Do Shareholders value Corporate Philanthropy?: An Event Study Test
Paul C. Godfrey
Craig B. Merrill
Marriott School of Management
Brigham Young University
Jared M. Hansen
Texas Tech University
Contact Author Information
Paul C. Godfrey
789 TNRB
Brigham Young University
Provo, UT 84602
(801) 422-4522
(801) 422-0539 FAX
This draft is intended for review and feedback. Before citing results reported in this paper, please contact the authors for updated information.
ABSTRACT
Why should rational, profit maximizing managers invest in philanthropy? Godfrey (2005) proposes a new mechanism for shareholder value: Cash flow preservation vs. cash flow creation. The theory presents an intriguing rationale for philanthropy and CSR, but is as yet untested; the major contribution of this paper is to provide a direct test of the theory’s core propositions. Using a sample drawn from the KLD data base, we use an event study model to examine the effect of philanthropic giving on abnormal stock returns surrounding negative legal or regulatory events for 151 companies between 1993 and 2000. The results indicate strong support for the idea that philanthropic giving provides insurance value for shareholders. The study also finds that higher levels of participation in philanthropy do not provide extra insurance protection, and that more tangible forms of giving provide better insurance protection than less tangible forms. (144 words)
Key Words: Philanthropy, Event Study, Insurance, Corporate Social Responsibility
Do Shareholders value Corporate Philanthropy?: An Event Study Test
Why should rational, profit maximizing managers invest in philanthropy? Adherents of strict capitalism and fiduciary duties point out that philanthropy yields no tangible, transactional returns to the firm and is inconsistent with the managerial obligation to return cash to shareholders (Easterbrook & Fischel, 1991; Financial Accounting Standards Board, 1993; Friedman, 1970). The case for philanthropy comes from two different sources; strategic philanthropists argue that, while philanthropy may not generate direct economic returns, it will enhance the firm’s long term competitive position through intangible gains in reputation, legitimacy, or employee loyalty (Fombrun, Gardberg, & Barnett, 2000; Jones, 1995; Post & Waddock, 1995). Business citizenship scholars hold that businesses, granted the right to exist by the larger society, have an obligation to return part of their earnings to that society, philanthropy stands as an obligation borne by business citizens of a community (Logsdon & Wood, 2002; Waddock, 2001; Wood & Logsdon, 2002). The profitability of philanthropy is one concrete example of the larger, longer debate on the relationship of Corporate Social Responsibility (CSR) to Corporate Financial Performance (CFP).
This larger debate about whether or not CSR contributes to CFP has a long and tortured history. Godfrey (2005) proposes a new mechanism for shareholder value: Cash flow preservation vs. cash flow creation. While current models use a forward, or “front door,” model (where CSR leads to enhanced CFP), Godfrey’s model suggests a “back door” mechanism (where CSR protects and insures CFP); Godfrey’s model stands as a complement to instrumental theory[1]. The theory presents an intriguing rationale for philanthropy and CSR, but is as yet untested; the major contribution of this paper is to provide a direct test of the theory’s core propositions.
The empirical literature the CSR-CFP link is a quagmire of conflicting results. (Margolis & Walsh, 2001), in an exhaustive study of the empirical literature in this area, note that thirty years of work has not yielded a clear and unequivocal finding in either direction. Within this body of general work is a set of studies dealing with the value of behavioral signals in during crisis. This body of work is broadly consistent with the insurance model we test here. These studies consist of two broad types, both looking at idiosyncratic firm-specific profiles in a context of events common across the sample. First, broad, abstract use of CSR measures linked to economy-wide events. (WTO study, 2005; 1987 stock crash study), and second, firm specific, narrow measures against common industry events. The cotton dust study (1983); and the Chemical industry studies (Northcutt etc.)—social reporting, not activity
Both types show limited support for the Godfrey hypothesis about CSR. Neither study provides a direct test, however, as this work doesn’t specifically model philanthropy, nor does it investigate the phenomena at the level of individual firms. We employ a different level of matching: a set of firm-specific attributes (telescoping from broad to narrow) matched with firm-specific events. Thus, we look at idiosyncratic profiles in idiosyncratic contexts. A more stringent test of any central tendency and a more direct test of the idea that philanthropy or other CSR’s represent individualized insurance policies for firms.
This paper also advances and tests three theoretical extensions to Godfrey’s (2005) core model. First, does the insurance effect generalize beyond philanthropy? Philanthropy produces moral goodwill because it is a purely discretionary activity (Carroll, 1979). Godfrey asserts, but does not explore, that other discretionary CSR’s (diversity initiatives, labor practices, environmental preservation). Our data set allows us to examine the effect of several discretionary CSR’s on shareholder wealth.
