Productivity growth in Indian banking:

Who did the gains accrue to?

Rajeswari Sengupta∗ Harsh Vardhan†
March 2021

Abstract
In this paper we analyse the beneficiaries of productivity gains in the In-
dian banking sector during the period from 1992 to 2019. We document the
relative efficiency of different groups of banks by ownership. We find that
the Indian banking sector, particularly the public sector banks experienced
steady productivity growth from the mid 1990s till about 2010. We conduct
a detailed descriptive analysis to examine the various stakeholders that the
productivity gains have accrued to, over the years and across bank groups.
We conclude that most of the gains may have accrued to the shareholders
which for the public sector banks would mean the government. These gains
presumably helped reduce the burden on the government of capitalising the
public sector banks, especially during the 1997-2002 period of sharp rise in
non performing assets.

JEL classification: G21, G28, D24, D61

Keywords: Banking sector, Bank productivity, Beneficiaries, Efficiency gains.

∗
Assistant Professor of Economics at Indira Gandhi Institute of Development Research
(IGIDR), Film City Road, Goregaon East, Mumbai 400065. Email: [email protected]
†
Executive-In-Residence at the Centre of Financial Services, SP Jain Institute of Manage-
ment & Research, Munshi Nagar, Dadabhai Road, Andheri West, Mumbai 400058. Email ID:
[email protected].
A shorter version of this paper appeared as a blog article on Ideas for India.
Link: https://www.ideasforindia.in/topics/money-finance/productivity-growth-in-indian-banking-
who-gained.html

Electronic copy available at: https://ssrn.com/abstract=3857373

1 Introduction

An important source of economic growth is productivity growth. Productivity can

be studied at the level of the entire economy and also for specific sectors. The
banking sector is a significant pillar of the Indian economy and is the cornerstone
of financial intermediation. Since the deregulation and privatisation reforms of
the early 1990s, Indian banking has witnessed many significant changes such as
the rapid growth of privately owned banks, gradual penetration of computers and
information technology in all strata of banking, improved risk management prac-
tices, and so on. All these changes would have important implications for the
overall productivity of the banking sector.
One parameter that demonstrates these gains is employment in banking. Be-
tween 1991 and 2010, despite growing at over 15%, Indian banking did not add
to net employment. It employed about the same number of people in 2010 as it
did in 1991 (see figure 1). In fact, employment fell across both private and public
sector banks between 2000 and 2009. Yet during this period, the banking sector
balance sheet grew by 25%. Balance sheet per employee grew phenomenally be-
tween mid 1990s and 2010 (see figure 2). This hints at dramatic improvements in
productivity.1
Another indicator of productivity growth is consistent reduction in costs. Op-
erating costs as percentage of total assets began declining from mid 1990s on-
1 There are multiple categories of banks in India such as commercial banks, co-operative banks,

regional rural banks etc. In this paper we primarily focus on the scheduled commercial banks who
account for more than 95% of the credit and deposits in the entire banking system.

Electronic copy available at: https://ssrn.com/abstract=3857373

wards, implying substantial productivity gains. Since roughly 2010 however, it
appears that productivity growth in the overall banking sector has slowed down
and may have even stalled. The data also reveals some convergence that has taken
place between private and public sector banks (PSBs) in terms of productivity
gains.
In this paper, we throw some light on who have been the main beneficiaries of
these efficiency gains witnessed by the Indian banking sector in the mid 1990s to
2010 period. To the best of our knowledge, this is the first attempt to investigate
who have the productivity gains accrued to among the various stakeholders in the
banking sector.
There exist a number of studies that analyse efficiency in the Indian banking
sector in the post-1991 reform period. Majority of these studies have explored the
effect of the deregulation and liberalisation reforms of early 1990s on productivity
growth in the banking sector.2 However none of these studies has explored the
potential beneficiaries of the productivity gains. As the literature survey of Berger
and Humphrey (1997) shows, studies on bank efficiency can contribute to policy
making, academic research and can be useful for improving bank management.
The four main stakeholders who can benefit from productivity gains in bank-
ing are employees, depositors, borrowers, and shareholders. It is likely that parts
of the gains accrued to all these classes of claimants. We try to assess how much
2 Seefor example, Bhattacharyya et al. (1997), Bhattacharyya and Pal (2013), Sathye (2003),
Kumbhakar and Sarkar (2003), Das et al. (2005), Das and Ghosh (2006), Kumar and Gulati (2009),
Das and Shanmugam (2004), Sensarma (2006), Casu et al. (2010), Sanyal and Shankar (2011)
among others.

