purposes other than conducting this exercise, and are not comprehensive.41 Nonetheless, we are optimistic that the data can provide at least the correct order of magnitude
In 1969, we have deposits data for 45 private sector banks. Between 1969 and 2000, we are able to identify twenty-one cases of bank failure, which resulted either in the liquidation of the bank, or merging of the bank with a public sector bank. (An additional 20 banks were nationalized, 14 in 1969, and 6 in 1980. We do not count these twenty nationalizations as failures). The value of the deposits at the time the bank failed can be taken as an upper bound of the cost of a bank failure. Thus, we calculate the value (in 2000 Rs.) of the deposits of these 21 banks.42 The largest single failure was Laxhmi Commercial Bank, which was merged with Canara bank in 1985, and represents 18.5 percent of the share of real deposits of failed banks. The total value of deposits for banks that failed between 1969 and 2000 is approximately 45 Billion Rs., a substantial sum.43
The total cost of recapitalization is also unknown. We also conduct a back-of-the-envelope exercise, using figures from the 1999-2000 issue of “Trends and Progress of Banking in India.44” These figures give the capital contribution of the central government to nationalized banks, as well as the amount of capital written down by the central government. While interpretation of the write-off is straightforward, the recapitalization funding require a little work. Banks earned money from the recapitalization bonds. The recapitalization subscription will, at least in theory, be returned to the government (several public sector banks have already returned capital): thus, the true cost of recapitalization is best measured by the interest income forgone by the govern- ment. The 2000-2001 issue of “Trends and Progress of Banking in India” reports the income
41 For example, the data may not correctly account for the possiblity that banks change their names, or merge while healthy. We identify the failure of private sector banks by their disappearance from our data: in many cases, these failures can be confirmed by secondary sources, but it is possible (even likely) that we have missed some failures, or evaluated as bank failures some events that were not failures. We would welcome a more careful
study of this issue. 42 For all price adjustments in this section, we use the consumer price index from the International Financial
Statistics database of the International Monetary Fund. 43 We stress again that this is an upper-bound: while the banks that failed were insolvent, the banks had other
assets, such as reserves, other performing loans, and real property, as well as deposit insurance, upon which
depositors were able to draw. 44 Reserve Ban of India (2001).