Abstract
This study explores the impact of strategic, tactical, and internal orientations on greenwashing within Colombia’s manufacturing industry. Adopting a comprehensive perspective on corporate legitimacy within the cognitive domain, this empirical research utilizes data from the 2021 Industrial Environmental Survey. Through advanced supervised machine learning techniques, such as LASSO-regularized logistic regressions and double machine learning, the study identifies key predictors associated with greenwashing practices. The results emphasize that tactical and internal orientations (such as environmental impact measurement and education) are significantly related to reducing greenwashing practices, whereas strategic orientation (environmental plans and studies) has a limited impact. Additionally, sectoral differences are identified, with sectors of lower environmental impact like food and textiles being less prone to greenwashing, in contrast to more polluting sectors like oil refining and metallurgy. The study highlights the necessity for more stringent policies and regulations to guarantee the genuineness of sustainable practices and prevent environmental certifications from being exploited as marketing strategies. On a corporate level, it suggests implementing sustainability training programs and tools for measuring environmental impact. Theoretically, the research offers a comprehensive understanding of greenwashing in emerging markets and underscores the significance of cognitive, moral, and pragmatic legitimacy in fostering genuine business practices.
Introduction
In response to growing global concern for environmental sustainability, regulations have been established to raise employers’ awareness of the environmental impact of their activities (Montoya-Restrepo & Montoya-Restrepo, 2022). This imperative has led companies to adopt green marketing strategies, as research shows a direct link between environmental innovation and the growth of small and medium-sized enterprises (SMEs) in volatile market conditions (Cristancho-Triana et al., 2024; Ebrahimi & Mirbargkar, 2017; Zapata & Londoño, 2019).
Despite the significance of green marketing, few empirical studies have adopted a comprehensive organizational perspective to understand its full impact (Papadas et al., 2017). Existing research has primarily focused on highly polluting industries like mining and chemicals, where environmental challenges are often viewed as constraints and costs rather than opportunities (Geels, 2015). Meanwhile, the global shift toward healthier lifestyles, seen in increased demand for natural products, has prompted companies to recalibrate their strategies toward sustainability. However, some of these sustainability claims lack accuracy, which undermines the credibility of business practices (Kahraman & Kazançoğlu, 2019).
Greenwashing—deceptive communication practices aimed at creating an overly positive perception of a company’s environmental efforts (Torelli et al., 2020)—is a growing global concern as companies increasingly present exaggerated, or false environmental information to enhance their image. Such practices mislead consumers, divert resources from sustainability efforts, and undermine corporate credibility and trust in green initiatives. Studies indicate that social, regulatory, and market pressures have pushed companies to communicate sustainable practices, but discrepancies often exist between these claims and corporate actions (Free et al., 2024). Understanding greenwashing is particularly critical in emerging markets, where environmental regulations and sustainability awareness are still developing (Gidage et al., 2024).
Research on greenwashing is essential because it distorts public and investor perceptions, impedes genuine sustainable development, and affects corporate legitimacy (Yoganandham et al., 2024). As expectations for corporate sustainability grow and policies like China’s “Green Credit Policy” are enacted, it is vital to explore how companies respond and improve policy effectiveness in preventing misleading practices (Kozar & Bolimowski, 2024). A lack of transparency and consistency in sustainability practices leads to a misleading corporate image, which may hinder the adoption of authentic environmental responsibility (Gupta & Singh, 2024).
Similar trends are observed in emerging markets like China, where green financial policies have driven companies to enhance the presentation of their sustainability reports to access economic benefits without making real environmental improvements (Li et al., 2024). This situation is mirrored in Colombia, where pressures to adopt sustainable practices may prompt companies to use greenwashing to gain legitimacy and funding (Yoganandham et al., 2024).
In Colombia, data from the Industrial Environmental Survey (2021), conducted by DANE (2022), reveals that 70% of manufacturing companies with environmental certifications, or ecological labels lack effective programs for efficient water usage, pollution prevention, or responsible resource management. This suggests that these certifications may serve as a facade to enhance public image rather than reflect substantial sustainability efforts. Additionally, Colombia offers fiscal incentives for companies investing in environmental sustainability, such as VAT and income tax discounts, accelerated depreciation of assets, and tariff exemptions (Paneso & Restrepo-Escobar, 2022). This concerns that companies might exaggerate or distort their environmental efforts to benefit from these incentives.
This study aims to fill a significant research gap by exploring how companies in emerging markets, particularly within Colombia’s manufacturing sector, react to increasing sustainability and regulatory demands. While much of the existing research concentrates on developed nations, there is a limited understanding of greenwashing in developing markets, where weaker regulations and lower social scrutiny create distinct challenges (Gidage et al., 2024). To bridge this gap, the study investigates the question: what is the influence of strategic, tactical, and internal orientation on greenwashing in the Colombian manufacturing industry in 2021?
