Ethics-focused dynamic capabilities: a small business perspective

Fassin (2005, p. 271): “Without condoning, one can understand that even an honest entrepreneur, facing such difficulties and pressures alone, who can see a possible last-resort solution to what is seen optimistically as a temporary problem, may be tempted to an ethically questionable expedient as the least worst option.” Also, Hannafey (2003, p. 104): “Entrepreneurs experience powerful competitive market pressures so keenly that these forces may alter their perspectives on ethics”.

We have argued for our assumption that SMEs need to change their ethical stances (not infrequently) over time in order to survive and prosper. They must do so because any preliminary (and expectedly general) code is unlikely to foresee all possible relevant and significant changes in technology, regulation, and stakeholder expectations. As a result, it is likely that the firms better equipped to change their stances will do better in many ways. As Stansbury (2009, p. 50) suggests: “If the set of ethical challenges that an organization faces are ever-changing because of technological and social changes, or because of moral imagination… [it] is therefore normatively important to incorporate a mechanism for updating codes of conduct or other behavioral standards on the basis of feedback from parties affected by them…”.

Our definition of ethical performance is based on the literature (e.g., Fassin 2005; Fisscher et al. 2005; Neubaum et al. 2004; Victor and Cullen 1988); it is part ‘do no harm’ (i.e., nonmaleficence—not breaking laws, deceiving others, discriminating, destroying trust or the natural environment, etc.) and part ‘do good’ (i.e., beneficence—improving working conditions or social fabric or trust or understanding or environmental impact, etc.…).

For example, a debate remains about whether SMEs are more or less ethical than their larger peers (Jamali et al. 2009). Research (in the US) has even proffered the idea that entrepreneurs are serial rule-breakers (Zhang and Arvey 2009) who are ‘psychically numbed’ to any further questionable behavior in business; such ideas lead to media criticisms of SME managers as being single-minded in their pursuit of success to the point that they compromise moral values when needed (Fisscher et al. 2005).

A DC has been defined as a routine, repeatable method that is expected to produce the intended results regarding changing the firm’s OCs (Helfat et al. 2007), or as Zollo and Winter (2002, p. 339) explain, DCs are “…routinized activities directed to the development and adaptation of operating routines”. Note that this is not a tautological definition from a firm performance perspective because the goal of the change is a step removed from any final performance change (i.e., there is no direct cause-and-effect of OC enhancement on firm performance due to factors like initial conditions, process costs, and the ‘Red Queen Effect’, e.g., Barnett and Pontikes 2008).

Although much research on CSR has failed to measure ethical performance (Cardy and Selvarajan 2006), surveys remain the dominant method for investigating ethical behavior (Barraquier 2011), and scenario responses are the accepted method for CSR/ethics evaluation in the US (including for SMEs, e.g., Longnecker et al. 2006). Given the limited access to secondary performance and capability data for SMEs (as most are privately held), we followed previous research in obtaining such data from them (Christmann 2000; Newbert et al. 2008).

A complementary logic borrows from the resource-based view (RBV; Barney 1986, 1991; Penrose 1959; Peteraf 1993; Wernerfelt 1984). Specifically, a more thought-out method, like a DC, is more likely to not only be valuable, but also easy to appropriate, yet rare and hard-to-imitate. DCs are planned rationally, so they are more likely to leverage current resources better (improving appropriability and rareness-as-uniqueness) and more likely to be protected (improving inimitability). When a firm factor, like a DC, has these characteristics, the RBV (and the DCV) argue that the factor is likely to provide comparative performance benefits (Barney 1996; Collis 1994; Teece et al. 1997).

In order to justify our proposed analysis of ethics-focused DCs, it is worthwhile to show that even though ethics-focused DCs may be a good theoretical possibility, that they are common in reality (so we can get a sample size worthy of our empirical analysis). Ethics-focused DCs exist in the real world for two reasons: (i) because a need exists—i.e., most firms, including SMEs, need to change their ethical positions in order to perform better socially and financially (Perrini 2006); and, (ii) because an opportunity exists—i.e., with an ethics-focused DC to guide those changes the firm will be at an advantage to rivals without such capabilities (Hess et al. 2002; Quinn and Jones 1995; Porter and Kramer 2006).