Electronic copy available at: https://ssrn.com/abstract=3857373

each of these stakeholders benefitted from the gains. It is also likely that the ben-
eficiaries and/or the extent of gains that accrued to them varied across types of
banks. We attempt to examine this variation as well.
Studying productivity gains in banking is important as it is a critical sector of
the Indian economy and its performance determines to a considerable extent the
performance of the whole economy. Banking has been the largest service sector
employer and accounts for one of the largest shares of organised employment.
Banking is also a highly regulated sector and hence the government through reg-
ulatory policies can significantly impact the behaviour and performance of banks.
Understanding not only what drives productivity but also who benefits from it
could be an important guide to regulatory policy.
Unlike most other businesses, banking has customers on both sides of the bal-
ance sheet. Depositors as well as borrowers are customers of banks and hence
have legitimate claims on any productivity gain. Owners (i.e. shareholders in-
cluding the government, considering that 70% of Indian banking is owned by the
government) and employees are the critical ‘internal’ stakeholders who also have
legitimate claims on productivity gains made by banks. With sizeable productiv-
ity gains over two decades, it will be instructive to understand how the gains were
shared between these four main stakeholders. It would provide insights into the
relative bargaining power of the stakeholders.
Our data analysis reveals that the biggest beneficiary of the productivity gains
of the Indian banking sector have been the shareholders which in the case of the
PSBs implies the government. The other stakeholders i.e. depositors, borrowers

Electronic copy available at: https://ssrn.com/abstract=3857373

and employees do not seem to have benefitted substantially from the efficiency
gains. We argue that the gains that have accrued to the government as the primary
owner of the PSBs may have in turn been used to bail out the defaulting borrowers
during times of rising non performing assets on the banks’ balance sheets. We also
conjecture that one reason for such skewed distribution of productivity gains could
be the lack of competition in the banking sector. 70% of the commercial banks
in India are owned by the government and they essentially act like one bank with
different names thereby rendering the sector non-competitive. Therefore, one way
to distribute the efficiency gains more equitably among all the stakeholders could
be to reduce the market share of the government owned banks.
In the rest of the paper we briefly discuss the patterns of productivity growth
in Indian banking in section 2. We conduct a detailed analysis of the beneficiaries
of productivity gains in section 3. Finally we conclude in section 4.

2 Productivity gains in Indian banking

In 1991, India embarked upon a policy of economic liberalisation with the ob-
jective to gradually transition into a capitalist, free market economy. A strong
and stable banking system was critical to ensure economic growth. The policy
changes made between 1993 and 1995, ushered in a dramatic shift in banking in
India. New banking licenses were issued to private sector companies. These new
private sector banks started operations in the mid 1990s.
The new private sector banks introduced an element of competition in the In-

Electronic copy available at: https://ssrn.com/abstract=3857373

dian banking landscape. They started with superior technology platforms that
helped them scale rapidly and efficiently. Public sector banks (PSBs) responded
by adopting technology through efforts on implementing core banking solution,
branch networking etc. They also downsized their overstaffed organisations by
offering ‘voluntary retirement schemes’ (VRSs) to reduce headcount. The gov-
ernment also started the process of listing the shares of the banks it owned. Listing
of shares brought the PSBs under scrutiny of the equity market participants who
demanded greater focus on commercial objectives, including productivity.
These structural changes in Indian banking in the early and mid 1990s got
reflected in significant gains in productivity. An important metric for measuring
productivity of the banking system is the cost of intermediation. Banks, as fi-
nancial intermediaries between savers and borrowers, incur costs in transforming
savings into credit. The cost of intermediation mainly consists of costs of opera-
tions of the banking system i.e. costs of manpower and establishment and other
items such as information technology (IT) related costs. The cost of intermedia-
tion must come down for the system to enjoy productivity gains.
In this paper we measure the cost of intermediation as the ratio of total op-
erating costs to average total assets (or liabilities). We also present other related
measures such as the ratio of total operating costs to average total income (sum of
net interest income and other income). Figure 3 show the evolution over time of
both the measures.3 The figure shows a secular decline in the cost of intermedia-
tion from mid 1990s until about 2010 after which it remains mostly constant and
3 All data used in our analysis have been obtained from the Reserve Bank of India.

