Research Article |
|
Corresponding author: Mikhail V. Rodchenkov ( m.rodchenkov@gmail.com ) © 2026 Non-profit partnership “Voprosy Ekonomiki”.
This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits to copy and distribute the article for non-commercial purposes, provided that the article is not altered or modified and the original author and source are credited.
Citation:
Rodchenkov MV (2026) Impact of temperature on the reporting performance of energy sector companies. Russian Journal of Economics 12(2): 274-296. https://doi.org/10.32609/j.ruje.12.168448
|
This study examines the impact of extreme temperatures on energy-sector companies, highlighting the financial and economic consequences of these temperatures as an important aspect of financial performance analysis. The research methodology is based on the application of regression models to corporate financial statements. This research obtained statistically significant evidence of a relationship among asset structure, financial flows, and external temperature for 55 publicly traded energy companies from 20 countries. The transformation of asset and cost structures is a necessary adaptation measure for companies operating in regions with harsh climates; the results reveal a significant excess of capital expenditures over operating expenses under such conditions. This study identified changes in the structure of financial flows that increased the financial burden and reduced investment attractiveness for these companies. The J-factor’s nonlinear function reflects greater variability in operating conditions at low temperatures than at high temperatures, explaining contradictions in capitalization between regions with hot and extremely cold climates. The forced nature of the changes and the involvement of many sectoral players heighten the challenges in finding solutions to ensure market fairness and mitigate the uneven impact of climate change on businesses in colder regions. This study’s results partially explain inconsistencies in tariff setting and product cost coordination among companies in regions with extreme climates and those in more temperate areas and outline the contours of required management decisions at the global, national, and corporate levels.
The author expresses sincere gratitude to the attendees of the Lomonosov Readings in April 2025 and the 52nd EBES conference in Istanbul in July 2025 for their insightful comments. The author also thanks Professor Viktor Suyts for his continuous support and the companies that provided financial statements and valuable comments for this study.
extreme temperature, IFRS, sustainable development, climate impact, capital expenditure, operating expenses.
In recent decades, various academic studies have investigated how environmental changes affect economic and social outcomes. This growing interest is driven, on the one hand, by a significant increase in the frequency of natural phenomena with negative and even catastrophic consequences, such as floods, tsunamis, cyclones, and snowstorms (
The discussion of how production elements of the modern economy respond to climatic conditions is an important component of the academic literature (
Companies operating in regions with severe temperature conditions face higher adaptation costs, resulting in increased financial strain and less attractive financial metrics for investors due to the greater upfront investments required (
Our study examines the dominance of specific information-significant reporting indicators (ISRI) (
The academic debate on the relationship between weather effects and economic indicators addresses various issues related to identifying, accounting for, and mitigating the negative consequences of natural disasters. Consistent with the logic of identifying changes that companies undergo due to climatic factors, this process constitutes the internal basis of adaptation (
The content, pace, and cost of adaptation are context-dependent on climate-impact parameters, with temperature — economic dependence playing a decisive role (
One factor influencing adaptation costs is business size, which enables the diversification of climate risks (
Differences in business characteristics lead to significant disparities in adaptation capabilities. This suggests that technology and experience alone are unlikely to fully mitigate global economic losses (
The available data highlight several important aspects of this study’s methodological focus and the a priori perception of the phenomena considered. First, a significant portion of the empirical research focuses on assessing rising temperatures, noting that adaptation effects imply that warming can induce adaptive behavior aimed at better coping with changing climatic conditions (
Studies on the temperature — economic relationship by business size reveal a consistent pattern. Specifically, above-average temperatures increase energy costs and reduce productivity in small enterprises, with little to no effect on larger firms (
The issue of how production elements function under harsh temperatures falls largely outside the scope of existing studies. However, the geographic expansion of production activities into regions with harsh climates, especially in the energy sector, highlights the need to understand the effects of low temperatures on technological processes (
Second, research results are often considered and assessed as external effects, which imposes certain restrictions on their applicability and shapes their interpretation. Thus, some countries have successfully adapted to both hot (the United Arab Emirates) and cold (Canada) climates (
Therefore, defining and reliably assessing the nature of the relationship between severe temperatures and corporate reporting indicators — specifically the cost characteristics of production assets and the financial resources required to maintain their operability — is of significant academic and applied relevance. It is imperative to consider these aspects when making decisions related to financial flow management and selecting medium- and long-term investment targets capable of mitigating the negative impact of the J-factor and supporting corporate competitiveness.