Our objectives are to: (1) identify key predictors of greenwashing by utilizing data from the Industrial Environmental Survey (2021) and applying supervised machine learning algorithms, including LASSO-regularized logistic regressions and random forests; (ii) detect significant variables associated with organizational focus and the phenomenon of greenwashing through the application of double machine learning; (iii) analyze the influence of strategic, tactical, and internal orientations on the adoption of greenwashing practices, considering the perspective of corporate legitimacy in the cognitive domain.
To achieve these objectives, the study employs a methodological approach based on supervised machine learning algorithms. Specifically, LASSO-regularized logistic regressions and random forests are used to identify the most significant predictors of greenwashing, while the double machine learning method pinpoints variables associated with organizational focus. Ultimately, this research provides a more comprehensive understanding of greenwashing practices in the Colombian manufacturing industry, examining how green marketing approaches influence corporate legitimacy in emerging markets.
Literature Review
Emergence of Greenwashing
The definition of green marketing has evolved significantly over time, as shown by Dangelico and Vocalelli (2017). After analyzing 114 studies, they discovered that the concept of ecological marketing has evolved alongside the increasing focus on environmental sustainability. Groening et al. (2018) build on this concept, suggesting that green marketing encompasses consumer-oriented actions that integrate various marketing activities to showcase a company’s commitment to minimizing the environmental impact of its products and services.
Within this context, Papadas et al. (2017) identify three fundamental pillars of green marketing:
– Strategic orientation involves long-term initiatives and policies centered on a company’s environmental strategy, proactive approaches to sustainability, and active engagement with external environmental stakeholders.
– Tactical orientation: Short-term actions aiming to adapt the traditional marketing mix toward a more ecological strategy, including decisions to reduce environmental footprints and communicate the benefits of environmentally responsible products.
– Internal orientation: The internalization of environmental values across the company through employee training and environmental leadership.
Increasing concerns about the environment, particularly considering the climate crisis, have led some companies to employ “greenwashing” tactics, presenting themselves as environmentally responsible while occasionally spreading inaccurate or misleading information (Rahim et al., 2019).
In many developing countries, where environmental regulations are often less stringent, companies fail to live up to their ecological advertising claims. Despite the growing interest in emerging economies, empirical research on green advertising has been relatively scarce over the past three decades (Agarwal & Kumar, 2020).
Companies often use selective positive disclosures to highlight their environmental efforts, sometimes with the intent of diverting attention from activities that could harm their overall environmental performance. These disclosures can legally enhance the company’s perceived environmental practices (Hora & Subramanian, 2019).
Consequently, “greenwashing” has become embedded in market language, referring to companies’ dissemination of false, or incomplete information to appear environmentally responsible (Rahim et al., 2019).
Greenwashing is connected to the theory of corporate legitimacy, which is divided into three types: cognitive, pragmatic, and moral legitimacy. Cognitive legitimacy is rooted in shared beliefs within a company’s social environment. Moral legitimacy stems from ethical evaluations of the firm’s conduct, while pragmatic legitimacy is grounded in stakeholders’ perceived benefits from the company (De Freitas Netto et al., 2020).
Current literature explores greenwashing’s impact from various angles, such as its effects on corporate reputation (Hora & Subramanian, 2019), its relationship with corporate social responsibility (Li et al., 2024).
Measuring Greenwashing
According to Ghitti et al. (2023), measuring greenwashing presents an empirical challenge. Selective disclosure, which emphasizes the partial reporting of environmental metrics, frequently fails to account for the risk of greenwashing caused by exaggerated claims or overstatements in the reported data.
Testa et al. (2018) propose using ESG (Environmental, Social, and Governance) ratings, which assess companies’ performance in these areas to detect greenwashing. However, inconsistencies across standards and misalignments in key performance indicators related to environmental communication and practices limit the reliability of this approach.
To address these limitations, Ghitti et al. (2023) introduced a technique that compares ex-ante intentions and ex-post results of environmental policies. A significant disparity between green intentions and outcomes serves as an indicator of greenwashing.
Also, recent research underscores the critical impact of Digital Green Strategic Orientation (DGSO) on promoting sustainable business practices through digital transformation. By focusing on green market orientation, digital green technology, and alignment with government policies, organizations enhance their digital green innovation performance. The integration of DGSO with Digital Green Business Model Innovation (DGBMI) not only fosters efficiency but also strengthens the adaptability of firms to environmental challenges. Such frameworks demonstrate how companies can leverage digital capabilities to boost sustainability efforts, innovation capacity, and competitive advantage in an evolving market landscape (Yin et al., 2024).
