Threat Interpretation and Innovation in the Context of Climate Change: An Ethical Perspective

Abstract

The ability of managers to identify and interpret challenges in the external environment is one of the micro-foundations of dynamic capabilities. The underlying literature on strategic issue interpretation suggests that interpreting environmental challenges as opportunities rather than threats is more likely to lead to proactive and innovative responses, but there are also potentially positive effects of threat interpretation, for instance high levels of commitment and risk-seeking behaviour. In this paper, I use the context of climate change to explore the link between threat interpretation and innovation in more detail. I use exploratory cluster analysis and illustrative case studies to develop a set of propositions to explain when threat interpretation can in fact encourage innovation. I identify two ethical mechanisms that positively mediate the relationship between threat interpretation and innovation: enlarged concept of responsibility to society and moral legitimacy. The paper contributes to the literature by identifying the importance of ethics in linking managerial interpretation to innovation, particularly in the context of global environmental and social challenges.

Methods Appendix

Sample and Data Sources

The CDP is a non-profit NGO that aims to improve the quality of information available globally on corporate carbon emissions and climate change strategies. The initiative now represents 767 institutional investors holding US$92 trillion in assets.⁵ The annual survey is sent to senior management executives in the largest firms globally (Reid and Toffel 2009). The CDP survey between 2003 and 2005 focused exclusively on surveying the Global 500 firms. These are the largest multinational firms by market capitalisation and include a mixture of carbon-intensive and non-carbon-intensive firms. From 2006 onwards, the sample was significantly expanded. The 99 firms in the sample are all Global 500 firms. Table 5 summarises the number of firms by sector.

Table 5 Firms in the sample by sector

I focus on the years 2003, 2006 and 2009 for three reasons. Firstly, 2003 was the first year of the CDP survey and a good baseline against which to measure future developments. Secondly, I chose 3-year intervals to improve the likelihood that there are changes in responses between one period and the next. Thirdly, I conclude the analysis in 2009 because after this year there was a shift in public and political sentiment in the aftermath of the financial crisis and the Copenhagen climate summit. The period from 2003 to 2009 thus gives a relatively undistorted view at a time when pressures for firms to deal with climate change were on the rise.

The second source of data on firm-level carbon emissions comes from Trucost. Trucost was founded in 2000 to quantify and put a price on the natural capital dependency of firms. The company developed its input–output model in 2002 in order to be able to identify the degree of dependency of companies on carbon, water, land use, waste and pollutants across the value chain. The model has since been used to estimate the environmental performance of over 4000 companies. Many of these firms do not publicly disclose the details of their environmental impacts. As such, the data from Trucost allow for comparison between companies on the basis of their environmental performance without relying solely on self-reported data (Trucost 2011).

Dependent and Independent Variables

Tables 6 and 7 summarise the coding structures for the new P and S and improved P and S variables. To capture the variation across firms, I identify three main stages of P and S development from the capability literature: identification, R&D and action (Adner and Helfat 2003; Teece 2007). In the second and third stages, I differentiate responses based on whether or not they include an external partnership. In each case, I have shown a sample quote from the data with a focus on financial sector firms as this is one of the sectors from which illustrations are drawn in the analysis.

Table 6 New products and services

Table 7 Improved P and S

Table 8 illustrates the scoring scheme for different stages of P and S development. Firms that are undergoing R&D efforts in partnership with a firm in another sector for instance are given a score (3) that is above R&D efforts that are only focused internally (2). This is because external search is an important part of the capability reconfiguration process (Lavie 2006). I assume then that partnering with organisations outside the firm is an indication of broader search efforts for opportunities and alternatives.

Table 8 P and S coding scores

Independent and Control Variables

To measure the independent and control variables in the study, I use a textual analysis programme called Linguistic Inquiry and Word Count (LIWC). LIWC calculates how people use different categories of words in written or spoken text. One of the main benefits of using LIWC is that the programme has standardised dictionaries that have been developed over several years. The default LIWC 2007 dictionary has almost 4500 words and word stems. Based on text samples from a variety of technical and non-technical settings, the 2007 dictionary is able on average to capture 86 % of words used by people in writing and speech (Pennebaker et al. 2007).⁶ I use LIWC sub-dictionaries for positive and negative emotion to analyse each firm’s open-ended responses to the question:

Do you believe climate change, the policy responses to climate change and/or adaptation to climate change represent commercial risks and/or opportunities to your company? Please provide information to explain your answer.

I include the 2003 responses in measuring both innovation and interpretation of climate change because this is the first year of firm responses. As such firms are not influenced by others’ prior responses or by their own prior responses to the survey. For later years, I measure interpretation in 2005 and 2008, i.e. a year before the measurement of innovation. This accounts for the fact that there is likely to be some delay in the effect of interpretation and it also reduces the problem of common method bias.

Summary Statistics

Figures 2, 3 and 4 illustrate how new P and S and interpretations of climate change developed over time in the sample. The sectors have been grouped together because there are several individual sectors that have few observations. In all sectors, there is an increase in new P and S development between 2003 and 2009 (Fig. 2). Among non-carbon-intensive sectors (Communications, Technology and IT), there is an increase over the period in positive interpretation. In the Energy and Utilities sectors, there is a slight decrease over time in the extent of positive interpretation. In other sectors, positive interpretation peaks in 2005 and then reduces again in 2008 (Fig. 3). There is a generally decreasing trend in terms of negative interpretation between 2003 and 2008 (Fig. 4).

Fig. 2

New P and S over time by sector

Fig. 3

Positive interpretation of risks and opportunities over time by sector

Fig. 4

Negative interpretation of risks and opportunities over time by sector