Fiscal Policy and Economic Resilience: The Impact of Government Consumption Alongside Oil and Non-Oil Revenues on Saudi Arabia’s GDP during Crises (1969–2022)

Round 1

Reviewer 1 Report (Previous Reviewer 1)

Comments and Suggestions for Authors

The comments are included in the attached file

Comments for author File: Comments.pdf

Author Response

Thank you for your thorough review. We are pleased to note that Reviewer 1 has accepted our responses to Comments 1 through 3and Comment 6. The provided responses have been reviewed and are deemed thorough, addressing the reviewer's comments adequately.

Regarding Comment 3:

The appearance of non-oil revenues with a non-significant effect may be attributed to the presence of an econometric problem in the estimated model, especially since the value of the multiple determination coefficient is high and reached about 0.98. Here I would like to point out that the high value of the multiple determination coefficient with the appearance of one of the independent variables with a non-significant effect may be attributed to the presence of a problem (Multicollinearity), which the researcher neglected to test. Also, the value of the non-oil revenues coefficient was very small, which confirms the existence of a problem in the estimated model, so I advise the researcher to test the Multicollinearity problem.

Response 3: Thank you for your insightful comment. We have carefully reconsidered and refined our model to address the potential issues you highlighted, including the high value of the multiple determination coefficient (R²) and the non-significant effects of certain independent variables. Initially, our model exhibited a very high R² value of 0.98, suggesting that it explained a large proportion of the variance in GDP. However, this raised concerns about potential multicollinearity and overfitting, especially given the non-significant effects of non-oil independent variables.

To address this, we conducted a Variance Inflation Factor (VIF) test to detect the presence and severity of multicollinearity among the independent variables. The VIF values, as presented in Table 10 of the updated analysis, indicated moderate levels of collinearity for most variables, with some lagged variables of non-oil revenues and government consumption showing higher VIF values. Despite these findings, the overall results suggest that multicollinearity is not severe enough to undermine the model significantly, supporting the robustness of our regression estimates.

After rigorous diagnostic tests and model re-specification, the R² value in the updated model decreased to 0.560 (Table 6). This change reflects a more realistic and conservative estimate of the model's explanatory power, suggesting that the initial model likely suffered from multicollinearity or overfitting issues, which have been mitigated. The updated model provides a better balance between explanatory power and robustness, ensuring that the estimated effects are more reliable.

When we included interaction terms in the ARDL model to account for the effects of specific economic crises, the R² value increased to 0.838 (Table 12). This increase indicates an improved fit of the model, capturing more variability in GDP by considering the differential impacts of economic crises. The inclusion of interaction terms helped to isolate the effects of government consumption, oil revenues, and non-oil revenues during these crises, leading to more precise and significant estimates.

Regarding the non-significant effect of non-oil revenues in the initial model, the coefficient was very small and raised concerns about the model's validity. In the updated ARDL model with interaction terms, the coefficient for non-oil revenues improved significantly, reflecting a better specification of the model. This improvement suggests that the refined model more accurately captures the impact of non-oil revenues on GDP, addressing the reviewer’s concern about the existence of an econometric problem.

To further enhance the robustness and reliability of our analysis, we have revised the methodology section to address the reviewer's concerns and improve clarity. Specifically, we have clarified that GDP, Government Consumption, and Oil and Non-oil Revenues are now presented in real values. Previously, nominal values were used, which could have introduced bias into the results. By adjusting the variables to real values using the Consumer Price Index (CPI), we ensure that the data are comparable across time, thereby enhancing the robustness and accuracy of our findings. Additionally, we conducted a thorough reanalysis of the adjusted data, accounting for inflation, to enable meaningful comparisons over time.

We hope these modifications address your concerns effectively. Additionally, we would appreciate any further guidance or specific recommendations for improvement that you may have.

Thank you once again for your constructive feedback, which has significantly contributed to enhancing the rigor and clarity of our analysis.

