The Effect of Digital Washing on Firm Value: The Mediating Role of ESG Performance

1. Introduction

Digital transformation (DT) has become a structural priority for firms and economies worldwide, driven by rapid technological advancements and stronger competitive pressure [1]. As global competitiveness increasingly depends on digital readiness, countries - especially developing economies- face huge pressure to shift toward digitally enabled growth [2,3].

Within Southeast Asia, DT has spread rapidly but unevenly. The Association of Southeast Asian Nations (ASEAN) has formally recognized digitalization as a strategic pillar and sustainable growth, emphasizing investments in digital infrastructure, data governance, skills development, and digital financial inclusion [4]. These variations create an important context for comparative research on DT in key ASEAN economies.

This study focuses on the ASEAN-4 (Indonesia, Malaysia, Thailand, the Philippines) and Vietnam. Although ASEAN-5 often includes Singapore, it is excluded in this analysis as it is a developed country and the main financial flow, which accounts for about 38% of the total FDI in the 5 countries [5]. Vietnam is included because it has recently become one of the fastest-growing digitalizing economies in ASEAN, plays an importance in regional supply chains and digital innovation [6]. The Vietnamese Government launched the National DT Program to 2025, with Orientation to 2030 under Decision 749/QĐ-TTg in 2020, with the goals of developing a digital government, economy, and society [7]. Adapting to governmental decrees, by the end of 2024, the development of DT contributed 13.17% of national GDP [8].

Additionally, across ASEAN-4 and Vietnam, DT has accelerated in recent years. Indonesia continues to expand its digital economy and aims for its digital economy to reach approximately US$120 billion by 2025 [9]. In 2024, Malaysia’s ICT and e-commerce sectors accounted for 23.4% of national economic activity, supported by record-high digital investments of RM 163.6 billion, reflecting strong momentum in nationwide digitalization [10,11]. Thailand’s digital economy grew to 4.44 trillion baht in 2024 (a 5.7% increase from 2023), and the country rose to 52nd in the 2024 United Nations Digital Government Development Index, second in ASEAN, demonstrating substantial progress in public-sector DT [12,13]. In 2024, the Philippines’ digital economy expanded to PHP 2.25 trillion, equivalent to 8.5% of GDP, driven by growth in digital infrastructure and Information Communication and Technologies-enabled industries [14]. Vietnam’s 2024 national DT report highlights rapid progress in digital public services, cloud adoption, and emerging technologies such as AI and semiconductors [15]. Together, these trends indicate strong momentum in the region’s DT.

From a theoretical perspective, there was research about DT’s impact on Firm value (FV), which found that the relationship between DT and FV is different depending on the time period as well as the firm's background. Most studies show an overall positive and beneficial effect for companies adopting such initiatives. Based on a panel of 2254 Chinese manufacturing firms from 2010 to 2020 [16], the intensity of DT is positively correlated with operational performance; [17] revealed that DT generally enhances firms’ competitiveness. However, there has also been a study that found a negative relationship between DT and FV [18]. In parallel, DT has been linked to improved environmental, social, and governance (ESG) performance [19], and ESG Performance also increases FV [20].

However, despite its prevalence and potential impact, many enterprises choose Digital washing (DW) - the topic was underexplored in academic research. It overstates a firm’s digital progress and has emerged as a symbolic practice similar to greenwashing. Evidence from Chinese firms indicates it can temporarily boost market perception and liquidity but carries long-term credibility risks [21]. Prior studies have mainly focused on authentic DT in developed economies, overlooking how symbolic digital behaviors operate in developing institutional contexts. Hence, little is known about the mechanisms through which DW influences FV, such as its indirect effects via ESG performance.

Therefore, this study seeks to address these knowledge gaps by focusing on ASEAN-4 and Vietnam’s firms, aiming to clarify the direct and indirect effects of DW on FV. In addition, it examines how ESG performance mediates this relationship, both individually and in combination. By exploring these mechanisms across firms in different industries, this study aims to uncover how symbolic digital disclosure can shape real corporate outcomes and is expected to provide empirical insights for managers, investors, and policymakers in assessing the situation of DT and the real economic consequences of DW. Based on the findings, this study conducts a comparative analysis to formulate strategies for sustainable development in Vietnam.

