Exploring the Role of Brand Capital Investment in the Realization of Firm-Level ESG Benefits and Consequences on Firm Performance: An Empirical Study

Abstract

This study examines how environmental, social, and governance (ESG) occurrences relate to firm performance and how these relationships depend on firms’ investments in brand capital. Using firm-level data spanning more than two decades, we analyze the effects of positive and negative ESG events on market-based (sales) and accounting-based (return on assets; ROA) performance for firms with and without brand capital investment (BCI). Using panel data on U.S. firms from 1995 to 2019, we compare firms that invest in brand capital through advertising with firms that do not. The results reveal an interesting asymmetric pattern. Specifically, BCI firms experience greater sales gains following positive ESG occurrences but incur significantly larger losses following negative ESG events. Interestingly, non-BCI firms benefit less from positive ESG activities but face smaller penalties from negative ESG occurrences. This study contributes to the marketing literature by examining brand capital investment and how ESG activities translate into performance gains versus when they impose performance costs for firms.

1. Introduction

In this research, we examine how brand capital investment impacts firm performance when a firm is experiencing a positive sustainability event, specifically a positive ESG occurrence, and a negative sustainability event, specifically a negative ESG occurrence. ESG is a subset of corporate social responsibility (CSR) and measures positive corporate behavior in three areas of sustainability—environmental, social, and governance (Chabowski et al., 2011). Examples of positive ESG occurrences include encouraging emission reductions (environmental), engaging in humane labor practices (social), and conducting ethical codes of conduct (governance), while negative ESG occurrences include detrimental behaviors in these three areas.

Examining firms’ positive ESG and negative ESG occurrences is important because stakeholders evaluate these occurrences to assess ethical behavior and future financial performance (Dorfleitner et al., 2020; Drempetic et al., 2020; Rezaee, 2016). High levels of positive ESG occurrences are associated with increased firm valuation, a sustainable competitive advantage, and profitability (Becchetti et al., 2012; Berman et al., 1999; H. J. Lee & Rhee, 2023; Rehman et al., 2021; Sendi et al., 2024; Servaes & Tamayo, 2013; Singhal et al., 2024; Zhao et al., 2018). Also, consumers have more favorable opinions toward socially responsible firms, are more likely to pay a premium, and display higher levels of loyalty when compared to socially irresponsible firms (Fombrun & Shanley, 1990; Saeidi et al., 2015; Siegel & Vitaliano, 2007).

On the other hand, firms are punished for negative ESG occurrences. A negative ESG occurrence is a subset of corporate social irresponsibility (CSI) (Hanson & Sharpe, 2025). Negative ESG consists of a focused range of offenses in the environmental, social, and governance domains often with deleterious outcomes (Dorfleitner et al., 2020; Lin-Hi & Müller, 2013). Firms experiencing a negative ESG occurrence experience lower consumer trust, brand devaluation, reduced access to outside resources, such as capital funds for borrowing, lower market share, stock sell offs, punitive fines, negative word of mouth, and difficulty with employee retention (Bang et al., 2023; Hillman & Dalziel, 2003; Ilhan et al., 2023; Klein & Dawar, 2004; Krüger, 2015; Luo & Bhattacharya, 2006; Rau & Yu, 2024; Sharfman & Fernando, 2008; Turban & Greening, 1997).

To help mitigate these negative outcomes from CSI, advertising is a commonly used communication tool. Increasing advertising spend during a brand crisis is more effective for socially responsible firms than socially irresponsible firms, thus improving firm performance (Cleeren et al., 2013; Jo & Harjoto, 2011; Rubel et al., 2011; Servaes & Tamayo, 2013; Sharpe & Hanson, 2020; Van Heerde et al., 2007). Also, this mitigating effect of advertising offers protective effects for the firm in the year following the CSI event (Sharpe & Hanson, 2021).

This research examines if firms investing in brand capital investment (i.e., advertising) experience greater consequences in terms of a negative ESG occurrence on firm performance than for firms completely abstaining from brand capital investment. Because advertising builds brand awareness, advertising expenditures are a common way to measure brand capital investment (Belo et al., 2014; Bronnenberg et al., 2022). We create two unique samples, one for firms engaging in brand capital investments (BCI firms) and one for firms not engaging in any such investment (non-BCI firms).

While prior research has focused on the positive effects of advertising efforts on brands experiencing positive and negative ESG occurrences with regards to firm value, it remains unclear if there are potential benefits and consequences to such advertising investment or noninvestment on firm performance outcomes (see Hanson & Sharpe, 2025). Findings in the CSR literature are mixed on the benefits of advertising investment on ESG and NESG occurrences. Some research studies find that advertising efforts for socially irresponsible firms can backfire (Becker-Olson & Olson, 2006; Oh et al., 2017). Other studies find that advertising helps to reduce the negative effect of an NESG occurrence on firm value; however, this effect is mixed when looking across the individual ESG domains with environmental crises benefitting the most, while governance crises experience negative outcomes (see Hanson & Sharpe, 2025). However, it remains unclear how brand capital investment and non-investment impacts firm performance when a brand experiences positive ESG and negative ESG occurrences and, specifically, how this impact is broken out across the individual ESG domains. This research seeks to fill this gap in the literature.

Specifically, our research focuses on the following research questions: do BCI firms benefit more from positive ESG occurrences than non-BCI firms? Do non-BCI firms experience less of an impact from negative ESG occurrences than BCI firms? Are non-BCI firms better able to navigate the consequential outcomes from a negative ESG occurrence because they do not engage in a highly visible communicative strategy (e.g., advertising)? Is this difference more pronounced for certain negative ESG domains? For firms experiencing a positive ESG occurrence, what is the impact of BCI on firm performance? Are there differences across the ESG domains; do some domains benefit more than others? For firms experiencing a negative ESG occurrence, what is the impact of BCI on firm performance? Are certain negative ESG domains contributing more to the impact of BCI on firm performance?

The remainder of the paper is organized as follows: first, we provide a review of relevant literature; next we describe the data, key variables, and our methodological approach; we then present and discuss the empirical results and finally we conclude with future research, managerial insights, and conceptual contributions.

2. Theory Development and Hypotheses

To understand the relationship between brand capital investment, positive ESG, and negative ESG on firm performance metrics, we draw upon congruence theory. Congruence theory has been used extensively in prior research to study the effect of promotion on stakeholders’ brand expectations (Fleck et al., 2012; Hong & Zinkhan, 1995; M. S. Lee et al., 2020). To be effective, brand messaging needs to be consistent and aligned with stakeholders’ expectations. For congruence to exist, there must be a harmonious match between the brand’s value and the way it is portrayed via promotional efforts (Hong & Zinkhan, 1995; Pracejus & Olsen, 2004). A congruous fit means that consumers are more likely to trust, resonate, engage, recall, and purchase from the brand (Brown & Dacin, 1997; Pracejus & Olsen, 2004). Firm communications are compared against prior beliefs, expectations, and observed organizational behaviors. When signals are aligned (i.e., high congruence), firm communications are processed as more credible, coherent and authentic, eliciting favorable responses (Paruchuri et al., 2021; Pérez, 2019).

