CHAPTER 6

NPARecoveryMeasures
6.1 NPA Scenario in Indian Banks

6.2 Hypothesis H I and Background

6.3 Recovery Measures - Current Practices

6.4 Testing o f Hypothesis HI

6.5 Effective o f Recovery Factors

6.6 Conclusion
 Chapter 6
NPA Recovery Measures
6.1 NPA Scenario in Indian Banks.
The quality of assets is one major problem that continues to lurk within Indian
banks. The banking industry (scheduled commercial banks) is burdened with gross NPAs
to the tune o f Rs.580 billion; this is a huge threat to the industry’s health, as banks have
to create provisions from their current profits on these loans. In India, NPAs are
considered to be at higher levels than those in other countries. The data in respect o f NPA
levels available for countries like USA, Japan, Hongkong, Korea, Taiwan and Malayasia
reveal that it ranged from 1% to 8% during 1993-94, 0.9% to 5.5% during 1994-95 and

Table 6.1: Ratio of Non-Performing Loans to Total Loans

Countries 1998 1999 2000 2001 2002 2003
Brazil 10.2 8.7 8.4 5.7 5.3 5.7
Chile 1.5 1.7 1.7 1.6 1.8 1.8
Mexico 11.3 8.9 5.8 5.1' 4.6 3.7
United Kingdom 3.2 3.0 2.5 2.6 2.6 2.2
United States 1.0 0.9 1.1 1.4 1.6 1.3
Japan 5.4 5.8 6.1 6.6 8.9 7.2
France 6.3 5.7 5 5 5 4.9
Germany 45 4.6 5.1 4.9 5 -
Greece 13.6 15.5 12.3 9.2 8.1 8.4
Italy 9.1 8.5 7.7 6.7 6.5 -
Russia 17.3 13.4 7.7 6.3 6.5 6.1
Turkey 67 9.7 9.2 29.3 17.6 14.2
Argentina 5.3 7.1 • 8.7 13.2 17.5 22.7
China - 28.5 22.4 29.8 25.5 22.0
India 14.4 14.7 12.7 11.4 10.4 8.8
Indonesia 48.6 32.9 18.8 11.9 5.8 -
Korea 7.4 8.3 6.6 2.9 1.9 2.3
Malayasia 18.6 16.6 15.4 17.8 15.9 14.8
Philippines 11.0 12.7 14.9 16.9 15.4, 15.2
Thailand 42.9 38.6 17.7 10.5 15.8 15.5
Srilanka 16.6 16.6 15 16.9 15.7 13.9
Bangladesh 40.7 41.1 34.9 31.5 28 -
Pakistan 23.1 25.9 23.5 23.3 23.7 20.7
Source: Global Financial Stability Report, Apri 2004, IMF.

201
0.85% to 3.9% during 1995-96 as against 23.6%, 19.5% and 17.3% respectively for
Indian banks during these years (RBI Bulletin, July 1999). Table 6.1 in the previous page
shows the performance o f different countries in respect o f their non-performing loans.
The number o f Public Sector Banks (PSB) having net NPAs to net advances ratios
greater than 10 per cent, was 9 out o f 27 in 1996-97 (RBI, 1996-97), but the number
declined to 3, 2 and 1 in 2001-02, 2002-03 and 2003-04 respectively. In the year 2005,
except for one bank rest o f the PSBs had this ratio below 5 percent (Table 6.2). However,
banks cannot relax with this performance and has to give a concerted effort to reduce the
NPA levels further and be at par with international banks.

