by Hami Amiraslani*
Instead of perceiving ESG-oriented investments as merely a sign of “doing good”, perhaps we should accept them as an indicator of corporate social capital that engenders trust. Trust doesn’t fill coffers, but it can mitigate adverse selection and moral hazard concerns, which in turn, facilitate access to finance in capital markets.
To establish and earn trust, firms can make discretionary investments in the environmental and social (E&S) aspects of the well-established trio of Environmental, Social, and Governance. These two factors correspond to the relation between the firm and its stakeholders, which is at the heart of the notion of social capital. The governance element, on the other hand, is inherently concerned with the relationship between a firm and its shareholders.
In a paper forthcoming in the Review of Accounting Studies, my co-authors* and I examine the role of social capital in an economically important setting where managerial moral hazard is of particular concern: the corporate bond market. It is well understood that as a firm becomes financially distressed, managers, acting in the interest of shareholders, have incentives to expropriate bondholders by investing in risky projects even if these projects reduce firm value. This is often referred to as the risk shifting (or asset substitution) problem. Similarly, managers of distressed firms have incentives for cash diversion, that is to pay out cash to shareholders in the form of dividends or repurchases prior to bankruptcy.
Bondholders anticipate these agency costs and demand higher yields when contracting with the firm. Their concerns are alleviated, however, when trust is higher. If bondholders believe that managers care about the interests of a broad set of stakeholders, and not just shareholders, they expect less risk shifting or cash diversion as this could potentially jeopardise the firm’s survival. Thus, by mitigating bondholders’ agency costs, social capital can facilitate access to debt capital and lower the cost of debt financing, particularly for those firms prone to risk shifting and cash diversion.
A question of trust
We posit that social capital plays a key role in establishing trust between the firm and its broad stakeholder base. Trust, in our view, can be obtained in two ways. It can be externally acquired when a firm incorporates its activities or sets up its operations in a high-trust society or region, or it can be internally generated througinth a firm’s own investment in social capital. We refer to these two types of trust as endowed trust and earned trust, respectively.
There is evidence that documents a relation between operating in high-trust regions and certain firm-level outcomes. However, the focus of our study is on the bond market consequences of a firm’s discretionary investments in social capital that lead to the generation of earned trust. This form of social capital is driven by how a firm chooses to conduct its business activities and inherently stems from its environmental and social performance. Our study examines the role of social capital in a large sample of US publicly traded firms with secondary market bond trades and primary market bond originations between 2006 and 2019.
Examining corporate bond spreads in the secondary bond market over this period, at best we find a modest relation between E&S performance and bond spreads. Interestingly, once we adopt more stringent empirical specifications that account for general time trends, this modest relation disappears entirely, indicating that, on average, there is no relation between firms’ E&S performance and corporate bond spreads.
Nonetheless, the sensitivity of the relation between E&S performance and bond spreads to the overall time-series variation in spreads raises an important question: Are the bond market payoffs to E&S activities more pronounced during certain time periods? For instance, when overall trust is low, does a firm’s social capital becomes more valuable?
To examine this conjecture, we focus on the financial crisis. This timeframe combines an exogenous shock to firms’ default risk and an erosion of overall trust in firms, markets and institutions, thereby increasing the potential importance of firm-level social capital for bondholders. The characterisation of the financial crisis as a period during which trust in business declined is also consistent with survey evidence. For example, Edelman reports that trust in business in the US remained stable until early 2008 (58 percent in early 2008) but declined precipitously to 38 percent in early 2009.
For the financial crisis, consistent with past research, we identify two distinct periods: the credit crunch (July 2007 to July 2008), when the supply of credit suffered a shock, but general trust had not yet eroded; and the trust crisis (August 2008 through March 2009), when a shock to trust occurred.
Our results are unambiguous: During the “crisis of trust” period, secondary market bond spreads of high-E&S firms did not rise as much as the spreads of low-E&S firms. While we find some benefits associated with E&S performance during the credit crunch and the post-crisis period (mainly for financial firms), these are statistically and economically smaller than those we document during the trust crisis.
We also find that the crisis-period effect is stronger for firms that have more incentives or opportunities to engage in risk shifting or cash diversion when in distress, such as firms with a high probability of default, firms with fewer tangible assets and firms incorporated in states that do not impose payout restrictions on insolvent firms. For these firms, the implicit commitment that such activities are unlikely to occur, as captured by E&S investments, is most valuable. We also find larger effects for firms whose E&S efforts are more salient, as evidenced by the publication of a separate ESG report or the inclusion of an ESG section in their annual report.
Overall, our results suggest that corporate social capital, captured by firms’ discretionary E&S investments, affects bond pricing when it matters most: during a crisis of trust, when bondholders seek reassurance that they will not be expropriated. In such periods, a firm’s social capital is perceived as a quasi-insurance policy against excessive risk taking that can harm bondholders and other stakeholders.