Electronic copy available at: https://ssrn.com/abstract=3857373

goes up slightly in 2018 and 2019.
Measured as a percentage of average total assets the cost of intermediation
came down from around 3% to around 1.9% for the banking system as a whole
i.e. a gain of 110 basis points. As a percentage of the total income, the cost of
intermediation came down from around 60% in 1992 to roughly 45% by 2010 and
remained at that level until 2018 before going up in 2019. On this metric, there
is a gain of around 15% over this period. This suggests that the Indian banking
system witnessed productivity gains from 1992 until roughly 2010 after which
productivity appears to have stagnated.
As mentioned earlier, with the advent of the new private sector banks in mid
1990s, the PSBs faced acute competition.4 They responded by aggressively adopt-
ing technology. In figure 4 we show the evolution of the cost of intermediation
for four groups of banks over the period from 1992 to 2019. The four groups
are, the State Bank of India (SBI) group, PSBs other than SBI, private banks in-
cluding both old and new private banks, and foreign banks, which are branches of
international banks operating in India.5
Figure 4 reveals an interesting pattern in productivity gains. Government
owned banks including the SBI group, the old private sector banks and foreign
banks, all had similar levels of intermediation costs in early 1990s, at around 3%
4 Private banks, prior to 1995, comprised the so called ‘old private sector banks’ that were
considered too small to be nationalised and hence continued with private ownership but had or-
ganisations and operations very similar to the government owned banks.
5 The SBI group has several separate banks which were all subsidiaries of SBI until they merged

into SBI in 2018. We kept SBI and its subsidiaries as a separate group given that it is the biggest
and the most widely present bank (in terms of branches) and hence it might have some pricing
advantage.

Electronic copy available at: https://ssrn.com/abstract=3857373

of total assets. The new private sector banks started in the mid 1990s with struc-
turally lower cost levels, largely due to their superior technology platforms. As
these banks expanded rapidly into sizeable businesses from the mid 1990s on-
wards, they benefitted from the economies of scale and their operating costs came
down sharply.
The scale effects plateaued by mid 2000s as they reached a certain size. Also
the operating costs seem to have gone up from early 2000s onwards presumably
due to higher wages and salaries offered to their employees. As shown in figure
11, private banks see a sharp increase in real per employee cost between 2002
to 2008 that potentially muted their productivity gains. These costs decline in
the period thereafter but by the mid 2000s the productivity gains ran their course.
Overall operating cost levels remained more or less flat for these banks since then,
especially since 2010.
Government owned banks, both the SBI group and the non-SBI PSBs, wit-
nessed steep gains in productivity, beginning mid 1990s as reflected in the sharp
drop in the cost of intermediation. By the mid 2000s, the government owned
banks, especially the non-SBI PSBs, had lower cost of intermediation than the
private or foreign banks presumably as a result of their “catching-up” through
adoption of superior technology as well as some scale effect given that they were
much bigger in size compared to the private banks.
All four groups appear to have hit a stagnation in productivity gains by 2010.
Their cost of intermediation has stayed more or less constant since then. There is
also some convergence in the cost of intermediation across the four groups with

Electronic copy available at: https://ssrn.com/abstract=3857373

PSBs’ averaging around 1.65% in recent years, private banks at around 2.25%
with foreign banks and the SBI group in between.
Table 1 also portrays a similar picture. It reports the decade-wise averages of
the measures of efficiency gains for the different groups of banks. While for the
banking system as a whole as well as for the public sector banks there has been
a consistent decline in the cost of intermediation from 1990s to the most recent
period, the private sector banks on the other hand have experienced a stagnation
of productivity growth over the last decade or so.

3 Beneficiaries of productivity gains

The above discussion implies that in the post-reform period, the Indian banking
sector experienced significant productivity gains till about 2010. Stagnant produc-
tivity since 2010 suggests that the gains made have been permanently embedded
in the banking system i.e. they have not been diluted subsequently. It is therefore
important to understand who among the four categories of stakeholders (namely,
owners, depositors, borrowers and employees) benefitted from these producitivity
gains. In this section we analyse each of these four stakeholders and attempt to
assess their share of the productivity gains. In the absence of a plausible empirical
framework to assess this issue, we resort to descriptive analysis of the data, and
reserve a more rigorous investigation for the future.

