An Analysis of the Education Loan Portfolio of

Vijaya Bank

A Dissertation Submitted in Partial Fulfillment of the

Requirements for the Award of the Degree of

Master of Philosophy
in
Commerce

by
Sanjay Panikar
(Reg. No. 1431602)

Under the Guidance of

Theresa Nithila Vincent
Associate Professor

Department of Commerce

CHRIST UNIVERSITY
BENGALURU, INDIA
April 2016
 APPROVAL OF DISSERTATION

Dissertation entitled An Analysis of the Education Loan Portfolio of Vijaya Bank by

Sanjay Panikar, Reg. No. 1431602, is approved for the award of the degree of Master of
Philosophy in Commerce.

Examiners:

1. ______________ ____________

2. ______________ ____________

3. ______________ ____________

Supervisor:

______________________ ______________________

Chairman:

______________________ ______________________

Date:
Place: Bengaluru

ii
 DECLARATION

I, Sanjay Panikar, hereby declare that the dissertation, titled An Analysis of the Education
Loan Portfolio of Vijaya Bank is a record of original research work undertaken by me for the
award of the degree of Master of Philosophy in Commerce. I have completed this study under
the supervision of Dr Theresa Nithila Vincent, Associate Professor - Department of
Commerce, Christ University.

I also declare that this dissertation has not been submitted for the award of any degree,
diploma, associateship, fellowship or other title. It has not been sent for any publication or
presentation purpose. I hereby confirm the originality of the work and that there is no
plagiarism in any part of the dissertation.

Place: Bengaluru
Date:

Sanjay Panikar
Reg. No.: 1431602
Department of Commerce
Christ University, Bengaluru

iii
 CERTIFICATE

This is to certify that the dissertation submitted by Sanjay Panikar (Reg. No 1431602) titled
‘An Analysis of the Education Loan Portfolio of Vijaya Bank’ is a record of research work
done by him during the academic year 2014-2015 under my supervision in partial fulfilment
for the award of Master of Philosophy in Commerce.

This dissertation has not been submitted for the award of any degree, diploma, associateship,
fellowship or other title. It has not been sent for any publication or presentation purpose. I
hereby confirm the originality of the work and that there is no plagiarism in any part of the
dissertation.

Place: Bengaluru
Date:
Dr. Theresa Nithila Vincent
Associate Professor
Department of Commerce
Christ University, Bengaluru

iv
 Acknowledgement

I wish to express my sincere thanks to Rev. Dr (Fr) Thomas Aykara, Chancellor,

Dr (Fr) Thomas C Mathew, Vice-Chancellor, Dr (Fr) Abraham V M, Pro-Vice
Chancellor and other esteemed authorities of Christ University for facilitating me to
successfully complete this work.
I am extremely grateful to Dr Theresa Nithila Vincent, my supervisor for her
invaluable guidance. Her encouragement and advice inspired and motivated me immensely.
I am also grateful to all the faculty members of Commerce Department, Christ
University for their suggestions, encouragement and their valuable feedback. They helped me
to keep my focus.
My heartfelt thanks to various officials in Vijaya Bank who have provided me with
the guidance and support in undertaking this research. Their suggestions and guidance was
invaluable in completion of this research. I am extremely grateful to Mr Sathish Ballal C,
General Manager (Credit Department, Priority & FI), Mr Ramakrishnan K G, Chief
Manager, Mr G K Murthy, Assistant General Manager (RACPC), and Mr R
Narayanaswamy, Chief Manager. I wish to express my sincere thanks Mr Jaikishan M,
Chief Manager (Personnel) and Mr Anand R P, Senior Manager (Personnel) for
facilitating this research.
I am grateful to Dr N Elangovan, Department of Management Studies for his
suggestions and guidance during the course of the study. I am indebted to Dr V R Uma, Co-
ordinator MPhil Program and Prof. Chandrasekharan K A, General Research Coordinator,
M. Phil for their valuable guidance and assistance.
I am also grateful for the love and patience shown by my family who supported me in
innumerable ways.

Sanjay Panikar

v
 ABSTRACT

Traditionally education is considered to be an important factor in improving an

individual’s quality of life. In India, Higher education has remained largely a state funded
initiative. While basic education is the primary goal, it is increasingly felt that higher
education also has not really succeeded in reaching out to the majority of the population. In
the coming decades, with increased globalization, India can reap the benefits of a highly
educated workforce. However, the Indian government despite its best intentions and planning
has not made any concerted efforts for improvement in access or quality of higher education.

While the policy makers are in consensus about the impact of highly educated, young
workforce on the overall development of the country, there is considerable difference of
opinion about who should bear the primary responsibility. The Planning commission, in a
paradigm shift proposed changing the mode of funding of the higher education system.
Strategic funding of this sector has been strongly proposed in order to make a marked
difference in the overall resource endowment for the higher education sector (Ministry of
Human Resource Development, Government of India, 2013).

To ensure funding is available on the demand side by the students, a model education
scheme was designed by IBA (Indian Banks Association) in the year 2000. The scheme has
been running for the last 15 years. This study attempts to analyse the education loan portfolio
of Vijaya Bank, based in Bengaluru and study the borrowing patterns of a sample of
borrowers in Bengaluru region.

The study has analyzed the educational loan portfolio of Vijaya Bank from 2002 to
2012 and developed a demographic profile of the borrowers. The study has analyzed the NPA
rates of the branches and has attempted to identify correlations and relationships between the
students, parental, course of the study and the loan characteristics and their ability to predict
defaulting behaviour among the borrowers.

vi
 TABLE OF CONTENTS
Approval of dissertation……………………………………………………........ii
DECLARATION.…………………………………………………….................iii
CERTIFICATE……………………………...……………………......................iv
Acknowledgement……………………………………………….........................v
Abstract……………………………………………...……………......................vi
Chapter 1: Introduction……………….............……………..….......................1
1.1 Higher Education……………………………………….....................1
1.2 History of Education in Modern India………………........................2
1.3 Current Status of Higher Education………………............................3
1.4 Financing Higher Education.…………………......….............................8
1.5 Indian Banks Education Loan Scheme……………..........................12
1.6 Vijaya Bank …………………………………………......................17
Chapter 2: Review of Literature …………………….....................................25
2.1 Introduction……………….......................…………........................25
2.2 Need for Review of Literature.......………………….......................26
2.3 Human Capital………………………………………………….......27
2.4 Equity in Higher Education…………………...................................32
2.5 Loan Defaults…………………………………................................35
Chapter 3: Report on the Present Investigation………….............................49
3.1 Introduction………………………………………………………..49
3.2 Statement of Problem………….............…………..........................49
3.3 Operational Definitions ………………………...............................50
3.4 Variables of the Study….................................…….........................51
3.5 Objectives….....................................................................................53
3.6 Hypotheses Framed & Tested…………………..............................54
3.7 Sampling procedure……………………………………..................54
3.8 Reliability………………………….....…………............................55
3.9 Validity………………………….............………...........................56
3.10 Conclusion…………………………….………............................56
Chapter 4: Results and Discussions …………………....……......................58
4.1 Introduction……………………………...…….............................58

vii
 4.2 Descriptive Analysis………………………………………………....58
4.2.1 Vijaya Banks Education Lon Advances………………………...58
4.2.2 Applicant Characteristics……………………………………......65
4.2.3 Family and Parent Characteristics………………….…………...67
4.2.4 Characteristics of the Education Program………………….....68
4.2.5 Loan Characteristics………………………………………..……....71
4.3 Hypothesis Testing and Interpretation…………………....73
4.3.1 NPA Amount and Total Cost of Education………………….........73
4.3.2 Parent’s Employment Type and Payment Behaviour…………..74
4.3.3 Gender and Payment Behaviour…………………………….….76
4.3.4 Applicants Marks and Choice of Study………………….……..77
4.3.5 Regression……………………………………………………....80
Chapter 5: Summary and Conclusion……………………….……………..92
5.1 Introduction…………………………………………………….....92
5.2 Findings …………………………………………………………..92
5.3 Recommendations………………………………………………...94
5.4 Limitations of the Study ……………………………………….....96
5.5 Scope for Further Research ………………………………………96
5.6 Conclusion………………………………………………………...98
Appendix I – Model Education Loans Scheme……………………………...100
Appendix II – Loan Application Form ………………………………...…....108
Literature Cited.………………………………………………………….……....117

viii
 LIST OF FIGURES

1.1 Gross Enrolment in Millions .................................................................................................................. 4

1.2 Growth in Number of Universities ....................................................................................................... 4
1.3 Growth in Number of Colleges ............................................................................................................. 5
1.4 Growth of Educational Loans (All Banks) ....................................................................................... 14
1.5 Profitability Indicators (CoD & YoA) ............................................................................................... 24
4.1 Growth of Education Loans (Accounts) ........................................................................................... 60
4.2 Growth of Education Loans & Gross Advances ............................................................................. 62
4.3 Trends in Gross & Net NPAs (All Advances) ................................................................................. 62

ix
 LIST OF TABLES
1.1 GER of Selected Countries ........................................................................................... 6
1.2 GER by Preference of Stream........................................................................................ 7
1.3 Expenditure on Education ............................................................................................. 8
1.4 Sample Indicative List of Interest Rates ...................................................................... 15
1.5 Bank-Wise Education Loans (31.03.2013)................................................................... 16
1.6 Region & Branch Structure .......................................................................................... 18
1.7 Banking Products & Services Offered ......................................................................... 21
1.8 Credit Advances Portfolio ........................................................................................... 22
1.9 Asset Quality (31st March 2014).................................................................................. 23
3.1 Sample Distribution ..................................................................................................... 52
4.1 Growth of Education Loan Portfolio ........................................................................... 59
4.2 Education Loans (Bengaluru) ...................................................................................... 61
4.3 Gross Non-Performing Assets (Education Loans) ...................................................... 63
4.4 Sample Branches (Education Loans & NPA Rate) ..................................................... 64
4.5 Frequency Distribution (Applicant Characteristics) .................................................... 66
4.6 Frequency Distribution (Family & Parental Characteristics) ...................................... 68
4.7 Frequency Distribution (Education Program Characteristics)...................................... 69
4.8 Frequency Distribution (Loan Characteristics) ........................................................... 71
4.9 Range, Mean & Standard Deviation ............................................................................ 72
4.10 Cost of Education……………………………………………………………………. 73
4.11 ANOVA – NPA Amount & Total Cost of Education………..…………………….... 74
4.12 Employment Type…………………………………………………………………… 75
4.13 ANOVA - Parents Employment Type & Payment Behaviour..................................... 76
4.14 Chi-Square Test – Gender & Payment Behaviour…………………………………... 77
4.15 Chi-Square Test - Applicants SSLC Marks & Choice of Study................................... 79
4.16 Model Summary - Applicant Characteristics & NPA Status ...................................... 81
4.17 Model Summary - Parent Characteristics & NPA Status ............................................ 83
4.18 Model Summary - Education Characteristics & NPA Status ...................................... 85
4.19 Model Summary - Loan Characteristics & NPA Status .............................................. 87
4.20 Model Summary - All Characteristics & NPA Status ................................................. 89

x
 INTRODUCTION

1.1 HIGHER EDUCATION

Various scholars have studied the importance of higher education and its impact on human
societies. The first major theory, on the impact of higher education was advanced by Becker
in 1962. It postulated that education provides people with higher abilities to manage the
economy and related services. Therefore from a student's perspective, investment in higher
education leads to higher economic rewards (Becker, Investment in Human Capital: A
Theoretical Analysis, 1962). In 1973, Arrow put forward an alternative theory, where higher
education is taken as a method of improving the skills for performing different vocational
tasks, as well as developing social skills for upward mobility (Arrow, 1973). A third view is
that higher education does not affect productivity, as it is a function of the job and not of the
person. People select jobs because of other criteria and not primarily because of their
education (Carnoy & Lobo, 1979). While the theoretical implications and the practical impact
on human societies due to higher education are still not clear, traditionally Indian societies
have looked at education as key differentiator in status and wealth. Various studies have
shown that an educated workforce leads to an increase in productivity, higher competitive
levels and innovation, higher earnings for the individual leading to a higher national economic
growth.

While there might be other factors at play, social and psychological desires have also
contributed to a stronger demand for higher education in India. The overall wealth and income
levels were holding back the development of higher education in India. However, with the
relaxation of economic policies combined with the faster integration with world economies,
the demand for Higher education in India has increased exponentially over the last few years.
The advent of the Internet and mobile services has increased the access to global views among
the youth in India. The number of students enrolling for graduation and higher levels has seen
a marked increase. The effects of the globalised workplace and its benefits in terms of social
mobility, higher incomes and access to health has led to an unprecedented increase in students
opting to pursue their education beyond basic primary education. This increase in enrolment is
not commensurate with funding the education sector has been getting in the last few decades.
The result has been deterioration in the overall access and quality of the education sector.
While India produces one of the largest numbers of graduates in the world, by some industry
estimates, almost 45% of them lack basic employment skills.
 The education system as a whole faces many challenges. The higher education system
is particularly faced with supply side challenges, like financing of the colleges and
institutions, qualified managers to run these institutions, qualified teachers, providing
equitable access to all deserving students, course’s relevance in the modern world. These
institutions are of critical importance as the quality of the education decides the future of the
country in terms of progress and development. The quality of education also affects the
country's integration into the globalised economy.

Successive governments since the time of independence have recognized the

importance of developing the Indian workforce through imparting modern education. They
have recognized that colleges and universities perform varied roles like creating new
knowledge through research and development; train the students in basic social and
interpersonal skills to work in a common environment and generally producing an intelligent
and dynamic human resource pool. Policy makers have realized that higher education should
be invested optimally, with a long-term perspective for the development of culture, social
cohesion, equity and justice among the country's citizens. With the advent of globalization
with the easing of boundaries, and flow of ideas and goods, has essentially necessitated the
development of a higher quality of education so that the Indian students can compete globally.

1.2 HISTORY OF EDUCATION IN MODERN INDIA

Post-independence, the government of India constituted a committee to study the

status and to recommend the future course of action for the development of higher education
in India. The Radhakrishnan Commission on University Education (1948-49) listed down
several goals for the development of higher education in India. The commission stated the
reforms needed in the education sphere in the following words: “The most important and
urgent reform needed in education is to transform it, to endeavour to relate it to the life, needs
and aspirations of the people and thereby make it the powerful instrument of social, economic
and cultural transformation necessary for the realization of the national goals. For this
purpose, education should be developed so as to increase productivity, achieve social and
national integration, accelerate the process of modernization and cultivate social, moral and
spiritual values.”

Subsequently in 1964-66, the 'University Education Commission Report' was prepared

and introduced by Kothari. These two reports laid down the basic framework for the National
Policy of 1986 for higher education in the country. The policy for the development of higher

2
education has primarily been based on the ‘National Policy on Education’ of 1986 (and
modified in 1992) and its Programme of Action adopted in 1992. The introduction of the
National Policy on Higher Education in 1986 brought together the objectives as listed down
by the Radhakrishnan and the Kothari commission and coalesced into a concrete policy by
setting five main goals for higher education, as enumerated below:

 Access: Greater access requires an enhancement of the education institutional

capacity of the higher education sector to provide opportunities to all those who
deserve and desire higher education.

 Equity: Equity involves fair access to the poor and the socially disadvantaged
groups to higher education.

 Quality and Excellence: involves provision of education in accordance with
accepted standards so that students receive available knowledge of the highest
standard that helps them to enhance their human resource capabilities.

 Relevance: involves promotion of education so as to develop human resources
keeping pace with the changing economic, social and cultural development of the
country

 Value Based Education: involves inculcating basic moral values among the youth

1.3 CURRENT STATUS OF HIGHER EDUCATION

To meet the objectives as set out by the Radhakrishnan Commission and Kothari
Commission and finally by the National Policy of 1986 for higher education in the country,
the government has been actively encouraging opening of new colleges and universities. India
has witnessed significant progress in the establishment of colleges and institutions for higher
learning in the last few decades. Figure 1.1 shows the increase of the GER (Gross Enrolment
Ratio) over the years from the time of independence to present.

3
 25

20

15
in Millions

10

Source : UGC Annual Report

Figure 1.1 Gross Enrolments in Millions

As can be seen from the figure 1.1, the country has witnessed an exponential growth in
students pursuing higher education. As per an observation from the ministry of human
resources, India has moved from an 'elite' system of higher education to a 'mass' system of
higher education as of the end of the year 2009 when we crossed the 15% gross enrolment
mark. This exponential growth has resulted in the number of universities increasing from 20
universities at the time of independence to 624 as of 2013 (Figure 1.2).

700

600

500

400

300

200

100

0
1947 1950 1960 1970 1980 1990 2001 2002 2006 2008 2009 2010 2011 2013
Source: UGC Annual Reports

Figure 1.2 Growths in Number of Universities

4
 The increase in universities has also been commensurate with the growth of the
number of colleges over the last six decades. At the time of Independence of India, there were
only 20 Universities and 500 Colleges in the country with 2.1 lakhs students in higher
education. The numbers now have increased to 30 times in the case of the Universities, 74
times in the case of Colleges and the students enrolment has gone up to over 100 times in the
formal system of higher education in comparison to the figures at the time of
independence(University Grants Commission, Government of India, 2014). As of end of
March 2013, there were 44 Central universities, 286 State, 151 State Private, 129 deemed to
be Universities and four Institutions established under State Legislation.

37,204
40,000
33,023
35,000
30,000
25,951
25,000

20,000 18,064

15,000
11146
10,000 7346
4577
5,000
496 1819
0
1947 1950 1960 1970 1980 1990 2001 2002 2006 2008 2009 2010 2011 2013

Source: UGC Annual Reports

Figure 1.3 Growths in Number of Colleges

Higher education is gaining additional significance in comparison to the aging

population in many of the developed countries. As per the estimates brought out by the
International Labour Organisation, there will be around 116 million workers in the age group
of 20 to 24 in 2020. In comparison China is expected to have only 94 million, giving us a
significant advantage in the global labour markets. The average age of the Indian population
in 2020 is expected to be 29 while many other countries are grappling with a rapidly aging
population. However, while India has made significant progress in gross enrolment in higher
education, we still lag behind in comparison with some of the developed countries.

5
 Table 1.1 GER of Selected Countries

Country GER in %
India 22%
South Africa 15%
Brazil 26%
Russia 76%
China 25%
USA 95%
UK 59%
Sweden 74%
Canada 60%
Argentina 71%

Source: The Global Competitiveness Report 2012-2013

India's GER at 22% is significantly lower than other BRICs countries (shaded) with
the exception of South Africa. If we expand the list further to include other emerging
countries like Argentina, we are among the lowest. The report presented by Goldman Sachs
indicates that the actual amount of money spent on higher education in India is among the
lowest among other BRICs countries.

The enrolment for higher education by the level of the education in the year 2013
indicates that 85.90% of the students are at Graduate level, 12.15% are at Post Graduate level,
0.84% are pursuing research levels (includes M.Phil. and PhD) and 1.11 per cent have
enrolled for diploma level courses.

Another negative development in the growth of the enrolment in higher education is

the wide variation of enrolment in different states. It varies from 41.4% in Chandigarh to as
low as 8.1% in Jharkhand. These create regional imbalances to the development of the human
resources and will increasingly lead to social unrest and large-scale migrations to areas with
more employment opportunities.

The maximum preference in the choice of streams in Indian higher education is clearly
in Arts and related subjects with 37.94% of the students opting for the same. The next highest
is in Science with 18.56% of the students selecting a science related field of study. The third
preference of the students in choosing their line of study in higher education is in Commerce
or Management. However, in the study of the educational loans undertaken the bigger share
was by students in the engineering field. This might be, because of the cost of the pursuing
higher education in engineering fields is considerably higher in comparison to other streams
6
of higher education. On average by some estimates the cost of a professional degree is almost
double the cost of a non-professional degree.

Table 1.2 GER by Preference of Stream

Stream Total Enrolment % to Total

Arts 8,157,308 37.9%

Science 3,992,007 18.6%
Commerce/Management 3,762,003 17.5%
Education 741,905 3.5%
Engineering/Technology 3,333,165 15.5%
Medicine 752,277 3.5%
Agriculture 103,013 0.5%
Veterinary Science 29,325 0.1%
Law 400,815 1.9%
Others 229,336 1.1%

Total 21,501,154 100.0%

Source: UGC Annual Report

While the number of students pursuing engineering related higher education is only
15.55% of the total, in number terms it comprises of 3.33 million students. In addition to
disparities in GER among states the access to institutions of higher education is also very
disparate. States like Haryana, Tamil Nadu and Kerala etc. have a higher number of colleges
and thus provide more access to higher education for the population of that geographical area.

The amount of money allocated to the education sector by the government is

significantly lesser than other comparable countries. India lags behind most of the developed
countries. In case of South Asian countries, while India spends only 4.1% of its Gross
Domestic Product (GDP) on education, countries like Thailand and Malaysia spend
considerably more. The GDP of Thailand and Malaysia are considerably less than India,
thereby, in real terms providing us with a bigger pool of money to invest.

7
 Table 1.3 Expenditure on Education

Spending on Spending on
Country Education as a % Country Education as a %
of GDP of GDP

Switzerland 5.8% South Africa 5.3%

U.S. 5.7% Thailand 5.2%
France 5.6% Chile 4.2%
U.K. 5.3% Brazil 4.2%
Malaysia 8.1% India 4.1%
Mexico 5.3% Russia 3.8%

Source: United Nations Human Development Program

As seen in the table 1.3, the spending on higher education as a percentage of its Gross
Domestic Product is very low in India. There remain numerous challenges for India in
reaching a higher gross enrolment ration in terms of financing these institutions. The main
challenges can be broken up into three areas of improvement, namely, financial access to
higher education, equity in higher education in terms of ownership of these institutions and
excellence of the education. Improvement of access to education is complex with many social,
religious, and economic factors influencing it. Excellence in education is similarly complex
due to varied standards and difficulty in measurement.