This study’s research approach is based on a comprehensive examination of the relationship (interdependence) between reporting indicators reflecting the structure of FA and business financial flows, and the severe temperature conditions in regions where the main production activities of energy companies are located across various jurisdictions. The choice of sector is driven by stronger incentives for adaptation among companies in extractive industries because specific technological processes are required in external environments (
(a) An enterprise’s adaptation activities (related to FA, technological processes, and labor resources) require greater investments and expenses than those of competitors operating in more moderate conditions within the same sector. These investments lead to higher costs and changes in the asset structure, while adaptation-related costs alter corporate cash flows and expenditure patterns. This forced increase in investment and operating spending affects the company’s financial position, operational capabilities, investment attractiveness, and overall competitiveness.
(b) The specificity of severe temperatures lies in the predominant increase in FA costs, which primarily leads to changes in the structure and volume of expenses. Extremely cold climates require strict design standards for permanent facilities and shortened construction periods, discouraging an ad hoc approach and establishing a foundation for the long-term use of such infrastructure. This situation contributes to a relative reduction in future operating expenses. In other words, the share of capital expenditures (CapEx) is higher and the share of operating expenses (OpEx) is lower for companies operating in harsh climates than for those in regions with more favorable temperature regimes. Therefore, the net value of FA is used to assess changes in the structure of tangible assets. This indicator comprises a significant portion of the balance sheet for industrial companies, especially those involved in mining and processing in the energy sector. Changes in FA value significantly affect such businesses’ financial models. Accrued depreciation is used to assess changes in the cost structure associated with the use of FA, taking adaptation into account. Additionally, the amount of capital included in the analysis is both the initial source of CapEx and an important factor in the parameterization of the adaptation process.
(c) Changes in business financial flows are assessed using expense measures, unlike previous studies that used income as a target indicator (
(d) Panel data from official corporate financial statements are used for the analysis. These data are considered more reliable because independent professional auditors verify their accuracy. Additionally, discrete financial data are more effective for identifying and evaluating economic features, specifically changes in the structure of reporting indicators under the influence of the temperature factor (J-terms). Geoclimatic diversification was applied by temperature conditions to improve the validity of the results.
(e) Following previous research on temperature — economic dependence (
(f) Double testing ensures the reliability of identifying the relationship between FA value and temperature. Forward testing treats the severe temperature variable as the primary independent variable, and its impact on FA value is analyzed. Backward testing evaluates the extent to which changes in cost indicators (predictors) correspond to severe temperatures.
The empirical strategy for regression modeling involves initially constructing a model with the maximum set of independent variables. This step is followed by backward selection to isolate the most significant predictors for the dependent variable; thus, the research approach effectively addresses the study’s objectives and objectively validates the research hypothesis.
The research followed a systematic approach to address the outlined objectives. The research stages included the following:
(1) To form an array of primary data based on official corporate reports;
(2) To develop initial assumptions and select reporting indicators that reflect material and cost characteristics of business operations;
(3) To conduct regression and multivariate analyses of the relationship between the selected reporting indicators and the severe temperature conditions typical of the sampled companies’ production areas;
(4) To present conclusions and recommendations based on the results of hypothesis testing.