Table 6.2: Ratio of Net NPA to Net Advances of PSBs

Name of Banks 2001-02 2002-03 2003-04 2004-05
Allahabad Bank 10.55 7.07 2.37 1.28
Andhra Bank 2.45 1.79 0.93 0.28
Bank of Baroda 5.68 4.81 4.95 2.82
Bank of India 6.01 5.36 4.49 2.77
Bank of Maharashtra 5.81 4.83 2.46 1.83
Canara Bank 3.89 3.59 2.89 1.88
Central Bank of India 7.98 6.74 5.57 2.98
Corporation Bank 2,31 1.65 1.80 1.12
Dena Bank 16.31 11,82 9.40 5.23
Indian Bank 8.28 6.15 2.71 1.42
Indian Overseas Bank 6,32 5.23 2.85 1.27
Oriental Bank of Commerce 3.21 144 - 1.29
Punjab and Sind Bank 11.68 10.85 9.62 8.09
Punjab National Bank 5.27 3.80 0.98 0.20
Syndicate Bank 4.53 4.29 2.57 1.59
UCO Bank 5.65 4.38 3.65 2.93
Union Bank of India 6.26 4.91 2.87 2.64
United Bank of India 7.94 ■5.52 3.75 2.43
s/ijaya Bank 6.02 2.61 0.91 0.59
State Bank of India 5.64 4.49 3.45 2.65
State Bank of Bikanar & Jaipur 5.77 4.16 1.24 1.61
State Bank of Hyderabad 4.96 3.26 0.65 0.61
State Bank of Indore 3.58 2.66 - 1.01
State Bank of Mysore 7.36 5.19 2.96 0.92
State Bank of Patiala 2.94 . 1.49 - 1.23
State Bank of Saurashtra 4.95 3.53 - 1.42
State Bank of Travancore 5.72 3.05 1.39 1.80
Source: Report on trend and Progress of Banking in India, 2004-05

202
6.2 Hypothesis III and Background
The growing blockage of huge funds in non-performing assets and the poor
recoveries has compelled bankers to consider recovery as a key performance area.
Consequently, banks have taken up various measures (refer to section 6.3 for details) to
speed up the recovery of NPAs. However, very limited empirical studies are available as
to what extent these measures are effective. Parmer (1995) has considered nine
S

independent controllable variables indicating recovery measures and found that one
single variable that is number of follow-up visits, has positive correlation with recovery.
He discovered that bank branches relied mainly on issuing notices instead of making
sincere efforts to improve recovery by personally contacting the borrowers and suggested
that maintaining continuous liaison with the borrowers is imperative.
Banks recovery action starts with a reminder letter to the borrowers. Patel &
Kaveri, 1998 observed that all borrowers do not respond and therefore it becomes
essential to visit them personally which help in recovery. Compromise and out of court
settlement has also been cited as effective measures o f recovery by Murty (2000-2001).
On the other hand, suit filing as a measure o f recovery was found to have a diminishing
impact. A study by RBI on thirty-three banks relating to measures taken for recovery
reveals that the recoveries made out of suit-filed cases during 1996, 1997 and 1998 were
7.33%, 4.74% and 4.32% of the suit-filed amounts (RBI, 1999). This shows a decreasing
trend of recovery through this route.
Thus, it appears that there is a limited research testing the effectiveness of various
measures of recovery. In the light of the above, it has been found necessary to find out
the effectiveness of the various recovery measures that are in practice. Thus, the study
has taken the following hypothesis and research question. *
Hypothesis HI: The number o f follow-up visits by bank staff has positive
correlation with recovery of NPAs.
Question 3 : What is the level of effectiveness of the various measures taken up
to recover non-performing assets (NPA) and to reduce the incidence of NPA?
The rest of the chapter is organized as follows. Section 6.3 briefly discusses the
different recovery measures currently banks use. Section 6.4 tests the hypothesis. Section
6.5 presents the findings of the research question mentioned above. Section 6.6
concludes.