With the understanding that an economy-level shock unveiled the importance of E&S activities for firms, we also investigate industry-specific shocks. In particular, we study the change in bond spreads of oil and gas firms around the BP Deepwater Horizon oil spill and of financial firms around the Wells Fargo cross-selling scandal.
The Deepwater Horizon oil spill, caused by an explosion on the firm’s drilling rig in the Gulf of Mexico in 2010, led to a temporary loss of trust in all oil and gas companies and the energy sector as a whole. The Wells Fargo cross-selling scandal was brought about by the creation of millions of fraudulent accounts on behalf of the bank’s clients without their consent. As a result of this scandal, trust in banking, which had started to recover after the financial crisis, suffered yet another breakdown.
Following these two episodes, due to an erosion of trust, bond spreads associated with firms operating in the oil and gas sector and in the financial sector increased. But interestingly, firms that had higher investments in social capital before those two episodes unravelled experienced, on average, less of an increase in their bond spreads.
We consider two additional localised shocks to trust. The first event occurred in 2013, when the Rana Plaza building collapsed in Dhaka, Bangladesh, killing more than 1,100 workers. The building housed garment factories employed by American, European and Asian apparel manufacturers and retailers, and led to a loss of trust in these companies. The second event occurred in the pharmaceutical industry. Since 2016, the industry has been plagued by claims of price gouging, and in 2017 the US Department of Health and Human Services declared the opioid crisis a public health emergency.
These events had a profound effect on the overall level of trust in the industry. Our analyses using these events also show that the bond spreads of high-E&S firms experienced less of an increase during these two low-trust periods.
Direct benefits of E&S performance
To assess whether social capital delivers direct benefits to firms, we examine the effect of E&S investments on firms’ ability to access and raise new debt capital during the crisis. We also study whether the terms of bonds originated during this period vary between high- and low-E&S firms.
Our results indicate that high-E&S firms that accessed the corporate bond market during the crisis were able to raise more funds than low-E&S firms. In terms of economic significance, high-E&S firms were able to almost double the size of their bond issues during the crisis relative to the average firm. In the post-crisis period, on the other hand, E&S efforts have no statistically detectable impact on the amount of debt capital raised in the bond market.
We also examine the effect of E&S performance on the pricing and maturity of new bond issues, both during and after the financial crisis. Consistent with our secondary market results, the analysis shows that high-E&S firms were able to raise more debt during the crisis at a lower cost of debt on average.
At the same time, by studying their bond contracts, we examine whether these firms were able to attract more favourable maturity terms during the crisis. Given bondholders’ limited flexibility in recontracting due to unanimous consent requirements, maturity can be viewed as the ultimate covenant. If E&S activities engender trust, high-E&S firms may be able to issue bonds with relatively longer maturities when prevailing trust levels have been eroded. Our results corroborate this view and show that high-E&S firms were not only able to originate bonds at lower credit spreads, but they also did so for longer maturities.
Collectively, our tests in the primary bond market provide further evidence that bondholders value the trust earned from building social capital via E&S investments. During the crisis, high-E&S firms were able to raise more debt, at more favourable interest rates, and for a longer period.
Reassessing the benefits of E&S investments
Our results provide compelling evidence that firms’ E&S efforts paid off on the bond market during the financial crisis, when trust in corporations, markets and institutions declined. Prior research has documented that similar benefits also accrued to shareholders during this period. So, a natural question is whether firms reassessed the benefits of E&S investments after the financial crisis. E&S efforts are costly, and if the associated benefits materialise only in certain crisis situations, the expected payoffs may not be sufficiently high to warrant these investments.
Our analyses indicate that firms in the lowest terciles of E&S performance prior to the financial crisis adjusted their E&S activities subsequently to catch up with firms in the top tercile. This finding is consistent with the notion that the financial crisis led to learning effects associated with the benefits of E&S investments. It also conforms to evidence on increased pressure from institutional investors in recent years for E&S investments. In the post-crisis years, firms with the weakest (pre-crisis) E&S performance levels made the largest adjustments to their E&S efforts, particularly before accessing the primary bond market.
Overall, our results suggest that earned trust delivers bond market payoffs when general levels of trust are low. Since firms have discretion in enhancing their social capital through investments in E&S activities, they can exert some influence on their cost of debt, particularly when bondholders’ potential agency costs are higher. Combined with the findings on the equity market benefits of E&S performance, our results suggest that a firm’s social capital can act as a quasi-insurance policy and increase the enterprise value of a firm when overall trust is low.
*INSEAD Assistant Professor of Accounting and Control
**first published in: knowledge.insead.edu