The research methodology is based on regression and multivariate analysis, providing clear algorithms for processing temperature — economic relationship data (
A specific form of GLM (binomial regression) was employed to assess the probability of severe temperature conditions in response to changes in cost indicators. The limitations typically associated with multiple regression analysis for such tasks are addressed by reframing the problem. Rather than predicting a binary outcome, the model estimates a continuous dependent variable with values in the interval [0, 1] based on given values of the independent variables. This approach uses the logit transformation (1):
P = 1/(1 + e – y), (1)
where P is the probability of event occurrence (correspondence to severe temperatures), e is the base of the natural logarithm (2.718…), and y is the standard regression equation.
If Pi reflects the probability of success for the i-th observation, i = 1, ... , N; Xβ is a linear predictor, the link function connects the covariates of each observation to its corresponding probability via the linear predictor. In logistic regression, the logit link (2) is used:
Xβ = ln(P/(1 – P)). (2)
The regression coefficient (βk) represents the change in the logarithm of the odds associated with a one-unit change in the value of the Xk covariate; thus, exp(βk) is the odds ratio associated with a one-unit change in Xk. Presenting results as odds ratios makes them easier to interpret. If exp(βk) > 1 (equivalently, βk > 0), then the odds of the outcome increase as the predictor variable increases. If exp(βk) < 1 (equivalently, βk < 0), then the odds of the outcome decrease as the predictor variable increases.
The target indicator’s predicted probability (marginal effect) was calculated for each independent variable to refine the boundaries and strength of the relationship between the severe temperature indicator and the explanatory variables, assuming that mean values of all other variables in the model were held constant. To evaluate the regression results, a statistical significance level of α = 0.1 was set, ensuring the required degree of reliability (
The regions were classified by temperature indicator, considering the following criteria. First, following the methodology of
Second, according to publicly available data on average annual temperatures provided by international aggregators
Given the overlap between category three and the characteristics of cold regions, all sample elements were classified into three categories to streamline the structuring of elements within the sample, recognizing that the lower limit of severe temperatures is substantially wider than the upper limit (J-terms).
This study’s dataset consists of financial reporting indicators for the public companies included in the sample. The sources of the values were statements of financial position and statements of comprehensive income prepared in accordance with IFRS or US GAAP; if these were unavailable, we used financial statements prepared according to national standards.
The research objects were public companies in the energy sector engaged in hydrocarbon production, transportation, trade, and processing. Our study focuses on sectoral public companies due to their higher discipline in reporting and disclosure, driven by regulatory frameworks and the requirements of organized market administrations. Furthermore, focusing on a single sector helps minimize the risk of bias in estimates commonly observed in cross-sectoral studies (
The study utilized corporate reporting data from 2019 and 2020. These periods were selected to evaluate the relationships under normal (growth) market conditions and during a downturn caused by the COVID-19 pandemic. We aggregated the data from official corporate disclosure resources and internationally recognized financial and economic data aggregators, including Bloomberg and Yahoo Finance.
It is standard practice to include data from companies in various economies. For example, a study on the relationship between temperature and income included data from 12 countries (
This study constructed a conceptual model for the relationship between the severe temperature indicator and corporate financial reporting indicators based on modified models of the relationship between income and temperature (
(1) The climatic variable (severe temperature indicator TemperatureCold);
(2) Indicators (log-transformed using the natural logarithm) of corporate financial statements reflecting the structure of assets and financial flows most sensitive to adaptation processes:
(3) Indicators of ISRI dynamics — deltas for the period from 2019 to 2020:
ΔOpExln = OpEx2020ln − OpEx2019ln;
ΔDD&Aln = DD&A2020ln − DD&A2019ln;
ΔSEln = SE2020ln − SE2019ln;
(4) An indicator capturing the vector of unaccounted variables (cumulative error).
The conceptual assumptions were built by considering the importance of ISRI in assessing the relationship of indicators with severe climatic conditions.
(a) Companies planning to operate in harsh temperature conditions must consider increased requirements related to equipment and workforce performance when forming and implementing capital investments (CapEx) in the necessary equipment and production assets. The value of FA serves as an aggregate reflection of these investments and is reported accordingly. A change in FA value (PP&E) and its correlation with the severe temperature indicator can be used to assess the validity of the first part of the hypothesis.