Congruence is important when evaluating sustainability practices, such as positive ESG and negative ESG occurrences. Increased brand credibility results when a company’s social responsibility positions align with their core values, thus creating a strong brand fit (Moosmayer & Fuljahn, 2013). When social behavior is highly congruent with the core business, stakeholders view such firms more favorably than when there is a disconnect (Brown & Dacin, 1997; Sen & Bhattacharya, 2001). Recent research reveals that when congruence is weak, ESG communications can backfire, producing negative stakeholder reactions. When consumer expectations about CSR exceed perceived CSR performance, consumers develop perceptions of hypocrisy, fueling negative emotions, which can lead to negative attitudes and behaviors including negative word-of-mouth, complaints, and boycotting of the firm (L. Lee & Hur, 2024; Z. Wang & Zhu, 2020). Congruence theory acts as a key underlying mechanism explaining why ESG communications succeed and when such messaging is misaligned (Liu et al., 2023; Xu et al., 2025).

To develop our hypotheses, we discuss the expected relationships for brand capital investment, positive ESG occurrences, negative ESG occurrences, and firm value. While ESG and NESG are beginning to be explored more in the literature (see Dorfleitner et al., 2020; Drempetic et al., 2020; Seok et al., 2024; N. Wang et al., 2024), the literature on CSR and CSI has a more established history in the sustainability literature, so we will also discuss relevant CSR and CSI findings below in order to more thoroughly develop our hypotheses.

2.1. Brand Capital, ESG, and Firm Value

The positive impact of ESG management on sustainability is substantiated by a wide range of research studies (see Alsayegh et al., 2020; Eccles et al., 2012; Finger & Rosenboim, 2022; Servaes & Tamayo, 2013; N. Wang et al., 2024). Increasingly, stakeholders are turning to ESG metrics as a standardized approach for understanding a firm’s level of ethical compliance (Drempetic et al., 2020; Li et al., 2024; Rezaee, 2016). Firms that score high on ESG activities are better able to create enhanced customer loyalty, sustainable competitive advantages, and positive brand reputations, and they can support as risk management activities (Cippiciani et al., 2025; H. J. Lee & Rhee, 2023; Rehman et al., 2021). Furthermore, companies that disclose their ESG information have higher firm valuations when compared to firms that do not disclose their ESG behavior (Zhao et al., 2018).

Additionally, investment in promotional efforts signaling sustainability practices results in higher levels of profitability and firm value for socially responsible firms than for socially irresponsible firms (Ajour El Zein et al., 2019; Becchetti et al., 2012; Berman et al., 1999; Jo & Harjoto, 2011; Seok et al., 2024; Servaes & Tamayo, 2013). The existing literature strongly supports that socially responsible firm behavior is a strong positive predictor of firm valuation. Therefore, we expect to see BCI firms benefitting more from ESG occurrences than non-BCI firms when evaluating the impact on firm performance, specifically, Tobin’s q and sales. We suspect that BCI firms will be able to leverage their ESG initiatives via brand promotion to enhance their brand image, thus improving Tobin’s q and sales. This leads us to our first hypothesis:

H1. 

The benefits of positive ESG (PosESG) occurrences on firm performance are greater for brand capital investment firms than for non–brand capital investment firms.

2.2. Brand Capital, Negative ESG, and Firm Value

Research strongly shows that sustainability violations consistently erode firm value (Jo & Na, 2012; Luo & Bhattacharya, 2006; Sharma et al., 2022). When stakeholders’ prior positive associations become violated, brand incongruence results (Eklund & Helmefalk, 2022; Sjödin & Törn, 2006). As a result, firms experience a multitude of negative reactions from stakeholders including selling off shares, limiting or stopping purchases, and ending employment (Mackey et al., 2022; Valor et al., 2022). Thus, the relationship becomes strained, significantly impacting the brand and its value (Frooman, 1997; Miller et al., 2020; Valor et al., 2022).

Investing in brand capital helps a firm mitigate sustainability crises before, during, and even after a brand crisis (Gao et al., 2015; Sharpe & Hanson, 2021). By engaging in frequent brand communications, brands can establish strong pre-crisis relationships, potentially mitigating negative brand consequences (Burgoon et al., 1995; Cowden & Sellnow, 2002). Increasing advertising after a brand crisis allows the offending firm to control the message, project this message across multiple communicative platforms, and restore trust with consumers, leading to increased product demand (Cleeren et al., 2008; Coombs, 2014; Grullon et al., 2004; Joshi & Hanssens, 2010). Thus, communicative messages, following a brand crisis can lead to a restorative effect on firm performance (Cleeren et al., 2013; Cowden & Sellnow, 2002; Grullon et al., 2004; Joshi & Hanssens, 2010).

However, recent research finds an increase in consumer skepticism regarding the validity of sustainability claims (Carlos & Lewis, 2018; Farooq & Wicaksono, 2021). Also, when a firm engages in a brand scandal, trust violations occur, negatively impacting advertising effectiveness and purchase intention, leading to increased brand incongruence (Gistri et al., 2019; Lin et al., 2015). Thus, it remains unclear if non-BCI firms engaging in a negative ESG occurrence are in a better position when compared to BCI firms experiencing a negative ESG occurrence. Based on this reasoning, we expect that when a BCI firm experiences a negative ESG occurrence, such firms will be more exposed to the potential fallout from the negative ESG occurrence than firms without BCI. This leads to our next hypothesis:

H2. 

The consequences of negative ESG (NegESG) occurrences on firm performance are greater for brand capital investment firms than for non-brand capital investment firms.

3. Data and Method

We utilize two key sources to compile our comprehensive dataset—KLD Research & Analytics, Inc.’s Stats database and the Standard and Poor’s Compustat industrial files. The KLD dataset is one of the most widely used sources for academic research on corporate social responsibility (CSR), ESG performance, and corporate misconduct. The database provides annual, firm-level binary indicators that classify a range of social and environmental “strengths” and “concerns” across domains such as environment, community, diversity, employee relations, and governance. The comprehensive coverage provided by KLD data has made it a foundational dataset widely utilized for empirical analyses of how firms’ social practices relate to financial performance, stakeholder responses, and broader marketing outcomes in the CSR, CSI, positive ESG, negative ESG, sustainability, and corporate governance literature.