This study therefore focuses on the financial accessibility for borrowers in India for
pursuing higher education.

1.4 FINANCING HIGHER EDUCATION

The improvement of equity, in higher education has to be targeted separately from

both the supply and the demand side. On the supply side, the government has launched many
initiatives to improve the area of higher education. The government has allowed private
investments in education, has encouraged the opening of private colleges and universities and
have launched open courses. To increase access to students in remote places, the government
in addition has launched distance education modelled on the famous Open University
education in Britain.

8
 While all these have considerably helped access to higher education, on the demand
side one of the biggest challenges has been of providing finance to the poor but meritorious
students. This challenge is not restricted only to India, but affects every developed economy.
Governments have tried various models of providing finance to students but have always had
a problem in reconciling the costs of the finance. Student loans have traditionally been used as
the vehicle to bridge the gap.

Various countries have tried different models of providing access to higher education,
from subsidising the education by administering the subsidy through the institutions to
directly providing a loan to the students with a deferred payment mechanism. Governments
have always grappled with the problem of providing the highest quality of education while
containing the fiscal expenditures incurred for providing the same. Some studies have
indicated, that passing the burden of funding, to the students, partly increases the efficiency of
the whole objective of increasing quality of education by increasing the students equity
(Psacharopoulus, Tan, & Jimenez, 1985). Recovering the cost of higher education from the
students while leading to higher efficiency in managing public funds is fraught with political
considerations. Other main drawback in this approach is the inability to provide educational
opportunities to the meritorious but poor students. In socio-economic terms, this approach can
lead to inequity in access to higher education.

To overcome the problem of balancing the fiscal expenditure while providing good
quality tertiary education to the deserving students, many countries have adopted financing
the higher student education through a loan based instrument. Such schemes defer the
payment from the students till they graduate and find economic employment. Many studies
have been conducted on the efficacy of the deferred loan programs in developing countries
(Woodhall, 1983).

Countries have implemented student loan financing programmes in various forms with
varying features. Some programs have focussed on the repayment structure while some have
explored the delivery of efficient credit at cheap rates. Still other countries have explored the
mechanism of delivery with some countries resorting to state run institutions while others
subsidising commercial banks to deliver the financing. In terms of student financing, India has
been lagging in implementation of financing schemes. Many developed countries like USA
and various European countries had implemented student financing programs starting from
the 1950s. Some Latin American countries started such programs post the decade after the
world war, for e.g. Colombia started in 1953.
9
 Most of student financing programs have been fixed instalment repayments, with a
moratorium period, provided to the student during the initial years. The period of moratorium
varies between different countries but usually is given up to a year after graduation.
Repayment of the loan is made for a fixed period or tenor. The amount of the instalment
depends on the period of the repayment period and the interest rate charged under the scheme.
Some countries have implemented loan repayment schemes where the instalment amount is
fixed at the time of granting of the loan and gradually increases over a period of time. This
helps the borrower to manage his repayments so that his initial payments will be
commensurate with his earning.

Some programs implemented in countries like Sweden etc. are dependent on the
income and are known as income-contingent loan. These are generally more efficient from a
social point as it does not burden the student excessively. The repayments are fixed, as a part
of the income he earns after graduation. However, the implementation of such income-
contingent programs are complex, as in addition to the delivery of credit, the administrators
have to develop a mechanism, to track the student’s income after completion of the course.
However income-contingent loans are more beneficial to the poorer students than fixed
instalment loans. The future earnings from completion of an educational course are
unpredictable, and therefore the risk of not earning as per the students initial forecast is very
high. In addition, generally, poorer students are more risk averse and will not prefer any
repayment program, which commits them to a fixed repayment amount. Income contingent
loans are more efficient in providing a systematic benefit to the student borrowers. Besides,
the monitoring issues as mentioned earlier, income-contingent loans might discourage
earnings.

There are variations between the different programs on what is covered under the
terms of student financing. Some countries provide for tuitions and living expenses. Other
countries only finance tuitions, leaving the burden of financing the living expenses to the
student’s own resources.

The administration of financing program also varies between countries. Some

countries have enabled, usually through legislation, separate autonomous lending institutions.
They are usually capitalized through public funds, which are expected to reduce as once the
students start repaying their loans. The borrower’s repayments with the higher interest rate to
the cost of the capital (spread) are expected to continue the funding of these institutions. In
India, initially the government tried financing higher education through this method. The
10
National Scholarship scheme started in 1961-62 was publicly funded with a ceiling of Rupees
25,000 for meritorious students. The scheme was discontinued in the 1990s after having
successfully financed more than 475,000 students. Low repayment rates combined with
cumbersome bureaucratic procedures was the main reason for the failure of the scheme. A
similar program for higher education started in 1963 was the National Loan Scholarship
scheme. It was a fixed instalment repayment scheme administered through the state
governments with a target of 20,000 loans per annum. However this scheme was discontinued
in 1991, due to non-viability of the scheme due to minimal returns, and lack of resources for
administering the loans. This program can be called the pre-cursor to the educational loan
scheme being provided by commercial banks. Lack of government and public funding with
poor rate of returns made this government scheme unprofitable and unsustainable. In India,
instead of separate institutions for managing these loans, the government provided these
through various government departments.

Another option adopted by many countries is to administer these financing through

commercial or specialised banks. The banks could be government, public or privately owned.
However the enthusiasm for providing such financing is lower in private and commercial
banks due to lower rate of returns such programs provide. Due to banks being hesitant to
manage these programs, due to better opportunities to invest their funds, governments in some
countries guarantee these advances. In the United States, the banks manage funding with the
government subsidising the costs. The banks act as the grantors and collections agents for
these loans. The drawback of the program in United States is the insistence of banks to lend
only to credit worthy borrowers instead of providing financing to the needy. While credit
worthiness and the needy, do not necessarily exclude each other, several studies have shown
that the students who most need the loans are the poor, who may or may not have good credit
history. Commercial banks generally prefer to avoid student financing as it is funding an
intangible asset. Human capital, in financial terms, is the ability of the student to generate
income post-graduation and is inherently risky. Many students do not complete their studies
while others do not find equitable employment post studies. Another deterrent to banks for
financing student loans is their small loan size, which makes it unprofitable to process and
monitor these loans.

The sustainability of any student-financing program is dependent, on how much of it is

repaid. Most lending institutions apart from their initial capitalization by public funds depend
on repayments to continue granting of loans. With high default rate it becomes unsustainable

11
for public financing of such student aids. Private goods cannot be procured at the expense of
public money indefinitely due to spiralling costs and lack of budgetary allocations.

1.5 INDIAN BANKS EDUCATION LOAN SCHEME

The current loan scheme was envisaged and developed in the year 2000. It was a
replacement for the National Loan scheme, which was discontinued, in the early 1990s. On
the 13th of June 2000, the finance minister of India had requested the commercial banks to
provide financial assistance to the poor and meritorious students. Under this directive a study
group was formed under the chairmanship of Mr. R. J. Kamath, the Chairman and Managing
director of Canara Bank. The recommendations of the study group were used to frame the
comprehensive model educational loan scheme developed by the Indian Banks Association
(IBA). The scheme was announced in the Union Budget of 2001-2002 and launched on April
28, 2001 (refer Annexure 1).

The scheme was designed to provide loan assistance to the poor and meritorious
students and to be administered through the commercial banks. The government did not
provide any assistance or guarantee for the loan. The scheme was comprehensively revised in
September 2002. The loan amounts were divided into three slabs with:

 A clean and unsecured credit facility of up to Rs. 400,000.



 For amounts above that up to Rs. 750,000, the borrower had to provide a third
party guarantee

 For amounts above Rs. 750,000 the loan had to be collateralized with a suitable
asset.

The students were eligible for up to Rs. 1,000,000 for studying in India and Rs.
2,000,000 for studying aboard. The margin requirements were nil up to Rs. 400,000 and 5%
and 15% for amounts above that for studying in India and outside India respectively. The
Indian educational loan scheme used the Canadian model and provides assistance for tuitions,
examination fees, library fees, travel expenses and insurance if required. In addition the
student is eligible for purchasing books and laptops etc. through this financing.

However the loan scheme set a list of courses / institutions which were eligible for
availing of these loans namely, in India:

12
  Approved courses leading to graduate / postgraduate degree and P G diplomas
conducted by recognized colleges / universities recognized by UGC / Govt. /
AICTE / AIBMS / ICMR etc.

 Courses like ICWA, CA, CFA etc.

 Courses conducted by IIMs, IITs, II Science, XLRI. NIFT, NID etc.


 Regular Degree/Diploma courses like Aeronautical, pilot training, shipping,
Nursing

 Approved courses offered in India by reputed foreign universities.

In case the student is intending to study abroad, the following courses were eligible for
the loan:

 Graduation: For job oriented professional / technical courses offered by

reputed universities

 Post-graduation: MCA, MBA, MS, etc.

 Courses conducted by CIMA- London, CPA in USA etc.

 Degree/diploma courses like aeronautical, pilot training, shipping etc.

The scheme also made it mandatory for the borrower to be an Indian national. To ease
the burden on students, the banks have been urged to provide a moratorium or a repayment
holiday for the duration of the course plus 1 year, or six months after getting a job whichever
is earlier. The scheme has been designed to provide maximum facilities to assist the education
of the student.

The borrower can repay the loan in equated monthly instalments (fixed instalment
loans) in a period of 10 years in case the loan is below Rs 750,000 and in 15 years in case the
loan is above Rs 750,000. In addition the banks have been prohibited from charging any kind
of prepayment penalties for educational loans. In addition, the banks have been prohibited
from charging processing fees for course in India. In case the loan is being taken for study
abroad, the banks can at their discretion charge a processing fee.

13
 The Educational loan scheme launched by IBA in consultation with the Finance
ministry and Reserve Bank of India has been very successful in providing means of financing
for funding in higher education. The educational loan portfolio, of commercial banks, has
been estimated, as below 500 Crores of rupees before 2000. Since the launch of the new
scheme, the disbursement of the education loans has grown at an exponential rate from
estimated disbursement of Rs 500 Crores before 2000 to almost Rs. 57,500 Crores as of 2013.

60,000 50
45
50,000
40
35
40,000
Outstanding in Crores 30
30,000 25
# of Accounts in Lakhs
20
20,000
15
10
10,000
5
0 0
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
Source: Reserve Bank of India

Figure 1.4 Growths of Educational Loans (All Banks)

The number of loans sanctioned has increased from 2.49 Lakhs in 2003 to 25.70 Lakhs
in 2013. It was estimated to be less than 1 Lakh before the launch of this scheme. It should
also be noted that the spurt in education has coincided with the opening up of the Indian
economy to globalisation, leading to better and less restrictive policies and procedures. This
has also contributed to the spurt in the credit off take of education loans. Foreign exchange
flows were highly restrictive before opening of the Indian market, due to limited foreign
exchange inflows and high sovereign loan repayments.

The interest rates offered by the banks are based on market rates. In an ideal situation,
to maximise the off-take of educational loans and to encourage financially and socially
disadvantaged youths to pursue higher education, a scheme where the interest is subsidised
would be beneficial. However, in absence of any enthusiasm from the policy makers in
providing public funding, banks have been asked to provide these loans with their own funds

14
at their own risk. To compensate these risks, restrictions have not been placed on the banks
for pricing of these products. Table 1.4 shows a sample, of the interest rates being charge by a
random selection of banks. The list is not comprehensive.

Table 1.4 Sample Indicative Lists of Interest Rates

Interest Rate (in %)

Interest Up to 4 to Above
Bank
Rate Type 4 lakh 7.5 lakh 7.5 lakh
SBI Floating 11.25 12.75 11.75
Bank of Baroda Floating 10 12 12
IDBI Bank Floating 11.25 11.5 11.5
Allahabad Bank Floating 11.25 11 11
HDFC Bank Fixed 12-14 12-14 12-14
Andhra Bank Fixed 11 12.75 12.75
Bank of Floating 9.5 10 10.75
Bank of Maharashtra Floating 10.25 11 11
Canara Bank Floating 11 12 9.75
Central Bank of Floating 10 10 10
Federal Bank Fixed 13.25 14.75 14.75
Indian Bank Floating 12.5 12.5 12.75
Indian Overseas Bank Fixed 10.25 11.5 11.5
Karnataka Bank Fixed 12.25 13.25 13.25
Oriental Bank Floating 10.75 11.5 11.5
Punjab National Bank Floating 10.5 11.25 11.25
State Bank of Mysore Floating 11.25 12.75 NA
UCO Bank Floating 11.25 11.75 11.75
Union Bank of Fixed 11.75 12.25 11.5
Vijaya Bank Floating 11 11 11
City Union Bank Fixed 13.75 14.5 14.5
Punjab & Sind Bank Floating 12 12 NA
Syndicate Bank Floating 10 10 9.5
United Bank of India Floating 10.25 11 NA

Source: Multiple websites (February 2015)

15
 The pricing of the loan is range bound as most of the banks have similar cost of funds
for educational advances.

Banks regularly report the status of the advances to the Reserve Bank of India.
Advances in Education sector is considered to be part of Priority sector advances. As per RBI
norms, every bank is mandated to advance at least 40% of their advances to the priority
sector. Therefore as an added incentive to the commercial banks, educational loan can be
classified as priority sector lending by the banks.

The number of educational loans and the amounts disbursed by a sample of

commercial banks is given in table 1.5.

Table 1.5 Bank-Wise Education Loans (31.03.2013)

Number of Amount
Name of Bank
Accounts (in Crores)
Allahabad Bank 47,610 1,262
Andhra Bank 61,542 1,427
Bank of Baroda 88,743 1,970
Bank of India 122,839 2,412
Bank of Maharashtra 27,218 553
Canara Bank 217,434 4,267
Central Bank of India 109,762 2,527
Corporation Bank 49,897 1,150
Dena Bank 15,391 328
Indian Bank 204,691 3,650
Indian Overseas Bank 201,285 2,978
Oriental Bank Commerce 48,449 1,227
Punjab National Bank 155,879 3,588
Punjab & Sind Bank 7,003 219
Syndicate Bank 113,138 2,556
UCO Bank 50,571 1,141
Union Bank of India 90,807 2,082
United Bank of India 23,285 552
Vijaya Bank 34,393 670

16
 State Bank of India 604,339 13,753
State Bank of Bikaner 22,449 501
State Bank of Mysore 29,883 614
State Bank of Patiala 15,020 405
State Bank of Hyderabad 51,425 1,123
State Bank of Travancore 109,705 2,394
IDBI Bank 6,707 171
Others 60,535 3,980
Total 2,570,000 57,500

Source: RBI &Banks Websites

The largest advance is by State Bank of India. SBI is the largest bank in India with the
maximum number of branches. The second largest lender is Canara Bank, due to it being the
nodal agency for implementing the scheme.

To further ease the burden of repayment, the government has provided tax benefits to
the graduates. This is based on the assumption that students who have just passed out need
time to settle and to improve their economic conditions. The repayment of an educational loan
is deductible under section 80E of the Income Tax Act. The annual limit for deduction is Rs.
40,000 (for both the principal and the interest). Only loans borrowed for higher education (full
time studies in any graduate or post-graduate, professional, and pure and applied science
courses) may claim deduction. The deduction will be available for a maximum of eight years
starting from the day the borrowers starts his repayments.

1.6 VIJAYA BANK

Vijaya Bank is a medium sized commercial bank with its Head Office in Bengaluru. A
few people from the agrarian community in Mangalore founded it, in October 1931. The
objective of the founder was to promote banking habits, thrift and entrepreneurship among the
farming community of Dakshina Kannada district in Karnataka State. The bank became a
scheduled bank in 1958 (Vijaya Bank, 2015). During the years between 1963 and 1968, the
bank took over nine smaller banks thus growing inorganically to become an all India bank.
The bank was nationalised on 15th April 1980. As of 31st of March, 2014, the bank has a
network of 1565 branches and 1528 ATMs, that span all 28 states and 4 union territories in
the country.

17
 The bank is organised into twenty-seven regional offices spread across all the states of
India. The regional offices are based on the number of branches and cities that it handles. The
aim of the bank is to ideally establish one regional office for 50 branches. However, for
historical and geographical reasons, many regional offices handle more or less than 50
branches.

The complete list of the regional offices and its branches is as per table 1.6.

Table 1.6 Region & Branch Structure

Number of Area / City

Regional Office
Branches Covered

Ahmedabad 79 Ahmedabad
Baroda
Jamnagar
Porbandar
Rajkot
Surat
Nawsai
Bengaluru North 64 Bengaluru
Bengaluru South 63 Bengaluru
Bhopal 44 Bhopal
Gwalior
Indore
Jabalpur
Chandigarh 66 Chandigarh
Amritsar
Jalandhar
Ludhiana
Patiala
Rohtak
Srinagar
Chennai 55 Chennai
Coimbatore 51 Coimbatore

18
 Madurai
Salem
Trichy
Tuticorin
Erode
Delhi 66 Delhi
Faridabad
Gurgaon
Guwahati 42 Guwahati
Shillong
Tripura
Hanagar
Aizawal
Kohima
Imphal
Hasan 62 Hassan
Hubli 79 Hubli
Belgaum
Hyderabad 78 Hyderabad
Jaipur 40 Jaipur
Udaipur
Jodhpur
Ajmer
Kochi 48 Kollam
Kottayam
Ernakulam
Alapuzha
Kolkata 55 Kolkata
Howrah
Bhubhaneshwar
Kozhikode 58 Kozhikode
Lucknow 52 Lucknow
Allahabad
Kanpur

19
Mangalore 75 Mangalore
Meerut 41 Meerut
Agra
Aligarh
Bareilly
Dehradun
Ghaziabad
Muzafarnagar
Saharanpur
Mumbai 72 Mumbai
Mysore 76 Mysore
Nagpur 43 Nagpur
Raipur
Patna 48 Patna
Dhanbad
Gorakhpur
Jamshedpur
Ranchi
Varanasi
Pune 38 Pune
Solapur
Panaji
Madgaon
Kolhapur
Shimoga 41 Shimoga
Bhadrawati
Udupi 67 Udupi
Vijayawada 62 Vijaywada
Vishakhapatnam
Kakinada
Guntur

Source: Vijaya Bank

20
 The bank has a total of 1,565 branches, spread across all the states in India. Out of
these the majority of the branches are in their home state of Karnataka. Five hundred branches
are in Karnataka, which is 32% of the total number of branches. Of the total regional offices, 8
are based in Karnataka. Almost half of the branches are based in South India with 47% of the
branches being in the four states of South India. The bank has twelve regional offices in South
India. The educational loan advances show a similar pattern based on their geographical
presence. The bank has a very robust business per employee with a metric of Rupees 14.86
crores per employee and business per branch of Rupees 129.83 crores.

The bank has been very proactive in adoption of technology and has launched various
technology intensive products like ATM's with its related network, Cash Deposit Machines,
Credit Cards, Internet Banking, Mobile Banking, Phone Banking and other all India network
services like NEFT and RTGS. The bank has an employee count of 12,000 (Vijaya Bank,
2015).

The bank is active in all the different types of conventional banking. Their product
bouquet is wide and deep with the bank catering to all the different types of banking needs.

Table 1.7 Banking Products & Services Offered

Category Product Category Product

Loans & Advances Vijay Home Loan Savings &Deposits Savings Bank

Education Loan Current Account

Education Loan
Term Deposits
Vocational
Personal Loan

Vehicle Loan NRI Services International banking

Loans to Transport
Foreign Exchange Services
Operators
V Vehicle Treasury Services

Jewel Loans Remittance Services

NRI/ NRO/FCNR
Vijaya Gold Cash Credit
Accounts
V Reverse Mortgage

Loan for Traders Card Services Domestic Cards

V Restaurant Verified by Visa Cards

V Secured Overdraft Global Cards

21
 V Rent Debit Cards

Loan Against Property V Secured Overdraft

Loan for Medical

Practitioners V Rent

V Equip

V Cashew

Vijaya Mangala

V Solar

Source: Vijaya Bank

The credit portfolio of the bank is very strong and growing rapidly. As can be seen in
Table 1.8, the gross advances have grown 8.3% during the year. The bank’s priority sector
advances, which are loans mandated by the regulators, has also grown substantially during the
year. They have witnessed a growth of almost 25%. During the year from April 2013 to
March 2014, their education loan advances has also grown by 18%.

As part of its profit maximisation, risk expansion and asset class dilution strategy, the
bank has been growing its retail banking portfolio substantially. The bank is able to get better
spreads in the Retail loan sector. The spreads are the interest income that the bank makes and
is the difference between the yields on its loans minus the cost of funds. The growth of non-
retail loans has been poor in India in the last few years mainly due to the slowdown of
manufacturing activity. The bank has also been affected by the slow and sluggish demand for
industry credit. The education loan portfolio shown in Table 7 below is as of 31st December
2014.

Table 1.8 Credit Advances Portfolio

Dec-2013 Dec- 2014

Item Growth %
(in Rs. Cr) (in Rs. Cr)

Gross Advances 73,100 79,136 8.3%

Priority Sector Advances 22,008 27,450 24.7%
Total Agriculture 6,903 10,917 58.1%
MSE 10,255 11,005 7.3%
MSME 12,953 12,737 -1.7%
Retail 15,090 17,609 16.7%

22
 Housing Loans 5,435 6,066 11.6%
Education Loans 742 874 17.8%

Source: Vijaya Bank

The bank's asset quality is acceptable. The bank has a proactive risk management
department, which uses the latest risk management tools. The bank has also developed in
association with its information technology partners, a comprehensive asset tracking module,
which provides near real time dashboards of their asset portfolios and other performance risk
metrics. In addition the bank actively manages risk by maintaining adequate provisions for
bad debts. Their provisions coverage ratio as of 31st Marc 2014 is at 64%. Table 1.9 gives an
overview of their asset quality and Non-Performing asset rates as of the end of financial year
of 2013-2014.