Causes of Greenwashing
Competitive Pressure
Yang et al. (2020) suggest that companies use greenwashing to gain a competitive edge by projecting an eco-friendly image, thereby increasing market share. Consumer demand for environmental responsibility further pressures companies to adopt green behaviors (Farooq & Wicaksono, 2021; Testa et al., 2018).
Environmental Regulation and Government Policies
K. Zhang et al. (2023) examine the influence of political connections and environmental subsidies on greenwashing in China. Their findings reveal that companies receiving environmental subsidies can inhibit greenwashing, using various rhetorical tactics and prioritizing resource and waste management to enhance their Corporate Social Responsibility (CSR).
Huang et al. (2022) argue that to effectively control greenwashing, central governments should standardize and enforce environmental reporting, unifying disclosure formats to enhance credibility and comparability. At the local level, governments need to enhance corporate oversight, increase penalties for violations, and encourage companies to adopt truly sustainable practices for social recognition.
It is crucial to recognize that environmental conservation efforts are often specific and mediated by conflicts of interest (Vucetich et al., 2018). Addressing greenwashing involves not only corporate strategies but also ethical, regulatory, and participatory aspects.
Companies take advantage of market opportunities by responding to the rising demand for eco-friendly products, using greenwashing to attract environmentally conscious consumers, and generate positive word-of-mouth (Yang et al., 2020).
Interaction Between Corporate Governance and Environmental Practices in Emerging Markets
Recent literature has increasingly focused on the connection between corporate governance and environmental practices in emerging markets. These studies highlight how the structure of corporate governance can play a significant role in shaping the adoption and effectiveness of environmental practices, offering a vital framework for understanding sustainability mechanisms within emerging economies.
Influence of Corporate Governance on Environmental Sustainability
Mishra et al. (2024) carried out a systematic review of the role of corporate governance in promoting environmental sustainability within emerging markets. They assert that robust corporate governance enhances sustainable practices by establishing clear policies, ensuring transparency, and fostering accountability. Effective governance plays a crucial role in aligning environmental objectives with stakeholder interests and promoting the integration of sustainable practices into corporate operations.
In emerging markets, where environmental regulations may be less stringent, corporate governance plays an even more vital role. Companies with strong governance structures are better equipped to implement proactive environmental policies, compensate for regulatory gaps, and respond to consumer and stakeholder demands for improved environmental performance.
Role of Corporate Governance in Promoting Sustainable Practices
Ali et al. (2024) explore how corporate governance drives the adoption of sustainable practices in emerging markets. Their study identifies key factors, such as board composition, board independence, and the establishment of dedicated committees (e.g., sustainability committees), as instrumental in the uptake of environmental practices. Companies with diverse and independent boards are more likely to adopt environmental initiatives, resulting in improved sustainable performance.
The study also highlights the significant role of organizational culture and leadership in promoting sustainability. Leaders who are committed to sustainability can shape corporate decisions and instill a culture of environmental responsibility. In emerging markets, where sustainable practices may not be deeply rooted, leadership and corporate governance support are essential to driving substantive change (Tansan et al., 2023).
Impact of Corporate Governance on Sustainability Performance
Javeed and Lefen (2019) provide empirical evidence on the connection between corporate governance and sustainability performance in emerging markets. Their analysis shows that companies with superior corporate governance practices not only achieve better financial outcomes but also exhibit stronger environmental performance. This suggests that sound governance is a key enabler of sustainability, driving efficient resource management, reducing environmental risks, and enhancing corporate reputation.
Moreover, the study underscores that corporate governance practices assist companies in navigating the complex regulatory environments of emerging markets, facilitating compliance with environmental standards and improving the company’s capacity to attract sustainable investments.
Methodology
This study examines the influence of strategic, tactical, and internal orientations on greenwashing within Colombia’s manufacturing industry in 2021, drawing on data from the Industrial Environmental Survey. An advanced methodological approach is employed, featuring LASSO-regularized logistic regressions, and double machine learning (DML). The process involved data collection, and preparation, selection of key environmental variables, application of LASSO to pinpoint significant predictors, and the use of DML to enhance the precision of causal estimates. Cross-validation techniques were applied to ensure model robustness.
The methodology is grounded in supervised machine learning algorithms, such as LASSO-regularized logistic models, and random forests. These algorithms enabled the identification of the most relevant predictors related to strategic, tactical, and internal orientations concerning greenwashing within Colombia’s manufacturing sector in 2021. Additionally, the DML method was employed to identify significant variables within organizational focus and greenwashing.
