Causal Analysis Article Critique
In Jo and Harjoto’s (2012) article The Causal Effect of Corporate Governance on Corporate Social Responsibility, the authors argue that there exists a causal relationship between corporate governance (CG) and corporate social responsibility (CSR) outputs. The study finds that a lag of CSR has no effect on CG variables. However, the lag of CG variables positively effects a company’s engagement in CSR (or CSR output). Firm characteristics were appropriately controlled for, and the study was grounded in two theories, stakeholder theory and agency theory, which are often applied to international business, international relations, and international politics.
The central causal argument proposed by Jo and Harjoto (2012) is that CG causes companies to either engage or not engage in CSR practices and, therefore, influences their CSR outputs. There is a positive correlation between these two variables, indicating that, in companies where CG members are more concerned with ethical behaviors and obligations to the community, CSR measures and practices are greater. Conversely, in organizations where CG members are less concerned with these community and social priorities and obligations, CSR measures and practices decrease in number and impact. Ferejohn (2004) describes this as a structural explanation because it possesses “some underlying causal processes standing ready to police structural irregularities” (Ferejohn, 2004, p. 151). In this example, the causal processes are the CG social and communal concerns and the structural irregularities are the degree to which the company participates in CSR activities.
Fearson (1991) notes that counterfactual propositions of arguments are vital in any effort of political scientists to assess properly their causal hypotheses. In this article, Jo and Harjoto (2012) provide both implicit and explicit counterfactual arguments. Their explicit counterfactual statements are: 1) it not the case that companies with low CG scores regarding concern for communal and societal obligations also score high in CSR output; 2) it is not the case that their research revealed instances where companies with high CG scores in concern for communal and societal obligations also scored low in CSR output; and 3) it is not the case that CSR impacts CG (even though CG impacts CSR). They state that the evidence shows that there is a definitive and undeniable positive correlation between the two variables, but the relationship is one-directional with CG impacting CSR. Only one implicit counterfactual is offered by the authors and is as follows: 1) it is not the case that companies’ CG members fail to play a role in making key decisions pertinent to CSR. Therefore, it makes logical sense that their commitment to communal and social systems would be represented in their policies and decision-making priorities.
Furthermore, Jo and Harjoto (2012) identify the causal mechanism. Specifically, the mechanism responsible for causing a positive causal relationship between CG and CSR is the priorities of the people who make up CG. If the people who form, collectively, the CG board of directors (or whatever governing body the organization has) are concerned with CSR efforts, they will push for policies and initiatives that increase the company’s CSR output. Conversely, if CSR efforts are not a high priority for the company’s decision-makers and board members, then the company’s CSR output will be lacking or even nonexistent, proportional to the degree to which CSR efforts are prioritized and privileged by the CG body.
The argument put forth by Jo and Harjoto (2012) is moderately compelling. On the one hand, it accurately describes the positive relationship, to include the counterfactual and casual mechanism, between CG and CSR output within a company. This argument adheres to basic principles of logic and is backed by empirical research, with the findings appropriately used for the discussion portion of the paper.
On the other hand, there are two shortcomings in their argument. First, as Mill (1884) states “it will be necessary to bear in mind the twofold character of inquiries into the laws of phenomena; which may be either inquiries into the cause of a given effect, or into the effects or properties of a given cause” (Chapter VIII). It is not clear that the authors looked at the latter. Rather, it only seems that the cause of the effect was assessed without looking into the properties of the cause. A second, but related, criticism is that the argument proposed by the authors do not explain all the causal mechanisms for why CG positively impacts CSR. For example, it is elementary to say that there is a causal relationship between CG and CSR and that the mechanism is priorities of the CG members. A richer argument would show the causal relationship between certain CG members’ characteristics (i.e. stinginess, compassion, empathy, etc.) and CSR output. King et al. (1994) note that there are often a “plurality of causes” (p. 87), and that by understanding the various causes, a more accurate a holistic picture emerges. Multiple causes are not listed in the study and, instead, the authors only focus on the priority mechanism when, likely, the worldview, traits, characteristics, and values of the CG members are equally important (and feed into the priority mechanism). A follow-up study should candidly explore what Mill referred to as the “properties of a given cause” (Chapter VIII) to gain a richer understanding of the variables and their relationship to one another.
Fearon, J. D. (1991). Counterfactuals and hypothesis testing in political science. World politics, 43(2), 169-195.
Ferejohn, J. (2004). External and internal explanation. Problems and Methods in the Study of Politics, 144.
Jo, H., & Harjoto, M. A. (2012). The causal effect of corporate governance on corporate social responsibility. Journal of Business Ethics, 106(1), 53-72.
King, G., Keohane, R. O., & Verba, S. (1994). Designing social inquiry: Scientific inference in ualitative research. Princeton university press.
Mill, J. S. (1884). A system of logic, ratiocinative and inductive: Being a connected view of the principles of evidence and the methods of scientific investigation (Vol. 1). Longmans, green, and Company.