Table 1.9 Asset Quality (31st March 2014)

31st Mar-2013 31st Mar-2014

Item
(in Rs. Cr) (in Rs. Cr)

Gross NPA 1,533 1,986

Gross NPA (%) 2.17 2.41

Net NPA 910 1,262

Net NPA (%) 1.30 1.55

Provision Coverage Ratio (%) 68.31 64.05

Source: Vijaya Bank

The bank’s spread has been under pressure for the last few quarters. Macroeconomic
conditions, especially high inflation rates, have made the Reserve Bank of India keep the repo
rates and other quantitative tools stringent. The credit growth in the country has been sluggish
for the past few years. However, the bank is able to maintain an average spread of 2% - 4%
for the past few quarters as shown in Figure 1.5.

23
 15%
Yield on Advances

Cost of Deposits
13%

11%

9%

7%

5%
Mar 2013 Jun 2013 Sep 2013 Dec 2013 Mar 2014

Source: Vijaya Bank

Figure 1.5 Profitability Indicators (CoD & YoA)

The Bank has managed to keep the cost of servicing their customer’s deposits within a
range between 7% and 9% in the last few quarters. Despite intense competition in the market
and stringent regulatory framework regarding interest rates the bank has managed to remain
competitive in relation to other public sector banks. At the same time their advances across all
loan types has consistently been above 11%. Therefore the bank has managed to keep their
spreads on a positive side and it compares favourably too many other similar sized public
sector banks.

24
 REVIEW OF LITERATURE

2.1 INTRODUCTION

As part of academic research, the researcher has to review the existing literature in the
area of his study. It gives an idea of the overall research conducted in the similar field, by
other researchers previously. It provides an important basis for formulating the hypothesis and
ensures that the work is not duplicated. A literature review can be defined as a selective
analysis of existing research that is relevant to the topic, showing how it relates to your
investigation. It explains and justifies how the investigation may help answer some of the
questions or gaps in his area of research.

According to Watson and Webster “A review of prior, relevant literature is an essential

feature of any academic project. An effective review creates a firm foundation for advancing
knowledge. It facilitates theory development, closes areas where a plethora of research exists,
and uncovers areas where research is needed (Webster & Watson, 2002). Review of literature
provides the researcher with the background to the problems and the prior work and issues
faced by previous researchers.

Mouly (1984) has explained the importance of the literature review in the academic
process as ‘It is a critical step which invariably minimizes the risks of dead ends, rejected
studies, wasted efforts, trial and error activity-oriented approaches and even more importantly
enormous findings based on faulty research designs.’

According to Hart, "The need to uncover what is already known in the body of
knowledge prior to initiating any research study should not be underestimated (Hart, 1998).
Shaw noted that the process of the review should ‘explain how one piece of research builds on
another’ J. Shaw (1995).

According to Levy & Ellis, an effective literature review should include the following
characteristics: a) methodologically analyse and synthesize quality literature, b) provide a firm
foundation to a research topic, c) provide a firm foundation to the selection of research
methodology, and d) demonstrate that the proposed research contributes something new to the
overall body of knowledge or advances the research field’s knowledge base (Levy & Ellis,
2006).

25
 The chapter has been divided into theme-based sections. Section 2.2 analyses the need
for literature review in the academic process. Section 2.3 looks at the existing literature on the
concept of 'Human Capital' and the contribution of college education to its development.
Section 2.4 reviews the existing literature on equity or funding of higher education in an
Indian context.

Section 2.5 analyses the existing literature regarding student loan defaults. The
existing research in the area of educational loan defaulters is mainly from the United States of
America. This study has attempted to use a similar approach to analysing the loan defaulters
of Vijaya Bank pertaining to Bengaluru region. The existing literature on student loan
defaulters has been analysed and reviewed to help in developing a model using available
variables, which can provide the researcher with a method to predict future defaulters.

There is very limited literature on the sources of financing from the supply side of
funding, as this is a relatively new area of public finance, and education has traditionally been
accepted as a governmental responsibility, unlike more developed capitalistic economies.

2.2 NEED FOR REVIEW OF LITERATURE

As Shaw noted (J. Shaw, 1995) a review of literature, besides allowing the researcher
to get acquainted with current knowledge in the field or area, in which he is going to conduct
the research, also serves to explain how this research has been built upon the research work of
others.

The process of reviewing the existing literature helps by:

i. The review of literature helps the researcher to formulate the problem and
define the limits of the research work.

ii. A study of the existing research in the researcher’s field of study helps the
researcher to understand the depth of research already done in the field and
helps to avoid the perils of dead ends as commented by Mouly (Mouly, 1984).

iii. An analysis of the existing literature helps the investigator to clearly define the
scope of his study.

iv. With the concrete definition of the problem, it helps the researcher to state the
objectives of the study clearly and concisely.

26
 v. The investigator, by conducting a thorough review of the literature can avoid
duplication of well-established findings, unintentionally.

vi. A comprehensive study of the existing work provides a better understanding to

the researcher about the research tools and methodologies available and used
by previous researchers and the outcomes. It helps the researcher in developing
his research methodology.

vii. It gives the researcher a better understanding of the issues at hand as it

provides an overall understanding of the future areas of research.

2.3 HUMAN CAPITAL

The term 'Human Capital' has been one of the most controversial issues in
understanding the economics of higher education. As Schultz posits in his seminal study titled
'Investment in Human Capital', it is not clear that skills and knowledge are a form of human
capital. He says that while people acquire skills and other related abilities in managing to live
their lives, the thinking that this is something we can systematically develop through
investment is not clear. Human capital has quantitative and qualitative dimensions. Where the
output can be measured in monetary terms like number of products created per day by a
worker, it is harder to estimate the human capital value. The number of people working in the
economy and the hours they work are quantitative dimensions which be estimated with some
degree of accuracy.

However when it comes to the qualitative aspects of investing in human capital, there
are problems in clearly defining expenditures for consumption and expenditures for
investment. According to Schultz, there are conceptual and practical difficulties in segregating
these investments. He has divided such investments into three classes, namely:

i. Pure consumption investments that only satisfy consumer preferences and does
not enhance the skill or abilities of the individual

ii. Pure investments which does not satisfy any immediate consumer preference or
need and only enhances the abilities of the individual for future consumption

iii. A mix of pure consumption and pure investments. Most economic activities are
in this class where the activity is undertaken with an intention to satisfy
immediate consumer preference at the same time on the expectation that there
27
 will be future rewards.

Since most human activities fall in the third class where it is part consumption and part
investment, the measurement of the capital investment component is highly difficult. In
comparison to measurement of investment in physical goods or a common set of desired
goods like share market etc., the quantification of investment in human capital is difficult to
ascertain with any degree of certainty

According to Schultz, despite the inability to measure the human capital with any
certainty, it is possible to understand the five areas where it has major implications namely:
(1) in the area of health and lifestyle; (2) apprenticeship and on the job training; (3) education,
both basic and higher education; (4) non formalised study programs for adults or the
availability of opportunities for adults to improve their skills and knowledge outside the
formal training sector; (5) individuals and family migrations to better economic opportunities.

Of the five areas, identified by Schultz, the two relevant categories for this study are:

 Apprenticeship and on the job training


 Education as understood in modern societies

Traditionally, human skill development, especially in speciality skilled jobs (for e.g.
blacksmith, ironsmith etc.) depended on a process of on-the-job training or apprenticeships. In
Europe, from the late 13th century onwards, such apprenticeship agreements were codified
into general laws and regulations. A blacksmith’s apprentice for example, would be
indentured to his master for a period of seven years. After the period of service, the apprentice
was free to leave the services and start his own business. Therefore, in earlier times human
capital was created through a more direct process. Skills were transferred and the economic
and social skills required for managing the activity was learned on the job. The economics of
such investments is straightforward. The apprentice works during the initial period learning
on the job. But at this stage his productivity and contribution is less and the master bears the
cost of maintaining the apprentice. Later as the apprentice develops his skills, the master
expects to recoup his initial investment. However, in modern society in India, the concept of
apprenticeship is reducing and is almost disappearing. One of the reasons attributed to the
disappearance of this mode of human investment is the rise of the schooling system and the
modern education system.

28
 Schultz is of the view that investment in education has risen considerably since the
beginning of the 20th century and might explain the rise in earnings. This is borne out by the
Indian experience in the number of the students enrolling and the growth of universities and
colleges. To study the impact of education in understanding the concept of human capital, it is
necessary to understand the cost of the education system. The salaries of then teachers and
other ancillary service providers, the cost of establishing the infrastructure, libraries,
administrators, and the cost of maintaining and operating the education system needs to be
ascertained. This can be calculated to a certain degree. The other component of the human
capital in education is the cost of the income foregone by the student. This is difficult to
ascertain but the general rise in education costs indicate that there is a high-income elasticity
for the demand for education. By some estimates, the cost of completing a medical course in
India from a reputed private college has increased three fold in the last two decades. Yet, the
demand for such courses has not reduced, thus providing a convoluted proof that the returns
on education in terms of investment is more attractive than investments in other non-human
goods (W.Schultz, 1961).

Therefore it is clear from this study that investment in education can be considered as
a form of investment in the development of human capital and can be to some extent
considered as an investment with measurable rate of returns.

This has implications for the Indian policy makers in their social and economic
development agenda for the student population through providing access to easy financing for
higher education.

The second, important development, in understanding of the concept of 'Human

Capital' was developed by Becker (Becker, Investment in Human Capital: A Theoretical
Analysis, 1962). According to Becker, Human capital refers to the knowledge, information,
ideas, skills, and health of individuals. This is the 'age of human capital' in the sense that
human capital is by far the most important form of capital in modern economies. The
economic successes of individuals, and also of whole economies, depend on how extensively
and effectively people invest in themselves. Becker was one of the first economists to develop
a quantitative framework for measuring the human capital. He developed a model, which
could measure the rate of returns in investment in human capital. A student enrols for an
education program, especially a higher education degree, with an expectation of future
rewards. Becker's basic model is simple. He says, like a machine, a human agent of
production is developed through application of outside resources into the individual (for e.g. a
29
teacher teaching the students in mathematics). Such input results in the output of a trained
human agent or worker who can provide enhanced value due to his additional training.
However the process of imparting this additional value can take time and its benefits are
spread out over a longer period of time, the investment in such imparting of skills can be
considered as an investment. In simple terms this can be compared to a baker buying an
additional machine to bake his bread. While he invests in the machinery today, the expected
returns are over a long period of time and it will take considerable time for him to recoup his
investment. If we were to take this analogy further to his example of a medical student
investing in his studies, the student expects to recoup his investment over a long period of his
practice.

In the development of his framework, Becker has further distinguished between

general and specific training. College and higher education no doubt develop the social and
inter-personal skills of its students, but that by itself does not provide a return on the
investment. The general training provided is useful to multiple employers, whereas the
specific training is useful only to one employer. This explains, according to him, the incentive
for an employer to invest in this training. This same framework explains the incentive for the
governments and the industry to invest in the students in India. Becker's framework takes the
economic argument of the inputs of modern economy like land, labour etc. and expands this
framework using a quantitative methodology. He has explained the variables that can be used
for measuring the rate of return in human capital investment. However, for our investigation,
on the overall development of human capital through higher education, this is not relevant.

In 'What Matters in College: Four Critical years Revisited', Astin (Astin, 1977), has
investigated the effects of college education on the students and the psychological and
sociological effects of such education. He has studied more than twenty four thousand
graduates who had passed in the late 1980s and the effect their college education had on their
subsequent lives. He has identified 192 measures of evaluating the college environment,
which has been classified under five categories, namely:

 Institutional characteristics (Institutional)


 Curricular characteristics (Institutional)

 Faculty environment (Institutional)

 Student environment (Student body)

30
  Individual involvement (Student body)

His study concluded that students who pursue higher education develop a greater sense
of interpersonal and intellectual competence and a greater commitment to developing a
meaningful philosophy of life. The students post completion of higher education displayed
substantial increase in their knowledge and skills. The study also found that colleges, which
had a strong student orientation by the faculty, had a greater impact with the student’s
development.

The study has found that student attitudes and behaviours are shaped by -

 The type of the institution does not directly impact the effectiveness of the
under graduate education and what matters is the environment prevalent in the
college and the attitude of the faculty.

 The students performance in standardized test are different from those that
enhance retention and other cognitive and affective outcomes

 The most important factor in developing the student’s attitude and perception
is the peer group.

Astin, analysed the attitude of students in reference to their aspirations, concepts of

their self, values systems, attitude towards achievements, and attitudes towards contributions
to the development of a healthy society. The study shows how contrasting types of colleges
affect outcomes .The study also analysed the relative impact of institutions versus the effects
of developing maturity in the development of the student’s attitudes. Astin suggests that a
college's judicious and imaginative use of peer groups can substantially strengthen its impact
on student learning and personal development.

In their study on ‘How College Affects Students’, Pascarella and Terenzini analysed
large number of students and concluded that college seniors have a good grasp of the labour
markets. College graduates have a high degree of pressure to find gainful and satisfying
employment post their studies. The inter-generational study spread over a period of thirty
years found some general conclusions on the effect of higher education on subsequent career
and other life events.

 The study found that higher education in general provided and occupational
advantage over non-graduates (in their study the advantage was quantified as
31
 9% to 17% advantage over non graduates). A vocational degree provides the
individual a net occupational advantage of around 5% to 9% advantage over
high school diploma.

 The study found evidence to suggest that higher education enables greater
workforce participation and decreases the chances of unemployment.
However, higher education does not guarantee suitable employment.

 Their study also found the college education tends to have a positive, though
indirect effect on job satisfaction through its impact on job prestige and
earning, job autonomy and non-routine work.

 The study also analysed the earnings ability of the different categories of
students. They found a direct difference in earnings between college graduates
and non-graduates in terms of their dollar earnings. . In men, on average they
found that the college graduates earned 37% more and for women the
difference was around 39%.

The study concluded that while the causal link between college graduation attainment
and occupational advantages and earning capacity is complex, there is some evidence to
suggest that on average the human capital post college graduation yields a higher rate of
return. In their study they estimate, the investment yields to be in the range of 9% to 16%
(Pascarella & Terenzini, 2005).

2.4 EQUITY IN HIGHER EDUCATION

The past few decades has seen a fundamental change in access to higher education.
Previously, it was reserved for a few of the richest and the brightest individuals, often through
hereditary advantages of birth and wealth. The democratization of access has led to a rapid
growth of college enrolments. However, this rapid growth has led to issues in providing
quality access to student aspirations. While the supply side issues were handled through
liberalisation of school licensing requirements, creation of autonomous universities etc., the
issue of financing on the demand side has still not been satisfactorily resolved. Governments
have been traditionally reluctant to assume responsibility due to resource issues. Other public
needs like health, infrastructure and public distribution systems demand higher resources and
due to its political sensitivities, is more catered to by the political establishment.

32
 Financing of educational and educational institutions has been a sate funded initiative
for the last few decades, with the central government bearing almost three quarters of the cost
(Tilak, 1993) .The share of private sector though increasing in the last few years, is still
geographically limited with limited excellence in the institutions. Public funds, through state
and central government grants continue to increase, but increasingly there are indications of
resource shortages in future.

Tilak has analysed the issue of education funding in India and concludes that while
'privatisation' is one solution, it is not the best solution. He argues that there are four critical
decisions in respect to funding higher education:

 How much share of the national budget should be allocated to higher education


 How much should higher education be allocated out of the total education
budget

 The methodology used in allocating these resources among various institutions

 The distribution of the funds among function.

The study also identified that the expenditure per pupil has not kept pace with the total
increase in education expenditure. The total expenditure on education has shown double digit
growth over the last few decades whereas the expenditure per pupil has only increased by 1.3
times. If we take inflation into consideration the real expenditure per pupil has not increased
significantly. He also argues that the allocation of cost between recurrent and capital
investments is not optimum. Higher education being a labour and equipment intensive activity
due to higher teacher to student ratios, higher laboratories and library expenses etc makes the
system expensive. The recent growth in distance education is a result of these increased
expenses. The distance education courses provide a dual advantage to the institution as the
share of the teachers’ salaries are considerably lower in comparison to the higher growth in
fees. The drawback with distance education is the lack of quality in the education imparted.
Serious research and innovations come from brick and mortar institutions. Therefore
consensus exists between administrations and academia that qualitative improvements in
education can only come from non-distance learning institutions.

Tilak has identified sources of finance for higher education in India as:

 Central government
33
  State governments

 Local governments

 Non-Government sector or private sector

 Other community driven initiatives

The data as of 1983-84 shows the government contributing 79.6%, local bodies
contributing 1.1%, fees charged to students contributing 13.8% and others sources
contributing to 5.4% of the total costs of higher education.

The limitations he has identified in public financing of higher education are:

i. The rate of return from public institutions is lower than rate of return from
private entities (including semi-autonomous entities like IIMs etc.)

ii. Even though the democratization of education has opened up opportunities

to all sections of the society, the majority is still from the relatively higher
economic class, whose actual economic wealth is higher than what they are
actually charged.

iii. This creates unequal and unnecessary subsidies which the current state of
public finances should avoid bearing. Government revenues are largely
through indirect taxes with direct income tax being a small part of the total
collections. Hence given that more economically well off sections are
opting for higher education, the government is effectively cross subsiding
the rich with collections from the poor. He further states that such large
public subsidizations will lead to higher education becoming highly
inequitable with serious regressive effect on income distribution.

iv. Due to limited public resources in our country, the growth of investment in
education will be at cost of investment in elementary education. However
several studies have shown that in terms of country's development, returns
from elementary education is higher than higher. Hence it does not make
economic sense to increase the share of higher education at the expense of
elementary education.

34
 In his study, Tilak has explored the possibility of financing higher education through
the levy of graduate tax (an education tax on employers who use educated manpower). The
argument for graduate tax rests on the premise that employers who employ this highly capital
invested resource should share the cost burden in the production of this 'human capital'. The
main drawback, he argues is that employers will be negatively incentivised to employ
graduates and would prefer to use less educated or uneducated man power. Tilak has also
suggested levying a cess for higher education. Currently, many states levy cess but for general
education. Perhaps, the government can introduce a higher education cess with the funds
clearly allocated for the development and financing of institutions of higher education.

He further explores the different forms of privatization as an alternative to government

funded financing. He has classified privatisation into; (a) extreme privatisation with total
privatisation of everything with negligible government intervention; (b) strong form of
privatisation leading to full recovery of fees from students; (c) weak form of privatisation with
public funding of higher education with partial recovery from other sources including
students; (d) pseudo privatisation implying private administration of education with public
finances. He further states that only option (a) and option (c) are feasible in India.

In conclusion, he states that India being a resource poor country needs to look at
innovative and sweeping reforms, which might be necessary to keep funding higher
education.

2.5 LOAN DEFAULTS

There have been numerous studies, mainly in the North American continent about
student loans and defaults. Loan defaults result in enormous costs to the governments and
other commercial institutions. In the US, the federal government has provided guarantees to
commercial banks that act as commercial collection agents.

Flint (Flint, 1997) predicts that student loan defaults are set to rise and eventually will
surpass government spending on many more needed areas of public expenditure. The large
student loan defaults have resulted from the growing use of student loans for financing higher
education. This experience is similar in India with the exponential growth of student loans.
The study explores how the increase of loan defaults in the American universities etc. is
having an effect on government policies and school administrators. Flint has utilised a large
sample of the Student Loan Recipient Survey (SLRS) conducted by National Post Secondary

35
Study Aid study (NPSAS: 87) in 1987. The SKRS surveyed 11,847 students from 1,412
colleges who had left the school between 1976 and 1985. In addition, Flint has merged data
from 14,000 academic transcripts to include school performance in the variable list.

As suggested by Flint, student loan defaults can be explained using theoretical

perspectives from three disciplines: economics, sociology and psychology.

i. Economics perspectives are based on two economic perspectives of student

behaviour, firstly, based on the human capital theory which explains student
behaviour of investing in higher education with the expectation of future
financial returns which will outweigh the current cost. Human capital theory
has been covered earlier in Section 2.2 above. The second economic
perspective is provided by the 'Ability to Pay' theory, which posits that student
repayment behaviour is a function of prioritizing consumption categories. The
theory suggests that people’s first priority is for essential expenses related to
subsistence and medical expenses. The residual income, if any, is the
discretionary income, which the borrower may use for repaying loans and other
financial obligations.

ii. Sociological perspectives have looked at status attainment and social

integration models to explain student’s behaviour. Status attainment model
developed by Blau & Duncan (Blau & Duncan, 1967)explain the role of a
student's educational experience in social mobility. Another sociological
perspective is provided by the social integration models, which are, derived
from the person/environment interaction theories ((Baird, 1990), (Walsh,
1973)). The social integration models posit that the physical and social
characteristics of one's surroundings exert significant influence on an
individual’s behaviour.

iii. Student defaults depend on the degree to which student’s future behaviour can
be influenced; psychological perspectives have looked at theories of attitude
formation to understand defaults. Counselling affects the behaviour through
direct intervention in attitude formation. In this study, Flint has studied the
influence of informants (usually friends and credit professionals, university
loan administrators etc.) to the financing and repayment process. Attitudes
towards student loans and later repayment behaviours may be the effect of their

36
 satisfaction or dissatisfaction with the college experience.

Flint has used these three perspectives to devise a framework for analysing the data
from the SLRS study. He has studied the prior work done by other researchers by classifying
them into one of the above three categories. However, he concludes that while each of these
studies has its strengths, none of them have looked at quantitative method to predict future
defaults. The study has therefore grouped all variables into different categories representing
the student’s characteristics before joining the college, characteristics of the college and
characteristics after completion of the course. The study has used many variables and to name
a few, it has used gender, race, age, father's income, father’s education and institutional
variables like the level of degree offered, composition of the college management, amount
borrowed etc. The variables have been classified as; (a) Student background characteristics;
(b) Institutions Choice characteristics; (c) Student Academic Characteristics; (d) Loan
origination, counselling and award packaging variables; (e) Exit counselling and end-of-grace
period variables; (f) Point-of-Survey variables

The study has arrived at a regression model, which predicts repayment status in around
87% of the cases. In conclusion, Flint argues that Economic and sociological variables do not
influence repayment behaviour to a significant extent. Psychological factors including student
attitudes are the strongest indicator of student behaviour during the repayment period. In
banking industry this is commonly referred as the 'the willingness to pay'.