Double Machine Learning (DML) is an advanced statistical technique designed to estimate the causal effect of one variable on another in the presence of confounding factors. Traditional causal inference models often face challenges like high dimensionality, and multicollinearity, which DML effectively addresses. This method combines machine learning techniques—such as LASSO regression, and random forests—with causal inference to adjust models, and account for confounding variables. Unlike traditional linear models, DML does not presume linearity between variables, allowing for the capture of complex, and nonlinear relationships (Ling et al., 2024).
In prior environmental research, Zhao et al. (2024) used DML to explore the impact of green finance on energy efficiency, while Wang et al. (2024) employed DML for a rigorous empirical analysis of how China’s Environmental Information Disclosure Pilot Policy affects carbon emission efficiency. Similarly, Ling et al. (2024) applied DML to investigate environmental issues related to pollutant emissions, and climate change in the context of China’s transportation sector.
Data
The data used in this analysis come from the Industrial Environmental Survey conducted by the Departamento Administrativo Nacional de Estadística (DANE, 2022). This sample covers six regions of Colombia: Amazon and Orinoquia, Bogotá, Caribbean, Central, Eastern, and Pacific. The representation of each region varies, with the highest concentration of companies in Bogotá (176), and the lowest in Amazon and Orinoquia (7), reflecting the business distribution across the country (DANE, 2022). The design also captures a diverse range of economic environments, including both urban and rural areas, and includes companies that are environmentally certified, or in the process of certification (Figure 1).
The sample of 681 companies is distributed across six regions of Colombia, each with varying levels of representation. Bogotá: The largest share, with 176 companies, representing a significant concentration of businesses in the capital. Central Region: A substantial portion, with 159 companies, reflects the economic activity in the heart of the country. Caribbean Region: Comprising 89 companies, this region reflects the business dynamics of the northern coastal areas. Eastern Region: Comprises 138 companies, providing insights into the economic landscape of the region. Pacific Region: Consists of 112 companies, representing the economic environment of the western coastal areas. Amazon and Orinoquía: The smallest representation, with seven companies, highlighting the lower concentration of businesses in this largely rural, and less industrialized region.
The sample of 681 companies was selected using a probabilistic, and stratified approach, considering variables such as the region of the country, the establishment’s main economic activity according to the CIIU Rev. 4 A.C. code, and the size of the establishment, measured by the number of employees, and industrial production. These variables were used to differentiate companies into categories of higher, and lower value. The combination of the first two variables led to the creation of 49 strata, along with additional clusters. The sample design aimed to achieve estimates with a coefficient of variation (sampling error) below 5% nationally within the target population (Methodology and Statistical Production Directorate [MSPD] & National Administrative Department of Statistics [DANE], 2022).
Table 1 provides a breakdown of the number of firms within various sectors of the manufacturing industry. Here’s an analysis of the distribution of companies across these sectors:
Food, beverages, and tobacco (157 firms): This sector is the most represented in the sample, comprising approximately 23.05% of all firms. This prominence likely reflects the significant role of this sector in Colombia’s manufacturing landscape, contributing to domestic consumption and export.
Manufacture of chemicals and chemical products (112 firms): This is the second-largest sector in the sample, representing around 16.45% of all firms. The chemical industry strong representation suggests its importance in Colombia’s economy, particularly in pharmaceuticals, agrochemicals, and industrial applications.
Manufacture of wood and paper products (72 firms): This sector comprises wood-based products and paper goods, accounting for approximately 10.57% of the sample. Its representation highlights the importance of forestry, and paper-related industries in Colombia.
Manufacture of rubber and plastic products (69 firms): This sector represents around 10.13% of the sample. The substantial presence of firms in rubber and plastic manufacturing reflects the industry’s role in supplying packaging, and other plastic products across multiple sectors.
Manufacture of basic metals and fabricated metal products (57 firms): Representing around 8.37% of the sample, this sector indicates a significant presence of metallurgy and metal fabrication activities, which are essential for industries such as construction, machinery, and infrastructure development.
Manufacture of non-metallic mineral products (56 firms): This sector makes up roughly 8.22% of the sample, and includes products such as ceramics, glass, and cement. The presence of these firms indicates the relevance of construction materials, and other mineral-based products in Colombia’s manufacturing sector.
Coking, refined petroleum products, and nuclear fuel (30 firms): Accounting for about 4.41% of the sample, this sector encompasses energy-intensive industries such as fuel refining and processing. These firms are likely to have considerable economic value and environmental impact.
Manufacture of textiles, wearing apparel, leather, and footwear (25 firms): Comprising around 3.67% of the sample, this sector is smaller in terms of the number of firms. However, textiles and apparel have historically played a significant role in Colombia’s manufacturing industry, potentially signaling a shift toward fewer but more consolidated or specialized companies.
Other manufacturing (103 firms): This broad category, comprising around 15.12% of the sample, includes various smaller manufacturing subsectors. The diversity within this group may suggest a wide array of niche industries or firms that do not fit into the more clearly defined sectors.