In this dataset, corporate social performance activities, including but not limited to ESG and NESG specific events, are tracked and rated annually for 3928 firms between 1995 and 2019. When matched at the firm- and year-level with the Standard and Poor’s Compustat industrial files, we identify 767 KLD firms, with ESG- and NESG-specific events, with available annual firm-level financial accounting data. Consistent with previous studies, brand capital investment (BCI) is measured using annual advertising expenses data from Compustat—see Belo et al. (2014, 2022) and Vitorino (2014). Upon merging the data from these two sources, our final sample consists of 13,603 firm-year observations and two primary categories of firms—those that invest in brand capital (BCI firms, N = 426) and those that do not (non-BCI firms, N = 243). The size of our sample is further reduced by the exclusion of firms in the financial and regulated utilities industries (SIC codes 6000–6999 and 4900–4999) and firms missing the dependent and/or the control variables. Following this selection process, the resulting final sample consists of 5350 BCI firm-year observations and 3658 non-BCI firm-year observations across the 24-year period between 1995 and 2019.

This sample is broadly diversified across industries and years for both BCI and non-BCI firms. BCI firms draw from a wide range of sectors, with service-related industries representing the largest share (39%), while non-BCI firms are primarily concentrated in the manufacturing- and service-oriented industries (59%). For both BCI and non-BCI firms, the data are evenly distributed over the sample timeframe (1999–2019), with most years contributing a similar proportion of observations. This balanced industry and year structure supports reliable panel analysis by preventing overrepresentation from any single industry or year.

3.1. Variable Definitions

3.1.1. Measuring BCI, PosESG, and NegESG Occurrence

Following Belo et al. (2014, 2022) and Vitorino (2014), brand capital investment (BCI) is measured by annual firm-level advertising expenditures. In Compustat, advertising expense is defined as the annual cost of advertising media (inclusive of radio, television, periodicals, etc.) and promotional expenses. As noted by Simon and Sullivan (1993), advertising impacts a company’s brand name through brand associations, perceived quality, and user experience. For instance, advertising that highlights verifiable attributes can influence brand associations. Additionally, frequent advertising can improve the perceived quality of experience goods—those whose quality cannot be determined until after purchase.

We closely follow previous work when constructing the aggregate measurements of PosESG and NegESG occurrences (see Kim & Li, 2021; Surroca et al., 2020). To create the ESG occurrence index, we add the number of strengths found in each category—environmental, social, and governance; the same methodology is applied to create the NegESG occurrence index except we are interested in the number of concerns found in each category. Because we are focusing on PosESG, which is a subset of CSR, we only calculate strengths and concerns in the following CSR dimensions: community relations, diversity and inclusion, employee relations, environment, human rights, and corporate governance. For each year of KLD data, strengths and concerns are calculated across the six dimensions of CSR.

When constructing the PosESG and NegESG variables, all domains are weighted equally in the aggregation process. More specifically and consistent with prior literature (see Kim & Li, 2021; Surroca et al., 2020), no additional scaling or domain-specific weighting is applied. Given that this approach treats each ESG dimension contributing equally, we recognize that applying alternative weighting schemes could lead to variation in how the index is interpreted. We address this concern by providing additional methodological detail and conduct robustness checks presented and discussed in the next section.

3.1.2. Key Dependent Variables

Performance measures, return on assets (ROA), and sales are the key dependent variables in our study. ROA measures the annual profitability of each firm in the sample relative to its total assets. For each firm, this variable is calculated as operating income before depreciation divided by total assets. Sales is another important indicator of firm performance and measures the total revenue generated from the sale of goods and services annually for each firm. This measure reflects a firm’s ability to attract customers and generate business volume.

3.1.3. Other Variables

To account for relevant firm- and industry-level factors, we control for annual industry competition using the Herfindahl–Hirshman Index (HHI), market value of equity (MVE), firm size (total assets), number of employees (EMP), firms operating in consumer industries (Consumer), years of advertising expenditures, and firm age. To reduce the possible effect of spurious outliers, all ratio and continuous variables are winsorized at the 1st and 99th percentile levels. Also, to address potential market effects and business cycle fluctuations, we include industry and year dummy variables.

4. Empirical Models and Results

4.1. Summary Statistics

The summary statistics for our two samples of firms are presented in

Table 1. Summary Statistics. This table reports the summary statistics for the key measures and other firm characteristics for the 5350 firm-year observations in the final sample of 426 BCI firms and for the 3658 firm-year observations in the final sample of 243 Non BCI firms. To reduce the effect of outliers, all accounting variables are winsorized at the 1st and 99th percentile.

Panel A: ESG Firms with Available Brand Capital Investment Data in Compustat
Variable	n	Mean	S.D.	Min	Mdn	Max
PosESG occurrences	5350	1.288	2.854	0.000	0.000	20.000
NegESG occurrences	5350	0.882	1.743	0.000	0.000	13.000
PosENV occurrences	5350	0.225	0.712	0.000	0.000	5.000
PosSOC occurrences	5350	0.986	2.143	0.000	0.000	15.000
PosGOV occurrences	5350	0.078	0.307	0.000	0.000	3.000
NegENV occurrences	5350	0.131	0.506	0.000	0.000	5.000
NegSOC occurrences	5350	0.504	1.089	0.000	0.000	8.000
NegGOV occurrences	5350	0.247	0.558	0.000	0.000	4.000
Sales [$M]	5350	11,000.000	18,000.000	4.727	3591.604	91,000.000
HHI	5350	0.077	0.047	0.025	0.066	0.563
MVE [$M]	5350	18,000.000	34,000.000	11.405	3944.077	140,000.000
Total Assets [$M]	5350	14,000.000	30,000.000	8.000	3516.724	180,000.000
ROA	5350	0.152	0.101	−0.386	0.147	0.425
Employees	5350	42.824	68.896	0.040	13.700	308.000
Consumer	5350	0.171	0.376	0.000	0.000	1.000
Advertising Expenditures [$M]	5350	315.359	615.585	0.029	63.591	2998.000
Years of advertising expenditures	5350	26.404	12.782	1.000	24.000	53.000
Firm age [Years]	5350	30.223	18.515	2.000	26.000	70.000
Panel B: ESG Firms Without Available Brand Capital Investment Data in Compustat
Variable	n	Mean	S.D.	Min	Mdn	Max
PosESG occurrences	3658	0.735	1.818	0.000	0.000	16.000
NegESG occurrences	3658	0.956	1.995	0.000	0.000	16.000
PosENV occurrences	3658	0.143	0.507	0.000	0.000	5.000
PosSOC occurrences	3658	0.534	1.405	0.000	0.000	11.000
PosGOV occurrences	3658	0.059	0.254	0.000	0.000	2.000
NegENV occurrences	3658	0.246	0.758	0.000	0.000	5.000
NegSOC occurrences	3658	0.516	1.094	0.000	0.000	8.000
NegGOV occurrences	3658	0.193	0.471	0.000	0.000	3.000
Sales [$M]	3658	8657.737	17,000.000	4.727	2236.002	91,000.000
HHI	3658	0.078	0.052	0.025	0.063	0.283
MVE [$M]	3658	9094.381	20,000.000	11.405	2416.240	140,000.000
Total Assets [$M]	3658	9183.939	20,000.000	9.039	2391.447	180,000.000
ROA	3658	0.130	0.091	−0.386	0.132	0.425
Employees	3658	21.471	30.687	0.040	9.846	254.000
Consumer	3658	0.065	0.246	0.000	0.000	1.000
Advertising Expenditures [$M]	0	.	.	.	.	.
Years of advertising expenditures	3658	0.000	0.000	0.000	0.000	0.000
Firm age [Years]	3658	26.663	18.538	2.000	20.000	69.000