Second, is the insurance effect subject to decreasing marginal returns? The economics of insurance stipulates a clear optimal premium level with investments beyond the optimal level of involvement representing dead weight losses. In terms of CSR helping insure the economic value of a firm’s intangible asset base the hard logic of optimality may break down. Uncertainty surrounding the value of the intangible (relational) assets, and the exact amount of goodwill generated through philanthropy means that rational executives may insure at some level in the neighborhood of the optimum and may err on the side of over-involvement in the hope of providing adequate “coverage.” We hope to shed light on this issue by raising the question Does increasing the level of commitment to philanthropy result in greater levels of shareholder wealth protection?
Third, we extend the insurance model by asking Does the type of giving matter? Philanthropy, in particular, is problematic because its noise/signal ratio may be rather high; does a firm donate to the symphony to get prime seats for executives and spouses, or to support community activities? If philanthropy provides a signal for a firm’s moral intentions, commitments, and coloration (Jones, 1995), then do different types of philanthropic activities have different signal values?
We present our argument and evidence as follows. We first provide the theoretical rationale and hypotheses we test. We then describe our methodology, data set, and the variables used to test our hypotheses. We next present the results of our analysis and the findings in terms of support for, or rejection of, our hypotheses. We conclude by noting the limitations of our work and considering the implications of our findings for both academics (theorists and empiricists) and practicing managers.
THEORY AND HYPOTHESIS DEVELOPMENT
Key Theoretical Constructs
Philanthropy and CSR Philanthropy is “an unconditional transfer of cash or other assets to an entity or a settlement or cancellation of its liabilities in a voluntary nonreciprocal transfer by another entity acting other than as an owner (Financial Accounting Standards Board, 1993); emphasis original)”. Corporate social responsibility is defined as actions that are not required by law but that appear to further some social good, and extend beyond the explicit transactional interests of the firm (McWilliams & Siegel, 2001). Both philanthropy and CSR constitute voluntary and discretionary actions by firms, around which stakeholders can make moral attributions of intention.
Intangible (relational) assets. Many of a firm’s most valuable assets are resources are intangible, idiosyncratic to the firm, and may have been developed over a number of years (Barney, 1991; Dierickx & Cool, 1989). A number of a firm’s resources are relationship-based because the earning potential of these assets depends on the relationships a firm has with its stakeholders and the related assessments these stakeholders make regarding some (or all) elements of the firm’s activities (Wood & Jones, 1995). Examples of these assets are reputation, loyalty, legitimacy, and trust. These relationship-based intangible assets are termed relational wealth in the Clarkson Principles of Stakeholder Management (Business Ethics Quarterly, 2002).
Moral capital/ insurance Jones (1995) offers an account of a firm’s moral reputation wherein stakeholders assess interactions between the firm and its constituents within the overall strategic context of the firm —its visions, strategies, policies, systems, etc.— that all reflect some degree of “moral coloration” by the firm’s policy maker. From these morally colored activities and contexts stakeholders impute moral values, principles, intentions, and character elements. Philanthropy constitutes an action (or interaction) between the firm and its constituents within a context rich in moral intention and coloration; the resulting moral capital represents the outcome of the process of assessment, evaluation, and imputation of moral values and character by stakeholders and communities of a firm based on its philanthropic activities.
Shareholder Value Shareholder wealth is the expected discounted value of a firm’s anticipated cash flow stream from the employment of its tangible and intangible assets, consistent with the prescriptions of the Capital Asset Pricing Model (Brealey, Myers, & Marcus, 1995). Insurance adds value to a firm’s equity price by protecting shareholders against risks, such as bankruptcy, that may make it difficult to liquidate a position profitably (Stultz, 1996).
Negative Events organizational acts adversely affect stakeholder groups and the communities to which they belong; organizational action or conduct that creates adverse impact on a stakeholder group constitutes the bad act element of an offense.
Philanthropy/CSR as Insurance
The mens rea effect of CSR. Under the common law tradition, two elements must be present for an offense to occur: a bad act and a bad mind (LaFave, 2000). A bad act requires that some action or conduct be performed that creates harm or adverse impact on another, be it an individual, group, or community. Bad acts must be accompanied by a bad mind in order to constitute an offense, for “Actus not facit reum nisi mens sit rea (an act does not make one guilty unless his mind is guilty)” (LaFave, 2000: 225). This is the doctrine of Mens Rea. The logic of corporate Mens Rea parallels organizational scholarship that views corporations as secondary moral agents (Wherhane, 1985) or as moral agents but not moral actors (Werhane, 1985; Wood & Logsdon, 2002). The doctrine of Mens Rea also underpins the U. S. sentencing guidelines, both the 1991 original and the 2004 revisions. Jurists seek for evidence of proactive ethical and preventive behaviors, as well as considering the firm’s history of ethical and compliance performance when considering punitive sanctions for corporate misdeeds.