Distribution of the Sample by Sectors.
Source. Own elaboration using Stata 18 (StataCorp, 2023).
Variables
The target variable is dichotomous, taking the value of 1 for companies with environmental certifications, or eco-labels (such as ISO 14001, the Colombian Environmental Seal, District Environmental Excellence Program, or international/regional environmental certifications) that do not meet specific conditions.
These conditions include (i) investment in the management of mineral and energy resources, (ii) investment in air pollution prevention, and (iii) having a program for efficient water use and savings. The use of this variable as a measure of greenwashing is justified by its focus on certified companies that fail to comply with key practices in managing energy resources, preventing air pollution, and ensuring efficient water use. These conditions are critical for assessing the authenticity, and effectiveness of environmental practices among certified companies.
Non-compliance with these conditions may signal potential greenwashing, as companies could be leveraging certifications as part of an image strategy without substantiating them with substantial actions in crucial environmental areas. The predictors were categorized following the framework of Papadas et al. (2017) to identify which of their proposed pillars has the greatest influence on greenwashing among Colombian industrial companies (Table 2).
(1) Strategic orientation: Questions addressed the company’s commitment to sustainability and the extent to which it is integrated into its business strategy.
(2) Tactical orientation: Included questions about the implementation of cleaner production practices.
(3) Internal orientation: Focused on the corporate culture promoting sustainability, and whether employees are encouraged to adopt environmentally friendly practices.
Predictors.
Source. Own elaboration using Stata 18 (StataCorp, 2023).
Additionally, economic activity was included as a control variable due to its potential relationship with greenwashing.
LASSO is a regularization technique used to reduce overfitting in regression models. In the context of logistic regression, LASSO selects a subset of key predictors and shrinks the coefficients of less significant predictors toward zero, enhancing model performance, and interpretability.
For the double machine learning (DML) model, the configuration was as follows:
(i) Partial linear model with cross-fitting: The model uses a partial linear structure with cross-fitting, utilizing 4 folds, and 1 resampling. This approach allows for performance evaluation across different data subsets.
(ii) Multiple learners for the outcome equation: The model employs various learners:
– Y1_logit: Logistic regression.
– Y2_rforest: Random Forest with a “class” type.
– Y3_rforest: An additional random forest with a “class” type.
(iii) Learners for the treatment equation:
– D1_logit: Logistic regression.
– D2_rforest: Random Forest with a “class” type.
(iv) Cross-fitting and short-stacking:
– Cross-fitting was performed for both the outcome and treatment equations to enhance model robustness.
– The final model uses short-stacking to combine predictions from different folds, leveraging the strengths of each learner.
(v) Estimation methods: Three estimation approaches were applied:
– Robust: Uses robust standard errors for more reliable results.
– Allcombos: Estimates all possible combinations of learners to identify the optimal model.
– Spec (1) replay: Repeats the specification with the lowest mean squared error (MSE) from the cross-fitting process.
Across all algorithms applied, 90% of the sample data was designated for training purposes, with the remaining 10% set aside for validation.
Results
Logistic Regression
Table 3 presents the standardized coefficients from the LASSO-regularized logistic regression, indicating the change in the log odds of the dependent variable when a predictor increases by one standard deviation. These coefficients offer insight into the relative strength of each predictor’s effect on the target variable. The model demonstrated an ROC of approximately 77% on the training sample.
Standardized Coefficients of LASSO Logistic Regression.
Source. Own elaboration using Stata 18 (StataCorp, 2023).
The results of the LASSO logistic regression (Table 3) highlight several variables that significantly affect the likelihood of an industrial company engaging in greenwashing in Colombia:
– Expenditure on personnel for environmental protection (c1acyggg): An increase of one unit in spending on personnel focused on environmental protection activities is linked to a 0.29 standard deviation decrease in the likelihood of the company engaging in greenwashing.
– Expenditure on training and environmental education (c1acyggb): A one-unit increase in spending on training and environmental education is linked to a decrease of 0.23 standard deviations in the likelihood of greenwashing.
– Expenditure related to management processes (c1acyggc): A one-unit rise in spending on management processes corresponds to a decrease of 0.15 standard deviations in the probability of greenwashing.
– Payments for carbon footprint measurement (PTREDEMIS): A one-unit increase in payments related to carbon footprint measurement is associated with a decrease of 0.13 standard deviations in the likelihood of greenwashing.
– Expenditure on research and development (c1acygge): A one-unit rise in spending on basic, experimental, or applied R&D is associated with a 0.09 standard deviation reduction in the likelihood of greenwashing.