2. Theoretical Background and Literature Review

2.1. Theoretical Background

This study uses some background theories, as follows: the theory of planned behavior (TPB), resource-based view theory (RBV), signaling theory, and stakeholder theory.

TPB indicated the factors affecting behavior are: attitude variables, subjective norm variables, and perceived behavioral control variables, which combine to indirectly impact behavior through intention [22]. Firms have a favorable attitude toward disclosing DT because it has a positive impact on FV [23]. Moreover, due to social pressure - a subjective norm, they expect enterprises to undergo DT, and they just evaluate firms from an external perspective [24]. However, firms have struggled with DT in aspects of human resources, a lack of a business-oriented digital workforce [25,26], insufficient financial resources, and technological requirements [27]. Together with these variables, DW produces a similar impact to DT, yet enterprises could easily achieve it with fewer resources.

RBV suggests that the value and development of firms should be their heterogeneous resources [28] which give firms a competitive advantage. Through DT, firms are able to create rare and inimitable digital capabilities, thereby increasing FV. ESG performance is considered to be firms’ intangible resource [29], different firms have competitive advantage by applying alternative ESG strategies.

Signaling Theory provides a framework for understanding how parties in an information asymmetric environment convey credible information to reduce uncertainty [30]. Firms increasingly utilize technologies to improve operational efficiency and signal transparency, innovation, and governance quality to the market [31]. Adopting digital technologies is costly, requires long-term commitment, and is easily observable by external stakeholders-characteristics, it makes a strong signal according to the conditions outlined by [32]. Thus, DT can serve as a strategic signal that a firm possesses advanced capabilities in data management and information disclosure, which reassures investors of the firm’s integrity and reduces perceived information risk.

Stakeholder Theory suggested that firms achieve long-term sustainability by aligning the interests of all stakeholders [33]. As a strategic framework, ESG metrics provide a means to assess a firm’s performance across dimensions that matter to a wide range of stakeholders, in a manner similar to how financial indicators measure outcomes for shareholders [34].

2.2. Literature Review

2.2.1. Digital Washing and Firm Value

According to current literature, much research studied the impact of DT on FV there was no study examining the relationship between DW and FV.

On the one hand, in the short term, DT has a negative impact on FV when it is measured by Return on Assets, this is explained by [18] that investments in digital maturity may take more years to be reflected in the firm’s financial indicators. Moreover, because the information technology assets are either not depreciated or their depreciation is shown slowly, this affects a firm's financial indicators negatively in the short term.

On the other hand, also in [18], in the long-term, by utilizing DT, the company can maximize FV. DT is not only about fostering technological innovation [35], but also changing internal structure, enhancing collaboration across various functions, and boosting efficiency in production and organizational processes [36]. Moreover, DT redefines how companies deliver value to customers [37], then promotes the value of their products and services to external stakeholders, as a result increasing FV [38]. Although the DT requires a long-time period, it is a core element driving future competitiveness for firms [39], creating sustainable competitive advantages for firms, thereby forming long-term value to customers and enhancing overall FV [31].

However, many firms face challenges in DT due to financial constraints and technological gaps [25,26,27]. As a result, some turn to DW - the act of exaggerating and fabricating the level of DT [40]. To better understand how firms may exaggerate their digital progress, we break DW into three parts to mirror the main layers of DT. First, digital-technology adoption reflects the use of visible tools such as cloud systems, IoT devices, AI applications, or advanced computing [41,42]. Second, business-model innovation illustrates more strategic, customer-facing changes, including platform-based services, digital payments, and data-driven offerings [31,43]. Third, digital-strategy implementation refers to broader plans and structures, such as digital roadmaps, Enterprise Resource Planning, and Application Programming Interface integration, or governance mechanisms, that guide DT at the organizational level [41,44]. [42] shows that higher levels of DT improve firm performance, suggesting that investors may react positively when they perceive DT as a driver of operational and financial gains, boosting FV even when actual DT remains limited. Therefore, together with TPB:

H1. 

Digital Washing has a positive effect on Firm Value.

H1a. 

Digital-technology-adoption Washing has a positive effect on Firm Value.

H1b. 

Business-model-innovation Washing has a positive effect on Firm Value.

H1c. 

Digital-strategy-implementation Washing has a positive effect on Firm Value.