In their article titled, 'What Matters in Student Loan Default: A Review of the
Research Literature' by Jacob P. K. Gross, Osman Cekic, Don Hossler, and Nick Hillman
(Gross, Cekic, Hossler, & Hillman, 2010) have analysed the existing literature on student
behaviour towards their loan repayments. They have posited that the higher education policy
in the United Sates has shifted from a student grant based aid, to loans as an aid for student
financing. As we have seen earlier, the reluctance of the governments to fund a private good
(student’s education) with public finance is causing this shift in the means of funding higher
education. Such shift has necessitated a deeper study of the reasons for defaults. Are defaults
a function of students inherent values or is it dependent on the institutions characteristics? The
authors have reviewed the studies in this field and have presented their finding in their paper

Their finding indicate that most of the empirical research in loan defaults have used

37
the grouping methodology adopted by Flint ((Flint, 1997) and Pascarella and Terenzini
(Pascarella&Terenzini, 2005). A study of institutions characteristics show that student who
have chosen a four year course have a lower default rates than their peers who have selected
two-year course (Podgursky, Ehlert, Ryan, Donald Watson, & John, 2002). Various studies
indicate that education loan borrowers who pursue less than bachelor degrees (four year
education) tend to borrow more. They also come from lower income households belonging to
minorities, characteristics that show an increase of probability of defaulting (Flint, 1997;
Gladieux & Perna, 2005; Goodwin, 1991).

Similarly some of the studies, which have analysed the age of the students, have
concluded that as age increases, the likelihood of defaults increase. Several explanations for
this inverse relationship between age and default rate have been hypothesised. Herr and Burt
(2005) posit that older students tend to have greater financial obligations and families to
support. Studies, which have analysed the impact of gender on defaults rates, have not found
any conclusive proof of any correlation between the likelihood of default and gender (Harrast,
2004; Volkwein&Szelest, 1995). However variables like family structure and the number of
dependants seems to indicate a relationship to likelihood of defaults. Student who claim a
higher number of dependants tend to default more (Dynarski, 1994; Volkwein&Szelest, 1995;
Woo, 2002). Every increase in a dependent child increased the probability of default by 4.5%
(Volkwein&Szelest, 1995).

Studies, which have looked at the debt burden of the students, suggest that as debt
burden increases, so does the likelihood of default (Choy & Li, 2006; Dynarski, 1994;
Lochner & Monge-Naranjo, 2004.

They have also reviewed research, which have studied college experiences in terms of
academic enrolment and intensity, educational attainment, academic preparation program of
study. For the above characteristics when the attainment or degree of difficulty is higher, the
likelihood of default decreases. Where the student has to prepare hard for the admission due to
competition etc. the likelihood of default reduces.

The review concludes by suggesting that the default behaviour of students is a

complex behaviour rooted in multiple variables relating to the students attitudes, peer group,
support of friends and families, economic climate post completion among others.

38
 Wilms et al (Wilms, Moore, & Bolus, 1987) in their study on student loan defaults in
the state of California have researched into the behaviour of defaulters to ascertain if the
reasons for defaults are with the students or the institutions they were enrolled in. They
analysed the borrowers who have availed of loans under the Guaranteed Student Loan
program (GSL) under the Higher Education Act of 1965. GSL programs are administered
through a partnership between federal government and private banks. They have accessed
data from 93 colleges and 140 accredited vocational colleges. For the institutions part of the
data analysis the colleges were segregated into three categories; (1) High Default rates; (b)
Medium Default rates; (c) Low Default rates.

The sample size for the student data was targeted at 6,000 students (from a population
of 17,000). After further cleaning of the data, a final sample of 5,926 records was selected. As
certain characteristics like ethnicity and prior education data was not available, the researchers
collected additional data from the school for these students. The researchers used cross
tabulations and chi-square tests. Chi-square test was used to test the relationships between the
variables. The researchers have used standard demographic variables like age, gender,
income, education etc. for developing their model. They have used discriminant analysis to
further identify the relationships between these variables.

They have concluded that the most critical indicator of future defaulting behaviour is
whether the student completes the program. They posit that the completion (or non-
completion) of the course indicates behavioural aspects and is linked to personality traits like
persistence, lack of motivation and low expectations. The second important finding of their
research is the relationship between ethnicity, family income and educational attainment. All
these variables have a direct correlation with student repayments. Unlike other researchers
they have also analysed the relative importance of the groups i.e. institutions versus the
student characteristics. They conclude that the student characteristics are significantly more
important than any other characteristics like college, location etc.

Becker in research on the reasons for people's behaviour (Becker, Nobel Lecture: The
Economic Way of Looking at Behaviour, 1993) has posited theories of behaviour based on
economic reasoning. The analysis assumes that the primary goal of individuals is to maximize
their individual conception of welfare. It could be masochistic, spiteful, and altruistic or plain
selfish but in all case the individuals behaves with forward-looking intentions (similar to
concept of future rewards). However he suggests that forward-looking behaviour is a function
of past experiences, past circumstances, childhood integration with parents and society etc.
39
Becker posits that student defaulting behaviour while inconsistent with the expectation of the
society is a result of the past circumstances. The behaviour is constrained by incomes,
dependants, time, and imperfect memory and is also dependant on opportunities available for
earning money. The research has implications for the current research in terms of
understanding the characteristics of the borrowers and their default patterns.

Steiner &Tym (Steiner &Tym, 2005) analysed student borrowers and defaulters who
attended the University of Florida during the period from 1999 to 2002. Their sample size was
around 17,036 students. The default percentage for these students was 4.5 per cent (766
students defaulting out of the total of 17,036). They have used logistic regression to identify
the independent variables and its relation to the default probability. Their study found that
similar to Wilms et al (Wilms, Moore, & Bolus, 1987), the students who completed their
course were three percentage points less likely to default that the students who did not
graduate. The default rate logically is directly correlating to family income i.e. the higher the
family income of the student; the less likely the student would default. This is due to
economic family support provided by the family and friends through temporary grants etc.
Unlike other similar studies where the researchers have used gender as a variable and did not
find any differences between genders, they found that females were less likely to default than
males. Their study found evidence consistent with (Flint, 1997; Wilms, Moore, & Bolus,
1987) of African Americans having a higher propensity to default in comparison to Hispanics
and Whites. Another finding of their study is also consistent with earlier research where it was
found that students who have a higher GPA tend to have lower probability of default. The
researchers have also analyzed whether parental marriage status have any bearing on student
behaviour and concluded that students who come from a divorced or single parent family
structure tend to default more.

In the research on the role of finance in student’s attitude to education persistence, the
authors Cabrera et al (Cabrera, Nora, & Castaneda, 1992) have analysed the significance of
finance in the student’s college persistence. They have used theoretical frameworks to
develop the causal relationships to explain students’ performance in college. Their study is
based on the theoretical model introduced by Tinto (1975, 1987) known as the student
integration model. According to this model, student persistence is a combination of
interactions between the student and the academic and social components of the institutions.
The model developed by Cabrera et al, is based on the assumption (Tinto, 1987) that the
experiences that the student consumes of the academic and social life in the institutions

40
affects their commitment to completion of the course. Their commitment then enhances their
desire to persist. Their model has also analysed the role of finances in the students’ integration
into the social environment by removing their financial barriers to full participation. The
sample used in their study was about 450, consisting of college students enrolled during the
period between 1989 and 1990. Questionnaire developed from the research done by Bean
(1982, 1985), Metzner and Bean (1987) among others, were mailed to about 2,500 students
out of which, responses were received from around 450 students.

The authors developed scores based on their data to measure overall persistence score,
namely:

i. Institutions Persistence (P) was measured based on the students’ continuance

in college as of fall of 1989.

ii. Intent to Persist (IP) was based on the student intent as stated by the student to
continue enrolling in the college

iii. Institutions Commitment (IC) was a composite score arrived by averaging

eight items like feelings of belongings, students perceptions of attainment in
graduating from the college, institution prestige etc.

iv. Goal Commitment (GC) was measured using students statements based on the
items developed by Pascarella&Terenzini (1980) and Dunham (1984)

v. Academic Integration (AI) was arrived at by merging the student's academic

performance as measured by GPA and the scale developed by
Pascarella&Terenzini (1980).

vi. Social Integration (SI) was measured by analysing the students peer group
relations (Pascarella&Terenzini (1980)

vii. Precollege Academic Performance (HSQ) was recorded from the student
previous school records.

viii. Significant Others Influence (SO) was developed the researchers to measure
the support from family and friends.

ix. Finances (FI) measured if the students received aids and loans

41
 The researchers used structural equation modelling to estimate the relationships
between these variables to predict student’s persistence. They concluded that financial aid
helps in the commitment that students display towards their course. Loan repayment
obligations, incentivises the students to complete the course.

Their study also looked at the relationship between student precollege academic ability
and the availability of student financing. In an Indian context while the banks are not
explicitly allowed to discriminate, the researcher has identified biases in the bank personnel
during informal interviews. The tendencies, to reduce availability, or deny availability, of
finance to academically weaker students continue to persist, though sometimes
unintentionally.

Brugel et al (Brugel, Johnson, & Leslie, 1977) have studied the preferences and
attitudes of post-secondary students towards different type of education financing. Their study
tried to ascertain student attitudes towards the level of debt, the period that the students are
willing to consider for repayment and what amount of the future income is reasonable as
monthly instalments as per their current perceptions. This study has explored the
psychological attitudes of the students towards loan programs. They have randomly selected
around 220 students to administer the questionnaire. The study was designed based on the
economic theory of demand which states that the demand for any commodity like student
loans is influenced by several factors like the income, the level of income of average
households, the future earning from the commodity etc. To understand the relative
preferences, the researchers asked the students to rank the various loan features. The results
were ranked as:

i. Length of repayment (7%)

ii. The amount of repayment instalment (31%)

iii. Total amount repaid at the end of repayment (29%)

iv. Interest rate (21%)

v. Income protection or guarantee (13%)

Srinivasan in his analysis of educational loans has attempted to find the significant
factors that contribute to the sanction of the loan by the banks. The researcher has used a
structured questionnaire to collect the data. The study was conducted during 2009 in the

42
National Capital region comprising of New Delhi, Gurgaon, Faridabad, Ghaziabad, Noida and
Greater Noida. Srinivasan has used a probabilistic probit model to arrive at a dichotomous
outcome i.e. loan granted or not granted. The variables, the researcher has used in the analysis
include, loan granted (yes, no), type of bank (public sector, private sector), work experience
(yes, no), type of course (post graduate, graduate), type of institute (government, private),
family income to loan ratio. The researcher has concluded that it is easier for a student
applying for a postgraduate course, in approved institutions with a high family income to get a
loan (Srinivasan & Das, 2011).

Gaikwad&Solunke in their study of the growth of higher education in India have

analysed higher education in India. They have looked at the region wise distribution of the
colleges and have looked at the infrastructural and quality lacunae evident in the higher
education. The authors have concluded that while the number of colleges and universities are
growing, the quality of these institutions is very low (Gaikwad & Solunke, 2013).

Puttaswamaiah in his study of educational loans in India has analysed the trends and
patterns in the growth of educational loans. He has also analysed the pattern of education loan
credit over the last decade. Specifically he has analysed the educational loan trends in Canara
Bank and State Bank of Mysore and the contribution of education loans to their total priority
sector portfolio over the years. The study found that only 3.7 per cent of the total priority
sector lending by public sector banks is being channelled to educational loans.
(Puttaswamaiah, 2010).

Dewan et al (Dewan, Goel, & Malhotra, 2013) have researched the student borrower’s
perception to educational loans in the state of Haryana. The sample consisted of 55
individuals in both undergraduates and graduate levels. The researchers have defined
variables relating to the loan features like: (a) Loan Value Additions (b) Mortgage &Effective
(c) Eligibility criteria (d) Procedure & convenience (e) Disbursement and (f) Rate of Interest.
The study found that the students ranked rate of interest as the most important loan features
(in contrast with this similar factor being the fourth most important feature in a loan program
in Brugel's study). The study also found that the students were unhappy with the security of
immovable property being taken as collateral.

Janetius, Shilpa and Eswaran in their analysis of wilful defaulters among rural students
have analysed the reason for defaults among the students coming from a rural and agrarian
background. They have done a longitudinal descriptive study using a sample size of 76

43
students (55 % males) with 42 per cent applying for undergraduate loans. They have studied
the salaries of these students post completion of their education. The salaries range from
Rupees 60,000 to Rupees 264,000 (mean Rupees 150,000) per annum. They have categorised
the defaulters into (a) Defaulters with genuine reasons for defaulting; (b) Wilful defaulters; (c)
Defaulters with unconventional reasons. They found that around 2% of the sample have
defaulted due to genuine reasons of less or insufficient income. In their study, 16% of their
sample consisted of deliberate wilful defaulters. Almost 82% of their sample has defaulted
due to unconventional reason. They find that the rural students have a mind-set of expecting
loan waivers and other freebies from the government. They posit that the previous loan
waivers by the government due to political considerations have created a poor financial
management mind-set among the rural borrowers (Janetius, Shilpa, & Eswaran, 2013)

Gandharhas reviewed the growth & performance of the commercial banks in India and
their educational loans portfolio. Gandhar has looked at the deficiencies in the current loan
scheme by the scheduled commercial banks. He has concluded that loans being provided
under the educational loan schemes are on commercial basis. The banks are unable to identify
and provide attractive options to the needy but meritorious students (Gandhar, 2010).

Eckwert&Zilcha have analysed the different types of student financing using an

overlapping generation’s framework in the education sector for human capital formation and
economic welfare. As per their study, while there is consensus on the increase in capital
formation through higher education, the expansion of higher education has collided with fiscal
pressures on the government. The tendency among policy makers has been to transfer the
burden to other players. The shift to private funding is based usually justified on the principles
of egalitarian income distribution. Students who attain higher education on average have
higher income prospects and earn more. In such a scenario, public funding constitutes an
implicit transfer of wealth from the poor to the more affluent individuals. They have studied
three different regimes of funding; (a) Regime 1 - Unrestricted access to Credit markets; (b)
Regime II - Unrestricted Insurance of Loans; (c) Regime III - Restricted Insurance of loans.
Based on their empirical research, they have concluded that the individual’s incentive to
invest in higher education is affected by the financing regime, available to the student. They
found that regime three, where the risk is pooled with the restricted insurance by the
government is the best scheme in its ability to generate capital growth.

Calero in his study on market reforms and equity in the financing of higher education
suggests that based on the OECD study on higher education (OECD, 1990), financing systems
44
should not be seen only as a resource allocation mechanism for the governments but also a
method to develop a dialogue between the needy and the funders. According to him, the
European experience has been towards greater market decentralisation and cost
consciousness. Therefore the last few years has seen greater autonomy for the educational
institutions in their funding. This is reflected in their shift from traditional fees based funding
to royalties, greater collaboration with industry, patent filing etc. He notes that this
decentralisation and the importance of generating funds by the institutions itself without
resorting to government funding has led to growing diversity in the type of course being
offered by universities. Calero provides suggestions on how to improve funding in education
from a supply and demand side perspective i.e. how to fund the colleges. He suggests that
institutions look at competitive research funding for generating resources. Institutions with
low levels of original output in terms of research will be in an unfavourable position unless
they step up their research outputs. These will incentivise institutions to seek higher levels of
quality in their students and faculty. He also suggests that governments should get into long
term agreements with the companies. On the demand side, Calero, suggests that to improve
demand side economics, tuition fees should be subsidised and educational loans should be
provided to the needy and deserving (Calero, 1998).

Narayana (Narayana, 2005) has studied the budgetary allocation of government funds
to higher education in India and has come to the conclusion that the government spending on
higher education disproportionately higher in India. He argues that this is not sustainable and
has suggested ways to shift the burden of funding to the private sect and the students. He has
probed various ways for the state governments to reduce their support for higher education.
He has concluded that: -

i. The current form of student financing through commercial banks is wide and
responsive to the changing needs of the students. The scheme is meeting the
requirements of the student for their studies in India and outside the country
(fees, tuitions, housing and living expenses, library and computer expenses
etc.)

ii. The current budgetary subsidies being provided through grants and salaries
cannot be completely replaced by student fees. In this scenario, the student fees
with become huge and uneconomic for the students. Therefore student fee
revision cannot be an appropriate mechanism for cutting subsidies.

45
 iii. Narayana posits that the empirical results suggest that commercial student
loans may not be a perfect substitute for government subsidies for the poor and
needy students in collegiate education. He suggests that student loans which
are a form of credit instrument needs to be complemented by budgetary
subsidy which is a form of fiscal instrument along with other instruments like
scholarships and free education for the more meritorious candidates.

Shen & Zidermanhave analysed the student loan schemes prevalent in around seventy
countries. They posit that almost all these schemes contain significant government assistance.
In addition the administrative costs and credit losses are usually borne by the government or
other intermediaries and not passed on to the students. The researchers have probed two
aspects of the student funding system; (a) the repayment ration or the amount the student is
required to repay; (b) recovery ratio or the amount the lenders, expect to receive aback in
repayments. They find that in many countries the repayment ratio is below 40% even when
not explicitly stated by the lender. This is due to the fact that many of the state sponsored
loans have inherent loan structure that provides hidden subsidies to the students like: -

i. Below market interest rates for the students,

ii. Long grace periods where the student does not have to repay any amount

iii. Loan period where interest is not levied on the loans and

iv. Repayments not being linked to inflation.

The Indian loan schemes being offered by the commercial banks feature two out of the
above four features i.e. lower interest rates when compared to other types of commercial
unsecured loans and the risk being borne by the banks and instalments are not being linked to
inflation. The equated monthly instalment that the student has to pay on completion of the
course and on getting employment is not linked to the inflation and remains fixed throughout
the loan repayment period.

The researchers have used a sample of 44 loan programs from 39 countries. The
countries considered were twelve in Europe, two in Australia, eight in the Americas, fourteen
in Asia and eight in Africa. They found that most of the schemes are based on mortgage loan
type of repayments i.e. fixed monthly instalments. They also found great variation in the
scheme structure and duration: forty years in Egypt to five years in Latvia. Some countries

46
like Australia, England, Ethiopia, Ghana, New Zealand, Sweden and South Africa have
structured their loan schemes as an income contingent loan. The loan repayments are a
percentage of the student’s income and not a fixed amount. When the students income goes
down the repayment burden accordingly reduces in absolute monetary values though might
not reduce in percentage terms. Their study found that the highest repayment ratios are in
countries like Canada, Netherlands and Namibia reaching almost 100% of the loan amount.
These countries do not have any hidden subsidies built in their loan design. At the other
extreme, countries like Egypt, Russia and Nigeria expect only about 10% of the loan amount
in repayment with the rest 90% being processed as hidden and open subsidies. Their analysis
of the loan scheme of India shows a repayment ratio of only 80.23% with 19.77% of the loan
amount being waived as subsidies (Ziderman & Shen, 2009).

Brown & Zehnder in their study of information sharing between lenders and its
influence on repayment behaviour found that pooling of such common individual
characteristics helps the bankers in controlling defaults. The borrowers attempt to keep their
credit history clean in anticipation of future needs. They have analysed the behaviour of
borrowers in terms of disclosure of information when borrowing, its impact on the lenders
decision-making abilities and the ability to enforce contracts. They found that in most market
driven credit markets there is an asymmetry in terms of information between the borrowers
and lenders. Borrowers know more about their needs, their past behaviours and their moral
character. Lenders in absence of these inputs have credit granting problems with lenders
needing to constantly review their policies. They also state that in advanced countries in
absence of strong enforcing laws, such inefficient lending increases the cost for the lenders
(Levine 1998; Jappelli, Pagano and Bianco 2005). They quote previous studies, which have
proved that information sharing between lenders help in reducing inefficient credit selection
(Pagano, Jappelli, 1993). A World Bank study in 2006, posited that information sharing
(colloquially known as Credit Bureaus) exist in more than 100 countries. Such Bureaus can be
public or private. The government creates public registries, whereas a grouping of interested
parties usually creates the private ones. Their study suggests that information sharing
mechanisms by itself does not contribute to lower rates as in absence of an adverse
disciplining mechanism; borrowers ignore their repayment obligations. However, they find
that if information sharing processes are used along with relationship banking (regular and
formal contact between the borrower and the lender), the likelihood of defaults reduce
drastically (Brown & Zehnder, 2007).

47
 The first information sharing bureaus in India was setup by Trans Union along with
SBI, HSBC, IOB, Union Bank of India, Bank of India and various other banks. They set up,
Credit Information Bureau (India) Limited or CIBIL as a Credit Information Company (CIC)
and started functioning from August 2000. Trans Union, an American company based in
Chicago is the majority shareholder holding around 55% of the equity. Next biggest
shareholder is State Bank of India with 6.1% with the balance being held by various banks
and financial institutions. They hold on average around 4% to 6%, in the company.

Rothstein & Rouse, analysed the student behaviour in a specific university, which in
the year 2000, decided to shift from a commercial student loan regime to a system of
financing higher education to meritorious students through grants. They find that while the
returns from higher education (college degree) have risen considerably in the last few years
the cost of this education has risen faster. The earnings or the premium on wage for graduates
rose by twenty seven per cent, while the cost during the similar period increased by sixty three
per cent. As per their study the proportion of students who apply for some aid is almost 65 per
cent in 2004. They have analysed the student choice in employment post completion of their
studies and find that student who do not have any debt tend to choose lower paying jobs in
public service industries. In contrast, students who have high level of debt choose to work in
high risk and high reward jobs. They suggest that the most plausible explanation of the
students’ behaviour is the growing averseness to holding debt. They also measured the
student’s altruistic inclinations towards their college later and find that students who held
financial debt during their college experience in general tend to lower their donations to the
college (Rothstein & Rouse, 2011).

