Energy-related Contextual Determinants of Climate Change Reported to the CDP: Evidence From BRICS Nations

I. INTRODUCTION

In this paper, we examine how energy-related contextual determinants influence the climate change reporting of Forbes Global 2000 companies from BRICS (Brazil, Russia, India, China, and South Africa) nations to the Carbon Disclosure Project (CDP). BRICS nations bring together the major emerging economies that have been the engines of global economic growth. This study examines the status and level of CDP reporting based on national regulations, the Paris Climate Agreement, and energy-related contextual determinants.

Corporate disclosure studies have used institutional theory to understand how companies respond to external processes built on the assumption regarding how structures, such as schemes and rules, become authoritative guidelines for firm behavior (DiMaggio & Powell, 1983; Scott, 2005). Yang and Farley (2016) investigated how international and domestic guidelines influenced Chinese companies’ climate change reporting. They found that Chinese national reporting guidelines have a greater impact than international guidelines. Narayan and Smyth (2007) studied the stationarity of per capita energy consumption, and Narayan et al. (2016) found evidence for an environmental Kuznets curve in 181 countries. Grauel and Gotthardt (2016) found that country-level environmental regulations and national contexts determine the voluntary carbon reporting of large companies.

This study looks at major environmental reporting regulations in BRICS as follows: (a) Brazil: resolution on Socio-environmental Responsibility Policy (2014); (b) Russia: no environmental legislations on carbon emission disclosure; (c) India: Business Responsibility Report (2012); (d) China: National Development and Reform Commission–mandated greenhouse gas (GHG) reporting (2014); and (e) South Africa: National Greenhouse Gas (GHG) Emission Reporting (2017).

Based on the above discussion, we hypothesize that (a) different energy-related national contextual and regulatory regimes influence the CDP reporting status of companies in BRICS; and (b) different energy-related national contextual and regulatory regimes influence the extent of the CDP disclosures by companies in BRICS.

These hypotheses are important because organizations are increasingly concerned about climate change risks to maintain their license to operate and to ensure their long-term success in a competitive business environment, as well as comply with national or regional policies aimed at reducing GHG emissions. Developed countries introduced mandatory corporate GHG reporting regulations in the early 2000s (Brouhle & Harrington, 2010; Freedman & Park, 2017; Stolaroff et al., 2009; West & Peña, 2003), but emerging economies followed with increased economic activity and resultant emissions.

Our study attempts to understand the underlying energy-related contextual factors determining corporate reporting and disclosure on climate change aspects to the CDP by companies in BRICS. Voluntary frameworks/guidelines positively influence higher-level categories of disclosure, while mandatory regulations negatively influence them. The Paris climate agreement has influenced companies’ reporting status and all levels of CDP reporting. This study adds to the literature in two ways. First, it examines country-specific energy-related contextual determinants to investigate CDP reporting in the context of emerging economies. Second, this research validates institutional theory in the context of CDP reporting by companies in BRICS.

Section 2 covers the data and results, and Section 3 presents concluding remarks.

II. DATA AND RESULTS

A. Data and Model

This is an empirical study. It uses secondary data such as a) participation in CDP questionnaires and b) the CDP scores of companies based on their CDP questionnaire responses. As mentioned in the companies’ CDP questionnaires, the scoring methodology measures the respondents’ progress toward environmental stewardship (CDP, 2020). A company’s environmental stewardship is assessed on four levels: 1) disclosure, 2) awareness, 3) management, and 4) leadership.

This study also considers national regulations on climate change reporting by companies and categorizes them into a) mandatory reporting regulations, b) voluntary reporting frameworks/guidelines, and c) the absence of regulations. The sample includes the Forbes list of 2000 global companies in BRICS, resulting in 281 companies from BRICS countries (Brazil, 19; Russia, 22; India, 53; China, 177; and South Africa, 10), with 1,931 firm–year observations for the period of 10 years from 2010 to 2019.

The logit model is ideal for studying categorical dependent variables. The logit model restricts estimated values to zero (non-reporting) and one (reporting). A binomial logit model can be used to assess CDP reporting status. A multinomial logit model can also be used to determine the CDP score, since it has multiple categories. The model can thus be specified as follows:

$$\begin{align}CDPST_{it} =&\ 1 \div \lbrack 1 + e - (\beta_{0} + \beta_{1}PCA_{it} \\&+ \beta_{2}MAND_{it} + \beta_{3}VOL_{it} \\&+ \beta_{4}LnCE_{it} + \beta_{5}LnED_{it}\ \\&+ \beta_{6} LnFE_{it} + \beta_{7}LnFI_{it} + \varepsilon_{it})\rbrack\end{align}\tag{1}$$

$$\begin{align}CDPSC_{it} =&\ 1\div \lbrack 1 + e - (\beta_{0} + \beta_{1}PCA_{it} \\&+ \beta_{2}MAND_{it} + \beta_{3}VOL_{it} \\&+ \beta_{4}LnCE_{it} + \beta_{5}LnED_{it}\ \\&+ \beta_{6}LnFE_{it} + \beta_{7}LnFI_{it} + \varepsilon_{it})\rbrack\end{align}\tag{2}$$

where

CDPST = CDP reporting status

CDPSC = CDP reporting score

PCA = dichotomous variable, where one indicates the country’s ratification of the Paris Climate Agreement, and zero otherwise