2.2.2. Digital Washing and ESG Performance:

Some studies found a complex relationship: the relationship between DT and ESG performance is U-shaped [45]. Besides, applying DT in the early stage has a struggle related to financial and operational resources and crowds out short-term ESG investments [46], which could affect sustainability long-term goals [19].

However, most previous research indicates the positive relationship between DT and ESG Performance. DT is a key factor in enhancing ESG performance by applying digital technologies to bring major improvements, with the goal is lower operating costs and increased social responsibility, helping firms meet stakeholder expectations and enhance their overall enterprise value [19]. It enhances ESG performance by promoting transparency [47], innovation [19], and efficiency in corporate processes [45].

From an environmental perspective, in the research of [48], DT has a negative effect on the environment because of more consumption of resources and energy, and the waste and emissions from production. However, DT improves a business’s development sustainability [49], with new business models fundamentally focusing on the environment [50], and the efficiency of information use through adoption of technologies across the product life cycle [48].

Regarding the social pillar, some studies argue that the development of DT can challenge security, privacy, misinformation, and social division [51]. On the other hand, DT improves operational competencies and customer experience [52]; it also enhances the way people connect and access information [53].

In terms of governance, DT influences firm governance by increasing information transparency and reducing information asymmetry [54], which raises the efficiency of monitoring the production process, resulting in improved enterprise’s operation [45]. Moreover, DT enhances the sharing of resources to external enterprises, helps meet the needs of multi-stakeholders and affects stakeholder governance [55]. Additionally, DT increases accounting reliability, which improve overall level of firm governance, and results in reinforced sustainable practices [56].

Together with TPB, this research proposes:

H2. 

Digital Washing has a positive influence on ESG Performance.

H2a. 

Digital Washing has a positive influence on the environment.

H2a.1. 

Digital-technology-adoption Washing has a positive influence on the environment.

H2a.2. 

Business-model-innovation Washing has a positive influence on the environment.

H2a.3. 

Digital-strategy-innovation Washing has a positive influence on the environment.

H2b. 

Digital Washing has a positive influence on Social.

H2b.1. 

Digital-technology-adoption Washing has a positive influence on Social.

H2b.2. 

Business-model-innovation Washing has a positive influence on Social.

H2b.3. 

Digital-strategy-innovation Washing has a positive influence on Social.

H2c. 

Digital Washing has a positive influence on Governance.

H2c.1. 

Digital-technology-adoption Washing has a positive influence on Governance.

H2c.2. 

Business-model-innovation Washing has a positive influence on Governance.

H2c.3. 

Digital-strategy-innovation Washing has a positive influence on Governance.

2.2.3. ESG and Firm Value

When approaching FV, there are wide debates about the effect of ESG on it. By using meta-analysis of more than 2000 prior research papers, [57] found that 48% papers written 1970 to 2015 indicate the positive relationship between ESG and firm performance, 11% negative and 23% neutral.

However, most studies indicate positive correlation between ESG Performance and FV [20]. Many external stakeholders claim ESG as a signal to evaluate the sustainability of business [58], because it enhances efficiency of operation [59], improves corporate reputation [60], increases productivity of employee and capital market benefits [61], prevents risk problems, so can be considered as an advantage in competitiveness and sustainable management strategy [20,62].

Specifically, each component of ESG performance also impacts FV. ​Research indicates that corporate performance regarding environmental responsibility can significantly boost corporate financial performance by reducing operating costs, capital cost, and increasing revenue [63]. ​Bad environmental performance correlates negatively with a firm's intangible asset value [64]. ​Environmental performance demonstrates a company’s environmental and social concern, which expressed through corporate social responsibility, is positively received by the society, raising the FV [65]. ​

Societal factors are often studied under the heading of Corporate Social Responsibility (CSR). CSR participation has been shown to increase FV [66]. ​In particular, CSR activities that concentrate on internal social enhancement and product quality, tend to enhance FV effectively [66]. ​Social capital, including civic cooperation and interpersonal trust emerges as an important determinant of FV [67].

Strong corporate governance is often associated with improved financial performance and greater investor attraction [68]. ​Numerous studies have found firm-level corporate governance and firm valuation are strongly and favorably correlated [69]. ​Effective corporate governance can positively impact FV, with better-governed firms having lower capital cost [70]. ​Internal governance factors, such as board leadership and independence, institutional ownership, and analyst following, are related to FV [71]. ​Combine with TPB:

H2a.1. 