203
6.3 Recovery Measures -Current Practices
Recovery o f NPA has become the critical performance area (CPA) for banks in
India since NPA reduces the profitability of a bank, weaken its financial health and erode
its solvency. The following paragraph discusses the various measures that the banks
adopt to recover the non-performing assets.
a) Reminders System - This is the simplest way to follow-up default cases.
Generally, response' to this recovery measure particularly from honest borrowers is
encouraging.
b) Personal Visits / Contacts- It is an effective and simple mode for recovery.
Banks experience has shown this as an effective measure especially in the case o f small
borrowal accounts. Involvement of staff at all levels in the bank branch is necessary.
Frequent visits are called for in case of hardcore borrowers. Over the years it is observed
that the number of visits to the borrower is going down due to cut in workforce in banks.
c) Recovery Camps - In respect of agricultural advances, recovery camps are
organized during the harvest season. In organizing recovery camps, bankers take the help
o f outsiders, particularly revenue officers in the state government, local panchayat
officials, etc. Banks give publicity of the recovery camp to be organized in the area.
d) Recovery Agents- Banks appoint outside professional agencies whose services
are utilized to ascertain the whereabouts of the borrowers and enforcement of securities.
There is some hesitancy on the part of public sector banks due to the unpleasant
experience of some financial institutions in certain cases.
e) Restructuring / Rehabilitation
The rehabilitation programme is for raising the unit’s capacity to generate
adequate internal surplus so that a certain portion of internal surplus could then be used
for reducing the irregularity in the account. The bank undertakes the following actions as
part of this programme - developing information system, review o f performance,
monitoring areas of weakness.
f) Corporate Debt Restructuring
For large borrowers enjoying multiple banking accounts and with, credit facilities
above Rs.20 crores, the Corporate Debt Restructuring (CDR) body has been created to
facilitate debt restructuring. Cases of DRT, BIFR and willful defaults, doubtful and loss
accounts and suit filed cases are outside the purview of the CDR. Thus, standard and sub-

204
 standard accounts are only eligible to seek CDR shelter. No banker/ borrower can take
recourse to any legal action during the ‘stand-still’ period of 90-180 days.
g) Loan Compromise
Once a NPA unit is evaluated and found that it cannot be revived, it becomes
incumbent on the branch to initiate recovery proceedings. Since money has got time
value whatever recovery means a branch adopts should result in quick repayment. As a
recovery means, the, bank can either go to a court praying for a decree or negotiate with
the customer for a compromised settlement. In general, it is experienced that filing suits
and recovery of dues of banks through the process of courts is cumbersome, expensive
and time consuming. A compromise in bank loans means agreeing to borrower’s request
of accepting a part of outstanding dues in the books of the bank as full and final payment
or allowing for the non-compliance of the terms of the loan, after analyzing the
alternative courses of action, genuineness and capacity o f the borrower to repay. It is also
called as voluntary debt reduction or scaling down o f dues. Loan compromise helps the
bank to recover dues with least passage of time and minimize further the loss of earning
opportunities.
h) Security Adjustment or Appropriation of Subsidy
Where readily encashable securities ^ such as Fixed deposits, LIC Policies,
Government Securities, etc. are available and if borrowers are not responding to request
for regularization of accounts, such securities are encashed after giving due notice to the
borrowers, guarantors, etc.
In case of sponsored schemes, the subsidy amount is kept in deposit accounts of
the bank for appropriation after a specified lock-in period. In such cases where subsidy is
available and the account is likely to become NPA, the subsidy amount is appropriated in
the loan account.
i) Recalling of Advances - When rigorous follow-up efforts do not yield any
fruitful result, a final recall notice is usually served asking the party to rectify the
irregularity in the account immediately. Recall notice should be sent only if the borrower
is a willful defaulter, bank liability is significantly uncovered, borrower refusing to renew
the credit limit, etc. The exercise of recalling of advances is one of the recovery strategies
which have been found effective in respect of genuine borrowers who would avoid facing
legal action through the courts.
j) Sarfaesi