Regarding Comment 5, we have carefully revised our manuscript to address the structural changes that appeared in our research results, as suggested. The necessary revisions have been added and integrated into the sub-section titled "5.5 Diagnostic Tests and Robustness Checks." It now includes the following paragraph: “Additionally, our study acknowledges the significant influence of global economic fluctuations, such as oil price shocks, on Saudi Arabia's economy. This is evident in several key findings. The historical data analysis reveals significant fluctuations in oil revenues corresponding to global oil price shocks, such as the sharp decline in oil revenues during the 2014-2016 oil price collapse, which had a pronounced negative impact on GDP growth. Stability checks using cumulative CUSUM and CUSUM of Squares tests indicate some instability in the coefficients, suggesting structural breaks that align with significant global economic events, such as the 1973 Oil Crisis, the 1990 Gulf War, and the 2008 Financial Crisis, all of which had substantial impacts on the Saudi economy. The increasing trend in non-oil revenues, particularly post-2010, reflects Saudi Arabia's efforts to mitigate the impact of global oil price volatility through economic diversification, which is crucial in reducing economic dependency on oil and enhancing resilience to global economic fluctuations.

By incorporating these analyses, our research aligns with the observation that oil-dependent economies, such as Saudi Arabia, are significantly impacted by global economic events. The structural imbalances observed in our study underscore the importance of considering global economic conditions in macroeconomic analyses of oil economies. These findings provide a comprehensive understanding of the economic dynamics in Saudi Arabia, highlighting the critical role of global economic fluctuations and the ongoing efforts towards economic diversification.

Reviewer 2 Report (Previous Reviewer 2)

Comments and Suggestions for Authors

The authors have addressed all my comments. I still recommend one more round of proofreading and to mention the base year used for CPI deflation of nominal variables. 

Comments on the Quality of English Language

Another round of proofreading is needed. Figures' captions can be improved for clarity and for being more succinct. 

Author Response

Thank you for your thorough review and for confirming that we have addressed all your comments. We appreciate your recommendation for an additional round of proofreading and the suggestion to mention the base year used for CPI deflation of nominal variables.

We have carefully reviewed the entire manuscript to correct any typographical errors, grammatical mistakes, and inconsistencies, ensuring that the writing is clear and accurately conveys our findings. Additionally, we have updated the methodology section to specify the base year used for CPI deflation. The nominal variables, including GDP, Government Consumption, and Oil and Non-oil Revenues, have been adjusted to real values using the Consumer Price Index (CPI) with the base year set to 2010.

To address the suggestion of improving figures' captions for clarity and succinctness, here are revised versions of the captions in this manuscript:

Figure 1. Oil Revenues (1969-2022): This graph shows the time series data of oil revenues in Saudi Arabia, highlighting significant peaks and troughs corresponding to global oil price fluctuations.

Figure 2. Non-Oil Revenues (1969-2022): This graph displays the time series data of non-oil revenues, illustrating a steady increase over time, particularly accelerating post-2000 due to diversification efforts.

Figure 3. Government Consumption (1969-2022): The graph depicts the trend of government consumption over the years, showing fluctuations that align with oil revenue changes.

Figure 4. Real GDP (1969-2022): This graph represents the real GDP of Saudi Arabia over the sample period, adjusted to 2010 CPI values.

Figure 5. CUSUM Test for Model Stability: This graph shows the cumulative sum (CUSUM) of residuals, used to assess the stability of the model's coefficients over time.

Figure 6. CUSUM of Squares Test for Model Stability: The graph presents the CUSUM of squares of residuals, indicating potential structural breaks or instability in the model's coefficients.

These revisions have been incorporated to enhance the transparency and robustness of our analysis.

Thank you once again for your valuable feedback and for helping us improve the quality of our manuscript.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

-         Note (there is a problem displaying the tables). The tables are not clear and do not clearly express the results of the analysis.

-         There is a problem in translating some econometric  terms, for example, serial correlation should be Auto correlation as is known in econometric  analysis.