– Economic activity: Companies operating in the textiles, apparel, footwear, and leather sectors (coded as 8) and food, beverages, and tobacco sectors (coded as 1) demonstrate a lower likelihood of engaging in greenwashing.
Additionally, Table 3 identifies variables that are associated with a higher risk of greenwashing:
– Lack of waste measurement instruments (REGRESIDGEN): Establishments without proper instruments for measuring waste.
– Absence of an environmental contingency plan (c4ga3ip4): Lack of a contingency plan as part of environmental planning.
– No environmental risk study (c4ga3ip6): Failure to conduct an environmental risk study as part of planning.
– Lack of comprehensive solid waste plans (c4ga3ip8): Absence of solid waste management plans as a planning instrument.
– Economic activities at higher risk: Companies in sectors such as coking, refined petroleum product manufacturing, and nuclear fuel (coded as 2), non-metallic mineral products industry (coded as 6), and metallurgy and metal product manufacturing (coded as 7) exhibit a higher probability of engaging in greenwashing.
These findings underscore the importance of specific expenditures on environmental protection, and planning instruments, as well as sectoral characteristics, in influencing the propensity for greenwashing practices.
Double Machine Learning
The double machine learning (DML) model effectively estimates the relationship between greenwashing, and other variables while controlling for potential confounding factors. The results demonstrate robustness across various estimation methods.
To examine the statistical associations between strategic orientation, and greenwashing, the variables identified as key predictors by the LASSO-regularized logistic regression were selected. Specifically, the LASSO algorithm highlighted three strategic orientation variables: the existence of an environmental contingency plan as a planning instrument (c4ga3ip4), the inclusion of an environmental risk study as part of planning (c4ga3ip6), and the implementation of comprehensive solid waste management plans (c4ga3ip8).
The analysis showed that having an environmental contingency plan, an environmental risk study and comprehensive solid waste management plans were not statistically significantly associated with the likelihood of firms engaging in greenwashing. Consequently, there is insufficient evidence to suggest that these strategic orientation variables significantly impact the probability of companies adopting greenwashing behaviors (Table 4).
Double Machine Learning Predicts Predictors Associated with Tactical Orientation.
Source. Own elaboration using Stata 18 (StataCorp, 2023).
Note. Min MSE DDML; model, specification 1; y−E[y|X] = y−Y1_logit_1; D−E[D|X] = D−D1_logit_1.
The selected predictors related to tactical orientation were personnel expenses dedicated to environmental protection activities (c1acyggg); expenses associated with management processes (c1acyggc); the establishment’s use of a waste measurement instrument (REGRESIDGEN); payments for measuring the carbon footprint (PTREDEMIS); and expenditures on research, and development (basic, experimental, or applied) (c1acygge).
The results of the Double Machine Learning (DML) model, which examines the relationship between tactical orientation, and greenwashing across these predictors, indicate that not all variables are significantly associated with the likelihood of greenwashing. Specifically, “Personnel Expenses Dedicated to Environmental Protection Activities,” and “Expenses Related to Management Processes” show no significant associations, with p-values of .361 and .166, respectively. However, the presence of a “Waste Measurement Instrument,” and “Payments for Measuring the Carbon Footprint” exhibit a negative and statistically significant association at the 10% significance level (see Table 5).
Double Machine Learning Predictors (Tactical Orientation).
Source. Own elaboration using Stata 18 (StataCorp, 2023).
Note. Min MSE DDML; model, specification 1; y−E[y|X] = y−Y1_logit_1; D−E[D|X] = D−D1_logit_1.
These findings suggest that investment in carbon footprint measurement may be linked to more sustainable business practices and reduced susceptibility to greenwashing.
The variable associated with internal orientation is “Expenditures on environmental training, and education activities (c1acyggb).” The negative coefficient suggests an inverse relationship between spending on environmental training and education and the likelihood of greenwashing. The statistical significance (p = .023) indicates that this association is robust. In practical terms, an increase in spending on environmental training and education is linked to a lower probability of a company engaging in greenwashing practices. This implies that companies investing more in environmental education are less likely to exhibit greenwashing behaviors, reflecting a more genuine, and responsible approach to environmental practices (see Table 6).
Double Machine Learning Predictors Associated with Internal Orientation.
Source. Own elaboration using Stata 18 (StataCorp, 2023).
Note. Min MSE DDML; model, specification 1; y−E[y|X] = y−Y1_logit_1; D−E[D|X] = D−D1_logit_1.
Discussion
This study contributes to the growing body of literature on greenwashing by providing empirical evidence from the Colombian manufacturing sector. Responding to the call by Montgomery et al. (2023) for both identifying and preventing greenwashing, the findings shed light on organizational factors linked to the practice, specifically in environmentally certified Colombian firms. By employing an innovative methodological approach with LASSO-regularized logistic regression, and Double Machine Learning (DML), this study effectively captures the predictors associated with greenwashing, enhancing the reliability, and accuracy of the results.
