The Response of Global Oil Inventories to Supply Shocks

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The paper extend the GVAR to 36 economies, add an OECD crude-inventory series and a critical-minerals price index, and updates the sample to 2022 Q2.

The introduction provides only cursory theoretical context on how inventories mediate price shocks. Clarifying the conceptual framework would strengthen the background section.

The claim to be the first GVAR with global & country-specific inventories is unsubstantiated. Re-position the contribution by: (a) surveying recent inventory-inclusive VARs/GVARs (e.g. Kim et al. 2020; Zhao et al. 2023); (b) state explicitly what is incremental versus novel.

Table 2 shows that only the global shock under tight conditions achieves significance at the 90 % level; most country shocks are insignificant. Discuss statistical power and possible over-parameterisation.

the foreign variable trade weights are anchored on the pandemic years 2019-21 (t2019,2020,2021) (Eq 5B), which may bias spill-overs.

Strengthening robustness checks and clarifying which findings are genuinely supported would align the conclusions more closely with the results.

The manuscript is promising, principally because of its updated dataset and policy-oriented scenarios, but the empirical evidence is not yet convincing, and several modelling choices require justification. 

Author Response

• The introduction provides only cursory theoretical context on how inventories mediate price shocks. Clarifying the conceptual framework would strengthen the background section.

 

We have addressed this issue by expanding the introduction to include the following points:

 

Herrera (2018) describes the basic transmission mechanism.  Industry-level inventories and sales often respond faster to an oil price shock than gross domestic product (GDP).  The initial price shock acts as a negative demand shock which catches firms by surprise resulting in higher than anticipated inventory builds.  Firms desire to smooth their output and use existing inventories to spread production out over time, thereby delaying the reduction in GDP  The intuition underlying the basic theory is that, given a non-zero and negative price elasticity of the demand for oil, information concerning a future shortfall of oil relative to demand will cause an increase in the demand for oil inventories, and hence the real price of oil (Kilian and Murphy, 2014).  At the same time, knowledge of oil supplies heading into inventories today will reduce the rational expectations of oil prices next year (Hamilton, 2009).  The relationship is so well established that a simple announcement of a change in inventories can be shown to have significant, and lasting implications for world oil prices. Bu (2014) investigates the effect of EIA inventory reports on crude oil futures prices and finds that negative shocks to inventory information lead to an increase in crude oil prices.

 

• Bu, Hui. “Effect of Inventory Announcements on Crude Oil Price Volatility.” Energy Economics 2014, 46 (November): 485–94. https://doi.org/10.1016/j.eneco.2014.05.015.
• Hamilton, James. “Understanding Crude Oil Prices.” The Energy Journal 2009, 30 (2): 179–206.
• Herrera, Ana Maria. “Oil Price Shocks, Inventories, and Macroeconomic Dynamics.” Macroeconomic Dynamics 2018, 22(3). https://doi.org/10.1017/S1365100516000225.

 

 

• The claim to be the first GVAR with global & country-specific inventories is unsubstantiated. Re-position the contribution by: (a) surveying recent inventory-inclusive VARs/GVARs (e.g. Kim et al. 2020; Zhao et al. 2023); (b) state explicitly what is incremental versus novel.

 

We substantiate the claim that “To the best of our knowledge, the model we construct for this study is the first one that includes both global and country-specific oil inventories in a GVAR model” by incorporating the following revisions:

 

Kim et al. [26] use a GVAR framework to establish that the buffer responses of inventories – globally, in the U.S., and in Cushing, Oklahoma – tend to be immediate for demand shocks but gradual for supply shock. Zhao et al. [27] also utilize the U.S. oil stocks as a benchmark for the inventory variable but find that the buffering effect of inventory tends to appear in the long-term, while the short-term dynamics is primarily driven by the speculation effect. The authors use a hybrid Wavelet-ARDL-SVR (WAS) model to demonstrate that speculation on the supply side is more likely to cause market risk.  Besides the lack of consensus on the drivers and time horizon of the buffer and speculative patterns of oil inventories, contemporary modelling literature on the subject is limited in the geographical or country coverage – primarily focusing on the U.S. or global stocks. However, as shown in earlier studies, countries have demonstrated varied approaches to managing their inventories, especially in response to oil market shocks [7].

 

“To the best of our knowledge, the model we construct for this study is the first one that includes both global and country-specific OECD oil inventories in a GVAR model”

 

• Table 2 shows that only the global shock under tight conditions achieves significance at the 90 % level; most country shocks are insignificant. Discuss statistical power and possible over-parameterization.

 

 

Power: We have bootstrapped the results to assess how likely it is to detect significant transmission paths. Wide or unstable intervals indicate low power. These are clearly indicated.

Over-parameterization Diagnostics: We have used the AIC tests to determine lang length for each country specific VAR. These are reported in the Appendix. We also report rank weak exogeneity Likelihood ratio F tests comparing restricted and unrestricted models. This provides verification that the foreign variables are weekly exogenous in every block. Residual diagnostics and normality tests have been performed on the analysis.