-         The appearance of non-oil revenues with a non-significant effect may be attributed to the presence of a econometric  problem in the estimated model, especially since the value of the multiple determination coefficient is high and reached about 0.98. Here I would like to point out that the high value of the multiple determination coefficient with the appearance of one of the independent variables with a non-significant effect may be attributed to the presence of a problem(Multicollinearity), which the researcher neglected to test. Also, the value of the non-oil revenues coefficient was very small, which confirms the existence of a problem in the estimated model, so I advise the researcher to test the Multicollinearity problem.

-         I believe that the discussion of the results of the ARDL model was a great exaggeration on the part of the researcher, meaning that it is not reasonable that the results of the estimated model explain all the discussions raised by the researcher about measures to counter cyclical fluctuations, public investments, the challenges of pro-cyclical spending, and the social and political effects of fiscal adjustments. I believe that there is a great exaggeration in relying on the results of the ARDL model in arriving at the conclusions proposed by the researcher, and I believe that the matter involves theoretical conclusions, not practical ones, based on the results of the estimated model.

- In the Bound test, there is a set of drawings showing structural imbalances, and the researcher neglected to include them in the research.

         -  There is a large theoretical framework that can be summarized.

I suggest that the analysis be repeated again and that econometric problems be tested, especially the Multicollinearity problem

Comments on the Quality of English Language

The research needs a slight modification in terms of language

Reviewer 2 Report

Comments and Suggestions for Authors

Summary:

The paper aims to uncover the impact of government consumption on GDP growth in Saudi Arabia. The paper also discusses the impact of oil and non-oil revenues on GDP growth. The paper uses data from the World Bank covering the period 1990 until 2022. The author(s) apply(ies) an ARDL model to probe the relationship mentioned. Results reveal that government consumption protects the economy during shocks and stimuli the economy.

Evaluation:

The paper offers an already well-established and discussed topic in the literature in general. The novel approach is distinguishing between oil and non-oil revenues in an oil-dependent economy. The paper overall is decently written. However, it entails several unnecessary details and lacks important ones. The first main issue is the small number of observations. The paper mentions in the abstract that the years observed are between 1990 and 2022. Yet, the model includes only 30 observations (as seen in Table 4.5) rendering the results statistically biased. Another main issue is discussing Keynesian theory in the literature instead of sticking to empirical findings and evidence from different countries/region. The Keynesian theory is well-known and understood by all economists there is no need to dwell on its details. The conclusion also should be shortened since it includes unnecessarily long discussion and must be shortened. I recommend rejecting the paper for the mentioned reasons and due to the major comments below.

Major Comments:

-Make the literature about related evidence instead of explaining Keynesian Economics.

Sections 2.2 until the first two paragraphs in section 2.8 should be removed since the discussion is common knowledge among economists. The first two paragraphs in section 2.10 discuss the importance of diversification which is also a well-known and agreed-upon topic, that part should be removed.

On page 5 line 199, the paper mentions the research examination of Shaheen (2019) without mentioning its results. Include previous findings in the literature to have an added-value in it.

-Methodology lacks clarity.

The equation of the ARDL model is missing. Moreover, it is not clear if GDP, Government Consumption, and Oil and non-oil revenues are in nominal or real values. If the former is the case, then the results will be biased. Apply real values and state that they are real to make a comparison across time.

-Issues in Figures 1 and 2.

Figure 1 plots the variables overtime, which is not needed.

Figure 2 shows each variable in a plot on its own overtime and underneath a paragraph that mentions the absence of a time-trend ‘The above figures 2 show that the time series do not have a general trend over time.’. It is clear that at least GDP and GGEX have a positive time trend. The explanation underneath Figure 2 is misleading.

-30 observations will render the results bias.

If you want to stick to the same time period use monthly or at least quarterly data if you want to apply the ARDL model.

-The conclusion includes many repetitions.