Quantifying Greenwashing and Corporate Legitimacy
In addressing the empirical challenge of measuring greenwashing highlighted by Ghitti et al. (2023), this study created an indicator focusing on mineral and energy resource management, efficient water management, and investments in preventing atmospheric pollution. This allowed for a more precise understanding of the alignment—or lack thereof—between corporate actions and the reputation of manufacturing firms. In line with Hora and Subramanian (2019), the study confirms that corporate reputation does not always match genuine social responsibility efforts. Furthermore, consistent with Liute and De Giacomo (2022), the research reveals that certifications in Colombia do not necessarily guarantee environmental responsibility.
This finding has significant implications for environmental certification entities and policymakers, providing valuable insights into areas that require attention to ensure authentic sustainability practices. Specifically, the study’s results offer guidance on how strategic, tactical, and internal orientations influence greenwashing, revealing the impact on corporate legitimacy in its cognitive, moral, and pragmatic domains as defined by De Freitas Netto et al. (2020).
Strategic, Tactical, and Internal Orientations: Key Insights
The study finds that strategic orientation, such as the development of environmental contingency plans, risk studies, and solid waste management plans, does not significantly influence greenwashing practices. This suggests that while strategic plans may exist on paper, they do not necessarily translate into meaningful actions or deter deceptive practices. It indicates that the existence of these plans may be more symbolic than substantive, which aligns with previous literature indicating that such measures in emerging markets often serve legitimate purposes without affecting real environmental change (Yoganandham et al., 2024).
Conversely, tactical, and internal orientations show significant relationships with greenwashing. The presence of waste measurement instruments and payments for carbon footprint measurement are negatively associated with greenwashing, suggesting that companies committed to tangible, transparent actions are less likely to engage in misleading practices. This aligns with Testa et al. (2018), highlighting the importance of credible monitoring, and disclosure practices. Such findings emphasize that tactical approaches focusing on measurable activities can reduce the likelihood of greenwashing.
Furthermore, internal orientation, particularly expenditures on environmental training, and education, is significantly associated with a reduced likelihood of greenwashing. This suggests that fostering an environmentally conscious corporate culture through education, and training can deter greenwashing behaviors, as companies with informed and responsible employees are more likely to adopt authentic sustainability practices. This observation supports the work of Papadas et al. (2017), reinforcing internal values, and employee engagement are critical to genuine environmental responsibility.
Implications for Cognitive, Moral, and Pragmatic Legitimacy
This research delves into the cognitive, moral, and pragmatic domains of corporate legitimacy (L. Zhang et al., 2021; J. Zhang et al., 2024). Regarding cognitive legitimacy, which relies on social acceptance based on shared assumptions and perceptions, the study reveals that strategic orientation alone is not enough to influence greenwashing practices. The significant negative association between environmental training expenditures and greenwashing highlights the role of internal orientation in fostering cognitive legitimacy, as companies investing in employee awareness and training are perceived as more authentically sustainable (Papadas et al., 2017). To strengthen cognitive legitimacy, companies should ensure transparent communication about their sustainability efforts and integrate them into daily operations with quantifiable data and internationally recognized standards.
Regarding moral legitimacy, which relates to ethical perceptions of corporate behavior, the absence of environmental planning instruments like contingency plans and risk assessments is associated with a higher risk of greenwashing. This indicates a potential lack of moral commitment to sustainability, as these companies may fail to undertake necessary measures to responsibly manage their environmental impact. To enhance moral legitimacy, firms should not only implement robust environmental planning but also actively involve stakeholders, and communities in their sustainability efforts to reflect shared values, and social justice.
Pragmatic legitimacy, based on the practical benefits, and utility that sustainable practices provide to stakeholders, is evidenced in sectoral differences. The study finds that sectors such as textiles and food are less susceptible to greenwashing, whereas industries like refined petroleum and metallurgy are at higher risk. This variation could be due to pragmatic expectations placed on different industries, especially those with a higher environmental impact. Companies can build pragmatic legitimacy by transparently communicating the material benefits of their sustainable actions, providing detailed sustainability reports, and emphasizing how these practices contribute to both operational efficiency and community well-being.
Implications for Policy and Corporate Practice in Emerging Markets
This research supports the challenge posed by Agarwal and Kumar (2020) to conduct more empirical studies in emerging economies, where regulations are less strict, and greenwashing is more prevalent. The study shows that in Colombia, environmental certifications can be obtained without substantive environmental actions, suggesting a regulatory environment that may not effectively ensure genuine corporate sustainability. The findings indicate that efforts to improve environmental standards must go beyond certification and focus on implementing and enforcing actual sustainable practices.