205
 The Securitization and Reconstruction of Financial' assets and Enforcement of
Security Interest Act 2002 (SARFAESI) came handy for banks to accelerate the recovery
process. The Act empowers the banks to take possession, manage and sell securities
without the intervention o f a court/ tribunal.
k) Filing Suits
Filing of suit is the last resort available to a bank for recovery considering the
disadvantages of filing suit. It involves incurring court fees and expenses for providing
the witness of the bank. The long drawn process of legal proceedings results in delay in
disposal of securities, by which time, the value and conditions of the same may
deteriorate. Generally, it may take 6-7 years to get the suits decreed. The money value
would have eroded substantially by that time. Sale through court turns out to be distress
sale on account of lack o f bidders and hence no fair return is guaranteed. After filing the
suit, borrowers/ guarantors start disposing of their personal assets and hence recovery of
bank dues is likely to prove difficult. Moreover, the realized value sometimes does not
meet even the expenses towards maintenance, safe keeping of the securities and
liquidator’s/ Commissioner’s expense/ fees. (Canara Bank, 2001). In this light, the issue
that needs deliberation is that despite banks resorting to filing of suits, whether the legal
process has been beneficial in recovery of dues of banks and enforcement of credit
discipline (RBI, 1999).
l) Write-off
If it is un-remunerative either to file suit and / or continue the account in the
bank’s books, banks waive off legal action and/ or write-off outstanding dues. By waiver
of legal action, banks may take a decision not to pursue recovery through court of law.
But for write-off, the banks decide to close the account by transfer of funds from their
profits to the loan account. Write-off is proposed under certain circumstance such as : (i)
borrowers are adjudicated as insolvents, (ii) revenue authorities under the State Public
Recovery Act have recovered some amount and there is no further chances of any
recovery, (iii) both borrower and guarantor are untraceable after selling their assets, and
(iv) decrees remain unexecuted several times due to reasons beyond the control of the
bank. However, even after write-offs, a bank continues its recovery efforts.

6.4 Testing of hypothesis III

6.4.1 Methodology

206
 For testing the third hypothesis, the percentage of recovery for the NPA cases as
on the latest date has been taken from the records maintained by the recovery department
of the respective banks on their NPA cases. Altogether 75 cases have been considered.
For each of the NPA cases considered in the sample, frequency / intensity of follow-ups
was found out through interviews of recovery officers handling these cases. In particular,
the intensity was found out for personal visits made by concerned bank officials as a
follow-up measure for each of the 75 cases. The correlation between the NPA recovery
percentage and the frequency of follow-up visits for the 75 non-performing cases was
observed through the Pearson correlation co-efficient.
Further we have examined how frequency of visits influences the recovery
measures mainly compromise, compromise with legal action and legal action. We assign
1 to the case if recovery is made through compromise, 0.5 if recovery is made through
compromise with legal action and 0 if it is through legal action. The recovery measures
are then regressed against frequency of visit.
6.4.2 Analysis and Findings
The Pearson correlation coefficient between the frequency of visits and
percentage recovery is 0.45 indicating that percentage of recovery depends on the
intensity of visits. Higher the number of visits higher is the percentage of recovery. The
correlation results are given in Table 6,3 (Appendix 6.1).
An analysis of the regression between number of visits and recovery measures
reveal that higher number of visits leads to compromise between banks and the defaulting
borrowers which in turn leads to higher recovery (Table 6.4).

Table 6.4: Regression between Frequency of visit and Recovery Measure

F= 41.23(.000); R= 0.601; R2 =0.361; Adj R2 = 0.352
B t significance
Constant ' -1.24 .216
Frequency of visit 01601 6.4 .000

As can be seen from the Tables 6.5a & 6.5bb (Appendix 6.2), compromise leads
to average recovery of 57% against 17% recovery in case o f non-compromise cases.
Thus intensity of visits help the banks in recovering the NPAs. As can be seen
from the frequency distribution (Table 6.3), out of 75 cases, in 45 cases intensity of the

207
visit was satisfactory and for the rest, intensity of the visit w'as up to the expectation of
the bank officials,

