The Financial Sector Assessment Program (FSAP) performs bank stress testing to evaluate the resilience of the banking sector to different unexpected shocks, including sharp changes in the interest rate or exchange rate. In addition to macroeconomic shocks like these, the soundness of the banking sector also depends on the soundness of bank borrowers: systemic shocks to borrowers’ ability to repay loans is transmitted to banks through corporate defaults.
For example, an interest rate shock may affect banks directly, through its impact on the income and expenses from their lending practices. In addition, if the interest rate shock affects borrowers’ ability to repay, the shock will also be transmitted to the banking sector through an increase in corporate defaults. Similarly, a negative shock to corporate earnings will manifest as higher default rates and also adversely affect bank stability.
Assessment of corporate vulnerability thus would strengthen the analysis of bank vulnerability to shocks and should play an important role in bank stress testing. Unfortunately, assessment of corporate vulnerability is rarely included in the FSAP’s standard bank stress testing.
One way to link corporate and bank vulnerability is through assessment of corporate vulnerability on an industry level. In many countries banks are required to report loan concentration by industry and limits (for example, on real estate lending) exist to cap excessive risk taking. Because different banks will have different exposures to different industries, they will be subject to different vulnerabilities, depending on the relative health of the industries they are exposed to.
Linking corporate vulnerability assessment and bank stress testing requires detailed data on corporate balance sheets to evaluate corporate vulnerability. One possible source of such data is the annual accounting statements that stock exchanges require listed companies to file.
In a recent paper I show how corporate vulnerability can be evaluated using data from publicly listed firms in Jordan. In that paper I present several measures of corporate vulnerability that can be calculated on an industry level and linked to bank stress testing through the exposures of individual banks to different industries.
One measure of corporate vulnerability on an industry level can be calculated as the proportion of firms in the industry that find it difficult to meet their interest payments with current earnings—that is, those for which the interest coverage ratio (ICR) falls below a certain threshold. For example, if the ICR falls below 1, this means that the firm does not have enough cash flow to cover its interest payments. The larger the share of firms in the industry that find it difficult to meet their interest payments, the more vulnerable that industry is, and the greater the risk that banks with higher exposure to that industry will bear.
A second useful measure of corporate vulnerability is the uncovered debt ratio (UDR), which shows the proportion of industry debt for which earnings do not adequately cover the interest payments. The measure provides a useful proxy for underlying credit quality in the corporate sector and can be calculated on an industry level. The definition for UDR that I use in the paper follows Jones and Karasulu (2006), who define it as the proportion of total industry debt for which the ICR is less than 0.75. Figure 1 shows the proportion of firms with an ICR below 0.75 and industry-level UDRs for Jordan in 2006, obtained using publicly available data on listed firms.
Figure 1. Share of firms with an interest coverage ratio below 0.75 and uncovered debt ratios by sector in Jordan, 2006
Next, corporate stress testing can be performed by making some assumptions about shocks to earnings or interest payments and calculating the impact of such shocks on the ICR and UDR. Shocks to interest rates will increase the interest payments that firms are required to pay (if their loans are short term or carry floating rates, which are common in developing countries). Thus interest rate shocks will affect firm-level ICRs and therefore industry-level UDRs. Similarly, shocks to earnings will affect firms’ cash flows and will also affect ICRs and UDRs.
Figure 2 presents two scenarios for UDRs in Jordan in 2006, based on publicly available data. The first scenario assumes a 4 percent increase in interest rates (that is, an increase of 400 basis points), and the second scenario an 8 percent increase. The figure shows that firms in the construction industry as well as those in the utilities and energy and the financial industries were vulnerable to interest rate shocks.
Figure 2. Uncovered debt ratios by sector under different interest rate scenarios in Jordan, 2006
After corporate stress testing is done, the final step is to link corporate stress testing with bank stress testing. To do that requires data on bank exposure to different industries. While these data are collected by many bank regulators, the information is often confidential. Ideally, this exercise is performed using data on nonperforming loans (NPL) by sector. But if data are only available for bank aggregate NPLs, as is most always the case, these can be used to estimate sectoral NPLs by assuming that all sectors have same baseline NPLs for each bank. Next, assume that changes in UDR as a result of stress-tests are proportional to changes in NPLs. For example if a 4% interest rate shock increased sector UDR by 10%, we assume that the sector's NPLs would increase by 10%. Thus, after the shock different sectors will have different NPLs even though the baseline was assumed to be the same. Because each bank has different sectoral exposures, they will have different increases in their NPLs as a result of the shock.
To summarize, corporate stress testing can enhance bank stress testing and provide useful information for monitoring bank stability. Because different banks have different sectoral exposures, and because different sectors exhibit different vulnerability at different times, bank stress testing that includes corporate exposures will provide a more precise evaluation of bank soundness.
Further reading:
Jones, Matthew, and Meral Karasulu. 2006. “The Korean Crisis: What Did We Know and When Did We Know It? What Stress Tests of the Corporate Sector Reveal.” IMF Working Paper 06/114, International Monetary Fund, Washington, DC.
Love, Inessa. 2010. “Corporate Vulnerability and Bank Stability: Evidence from Jordan.” Policy Research Working Paper 5502, World Bank, Washington, DC.
Join the Conversation