Electronic copy available at: https://ssrn.com/abstract=3857373

3.1 Depositors

Indian commercial banks offer three types of deposits - (i) demand deposits (com-
monly known as current account deposits and are similar to checking accounts in
the US), (ii) savings account deposits and (iii) term deposits (also known as fixed
deposits). Of these, demand deposits do not carry any interest. Until 2012, inter-
est rate on savings account deposits (accounting for roughly 25-30% of a bank’s
deposit base) was regulated by the government at 4%. Even after deregulation
most banks adhered to the old levels of 4%.6 Only the term deposits are priced
based on market conditions and their pricing reflects prevailing interest rates and
competitive dynamics in the deposit business.
This implies that if the depositors benefitted from productivity gains, we would
expect the term deposit pricing to improve over time relative to a suitable refer-
ence rate. Accordingly, in figure 5 we plot the annual average interest rate on term
deposits and the 5 year government security yield from 1997 to 2019.7 Average
maturity of term deposits in Indian banks is around 2.6 years (ranges between 2.5
to 2.8 years), and hence 5 year government security is the closest reference yield
with reliable data.8 The reason for using data from 1997 is that reliable data on
6 With the exception of two small private banks whose deposit market share was less than
2%. Presumably banks continued with the low deposit rates because no bank had the incentive to
compete on price. The gain in market share was potentially seen to be minimal and losses quite
high as the entire stock of savings accounts would have to be repriced.
7 Interest on term deposits is calculated as the difference between total interest on deposits and

interest on savings account (calculated at about 3.8% on average deposits). Interest rate of term
deposits is obtained as the ratio of interest on term deposits and average term deposits across two
consecutive years. While the rate on saving account is 4% effectively the rate works out to about
3.8% as banks take the average balance between the 5th and the 25th of each month. The analysis
does not change if we take the rate to be 4%.
8 The pattern of relative pricing and hence the conclusions do not change if we use the 10 year

10

Electronic copy available at: https://ssrn.com/abstract=3857373

government security yields prior to 1997 is not available. The objective is to com-
pare the pricing of deposits with a reliable reference rate which is independent of
the competitive dynamics of the banking sector.
We find that during the period of study, especially during the mid 1990s to
2010 when the banking sector experienced producitivity growth, there was no
appreciable improvement in the term deposit pricing relative to the risk-free refer-
ence rate i.e. the G-Sec yield. This suggests that depositors may not have received
any part of the productivity gains. Average term deposit has been priced during
this period at a discount of around 50 to 70 basis points and the discount has
stayed in the same range. This implies that depositors would be better off invest-
ing their savings in government securities as opposed to putting their money in
term deposits with banks.9 The only exception was the period between 2001 to
2004 when the deposits were priced at premium to the reference yield.10
If we look at the same data across the various groups of banks, we see no
perceptible differences in deposit pricing except for foreign banks which appear to
be pricing deposits at even higher discount to the reference rate. Figure 6 presents
the relative deposit pricing for the four bank groups.
yield or the 1 year yield.
9 We would normally expect the bank deposits to be priced slightly higher than the G-Sec yield

of similar maturity to account for the credit risk in the bank.

10 This could be attributed to the intense competition and hence higher pricing of term deposits

due to the merger of a large Development Finance Institution (DFI) into a commercial bank. This
merger compelled the bank to raise large amount of deposits to meet its reserve requirements
(merger of erstwhile ICICI Ltd into ICICI Bank).

11

Electronic copy available at: https://ssrn.com/abstract=3857373

3.2 Borrowers

Bank borrowers are important stakeholders in the banking business and rightful
claimants to productivity gains. As in the case of the depositors, any sharing
ofproductivity gains by the borrowers should get reflected in the pricing of loans.
Loan prices should decline relative to an appropriate reference rate. In figure 7 we
plot the average yield on loans for all banks and the 10 year government security
yield for the period from 1997 to 2019.11
The figure shows that the loan prices have not declined relative to the G-Sec
rate and infact the loan pricing premium over the risk-free reference rate has been
more or less constant from 2005 onwards. This suggests that as in the case of
depositors, the borrowers too may not have received any share of the productivity
gains enjoyed by the banking system.
Loan prices typically include a risk premium and a maturity premium. Any
change in the average riskiness of borrowers or in the average maturity of the
loans would impact the loan premium. This implies that if during the period
under review, the average maturity of loans went up or the riskiness of the average
borrower increased, then even constant loan price premium might represent gains
to borrowers. However, the data shows that there has been no perceptible change
in average loan maturity, which has stayed constant at around 3 years. Riskiness
of the average loan is hard to assess purely from publicly available data. There is
however a case to be made, looking at the rapid build up of non performing loans
(NPLs) in the period from 2014 to 2018 that riskiness of loans was increasing
11 The pattern does not change if we use the 5 year G-Sec yield instead.

12

Electronic copy available at: https://ssrn.com/abstract=3857373

in the period from 2005 onwards which was not accompanied by a perceptible
increase in loan price premium. In the previous years (from 1997 until 2005) the
loan pricing premium appears to have increased.