(b) As the primary source of initial capital investments (CapEx), a high volume of capital (SE) offsets the impact of harsh temperatures on the value of FA (PP&E) by enabling higher investment levels regardless of climatic conditions. Consequently, the predictive value of capital is assumed to be low, making it reasonable to exclude this indicator in subsequent model constructions.
(c) The operating expenses indicator (OpEx) is more important during a company’s operational phase, as the impact of severe temperature conditions is assumed to have been addressed during the design and preparation stages. Moreover, the relevance of operating expenses is reinforced by including FA depreciation. This inclusion justifies the exclusion of accrued depreciation as a separate variable during the systematic reduction of predictors.
(d) Depreciation charges (DD&A) are a derivative of the value of FA but allow assessment of changes under the influence of temperature; however, given their inclusion in operating expenses, they can be excluded from further model constructions in favor of OpEx.
According to this research approach, two models were developed for “forward” and “backward” testing.
Following the above research approach, the conceptual model with a complete set of predictors is described by equation (3):
PP&Eln = β0 + βi1 OpExln + βi2 DD&Aln + βi3 Temperature3 + βi4 SEln +
+ (βi5 ∆SEln + βi6 ∆DD&Aln + βi7 ∆OpExln) + ε, (3)
where β0 is the constant; βij is the variable coefficient; ε is the vector of omitted variables (error term).
The model includes (Table
| Variable | Obs. | Mean | Std. dev. | Min | Max |
|---|---|---|---|---|---|
| 2020 | |||||
| PP&Eln | 55 | 7.661 | 1.401 | 4.708 | 10.085 |
| OpExln | 55 | 7.608 | 1.268 | 4.389 | 9.960 |
| ΔOpExln | 55 | −0.080 | 0.121 | −0.376 | 0.113 |
| ΔDD&Aln | 55 | 0.235 | 0.287 | −0.292 | 1.262 |
| ΔSEln | 55 | 0.130 | 0.218 | −0.309 | 1.088 |
| Temperature3 = 1 | 55 | 0.455 | 0.503 | 0 | 1 |
| Temperature3 = 3 | 55 | 0.291 | 0.458 | 0 | 1 |
| 2019 | |||||
| PP&Eln | 55 | 7.518 | 1.391 | 4.503 | 9.693 |
| OpExln | 55 | 7.687 | 1.279 | 4.389 | 9.993 |
| Temperature3 = 1 | 55 | 0.455 | 0.503 | 0 | 1 |
| Temperature3 = 3 | 55 | 0.291 | 0.458 | 0 | 1 |
Finally, the model includes a severe temperature conditions indicator, reflected by the ordered variable Temperature3, which has three ordered values: 1 = comfortable, 2 = moderate (average), and 3 = extreme (severe conditions). Temperature3 = 2 (moderate temperatures) is assigned as the base category in this model. This approach allows us to control the influence of changes in temperature conditions on changes in the dependent variable, PP&Eln.
The set of coefficients (βi5, βi6, and βi7) for the ISRI dynamics variables reflects the direction of change in the variables, allowing assessment of the emerging trend in the transformation of the structure of assets and financial flows.
The descriptive statistics indicate that the ISRI dynamics variables are not normally distributed (see Table
The OLS regressions built to test the conceptual assumptions (Table
| Variable | Conceptual (preliminary) | |
| OLS1 | OLSBeta | |
| OpExln | 0.974 | 0.945*** |
| DD&Aln | 0.386*** | |
| SEln | 0.494*** | |
| Temperature3 = 1 | −0.005 | 0.228 |
| Temperature3 = 2 | base | base |
| Temperature3 = 3 | 0.152 | 0.665** |
| ΔOpExln | 0.750 | |
| ΔDD&Aln | 0.245 | |
| ΔSEln | 0.001 | |
| const | 1.112** | 0.173 |
| Obs. | 55 | 55 |
| aic | 71.429 | 136.999 |
| bic | 83.473 | 151.050 |
| rank | 6 | 7 |
The results of the OLSBeta model (see Table
The model aims to identify changes in the probability of severe temperatures in response to changes in cost indicators reflected by the predictors. To test the relationship between severe temperatures and a company’s material and cost indicators using the backward method, TemperatureCold is specified as the dependent variable and coded as a categorical variable (1 = “severe temperatures,” 0 = “other, non-severe conditions”), reflecting compliance with the target climatic conditions.
