and open-economy spillover conditions, but the sensitivity analysis shows that the main findings remain robust when it is excluded.

6. Discussion

The findings from the ARDL–ECM analysis indicate that Saudi Arabia’s economic growth is shaped by lagged and non-monotonic transmission patterns of monetary and liquidity conditions, a feature widely documented in small, open, resource-dependent economies (Mishra et al., 2012 [12]; Alaskar et al., 2024 [25]). The observed first-lag persistence in GDP growth underscores the economy’s inherent momentum, which is plausibly linked to the continuity of fiscal expenditure, credit conditions, and the multi-period nature of public and private investment projects. The subsequent negative third lag suggests that such momentum eventually tapers off, potentially due to adjustment costs, project completion cycles, inventory corrections, and the dissipation of short-run stimulus, paralleling the cyclical dynamics reported by Alharbi (2023) [26] in the Saudi context. In the Saudi economy, this lagged pattern is economically plausible because growth is strongly influenced by public expenditure, oil-revenue cycles, and large-scale investment programs. These channels usually affect output gradually through government procurement, infrastructure spending, bank credit, and private-sector confidence, rather than through an immediate one-month response.

The oscillatory association between SAIBOR (interest rate) and growth—weakly contractionary contemporaneously and alternating in sign across subsequent lags—is consistent with delayed and uneven pass-through under a fixed exchange-rate regime (Al-Jasser, 2008 [27]). In pegged systems, domestic short-term rates largely track the anchor currency stance, which can generate delayed transmission to domestic borrowing rates and credit supply, limiting the central bank’s capacity to respond countercyclically in real time (Bernanke & Gertler, 1995 [10]; Tonghui & Cho, 2024 [28]). Accordingly, the estimated lag pattern is best interpreted as evidence of gradual transmission and potential overshooting rather than a single immediate policy effect. This interpretation is particularly relevant for Saudi Arabia because the SAR/USD peg links domestic interest-rate conditions to U.S. dollar monetary conditions. Consequently, SAIBOR movements may reflect both domestic liquidity needs and external monetary policy pressures. The alternating signs across lags suggest that banks, firms, and households adjust lending, borrowing, and investment decisions gradually, producing delayed and non-linear effects on growth.

Broad money supply (M2) displays a dynamic pattern consistent with short-lived liquidity support followed by delayed adjustment, where the contemporaneous effect is weak while the first lag is negative and statistically significant. This suggests that liquidity surges may coincide with short-run activity but subsequently be associated with inflationary pressure, portfolio reallocations toward non-productive assets, or temporary misallocation of credit that weakens real-sector expansion (Tazhibayeva et al., 2008 [29]; Bordo & Duca, 2023 [30]). This interpretation aligns with the “liquidity overhang” mechanism observed in other oil exporters (Ben Cheikh et al., 2020 [31]) and is consistent with the broader GCC evidence that liquidity expansions may generate delayed macroeconomic costs when not accompanied by strong macroprudential and sterilization frameworks (Chaabouni & Abid, 2025 [32]). In Saudi Arabia’s bank-based financial structure, the effect of money supply depends not only on the amount of liquidity available, but also on the way this liquidity is transformed into productive credit. If liquidity is directed toward consumption, imports, precautionary deposits, or asset-market activity rather than productive investment, its growth effect may weaken after a short delay. This helps explain why monetary expansion does not generate a uniformly positive response across all lags.

Inflation (CPI) shows dual effects—positive at early lags and negative at later lags—suggesting a short-run demand-linked association followed by delayed real-income erosion. In the short run, moderate inflation may co-move with demand expansion and nominal spending, whereas persistent inflation can weaken real purchasing power and elevate uncertainty, thereby dampening real activity over subsequent months. This pattern is consistent with the Phillips curve’s short-run trade-off and its tendency to flatten over time (Blanchard, 2016 [22]; Rubbo, 2023 [33]). For Saudi Arabia, this mixed response may reflect the interaction between domestic demand, imported inflation, administered prices, and oil-related liquidity. A positive short-run inflation–growth association can occur when stronger demand raises both nominal activity and prices. However, once inflation persists, it can reduce household purchasing power, raise business costs, and weaken private-sector activity, explaining the negative delayed effect.