When bad acts occur, Godfrey (2005) argues that stakeholders invoke the cognitive template suggested by the Mens Rea doctrine to help determine appropriate sanctions. As stakeholders consider possible punishments and sanctions, positive moral capital acts as character evidence on behalf of the firm. Positive moral capital provides counterfactual evidence to mitigate assessments of a bad mind; positive moral capital reduces the probability that the firm possessed the evil state of mind that justifies harsh sanctions (Strong, 1999). Positive moral capital encourages stakeholders to give the firm the benefit of the doubt regarding intentionality, knowledge, negligence, or recklessness. Positive moral capital also addresses the relevant issue of a firm’s history of moral behavior. If stakeholders follow the Mens Rea template found in the United States Sentencing Guidelines, then the nature and severity of punishments for bad acts will be significantly influenced by stakeholder assessments of a bad mind. Negative sanctions may aim to remedy the causes or consequences of adverse impacts, they may seek compensation for adverse impacts, or sanctions may aim to punish the firm and deter future adverse impacts. Remedial sanctions may include new regulations or laws aimed at limiting behavior or establishing future liability, or simply increased scrutiny and monitoring by affected stakeholder groups. Compensatory sanctions may include fines, lawsuits, or other actions aimed at financially compensating impacted groups. Punitive sanctions may include fines, incarceration for key individuals, negative publicity campaigns, or boycotts of the firm’s products or services. Punitive actions may also come in the form of decreased willingness to do business with the firm, or loss of reputation, loyalty, trust, or legitimacy.
Moral capital provides insurance-like protection to relational wealth because it fulfills the core function of an insurance contract: it protects the underlying relational wealth and earnings streams against loss of economic value arising from the risks of business operations (Trieschmann & Gustavson, 1998). Positive moral capital provides a reservoir of positive attributions that can be drawn upon to “indemnify” relational wealth against loss of value when stakeholders are adversely affected. This logic gives rise to
H1a: In the context of a negative event, declines in shareholder value will be smaller for philanthropic than non-philanthropic firms.
Godfrey’s (2005) argument centers specifically on philanthropy; however, the logic of Mens Rea and insurance value should apply for other discretionary CSR’s as well. The discretionary aspect of other CSR activities—those exceeding legal or ethical mandates—allows stakeholders to make attributions of moral coloration or character in relation to these actions as well. If discretion is the driver of moral attribution then firms having exemplary labor relationships, strong efforts to promote diversity in the workplace, a commitment to quality products, and/or a solid record of pro-active environmental stewardship should enjoy the same accumulation of moral capital. We aim to test this proposition directly through
H1b: In the context of a negative event, declines in shareholder value will be smaller for firms engaging in other discretionary CSR’s (employee relations, diversity, product quality, and environmental stewardship) than firms not engaging in discretionary CSR’s.
Philanthropy and Intangible (Relational) Assets
Firms have two types of value producing assets—tangible and intangible. Tangible assets, such as property, plant, and equipment, can be insured through traditional indemnity contracts. Intangible assets, such as brand, loyalty, trust, or legitimacy, do not possess the criteria needed for indemnification in traditional insurance markets. Rejda ((Rejda, 1992): 24) outlines a number of criteria for a functioning insurance market to exist that intangible assets do not meet. First, there must be a large number of homogeneous exposure units (objects to be insured); however, relational wealth—like trust or brand—is not homogenous but rather idiosyncratic to a particular relationship between a given firm and a given set of stakeholders. Second, probable losses must be accidental and unintentional; while some events that cause loss to the value of relationship-based intangible assets may be accidental or unintentional many of the events that negatively impact firm-stakeholder relationships are conscious and deliberate decisions (e.g., closing a plant, discontinuing a product or a product line, stretching out suppliers’ payment terms beyond reasonable limits, cutting philanthropic activity to community groups). Third, the loss must be determinable and measurable; the magnitude of loss to relational wealth is difficult to ascertain both ex ante—no original invoice exists for brand or loyalty—and ex post—losses may occur over a broad geographic region or persist through time.