The model with a complete set of predictors for the conceptual regression TemperatureCold is described by equation (4):
TemperatureCold = β0 + βi1 OpExln2 + βi2 DD&Aln2 + βi3 PP&Eln2 +
+ βi4 SEln2 + ε, (4)
where β0 is the constant; βij is the variable coefficient; ε is the residual (error term).
According to the conceptual assumptions, the model includes (Table
| Variables | Obs. | Mean | Std. dev. | Min | Max |
|---|---|---|---|---|---|
| 2020 | |||||
| TemperatureCold | 55 | 0.286 | 0.456 | 0 | 1 |
| OpExln2 | 55 | 59.457 | 18.935 | 19.262 | 99.161 |
| DD&Aln2 | 55 | 30.847 | 15.730 | 4.849 | 68.370 |
| PP&Eln2 | 55 | 60.614 | 20.657 | 22.167 | 101.708 |
| SEln2 | 55 | 58.206 | 20.286 | 18.102 | 99.824 |
| 2019 | |||||
| TemperatureCold | 55 | 0.291 | 0.458 | 0 | 1 |
| OpEx2019ln2 | 55 | 60.702 | 19.328 | 19.260 | 99.868 |
| DD&A2019ln2 | 55 | 28.335 | 14.754 | 4.502 | 57.675 |
| PP&E2019ln2 | 55 | 58.425 | 20.199 | 20.277 | 93.949 |
The constructed model uses binary regression (binreg) with TemperatureCold = 1 as the event. A step-by-step change in the set of predictors (squared natural logarithms) is used to increase the contrast of the estimates.
Testing with a systematic reduction in the set of regressors confirms the validity of the initial assumptions (Table
| Base | 2020 | 2019 | |||||||
| TemperatureCold | BinReg1 | BinReg2 | BinReg2019 | ||||||
| Scale | Odds ratio | Coef. | Odds ratio | Coef. | Odds ratio | Coef. | |||
| OpExln2 | 0.920** | −0.083** | 0.927* | −0.076* | 0.929** | −0.073** | |||
| DD&Aln2 | 0.851* | −0.161* | 0.857* | −0.155* | 0.877* | −0.131* | |||
| PP&Eln2 | 1.097 | 0.092 | 1.153** | 0.142** | 1.136** | 0.128** | |||
| SEln2 | 1.064 | 0.062 | |||||||
| const | 0.510 | −0.674 | 0.519 | −0.655 | 0.586 | −0.535 | |||
| df | 50 | 51 | 51 | ||||||
| Obs. | 55 | 55 | 55 | ||||||
The odds ratio for PP&Eln2 is positive in the models with the adjusted predictor set based on 2020. This outcome indicates that, for a one-unit increase in the predictor, the odds of severe temperatures increase by about 15% (OR = 1.153; P > |z| = 0.013; coef = 0.142). The odds ratios in the BinReg2 model for OpExln2 (OR = 0.927) and DD&Aln2 (OR = 0.857) indicate that the odds of severe temperatures decrease as these variables increase by one unit, holding the remaining variables constant. That is, increasing OpExln2 by one unit in the 2020-based models is associated with a ~7% decrease in the odds of severe temperature conditions (coef = −0.076). A similar one-unit increase in DD&Aln2 reduces the odds by about 15% (coef = −0.155).