Environmental has a positive influence on firm value.

H2b.1. 

Social has a positive influence on firm value.

H2c.1. 

Governance has a positive influence on firm value.

2.2.4. The mediating role of ESG Performance:

Prior research indicates the mediating role of ESG: [72] showed that ESG performance mediates the influence of DT on high quality development of enterprises, while [73] pointed out the mediating role of ESG performance in the relationship between open innovation and enterprise value. Especially, [74] studied the mediation of ESG performance in the impact of DT on FV.

Therefore, ESG and its each component are proved to play an indirect role in the influence of DT on FV. Together with TPB:

H3. 

ESG mediates the impact of digital washing on firm value.

H3a. 

Environmental mediates the impact of digital washing on firm value.

H3a.1. 

Environmental mediates the impact of Digital-technology-adoption Washing on firm value.

H3a.2. 

Environmental mediates the impact of Business-model-innovation Washing on firm value.

H3a.3. 

Environmental mediates the impact of Digital-strategy-innovation Washing on firm value.

H3b. 

Social mediates the impact of digital washing on firm value.

H3b.1. 

Social mediates the impact of Digital-technology-adoption Washing on firm value.

H3b.2. 

Social mediates the impact of Business-model-innovation Washing on firm value.

H3b.3. 

Social mediates the impact of Digital-strategy-innovation Washing on firm value.

H3c. 

Governance mediates the impact of digital washing on firm value.

H3c.1. 

Governance mediates the impact of Digital-technology-adoption Washing on firm value.

H3c.2. 

Governance mediates the impact of Business-model-innovation Washing on firm value.

H3c.3. 

Governance mediates the impact of Digital-strategy-innovation Washing on firm value.

3. Methodology

3.1. Data and Sample

The sample comprises the listed firms in countries of ASEAN-4 and Vietnam. The initial sample size is 3695 in 6 years. To ensure data integrity and consistency, all observations with missing values in any variables are excluded [75], and continuous variables are winsorized at 1 and 99% quantiles. After the process, the dataset comprises 530 firms (2,529 samples in total) spanning 6 years.

The data for the mediating, dependent, and control variables are collected from the Bloomberg database. For the independent variables - DW and its dimensions, this research adopts the method of text mining and machine learning to measure according to the methods in similar research on DW and earnings management in corporate finance [40,76,77,78].

3.2. Variables

3.2.1. Dependent Variable

Firms’ value refers to the firm’s ability to generate sustainable performance internally and the extent to which external stakeholders recognize this performance. This study uses Tobin’s Q as a measurement of firms’ value.

3.2.2. Independent Variable

DW is the act of exaggerating or fabricating a firm’s level of DT, as indicated by the gap between its goal-oriented promises and its limited actual performance [40]. Therefore, to indicate the degree of DW and its dimensions, the promises and actual actions must be measured separately, denoted by DTA and DTW (DT actions and DT words), respectively.

Since the manifestation of DW and DT is similar, this research adopts the available methods of measuring DT degree in combination with machine learning to measure the DW level. Text mining is an available method and has been proven to be effective through being utilized in much research on DW in divergent contexts [36,79,80]. Text mining using Python, in this research, is used to analyse the annual reports of firms to measure their promises and actual actions in terms of DT. Before the annual reports are analysed, the texts are standardized by tokenization, removing punctuation and lowercasing. As the annual reports comprise both the statement on the past year's actual actions and the promises for the following year, the report is identified into parts of “actual actions”, “promises”, and “others” (which is neither actions nor promises). This separation is conducted by identifying which parts, including words related to plan, strategy, or reported actions, or neither, by Python. The document is processed using a rule-based, heading-driven segmentation approach. Pages are classified into 3 sections based on predefined heading keywords. After segmentation, keyword frequency analysis is applied to measure the occurrence of relevant terms in each section.