(Panel A and Panel B). Panel A consists of 426 sample firms engaging in BCI activity between 1995 and 2019. Panel B consists of 243 sample firms not engaging in BCI activity during the sample period. On average, BCI firms have a higher likelihood of engaging in positive ESG activities at 1.28 versus 0.73 for non-BCI firms—with much of the positive activities reflected in the area of social performance (0.98) compared to positive activities in environmental (0.22) and corporate governance (0.78). In terms of a negative NESG occurrence, non-BCI firms have a slightly higher rate of occurrence at 0.95 versus 0.88 for BCI firms. Positive social-related activities are also the most common sustainability effort among non-BCI firms, with a likelihood of 0.53 compared to positive efforts in the areas of environmental (0.14) and corporate governance (0.06). For BCI firms, average annual advertising spend is $315.39 million across an average of 26.14 years.

When comparing the financial and firm-level metrics across the two panels, some interesting findings emerge. On average, BCI firms have more sales ($11.00 billion versus $8.65 billion) and employees (42.91 million versus 21.45 million) and tend to be older than non-BCI firms (30.23 years versus 26.67 years). BCI firms also show higher average profitability percentages (ROA) compared to non-BCI firms, 15.20% and 13.00%, respectively. Both of our samples have similar Herfindahl–Hirschman Indices (HHI)—0.077 for BCI firms and 0.078 for non-BCI firms, suggesting that firms in our sample operate in highly competitive markets.

The correlation matrix for BCI firms is presented in

Table 2. Correlation Matrix. This table reports the pairwise correlations for our ESG measures and other firm characteristics for the 5350 firm-year observations in the final sample of 426 BCI firms (Panel A) and for the 3658 firm-year observations in the final sample of 243 non-BCI firms (Panel B).

Panel A: ESG Firms with Available Brand Capital Investment Data in Compustat
		a	b	c	d	e	f	g	h	i	j	k	l	m	n	o	p	q	r
a	PosESG activities	1.000
b	NegESG activities	0.593	1.000
c	PosENV activities	0.799	0.420	1.000
d	PosSOC activities	0.974	0.595	0.660	1.000
e	PosGOV activities	0.644	0.386	0.508	0.545	1.000
f	NegENV activities	0.401	0.642	0.371	0.372	0.267	1.000
g	NegSOC activities	0.515	0.912	0.342	0.524	0.335	0.387	1.000
h	NegGOV activities	0.485	0.762	0.309	0.500	0.308	0.343	0.548	1.000
i	Sales [$M]	0.402	0.357	0.283	0.411	0.211	0.238	0.334	0.246	1.000
j	HHI	0.026	0.006	0.025	0.027	−0.005	0.036	0.008	−0.029	0.167	1.000
k	MVE [$M]	0.390	0.242	0.271	0.398	0.215	0.179	0.189	0.224	0.752	0.087	1.000
l	Total Assets [$M]	0.354	0.292	0.249	0.361	0.194	0.233	0.255	0.202	0.868	0.236	0.776	1.000
m	ROA	0.087	−0.021	0.058	0.093	0.027	−0.010	−0.029	0.000	0.091	−0.011	0.210	0.027	1.000
n	Employees	0.337	0.335	0.195	0.363	0.144	0.160	0.354	0.210	0.797	0.067	0.523	0.589	0.129	1.000
o	Consumer	0.031	0.007	0.028	0.029	0.017	0.053	−0.003	−0.020	−0.018	0.368	−0.006	0.006	0.065	−0.064	1.000
p	Advertising Expenditures [$M]	0.414	0.295	0.304	0.413	0.257	0.225	0.251	0.228	0.724	0.183	0.720	0.744	0.118	0.557	0.074	1.000
q	Years of advertising expenditures	0.251	0.102	0.197	0.250	0.134	0.131	0.081	0.042	0.328	0.077	0.286	0.242	0.185	0.354	0.274	0.380	1.000
r	Firm age [Years]	0.231	0.165	0.207	0.223	0.109	0.207	0.128	0.079	0.380	0.125	0.290	0.297	0.086	0.326	0.245	0.340	0.655	1.000
Panel B: ESG Firms Without Available Brand Capital Investment Data in Compustat
		a	b	c	d	e	f	g	h	i	j	k	l	m	n	o	p
a	PosESG activities	1.000
b	NegESG activities	0.606	1.000
c	PosENV activities	0.702	0.400	1.000
d	PosSOC activities	0.954	0.589	0.488	1.000
e	PosGOV activities	0.476	0.280	0.337	0.313	1.000
f	NegENV activities	0.556	0.850	0.381	0.535	0.256	1.000
g	NegSOC activities	0.515	0.924	0.342	0.493	0.271	0.648	1.000
h	NegGOV activities	0.479	0.724	0.285	0.489	0.148	0.491	0.551	1.000
i	Sales [$M]	0.340	0.335	0.201	0.350	0.092	0.328	0.256	0.298	1.000
j	HHI	0.037	0.054	0.057	0.013	0.080	0.082	0.036	0.014	0.133	1.000
k	MVE [$M]	0.374	0.323	0.194	0.388	0.139	0.342	0.259	0.219	0.698	0.185	1.000
l	Total Assets [$M]	0.400	0.416	0.225	0.408	0.152	0.438	0.340	0.269	0.762	0.200	0.894	1.000
m	ROA	0.031	0.026	0.023	0.032	0.001	0.028	0.015	0.032	0.008	0.065	0.109	0.021	1.000
n	Employees	0.350	0.361	0.222	0.355	0.098	0.319	0.301	0.317	0.593	−0.022	0.535	0.523	0.015	1.000
o	Consumer	−0.024	−0.018	0.028	−0.049	0.044	−0.007	−0.025	−0.004	0.004	0.503	−0.059	−0.035	0.001	−0.035	1.000
p	Firm age [Years]	0.234	0.270	0.206	0.211	0.099	0.303	0.210	0.172	0.379	0.033	0.372	0.410	0.069	0.438	0.022	1.000

(Panel A) and shows that positive ESG activities are highly correlated with their underlying components, specifically PosSOC (0.974) and PosENV (0.799), indicating that positive ESG actions tend to occur jointly across dimensions. Similarly, negative ESG activities cluster strongly, with NegESG highly correlated with NegSOC (0.912) and NegGOV (0.762), suggesting that ESG controversies often span multiple domains. Similarly, the correlation matrix for non-BCI firms, also presented in Table 2 (Panel B), reflects that positive ESG activities are highly correlated with PosSOC (0.954) and PosENV (0.702), while negative ESG activities cluster strongly across all three of its underlying domains, NegESG (0.850), NegSOC (0.924), and NegGOV (0.724). Overall, correlation patterns are similar across BCI and non-BCI firms, with strong clustering among ESG subcomponents and firm size variables. In both groups, a positive ESG aligns most closely with environmental and social dimensions, while a negative ESG clusters across social and governance concerns. Additionally, ESG activities are more strongly associated with sales and advertising among BCI firms, suggesting greater market relevance of ESG for firms with brand capital investment.