Similarly, oil prices exert a strong contemporaneous positive association with growth, reflecting immediate fiscal windfalls, liquidity expansion, higher public spending, and improved external balances, while the negative second lag suggests partial normalization through fiscal adjustment, import leakage, or demand cooling after the initial boom, a mechanism documented in hydrocarbon economies (Esfahani et al., 2012 [34]). This result reflects the central role of oil in Saudi Arabia’s macroeconomic structure. Higher oil prices rapidly improve fiscal revenues, external balances, banking-system liquidity, and market confidence. These effects can stimulate public expenditure and private-sector expectations in the short run. However, the negative lagged response suggests that the initial oil-price impulse is partly reversed over time through fiscal smoothing, import leakage, inflationary pressures, or delayed adjustment in expenditure and investment decisions. Therefore, the oil-price effect is lag-dependent and non-monotonic: it is strong and positive immediately but followed by partial correction in later months.

The nominal exchange rate exhibits a significant negative coefficient at the second lag. Given the stability of the SAR/USD peg, this result should be interpreted primarily as reflecting external monetary and financial conditions rather than discretionary exchange-rate policy. The sensitivity analysis excluding EXCR confirms that the core findings remain unchanged, supporting the robustness of the estimated monetary and oil-price transmission patterns. Nevertheless, even small peg-consistent fluctuations may be associated with changes in liquidity conditions and external financing costs that influence private-sector decisions with a lag (Maghyereh & Al-Kandari, 2007 [35]; Matallah, 2022 [36]). Therefore, EXCR is retained as a control variable capturing the institutional role of the exchange-rate peg and related external monetary spillovers.

The ECM results indicate that approximately 36% of disequilibrium is corrected each month, highlighting relatively rapid adjustment dynamics. This speed of adjustment suggests that deviations from long-run equilibrium are not persistent, potentially reflecting Saudi Arabia’s strong institutional capacity for liquidity management and fiscal stabilization. The magnitude is comparable to evidence for other GCC economies, where ECM adjustment speeds commonly fall between 25 and 40% monthly (Razia & Omarya, 2022 [21]). Economically, this adjustment speed means that more than one-third of the gap between actual growth dynamics and the long-run equilibrium relationship is corrected within one month. This relatively fast convergence is plausible in Saudi Arabia because fiscal policy, oil revenues, public-sector deposits, and SAMA’s liquidity-management tools can help absorb short-run shocks. The ECM coefficient therefore reflects not only statistical correction, but also the ability of Saudi institutions and the banking system to stabilize deviations from equilibrium.

The VAR-based robustness checks reinforce these conclusions, showing consistency in the signs and significance of key variables—particularly oil prices, SAIBOR, and M2—across alternative dynamic specifications. The stability diagnostics further indicate that the results are unlikely to be statistical artifacts driven by misspecification but instead reflect systematic macroeconomic linkages in the Saudi economy. The consistency of these patterns with prior studies provides additional external validation and strengthens confidence in the transmission interpretation. The VAR results are especially useful because they treat the variables as jointly endogenous, which is appropriate for a macroeconomic system where oil prices, monetary conditions, liquidity, inflation, exchange-rate-related factors, and growth interact simultaneously. The similarity between the ARDL–ECM and VAR results confirms that the findings are not dependent on one specific modeling framework.

Overall, these findings position Saudi Arabia within the broader empirical literature on monetary transmission in oil-exporting economies, confirming both the structural constraints imposed by the peg and the outsized role of commodity prices and liquidity conditions in shaping high-frequency growth dynamics (Abid et al., 2024 [37]; Abid, 2026 [38]). This conclusion is consistent with recent evidence showing that oil dependence continues to shape macroeconomic and financial market dynamics in Saudi Arabia despite ongoing economic diversification efforts. The main economic implication is that monetary transmission in Saudi Arabia differs from that in more diversified economies with flexible exchange rates.