 

The results of all of these tests are reported in Online Appendix A. Unit root, weak exogeneity, and structural stability tests. The section of lag orders, cointegrating relationships and persistence profiles is presented in Online Appendix B.

 

 

• The foreign variable trade weights are anchored on the pandemic years 2019-21 (t2019,2020,2021) (Eq 5B), which may bias spill-overs.

 

This is an excellent point, thank you so much for bringing it up. In fact, given the three-year calculation window for the trade weights, and volatility of the world oil market, there are relatively few options that do not suffer from major disruptions. These include but are not limited to the 1990 Iraq invasion of Kuwait, the 1997 Financial crisis, 1998 Asian financial crisis and OPEC production increase, 2001 September 11, 2003 Iraqi invasion of Kuwait, 2005 Hurricane Katrina and Rita, 2008 Financial crisis, 2011 Arab Spring, 2014 US Shale shock, 2016 price collapse, 2020 Covid, 2022 Russia invasion of Ukraine.

 

To address these concerns, we have added a footnote:

 

It is important to notice that the trade weights for the loose market scenario were calculated using the trade flows from 2019-2021, through Covid, U.S. China trade tensions and deglobalization and the Russian invasion of Ukraine.  Trade weights for the loose market scenario, on the other hand were calculated using trade weights from 2015-2018, which experienced the rise of US Shale, and 2016 price collapse.   While the spillovers may be over or underestimated relative to ‘todays’ market, the weights were specifically designed to emulate conditions in two different states of the market--tight and loose markets—and not to represent or reflect current interdependencies.

 

In addition, while we do not utilize time-varying trade weights in this specification of our model, we do run a similar scenario --Tight markets—using trade weights from 2015-2018—and note that there is no appreciable difference in the power of the tests.

 

 

• Strengthening robustness checks and clarifying which findings are genuinely supported would align the conclusions more closely with the results.

 

As mentioned above, we have bootstrapped the results to assess how likely it is to detect significant transmission paths.  Wide or unstable intervals indicate low power.  These are clearly indicated, and the bootstrapped confidence intervals are clearly within an acceptable range.  To be specific, clearly within a relative width of plus or minus 2-3 times the point estimates, sign stability (no frequent crossing of the horizontal axes), fast convergence of the bootstrap.

 

“The statistical significance of the findings is clearly presented in Tables 2 and 3, and we note  However, as shown in Table 2, the simulation results are not statistically significant for the world and most countries under investigation.  Interestingly, the results for the US and South Korea are significant at the 90% confidence level…. It should be noted that the results are statistically significant in only two cases, Iran (tight market scenario) and Mexico (loose market scenario). The model output demonstrates that such varied effects can’t be solely explained by any potential single factor, such as the country’s total output, its share in oil trade flows, or current market conditions. Rather, the output reflects the complex interconnections and interdependencies captured by the GVAR modeling approach.”

 

 

• The manuscript is promising, principally because of its updated dataset and policy-oriented scenarios, but the empirical evidence is not yet convincing, and several modelling choices require justification. 

 

Our claim has been clarified by the addition of the following We extend the temporal coverage to 2022Q2, making our model one of the most up-to-date open-access GVAR models at the time of writing.   And the addition of the following footnote: The open access GVAR Modelling site has an updated data set to 2023Q3 (Smith, 2025)

Reviewer 2 Report

Comments and Suggestions for Authors

This paper explores asymmetric shocks in oil markets under tight or loose market conditions and find the magnitude of and significance of oil inventory responses under tight conditions greatly outsize the responses to the same shocks under looser conditions. 

I find the novelty and relevance of the paper to be of interest to readers and researchers. Furthermore, the empirical results seem sensible and sound given the authors' identification strategy. 

My only comment would be that the presentation of results could be slightly improved. The tables and figures are presented in a way that could be somewhat off-putting to readers. Table 2 in particular could be improved/rectified to be more aesthetically appealing and professional. The Figure quality also seems a bit off for Figures 1a and 3, where parts of the black border on the righthandside seem to "disappear." Fixing the or standardizing the Figure dimensions would improve readability. 

Author Response

This paper explores asymmetric shocks in oil markets under tight or loose market conditions and find the magnitude of and significance of oil inventory responses under tight conditions greatly outsize the responses to the same shocks under looser conditions. 

 

I find the novelty and relevance of the paper to be of interest to readers and researchers.

 Furthermore, the empirical results seem sensible and sound given the authors' identification strategy. 

My only comment would be that the presentation of results could be slightly improved. The tables and figures are presented in a way that could be somewhat off-putting to readers. Table 2 in particular could be improved/rectified to be more aesthetically appealing and professional.

 

The Figure quality also seems a bit off for Figures 1a and 3, where parts of the black border on the right-hand side seem to "disappear." Fixing the or standardizing the Figure dimensions would improve readability. 

 

We have updated the charts and tables improving their quality and visibility.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

I have no further comments