3.3 Employees

The next stakeholders that we consider are employees. Employee costs (wages
and salaries) are a component of the overall cost of intermediation. Hence pro-
ductivity growth measured as a reduction in the cost of intermediation would im-
ply that employee costs would come down but the important question is, by how
much on a relative scale. In figure 9 we plot the employee costs for the banking
system as a percentage of average total assets, and total income.
We find that there has been a secular decline in employee costs from about 2%
of average total assets to around 1%. As a percentage of total income, they have
come down from around 40% to roughly 30%. It is worth noting that the share of
employee costs in the total operating costs (not shown here) also came down from
around 70% to 50%.
Employee costs have come down largely because the number of employees
declined between 1991 and 2010. To evaluate the possible share of employees in
productivity gains of the banking system, we plot the real per employee wage cost
relative to the growth in real balance sheet of the entire banking sector from 1992
to 2019 in figure 10. 12 Our hypothesis is that if the employees had shared some
12 The real parameters are computed by deflating the nominal numbers by the consumer price
index.

13

Electronic copy available at: https://ssrn.com/abstract=3857373

productivity gains, the total wage bill would come down but the per employee real
(or nominal) wages would improve by at least as much as the growth in banking
income or assets, if not by more.
The figure shows that during 1992-2019, the growth in real per employee
wages has lagged behind the growth in banking sector balance sheet. While the
real balance sheet continues to grow, real employee wage remains flat. The com-
pounded annual growth rate (CAGR) of real per employee compensation during
this period was 3.6% whereas the growth in real balance sheet of the banking sys-
tem was 7.7%. The CAGR in real wages was faster from 1992 to 2007 at 5.9%
when the real balance sheet grew at 9.4%. From 2007 to 2019 the real per em-
ployee wages have been almost stagnant, growing at only 0.7% annually when the
real balance sheet grew at 4.9%.
Figures 11 and 12 show that while for the private sector banks employee cost
went up sharply in the 2002-2008 period and declined thereafter, for the public
sector banks employee compensation grew more slowly until about 2010 after
which it become stagnant.
This suggests that employees may not have benefitted from the productiv-
ity gains given that wage growth has been consistently lower than the growth in
banking business and has been almost stagnant for the past 12 years or so.

3.4 Shareholders

The last stakeholders we consider are shareholders of the banks. For the PSBs,
the majority owner is the government which, by law, has to maintain at least 51%

14

Electronic copy available at: https://ssrn.com/abstract=3857373

ownership in these banks. For private banks, the ownerships tends to be widely
distributed as the current regulations limit any individual or entity to a 5% owner-
ship (which can go upto 10% with special approval of the regulator). Most foreign
banks operate in India as branches of the parent international banks.
Figure 13 depicts the pre-provisioning operating profits (PPOP) of the banking
system and the cost of intermediation. PPOP, also referred to as operating profit
is the profit before risk related provisioning and hence measures the gains to the
shareholders before paying for risk. As discussed earlier and shown in this figure
as well, the operating cost of the banking system has been declining during the
first part of the period under review. The productivity gains added to PPOP until
about 2007 and then operating profits stagnated as there were negligible gains
in productivity from then on. The evolution of these two parameters therefore
suggests that the cost of intermediation saved by the banks went directly into the
operating profits implying that it accrued to the shareholders. As discussed above,
it seems that other stakeholders in banks- depositors, borrowers, and employees
did not receive much of the productivity gains.
Among other uses, increases in PPOP are potentially utilised to make provi-
sion for the credit losses (losses on NPLs) that banks incur. After providing for
these losses, paying taxes, and paying dividend to shareholders (if any), the resid-
ual profits add to the capital base of the bank. In order to understand the pattern of
use of the gains in PPOP towards credit loss provisioning, ideally we would like
to look at the data on credit loss provisioning over the same time period. Unfortu-
nately this data in not available on a consistent basis. Hence, we look the level of