A dictionary of digital-transformation-related keywords with 3 dimensions is constructed to serve for textual analysis in an academic and practical approach (Table 1). From an academic approach, digital technologies bring about disruptions, which require firms’ strategies and adaptation, therefore leading to business model changes and value redefinition of firms [31]. In the view of a practical approach, DT is about 4 key aspects: AI, big data, cloud computing, and blockchain [43]. The dictionary is used to count the frequency of keywords’ occurrences in each category: actual action, promises, and others. The results are then divided by the total words counted in the whole annual report to be standardized; thereby, DTA and DTW at a particular year are measured.

A simple machine learning linear regression approach with scikit-learn library in Python is employed in this research to predict the expected level of digital transformation time t (denoted by DTAf) based on the level of digital transformation commitment (DTW) at time t-1. The sample of this research ensures a sufficiently large sample size for splitting the data into training and testing sets (80:20), guaranteeing that the matrix is non-singular and helps avoid over-fitting. The forecasted value of DTA at time t (DTAf) is then subtracted from the actual DTA at time t; thereby, the degree of DW at time t is measured. The same method is used.

3.2.3. Mediating Variable

The ESG performance of firms is measured by the ESG score retrieved from the Bloomberg database (Table 2).

3.3. Model

To test whether DW has a positive effect on FV (H1), this research uses the OLS model for regression analysis. The equations are as follow:

$${FV}_{it}={\beta }_0+{\beta }_{1}D{W}_{it}+{\beta }_{2}B{M}_{it}+{\beta }_{3}LE{V}_{it}+{\beta }_{4}RO{A}_{it}+{\beta }_{5}SIZ{E}_{it}+YEA{R}_{fe}+{\epsilon }_{it}$$

(1)

$$F{V}_{it}={\beta }_0+{\beta }_{1}A{W}_{it}+{\beta }_{2}IN{W}_{it}+{\beta }_{3}IM{W}_{it}+{\beta }_{4}B{M}_{it}+{\beta }_{5}LE{V}_{it}+{\beta }_{6}RO{A}_{it}+{\beta }_{7}SIZ{E}_{it}+YEA{R}_{fe}+{\epsilon }_{it}$$

(2)

To test ESG’s mediation in the effect of DW and FV (H3), this study uses the causal mediating effect model. Initially, the effects of independent variables on mediating ones are tested using OLS models:

$$ES{G}_{it}={\beta }_0+{\beta }_{1}D{W}_{it}+{\beta }_{2}B{M}_{it}+{\beta }_{3}LE{V}_{it}+{\beta }_{4}RO{A}_{it}+{\beta }_{5}SIZ{E}_{it}+YEA{R}_{fe}+{\epsilon }_{it}$$

(3)

$$E_{it}={\beta }_0+{\beta }_{1}D{W}_{it}+{\beta }_{2}B{M}_{it}+{\beta }_{3}LE{V}_{it}+{\beta }_{4}RO{A}_{it}+{\beta }_{5}SIZ{E}_{it}+YEA{R}_{fe}+{\epsilon }_{it}$$

(4)

$$S_{it}={\beta }_0+{\beta }_{1}D{W}_{it}+{\beta }_{2}B{M}_{it}+{\beta }_{3}LE{V}_{it}+{\beta }_{4}RO{A}_{it}+{\beta }_{5}SIZ{E}_{it}+YEA{R}_{fe}+{\epsilon }_{it}$$

(5)

$$G_{it}={\beta }_0+{\beta }_{1}D{W}_{it}+{\beta }_{2}B{M}_{it}+{\beta }_{3}LE{V}_{it}+{\beta }_{4}RO{A}_{it}+{\beta }_{5}SIZ{E}_{it}+YEA{R}_{fe}+{\epsilon }_{it}$$

(6)

$$ES{G}_{it}={\beta }_0+{\beta }_{1}A{W}_{it}+{\beta }_{2}IN{W}_{it}+{\beta }_{3}IM{W}_{it}+{\beta }_{4}B{M}_{it}+{\beta }_{5}LE{V}_{it}+{\beta }_{6}RO{A}_{it}+{\beta }_{7}SIZ{E}_{it}+YEA{R}_{fe}+{\epsilon }_{it}$$

(7)

$$E_{it}={\beta }_0+{\beta }_{1}A{W}_{it}+{\beta }_{2}IN{W}_{it}+{\beta }_{3}IM{W}_{it}+{\beta }_{4}B{M}_{it}+{\beta }_{5}LE{V}_{it}+{\beta }_{6}RO{A}_{it}+{\beta }_{7}SIZ{E}_{it}+YEA{R}_{fe}+{\epsilon }_{it}$$