48
 REPORT ON THE PRESENT INVESTIGATION

3.1 INTRODUCTION

The previous chapter consisted of a review of the existing literature in the field of
higher education in reference to development of human capital and the need for student
financing to achieve that objective. The literature also provided glimpses into the problems of
student financing in terms of non-repayment (leading to credit losses by the banks and thereby
leading to higher costs), the cost and the modalities of administering student financing
programs and the general reluctance of the governments to use their public funds for such
private benefits. The existing literature clearly demonstrates and provides support to the
researcher to pursue the research study. The current chapter discusses the methodology used
in conducting the research, the sources of the data and the hypothesis being tested.

3.2 STATEMENT OF PROBLEM

In their study on student loans and the causes of defaults ‘What matters most in
Student Loan Default: A Review of the Research Literature’, Gross et al have argued that the
change in how student loans financing have changed from a government directed social
development cost to a commercial cost borne by banks (Gross, Cekic, Hossler, & Hillman,
2010). This shift in costing of the student financing has been used to analyse the student loan
financing in an Indian context.

Education is commonly acknowledged to increase the quality of life in the country.

The development of modern nations with improved health of its citizens; longer life spans and
generally better social conditions have been attributed to improvements in education in the
last two centuries. There are dissenting views to this theory (Becker, Investment in Human
Capital: A Theoretical Analysis, 1962) with some economists and social investigators, not
finding a strong causal relationship between an educated workforce and the overall quality of
living (educated in this context, is meant as higher education and elementary education is not
generally in dispute). Therefore the cost of funding has been a contentious issue with a
gradual shift from governments delinking their roles from providing higher education. The
gap has been filled with commercial lenders but under government supervision and control
(i.e. priority sector lending).

49
 Due to this quasi market driven financing, the data on returns and sustainability is
diluted and in some cases unavailable. The government has, in the past for political gains,
waived off student loans. These are popular among voters, but add costs to the credit system.
Such waivers lead to expectations among future borrowers and influence their credit
behaviour.

There are has been a limited number of studies done in India to understand the
persistence of defaults among student borrowers. Most of the research in this behaviour has
been done in context of American students. This research attempts to understand the
characteristics of the educational loan borrowers specifically borrowers who have availed of
such loans from Vijaya Bank, to find out the profile of the students and their parents and the
characteristics of the courses they are attending. While empirical studies have shown the
importance of higher education for a country’s development and human capital, the lack of
repayment of these loans by the borrowers post-graduation is worrying trend. The
sustainability of these public programs is doubtful in case such programs continue to be
unprofitable to the institutions administering in the absence of government funds and
subsidies. The study attempts to identify relationships, if any, that exists between the borrower
demographics and default patterns. Industry and academic sources have been utilised
wherever possible in understanding the issues related to student financing.

Using some of the models used by scholars in analysing and studying the student loan
defaulters in the USA, this study has also attempted to derive a group of variables which can
be used to analyse the demographics and psychographic profiles of the students who have
availed the loans from Vijaya Bank. Vijaya bank is a typical public sector bank and the
customers of the bank while largely based in the south, are very similar to customers across
other banks and regions. The bank uses a student loan financing model as prescribed centrally
by IBA and is very similar to the loan offered by other public sector banks. The acceptance
credit criteria and interest rates are competitive and similar to other banks. Hence, the study
has assumed that the students who have availed the loan from Vijaya bank are a reasonably
good representative sample of educational loan borrowers in South India.

3.3 OPERATIONAL DEFINITIONS

NPA - RBI has classified an asset, including a leased asset, as non-performing when it ceases
to generate income for the bank. In case of Educational loans, Non-performing asset is a loan

50
or an advance where interest and/or instalment of principal remain overdue for a period of
more than 90 days in respect of a term loan.

Asset Classification - Banks are required to classify nonperforming assets further into the
following three categories based on the period for which the asset has remained
nonperforming and the probability of realising of the dues:

i. Substandard Assets - With effect from March 31, 2005, a substandard asset would be
one, which has remained NPA for a period less than or equal to 12 months.

ii. Doubtful Assets - With effect from March 31, 2005, an asset would be classified as
doubtful if it has remained in the substandard category for a period of 12 months.

iii. Loss Assets - A loss asset is one where the bank or internal or external auditors have
identified loss but the amount has not been written off wholly.

Education Loans - Educational loans include loans and advances granted to individuals for
educational purposes up to Rs.10 lakhs for studies in India and Rs.20 lakhs for studies abroad,
and do not include those granted to institutions. For this research the education loan is
restricted to the loans advanced by banks as per the policy guidelines issued by Reserve Bank
of India dated April 28th, 2001 (RPCD.PLNFS.BC.NO.83/06.12.05/2000-01)

Gross Enrolment Ratio -The United Nations Educational, Scientific and Cultural
Organization (UNESCO), describes 'Gross Enrolment Ratio' as the total enrolment within a
country ‘in a specific level of education, regardless of age, expressed as a percentage of the
population in the official age group corresponding to this level of education.’

RACPC– Retail Assets Centralised Processing centre is the specialised department setup by
the bank to handle complex and larger retail loans. They process mortgage loans, personal
loans, educational loans etc. All educational loans above Rs 750,000 are processed by the
RACP centre.

3.4 VARIABLES OF THE STUDY

The study uses eighteen demographic and economic variables taken from the student
loan application forms. The research attempts to identify if there is a common set of attributes
influencing the repayment behaviour of student borrowers. The link between students’
behaviour, post completion of the study and the type of study, parental backgrounds etc. are

51
highly complex and inter-connected. Do life circumstances—like the types of jobs and
income levels of students after they graduate—have an impact on default rates?

The study methodology, group student behaviour into various categories on the
assumption that student’s repayment is dependent on the initial student’s background and
experiences during and post-graduation. Several empirical studies have shown that the choice
of the institution is an important factor in predicting post completion repayment
behaviour(Hossler, Braxton, & Coopersmith, 1989). It is a frequent practice to group variables
into different life stages to study student behaviour (Astin, 1977). The variables have been
taken from borrowers of the bank in the last few years and have been collated from the
physical loan application forms and other related documents.

The variables have been grouped as follows:

Category Variable Type of Data

Gender Nominal

Age Scale

Prior Work Experience Nominal

Applicant
Characteristics
Year of Applying Nominal

Payment Behavior Nominal

NPA Status Nominal

Stream of Education Nominal

Education Program
Characteristics
Cost of the Program Scale

52
 College Profile Nominal

Country of College Nominal

Family Income Scale

Family Total Income (including Guarantors) Scale

Characteristics

Parents Highest Educational

Nominal
Attainment

Branch Nominal

Loan Amount Scale

Loan Related Loan Interest Bracket Nominal

Secured or Unsecured Loan Nominal

Repayment Period Nominal

3.5 OBJECTIVES

The major objectives of the study are as listed below:

• To analyze the educational loan portfolio of Vijaya Bank from 2002 to 2012
and develop a demographic profile of the borrowers

• To study the educational loan growth rate and default rate of Vijaya Bank in
the past five years.

• To analyze the educational loan portfolio of Vijaya Bank, Bengaluru region in

53
 terms of the educational loan outstanding and NPA rates

• To analyze the relationships between Applicant, Education Program, Family,

Loan characteristics and Loan repayment behavior.

• To identify the key characteristics of the educational loan Defaulters.

3.6 HYPOTHESES FRAMED & TESTED

To meet the objectives as stated in Section 3.5 above, a combination of descriptive and
inferential statistics have been used in the study. In inferential statistics, the following
Hypotheses were framed and tested.

• There is no difference between the bad debt amounts (NPA) and the cost of the
education course chosen by the borrower.

• There is no difference on the payment behaviour of the borrowers based on

their parent’s employment type

• There is no relationship between gender and their repayment behaviour.

• There is no relation between school scholastic achievement and choice of study

in higher education

In addition to the above hypotheses, regression analysis has been run individually, to find
relationships, if any, between defaulting behavior and Applicant, Parental, Educational and
Loan characteristics. Finally a regression analysis has been done to find relationships if any,
between defaulting behaviour and all the characteristics.

3.7 SAMPLING PROCEDURE

The sample consists of secondary data taken from the loan applications from three
branches of Vijaya bank in Bengaluru.

Vijaya Bank has instituted a specialised loan processing centre for credit evaluation of
applications above Rs 750,000. As per the model scheme launched by Indian Banks
Association, educational loans above Rs 750,000 need to be secured with assets, usually
immovable property. Since processing of these applications is complex due to requirements of

54
legal opinions and surveys to identify the worth of the collaterals provided, the bank has setup
a specialised cell (RACPC) to handle these loans. The customer processing is handled by the
respective branches with the branches sending the documents to the RACP centre for loan
approval. The account setup and disbursement is handled by the branch post approval.

Table 3.1 Sample Distribution

Process Centre Amount Sample Size

RACPC
Above Rs. 750,000 91
(Bengaluru North)

Indiranagar Branch Below Rs. 750,000 74

Hanumanth Nagar
Below Rs. 750,000 69
Branch

The investigator has also conducted interviews with the bank managers and officials
from the head office to ascertain the main issues and challenges with the loan program. The
study is restricted to student borrowers, from the Bengaluru region. The RACPC and branches
primarily cater to borrowers from their locality and therefore the sample consists of students
resident in Bengaluru city at the time of application.

The selection of the branches has been random, with one branch from the east part of
Bengaluru (Indiranagar) and the other from the west of Bengaluru (Hanumanth Nagar). Both
the branches are residential branches located in typical representative residential localities in
Bengaluru.

To ensure loan repayment behaviour is available, loans disbursed before 2012 have
been taken. The average time for completion of the education course is between 2 to 3 years.
Hence some of these loans disbursed in the latter half of 2012 might not have matured in
terms of their repayment cycle.

3.8 RELIABILITY

The study has used only secondary data. The data has been collated from the physical loan
application forms and documents submitted by the applicant at the time of seeking the loan.
55
The bank requires extensive documentary proof and supporting documents along with the
application form. The applicants have to submit:

i. completed application form

ii. address proof

iii. identity proof

iv. Income documents (salary certificates/ income tax returns etc.)

v. proof of ownership in case of immovable property being provided as security

vi. letter of admission from the college

In addition, the bank also uses the credit bureau (CIBIL) reports to check the amount
of debt and delinquent behaviour of the applicant, if any. The reliability of the data is
established through documentary evidence and verifications undertaken by the banks.

Therefore standard reliability tests like test-retest reliability, inter-observer reliability,

parallel forms reliability and internal consistency reliability have not been checked.
Cronbach's alpha is usually used in psychometric tests to test the validity of multiple
measures, which measure the same construct. In this study due to the data being factual and
verified, Cronbach's alpha has not been checked.

3.9 VALIDITY

Test validity is a measure used to test if the data purports to measure what it is
intended to do. It is defined as “the degree to which the evidence and theory support the
interpretations of test scores entailed by proposed uses of tests”. However, as this study does
not measure any opinion based data, validity tests have not been conducted. The research
methodology has not used any constructs to test the hypothesis. Therefore content, criterion
and construct validities have not been checked.

3.10 CONCLUSION

The chapter described the methodology adopted in conducting the study. Due to the
secondary nature of the data, the reliability of the data did not need to be tested. All available
demographic data has been taken from the application forms and other documents available

56
with the bank. Post disbursement data, in the form of repayments has been taken from bank’s
monthly report as of 31 of January 2015, for the branches. The repayment data for the
RACPC accounts have been taken from the bank’s core banking system on the 15 of February
2015.

The dependencies and other relationships have been tested has then been analysed
using appropriate statistical techniques for description and inference. The details of the
statistical data analysis and interpretation are presented in the next chapter.

57
 RESULTS AND DISCUSSIONS

4.1 INTRODUCTION

The analysis and interpretation has been divided into two parts. The first part consists
of an analysis of the educational loan portfolio of the bank and its growth of the last five
years. The study also analyses the demographic characteristics of the sample of the banks
borrowers from the Bengaluru region. Descriptive measures have been used to depict the data.
Standard statistical measures like frequencies mean and standard deviations have been used in
the analysis. Standard deviation has also been used wherever applicable, to indicate the spread
of the data and to show the clustering of the data around the mean.

The second part analyses the dependencies that were tested in the sample. Various
relationships between the variables and the influences of these on other characteristics have
been hypothesized and tested using the relevant statistical tools.

4.2 DESCRIPTIVE ANALYSIS

The educational loan portfolio of the bank has seen significant growth year on year in
the last few years. The advances in this sector though classified as a part of priority sector
lending has been administered as a part of commercial retail lending schemes.

To ensure a structured analysis, the various factors affecting student financing has
been grouped based on the assumption that student demographic and family demographics in
addition to institutional choices influence behaviour. In addition, the research has also
collected loan related factors to identify if any dependencies exist between the loan
characteristics and subsequent behaviour. Grouping of variable is a standard practice in
similar studies (Astin, 1977, 1993; Pascarella & Terenzini, 1991; Flint, 1997). The variables
have been grouped into four categories, based on nature of the characteristics; (a) Applicant
Characteristics; (b) Education Program Characteristics; (c) Family Characteristics; (d) Loan
Related.

4.2.1 Vijaya Banks Education Loan Advances

The bank started providing loans to students after the launch of the scheme by Reserve
bank of India in 2001. However, similar to experience of other banks, the advances were
minimal in the early years. The advances started growing significantly in the last few years.

58
Table 4.1 shows the growth of the educational advances of the bank since 2010. Except for
2012 and 2013, the portfolio has witnessed double-digit growth. On average, the bank has
been disbursing 129 crores every year since 2010.

Table 4.1 Growth of Education Loan Portfolio

Amount % Number of Disbursements

Year
(in Crores) Growth Accounts (in Crores)

March 31st, 2010 534.5 24% 28,475 144

March 31st, 2011 603.0 13% 31,191 129

March 31st, 2012 643.0 7% 32,985 123

March 31st, 2013 671.0 4% 34,331 110

March 31st, 2014 760.0 13% 38,013 137

Source : Vijaya Bank Reports, All India

The number of loans granted has been growing every year. On average they have been
booking 2,385 loan accounts every year for the past 5 years. Their total number of customers
st
who have been granted loans stand at 38,013 as of 31 March 2014 (Figure 4.1). The growth
in advances has been increasing YoY on average at roughly a rate of 12%. Between end of the
financial year of March 2013 and March of 2014, the loan portfolio has grown by 13% from
Rupees 671 crores to Rupees 760 crores. Similarly the number of customer with loans has
increased from 34,331 as of March 2013 to 38,013 as of March 2014. The annual
disbursements in the financial year 2014, was Rupees 137 crores.

59
 38,013
39,000

37,000
34,331
35,000
32,985
33,000 31,191

31,000 28,475

29,000

27,000

25,000
March 31st, March 31st, March 31st, March 31st, March 31st,
2010 2011 2012 2013 2014

Source: Vijaya Bank Reports, All India

Figure 4.1 Growths of Education Loans (Accounts)

The bank has two regional offices in Bengaluru due to the size of the city. The Table
15, below shows the educational loans advanced by the two regions separately since 2010.
The advances of these two regions together, as a per cent of total All India advances stands
around 15% in terms of rupee value and 11% in terms of number of accounts. Bengaluru
North region has been more active in granting of educational loans with on average a 20%
more number of loans being granted.

The All India average loan amount is Rupees 1.99 lakhs whereas in Bengaluru it
stands at Rupees 2.77 Lakhs, indicating that the borrowers are on average availing higher
quantity of loan amount. There is a slight difference between the average values between
Bengaluru North and Bengaluru South; with Bengaluru South region have a slightly higher
average loan amount. The difference is Rupees 2,923 for the year ending March 2014, but this
is considered insignificant.

60
 Table 4.2 Education Loans (Bengaluru)

Bengaluru Bengaluru Bengaluru % to Total All-

North South Total India Advances
Year No Amt No Amt No Amt No Amt

Mar 2010 1,839 52 1,463 41 3,302 93 11.6% 17.5%

Mar 2011 2,020 57 1,588 47 3,608 103 11.6% 17.2%

Mar 2012 2,119 59 1,672 48 3,791 106 11.5% 16.5%

Mar 2013 2,121 57 1,724 49 3,845 106 11.2% 15.7%

Mar 2014 2,234 62 1,829 51 4,063 113 10.7% 14.9%

Source: Vijaya Bank, All Bengaluru

The loan advances in the Bengaluru region is on average around 11.3% of the total
educational loan advances in terms of number of accounts. However, due to the average loan
amount being granted in Bengaluru being higher than the all India loan ticket size, in amount
terms Bengaluru advances on average 16.4% of the total education loan advances.

The last few years has witnessed a tremendous growth in Vijaya Banks overall
advances, fuelled by an exponential growth in Retail loans. The bank in its goal of spreading
risks and earning higher yields has identified retail loans as a higher profitable sector. This has
led to change in focus towards personal and housing loans. This is reflected in the sluggish
growth of education loan advances in the year 2012 and 2013. We see in figure 4.2, the
growth in percentage of educational loan advances and the gross advances of Vijaya Bank
over the past five years. As can be seen the growth of their advances towards the education
loan sector has not been consistent. 2012 and 2013 saw a slowdown in the granting of credit
advances, but it has picked up in the year ending March 2014. They had a 24.07% in growth
between 2009 and 2010. It reduced to 12.80% during the year ending 2011.

61
 26%
24.07%

20.2%
21% 19.2%

16.9% 17.4% 16.9%

16%
12.80% 13.27%

11%

6.65%

6% Growth of Education Advances 4.35%

Growth of Gross Advances

1%
Mar 2010 Mar 2011 Mar 2012 Mar 2013 Mar 2014

Source: Vijaya Bank Reports, All India

Figure 4.2 Growths of Education Loans& Gross Advances

The bank has been managing the losses very prudently with adequate amount of
provisioning being carried in their books. Figure 4.3 shows the trends in the Gross and Net
Non-Performing assets in their books as a percentage to total assets.

3.5%

3.0%

2.5%

2.0%

1.5% Gross NPA

Net NPA
1.0%
Dec 2013 Mar 2014 Jun 2014 Sep 2014 Dec 2014

Source: Vijaya Bank Reports,

Figure 4.3 Trends in Gross & Net NPAs (All Advances)

The Net non-performing assets are within 2% for the last five quarters indicating that
the bank is booking higher quality credit and is also managing the collection aspects of the
credit. The non-performing assets in the Education sector stands at 36 crores as of December

62
2014. The gross advances to the educational sector as of December 31st, 2014 stood at
Rupees 874 crores.

The banks advance to the student and higher education is has a slightly higher gross
non-performance rate. However, the bank has been resorting to regular follow-up to ensure
defaults are reduced. High value cases are followed up with regular contacts with the
customers. In case these are ineffective, the banks resort to legal filing of claims. During
interviews with the bank managers, they informed that the legal process has been simplified
and made stronger with the introduction of the SARFAESI act of 2002 (Securitisation and
Reconstruction of Financial Assets and Enforcement of Security Interest Act).

Table 4.3 Gross Non-Performing Assets (Education Loans)

Number
Amount
Category of
(in Crores)
Accounts
Up to 10 Lakhs 65.95 5,098

Above 10 Lakhs 2.48 26

Total Education Loan Portfolio 760.31 38,013

Gross NPA in % 9.00% 13.48%

Source: Vijaya Bank Reports

The study has also analysed a sample of the branches and their educational loan
advances. The selection of the branches is random and covers a representative sample of all
types of branches in the Bengaluru region (residential, commercial, market areas, and institute
extension counters).

63
 Table 4.4 Sample Branches (Education Loans & NPA Rate)

Total Educational Total NPA

NPA Rate
Loans Portfolio Portfolio

Branch Number Number

Amount Amount
of of Accounts Amounts
(in (in
Accounts Accounts (in %) (in %)
Lakhs) Lakhs)
Trinity Circle 126 407.00 11 8.51 8.73% 2.09%

Jayanagar 74 327.00 12 25.83 16.22% 7.90%

Indirangar 74 364.00 5 3.95 6.76% 1.09%

Jalahalli 106 330.00 3 2.22 2.83% 0.67%

Peenya 90 172.00 3 1.88 3.33% 1.09%

Nagappa 38 148.00 2 4.75 5.26% 3.21%

Vijaynagar 123 384.00 18 29.80 14.63% 7.76%

Mudalapalya 46 128.00 8 9.63 17.39% 7.52%

MSR Technology
88 228.00 8 12.50 9.09% 5.48%
Institute

Hanumanthnagar 58 156.00 4 3.80 6.90% 2.44%

Malleshwaram 42 170.00 5 9.20 11.90% 5.41%

Rajajinagar 55 188.00 8 20.62 14.55% 10.97%

Sanjaynagar 39 169.00 3 3.73 7.69% 2.21%

Shantinagar 41 88.00 7 6.54 17.07% 7.43%

Ulsoor 47 160.00 4 4.39 8.51% 2.74%

Victoria Road 27 106.00 1 0.70 3.70% 0.66%

Whitefield 25 51.00 6 6.14 24.00% 12.04%

64
 Yelahanka 38 126.00 0 0.00 0.00% 0.00%

Yeshwanthpur 31 140.00 2 2.03 6.45% 1.45%

Cox Town 31 67.72 4 5.59 12.90% 8.25%

Domlur 42 158.00 8 5.77 19.05% 3.65%

Adugodi 41 78.56 8 14.88 19.51% 18.94%

Brigade Road 43 139.00 1 2.72 2.33% 1.96%

City Market 17 59.73 0 0.00 0.00% 0.00%

Dasarahalli 141 253.00 31 31.14 21.99% 12.31%

Gandhi Bazaar 56 250.00 10 18.06 17.86% 7.22%

HBR Layout 61 224.00 5 6.80 8.20% 3.04%

HSR Layout 35 124.00 4 6.28 11.43% 5.06%

K G Road 37 162.00 3 2.28 8.11% 1.41%

Infantry Road 24 99.02 0 0.00 0.00% 0.00%

K R Puram 100 226.00 5 4.94 5.00% 2.19%

Koramangala 41 193.00 5 10.65 12.20% 5.52%

Residency Road 33 175.00 3 3.63 9.09% 2.07%

Total / Average 1,870 6,051.03 197 268.96 10.08% 4.66%

Source: Vijaya Bank

In the Bengaluru regions, the highest advance is by the Trinity Circle branch (4.07
Crores), followed by Vijayanagar branch (3.84 Crores). In terms of bad loans, the highest
non-performing loans belong to Adugodi branch (18.94%) followed by Dasarahalli branch
(12.31%).