15

Electronic copy available at: https://ssrn.com/abstract=3857373

non performing assets (NPAs) of the banks which would be a crucial determinant
of the credit loss provisioning needed. For this we analyse three separate groups
of banks: government owned banks (PSBs and SBI group together), private banks
and foreign banks.
Figure 14 shows the three variables- cost of intermediation, PPOP and gross
NPAs of government owned banks as a group over the period of 1997 to 2018.
The figure can be interpreted as follows. PSBs enjoyed a steady growth in pro-
ductivity from mid 1990s onwards till about 2010 as reflected in the decline in
their operating costs. PPOP as a proportion of average total assets increased with
the productivity gains in these banks. During 1997-2005, part of the increased
PPOP was presumably used to bring down the high level of gross NPAs thathad
built up in the PSBs in the mid 1990s. This would imply that the productivity
gains enjoyed by these banks during this period effectively went to the defaulting
borrowers via PPOP and NPA provisioning. Consequently the government as the
owner of the PSBs had to put in less additional capital into these banks.
With the stagnation in productivity from 2010 onwards, the growth in PPOP
also got stalled. When the next round of NPA buildup started from 2011 onwards,
there were not enough productivity gains to address the rise in gross NPAs. As a
result, these banks had to raise large amounts of external capital from the govern-
ment as the internally generated capital was not enough to provide for the NPAs.
In figure 15 we look at the same parameters for private banks. Productivity
gains for the private banks as reflected through lower cost of intermediation were
mostly during the period 1997-2005 and even then, were not as substantial as

16

Electronic copy available at: https://ssrn.com/abstract=3857373

those for PSBs. The productivity gains enjoyed by the private sector banks were
lower than the PSBs during the 2002-2010 period and stagnated from 2010 on-
wards. Compared to their public sector peers, the private banks also had lower
levels of NPAs (Sengupta and Vardhan (2017)). Hence they would have incurred
lower credit loss provisions and the issue of productivity gains going to defaulting
borrowers is less relevant for them.
The patterns of productivity gain, PPOP, and NPAs appear to be different for
the foreign banks. They did not gain much in terms of productivity in the mid
1990s-2008 period, compared to the PSBs and private banks. They saw substantial
productivity gains in the subsequent years during which their PPOP also went up.
They also had much smaller NPA levels.

4 Conclusion

Our descriptive analysis from 1992 to 2019 reveals that banking sector in India ex-
perienced substantial productivity gains till about 2010 after which the gains seem
to have plateaued and even declined marginally in recent years. Among the bank
groups, private sector banks which primarily began operating from 1994-95 on-
wards in the post-liberalisation era, witnessed sharp reductions in their operating
costs even as their balance sheets expanded rapidly, indicating steady productiv-
ity improvement. These banks started on superior technology platforms which
may have contributed to the steep efficiency gains. However the benefits ran their
course by about mid 2000s. On the other hand, public sector banks witnessed

17

Electronic copy available at: https://ssrn.com/abstract=3857373

significant productivity gains from mid to late 1990s onwards till about 2010, and
their gains even surpassed that of the private banks.
When we analyse the different stakeholders of the banking system to under-
stand who benefitted from the productivity growth, we do not find any appreciable
amount of the gains accruing to the depositors, borrowers or employees. Instead
it appears that much of the efficiency gains enjoyed by the banking system went
to the shareholders which in the case of public sector banks would imply the gov-
ernment. It is also possible that the gains that accrued to the shareholders were
used to deal with the losses incurred by the public sector banks on their loan books
particularly during the period from 1997 to 2003 when these banks recorded high
levels of non performing assets.
It might be a matter of concern for policymakers that the other stakeholders
of the banking system have not been benefitting from the efficiency gains. The
uneven distribution of the gains might be due to the lack of competition in the
Indian banking sector. In a truly competitive banking system, all four stakeholders
would have nearly equal bargaining power and this would have resulted in a more
equitable distribution of the productivity gains.
The gains accruing to the borrowers would have lowered the cost of debt cap-
ital, for depositors it would have meant a better return on their savings and em-
ployees would have benefitted from higher wages. One may argue that the multi-
plier effect of the gains going to the (good) borrowers, depositors, and employees
wouldbe much greater. This does not seem to have happened in India.
There are close to 100 scheduled commercial banks in India by now but 70%

18

Electronic copy available at: https://ssrn.com/abstract=3857373

of the sector is owned by the government. In the 1990s, public sector banks ac-
counted for 90% of the total size of the banking sector. These banks look nearly
identical to each other in all aspects of banking business. Thus, despite there
beinga large number of banks in India there is hardly any element of competi-
tion among them. Government ownership has resulted in a near monopolistic,
non-competitive banking sector where any gains from efficiency improvement in-
stead of getting passed on to the broader economy, accrue to the owners i.e. the
government and are in turn used to effectively bail-out the defaulting borrowers.
To ensure a more equal and broad based distribution of the productivity gains
of the banking sector among all the relevant stakeholders, perhaps what is needed
is a gradual reduction in the market share of the government owned banks so that
there is greater heterogeneity among the banks and hence greater competition in
the banking sector.

19

Electronic copy available at: https://ssrn.com/abstract=3857373

References

Berger, Allen N. and David B. Humphrey, “Efficiency of financial institutions:

international survey and directions for future research,” European Journal of
Operational Research, April 1997, 98 (2), 175–212.

