Non-Performing Assets and Bailouts
Mounting Non-Performing Assets (NPAs) and resulting questions about the ability of the banks with high levels of NPA to honor their liabilities to their depositors has been an important concern in the 1990s.
Recent RBI figures suggest that public sector banks have substantially higher levels of non- performing assets than private banks. For example, for the year ending in March 2003, gross NPAs represented 4.6 percent of public sector banks total assets, compared to 4.3 percent of old private sector banks, and 3.7 percent of new private sector banks. However it is not clear how well these numbers represent the true situation in these banks. There is some skepticism about the accuracy of reported NPA numbers: banks may engage in creative accounting or “evergreening,” and the current classification norms mapping loan repayment delay to NPL do not yet meet international norms.
An informative check, conducted by Topalova, is to use data from corporate balance sheets to estimate the ability of firms to repay their loans.40 Firms whose income (defined as earnings before interest, taxes, depreciation, and amortization) is less than their reported interest expense represent firms that are either defaulting, are very close to default, or would be defaulting if their loans were not “evergreened.” This share of “potential NPAs” has increased significantly in the past five years, while banks reported level of NPAs have stayed fairly constant. Topalova also finds that banks are exposed to substantial interest rate risk: a 200 basis point increase in the rate of interest could result in a four percentage point increase in the share of NPLs in the banking system.
These high levels of NPAs raise obvious concerns about the stability of individual banks. However the government’s policy so far has been to allay these concerns by simply taking over the uncovered liabilities of the failing banks, whether nationalized or private. Therefore we will measure the cost of the NPAs in terms of resources that have gone into bailing out these banks.
We are not aware of a systematic accounting of all bank failures in India since 1969. To calculate the cost of bank failures, we use data collected from annual issues of the Statistical
Tables Relating to Banks in India, starting in 1969. Unfortunately, the data were collected for 40 Topalova (2004).