With regards to multicollinearity considerations, we observe that several correlations exceed conventional thresholds (≥0.800), notably among ESG dimensions and among size related variables (e.g., Sales and Total Assets, Sales and Employees). Theoretically, these high correlations are expected and reflect construct overlap rather than redundancy. To mitigate potential multicollinearity concerns, our empirical examinations include standard control variables, rely on aggregated ESG measures, and employ fixed effects where appropriate.

4.2. Positive ESG, Negative ESG, Brand Capital Investment (BCI), and Firm Performance

Using a series of statistical models, we explore the relationships between brand capital investment, PosESG, NegESG occurrence, and firm performance in the areas of ROA and sales. First, we examine the relationship among PosESG, NegESG, and firm performance for BCI firms. More specifically, our first hypothesis (H1) is tested by regressing ROA (ROA)_i,t and the log of Sales (Ln(Sales)_i,t) based on key focal and control variables, using the fixed effects ordinary least squares (OLS) regression models presented below:

$${\left(\mathrm{R}\mathrm{O}\mathrm{A}\right)}_{i,t}=\alpha +{\beta }_1{PosESG}_{i,t-1}+{\beta }_{2}Neg{ESG}_{i,t-1}+\gamma Z_{i,t-1}+{\mu }_i+{\lambda }_t+{\epsilon }_{i,t},$$

(1)

$${\mathrm{L}\mathrm{n}\left(\mathrm{S}\mathrm{a}\mathrm{l}\mathrm{e}\mathrm{s}\right)}_{i,t}=\alpha +{\beta }_1{PosESG}_{i,t-1}+{\beta }_{2}Neg{ESG}_{i,t-1}+\gamma Z_{i,t-1}+{\mu }_i+{\lambda }_t+{\epsilon }_{i,t},$$

(2)

where i, t, μ, and λ reference the firm index, time index, firm fixed effects, and year fixed effects, respectively. To control for potential endogenous factors among the sample firms, our models are estimated with firm and year fixed effects. In addition, standard errors are heteroskedasticicity-robust and clustered at the firm level.

The dependent variables for models 1 and 2 are (ROA)_i,t and (Ln(Sales)_i,t), respectively. In models 1 and 2, the PosESG and NegESG scores are the independent variables of interest. The coefficients associated with these variables capture the impact of the PosESG and NegESG scores in year t − 1 on ROA and sales in year t. Z is a vector of control variables. To account for firm-level, industry-level, and time effects, firm, year, and industry fixed effects are included.

The estimation results for Equations (1) and (2) are presented in

Table 3. Positive ESG, negative ESG, brand capital investment (BCI), and firm performance. This table shows OLS regression with firm fixed effects results of the effects of PosESG and NegESG on ROA and logSales. Coefficient estimates for industry dummies (based on the Fama-French 12 industry classification) and year dummies are not reported. T-statistics, shown in parentheses, are based on standard errors clustered at the firm level. ***, **, and * denote significance at 1%, 5%, and 10% respectively.

Panel A: ESG Firms with Available Brand Capital Investment Data in Compustat
	(1)	(2)	(3)	(4)
	ROAt	ROAt	logSalest	logSalest
PosESGt−1	−0.000	0.000	0.006 **	0.007 ***
	(0.001)	(0.001)	(0.002)	(0.003)
NegESGt−1	−0.001	−0.001	−0.010 **	−0.013 ***
	(0.001)	(0.001)	(0.004)	(0.005)
logSalest−1	0.019 ***	0.018 ***	0.771 ***	0.770 ***
	(0.006)	(0.006)	(0.019)	(0.019)
HHIt−1	0.243 **	0.248 **	−0.115	−0.074
	(0.121)	(0.124)	(0.383)	(0.384)
Consumert−1		0.000		0.000
		(0.000)		(0.000)
PosESGt−1 × Consumert−1		−0.002		−0.011 ***
		(0.002)		(0.004)
NegESGt−1 × Consumert−1		0.003		0.020 ***
		(0.003)		(0.006)
MVEt−1			0.000 ***	0.000 ***
			(0.000)	(0.000)
Assetst−1			0.000	0.000
			(0.000)	(0.000)
ROAt−1			0.231 *	0.226 *
			(0.134)	(0.133)
Year_dum	Yes	Yes	Yes	Yes
Industry_dum	Yes	No	Yes	No
_cons	−0.025	−0.025	1.940 ***	1.938 ***
	(0.062)	(0.062)	(0.168)	(0.168)
N	3569	3569	3569	3569
R2	0.042	0.042	0.834	0.835
adj. R2	0.036	0.036	0.833	0.834

. The results in column 1 show a negative insignificant relationship between PosESG occurrences and ROA for BCI firms. When the control variables are included in (column 2), the relationship becomes positive but remains insignificant. This suggests that positive ESG occurrences have a limited impact on the profitability of BCI firms. However, when looking at the relationship between PosESG occurrences and sales for BCI firms, in columns 3 and 4, we find a positive and highly significant relationship between PosESG occurrences and sales (β = 0.006, p < 0.05, and β = 0.007, p < 0.01 respectively). These results suggest that positive ESG events may have a more immediate impact on market-driven outcomes such as sales. One possible explanation for the observed relationship is that positive ESG activities can improve a firm’s public image, strengthen customer loyalty, or attract new consumers, thereby contributing to higher revenue generation.

In terms of the impact of a NegESG occurrence on ROA and sales for BCI firms, we find a negative and insignificant relationship between NegESG occurrences and ROA (column 1). The same relationship holds when the control variables are included in (column 2), suggesting that negative ESG occurrences do not translate to measurable declines in profitability for BCI firms. These findings suggest that for BCI firms, profitability measures, such as ROA, may be less sensitive to short-term reputational shocks related to negative ESG occurrences. It may also be argued that BCI firms are able to mitigate the immediate operational effects of ESG controversies, preventing them from appearing in profitability metrics. When looking at the relationship between NegESG occurrences and sales, reflected in column 3, we observe a negative and significant relationship (β = −0.010, p < 0.05). When the control variables are included, reflected in column 4, we see a strengthening of the negative relationship, as well as an increase in the significance (β = −0.013, p < 0.01). This pattern suggests that negative ESG activities may have more immediate market-facing consequences, linked to a potential reduction in customer perceptions and demand. Furthermore, the increased magnitude after adding controls also implies that this effect is robust and not driven by omitted firm characteristics.