Bhattacharyya, Aditi and Sudeshna Pal, “Financial Reforms and Technical Ef-
ficiency in Indian Commercial Banking: A Generalized Stochastic Frontier
Analysis,” Review of Financial Economics, November 2013, 22 (3), 109–117.

Bhattacharyya, Arunava, C. Lovell, and Pankaj Sahay, “The impact of liber-

alization on the productive efficiency of Indian commercial banks,” European
Journal of Operational Research, 1997, 98 (2), 332–345.

Casu, Barbara, Alessandra Ferrari, and Tianshu Zhao, “The impact of regu-
latory reforms on cost structure, ownership and competition in Indian banking,”
Journal of Banking & Finance, January 2010, 34 (1), 246–254.

Das, Abhiman and K. R. Shanmugam, “Efficiency of Indian commercial banks

during the reform period,” Applied Financial Economics, 2004, 14 (9), 681–
686.

and Saibal Ghosh, “Financial deregulation and efficiency: An empirical anal-

ysis of Indian banks during the post reform period,” Review of Financial Eco-
nomics, 2006, 15 (3), 193–221.

20

Electronic copy available at: https://ssrn.com/abstract=3857373

 , Ashok Nag, and Subhash C. Ray, “Liberalisation, Ownership and Effi-
ciency in Indian Banking: A Nonparametric Analysis,” Economic and Political
Weekly, 2005, 40 (12), 1190–1197.

Kumar, Sunil and Rachita Gulati, “Did efficiency of Indian public sector banks
converge with banking reforms?,” International Review of Economics, March
2009, 56 (1), 47–84.

Kumbhakar, Subal C and Subrata Sarkar, “Deregulation, Ownership, and Pro-

ductivity Growth in the Banking Industry: Evidence from India,” Journal of
Money, Credit and Banking, June 2003, 35 (3), 403–424.

Sanyal, Paroma and Rashmi Shankar, “Ownership, Competition and Bank Pro-
ductivity: An Analysis of Indian Banking in the Post-Reform Period,” Interna-
tional Review of Economics and Finance, 2011, 20, 225–247.

Sathye, Milind, “Efficiency of banks in a developing economy: The case of In-

dia,” European Journal of Operational Research, 2003, 148, 662–671.

Sengupta, Rajeswari and Harsh Vardhan, “Non-performing Assets in Indian

Banks : This Time It Is Different,” Economic and Political Weekly, 2017, 52
(12).

Sensarma, Rudra, “Are foreign banks always the best? Comparison of state-
owned, private and foreign banks in India,” Economic Modelling, July 2006, 23
(4), 717–735.

21

Electronic copy available at: https://ssrn.com/abstract=3857373

5 Tables and Figures

Table 1 Evolution of productivity growth in Indian banking

Operating Costs/Average Total Assets (%)
Periods All banks Non-SBI PSU SBI group Private Foreign
Banks Banks Banks
1992-1999 2.90 2.90 3.0 2.70 3.10
2000-2009 2.30 2.20 2.30 2.20 3.20
2010-2019 1.80 1.50 1.90 2.20 2.30
Operating Costs/Total Income (%)
Periods All banks Non-SBI PSU SBI group Private Foreign
Banks Banks Banks
1992-1999 63.5 72.9 58.9 55.7 45.6
2000-2009 51.4 53.0 51.6 49.8 45.8
2010-2019 47.2 47.7 55.1 44.5 40.9
Note: This table reports the average values of efficiency gains as measured by the ratio of the operating cost to total assets
and operating cost to total income, of different groups of banks as well as of the entire banking system, for the last three
decades.

Figure 1 Total number of employees in the Indian banking system, 1992-2018

This figure shows the total number of employees in private sector banks, public sector banks and
all other commercial banks in India.

22

Electronic copy available at: https://ssrn.com/abstract=3857373

Figure 2 Total balance sheet of the Indian banking system, 1992-2018
This figure shows the total balance sheet size of the private sector banks, public sector banks and
all other commercial banks in India.

Figure 3 Cost of intermediation of the Indian banking system, 1992-2019

This figure shows the evolution over time of two measures of productivity in Indian banking:
ratio of operating costs to total income and ratio of operating costs to average total assets of all
commercial banks.