Do not compare the case of Venezuela to Saudi Arabia without using additional citations. Comparability is difficult without controlling for multiple other variables.

-Absence of organization.

Page 17 has table 4.8.

Page 18 has table 4.5.

Label the tables in order.

Author Response

Comment 1: Make the literature about related evidence instead of explaining Keynesian Economics. Sections 2.2 until the first two paragraphs in section 2.8 should be removed since the discussion is common knowledge among economists. The first two paragraphs in section 2.10 discuss the importance of diversification which is also a well-known and agreed-upon topic, that part should be removed. On page 5 line 199, the paper mentions the research examination of Shaheen (2019) without mentioning its results. Include previous findings in the literature to have an added value in it.

Response 1: Thank you for your constructive feedback. We have revised the manuscript to address your concerns and suggestions comprehensively. To enhance the relevance and specificity of the literature section, we have restructured the review to focus on related evidence and empirical studies rather than explaining Keynesian Economics. This adjustment ensures that our discussion is directly pertinent to our research objectives and provides new insights. Consequently, sections 2.2 through the first two paragraphs of section 2.8, which contained general discussions commonly known among economists, have been removed. This streamlining helps maintain a clear focus on relevant literature, thereby strengthening our argument. We also acknowledge that the importance of economic diversification is a well-known and widely accepted topic. Therefore, the first two paragraphs of section 2.10, which discussed this concept, have been removed. By eliminating these widely recognized points, we ensure that our literature review remains focused on presenting new insights and added value to the discussion. Additionally, we have updated the discussion of Shaheen (2019) on page 5, line 199, to include the specific findings of the study. These revisions aim to provide a more focused and evidence-based literature review, significantly improving the clarity and impact of our research. We believe these changes address your suggestions and contribute to refining our paper.

Comment 2: Methodology lacks clarity. The equation of the ARDL model is missing. Moreover, it is not clear if GDP, Government Consumption, and Oil and non-oil revenues are in nominal or real values. If the former is the case, then the results will be biased. Apply real values and state that they are real to make a comparison across time.

Response 2: Thank you for your insightful feedback. We have revised the methodology section to address your concerns and improve clarity. Specifically, we have included the equation of the Autoregressive Distributed Lag (ARDL) model and clarified that GDP, Government Consumption, and Oil and Non-oil Revenues are now presented in real values. Previously, nominal values were used, which could have introduced bias into the results. By adjusting the variables to real values using the Consumer Price Index (CPI), we ensure that the data are comparable across time, thereby enhancing the robustness and accuracy of our findings. Additionally, we conducted a thorough reanalysis of the adjusted data, accounting for inflation, to enable meaningful comparisons over time.

Comment 3: Issues in Figures 1 and 2. Figure 1 plots the variables overtime, which is not needed. Figure 2 shows each variable in a plot on its own overtime and underneath a paragraph that mentions the absence of a time-trend ‘The above figures 2 show that the time series do not have a general trend over time.’. It is clear that at least GDP and GGEX have a positive time trend. The explanation underneath Figure 2 is misleading.

Response 3: Thank you for your insightful feedback regarding the issues in Figures 1 and 2. We have carefully considered your comments and made the following adjustments to address your concerns:

Firstly, we have removed the original Figure 1, which plotted the variables over time, as you indicated that it was not necessary. This helps streamline the presentation and focus on the more relevant analyses.

Secondly, we have revised the figures to ensure that each variable is plotted individually over time, providing a clear visualization of the trends for each variable. The time series plots now appear in four figures: Figure 1 displays the trends for Oil Revenues, Figure 2 shows the trends for Non-Oil Revenues, Figure 3 displays the trends for Government Consumption Expenditure (GC), and Figure 4 shows the trends for Real GDP.

The explanatory paragraph underneath these figures has been updated to accurately reflect the observed trends. The previous explanation incorrectly mentioned the absence of a time-trend, which was misleading. The revised explanation now accurately reflects these observations and avoids any misleading conclusions.