Table 4. Positive ESG, negative ESG, and performance for non-brand capital investment (BCI) firms. This table shows OLS regression with firm fixed effects results of the effects of PosESG and NegESG on ROA and logSales. Coefficient estimates for industry dummies (based on the Fama-French 12 industry classification) and year dummies are not reported. T-statistics, shown in parentheses, are based on standard errors clustered at the firm level. ***, **, and * denote significance at 1%, 5%, and 10%, respectively.

Panel B: ESG Firms Without Available Brand Capital Investment Data in Compustat
	(1)	(2)	(3)	(4)
	ROAt	ROAt	logSalest	logSalest
PosESGt−1	−0.002	−0.002	−0.012 ***	−0.013 ***
	(0.002)	(0.002)	(0.004)	(0.005)
NegESGt−1	0.000	0.000	−0.004	−0.003
	(0.002)	(0.002)	(0.004)	(0.005)
logSalest−1	0.034 ***	0.034 ***	0.813 ***	0.813 ***
	(0.009)	(0.009)	(0.021)	(0.021)
HHIt−1	0.714 ***	0.724 ***	1.739 ***	1.702 ***
	(0.219)	(0.220)	(0.571)	(0.584)
Consumert−1		0.000		0.000
		(0.000)		(0.000)
PosESGt−1 × Consumert−1		−0.006		0.017 *
		(0.005)		(0.009)
NegESGt−1 × Consumert−1		0.005 **		−0.016
		(0.002)		(0.012)
MVEt−1			0.000 ***	0.000 ***
			(0.000)	(0.000)
Assetst−1			−0.000 **	−0.000 **
			(0.000)	(0.000)
ROAt−1			−0.243	−0.241
			(0.201)	(0.201)
Year_dum	Yes	Yes	Yes	Yes
Industry_dum	Yes	No	Yes	No
_cons	−0.239 ***	−0.242 ***	1.470 ***	1.480 ***
	(0.085)	(0.085)	(0.203)	(0.203)
N	2704	2704	2704	2704
R2	0.085	0.086	0.848	0.848
adj. R2	0.079	0.079	0.846	0.846

presents the empirical results for non-BCI firms. The results in column 1 show a negative and non-significant relationship between PosESG occurrences and ROA for non-BCI firms (β = −0.002, n.s). When the control variables are included (column 2), the relationship remains unchanged, suggesting that positive ESG events do not influence the profitability of non-BCI firms. One possible explanation for this observed relationship is that ROA may not immediately capture the effects of ESG-related actions. Comparably, when looking at the relationship between PosESG occurrences and sales for non-BCI firms (column 3), we observe that a negative and highly significant relationship exists between them (β = −0.012 p < 0.01); when control variables are added in (column 4), we find that the negative and significant relationship between PosESG occurrences and sales remains (β = −0.013, p < 0.01). This indicates that positive ESG events are associated with significantly lower sales than for non-BCI firms, suggesting that such activities may face higher skepticism from consumers and are not immediately rewarded in the marketplace. These findings hold when missing observations of advertising expenditures are replaced with zero.

Overall, these findings reveal observable differences regarding the relationship between ESG occurrences and the tested performance measures (ROA and sales) across BCI and non-BCI firms. For BCI firms, positive ESG events do not impact ROA but are consistently linked to sales increases, suggesting quicker market-based responses compared to accounting-based ones. Similarly, negative ESG occurrences show no effect on ROA but can result in significant declines in sales, indicating market sensitivity to ESG-related controversies. Furthermore, for non-BCI firms, positive ESG occurrences show no impact on ROA and instead correspond to lower sales, indicating that these firms may not benefit from such initiatives—or may even be penalized for such events. Lastly, across all models, adding control variables does not meaningfully alter the results, thus underscoring the robustness of these observed patterns and generating partial support for Hypotheses H1 and H2.

To assess the economic significance of our main results, we translate key coefficients into interpretable magnitudes. More specifically, for BCI firms, each additional positive ESG occurrence increases sales by about 0.6–0.7%, whereas each negative ESG occurrence reduces sales by about 1.0–1.3%. That said, a one-standard deviation increase in positive ESG, implies roughly a 2.4–2.8% sales growth, while a one-standard-deviation rise in negative ESG implies a 3–4% sales decline. Overall, these findings represent meaningful short-term revenue effects. For non-BCI firms, positive ESG occurrences are associated with 1.2–1.3% lower sales per additional event, indicating that a one-standard-deviation increase would reduce sales by about 4.8–5.2%. We also observe that neither positive or negative ESG events have meaningful implications for ROA for both BCI and non-BCI firms.

Overall, BCI firms gain sales from positive ESG activity, while non-BCI firms lose sales from the same activities. Additionally, negative ESG events produce significant sales penalties for BCI firms but show no comparable effect for non-BCI firms. These differences underscore our central claim that the economic returns generated by ESG activities vary for BCI and non-BCI.

4.3. Robustness Analysis: Positive ESG Domains, Negative ESG Domains, Brand Capital Investment (BCI), and Firm Performance

For BCI and non-BCI firms, we conduct supplemental analyses exploring the relationship between our performance measures (ROA and sales) and the individual domains of positive and negative ESG occurrences, which are the following: positive environmental events (PosE), positive social events (PosS), positive governance events (PosG), negative environmental events (NegE), negative social events (NegS), and negative governance events (NegG).

The results of this analysis pertaining to BCI firms are presented in

Table 5. Positive ESG domains, negative ESG domains, brand capital investment (BCI), and firm performance. This table shows OLS regression with firm fixed effects results of the effects of PosE, PosS, PosG, NegE, NegS, and NegG on ROA and logSales. Coefficient estimates for industry dummies (based on the Fama-French 12 industry classification) and year dummies are not reported. T-statistics, shown in parentheses, are based on standard errors clustered at the firm level. ***, **, and * denote significance at 1%, 5%, and 10%, respectively.