It has been argued that regular DCs should affect ethical performance positively (Marcus and Anderson 2006), although such arguments tend to hinge on the assumption of the firm’s instrumentalist approach to ethics issues like CSR and the natural environment (Aragon-Correa and Sharma 2003; Maxfield 2007; Porter and van der Linde 1995). With that instrumentalist approach, the firms can actually find competitive advantage in exploiting synergies between doing things right (e.g., leveraging core capabilities) and doing the right thing (e.g., reducing environmental impact), as Nestlé did in India by not only locally sourcing but also by providing the infrastructure in the community for its supply chain. Unfortunately, there is no work on whether ethics-focused DCs are likely to have a similar positive cross-over impact, although there are hints that that is likely, given for a firm to do social good it must first do its investors well (Marcus and Anderson 2006; Williamson et al. 2006).

More formally, we expect firms to consider changing their ethical stance in only three general circumstances: (i) when doing so is good for both competitive and ethical performance; (ii) when doing so is good for ethical but not competitive performance; and, (iii) when doing so is good for competitive but not ethical performance. [Instrumentalists only pursue case (i) opportunities; moralists only pursue cases (i) and (ii), and, opportunists only pursue cases (i) and (iii).] Let’s consider each case in turn. Case (i)—the synergy case—defines a situation that requires planning, skills, patience, and dedicated resources; in other words, it involves a DC (see Hart 1995; Porter and Kramer 2006; Porter and van der Linde 1995). The ability to find opportunities that are both financially and socially beneficial, and the ability to then change operating routines so that both benefits are realized, all within a changing competitive environment, and do so more than once, is what defines the best in dynamic capabilities. We have assumed and now argued that the majority of case (i) observations will involve an ethics-focused DC.

There are only two reasonable cases left—i.e., case (ii) and case (iii)—where the majority of ethics changes that do not involve an ethics-focused DC can occur. Case (ii) entails a financially costly change that improves ethical performance. But at many SMEs, this change may involve putting the firm’s survival in jeopardy in the short-term because such firms generally have greater sensitivity to financial strain due to their lack of assets. In order to avoid that event, case (ii) implies that the SME, besides having patient investors, has a robust resource base in the short- to medium-term to sustain the firm’s short-term-costly investment in ethics. However, case (ii) also implies that there is a strong need to build ethics-related assets in the short-term. The problem is that these two implications clash because the latter means that the firm entered an industry weakly endowed in its social capital (i.e., in its legitimacy, its reputation, its trustworthiness, and its ethical integrity). But, that situation would have been unlikely if it entered with a robust resource endowment. In other words, case (ii) is an unlikely case because the firm would not be able to afford the investment in ethical and social capital now had it not built these earlier in order to obtain a sizable resource base to now invest—i.e., the ‘unlikelihood’ arises because of this apparent contradiction. That leaves us with the most likely case where ethics are changed without the use of an ethics-focused DC. Case (iii) occurs when the firm changes ethics in a manner to improve its financial performance at detriment to its ethical performance.

MarketTools’ sample respondents are scattered and mobile and consist of members that regularly complete on-line surveys. As well, this technique is less prone to non-sampling errors such as data collection and data processing. None of the common problems in electronic survey applications applied to our application: lack of universal coverage was not an issue given the validated representative population of MarketTools; bias in sampling frames due to users versus non-users of the Internet (and e-mail) was also not a concern due to MarketTools’ database; and, compatibility problems and technical problems simply did not exist with our application.