6.5 Effectiveness of Recovery Measures

6.5.1 Methodology
The third research question relating to the effectiveness of the recovery measures has
been examined based on the responses of the branch heads and officials handling
recovery matters. For answering this research question, these bank officials were asked to
rate the various recovery measures on a three point scale - high, medium and low.
Opinion of the respondents were seeked on two sets of recovery measures - 1) legal and
2) non-legal. A structured questionnaire given in Appendix 3.3 has been used for the
purpose (items 2 and 3). Though seventy five questionnaires,,had been distributed only
fifty-two had been returned and could be used. These respondents belonged to the banks
covered in the study.
The effectiveness has been examined for two sets of recovery measures- 1) initial
follow-up measures taken prior to any compromise/ restructuring /up gradation and 2)
legal and non-legal measures. The extent of effectiveness has been expressed as High.
Medium and Low by the respondents for each of the recovery measures enlisted under
the two sets o f measures mentioned above. The first set o f recovery measures that has
been listed in the questionnaire (question No. 2) are a) reminders through letters , b)
reminders through phone calls, c) personal visits, d) meetings with senior bank officers
and e) recall notices as initial follow-up measures. The second set of measures listed out
are a) restructuring / rescheduling , b) compromise, c) filing of suits through DRT and
civil courts and, d) securitisation.

6.5.2 Analysis and Findings

Out of the five initial measures that has been examined for their effectiveness,
Personal Visit/ Personal Contacts is rated to be the most effective initial measure by the
respondents as can be seen from Table 6.6a. This has further been supported by our
previous finding indicated under the hypothesis. The second measure that is found to be
effective is meetings arranged with senior bank officers. It appears that counseling by
senior officers motivate borrowers for working out effective recovery measures mainly

208
compromise on the loan amount, Reminders through letters / telephone calls appear to
bear little fruit as an initial measure and does not help the bank in recovering the NPAs.

Table 6.6a: Effectiveness of follow-up measures prior to up gradation / compromise.

Follow-up Measures Effectiveness (%)

High Medium Low
a. Reminders through letters 9.6 29 61.5

b. Reminders through phone-calls 19.2 40.4 40.4

c. Personal visits/contacts 73.1 25 1.9

d. Meetings with senior bank officers 36.5 42.3 21.2

e. Recall notice 17.5 40 42.5

In the second part, the legal and non-legal measures that has been examined for
their effectiveness are restructuring -or rescheduling, compromise, filing o f suits through
DRT and civil courts, securitisation (Table 6.6b). Amongst these measures, Compromise
is considered to be the most effective. It leads to higher recovery as can be seen from
Table 6.5a (57%). Securitisation is considered to be second most effective measure where
banks are authorized to seize the assets of the borrower without involving the legal
machinery.
Table 6.6b: Effectiveness of Recovery measures- Legal and Non-legal

Recovery Measures Effectiveness (%)

High Medium Low
a. Restructuring/ rescheduling 18.4 44.8 36.8

c. Compromise 66 26 8

d. Filing Suits 11.9 47.6 40.5

e, Securitisation 43.8 31.2 25

Restructuring/ Rephasement of capital as well as restructuring of the organization

also sometimes lead to good recovery though the strategy does not seem to work in most

209
of the cases. Filing of suits does not seem to help the banks in the recovery process. Both
the amount as well as the speed of recovery is low in the case of legal measures. As a
result bank officers consider Filing Suits as the last resort after all other measures fail to
bring fruit.

6.6 Conclusion
The results show that Personal Visits or Personal Contacts and Compromise are
the most effective recovery measures. This has been established through regression
analysis of facts and figures and also from the opinion survey o f the bank officials. Both
are interrelated and personal visits lead to compromise, which in turn leads to higher
recovery.
In view of the above findings, banks should intensify the visits with the intention
of bringing the borrowers to compromise on the amount of loan to be paid. This strategy
is found to be much better than resorting to legal measures. Legal measure is time-
consuming, expensive and does not lead to higher percentage of recovery even when
court verdicts are in favour of the bank. Further it should not be denied that sometimes-
legal action like filing suits, serving legal notice also compel borrowers to come forward
for compromise. This is especially true in the case of wilful defaults. At the same time
efforts should also be made not to encourage good borrowers to seek compromise route.
To add, the mindset of the borrowers from the beginning should be tuned in such
a way that he is willing to repay rather than turn into a willful defaulter. Banks should
come forward to nurture and promote this cultural change among the borrowing public in
order to ensure an effective recycling of the precious lendable resources.