These trends also occur in the models with an adjusted predictor set based on 2019 (BinReg2019), confirming the stability of the identified effect under different economic conditions. Nevertheless, given that depreciation amounts are aggregated within operating expenses, it is advisable to reduce the number of predictors for TemperatureCold to two, reflecting normalized operating expenses (OpExln2) and FA cost (PP&Eln2). As a result, an optimal model (BinRegOptimDM) was obtained for the basic testing.
The OLS regression models with PP&Eln as the dependent variable showed satisfactory results (Table
| Variable | 2020 | 2019 |
| OLS OptimDM | OLS OptimUM | |
| OpExln | 0.952*** | |
| Temperature3 = 1 | 0.154 | 0.262 |
| Temperature3 = 2 | base | base |
| Temperature3 = 3 | 0.574* | 0.626** |
| OpEx2019ln | 0.912*** | |
| const | 0.18 | 0.204 |
| Obs. | 55 | 55 |
| aic | 132.255 | 134.099 |
| bic | 148.284 | 142.128 |
| rank | 4 | 4 |
Based on the results of the preliminary OLS models, with step-by-step exclusion of variables, an optimal downturn-period model (OLSOptimDM) was obtained according to the conceptual assumptions. Its regressors are operating expenses (OpExln) and severe temperatures (Temperature3 = 3). The results for OLSOptimDM reflect a statistically significant positive relationship between FA cost and OpExln and Temperature3 = 3. Severe temperatures (Temperature3 = 3), compared with moderate temperatures (Temperature3 = 2; baseline), statistically significantly contribute to an increase in FA cost (β = 0.574; p < 0.1).
We also constructed a model with fixed effects (robust). Based on bootstrap results for 150 replications, the robustness of the model and the significance of the influence of the selected regressors on the dependent variable were confirmed for the 2019 upturn-period model (OLSOptimUM). The constructed and tested models passed the post-estimation regression tests. They did not reveal statistically significant evidence of specification errors, heteroscedasticity, multicollinearity, abnormal skewness, or excess kurtosis. Thus, the results of constructing OLS regressions for PP&Eln with a systematic reduction in the set of independent variables allowed us to:
(a) Reduce the number of predictors to two significant ones — the operating expenses indicator and the severe temperature conditions indicator (OpExln and Temperature3 = 3);
(b) Confirm the correctness of the conceptual assumptions: severe temperatures (Temperature3 = 3) increase FA cost relative to moderate temperatures in the operating areas of sectoral companies.
When expressing predictors as squared natural logarithms, which increases the contrast of the estimates, the specification quality of the regressions is higher. The constructed models were tested for heteroscedasticity, omitted variables, and specification errors (Table
| Base | 2020 | 2019 | ||||
| TemperatureCold | BinRegOptimDM | BinRegOptimUM | ||||
| Scale | Odds ratio | Coef. | Odds ratio | Coef. | ||
| OpExln2 | 0.904*** | −0.101* | 0.916** | −0.087** | ||
| PP&Eln2 | 1.063* | 0.061* | 1.056* | 0.055* | ||
| const | 3.314 | 1.198 | 2.751 | 1.012 | ||
| df | 52 | 51 | ||||
| Obs. | 55 | 55 | ||||
In the optimal model based on 2020, the odds ratio for PP&Eln2 is greater than 1, indicating that a one-unit increase in the predictor is associated with an approximately 6% increase in the odds of severe temperatures (OR = 1.063; P >|z| = 0.059). The coefficient for OpExln2 (OR = 0.904) indicates that the odds of severe temperatures decrease as OpExln2 increases by one unit, holding all other variables constant. That is, increasing OpExln2 in the 2020-based optimal model is associated with a 10% decrease in the odds of severe temperature conditions for the company (coef = −0.101). The same patterns occur in the 2019-based optimal model (BinRegOptimUM), confirming the stability of the identified effect under different economic conditions.
The identified patterns in the change in the odds of severe temperature conditions are confirmed by calculating marginal effects for each level of ISRI values: FA cost, operating expenses, and depreciation. Thus, the highest predicted probabilities of TemperatureCold (0.974) for nine levels of the composite variable (OpExln2#PP&Eln2) were obtained for level 3 (maximum FA cost (PP&Eln2 = 90) and minimum operating expenses (OpExln2 = 30; Fig.