Panel A: ESG Firms with Available Brand Capital Investment Data in Compustat
	(1)	(2)	(3)	(4)
	ROAt	ROAt	logSalest	logSalest
PosEt−1	−0.004	−0.004	0.007	0.003
	(0.005)	(0.005)	(0.008)	(0.009)
PosSt−1	0.001	0.002	0.007 *	0.012 ***
	(0.001)	(0.002)	(0.004)	(0.004)
PosGt−1	−0.000	−0.002	−0.005	−0.017
	(0.007)	(0.008)	(0.013)	(0.015)
NegEt−1	−0.002	−0.003	−0.015	−0.024
	(0.003)	(0.004)	(0.014)	(0.018)
NegSt−1	−0.001	−0.001	−0.007	−0.009
	(0.002)	(0.002)	(0.006)	(0.007)
NegGt−1	−0.001	−0.002	−0.015	−0.023 **
	(0.003)	(0.004)	(0.009)	(0.010)
logSalest−1	0.018 ***	0.018 ***	0.770 ***	0.770 ***
	(0.006)	(0.006)	(0.019)	(0.019)
HHIt−1	0.229 *	0.231 *	−0.132	−0.107
	(0.120)	(0.124)	(0.385)	(0.381)
Consumert−1		0.000		0.000
		(0.000)		(0.000)
PosEt−1 × Consumert−1		−0.001		0.032 *
		(0.009)		(0.017)
PosSt−1 × Consumert−1		−0.004		−0.029 ***
		(0.003)		(0.009)
PosGt−1 × Consumert−1		0.006		0.036
		(0.012)		(0.025)
NegEt−1 × Consumert−1		0.005		0.038 *
		(0.008)		(0.022)
NegSt−1 × Consumert−1		0.003		0.012
		(0.004)		(0.008)
NegGt−1 × Consumert−1		0.005		0.038 ***
		(0.008)		(0.014)
MVEt−1			0.000 ***	0.000 ***
			(0.000)	(0.000)
Assetst−1			0.000	0.000
			(0.000)	(0.000)
ROAt−1			0.230 *	0.223 *
			(0.134)	(0.134)
Year_dum	Yes	Yes	Yes	Yes
Industry_dum	Yes	No	Yes	No
_cons	−0.019	−0.018	1.944 ***	1.951 ***
	(0.062)	(0.062)	(0.169)	(0.167)
N	3569	3569	3569	3569
R2	0.042	0.043	0.834	0.835
adj. R2	0.036	0.035	0.833	0.834

. The results in column 1, pertaining to the positive ESG domains for BCI firms, show a negative and insignificant relationship between ROA and the individual domains of positive ESG occurrences for environmental (PosE) and governance (PosG). A positive and insignificant relationship exists for the social domain (PosS). This observed relationship remains unchanged when control variables are included (column 2). These findings could imply that disaggregating ESG occurrences into their individual domains does not substantively improve the explanatory power of positive ESG activities for firm performance as measured by ROA.

When looking at the relationship between the individual domains of PosESG occurrences and sales for BCI firms (column3), we observe a positive but insignificant relationship with positive PosE, a negative but insignificant relationship with PosG, and a significant relationship with PosS (β = 0.007, p < 0.10). When control variables are added (column 4), the results remain relatively consistent for PosE and PosG events. However, we find a positive and highly significant relationship with PosG events (β = 0.012, p < 0.01). This suggests that positive governance-related events may be more meaningful for revenue generation.

In reviewing the results in column 1 associated with the negative ESG domains, we find a negative but insignificant relationship between ROA and each of the individual domains. In column 2, when control variables are added, the observed relationships between ROA and the individual negative ESG domains remain unchanged. This indicates that negative ESG events, whether environmental, social, or governance-related, do not translate into immediate declines in ROA for BCI firms.

However, when evaluating the relationship between sales and the negative ESG domains (column 4), we observe a negative but insignificant relationship with NegE and NegS and a negative and significant relationship with NegG. Arguably, governance-related controversies often receive heightened public attention and scrutiny leading to a reduction in both investor and consumer confidence which may explain the stronger association with sales declines.

Next, we examine the results associated with non-BCI firms presented in

Table 6. Positive ESG domains, negative ESG domains, and firm performance for non-brand capital investment (BCI) firms. This table shows OLS regression with firm fixed effects results of the effects of PosE, PosS, PosG, NegE, NegS, and NegG on ROA and logSales. Coefficient estimates for industry dummies (based on the Fama-French 12 industry classification) and year dummies are not reported. T-statistics, shown in parentheses, are based on standard errors clustered at the firm level. ***, **, and * denote significance at 1%, 5%, and 10%, respectively.

Panel B: ESG Firms Without Available Brand Capital Investment Data in Compustat
	(1)	(2)	(3)	(4)
	ROAt	ROAt	logSalest	logSalest
PosEt−1	0.001	0.001	−0.018 *	−0.022 **
	(0.005)	(0.005)	(0.009)	(0.010)
PosSt−1	−0.004 *	−0.003	−0.010	−0.011 *
	(0.002)	(0.002)	(0.006)	(0.006)
PosGt−1	−0.000	0.002	−0.017	−0.015
	(0.008)	(0.009)	(0.019)	(0.022)
NegEt−1	0.003	0.002	0.018 **	0.021 **
	(0.003)	(0.004)	(0.009)	(0.009)
NegSt−1	−0.001	−0.001	−0.011 *	−0.011
	(0.002)	(0.002)	(0.006)	(0.007)
NegGt−1	0.000	−0.000	−0.020	−0.019
	(0.004)	(0.005)	(0.013)	(0.013)
logSalest−1	0.034 ***	0.034 ***	0.814 ***	0.814 ***
	(0.009)	(0.009)	(0.021)	(0.021)
HHIt−1	0.708 ***	0.719 ***	1.760 ***	1.725 ***
	(0.220)	(0.221)	(0.560)	(0.573)
Consumert−1		0.000		0.000
		(0.000)		(0.000)
PosEt−1 × Consumert−1		0.008		0.023
		(0.008)		(0.045)
PosSt−1 × Consumert−1		−0.014 *		0.025
		(0.008)		(0.029)
PosGt−1 × Consumert−1		−0.008		−0.016
		(0.012)		(0.045)
NegEt−1 × Consumert−1		0.008		−0.044
		(0.007)		(0.037)
NegSt−1 × Consumert−1		−0.004		−0.009
		(0.007)		(0.033)
NegGt−1 × Consumert−1		0.016		−0.002
		(0.011)		(0.052)
MVEt−1			0.000 ***	0.000 ***
			(0.000)	(0.000)
Assetst−1			−0.000 ***	−0.000 ***
			(0.000)	(0.000)
ROAt−1			−0.241	−0.240
			(0.201)	(0.202)
Year_dum	Yes	Yes	Yes	Yes
Industry_dum	Yes	No	Yes	No
_cons	−0.242 ***	−0.244 ***	1.474 ***	1.488 ***
	(0.084)	(0.084)	(0.203)	(0.204)
N	2704	2704	2704	2704
R2	0.086	0.088	0.848	0.848
adj. R2	0.078	0.077	0.847	0.846

. The findings in column 1, related to the positive ESG domains, reflect a positive but insignificant relationship between ROA and the positive environmental occurrences (PosE), a negative and insignificant relationship with positive governmental occurrences (PosG), and a negative and significant relationship with positive social occurrences (PosS) (β = −0.004, p < 0.10). When control variables are added in (column 2), the observed relationship between ROA and PosE and PosG remains insignificant, while the relationship between ROA and PosS becomes insignificant (β = −0.003, p < n.s). The observed findings suggest that positive ESG activities do not appear to enhance profitability for non-BCI firms and, in some cases, could coincide with slight declines in ROA.

When reviewing the relationship between sales performance and the individual domains of PosESG occurrences for non-BCI firms (column3), we observe a negative but insignificant relationship with PosS and with PosG and a negative and significant relationship with PosE (β = 0.018, p < 0.10). In column 4, when control variables are added, the intensity and significance of the negative relationship with PosE increases (β = 0.022, p < 0.05), and the negative relationship with PosS gains significance (β = 0.011, p < 0.10). These results imply that positive ESG actions, particularly those in the environmental domain, may be met with some skepticism from consumers due to a lack of reputational credibility often associated with non-BCI firms. Additionally, the increase in significance observed after adding control variables indicates that these findings are robust and not explained by firm-level differences.