Common method bias may pose problems for survey research that relies on self-reported data, especially when the data are provided by a single respondent, i.e., the same person at the same time. The usual concern is that these biases will artificially inflate observed relationships between focal variables. We used both procedural and statistical approaches to minimize the effects. The procedural methods we used included:

The statistical methods we used included:

When analyzing a change-related capability, effectively an ability that generates outcomes that occur later in time (i.e., when the changes have taken effect), there is some need to consider longitudinal data. This creates some difficulty for a survey-based study (unless it is done in multiple stages, years apart). We address this requirement in our one-stage survey through the relative-time referred to in the questions. Questions assessing dependent variables are referred to in current time, and questions assessing change-related capabilities are referred to in the firm’s past. This method of time-shifting is to capture the correlation of a dynamic concept with an outcome, and is not meant to prove causation. This technique provided more confidence in the discriminant validity of the constructs, separating the DCs from the final outcome measures; besides being in different time periods, they were worded as separate evaluative factors, and they did not have the highest correlation of all of the constructs involved in the analysis with each other.

Given the amount of correlation among variables, we tested for multi-collinearity in our empirical analyses. We found none, based on VIF values that were all below the 10.0 level in our regressions.

The three factors that compose the contingency variable were mean-centered prior to calculation in order to reduce the multi-collinearity effects in the OLS regression.

We also tested H3 using sub-samples and a means difference test. For SMEs with the greatest opportunism and trigger-sensitivity, those with an ethics-focused DC had a significantly higher average competitive_performance than those without (p < 0.0001). For SMEs with the least opportunism and trigger-sensitivity, those with an ethics-focused DC had a significantly higher average ethical_performance than those without (p < 0.01). Note that: SMEs with the least (greatest) opportunism and trigger-sensitivity where those firms that were both in the low-half (high-half) of the entrepreneurial_orientation score and the low-half (high-half) of the ethics_sensitivity_triggers score in the full sample; and, those firms with an ethics-focused DC had both a reported routinized_ethics_changes and an ethics-focused_dynamic_capability score in the high-half of the sample.

We also ran the six models with industry control dummy variables included; the main results remained.

The situational pressures that affected ethical performance (as shown in Table 2) varied from external to internal, and from real to perceived. Firms under pressure due to scarce capabilities performed worse competitively and ethically. Firms facing pressures due to industry hostility had worse ethical performance. Firms with self-imposed pressure to be more entrepreneurially-oriented benefited competitively at the expense of lowered ethical performance. Firms under pressure from an engaged founder benefited only competitively. For SMEs with less perceived pressure, specifically in the form of feeling more in control of their destiny, these were benefits in both competitive and ethical performance. It appears that the less pressure SMEs can put themselves under, the better balance they can find across competitive and ethical outcomes. [This set of results builds on previous work (e.g., Neubaum et al. 2004) that began to determine whether ethics were worse at new and small firms (e.g., those with less resources)]. Future work could consider some of these ‘controls’ as possible moderators. Preliminary analysis reveals a possibility of moderating effects for the ‘resource-level’ concerning financial performance and for the ‘ethics valued level’ concerning ethical performance.

Given the results of testing H3, it may be useful for policy-makers to consider SME heterogeneity, and perhaps intervene with a sensitivity to targeting those SMEs that are more opportunistic and reactive, in order to stem an overall decline in ethics. One way to do that proactively, without regulation, may be to act as procurer of their wares and to help them work on more ethical and socially beneficial products and processes (Williamson et al. 2006). A more reactive way to stem ethical declines in SMEs would be to reduce the costs and to increase the effectiveness of oversight, civil court (e.g., small claims), and mediation, in order to reduce opportunism, especially in business clusters that are deemed socially important or in activities that may involve large environmental harm. The correct policy when dealing with multiple SME types is left for future work.

We did not see any evidence of social desirability effects. For example, the distribution of the responses to the questions involving unethical behavior did not show an unusual skew to socially favorable positions. And, the results of our initial stage of survey development, where we could triangulate the financial measure responses, showed no skew towards the more favorable.