4.2.2 Applicant Characteristics

The demographics of the applicants show that most of them (72.6%) are regular in
their repayments. Only 8.1% of the number of borrowers in Bengaluru region in the sample is

65
in non-performing status. The rest are still in regular and doubtful status. The sample consists
of 67.5% males. Majority of the borrowers are in the age group of 21years to 25 years
(51.3%) and that only 10.3% of the applicants had prior work experience.

Table 4.5 Frequency Distribution (Applicant Characteristics)

Percent
Category Variable Values / Scale Frequency
to Total

2001 3 1.3%

2004 1 0.4%

2005 10 4.3%

2006 24 10.3%

2007 18 7.7%

Applicant Year of
2008 15 6.4%
Characteristics Application

2009 48 20.5%

2010 30 12.8%

2011 36 15.4%

2012 49 20.9%

Total 234

Regular 170 72.6%

Irregular 45 19.2%
Payment
Applicant
Behaviour of the No Payment 19 8.1%
Characteristics Borrowers

Total 234

Applicant NPA Status of 19 8.1%

Yes
Characteristics the Borrowers
No 215 91.9%

66
 Total 234

Male 158 67.5%

Applicant
Gender Female 76 32.5%
Characteristics

Total 234

20 & Below 86 36.8%

21 to 25 120 51.3%
Applicant Age of the

Characteristics Borrowers
More than 25 28 12.0%
Total 234

No 210 89.7%

Applicant Prior Work

Yes 24 10.3%
Characteristics Experience

Total 234

4.2.3 Family and Parent Characteristics

The analysis of the family background and the characteristics of the applicant’s
parents show that a majority of the borrowers come from a service or salaried families (65%).
Only 30.3% of the applicant’s parents are engaged in business. 3% of the applicants come
from an agricultural background based on the declaration of the income of their parents.
Majority of the students belong to the lower income category of annual income below Rupees
250,000 (35.9%). Twenty four students (10.3%) belong to very high income category with
annual family income above one million. One hundred and thirty students in the sample of
two hundred and thirty four belong to families where their parent’s highest educational
attainment is graduation. Only 22.6% of the students’ parents have completed post-
graduation.

67
 Table 4.6 Frequency Distribution (Family & Parental Characteristics)

Percent
Category Variable Values / Scale Frequency
to Total

Agriculture 7 3.0%

Business High Income 8 3.4%

Business Middle Income 24 10.3%

Business Low Income 36 15.4%

Homemaker 4 1.7%
Parents
Parent
Income Self Employed 3 1.3%
Characteristics
Levels
Service High Income 30 12.8%

Service Middle Income 62 26.5%

Service Low Income 54 23.1%

Service Pension
6 2.6%
Total 234

Up to 2.5 Lakhs 84 35.9%

Total Income
2.51 to 5 Lakhs 64 27.4%
Parent (including
Characteristics Guarantors if
5.01 to 10 Lakhs 62 26.5%
any)
> More than 10 Lakhs 24 10.3%

Graduate 138 59.0%

Parents Highest
Parent
Educational Under Graduate 43 18.4%
Characteristics
Attainment
Post Graduate 53 22.6%

4.2.4 Characteristics of the Education Program

The most popular course of study is engineering (29.1%) with science related subjects

68
coming second in the student’s choice of education. Computer science (19.2%) was coded as
a distinct subject, even though some of the courses are designed as engineering in computer
field. Science related subjects were the third most popular choice among the students availing
student loans. After STEM (Science, Technology, Engineering and Maths) based courses, the
next most popular choice was management, namely M.B.A degrees among the students with
33 students out of the sample of 234 choosing this stream in their higher education.

Table 4.7 Frequency Distribution (Education Program Characteristics)

Percent to
Category Variable Values / Scale Frequency Total

Arts 6 2.6%

Computer Science 45 19.2%

Engineering 68 29.1%

Education Stream Chosen Management 33 14.1%

Program for Higher

Characteristics Education Medical 20 8.5%

Science 45 19.2%

Others 17 7.3%

Total 234

1 to 3 Lakhs 47 20.1%

4 to 10 Lakhs 56 23.9%

11 to 20 Lakhs 40 17.1%

Education 21 to 30 Lakhs 32 13.7%

Cost of
Program
Education
Characteristics 31 to 40 Lakhs 35 15.0%

41 to 50 Lakhs 17 7.3%

More than 50 Lakhs 7 3.0%

Total 234

69
 Arts 3 1.3%

Business 27 11.5%

Engineering 103 44.0%

Education
Program College Profile General 69 29.5%
Characteristics
Law 2 0.9%

Medical 24 10.3%

Others 6 2.6%

Australia 10 4.3%

Canada 3 1.3%

France 5 2.1%

Education Germany 13 5.6%

Country of
Program
College
Characteristics India 131 56.0%

Sweden 1 0.4%

UK 9 3.8%

US 62 26.5%

44% of the students have enrolled in programs costing below 10 lakhs. This is
including the courses enrolled in Indian universities and colleges. 17.1% of the students have
enrolled in courses costing between 10 lakhs and 20 lakhs. 35.9% of the courses in the sample
cost above 20 lakhs. The profile of the colleges enrolled in shows similar characteristics as the
course selected. 44% of the colleges are oriented towards engineering programs and 69
(29.5%) of the colleges in the sample are general universities imparting all education in all
disciplines. Only 24 students (10.3%) of the borrowers are pursuing careers in medicine and
related fields. India, United States and United Kingdom are the top three choice of education
with 56%, 26.5% and 3.8% of the borrowers applying for studying in these countries
respectively.

70
4.2.5 Loan Characteristics

40% of the sample consisted of cases from the RACPC processing centre with 31.6%
from Indiranagar branch and the rest from Hanumanth Nagar branch. In monetary terms the
largest loan bracket is between 10 lakhs and 20 lakhs. The amount is the loan taken and does
not take into consideration the individuals own equity invested in his total cost of education.
59% of the loan availed is below 10 lakhs which includes the loans taken for studying in
India. The commercial nature of the education loan is reflected in 41.9% of the loan being
granted between 11% and 12/5 interest rate. The interest rates have been increasing in the last
few years and this also reflects the majority of loans granted is in the last few years. 54.3% of
the loans granted by the bank (in the sample) are against security, usually immovable property
provided by the borrowers. The total loan granted in the sample is about 23.15 crores and
therefore about 12.57 crores of the loans granted is backed by a security.

Table 4.8 Frequency Distribution (Loan Characteristics)

Percent to
Category Variable Values / Scale Frequency
Total

RACPC 91 38.9%

Hanumanth Nagar 69 29.5%

Loan Related Branch

Indiranagar 74 31.6%

Total 234

Upto 3 Lakhs 66 28.2%

3.01 to 10 Lakhs 72 30.8%

Loan Amount
Loan Related Taken by the 10.01 to 20 Lakhs 92 39.3%

Borrower
20.01 to 30 Lakhs 4 1.7%

Total 234

71
 10% to 10.99% 61 26.1%
Loan
Loan Related
Interest Slabs
11% to 11.99% 65 27.8%

12% to 12.99% 98 41.9%

13% to 13.99% 10 4.3%

No Collateral 127 54.3%

Collateralised
Loan Related Loan Property 107 45.7%

12 1 0.4%

60 144 61.5%

72 3 1.3%

EMI Period
Loan Related (in months) 84 4 1.7%

120 51 21.8%

180 31 13.2%

Total 234

The majority (61.5%) of the loans granted have a repayment term off 60 months.
However, during interviews with the bank officials it was clarified, that 5 years program was
the default standard program and many borrowers did not request for a longer duration due to
ignorance of other options. The borrower data shows variations in most of the variables. The
mean age of the applicants are 21.9 with an σ x of 3.270 indicating most of the students are
in the range of 18 to 24 years old.

Table 4.9 Range, Mean & Standard Deviation

Variable N Range Minimum Maximum Mean Std.

Deviation
Age 234 20.0 17.0 37.0 21.9 3.3

Cost of Education 234 64.0 1.0 65.0 18.2 16.0

Loan Amount (in Lakhs) 234 29.0 1.0 30.0 9.9 8.0

Interest Rate 234 3.5 10.0 13.5 11.7 0.9

Total Income (in Lakhs) 234 35.5 0.5 36.0 5.1 5.0

72
 The total cost of course these students are applying, shows great variation with the cost
ranging from one lakh to sixty-five lakhs. The mean is around 18.2 lakhs with a standard
deviation value of 16. The mean loan amount is 9.9 lakhs with the maximum loan value of 30
lakhs. This loan was granted on specific case with the borrower pledging substantial security.
The mean interest rate is 11.7% with the loans ranging from 10% to 13.5%.

4.3 HYPOTHESIS TESTING AND INTERPRETATION

4.3.1 NPA Amount and Total Cost of Education

The amount of loan availed is dependent on the total cost of the education. To
ascertain if there are any difference between the defaulting borrowers and the borrowers who
are paying regularly in terms of their total cost of education, a one-way ANOVA test was
performed. The total cost of education (IV) as declared by the applicants has a range of 64
Lakhs with a minimum of 1 Lakh and a maximum of 65 Lakhs in the sample. The mean of the
total cost of the education is 18.2 Lakhs with a standard deviation of 16 Lakhs.

ANOVA is used to compare differences of means between two or more groups.

Specifically, ANOVA compares the amount of variation between groups with the amount of
variation within groups. The test attempts to identify if any relationship exists between the
total cost of the education and the propensity of the student to default post completion of the
education.

Table 4.10 Cost of Education

Variable N Range Minimum Maximum Mean Std.

Deviation

Cost of Education 234 64.0 1.0 65.0 18.2 16.0

ANOVA tests are based on assumptions of the independence of observations. The

observations in the sample are all independent without any influence of any other member.
ANOVA tests also require that the dependent variables be normally distributed in each group.
To test the normality of the dependent variable, we used the Shapiro-Wilk’s Test of normality.
The dependent values were normally distributed with all the significance values above 0.05.

73
 The following null hypothesis is framed for the test:

H1: There is no difference between the bad debt amounts (NPA) and the cost of
the education course chosen by the borrower

The results of ANOVA test are presented in table 4.10. The results of the test reveal
that the amount of bad debts is highly dependent on the cost of the education. Higher the cost
of the education, the higher the loan amount availed by the borrower and therefore the
repayment amounts and the length of repayments would subsequently be higher. The results
indicate that in such a scenario the propensity for the borrower to default is higher. There was a
statistically significant effect on the amount of bad debts due to the cost of education at the
p<0.5 level for the three conditions [F (5, 13) = 9.590, p = 0.001].

Table 4.11 ANOVA - NPA Amount & Total Cost of Education

Sum of Squares df Mean Square F Sig.

Between Groups 294.354 5 58.871 9.590 .001

Within Groups 79.802 13 6.139

Total 374.156 18

From the results, the null hypothesis can be rejected and it can be concluded that the
cost of education program that the student enrols contributes to a difference on the amount of
the non-performing loan amount.

4.3.2 Parent’s Employment Type and Payment Behaviour

An ANOVA test was performed to compare the repayment behaviour of the borrower
in relation to their parent’s employment type. The parent’s employment type was grouped into
10 groups based on the declaration on the loan application form.

74
 Table 4.12 Employment Type

Employment Type Description

Agriculture Agricultural Income

Business High Income Annual Income > 12 Lakhs

Business Middle Income Annual Income > 3 Lakhs and < 12 Lakhs

Business Low Income Annual Income < 3 Lakhs

Homemaker Other Sources

Service High Income Annual Income >6 Lakhs

Service Middle Income Annual Income > 1.8 Lakhs and < 6 Lakhs

Service Low Income Annual Income < 1.8 Lakhs

Self Employed Other Source

Service Pension Pension

The test compared the differences in means among the various groups i.e. the type of
employment the parents of the borrowers were in and if the borrower was defaulting on
repayment of the loan. ANOVA is used to compare differences of means among more than two
groups by comparing the amount of variation between groups with the amount of variation
within groups.

The test was performed to identify any dependencies that may exist between parent’s
employment types (IV) on the borrower’s payment behaviour. The observation that students
from higher income families or from service oriented parental backgrounds will exhibit more
prudent repayment behaviour, is tested by framing and testing the null hypothesis below:

H2: There is no difference on the payment behaviour of the borrowers based on

their parent’s employment type.

75
 Table 4.13 ANOVA - Parents Employment Type & Payment Behaviour

Sum of Squares df Mean Square F Sig.

Between Groups 4.907 9 .545 1.409 .185

Within Groups 86.653 224 .387

Total 91.560 233

The results of ANOVA test are presented in table 4.11. The table reveals that the
repayment behaviour of the borrowers is not influenced by the parent’s employment type. The
nature of the employment of the parent has a statistically insignificant effect on the repayment
behaviour at the p<0.5 level, for the three conditions [F (9, 224) = 1.409, p = 0.185].

From the results, it can be concluded that parent’s employment type i.e. whether the
parent is engaged in business, service or is retired etc. makes no difference on the student’s
repayment behaviour and hence the null hypothesis can be accepted.

4.3.3 Gender and Payment Behaviour

During interviews with the bank officials, it was suggested that there were incidence of
higher defaults by females, due to marriage and unemployment out of choice and the
reluctance to repay due to under-utilisation of the educational attainment. To investigate the
dependence of payment behaviour on gender, cross tabulation and chi square test has been
done. Chi-square test is generally used to test if there are significant differences between
expected frequencies and the observed frequencies and requires that the data be ordinal
(categorical data). The Chi-square test does not give any indication about the strength of the
relationships and only confirms if a relationship exists or does not exist. The Chi-square test
will check if the expected frequencies of default behaviour match with the actual frequency of
female defaulters in the sample.

The following null hypothesis is framed for the test.

H3: There is no relationship between gender and their repayment behaviour.

76
 Table 4.14 Chi-Square Test - Gender & Payment Behaviour

Payment Behaviour

No
Crosstab Regular Irregular Total
Payment

Count 116 28 14 158

Gender Male
Expected Count 115 30 13 158

Count 54 17 5 76
Female
Expected Count 55 15 6 76

Count 170 45 19 234

Total
Expected Count 170 45 19 234

Asymp. Sig.
Chi-Square Tests
Value df (2-sided)

Pearson Chi-Square 0.945 2 .624

The results of the test are presented table 4.12. The relationship is found to be
2
insignificant at 5% level (λ =0.945 at p = 0.624). It can be inferred that the payment behaviour
is not dependent on the gender. It means men and women are the same in their tendency to pay
or not pay the loan obligations. From the cross tabulation, and a comparison of the actual
values and the expected values, it can be concluded that the sample does not show any
difference between genders and their likelihood of paying back their loans. Therefore, the null
hypothesis is accepted.

4.3.4 Applicants Marks and Choice of Study

To investigate if any relationships exist between students’ marks at end of their

secondary school and their subsequent choice of study (for e.g. Arts, Engineering etc.), cross
tabulation and chi square test has been performed. .

77
 Chi-square test is generally used to test if there are significant differences between
expected frequencies of student’s choice of topic of study and the observed frequencies of
actual choice of the number of students opting for that stream of study. Chi-square test requires
that the data be ordinal (categorical data) and our stream of study has been categorised into
seven different and distinct categories. All courses not falling in the previous six groups have
been grouped under ‘Others’. The Chi-square test is used to check if the expected frequencies
of the borrower’s defaulting behaviour match with the actual frequency of defaulters by
category in the sample. However the Chi-square test does not give any indication about the
strength of the relationships and only confirms if a relationship exists or does not exist. There
is a general perception and belief that academically inclined students prefer to pursue their
higher education in fields of science and engineering. To test that assumption, chi-square test
has been done and the following null hypothesis has been framed for the test.

H4: There is no relation between school scholastic achievement and choice of

study in higher education

78
 Table 4.15 Chi-Square Test - Applicants SSLC Marks & Choice of Study

Stream

Crosstabs Total
Computer
Arts Engg Mgt Medical Science Others
Science

Count 0 0 0 4 0 1 0 5
Less than
50 Expected
.1 1.0 1.5 .7 .4 1.0 .4 5
Count

Count 0 3 2 5 2 3 4 19
50.01 to
60 Expected
.5 3.7 5.5 2.7 1.6 3.7 1.4 19
Count

Count 2 5 13 1 1 5 3 30
60.01 to
70
Ma

Expected
rks

.8 5.8 8.7 4.2 2.6 5.8 2.2 30

Count
Applic
ants

Count 1 16 16 12 8 15 2 70
70.01 to
80 Expected
1.8 13.5 20.3 9.9 6.0 13.5 5.1 70
Count

Count 0 10 14 7 3 11 7 52
80.01 to
90 Expected
1.3 10.0 15.1 7.3 4.4 10.0 3.8 52
Count

Count 3 11 23 4 6 10 1 58
90.01 to
100 Expected
1.5 11.2 16.9 8.2 5.0 11.2 4.2 58
Count

Total Count 6 45 68 33 20 45 17 234

79
 Expected
6.0 45.0 68.0 33.0 20.0 45.0 17.0 234
Count

Asymp. Sig.
Chi-Square Tests
Value df (2-sided)

Pearson Chi-Square 55.173 30 .003

The results of the test are presented in the table 4.13. The relationship is found to be
2
significant at 5% level (λ =55.173 at p = 0.003). It can be inferred that the student’s marks at
the time of leaving the school strongly influences or is a factor in the student’s choice of
subject for higher education. Students with higher scholastic achievement in their school tend
to choose courses under science and engineering. At the top deciles between 90.01 and 100,
the expected count of students opting for pursuing higher studies in engineering 17 whereas
the actual count is 23 students.

The cross tabulation above shows the actual values and the expected values
indicating that students with higher scholastic achievement prefer to pursue their education
in science, computers and engineering. Therefore, the null hypothesis is rejected

4.3.5 Regression

Regression analysis is a statistical method for estimating the relationships among

variables. It is done to identify the relationship of independent variables (predictor variables)
on a dependent variable.

The main dependent variable analysed in the study is the Repayment status (NPA
Status) of the borrower. The research attempts to identify if it is possible to predict the
borrower’s future payments and his ability to continue repaying the loan till the entire loan is
repaid. To analyse and predict the borrower’s repayment behaviour the study has tested the
following variables namely Gender, Age, Work Experience, Total Income, Parents
Education Level, Stream, Total Cost of Education, College Profile, Country of College, Loan
Amount, Interest Rate, Collateral and the Tenor of the Loan.

80
 The regression test attempts to see if the above independent variables are able to
predict the likelihood of the borrower becoming a defaulter. The regression analysis was
done individually against each of the variables as per our grouping and finally with all the
variables.

4.3.5.1 REGRESSION – APPLICANT CHARACTERISTICS

The following independent variables namely Gender, Age and Prior Work Experience
were analysed to see their ability to predict the borrower’s repayment behaviour.

The measure of the strength of the computed equation is R-square, sometimes called
the coefficient of determination. R Square is simply the square of the multiple correlation
coefficients listed under R in the Summary table below, and represents the proportion of
variance accounted for in the dependent variable (NPA Status) by the predictor variables
(Gender, Age, Prior Work Experience). The multiple correlation coefficients (R) are 0.169,
and the R Square is 0.029. Therefore, the predictor variables explained only 2.9% of the
variance in the dependent variable of NPA status. The R value represents the simple
correlation between the dependant and the independent variable. It is 0.169, which indicates
very little correlation between these characteristics (Gender, Age and Work Experience) to the
borrower’s NPA status. The R Square value, which gives an indication of how much of the
variation in the dependant variable can be explained by the independent variable, is low at
0.029. This means that the very little variation in the NPA status is being explained by these
independent variables.

Table 4.16 Model Summary - Applicant Characteristics & NPA Status

Mode Adjusted Std. Error of

R R Square
l R Square the Estimate

1 .169 .029 .016 .272

ANOVA
Sum of
Model df Mean Square F Sig.
Squares

2.2
1 Regression .498 3 .166 .084
41

81
 Residual 16.953 229 .074

Total 17.451 232

Coefficients

Unstandardized Standardized
Coefficients Coefficients
Model t Sig.

B Std. Error Beta

(Constant) 2.081 .129 16.102 .000

Gender .021 .038 .036 .558 .577

1
Age -.002 .006 -.020 -.256 .798

Work
-.141 .070 -.154 -2.017 .045
Experience

The ANOVA table presents results from the test of the null hypothesis that R Square
value is zero. An R Square of zero indicates no linear relationship exist between the
independent and the dependent variable. The ANOVA table shows that the computed F
statistic is 2.241, with an observed significance level of 0.084. Thus, the hypothesis that there
is no linear relationship between the NPA status of the borrower and the borrower’s gender,
age or prior work experience is accepted. Therefore, it is concluded that the independent
variables (Gender, Age, and Prior Work Experience) cannot predict the future repayment
behaviour (NPA status) of the borrower. The results indicate that the regression model does
not predict the dependent variable significantly.

The Coefficients table shows the standardized Beta coefficient between the predictor
variables Gender, Age, Prior Work Experience and the dependent variables ‘NPA Status’. The
relationship is found to be statistically insignificant. The effect of gender is shown to be

82
positive but at a statistically insignificant level (Beta = 0.036, t = 0.558, p > 0.05). The effect
of age is negative but found to be statistically insignificant (Beta = -0.020, t = -0.256, p >
0.05). Similarly the predictor variable work experience has a negative effect on defaults.