(8)

$$S_{it}={\beta }_0+{\beta }_{1}A{W}_{it}+{\beta }_{2}IN{W}_{it}+{\beta }_{3}IM{W}_{it}+{\beta }_{4}B{M}_{it}+{\beta }_{5}LE{V}_{it}+{\beta }_{6}RO{A}_{it}+{\beta }_{7}SIZ{E}_{it}+YEA{R}_{fe}+{\epsilon }_{it}$$

(9)

$$G_{it}={\beta }_0+{\beta }_{1}A{W}_{it}+{\beta }_{2}IN{W}_{it}+{\beta }_{3}IM{W}_{it}+{\beta }_{4}B{M}_{it}+{\beta }_{5}LE{V}_{it}+{\beta }_{6}RO{A}_{it}+{\beta }_{7}SIZ{E}_{it}+YEA{R}_{fe}+{\epsilon }_{it}$$

(10)

Then, the effects of the mediating variables on the dependent variables are tested using the following OLS models:

$$F{V}_{it}={\beta }_0+{\beta }_{1}ES{G}_{it}+{\beta }_{2}B{M}_{it}+{\beta }_{3}LE{V}_{it}+{\beta }_{4}RO{A}_{it}+{\beta }_{5}SIZ{E}_{it}+YEA{R}_{fe}+{\epsilon }_{it}$$

(11)

$$F{V}_{it}={\beta }_0+{\beta }_1{E}_{it}+{\beta }_2{S}_{it}+{\beta }_3{G}_{it}+{\beta }_{4}B{M}_{it}+{\beta }_{5}LE{V}_{it}+{\beta }_{6}RO{A}_{it}+{\beta }_{7}SIZ{E}_{it}+YEA{R}_{fe}+{\epsilon }_{it}$$

(12)

After testing the two regressions, a bootstrap procedure is performed to test the significance of the mediation effect (H3); thereby, the indirect effect is also calculated.

$$F{V}_{it}={\beta }_0+{\beta }_{1}D{W}_{it}+{\beta }_{2}ES{G}_{it}+{\beta }_{3}B{M}_{it}+{\beta }_{4}LE{V}_{it}+{\beta }_{5}RO{A}_{it}+{\beta }_{6}SIZ{E}_{it}+YEA{R}_{fe}+{\epsilon }_{it}$$

(13)

$$F{V}_{it}={\beta }_0+{\beta }_{1}D{W}_{it}+{\beta }_2{E}_{it}+{\beta }_3{S}_{it}+{\beta }_4{G}_{it}+{\beta }_{5}B{M}_{it}+{\beta }_{6}LE{V}_{it}+{\beta }_{7}RO{A}_{it}+{\beta }_{8}SIZ{E}_{it}+YEA{R}_{fe}+{\epsilon }_{it}$$

(14)

$$F{V}_{it}={\beta }_0+{\beta }_{1}A{W}_{it}+{\beta }_{2}IN{W}_{it}+{\beta }_{3}IM{W}_{it}+{\beta }_4{E}_{it}+{\beta }_5{S}_{it}+{\beta }_6{G}_{it}+{\beta }_{7}B{M}_{it}+{\beta }_{8}LE{V}_{it}+{\beta }_{9}SIZ{E}_{it}+YEA{R}_{fe}+{\epsilon }_{it}$$

(15)

The composite variables (DW, ESG) and its sub-variables (AW, INW, IMW and E, S, G) are not incorporated in a single model due to the possibility of multicollinearity created by the potential correlation between them, which may undermine the reliability of the results [81]. _fe represents the fixed effect; ε denotes the random error terms; and _it represents the effect in firm i at time t.

3.4. Hypothesis Testing Method

Before the proposed hypotheses are tested, the descriptive and correlation matrix are calculated to provide overall characteristics of the sample as well as to detect potential correlation and collinearity among variables.

Subsequently, the regression models with Ordinary Least Squares (OLS), Random Effects (RE), and Fixed Effects (FE) estimators are estimated sequentially and selected to test the proposed direct effects. This research uses the Breusch-Pagan test to choose between the OLS and RE models [82]. If the OLS model is chosen, it is retained for the main analyses. If the RE model is selected, the researcher conducts a Hausman test to determine which of the RE or FE models is more appropriate.