The time series plots in Figures 1, 2, 3, and 4 have been carefully examined for trends over time. The visual inspection and subsequent analysis reveal the following trends for each variable:

• Figure 1: Oil Revenuesdisplay high variability with significant fluctuations, especially during key economic events. This variability reflects the volatile nature of oil markets and their susceptibility to geopolitical and economic disruptions.
• Figure 2: Non-Oil Revenuesexhibit a generally increasing trend, reflecting efforts towards economic diversification. This positive trend indicates progress in reducing dependence on oil revenues.
• Figure 3: Government Consumption Expenditure (GC)demonstrates an overall increase over time, indicating growing government spending. This positive trend suggests that government expenditure has been consistently rising, likely in response to expanding public sector needs and economic policies.
• Figure 4: Real GDPshows a generally increasing trend over time, indicating sustained economic growth. This positive trend suggests that overall economic activity has been expanding consistently.

Thank you again for your constructive feedback, which has significantly contributed to enhancing the rigor and clarity of our analysis.

Comment 4: 30 observations will render the results bias. If you want to stick to the same time period use monthly or at least quarterly data if you want to apply the ARDL model.

Response 4: The authors acknowledge the concern regarding the limited number of observations and potential bias in the results. To address this issue and ensure the robustness of the analysis, the authors have extended the study period to begin from 1969, thereby increasing the number of observations. By extending the period, the data set now includes a longer time span, which enhances the reliability of the results and reduces the potential for bias.

Comment 5: The conclusion includes many repetitions. Do not compare the case of Venezuela to Saudi Arabia without using additional citations. Comparability is difficult without controlling for multiple other variables.

Response 5: Thank you for your valuable feedback. We have revised the conclusion section to address the issue of repetition and to provide a more concise and focused summary of our findings and recommendations. We acknowledge that comparisons between Venezuela and Saudi Arabia require controlling for multiple variables, which was not within the scope of our current analysis. Each country’s economic context, political environment, and policy frameworks differ significantly, making direct comparisons challenging without a comprehensive control for these factors.

Comment 6: -Absence of organization. Page 17 has table 4.8. Page 18 has table 4.5. Label the tables in order.

Response 6: Thank you for your valuable feedback regarding the organization of the tables. We acknowledge the issue with the incorrect labelling of tables and have made the necessary adjustments to ensure clarity and proper sequencing throughout the manuscript. We have reviewed and relabelled all tables in the manuscript to ensure they are numbered sequentially and appear in the correct order. The tables are now numbered correctly without any discrepancies.

Reviewer 3 Report

Comments and Suggestions for Authors

Line 83 – “By examining these aspects, this study aims to contribute valuable insights into the formulation of more resilient economic strategies for Saudi Arabia,” The authors must specifically specify what type of insights.

Line 526 – “To ensure analytical rigor, the study conducts various diagnostic tests including checks for serial correlation, heteroskedasticity, and stationarity of residuals. Authors must describe these diagnostic tests and their methodological basis.

Line 519 – “The independent variables specifically include Government Consumption (GC), Oil Revenues(OR), Non-Oil Revenues(NOR).” Authors must describe what they understand by non-oil revenues.

Line 5 “This study investigates the impact of General Government Final Consumption Expenditure (GC) on GDP growth in Saudi Arabia during major economic crises from 1990 to 2022,” / line 553 “In this study, the selection of variables was strategically chosen to evaluate the impact of General Government Final Consumption Expenditure (GC) on GDP during economic downturns, specifically during the 2008 Financial Crisis and the 2020 Coronavirus Pandemic.” Authors must clarify the time period of the study, whether from 1990 to 2022 or from 2008 to 2020.

Line 816 – “This deviation suggests a critical area for policy reform, highlighting the need for more robust counter-cyclical measures that not only stabilize the economy during downturns but also build fiscal buffers during periods of high oil prices. Authors must describe, specifically, which ones?

Comments for author File: Comments.pdf