The Tracking of Environmental Costs: Motivations and Impacts

Abstract

Most accounting systems separately capture and accumulate one portion of the overall environmental costs of firms, while the remainder is embedded in other cost pools, such as general overhead costs or administrative costs. Little empirical evidence has been provided to explain the impacts of cost accounting systems that make a larger portion of firms' total environmental costs visible. The aim of this study is to conceptually and empirically examine the relationships among the tracking of environmental costs (TEC) by firms, their environmental motivations, and the impacts in terms of environmental and economic performance. Using survey data from a large sample of manufacturing firms, the results suggest two main conclusions. First, the TEC has an indirect influence on economic performance through environmental performance. Second, this indirect effect is influenced by the environmental motivations of the firm. More specifically, this indirect effect is greater (lesser) for firms whose motivations are predominately business-oriented (sustainability-oriented).

Acknowledgements

The authors would like to thank Marc Journeault, Maurice Gosselin, Ranjani Krishnan, and Claude Laurin, the participants at the 2010 Global Management Accounting Research Symposium (GMARS) Conference, 2011 North American Congress on Social and Environmental Accounting Research (CSEAR), and 2nd Conference of the Association de Contrôle de Gestion (ACG), as well as the two anonymous reviewers for their insightful comments and suggestions.

Notes

¹ These environmental costs include notably waste management and sewage services, pollution prevention processes, pollution abatement and control processes, reclamation and decommissioning, and environmental monitoring.

² A recent stream of research examines the valuation of external costs (societal costs) using different methods. Although very informative, these studies are outside the scope of our research that focuses specifically on internal costs (private costs). Based on the US Environmental Protection Agency (1995), private costs refer to the internal costs a business incurs or for which a business can be held accountable. Societal costs represent the external costs of a business's impact on the environment and society for which the business is not legally accountable.

³ The TEC represents one aspect of environmental costing. This aspect has been chosen because the increased visibility of environmental costs makes up a necessary condition to manage them (Anderson, 2007). Furthermore, it establishes a starting point for the integration of environmental costs within other management control systems and organisational routines.

⁴ Bansal and Roth (2000) also present a third motive for corporate ecological responsiveness, namely the legitimation motivation, which refers to compliance with social norms and regulations. It involves the desire of a firm to improve the appropriateness of its actions within an established set of regulations, norms, values, and beliefs (Suchman, 1995). This motivation is in line with the neo-institutional assumption that institutionalised practices tend to be ceremonial, symbolic, and intend to promote a legitimate image outside the organisation rather than to improve internal practices and performance (Meyer and Rowan, 1977; DiMaggio and Powell, 1983). This motivation is not considered in this study because cost accounting practices, such as TEC, represent first and foremost internal practices intended to provide managers with relevant information for purposes of decision-making and decision-facilitating (Demski and Feltham, 1976). It is less likely that TEC will be used predominately as a tool to promote a legitimate external image outside the firm like other tools, such as environmental reporting (Cho and Patten, 2007), ISO 14001 certification (Boiral and Henri, 2012), as well as other management practices such as establishing an environmental committee or environmental manager position (Bansal and Roth, 2000).

⁵ Environmental exposure refers to the firm's exposure to future environmental costs (Al-Tuwaijri et al., 2004).

⁶ The internal factors refer to (1) and (2), while the external factor refers to (2). The general organisational factors refer to (1) and (3), while the specific environmental factor refers to (2).

⁷ This response rate is similar to those reported in similar recent studies that are not based on a convenient sample. More specifically since 2000, more than 20 studies have been published in major accounting journals with response rates varying between 6.5% and 38%.

⁸ The extraction method used is maximum likelihood because it has a more formal statistical foundation than the principal factors methods (Fabrigar et al., 1999). An oblique rotation method is used (oblimin) because it permits correlations among factors (Fabrigar et al., 1999). Considering the expected correlations between the two types of environmental motivations, this method is more suitable than orthogonal rotations that constrain factors to be uncorrelated. To determine the number of factors, two methods have been used, namely the scree test and the Kaiser criterion.

⁹ This procedure was necessary in order to distinguish between a dominant score ensuing from a low-level of global intensity of environmental motivations compared with a high-level of intensity. For instance, a mean score of 2 can be obtained from a low-level of global intensity of environmental motivations (3 minus 1) or a high-level of global intensity (5 minus 3).

¹⁰ Data were collected from the NPRI provided by the federal government of Canada. This database contains information on more than 300 pollutants released and transferred from individual facilities across Canada (air, water, land and injected underground and transferred off-site to disposal, treatment, sewage, energy recovery, and recycling). Of the 319 firms in our sample, we have been able to identify 121 of them in the database. A negative and significant correlation has been established between the mean score of the eight environmental performance items as provided by the respondents of the survey and the natural log of the firm's total pollutants released minus the natural log of the average of its specific industry (−0.16, p < .05). Hence, the self-rated environmental performance is negatively correlated with the level of pollutants released against the industry. In other words, the firms that reported having good environmental performance are those that have fewer pollutants released.

¹¹ The threshold values recommended are (1) NNFI > 0.90 (Tabachnick and Fidell, 2001), (2) CFI > 0.95 (Hu and Bentler, 1995), and (3) RMSEA < 0.l0 (Browne and Cudeck, 1993).

¹² In order to validate the robustness of this indirect effect, structural equation modelling (SEM) has been conducted. As SEM allows for the control of measurement error, it can provide complementary evidence of the indirect effect. Similar to the current analysis, the results (not tabulated here but available upon request) suggest a negative non-significant direct effect of TEC on economic performance (−0.068; n.s.) and a significant positive indirect effect through environmental performance (0.180; p < .01). The only difference resides in the total effect of TEC on economic performance which is significant but only at the .10 level (0.112; p < .10).

¹³ A sensitivity analysis was conducted using a different computation of the dominant motivation score. The score is obtained simply by subtracting the mean score of business motivations from the mean score of sustainability motivations. The results remain virtually unchanged.