After model selection, potential model deficiencies are diagnosed, including heteroskedasticity using the Kezdi test [83], autocorrelation using the Wooldridge test [84], cross-sectional dependence using the Pesaran CD test [85], and multicollinearity assessed through the Variance Inflation Factor (VIF), which is calculated from the regression model after a demeaning transformation [84].

Appropriate corrective approaches are applied for each issue after identifying deficiencies. Specifically, when heteroskedasticity and autocorrelation are present, robust standard errors are employed to correct the estimations. If cross-sectional dependence occurs simultaneously with the aforementioned issues, the study applies Driscoll-Kraay (1998) [86] standard error estimators. In cases where VIF values exceed acceptable thresholds, variables are added or removed to ensure the stability of the estimations.

4. Results and Discussion

4.1. Empirical Results

4.1.1. Descriptive Result

Before the hypotheses are tested, the descriptive statistics of the data is calculated. The results of descriptive are shown in Table 3.

4.1.2. Correlation Matrix

To identify possibilities of multicollinearity issues, the correlation coefficients of all variables are examined. The correlation matrix is shown in Table 4. There are several possible multicollinearities, which will be further validated in following steps.

4.1.3. Baseline Regression

First, the GLS, RE and FE models were estimated. After the validation of BP LM tests and Hausman tests, the model of fixed effect was selected. Further validation methods reveal problems of heteroskedasticity and cross-sectional correlation (but no problem of Multicollinearity after Uncentered VIF validation). This research adopts the validation of Driscoll–Kraay standard errors to solve the problems. Despite its common use for datasets with large time periods and small numbers of cross-sectional units, it has been proven to be applicable for datasets with large numbers of cross-sectional units and small time periods, as there is no limitation for the scale of N in comparison with T [86]. Furthermore, this method can completely address the mentioned deficiencies, unlike clustered standard error, which cannot solve the issues of cross-sectional dependence.

The effect of DW and its dimensions on FV are tested based on the equations (1) and (2). The results are indicated in Table 5.

The coefficients of DW, AW and INW are significantly positive with p-value smaller than 0.05, meaning that the positive effects of DW, AW and INW on FV are statistically significant. Firms therefore in practice can greatly enhance their value by conducting those actions, confirming H1, H1a, H1b. However, the impact of IMW may not exist in real context.

4.1.4. Mediating Test

The paths of effect from independent variables on mediating ones are tested according to Eq. (3) - Eq. (10). The results are presented in Table 6. The results showed no significant effect.

Subsequently, the effects of mediating variables on dependent ones are tested; however there is no effect which is statistically significant (Table 7). Therefore H2 and its sub-hypotheses are rejected.

As the tests of effect paths from independent variables to mediating ones, and from the mediating to the dependent ones showed no significant results, the role of partial mediation of ESG and its pillars are not supported. Thus, further steps of calculating indirect effects are unnecessary as the effect is insignificant in real context.

4.2. Discussion

Our findings indicate that DW, when considered as a composite structure, has a positive and statistically significant effect on FV in both the ASEAN-4 and Vietnam.

Firstly, results illustrate that digital-technology-adoption washing - the first level of DW, exerts the strongest positive influence on FV. This category shows fundamental and highly visible forms of DT, which can be readily observed and easily communicated [41,42]. A reasonable explanation is that technology adoption sends a clear and tangible signal to the market when it suggests better operational efficiency, greater scalability, and more modernized processes. Investors tend to view these improvements as significant signs of short-term performance gains, therefore when firms overstate their adoption, the market still rewards these signals.

Secondly, business-model-innovation washing demonstrates a positive relationship with FV, these innovations are more complex, long-term and less immediately verifiable than digital-technology-adoption. These announcements often stimulate favorable expectations among investors and stakeholders, as business-model innovation is associated with future market expansion and improved competitiveness among other companies.

In contrast, digital-strategy-implementation washing does not influence FV. Strategic statements are often broad, qualitative, and lack observable evidence of execution [44]. Investors view these announcements as overly generic or symbolic, making them less credible signals of actual transformation. Thus, the market is unlikely to reward firms that merely claim digital strategies without real technological or business-model progress. Firms in these economies are still in early stages of DT, making immediate, tangible technological upgrades more influential than high-level strategic narratives.