23

Electronic copy available at: https://ssrn.com/abstract=3857373

Figure 4 Cost of intermediation for various groups of banks, 1992-2019
This figure shows the evolution over time and across groups of banks of the ratio of operating
costs to average total assets. The four bank groups are private sector banks which include the old
and the new private banks, foreign banks, State Bank of India and its associates and the non-SBI
public sector banks.
4.0
3.5
In percentage terms

3.0
2.5
2.0
1.5
1.0

1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

Non−SBI PSBs Foreign

SBI and associates Private

Figure 5 Term Deposit Pricing Relative to Risk Free Government Security Yield
This figure shows the annual average interest rate on term deposits of all banks and the yield of
5-year G-Secs for the period 1997 to 2019. Interest on term deposits is calculated as the difference
between total interest on deposits and interest on savings account (calculated at about 3.8% on
average deposits). Interest rate of term deposits is obtained as the ratio of interest on term deposits
and average term deposits between two years.
14
12
In percentage terms

10
8
6
4

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019

Average Interest rate on term deposits Yield on 5 year GOI Securities

24

Electronic copy available at: https://ssrn.com/abstract=3857373

Figure 6 Term Deposit Pricing Relative to Risk Free Government Security Yield
across Bank Groups
This figure shows the annual average interest rate on term deposits of the four groups of banks and
the yield of 5-year G-Secs for the period 1997 to 2019. Interest rate of term deposits is obtained
in the same way as in figure 5 and the four groups of banksare the same as in figure 4.
14
12
In percentage terms

10
8
6
4

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019

Yield on 5 year GOI Securities SBI and associates Foreign

Non−SBI PSBs Private

Figure 7 Average Loan Pricing relative to Risk Free Government Security Yield
This figure shows the annual average yield on loans for all commercial banks and the 10 year
government security yield. Average yield on loans is calculated as the ratio between interest
earned on loans and average total loans and advances across two consecutive years.
14
12
In percentage terms

10
8
6
4

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019

Loan yield all banks average Yield on 10 year GOI Securities

25

Electronic copy available at: https://ssrn.com/abstract=3857373

Figure 8 Average Loan Pricing relative to Risk Free Government Security Yield
across bank groups
This figure shows the annual average yield on loans for the four groups of banks and the 10 year
government security yield. Average yield on loans is calculated as the ratio between interest earned
on loans and average total loans and advances across two consecutive years. The four bank groups
are the same as in figure 4.

26

Electronic copy available at: https://ssrn.com/abstract=3857373

Figure 9 Employee Costs as a Percentage of Total Assets and Income
This figure shows the evolution over time of the share of employee wages and salaries in total
income as well as in average total assets of all commercial banks.

Figure 10 Real Per Employee Compensation and Total Balance Sheet of the
Banking System
This figure shows the evolution over time of the total wages and salaries of employees of all com-
mercial banks and the total annual assets of the banks, both parameters deflated by the consumer
price index of the respective years.

27

Electronic copy available at: https://ssrn.com/abstract=3857373

Figure 11 Real Per Employee Compensation and Total Balance Sheet of the pri-
vate sector banks
This figure shows the evolution over time of the total wages and salaries of employees of the private
sector banks and the total annual assets of the banks, both parameters deflated by the consumer
price index of the respective years.

28

Electronic copy available at: https://ssrn.com/abstract=3857373

Figure 12 Real Per Employee Compensation and Total Balance Sheet of the pub-
lic sector banks
This figure shows the evolution over time of the total wages and salaries of employees of the public
sector banks and the total annual assets of the banks, both parameters deflated by the consumer
price index of the respective years.

Figure 13 Operating cost and Operating Profits of Indian Banking, 1992-2019

This figure shows the ratio of operating costs to average total assets and the ratio of pre provision-
ing operating profits to average total assets of all commercial banks.
4.0
3.5
In percentage terms

3.0
2.5
2.0
1.5
1.0

1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017 2019

Ratio of operating cost to average total assets Ratio of PPOP to average total assets

29

Electronic copy available at: https://ssrn.com/abstract=3857373

Figure 14 Cost of Intermediation, Pre-provisioning Operating Profits and Gross
NPAs of PSBs, 1997 - 2018
This figure shows the evolution over time of the total wages and salaries of employees of the private
sector banks and the total annual assets of the banks, both parameters deflated by the consumer
price index of the respective years.

Figure 15 Cost of Intermediation, Pre-provisioning Operating Profits and Gross

NPAs of Private banks, 1997 - 2018

30

Electronic copy available at: https://ssrn.com/abstract=3857373

Figure 16 Cost of Intermediation, Pre-provisioning Operating Profits and Gross
NPAs of Foreign banks, 1997 - 2018

31

Electronic copy available at: https://ssrn.com/abstract=3857373