The coronavirus (COVID-19) outbreak and the accompanying decline in commercial activity have once again highlighted the interconnectedness of economies around the globe, impacting virtually all players in an increasingly interwoven world. The impacts of the outbreak are still evolving, as the hospitality industry, airlines, restaurants, and other retail businesses have taken the first hit, triggered by social distancing requirements and travel restrictions. Accordingly, larger banks have widely altered their loan loss reserves in both consumer and commercial loan portfolios, as defaults are expected to increase due to business shutdowns and historically high unemployment levels.
In the absence of an effective solution to contain the pandemic, the path of the economic recovery appears to be uncertain. Despite the orchestrated relief programs deployed by governments, potential scenarios for recovery vary significantly. One such scenario predicts a W-shaped path, with oscillating periods of expansion and contraction. In such non-monotonic recovery scenarios, vulnerabilities may materialize gradually, and it would not be surprising to observe economic spillovers within and across industries. These spillovers may adversely affect bank operations, causing loan portfolios to deteriorate abruptly, and may potentially lead to credit contagion (default clustering).
Credit contagion typically occurs through two channels: counterparty risk and funding liquidity shocks. Philippe Jorion and Gaiyan Zhang define contagion via the counterparty risk channel as “[T]he default of one firm causing financial distress for its creditors, in particular when a creditor is also pushed toward default.” The extent to which counterparty risk creates vulnerabilities generally depends on the form of contractual obligations, i.e., whether the firms are in a supplier-client or a pure creditor-debtor relationship. The latter imposes a relatively higher risk for the banking industry, as it is the underlying driver for correlated defaults.
Counterparty risk could also be amplified by endogenous demand or supply shocks, which could have further impact via exchange rates. For example, the pandemic has led to a significant reduction in the movement of both people and goods and caused a decrease in demand for oil. This decrease caused a decline in oil prices and resulted in a significant drop in oil revenue for countries that rely on oil exports. This drop further led to the devaluation of currencies and had an adverse impact on the cost of borrowing, which could have further implications for rolling over both sovereign debt and defaults in the private sector in those oil-dependent economies.
Alternatively, liquidity shocks are likely to occur when an increasing number of nonperforming loans cause creditors to deleverage their portfolios and trigger a flight to safety. Intensified through a feedback loop, the unwinding of risky positions can lead to correlated selling pressures across a wide variety of securities and can potentially create a slump in funding liquidity.
Ozzy Akay, et al., document that such a shock to funding liquidity is significantly linked to the occurrence of contagion. In the presence of funding shocks, loans that historically had low default correlations may experience increased co-movement in the likelihood of default and thus impose a larger risk for banks’ portfolios. Also, a reduction in risk appetite could cause a decrease in the number of market players willing to absorb the heightened risk, which may eventually make asset securitization harder to implement.
Banks may benefit from measuring credit contagion risk within their exposures, as such efforts would create timely business intelligence for decision-makers. Accordingly, institutions should consider augmenting enterprise risk models to account for these risks and modify their risk monitoring plan accordingly. The identification of credit contagion channels could also provide banks with actionable insight about where to build buffers and avert elevated losses.
While it is impossible to completely predict the wide-ranging implications of the pandemic, financial institutions should reexamine risk models considering emerging risks. These institutions should consider potential spillovers in their credit exposures within and across industries to mitigate further systemic lending issues.
Institutions should consider augmenting enterprise risk models to account for these risks and modify their risk monitoring plan accordingly.