The p value (p = 0.84) is above the accepted level of significance of 0.05 and hence it
is concluded that these independent variables (Gender, Age and Work Experience) do not
explain the dependent variable (NPA Default status) at statistically significant level.

4.3.5.2 REGRESSION – PARENT CHARACTERISTICS

The following independent variables grouped under the heading of borrower’s parents
and family characteristics namely, income of the parents and the highest education attainment
of borrower’s parents were analysed to see their ability to predict the borrower’s repayment
behaviour.

R Square, which is the square of the multiple correlation coefficients listed under R in
the Summary table below, and represents the proportion of variance accounted for in the
dependent variable (NPA Status) by the predictor variables (parents income and parents
education). The multiple correlations co-efficient R is 0.084, and the R Square is 0.007.
Therefore, parent’s income and education explained only 0.7% of the variance in the
dependent variable of NPA status. The R value of 0.084 indicates very little correlation
between these characteristics (parent’s characteristics) to the borrower’s NPA status. This
means that the very little variation in the NPA status is being explained by these independent
variables.

Table 4.17 Model Summary- Parent Characteristics & NPA Status

Adjusted Std. Error of the

Model R R Square
R Square Estimate

1 0.084 .007 -.001 .274

ANOVA
Sum of Mean
Model df F Sig.
Squares Square

Regression .124 2 .062 .826 0.439

1 Residual 17.333 231 .075

Total 17.457 233

83
 Coefficients
Standardized
Coefficients Coefficients
Model t Sig.
Std.
B Beta
Error

(Constant) 1.925 .041 46.639 .000

1 Total Income Lakhs .004 .004 .078 1.162 .246

Parents Education
-.017 .022 -.051 -.768 .443
Level

The ANOVA table presents results from the test of the null hypothesis that R Square
value is zero. An R Square of zero indicates no linear relationship exist between the predictor
variables of the parent’s income and education and the dependent variable (NPA status). The
ANOVA table shows that the computed F statistic is 0.826, with an observed significance
level of 0.439. Thus, the hypothesis that there is no linear relationship between the NPA status
of the borrower and the borrower’s parent’s education or the parent’s income level is
accepted. Therefore, it is concluded that the independent variables (parent’s income and
parent’s education) cannot predict the future repayment behaviour (NPA status) of the
borrower. The results indicate that the regression model does not predict the dependent
variable significantly. Their predictive ability was not found to be a significant.

The Coefficients table shows the standardized Beta coefficient between the predictor
variables of parent’s total income and their highest attainment in education level and the
dependent variables ‘NPA Status’. The relationship is not found to be statistically significant.
The effect of parents or family’s income is shown to be positive but at a statistically
insignificant level (Beta = 0.078, t = 1.162, p > 0.05). The other predictor variable grouped
under parental characteristics is the highest education attained by the borrower’s parents. The
effect of parent’s education is negative but found to be statistically insignificant (Beta = -
0.051, t = -0.768, p > 0.05).

The p value (p = 0.439) is above the accepted level of significance of 0.05 and hence it
is concluded that these dependant variables do not explain the default status at statistically
significant levels.

84
4.3.5.3 REGRESSION – EDUCATION CHARACTERISTICS

The following independent variables selected from the data and grouped under the
heading of education characteristics namely, stream of the education, the total cost of the
education, profile of the college and the country where the college is located were analysed to
see if these have any impact on the student’s future repayment behaviour.

The regression test was run using these four characteristics as independent variables
and the NPA status as the dependent variable. R Square, which is the square of the multiple
correlation coefficients, represents the proportion of variance accounted for in the dependent
variable (NPA Status) by the predictor variables. The multiple correlations co-efficient R is
0.126, and the R Square is 0.016. Therefore, the test shows that the student’s choice of the
stream, the cost of the course and the profile and country where the student did his education
has no impact on the behaviour of student in terms of repaying the loan. They only explain
1.6% of the variance in the dependant variable of NPA status.

The R value of 0.126 indicates very little correlation between these characteristics
(education characteristics) to the borrower’s NPA status. This means that the very little
variation in the NPA status is being explained by these independent variables.

Table 4.18 Model Summary - Education Characteristics & NPA Status

Adjusted R
Model R R Square Std. Error of the Estimate
Square

1 .126 .016 -.001 .274

ANOVA

Sum of
Model df Mean Square F Sig.
Squares

Regression .277 4 .069 .922 .452

1 Residual 17.181 229 .075

Total 17.457 233

85
 Coefficients

Unstandardized Standardized
Model
Coefficients Coefficients t Sig.

B Std. Error Beta

(Constant) 1.819 .084 21.756 .000

Stream .009 .011 .053 .770 .442

1 Total Cost of Education -.002 .001 -.100 -1.358 .176

College Profile .016 .015 .070 1.023 .308

Country of College .007 .011 .048 .659 .510

The results of the ANOVA table present the test of the null hypothesis that R Square
value is zero. An R Square indicates the linear relationship that exists between the predictor
and independent variables of Stream, Total Cost of Education, College Profile and the
Country of College and the dependent variable (NPA status). The ANOVA table shows that
the computed F statistic is 0.922, with an observed significance level of 0.452. Thus, the
hypothesis that there is no linear relationship between the NPA status of the borrower and the
borrower’s course characteristics is accepted. Therefore, it is concluded that these
independent variables cannot predict the future repayment behaviour (NPA status) of the
borrower.

The table showing coefficients of the equation shows the standardized Beta coefficient
between the predictor variables of grouped as educational characteristics namely the stream of
the course selected, the total cost of the course, the profile of the college and the country
where the college is located and its influence on the dependent variables ‘NPA Status’. The
relationship is not found to be statistically significant. The effect of the choice of the course is
shown to be positive but at a statistically insignificant level (Beta = 0.053, t = 0.770, p >
0.05). The total cost of the course is negative but found to be statistically insignificant (Beta =
-0.100, t = -1.358, p > 0.05). The profile of the college namely whether it is a Humanities or a
Science based college shows a positive effect but at statistically insignificant levels (Beta =
0.070, t = 1.023, p > 0.05). The last predictor variable grouped under the education

86
characteristics namely the Country where the college is located does not show a statistically
significant level of influence on the borrower’s defaulting behaviour (Beta = 0.048, t = 0.659,
p > 0.05).

The p value (p = 0.452) is above the accepted level of significance of 0.05 and hence it
is concluded that these dependant variables do not explain the default status at statistically
significant levels.

4.3.5.4 REGRESSION – LOAN CHARACTERISTICS

The loan characteristics namely, the amount of the loan, the interest rate of the loan,
whether security has been provided and the repayment period of the loan were tested to check
for their ability to predict future loan repayment behaviour of the student. The regression test
was run using these four characteristics as independent variables and the NPA status as the
dependent variable. The test shows an R value of 0.084. A measure of the strength of the
computed equation is R-square, sometimes called the coefficient of determination.

The square of the multiple correlation coefficients R Square indicates the proportion of
variance accounted for in the dependent variable (NPA Status) by the predictor variables (loan
characteristics). The multiple correlations co-efficient R is 0.084, and the R Square is 0.007.
Therefore, the test shows that the amount of the loan availed by the student, the Interest Rate
the borrowers has availed the loan at, collateral provided if any and the length of the loan are
not a very significant predictor of the repayment behaviour. These independent variables only
explain 0.7% of the variance in the dependant variable of NPA status.

Table 4.19 Model Summary - Loan Characteristics & NPA Status

Std. Error
Adjusted R
Model R R Square of the
Square
Estimate

1 0.084 .007 -.010 .275

ANOVA

Sum of Mean
Model df F Sig.
Squares Square

87
 Regression .123 4 .031 .406 0.804

1 Residual 17.334 229 .076

Total 17.457 233

Coefficients

Standardized

Unstandardized Coefficients
Coefficients
Model t Sig.

B Std. Error Beta

(Constant) 2.174 .262 8.284 .000

Loan Amount -.001 .005 -.015 -.092 .927

1 Interest Rate -.023 .024 -.075 -.927 .355

Collateral -.020 .083 -.037 -.245 .806

Tenor of the Loan .000 .001 .076 .778 .437

ANOVA tests the null hypothesis that R Square value is zero. An R Square indicates
the linear relationship that exists between the independent variables of loan characteristics
namely, the amount of the loan, the interest rate of the loan, and whether security has been
provided and the repayment period of the loan and the dependent variable (NPA status). The
ANOVA table shows that the computed F statistic is 0.406, with an observed significance
level of 0.804. Thus, the hypothesis that there is no linear relationship between the NPA status
of the borrower and the loan characteristics is accepted. It is concluded that these independent
variables cannot predict the future repayment behaviour of the borrower.

The table shows the Coefficients between the predictor variables of loan amount,
Interest rate of the loan, whether the loan has been availed against collateral and the length or
the tenor of the loan and its effect on the dependent variables ‘NPA Status’. As per the
guidelines issued by the reserve bank all loans above Rupees 750,000 needs to be granted
against suitable collateral. The relationship is not found to be statistically significant. The

88
effect of the loan amount is shown to be negative and at a statistically insignificant level (Beta
= -0.015, t = -0.092, p > 0.05). The effect of interest rate on the borrower’s behaviour is
shown to be negative but found to be statistically insignificant (Beta = -0.075, t = -0.927, p >
0.05). The effect of collateral or security provided on the borrower’s behaviour is shown to be
negative but found to be statistically insignificant (Beta = -0.037, t = -0.245, p > 0.05). The
effect of the length of the loan repayments or tenor of the loan or on the borrower’s behaviour
is found to be statistically insignificant (Beta = 0.076, t = 0.778, p > 0.05).

The p value (p = 0.804) is above the accepted level of significance of 0.05 and hence it
is concluded that the amount of the loan, interest rate, collateral and the repayment period of
the loan do not explain the default status at statistically significant levels.

4.3.5.5 REGRESSION – ALL CHARACTERISTICS

Finally to test if any of the independent variables had a significant ability to predict the
future NPA status or bad loan status, a regression was run using all the independent variables
against NPA status as the dependent variable. All the four different groupings were
simultaneously run namely the applicant’s characteristics, the applicant’s parents and family
characteristics, the course characteristics and the loan characteristics.
The test shows an R value of 0.256. The square of the multiple correlation coefficients
R Square indicates the proportion of variance accounted for in the dependent variable (NPA
Status) by the predictor variables (all characteristics). The multiple correlations co-efficient R
is 0.256 and the R Square is 0.066. These independent variables only explain 6.6% of the
variance in the dependant variable of NPA status. Therefore, the test shows that the
independent variables used in the test do not have a significant effect on the repayment status
of the borrower.
Table 4.20 Model Summary - All Characteristics & NPA Status

Adjusted R Std. Error of the

Model R R Square
Square Estimate

1 0.256 .066 .010 .273

ANOVA

Sum of
Model df Mean Square F Sig.
Squares

Regression 1.147 13 .088 1.186 0.291

1
Residual 16.303 219 .074

89
 Total 17.451 232

Coefficients

Unstandardized Standardized
Coefficients Coefficients
Model t Sig.
B Std. Error Beta

(Constant) 2.340 .335 6.991 .000

Gender .004 .042 .007 .097 .923

Age -.006 .007 -.074 -.888 .375

Work Experience -.142 .072 -.155 -1.978 .049

Total Income .005 .004 .093 1.182 .239

Parents Education -.004 .023 -.012 -.168 .867

Stream .011 .011 .065 .933 .352

1
Cost of Education -.003 .002 -.146 -1.223 .223

College Profile .019 .016 .086 1.186 .237

Country of College .005 .011 .031 .426 .671

Loan Amount .004 .007 .126 .633 .527

Repayment Period .001 .001 .138 1.399 .163

Interest Rate -.024 .026 -.079 -.914 .362

Collateral -.063 .089 -.115 -.714 .476

ANOVA tests the null hypothesis that R Square value is zero. The R Square value
indicates the linear relationship that exists between the independent (all characteristics) and
the dependent variable (NPA status). The ANOVA table shows that the computed F statistic is
1.186, with an observed significance level of 0.291. Thus, the hypothesis that there is no
predictive linear relationship between the NPA status of the borrower and the independent
characteristics is accepted. It is concluded that these independent variables do not predict the
future repayment behaviour of the borrower.

The table shows the Beta Coefficients of all the variables, which were grouped under
all the different groups (Applicant, Parental, Educational and Loan) and its effect on the
dependent variables ‘NPA Status’. The effect of the gender is shown to be positive but at a
statistically insignificant level (Beta = 0.007, t = 0.097, p > 0.05). The effect of the age is
shown to be negative but at a statistically insignificant level (Beta = -0.074, t = -0.888, p >

90
0.05). The effect of work experience is shown to be negative but at a statistically significant
level (Beta = -0.155, t = -1.978, p < 0.05). The effect of the parents total income is shown to
be positive but at a statistically insignificant level (Beta = 0.093, t = 1.182, p > 0.05). The
effect of the parents education is shown to be negative but at a statistically insignificant level
(Beta = -0.012, t = -0.168, p > 0.05). The effect of the course chosen by the student or stream
is shown to be positive but at a statistically insignificant level (Beta = 0.065, t = 0.933, p >
0.05). The effect of the total cost of the course chosen by the student is shown to be negative
but at a statistically insignificant level (Beta = -0.146, t = -1.233, p > 0.05). The effect of the
type of college chosen by the student is shown to be positive but at a statistically insignificant
level (Beta = 0.086, t = 1.186, p > 0.05). The effect of the location of the college (country) is
shown to be positive but at a statistically insignificant level (Beta = 0.031, t = 0.426, p >
0.05). The effect of the loan amount availed by the borrower is shown to be positive but at a
statistically insignificant level (Beta = 0.126, t = 0.633, p > 0.05). The effect of the length or
the tenor of the loan is shown to be positive but at a statistically insignificant level (Beta =
0.138, t = 1.399, p > 0.05). The effect of the interest rate of the loan is shown to be negative
but at a statistically insignificant level (Beta = -0.079, t = -0.914, p > 0.05). The effect of
collateral or security on the repayment behaviour is shown to be negative but at a statistically
insignificant level (Beta = -0.115, t = -0.714, p > 0.05).

The p value (p = 0.291) is above the accepted level of significance of 0.05 and hence it
is concluded that these independent variables (applicant’s characteristics, the applicant’s
parents and family characteristics, the course characteristics and the loan characteristics) do
not explain the default NPA status of the borrower’s at statistically significant levels. The
variables did not display statistically significant predictive capability of predicting defaults.

91
 SUMMARY AND CONCLUSION

5.1 INTRODUCTION

Summary and findings of the present study are detailed in this chapter. The chapter
explains the findings, interpretations of the study, implications, suggestions, recommendations
to the bank, limitations of the study and scope for further research.

There are has been a limited number of studies done in India to understand the
persistence of defaults among student borrowers. The chapter presents the results of the
research into finding correlations between the key attributes and demographics and the
students repayment behaviour.

5.2 FINDINGS

i. The study found that there is a direct correlation between the borrowers propensity
to default and the amount of the loan availed. Students who availed large loans
showed a higher likelihood of defaulting on their repayments. However, the study
also found that the amount of loan availed was dependent on the study program the
student had selected.
ii. The research attempted to identify relationships between parent’s employment
type and income to the students defaulting behaviour. The study analysed the
relationship between student’s repayment of the loan and the type of parent’s
employment (e.g. Business, Salaried, Homemaker etc.). It was found that the
parent’s employment has no bearing on the student’s behaviour towards his loan
repayments
iii. The study analysed if any relationship exist between the gender of the student and
the student’s repayments and NPA status. It was found that no such influences
exist and the likelihood of males and females defaulting is similar.
iv. Higher education helps in the developing the human capital of the country and
provides the competitive abilities that the globalised economy will require in the
future. India is poised to reap the benefits of the demographic advantages (large
number of young people joining the workforce) and cost advantages (lowered
production costs) over most of its rivals. However the burden of providing higher
education is being transferred or at the very least being shared between
government and the private sector. To provide funding for the deserving and

92
 meritorious student the Reserve Bank of India launched a student scheme in 2001.
The scheme is implemented through the commercial bank and has witnessed
significant growth over the last few years.
v. Vijaya Bank’s educational loan portfolio has touched Rupees 760 crores. They
have granted loans to 38,013 students till the end of March 2014. They have been
granting on average 2,385 loans every year for the past five years.
vi. The average loan amount granted by Vijaya Bank on an all India basis is Rupees
1.99 lakhs. In Bengaluru the average loan amount is Rupees 2.77 Lakhs, indicating
that the borrowers in Bengaluru are on average availing higher quantity of loan
amount. This could be due to the higher inspirational urban borrowers who are
applying for more expensive education programs. However this has not been
tested.
vii. Bengaluru region’s share in terms of number of loans granted, of the total
educational loan advances by the bank, is 11.3%. However the average loan
amount being granted in Bengaluru is higher than the all India average loan size.
Therefore in Rupee amount terms, Bengaluru region’s advances are on average
16.4% of the total education loan advances.
viii. The growth of the educational loan advances of the bank has not been consistent
over the past five years. The advances grew 24.07% in 2010 over 2009 but it
slowed down to 12.80% between 2010 and 2011. It further slowed down to 6.7%
and 4.3% in 2012 and 2013 respectively. The bank witnessed an upward swing in
growth to 13.3% between 2013 and 2014.
ix. The net non-performing assets of the bank in the Education sector stands at Rupees
36 crores as of December 2014. The advances to the educational sector as of
December 31st, 2014 stand at Rupees 874 Crores. The Net non-performing assets
of the bank’s advances (all advances) are within 2% for the last five quarters
indicating that the banks credit quality has been improving steadily. The bank has
developed an effective collection process with monthly reports being generated by
their specialized credit system.
x. The gross non-performing educational loan advances stands at 9% and amounts to
Rupees 68.43 crores. The number of loans in gross NPA status is about 5,124
accounts and constitutes 13.48% of the total number of educational loan advances.
xi. The analysis of a random selection of thirty branches belonging to the Bengaluru
regions shows the highest advance is by the Trinity Circle branch (4.07 Crores),

93
 followed by Vijayanagar branch (3.84 Crores). In terms of bad loans, the highest
non-performing loans belong to Adugodi branch (18.94%) followed by Dasarahalli
branch (12.31%). The average non-performing rate is 10.08% in terms of number
of accounts and 4.66% in terms of amounts in the selected branches.
xii. The cost of education program that the student is enrolling in has an effect on the
amount of the non-performing asset. The higher the cost of education, the higher
the loan amount taken and therefore higher the non-performance amount.
xiii. Relationships between student’s marks at the time of leaving secondary school
(SSLC) and their choice of educational stream (Arts, Engineering, Medicine etc.)
was tested in the study. The study found that students, who displayed higher
academic performance in their high school, tend to pursue science and engineering
subjects for their under-graduation and post-graduation.
xiv. The regression analysis attempting to identify the predictive abilities of the
student’s characteristics and their defaulting behaviour did not indicate any
relationship. The variables used in the study and grouped by four different
categories or groups like applicant related characteristics (gender, age, prior work
experience), Parental and family characteristics (parents income, parents highest
educational level), education program characteristics (stream, cost, college
orientation, country) and loan characteristics (amount, interest, collateralised
tenor) did not display capabilities in predicting defaults.

5.3 RECOMMENDATIONS

Vijaya Bank has a robust asset lending business. They have a wide network of
branches and regional offices. Overall, their NPA’s are comparable to other public sector
banks and their loan portfolio indicates similar levels of losses as other banks. They have very
strong risk management procedures and have computerised most of their operations.

However, like most other banks, Vijaya bank has been looking at educational loans as
a social development requirement, as controlled and directed by the government of India.
They have not looked at educational loans as a profitable line of business to be pursued on its
own merit. Due to the interest rate restrictions and the sensitivities involved in adopting
market rates of interest, the bank seems to be unable to maximize the profits from this
category of loans.

It is suggested that with better and timely monitoring reports and tools, the level of

94
losses in this category of loans can be reduced significantly. In addition it is recommended
that the interaction with the students and their parents be increased through mailers, written
communication or a monthly event at the branch. This will increase the participation of the
students, parents and the bank officials thereby creating a moral obligation for the students to
ensure prompt repayments. Therefore with better management of credit losses, the educational
loans can be a profitable at the same a socially rewarding enterprise for the bank.

The study also analysed the characteristics of the educational loan borrowers and their
defaulting behavioural patterns. Finding the key characteristics or attributes of the students,
which indicate a propensity to default on their loan repayments will assist the bank in
targeting the right kind of students to grant the loans. This will assist the financial and
banking sector in framing future education loan granting policies by identifying key and
significant borrower characteristics. The study used multiple characteristics as available in the
current process but could not find any significant predictive attributes.

Due to the inability of the current data (information) being captured by the banks to
predict loan behaviour, the banks should substantially augment the quality and quantity of
data being collected. Therefore it is recommended that the bank enhance the scope of the
information sought at the time of the application. The banks are currently following the
standard KYC norms as recommended by the regulators with little extra documentation. It is
suggested that the banks use additional tools like essays, psychological questionnaires etc. to
identify the sincerity, interest and affinity of the students to the course they are pursuing.

It is also recommended that the banks identify the compensations received by students
in various professions. The data will need to be refreshed yearly and there are specialised
agencies, which collate and provide this data. This will provide an indication to the repayment
capacity of the borrowers and the loans can be tailored accordingly. Based on this data the
loan amounts or the repayment schedule can be structured to ensure that the borrowers are not
put under duress due to unreasonable loan instalments, thereby causing them to avoid paying
the bank. There is significant evidence showing that once a student starts defaulting, it is very
difficult for the student to become solvent again in that debt.

The relationship between defaulting behaviour and the cost of the education is
significant. Therefore it is recommended that higher value loans should be evaluated with
more rigor and stringent lending criteria.