210
 Appendix 6.1
Table 6.3: Correlation Between Frequency of Visit and Recovery Percentage
Npa Recovery Freq.of Recovery Freq.of
Case percentage v/isit Case percentage Visit
1 21% Medium 57 0 Medium
2 50% High 58 0% Medium
3 6% Medium 59 20% Medium
4 80% High 60 44% High
5 0% Medium 61 30% High
6 81% Medium 62 39% High
7 0% Medium 63 48% Medium
8 ,0% High 64 50% High
9 60% High 65 0% Medium
10 60% High , 66 0 Medium
11 ' 50 High 67 40% High
12 100 High 68 0% Medium
13 99 High 69 80% High
14 0 High 70 100% High
15 0 Medium 71 40% High
16 50 Medium 72 40% High
17 ' 83 High 73 5% Medium
18 99 High 74 - 0% Medium
19 60 Low 75 90% High
20 0 High
21 100 High
22 50 High Correlation between X and Y = 0.45
23 7 Medium where X = Frequency of Vistt( 1- high, 0.5 -
24 70 High medium & 0 -Low )
25 55 High Y = Recovery percentage
26 0 Medium
27 92 High
28 93 High
29 50 High
30 46 Medium
31 0 Low
32 40 High
33 0 High
34 100 High
35 40 High
36 o' •High
37 23 High
38 41 High
39 0 High
40 37 Medium
41 0 Low
42 19 High
43 100 High
44 100 Medium
45 0 Medium
46 56 High
47 60 High
48 2 Medium
49 60 High
50 66 Medium
51 0 Medium
52 5 Low
53 10C High
54 14 Medium
5£ 6 High
55 11 High 21 !
 Appendix 6.2
Table 6.5a:
Average Percentage Recovery Table 6.5b: Average % Recovery
o jromise Cases in Non-Compromise Cases
c
o
E
Case Recovery % Recovery Measure Case Recovery % Recovery Measure
1 50% Compromise 1 21% Security adjustment+suit
2 80% Suit filed+Compromise 2 6% Securuity adjustment
3 0% Suit filed+Compromise 3 0 Suit filed,decreed
4 60% Compromise 4 81% Securuity adjustment
5 60% Compromise 5 0 Suit, hearing stage
6 50% Suit filed+Compromise 6 0 DRT
7 100% Security adjustment+compromise 7 0 DRT
8 99% Suit filed+Compromise 8 50% Security adjustment
9 83% Suit filed+Compromise 9 60% Security adjustment
10 99% Security adjustment+compromise 10 0 suit
11 90% Security adjustment+compromise 11 0 suit filed
12 50% Compromise 12 46% DRT
13 7% Compromise 13 0 Write-off(no. security)
14 70% Suit filed+Compromise 14 0 Write-off(no. security)
15 55% Compromise , 15 0 DRT
16 92% Suit filed+Compromise 16 0 DRT
17 .93% Securitisation notice+Compromise 17 100% Suit filed
18 50% Securitisation notice+Compromise 18 0 DRT
19 40% Suit filed+Compromise 19 56% Securitisation
20 0% Securitisation notice+Compromise 20 2% Suit filed
21 100% Securitisation +Compromise 21 66% Security adjustment
22 40% Suit filed+Compromise 22 0 Suit, decreed
23 23% Securitisation notice+Compromise 23 5% property attachment
24 41% Suit filed+Compromise 24 14% Securitisation Notice
25 37% Suit filed+Compromise 25 6% Security Adjustment
26 19% Security adjustment+compromise 26 0 DRT
27 100% Compromise 27 0 Securitisation Notice
28 60% Suit filed+Compromise 28 20% Securitisation Notice
29 60% Compromise 29 48% Suit filed
30 100% Security adjustment+compromise 30 0 DRT, not decreed
31 11% Compromise 31 0 DRT, not decreed
32 44% Compromise 32 0 DRT
33 30% Compromise 33 0 Suit filed
34 39% Compromise
35 50% Compromise Average Recovery =17.6
36 40% Suit filed+Compromise
37 80% Compromise
38 100% Compromise
39 40% Suit filed+Compromise
40 40% Suit filed+Compromise
41 5% Suit filed+Compromise
42 90% Compromise

Average Recovery = 57%

212