In column 1, the results associated with the relationship between negative ESG domains and ROA for non-BCI firms is shown to be insignificant with all three negative ESG domains. This relationship does not change with the inclusion of control variables (column 2). These findings imply that negative ESG occurrences do not have an immediate measurable impact on firm profitability (ROA). The relationship between sales and the negative ESG domains of non-BCI is presented in column 3 and is shown to be positive and significant with the NegE (β = 0.018, p < 0.05) domain and negative and significant with the NegS (β = 0.011, p < 0.10) domain. When control variables are added (column 4), the positive relationship between sales and NegE remains significant (β = 0.021, p < 0.10), while the negative relationship between sales and NegS becomes insignificant (β = 0.011, p < n.s.). These results suggest that negative environmental events may prompt increased sales activity—potentially the result of higher media attention and a strategic firm response. While negative social-domain-related occurrences may initially weaken sales, such occurrences lose significance once firm-level controls are added. The attenuation of the NegS effect in column 4 indicates that underlying firm characteristics, rather than the controversy itself, may be motivating the observed relationship.

In sum, the negative ESG events do not affect ROA for non-BCI firms, implying limited short-term financial sensitivity to controversies. However, sales responses are shown to vary by domain. Specifically, negative environmental occurrences are associated with higher sales, while negative social occurrences show weaker and less consistent effects. The observed patterns suggest that market reactions to ESG events may be nuanced and may depend on the type of controversy and individual firm characteristics.

5. Conclusions

This study explores the complex relationship between ESG occurrences and the impact on firm outcomes, specifically ROA and sales. We investigate both the benefits and consequences that result from positive and negative ESG activities for firms engaging in brand capital investments and compare these results to firms without such an investment. The comprehensive dataset, with over two decades of firm-level observations, adds to the robustness of these results.

Overall, the primary findings of our investigation indicate that firms with positive ESG activities that invest in brand capital will reap significantly positive benefits reflected in the form of higher sales. Conversely, BCI firms will face a significantly negative impact on sales performance as a consequence of engaging in negative ESG activities. However, when looking at the impact on ROA, we do not find a significant result. This finding suggests that positive ESG events may have a more immediate impact on market-driven outcomes, such as sales, instead of accounting-driven metrics, such as ROA. Positive ESG occurrences are visible to consumers and help strengthen customer loyalty, cultivate a positive public image, and attract new consumers to the brand. Conversely, negative ESG occurrences are perceived as incongruent by consumers, resulting in negative sales. Consistent with prior research, brand capital investment increases firm-level scrutiny from stakeholders, especially consumers (Forcadell et al., 2023; Lou, 2014). BCI firms are at an elevated risk of consumers noticing inconsistencies, which can result in decreased trust and purchase intentions (Delmas & Burbano, 2011; Fella & Bausa, 2024). Firms that highlight their ESG activities invite increased scrutiny, and if a mismatch exists between promotional claims and brand behavior, reputational costs ensure, resulting in decreased sales.

In comparison, non-BCI firms fail to realize the benefits of brand investment when engaging in positive ESG activities. However, when engaging in negative ESG activities, these firms are less negatively affected. This suggests that a lack of brand investment helps to reduce the negative impact of decreased sales from consumers disengaging with the brand. Although BCI firms experience a significant negative effect on sales, this impact becomes insignificant for non-BCI firms. Our findings show that BCI investment offers significant advantages for firms engaging in positive ESG activities by improving purchase outcomes. However, non-BCI firms benefit from being less vulnerable to the consequences of negative ESG events than BCI firms.

The insights generated by this investigation are interesting, unexpected, and illuminate theoretical and practical contributions to the field of marketing. Our results expand the empirical understanding of the nuanced relationship among brand capital investment, ESG, NESG, and firm performance, which is an underexplored area in the marketing domain. Conceptually, we explore the various relationships among positive ESG, negative ESG, brand capital investment, and firm performance outcomes.

Our empirical findings show that brand capital investment is necessary for maximizing the performance-enhancing benefits of positive ESG occurrences. Also, given that brand capital investment is measured by firm advertising expenditures, these results suggest that such investments can be informative about a firm’s expected performance in the area of sales. Furthermore, by interpreting firm-level advertising expenditures as an investment to create brand capital, this paper contributes to the growing stream of literature on financial research in marketing focusing on the effect of advertising and other marketing variables on firm performance outcomes (Bagwell, 2007; Hanson & Sharpe, 2025; Sharpe & Hanson, 2021; Srinivasan & Hanssens, 2009; Srivastava et al., 2006).

From a managerial perspective, our results inform marketing decision makers in actionable ways. For BCI firms, ESG practices can be a double-edged sword. When brand communication is aligned with ESG practices, it can enhance brand equity and stakeholder trust; however, when ESG practices are misaligned with communication efforts, it can lead to decreased sales. Managers should be mindful that their ESG practices closely align with outward facing communication messages as consumers punish firms who say one thing and then fail to deliver on their ESG initiatives. The finding that BCI firms are rewarded more for positive ESG activities suggests that marketing decision-makers should consider incorporating brand-building strategies into their ESG-related initiatives. Furthermore, managers should ensure that firms’ ESG efforts are effectively communicated through advertising and branding campaigns. Overall, marketing decision-makers can leverage these insights by making strategic investments in brand capital (e.g., advertising) a priority and coordinating such investments with ESG initiatives in order to effectively realize the potential of these efforts to enhance performance outcomes.

6. Future Research and Limitations

While this research examines brand capital investment around periods of positive ESG and negative ESG occurrences and the result on firm performance outcomes, additional research would be beneficial to explore further relational complexities. Specifically, classifying firms according to brand investment intensity levels and determining if diminishing returns exists across industries could be examined. Also, investigating the relationship among the individual domains, investment intensity, and industries would allow for a more nuanced understanding in the literature. Finally, an analysis into the frequency and intensity of positive ESG and negative ESG occurrences could yield helpful insights. For example, do firms with very high levels of ESG occurrences benefit even more from advertising frequency and intensity than firms with moderate or low levels of ESG occurrences and intensity? Similarly, future research could investigate the same questions surrounding negative ESG occurrences, frequency, and intensity.

Finally, while advertising spending is the primary used empirical measure of brand capital, we do acknowledge that as a singular dimension of brand capital, this is an imperfect proxy. In this study, we address this limitation by accounting for the persistence of brand capital over time. This limitation could potentially be further addressed in future research by incorporating alternative proxy normalizations and lag structures. These approaches, along with the addition of other accounting-based and non-accounting-based measures, will create a more specific operationalization of the brand capital investment construct. Overall, this domain presents numerous opportunities for future research opportunities and is an underexplored area in the current literature.