Moreover, the findings show that DW does not have an effect on ESG performance, aligning with the fundamental difference between the short-term nature of DW and the long-term trajectory of ESG outcomes. Operationally, scholars identify DW by applying yearly text-mining techniques to annual reports and sustainability disclosures, allowing them to capture rhetorical or promotional shifts that occur at the annual reporting cycle [87,88]. By contrast, ESG scores and the behavioural inputs underlying them typically evolve over multi-year horizons and are measured through lagged, often third-party, data streams-making them structurally insensitive to short-term fluctuations in disclosure tone or terminology [87,88].

Empirical evidence demonstrates that DT affects firms’ ESG performance positively, by enhancing information transparency, promoting green innovation, and strengthening internal governance mechanisms [74,89]. A review of empirical studies on the link between DT and ESG further clarifies this distinction: only sustained and substantive digital investments-such as long-term adoption of digital technologies or continuous digital innovation-generate persistent improvements in ESG performance [74,89,90].

Additionally, in this sample, ESG performance has no significant effect on FV. This result does not support prior research, which indicated that ESG performance can enhance a firm's reputation in investors’ sight [60], improve efficiency in labor level and operation, therefore increasing market capital benefits [61]. These results may be due to the difference in contexts of developing countries in comparison with the developed ones. Moreover, the market of developing countries may not put enough emphasis on sustainable development; therefore the impacts of ESG performance are not significant enough for the market to reflect in firms’ value.

However, this research indicates that ESG does not play a mediating role in the impact of DW on FV. This finding is opposite to prior research showing that ESG mediates the impact of DT on enterprises’ high quality innovation [72] and the effect of DT on FV [74]. The result stems from DW being characterized by symbolic short-term signaling rather than substantive technological investment. Unlike genuine DT, DW does not alter organizational processes, governance structures, or environmental and social practices conditions identified as necessary to generate measurable improvements in ESG performance. Therefore, the absence of a mediation effect is theoretically consistent: without real digital capabilities, firms lack structural, long-term changes required to improve ESG outcomes, even though DW may still generate a short-term direct impact on FV through impression management.

5. Conclusions and Implication

Firstly, this research found that DW when considered as a composite structure, is positively related with FV. However, this effect is driven by only two of its components. Secondly, DW does not have an influence on ESG performance. Thirdly, ESG and its pillar of it do not have a significant impact on FV. Finally, the study showed that ESG could not play a mediating role in the impact of DW on FV.

Remarkedly, this research contributes to the literature on DW’s direct effects on firm value by incorporating DW dimensions, which have not been examined before. Additionally, by investigating the mediation of ESG in the impact of DW on FV, this study expands on current research by indicating that ESG does not have a mediating role in the studied influence. Moreover, the research also contributes to the literature by providing updated empirical evidence in the ASEAN-4 and Vietnam contexts, where research on the impact of DW is limited, thereby supporting the comparison.

The findings offer several important learning points for Vietnam after the comparison in the results. For regulators and policymakers, they should consider strengthening standardized disclosure frameworks, enhancing digital reporting assurance, and developing regulatory guidelines for AI-generated or digitally inflated sustainability content.

For firms, enterprises are encouraged to shift from DW to substantive DT - using digital tools to strengthen ESG data systems, operational transparency, and sustainability outcomes to ensure that market value is supported by genuine organizational progress rather than rhetorical amplification.

Investors should have greater caution by focusing on the quality of digital sustainability disclosures and by complementing textual analysis with independent ESG data sources, third-party ratings, and substantive performance indicators. Investors with long-term horizons should prioritize firms exhibiting verifiable ESG improvements and transparent governance systems rather than those merely providing digitally enhanced sustainability communication.

The research still has limitations that future study can address. Firstly, the sample of this research was limited to 5 developing countries in ASEAN, therefore future research could expand the scope or focus on less developed countries. Secondly, due to limited time, we investigated firms in short-time from 2019 - 2024, while realizing the effects of sustainable development (ESG) requires a longer period to be clearly and measured. Finally, this research did not explore the differentiated impact across various enterprise types or different industries.