7. Conclusions

This study examined the dynamic relationships between monetary conditions, oil prices, and economic growth in Saudi Arabia using an ARDL–ECM framework complemented by VAR robustness checks. The results indicate that monetary transmission through SAIBOR is delayed and non-monotonic, reflecting gradual adjustment under the fixed exchange-rate regime. Liquidity conditions, captured by M2, exhibit significant lagged effects, while oil prices emerge as a key driver of short-run growth fluctuations. The ECM results reveal relatively rapid convergence toward long-run equilibrium, with approximately 36% of disequilibrium corrected within one month. Overall, the findings highlight the importance of monetary conditions, liquidity dynamics, and oil-market developments in shaping Saudi Arabia’s growth performance.

From a policy perspective, the findings call for more targeted and timing-sensitive policy coordination rather than broad monetary expansion alone. First, because SAIBOR effects are delayed and non-monotonic, SAMA and fiscal authorities should strengthen forward-looking monitoring of interest-rate transmission, especially during periods of U.S. monetary tightening or easing. Under the SAR/USD peg, domestic short-term rates partly reflect external monetary conditions; therefore, credit conditions, loan repricing, and bank liquidity should be monitored before policy shocks fully affect real activity. Second, the negative lagged effect of M2 suggests that liquidity expansion should be accompanied by macroprudential safeguards that direct credit toward productive sectors rather than speculative, consumption-driven, or low-productivity uses. Practical measures include closer monitoring of sectoral credit growth, loan concentration, real-estate exposure, and banks’ liquidity-to-credit conversion. Third, the strong contemporaneous effect of oil prices indicates that fiscal planning should continue to reduce the procyclicality of public spending. Oil-price windfalls should be partly saved or allocated through stabilization mechanisms, while expenditure programs should be smoothed over time to avoid short-lived demand booms followed by adjustment pressures. Fourth, given the central role of oil prices in short-run growth, Vision 2030 diversification policies should prioritize non-oil tradable sectors, private-sector productivity, and investment channels that reduce the sensitivity of growth to oil-price cycles.

The robustness checks using VAR analysis and the sensitivity test excluding the exchange-rate variable confirm that the main findings are not driven by a single model specification. The persistence of the main results after excluding EXCR supports the decision to retain the exchange-rate variable in the baseline model as a theoretically relevant control for Saudi Arabia’s fixed exchange-rate framework, while also confirming that the core conclusions are mainly driven by monetary, liquidity, and oil-price dynamics.

Several limitations should be acknowledged. First, the econometric framework identifies dynamic associations, predictive linkages, and equilibrium adjustment, but it does not establish structural causality or identify monetary policy shocks through impulse-response functions. Second, nominal GDP is temporally disaggregated from quarterly to monthly frequency, which allows high-frequency analysis but may introduce measurement uncertainty. Third, the model does not include an explicit fiscal-policy block, although fiscal expenditure, oil revenue, public-sector deposits, and fiscal balances are central to Saudi Arabia’s macroeconomic adjustment process. Fourth, the use of aggregate macroeconomic data does not allow the study to identify sector-specific transmission channels or bank-level credit responses.

Future research should extend this framework by incorporating fiscal variables such as government expenditure, oil revenue, fiscal balance, or public-sector deposits. It should also apply structural VAR, sign-restricted VAR, or local projection methods to estimate impulse responses to separately identified monetary policy, liquidity, oil-price, and oil-production shocks. Further work could also use bank-level or sectoral credit data to examine whether monetary transmission operates mainly through corporate lending, household credit, real estate, or public-sector-related liquidity channels. Finally, comparative studies across GCC economies would help determine whether the estimated patterns are specific to Saudi Arabia or reflect a broader regional feature of oil-exporting economies under fixed exchange-rate regimes.

In sum, this study contributes to the understanding of Saudi Arabia’s macroeconomic dynamics by providing high-frequency evidence on how monetary conditions, liquidity, and oil-price volatility are associated with growth fluctuations and long-run adjustment. The main policy implication is that Saudi Arabia’s stabilization strategy should combine forward-looking liquidity management, careful monitoring of bank credit allocation, fiscal smoothing of oil revenues, and continued diversification of non-oil productive sectors. Such a policy mix would help reduce the delayed costs of liquidity expansion, strengthen resilience to oil-price shocks, and support the long-run objectives of Saudi Arabia’s sustainable development agenda.