For policymakers, the results underscore the importance of monitoring and enforcing transparency in sustainability claims (González-Argote et al., 2024), particularly in emerging markets where greenwashing is more likely due to regulatory gaps. Policies that encourage environmental training, tangible measurements like carbon footprint tracking, and robust planning mechanisms could reduce greenwashing and promote authentic corporate responsibility.
For practitioners, the findings emphasize the need to integrate environmental concerns into core business activities rather than as peripheral, or symbolic acts. Companies should focus on tangible, measurable environmental actions and foster an internal culture of sustainability, as these efforts are linked to a lower probability of greenwashing and better align with social expectations for corporate behavior.
Conclusions
This study has explored the relationship between the strategic, tactical, and internal orientations of companies and the practice of greenwashing, providing a comprehensive and practical understanding of this phenomenon within the sector. Our findings indicate that companies dedicating significant resources to environmental protection, training, and education, as well as to management processes, are less inclined to engage in greenwashing. This supports the premise that genuine investment in sustainable, responsible practices serves as an effective deterrent against deceptive behaviors. Internal investments, particularly in employee environmental training, highlight the importance of cultivating an organizational culture committed to sustainability as a key strategy to prevent greenwashing.
For managers in Colombia’s manufacturing industry, these findings highlight the importance of developing and implementing continuous sustainability training programs across all levels of the organization, fostering transparency in environmental communications, and establishing internal systems to monitor sustainable practices. Additionally, investing in measurement tools, such as carbon footprint indicators, is crucial to ensure transparency and verifiability in environmental actions, which, in turn, will strengthen the company’s social responsibility and its relationship with stakeholders.
The study underscores the need for stricter regulations and governmental measures to ensure transparency and integrity in the disclosure of environmental certifications and eco-labels—areas that appear insufficiently addressed in the Colombian manufacturing sector. This issue is further complicated by Colombia’s fiscal incentives for environmental sustainability, which may encourage companies to exaggerate or misrepresent their conservation efforts to benefit from these advantages. This underscores the need for effective environmental regulation that reinforces the authenticity of certifications and prevents their misuse as mere marketing tools.
From a broader social perspective, reducing greenwashing and promoting authentic sustainable practices have substantial implications for society. The implementation of transparent, measurable strategies not only bolsters corporate legitimacy but also contributes to environmental protection and societal well-being by reducing pollution and promoting efficient resource use. Therefore, establishing clear regulations and adopting responsible business practices benefits not just the manufacturing industry but also positively impacts sustainable community development and consumer trust.
To apply the proposed automated methods, companies should consider developing predictive models that utilize relevant data from their operations and environmental practices. A concrete step in this direction would be to integrate data on environmental training expenditures, carbon footprint assessments, and waste management into machine learning systems to identify patterns linked to sustainable behaviors. The use of these models would enable companies not only to identify areas for improving their environmental performance but also to predict future trends and make better-informed strategic decisions to avoid greenwashing.
From a theoretical standpoint, this research significantly contributes to the holistic understanding of greenwashing within Colombia’s manufacturing industry, an area previously underexplored. By applying machine learning algorithms and double machine learning techniques, this study overcomes the limitations of traditional approaches, providing a more precise and dynamic perspective on the factors that drive or deter greenwashing practices. The application of these advanced techniques not only adds substantial scientific value by offering a detailed and robust analysis of the factors influencing greenwashing but also provides a methodological framework that can be adapted to other regions and sectors for identifying and addressing deceptive sustainability practices.
Concerning cognitive legitimacy, the findings indicate that tactical and internal orientations are crucial in fostering a perception of authenticity, whereas strategic orientation alone has a limited effect on reducing greenwashing. Moral legitimacy is challenged when key environmental planning instruments, such as contingency plans and risk assessments, are absent, suggesting a lack of ethical commitment to sustainability. Pragmatic legitimacy is reflected in the sectoral variations in the propensity for greenwashing, with sectors of lower environmental impact exhibiting a reduced likelihood of engaging in deceptive practices compared to more polluting industries like oil refining and metallurgy.
Future research should delve further into the interaction between the identified variables, investigating how particular contextual factors may influence these relationships to develop a more comprehensive understanding of sustainable practices and greenwashing in other emerging economies. Moreover, integrating both quantitative and qualitative approaches could provide a more holistic perspective on the motivations and challenges companies encounter when aligning their sustainability claims with actual practices, thereby enhancing corporate legitimacy and supporting authentic sustainable development.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by resources from Fundación Universitaria Konrad Lorenz.
Ethical Approval
Ethical approval was not required as the study did not involve human participants.
Informed Consent
Informed consent was not required as the study did not involve human participants.