MAND = dichotomous variable, where one indicates the existence of country-specific mandatory climate change reporting regulation, and zero otherwise

VOL = dichotomous variable, where one indicates the existence of country-specific climate change reporting frameworks/guidelines, and zero otherwise

LnCE = natural logarithm of annual carbon emissions by the country of origin

LnED = natural logarithm of the annual level of energy depletion by the country of origin, which is the ratio of the value of the stock of energy resources to the remaining reserve lifetime

LnFE = natural logarithm of annual fuel exports by the country of origin

LnFI = natural logarithm of annual fuel imports by the country of origin

B. Results

Table 1 presents descriptive statistics for the variables used in our study.

The variables CDPST, PCA, MAND, and VOL are dummy variables, where CDPSC has four scores categorized as 1 through 4. The variable LnCE ranges from a minimum of 6.02 to a maximum of 9.23, with a mean of 1.54; LnED has a minimum of 21.70 and a maximum of 26.19, with a mean of 24.64; LnFE ranges from a minimum of 0.20 to a maximum of 4.26, with a mean of 1.54; and LnFI ranges from a minimum of -0.32 to a maximum of 3.68, with a mean of 2.70.

Table 2 shows the analysis of variance results for country-wise differences in CDPST. The variable CDPST varies greatly across BRICS nations. A post hoc test shows that Chinese and Russian firms report less often than South African and Brazilian firms, due to varied national corporate reporting cultures and laws. South Africa’s early adoption of integrated reporting (King Code of Governance, 2009) improved CDP reporting. China and Russia (authoritarian regimes) hinder company participation in international reporting platforms.

Table 3 provides the results of the binomial logistic regression model. The variable CDPST is positively and significantly influenced by LnCE by country (at the 1% level of significance) and by LnFE (at the 1% level of significance). The PCA variable positively and significantly influences CDPST at the 10% level of significance.

Table 4 presents the results of multinomial logistic regression. The results show the determinants of a firm’s progress toward environmental stewardship under four categories of CDP scoring, namely, Disclosure, Awareness, Management, and Leadership. The variable PCA has a positive and significant influence (at the 1% level of significance) for Disclosure, Awareness, Management, and Leadership; VOL positively and significantly influences companies to be in the categories Management (1% level of significance) and Leadership (1% level of significance), whereas MAND negatively and significantly influences companies to be in the Leadership (5% level of significance) category. The variable LnCE by country of origin negatively and significantly influences CDPSC (1% level of significance) in all four categories. The LnED variable in the country of origin negatively and significantly influences the level of CDPSC in the Management (1% level of significance) and Leadership (1% level of significance) categories, whereas it is positive and significant at the 1% level of significance for the Disclosure category. The variable LnFE has a positive and significant influence on companies to be in the Disclosure category (1% level of significance), and LnFI has a positive and significant influence on companies to be in the Management (10% level of significance) and Leadership (1% level of significance) categories.

The positive relation of LnCE with CDPST and its negative relation with CDPSC indicates that companies from countries with greater emissions participate in CDP reporting, but their level of disclosure is lower. The positive relation of LnFE with CDPST and the Disclosure category of CDPSC indicates that more companies from fuel-exporting countries tend to participate and disclose more to the CDP. The influence of PCA on CDPST and all levels of CDPSC indicates the strong influence of international climate change protocols. Among national regulations, the positive influence of VOL and the negative influence of MAND on the higher-level category of CDPSC indicate that national voluntary reporting frameworks/guidelines make companies disclose extensive information more than mandatory reporting regulations.

III. CONCLUDING REMARKS

The BRICS nations exhibit rapid economic growth rates, large populations, and fast-growing markets for goods and capital. The resultant growth in carbon emissions has led to legislation requiring corporations to reduce GHG emissions. This study finds that national legislation (MAND and VOL) has little effect on CDPST, whereas LnCE, LnFE, and PCA influence CDPST. Further, national legislation (MAND and VOL), PCA, LnCE, LnED, and LnFI significantly influence CDPSC at higher levels of disclosure, such as in the Management and Leadership categories. The variable PCA and LnCE significantly influence the lower levels of disclosure, such as in the disclosure and awareness categories. By affirming a) the influence of PCA on CDPST and all levels of CDPSC, b) the positive effect of VOL, and c) the negative influence of MAND on the higher-level category of CDPSC, our study validates institutional theory in the context of CDP reporting by companies in BRICS. It is also clear that, as the carbon emissions in the country of origin increase, large companies in BRICS are taking on greater leadership roles in carbon disclosure than ever before.

ACKNOWLEDGEMENT

Authors would like to thank the anonymous referee and editorial team of Energy Research Letters for their valuable comments.