Since March 2020, large swaths of the global economy have been hit hard, not only by repeated lockdowns and quarantine rules but also by a very wide-ranging collapse of supply chains. There is every reason to expect that these developments could have a strong effect on business insolvencies and the probability of credit defaults and thus the stability of the financial system.

For central banks and commercial banks, a reliable estimation of probabilities of default is an indispensable prerequisite for many different tasks. In the field of accounting, the International Financial Reporting Standard 9 (IFRS 9) requires the creation of sufficient provisions to allow appropriate action to be taken in response to a deterioration in the quality of a credit portfolio. These provisions reflect expected credit defaults and also factor in the specific collateral structure of the credit involved. As probabilities of default increase, their effect on the size of provisions that need to be created and on the recoverability of loan collateral also increases.

In Germany, extensive financial policy, monetary policy and supervisory measures prevented a wave of insolvencies in the business sector. These measures included payment moratoria, deferment of credit valuation and more lenient rules for regulatory capital. As a result, insolvency reports were at a historic low in 2020/21. These arrangements will cease by the end of this year at the latest, however. Generally speaking, there exists a worrying possibility of a significant increase in business insolvency in the next few years, especially as the global economy is only slowly recovering from the restrictions of the last 18 months.

In light of the present circumstances, an accurate assessment or forecast of probabilities of default in the coming months is exceptionally difficult. Normally, probabilities of default are calculated using annual financial statement data. Time series enable the identification of trends and those businesses particularly at risk of default. As certain economic sectors saw drops in sales of up to 70% alongside dramatic slumps in profits, financial data from 2020 and 2021 are not comparable to data from previous years. The same is true of probabilities of default.

Over the course of a three-day expert panel, 26 specialists in the areas of financial stability and credit discussed unconventional methods of calculating probabilities of default. One central issue was the current insolvency trends in the 13 participating countries. This revealed that the same economic sectors were hit more or less everywhere: the retail, catering and hotel trades as well as agriculture. The participating experts presented a very eclectic mix of creative approaches to solving the problem of the lack of available data for estimating probabilities of default. Solutions suggested ran from a greater or even exclusive factoring in of cash flow and liquidity metrics, to the development of complex algorithms based on granular budgetary data.

All of the suggested solutions were met with great interest and sparked vigorous discussion. This expert panel could be held again at the end of next year, which would allow us to find out whether some of the proposed solutions were shown to work in practice.

Text: Charlotte Kimmel and Thomas Goswin