Page 246 - ES 2020-21_Volume-1-2 [28-01-21]

Regulatory Forbearance: An Emergency Medicine, Not Staple Diet! 229

Box 8: Inadequate Capital Raising by Banks

Using the below specification, Chopra, Subramanian, and Tantri (2020) document that banks,
both private and public, became undercapitalized after the AQR.
Y = α + β × Exposure + β × Exposure × Public Sector Bank + ν + δ + ϵ it
it
t
2
it
i
it
i
1
Data are organized at a bank-year level with years ranging from 2013 to 2019. There are two
dependent variables (i) total additions in the paid-up capital as a percentage of the bank’s
total assets, and (ii) total additions in paid-up capital minus the divergence in provisions
under the AQR as a percentage of the bank’s total assets. The independent variables are the
bank’s Exposure to AQR and an indicator variable for Public Sector Banks. Bank’s Exposure
to AQR equals the bank’s divergence in provisions due to the AQR (as a % of total assets).
and refer to bank and year fixed effects respectively. Bank fixed effects ensure that any bank-
specific time-invariant effect does not influence the results. Column (1) indicates that higher
the reported additional provisions due to the AQR, higher was the capital infusion. Column
(2) indicates that this is true only because of the capital infusion into public sector banks.
About the sufficiency of the capital infusion, the negative and much larger coefficient in
column (3) points out that the capital additions were vastly insufficient to offset the additional
provisions due to the AQR. In other words, when adjusted for additional provisions, banks’
capital actually declined.

(1) (2) (3)
Addition to Paid-Up
Capital after Adjusting
VARIABLES Additions in Paid-Up Capital for Divergence
(in %) (in %)

Bank’s Exposure to AQR 0.1859** -0.1247 -0.8141***
(2.2620) (-1.2679) (-9.9030)
Bank’s Exposure to AQR X
Public Sector Banks 0.3902**
(2.6037)

Observations 297 297 297
R-squared 0.3023 0.3133 0.4251
FE Bank & Year
Clustering Bank

Table 7: This table reports the OLS estimates of the impact of divergence in provisions
on the capital infusion using the equation above. Standard errors are clustered at the bank
level and t-statistics are reported in parentheses. *p<0.1; **p<.05; ***p<0.01. Source:
Chopra, Subramanian, and Tantri (2020).

Adverse impact on lending
7.39 As the banks were unable to raise adequate fresh capital after the clean-up, their lending
reduced. Figure 26 plots the percentage change in lending by each bank against the difference in
its gross NPAs in 2017 (two years after the commencement of the AQR) and 2015 (just before
the AQR). There was a sharp decline in lending post the increased NPAs that resulted from the