Forecast average probabilities of severe temperatures by levels of the composite variable OpExln2#PP&Eln2.
Source: Author’s calculations.
Similar constructions for the composite variable DD&Aln2#PP&Eln2 indicate statistically significant results for nine levels out of the 18 preset ones. The highest probabilities occur at levels 3, 6, and 9 at the maximum FA cost (90) and three minimum allocated depreciation levels (10, 20, and 30, respectively). Margin3 = 0.994; Margin6 = 0.973; Margin9 = 0.903 (Fig.
Forecast average probabilities of severe temperatures by levels of the composite variable DD&Aln2#PP&Eln2.
Source: Author’s calculations.
Marginal effects were calculated for FA cost and operating expenses with an increased number of preset levels (to enhance the analytical detail of testing) to clarify the distribution boundaries of the odds of severe temperatures given the allocated change in isolated ISRIs.
The estimated odds of severe temperatures for companies increase with FA cost and decrease as operating expenses increase (Fig.
Predicted probabilities (TemperatureCold = 1) by FA cost levels (PP&Eln2) and operating-expense levels (OpExln2).
Source: Author’s calculations.
The identified characteristics of the relationship between FA cost and operating expenses remain stable, even when considering the isolated influence of each indicator across all predefined levels of the analyzed variables.
Our study’s results support the following conclusions.
Severe temperatures are an important factor influencing the structure of physical and financial business indicators. Operating under such conditions increases companies’ financial investments in fixed assets (FA), reflected in higher FA values reported in corporate financial statements.
The shift in predicted probabilities toward lower operating expenses and annual depreciation values (based on the estimated relationship between the severe temperature regime and the selected ISRI) supports several key conclusions. First, this shift reflects a structural change in corporate costs. An increase in capital investments accompanied by a relative decrease in operating expenses confirms the validity of the second part of the research hypothesis. Second, CapEx typically exceeds OpEx by several multiples, resulting in a persistently elevated financial burden on businesses operating in severe temperature zones. This situation creates an intra-sectoral imbalance in the objective assessment and reporting of business efforts toward achieving SDGs. It also affects the comparability of reported financial indicators for companies operating under different climatic conditions. Third, the relationship between operating expenses and severe temperatures may also be indirectly influenced by other contextual factors, such as dividend policies, tax regimes, government support measures, and corporate strategies for maximizing profitability. Nevertheless, the study’s results provide strong evidence of a systematic influence of temperature on the structure of corporate assets and financial flows.
The results outline a broader problem of increasing financial pressure on companies primarily engaged in production activities, specifically hydrocarbon extraction, processing, and primary transportation. Many of these companies are subject to the most significant impact of severe temperature conditions (the J-factor). Without a well-developed and tested methodology for incorporating such climate-related risks into financial reporting, these companies will likely be required to allocate additional reserves. This allocation may, in turn, adversely affect key financial ratios and reduce their investment attractiveness compared to other market participants. Addressing these issues may require revising the coefficients, standards, and reserve thresholds currently used in reporting. Introducing progressive scales that account for the disproportionate burden faced by “producers,” as compared to “intermediaries,” in the global hydrocarbon market could help mitigate this form of systemic “social injustice.”
In corporate reporting, the International Accounting Standards Board has expanded initiatives to improve the accounting and disclosure of climate risks, including temperature risk; however, these initiatives are lagging.
In contrast, according to the Global Energy Monitor, the estimated cost of building a pipeline of comparable length (about 1,100 km) in Argentina is $4.1 billion (
The global nature of this problem necessitates a separate study to explore approaches for its effective resolution. The initial conditions for developing a solution are as follows. First, the interests of consumers and producers must be distinguished. Second, all stakeholders must understand that the impact of climate on the economy is objectively real and varies significantly. Third, the super-marginal nature of energy resource production under favorable conditions predetermines a significant competitive advantage for residents of certain economies compared to industry competitors with high production thresholds. Coordinated solutions that consider the interests of all industry participants are required (OPEC+ is a clear example of international coordination).
