Page 248 - ES 2020-21_Volume-1-2 [28-01-21]

Regulatory Forbearance: An Emergency Medicine, Not Staple Diet! 231

(1) (2) (3) (4)
VARIABLES Log(Loan Amount)
Exposure to AQR (%) -0.3484*** -0.2826* -0.2156** -0.2899*
(-3.6049) (-1.7049) (-2.3061) (-1.7165)
Observations 167,674 167,674 167,674 167,674
R-squared 0.3044 0.3393 0.3118 0.3453
Controls No No Yes Yes
Firm X Year & Firm X Year &
FE Firm X Year Lender Firm X Year Lender
Clustering Firm Lender Firm Lender
Table 8: This table reports the OLS estimates of the equation above. Standard errors
are clustered at the firm level in odd-numbered columns and the lender level in even-
numbered ones. t-statistics are reported in parentheses. *p<0.1; **p<.05; ***p<0.01.
Source: Chopra, Subramanian, and Tantri (2020).
The coefficients come out to be negative and statistically significant. This suggests that higher
the exposure of a bank, more is the contraction in its credit supply following the AQR. In
terms of the economic magnitude, a one standard deviation increase in the exposure due to
the AQR reduces a bank’s lending by as much as 4%.
Impact of lower capital on overall lending and zombie lending:
The below regression specification from Chopra, Subramanian, and Tantri (2020), looks at
the impact of undercapitalization on lending
Log (Loan Amount ) = α+β × Exposure to AQR + β × Zombie (j,t-1) + β × Capital
it
2
1
3
ijt
AdequacyRatio + β × Exposure to AQR × Zombie (j,t-1) + β × Exposure to AQR × CAR +
5
it
it
4
it
it
β × Zombie (j,t-1) × CAR + β × Exposure to AQR × Zombie (j,t-1) × Capital AdequacyRatio it
it
it
6
7
+ β × X + ν + δ + epsilon ijt
i
8
ijt
jt
The observations are at a firm-lender-year level. The dependent variable is the natural
logarithm of the total amount lent by a lender i to a firm j in a year t. The key independent
variables are: (i) Exposure defined as the lender’s divergence in provisions due to the AQR
(as a % of total assets), (ii) CAR defined as the bank’s capital adequacy ratio, and (iii) Zombie
defined as an indicator variable that equals one if interest coverage ratio (adjusted for any
income from related party transactions) is less than one and zero otherwise in the year t-1.
and control for any time-varying firm-level trend and time-invariant lender level effects
respectively. are variables controlling for time-varying firm-lender-year level trends.
(1) (2) (3) (4)
VARIABLES Log(Loan Amount)
Exposure to AQR (%) -0.6264*** -0.6476*** -4.8615*** -4.8462***
(-2.6718) (-2.7588) (-4.3438) (-4.5303)
Capital Adequacy Ratio 0.0076 0.0070
(1.5617) (1.5173)
Exposure to AQR (%) x Capital
Adequacy Ratio 0.3763*** 0.3724***
(3.8534) (3.9497)
Exposure to AQR (%) X
Zombie 0.7216*** 0.7509*** 3.9556*** 3.9325***
(3.3870) (3.6242) (3.3956) (3.5502)