Conceptually, the solution’s elements should include (a) global industry regulation measures, (b) easing the financial burden at the corporate level, and (c) increasing target companies’ investment attractiveness.
In this context, global industry regulation requires that J-terms be considered in the discussions and decisions of OPEC+, the IEA, BRICS+, and the G7. The WTO could serve as a logical communication channel, given the desire of many G20 economies to revive its functionality.
A framework for business — government relations focused on the accurate assessment and allocation of responsibility for the economic consequences of climate and environmental risks affecting businesses in extreme conditions could be a useful solution. Moreover, a global industry rating for the geoclimatic gradation of production could be used to increase transparency at the international level. This unified rating would serve as the basis for industry benchmarking and would allow for the identification and uniform assessment of the industry’s cost of J-term variability. Such data can help adjust climate and environmental risk assessment models. Companies should conduct climate risk assessments using standard industry methods. Expanding the use of agent-based models appears beneficial for improving the quality of modeling.
At the national level, solutions that ensure a timely response to challenges are appropriate. The development of a system of indicators to identify elevated financial risks in mining and production companies could be a key element of such a system. The excess of CapEx over OpEx relative to the estimated J-factor impact, based on a unified benchmark, could serve as a key indicator. This approach would help minimize environmental risks associated with abandoned real estate, structures, and equipment, similar to the abandonment of Arctic development assets following the collapse of the USSR. Comparable cases have been observed in the Canadian North and Alaska (
Implementing the proposed measures may require amendments to IASB and ISSB standards. The estimated scope of the required additions and clarifications covers at least three standards. Thus, it would be appropriate to supplement IFRS 1, “First-time Adoption of International Financial Reporting Standards,”
Following the guidelines of the Basel Committee on Banking Supervision, the financial burden on target companies could be eased by introducing reduced risk ratios for industry manufacturers and increased risk ratios for intermediaries when calculating the capital adequacy ratio of credit institutions.
The measures described should be considered conceptual and potentially feasible, requiring consideration of the national economic situations of relevant companies. Although they are not comprehensive recommendations, their practical utility lies in defining the contours of possible systemic management decisions.
The results are based on a sector-specific sample. As such, the findings are open to debate and may be confirmed or challenged through research using different sample characteristics or alternative methodologies. Nonetheless, the risk of a Type I error appears to be minimal.
Our study’s results confirm the hypothesis regarding the relationship between the high cost of fixed assets (FA) and severe ambient temperatures in the areas where energy-sector companies primarily operate. Severe temperature conditions are a significant adaptation factor and contribute to changes in the structure of assets and business financial flows, specifically an increase in FA costs with a relative decrease in operating expenses and depreciation. This pattern is not typical for companies operating under moderate or favorable temperature conditions.
The results highlight a broader issue of intra-sectoral imbalance in assessing company efforts to achieve SDGs. They also highlight the limited comparability of financial reporting indicators between companies operating in harsh climates and those in more temperate regions. The analysis revealed a persistently elevated financial burden borne by companies conducting production activities under severe temperature conditions. This problem requires urgent resolution at the sectoral level on a global scale and is particularly important given the widespread presence of the J-factor among many industry players and its global implications.
These results can provide a foundation for further research into external environmental factors. The findings can also support the development of a well-substantiated position for protecting national business interests in global markets. Additionally, they provide a basis for creating sector-specific comparison algorithms that promote a fair assessment of corporate efforts to achieve SDGs while helping mitigate the J-factor’s negative impact at the company level.
The author expresses sincere gratitude to the attendees of the Lomonosov Readings in April 2025 and the 52nd EBES conference in Istanbul in July 2025 for their insightful comments. The author also thanks Professor Viktor Suyts for his continuous support and the companies that provided financial statements and valuable comments for this study.