The bank currently uses some general websites to identify and rate the college that the
95
applicant is intending to join. However, this data is not being systematically captured and
utilised in the credit granting process. It is recommended that the bank subscribe to a
professional college ranking services which provide a comprehensive and exhaustive data on
most of the colleges and universities. It is also recommended that the banks develop a metric
internally, to identify if there are any particular college or region contributing to higher losses.
It has been found in previous studies that the cultural and social conditions of certain colleges
and universities, sometimes located in a particular region encourage defaulting behaviour
among their students.

5.4 LIMITATIONS OF THE STUDY

i. The study has been conducted on a small sample of borrowers due to data
availability limitations. Therefore the findings may not hold true for the entire
educational loan advances of Vijaya Bank.
ii. The geographical coverage is limited to Bengaluru region and the findings might
not reflect the all India education loan advances. There is some evidence to
indicate that the average loan size being granted in Bengaluru is higher than the
average all India loan size. Therefore by association, other demographics factors
might be different in other regions.
iii. Findings cannot be generalized across other bank institutions and their educational
loan products due to differences in the product characteristics and credit policies.
Results are only indicative and not conclusive.

5.5 SCOPE FOR FURTHER RESEARCH

i. The study can further expanded to other geographies like north, west, east and
south. The bank has twenty-seven regional offices and the study can be expanded
into other regions.
ii. The scope of the study can be enlarged, by collecting primary data from the
student borrowers through a structured questionnaire. The questionnaire can be
targeted to customers who are regular in their repayments and defaulters to identify
the primary reasons they are not able to keep their loan commitments.
iii. The demographic data used to analyse family and parental characteristics and its
influence on student’s repayment behaviour is limited (parental income, parental
education). Multiple studies have identified significant influences based on the
family structure and circumstances (Dynarski, 1994; Volkwein&Szelest, 1995;

96
 Woo, 2002). Further research can be undertaken to identify other familial and
parental factors that may influence student behaviour towards repayment of their
debts.
iv. Institutional characteristics have not been studied, though there is some evidence
that institution investment and faculty support is associated with a decrease in
likelihood of default (Volkwein & Szelest, 1995). The scope of the study can be
extended to including institution variables like nature of the institution, average
fees for the course, faculty to student ratio, graduation versus post-graduation
etc.(Pinto & Mansfield, 2006).
v. Students and their family’s debt burden have not been used as a variable in the
study. This information was not populated correctly in many cases and was
missing in most cases. However some research (Choy & Li, 2006; Dynarski, 1994;
Lochner & Monghe - Naranjo, 2004) suggests that the debt burden is an indicator
future default (higher the debt burden, the higher the likelihood of default). This
can be included in the scope of the study.
vi. Influence of friends and peer group has not been studied. Cultural and social
beliefs and practices can influence the attitude of students towards debt. This can
be included in future studies.
vii. Program of study has not been analysed in depth. Post-graduation earnings and its
influence on student behaviour have not been analysed. Harrast (2004) had found
that certain student streams tend to show a higher incidence of defaults, though the
reasons are unclear. There is considerable evidence to suggest that the choice of
study affects future earnings (Flint, 1997; Herr & Burt, 2005; Steiner & Teszler,
2005; Volkwein & Szelest, 1995). Future studies can include additional variables
associated with the stream of study.
viii. Employment prospects and earnings, post completion of the education, has not
been analyzed. It is logically expected that higher earnings would prompt better
repayment behavior. This can be tested empirically.
ix. The relationship between the quality and rankings of the College or University to
borrower’s behavior has not been analyzed. This can be undertaken to identify
relationships if they exist.

97
5.6 CONCLUSION

Sufficient evidence exists to show that higher education increases an individual’s

income and the country’s human capital (W. Schultz, 1961). But who should bear the cost is
in dispute. Critics of state involvement in higher education financing, object on grounds of
using public money for private good (Eckwert & Zilcha, 2012; Tilak, 1993). Among various
models suggested, India has decided to adopt a model of financing which can be termed as
semi-private i.e. a mix of government intervention along with student’s own contribution. To
this extent the government of India has been pushing the banking sector in India to extend
credit to students on commercial basis though with a multitude of hidden subsidies (softer
interest rates, repayment moratorium, inclusion in priority sector lending etc.).

This has resulted in an exponential growth in the educational loan portfolio of the
banks. This growth has been accompanied by increasing costs and non-performing assets
(credit losses). It is not clear if this kind of mandated granting of educational loans is
sustainable for the banks. Government protectionism has been reducing over the years causing
commercial banks to manage their business on market driven economics.

Therefore if the social objectives have to be continued on a sustainable basis, it is

imperative that commercial banks be allowed to manage their educational loan advances
independently, based on their costs and profit making objectives.

One element of these costs is the credit losses such advances have to contend with.
The propensity to default seems to be a complex behaviour and is dependent on much wider
set of an individual’s educational, social, peers and family background.

Therefore any model or framework which reduces credit costs by helping lenders
make more informed choices in their credit will help in making this socially important lending
more sustainable, thereby reducing pressure on public funds.

Vijaya bank has been looking at educational loans as a social service as directed by the
central government and finance ministry. The bank’s approach to educational loans has been
more as a requirement to fulfil priority sector lending norms than as to a separate profitable
line of business. They view is as a social development requirement as directed by the
government. Unfortunately this situation prevails in most of the banks, mainly due to social
development agendas, public relations issues and restrictions on interest rates.

98
 It is unlikely that there is going to be any change in the political viewpoint in India
regarding educational loans due to its popular social message. Therefore, as recommended in
the section above, the banks needs to look at educational loans as a profitable line of business
to be pursued on its own merit within the market constraints.

The standard demographical attributes used in loan processing by Public Sector banks
is inadequate in identifying future defaulting behaviour. A substantially augmented data
collection process and verification would be a good beginning in improving loan NPA rates.

The current application and data gathering mechanism used for the credit evaluation is
inadequate and does not provide any clear predictive abilities to identify problem borrowers.
The credit evaluation process will need to be enhanced to include student’s intentions,
sincerity to the college. It should also capture the rankings of the college, which will indicate
the employability of the student once he passes out.

The bank can tailor the educational loans to the individual borrowers, using these
additional evaluations tools, as long as they stay within the guidelines issued by the ministry.
With these enhancements, the bank might be able to reduce the losses from this category thus
ensuring higher returns and profitability of the bank.

99
 Appendix I - Model Education Loan Scheme

(RPCD.PLNFS.BC.NO.83/06.12.05/2000-01, 2001)

RPCD.PLNFS.BC.NO.83/06.12.05/2000-01 April 28, 2001

The Chairman / Managing Director All Scheduled Commercial Banks

Dear Sir

Educational Loan Scheme

1. The Finance Minister in a meeting with the Chief Executives of the public sector banks on
13 June 2000 had highlighted the role of commercial banks in facilitating pursuit of higher
education by poor, but meritorious students. In pursuance thereof the Indian Banks’
Association constituted a Study Group under the chairmanship of Shri. R. J. Kamath,
Chairman and Managing Director of Canara Bank to examine the issue in detail. Based on the
recommendations of the Study Group, a comprehensive model educational loan scheme was
prepared by the Indian Banks’ Association for adoption by all banks. The Scheme aims at
providing financial support from the banking system to deserving/meritorious students for
pursuing higher education in India and abroad. The scheme was announced in the Union
Budget for 2001-2002 and discussed in the meeting the Finance Minister had with the Chief
Executives of banks on 7 April 2001.

2. Government of India, Ministry of Finance, Department of Economic Affairs [Banking

Division] has considered and decided to accept the Model Scheme prepared by IBA for
implementation, subject to the following modifications:

(i) The condition of minimum qualifying marks in the last examination may be
dropped.
100
 (ii) No margin may be insisted upon for loans up to Rs.4 lakh. However, for loans of
higher amounts, the margin requirement may be 5% for inland studies and 15% for 
studies
abroad.

(iii) No security may be insisted upon for loans up to Rs.4 lakh. However, for loans
above 
this amount, collateral security of suitable value or co-obligation of parents /
guardians / third party along with the assignment of future income of the student for payment
of instalments may be obtained.

(iv) Loan up to Rs.4 lakh may be advanced at interest rate not exceeding PLR of
the bank. Above Rs.4 lakh, the interest rate may be PLR + 1%.

3. We accordingly, forward herewith a copy of the model scheme prepared by IBA for
implementation by banks after effecting the modifications indicated at [i] to [iv] of para 2
above, at the earliest so that its benefits are available to students from this academic session
itself.

4. It is clarified that this Scheme is separate and in addition to and not in super cession of the
scheme earlier circulated by RBI under Supreme Court orders vide our circular
st
RPCD.SP.BC.10/09.07.01/99-2000 dated 31 July 1999 issued to public sector banks.

5. Please acknowledge receipt.

101
 A MODEL EDUCATIONAL LOAN SCHEME

1. INTRODUCTION:

Education is central to the Human Resources Development and empowerment in any country.
National and State level policies are framed to ensure that this basic need of the population is
met through appropriate public and private sector initiatives. While government endeavors to
provide primary education to all on a universal basis, higher education is progressively
moving into the domain of private sector. With a gradual reduction in government subsidies
higher education is getting more and more costly and hence the need for institutional funding
in this area.

The scope of education has widened both in India and abroad covering new courses in
diversified areas. Development of human capital is a national priority and it should be the
endeavour of all that no deserving student is denied opportunity to pursue higher education for
want of financial support. Loans for education should be seen as an investment for economic
development and prosperity. Knowledge and information would be the driving force for
economic growth in the coming years.

The Honourable Finance Minister in a meeting with the Chief Executives of the Public Sector
Banks on 13th June 2000 had highlighted the role of commercial banks in facilitating pursuit
of higher education by poor, but meritorious students. He also expressed the need to have a
comprehensive educational loan scheme prepared that could be adopted by all banks.
Accordingly, a study group under the Chairmanship of Shri R J Kamath, Chairman and
Managing Director, Canara Bank was constituted to examine the issue in detail. This model
scheme has been prepared based on the recommendations contained in the report submitted by
the group in August 2000.

2. OBJECTIVES OF THE SCHEME:

The Educational Loan Scheme outlined below aims at providing financial support from the
banking system to deserving / meritorious students for pursuing higher education in India and
abroad. The main emphasis is that every meritorious student though poor is provided with an
opportunity to pursue education with the financial support from the banking system with
affordable terms and conditions. No deserving student is denied an opportunity to pursue
higher education for want of financial support.

102
In short, the scheme aims at providing financial assistance on reasonable terms:

 To the poor and needy to undertake basic education.


 To the meritorious students to pursue higher/ professional/ technical education.

3. APPLICABILITY OF THE SCHEME:

The scheme detailed below could be adopted by all Commercial Banks. The scheme provides
broad guidelines to the banks for operationalizing the educational loan scheme and the
implementing bank will have the discretion to make changes suiting to the convenience of the
students/ parents to make it more customer friendly.

The scheme details are as under:

4. ELIGIBILITY CRITERIA:

4.1 Courses eligible


a. Studies in India:

 School education including plus 2 stages.


 Graduation courses: BA, BCom., BSc., etc.

 Post-Graduation courses: Masters & PhD.

 Professional courses: Engineering, Medical, Agriculture, Veterinary, Law,
Dental, Management, Computer etc.

 Computer certificate courses of reputed institutes accredited to Dept. of
Electronics or institutes affiliated to university.

 Courses like ICWA, CA, CFA etc.

 Courses conducted by IIM, IIT, IISc, XLRI, NIFT etc.

 Courses offered in India by reputed foreign universities.

 Evening courses of approved institutes.

 Other courses leading to diploma / degree etc. conducted by colleges/
universities approved by UGC / Govt. / AICTE / AIBMS / ICMR etc.

 Courses offered by National Institutes and other reputed private institutions.
Banks may have the system of appraising other institution courses depending
on future prospects/ recognition by user institutions.


 b. Studies abroad:-
 Graduation: For job oriented professional / technical

103
 courses offered by reputed universities.
 Post-graduation: MCA, MBA, MS, etc.

 Courses conducted by CIMA- London, CPA in USA etc.

. 4.2 Student eligibility:

 Should be an Indian National


 Secured admission to professional / technical courses through Entrance Test /
Selection process.

 Secured admission to foreign university / Institutions.

 Should have scored minimum 60% (50% for SC / STs) in the qualifying
examination for admission to graduation courses.

. 4.3 Expenses considered for loan:

 Fee payable to college / school / hostel.


 Examination / Library / Laboratory fee.

 Purchase of books / equipment / instruments / uniforms.

 Caution deposit / building fund / refundable deposit supported by Institution
bills / receipts.

 Travel expenses / passage money for studies abroad.

 Purchase of computers - essential for completion of the course.

 Any other expense required to complete the course - like study tours, project
work, thesis, etc.

5. QUANTUM OF FINANCE:

Need based finance subject to repaying capacity of the parents/ students with margin and
the following ceilings.

 Studies in India - Maximum Rs.7.50 lakhs.


 Studies abroad - Maximum Rs.15 lakhs



6. MARGIN:

Up to Rs.2 lakhs: Nil

Above Rs. 2 lakhs: Studies in India: 15%

104
 Studies Abroad: 25%

 Scholarship/ assistantship to be included in margin.


 Margin may be brought-in on year-to-year basis as and when disbursements are
made on a pro-rata basis.

7. SECURITY:

Up to Rs.2 lakhs: No security:

Above Rs.2 lakhs: Collateral security equal to 100% of the loan amount or guarantee of third
person known to bank for 100% of the loan amount.

Note:-
 The document should be executed by the student and the parent / guardian.

 The security can be in the form of land / building / Govt. Securities / Public Sector
Bonds / Units of UTI, NSC, KVP, LIC policy, gold, shares / debentures, bank deposit
in the name of student / parent / guardian or any other third party with suitable margin.

 Wherever the land / building is already mortgaged, the unencumbered portion can be
taken as security on II charge basis provided it covers the required loan amount.


 In case the loan is given for purchase of computer the same to be hypothecated to the
Bank.

Banks who wish to support highly meritorious / deserving students without security may
delegate such powers to a fairly higher-level authority.

8. RATE OF INTEREST:

Up to Rs.2 lakhs: PLR

Above Rs.2 lakhs: PLR + 1%

 The interest to be debited quarterly/ half yearly on simple basis during the Repayment
holiday / Moratorium period.


 Penal interest @ 2% be charged for above Rs.2 lacs for the overdue amount and
overdue period.

105
9. SANCTION / DISBURSEMENT:

 The loan to be sanctioned as per delegation of powers preferably by the Branch nearest
to the place of domicile. 


 No application for educational loan received should be rejected without the
concurrence of the next higher authority.

 The loan to be disbursed in stages as per the requirement / demand directly to the
Institutions / Vendors of books / equipment’s / instruments to the extent possible.

10. REPAYMENT:

Repayment holiday / Moratorium: Course period + 1 year or 6 months after getting

job, whichever is earlier.

The loan is to be repaid in 5-7 years after commencement of repayment. If the student is not
able to complete the course within the scheduled time extension of time for completion of
course may be permitted for a maximum period of 2 years. If the student is not able to
complete the course for reasons beyond his control, sanctioning authority may at his
discretion consider such extensions as may be deemed necessary to complete the course.

 The accrued interest during the repayment holiday period to be added to the principal
and repayment in Equated Monthly Instalments (EMI) fixed.


 1-2% interest concession may be provided for loaners if the interest is serviced during
the study period when repayment holiday is specified for interest / repayment under
the scheme.

11. FOLLOW UP:

Banks to contact college / university authorities to send the progress report at regular
intervals in respect of students who have availed loans.

12. PROCESSING CHARGES:

No processing / upfront charges may be collected on educational loans.

13. CAPABILITY CERTIFICATE:

Banks can also issue the capability certificate for students going abroad for higher studies. For
this financial and other supporting documents may be obtained from applicant, if required.

106
(Some of the foreign universities require the students to submit a certificate from their bankers
about the sponsors’ solvency / financial capability, with a view to ensure that the sponsors of
the students going abroad for higher studies are capable of meeting the expenses till
completion of studies.)

14. OTHER CONDITIONS:

No due certificate need not be insisted upon as a pre-condition for considering educational
loan. However, banks may obtain a declaration / an affidavit confirming that no loans are
availed from other banks.

Loan applications have to be disposed of within a period of 15 days to 1 month, but not
exceeding the time norms stipulated for disposing of loan applications under priority sector
lending.

In order to bring flexibility in terms like eligibility, margin, security norms, banks may
consider relaxation in the norms on a case-to-case basis delegating the powers to a fairly
higher level authority.

107
Appendix II- Loan Application Form

108
109
110
111
112
113
114
115
116
 Literatures Cited
Arrow, K. J. (1973). Higher Education as a Filter. Journal of Public Economics .

Astin, A. W. (1977). Four Critical Years. Effects of College on Beliefs, Attitudes, and
Knowledge. San Francisco: Jossey-Bass Publishers.

Baird, L. L. (1990, June). The Undergraduate Experience: Commonalities and Differences

among Colleges. Research in Higher Education .

Basu, P. (2006). Improving Access to Finance for India’s Rural Poor. Washington: World
Bank.

Becker, G. S. (1962 October). Investment in Human Capital: A Theoretical Analysis.

Journal of Political Economy , 9-49.

Becker, G. S. (1993). Nobel Lecture: The Economic Way of Looking at Behaviour.

Journal of Political Economy , 385-409.

Beckhard, R., & Pritchard , W. (1992). Changing the Essence : The Art of Creating and
Leading Fundamental Change in Organizations. San Francisco: Jossey-Bass.

Bilton, C., & Cummings , S. (2010). Creative strategy: Reconnecting Business and
Innovation. Chichester, U.K: Wiley-Blackwell.

Birdsall, N., & Londoño, J. (1997). Asset Inequality Does Matter: Lessons from Latin
America. Inter-American Development Bank. OCE Working Paper.

Blau, P. M., & Duncan, O. D. (1967). The American Occupational Structure. New York: John
Wiley & Sons.
Brown, M., & Zehnder, C. (2007 December). Credit Reporting, Relationship Banking and
Loan Repayment. Journal of Money, Credit and Banking , 1883-1918.
Brugel, J. F., Johnson, G. P., & Leslie, L. L. (1977). The Demand for Student Loans in Higher
Education : A Study of Preferences and Attitudes. Research in Higher Education .
Cabrera, A. F., Nora, A., & Castaneda, M. B. (1992). The Role of Finances in the Persistence
Process: A Structural Model. Research in Higher Education .
Calero, J. (1998). Quasi-Market Reforms and Equity in the Financing of Higher Education.
European Journal of Education .
Carnoy, M., & Lobo, J. (1979). Can Educational Policy Equalise Income Distribution in Latin
America?. Saxon House .
117
Chakrabarty, D. (2011, October). Financial Inclusion : A Road India Needs to Travel.
Livemint .
Chakrabarty, D. (2013). Financial Inclusion in India: Journey So Far And Way Forward.
Finance Inclusion Conclave. New Delhi.
Choy, S. P., & Li, X. (2006). Dealing With Debt, 1992-93 Bachelor's Degree Recipients 10
Years Later. National Center for Education Statistics. Washington: U.S Department of
Education.
Collis, D. J. (2014). International Strategy: Context, Concepts and Implications.
Chichester, UK: John Wiley & Sons.
Commission on Growth and Development . (2008). Growth Report: Strategies for Sustained
Growth and Inclusive Development. World Bank.
Cooper, A., Alvarez, S., Carrera , A., Mesquita, L., & Vassolo, R. (Eds.). (2006).
Entrepreneurial Strategies: New Technologies in Emerging Markets. Wiley-Blackwell.
Dev, S. (2006). Financial Inclusion : Issues and Challenges. Economic and Political Weekly,
41 (41), 4310-4313.
Dewan, A., Goel, R., & Malhotra, R. (2013). Student's Perception regarding Student
Financing . International Journal of Social Sciences & Interdisciplinary Research .
Dollar, D., & Kraay , A. (2002). Growth is Good for the Poor. 7 (3), 195-225.
Dynarski, M. (1994). Who Defaults on Student Loans? Findings from the National
Postsecondary Student Aid Study. Economics of Education Review , 55-68.
Eckwert, B., & Zilcha, I. (2012 January). Private Investment in Higher Education: Comparing
Alternative Funding Schemes. Economica .
Flint, T. A. (1997 June). Predicting Student Loan Defaults. The Journal of Higher Education ,
322-354.
Gaikwad, B. R., & Solunke, R. S. (2013). Growth of Higher Edcaution in India.
International Research Journal of Social Sciences .
Gandhar, H. (2010). Educational Loan Scheme of Scheduled Commercial Banks in India : An
Assessment. International Journal of Business Economics and Management Research , 64-
94.
Gladieux, L., & Perna, L. (2005). Borrowers Who Drop Out A Neglected Aspect of the
College Student Loan Trend. San Jose: The National Center for Public Policy and Higher
Education.

118
Goodwin, D. (1991). Beyond Defaults: Indicators for Assessing Proprietary School Quality.
Washington, DC: Department of Education.
Gross, J. P., Cekic, O., Hossler, D., & Hillman, N. (2010). What Matters in Student Loan
Default: A Review of the Research Literature. Journal of Student Financial Aid , 19-29.
Hamel, G., & Prahalad, C. K. (1989, May-June). Strategic Intent. Harvard Business Review ,
63-76.
Harrast, S. A. (2004). Undergraduate Borrowing: A Study of Debtor Students and their
Ability to Retire Undergraduate Loans. Journal of Student Financial Aid , 21-37.
Hossler, D., Braxton, J., & Coopersmith, G. (1989). Understanding Student College Choice.
Higher Education: Handbook of Theory and Research , 231-288.
Janetius, S. T., Shilpa, S., & Eswaran, S. (2013). Change Management for Goal Setting. (T. C.
Mini, Ed.) Hikey.
John, V. (2001). A Model Educational Loan Scheme. Reserve Bank of India,
RPCD.PLNFS.BC.NO.83/06.12.05/2000-01. Mumbai: Reserve Bank of India.
Joshi, D. (2014). Financial Inclusion - The Indian Model, Challenges & Prospects.
Centre for International Development. Cambridge: Harvard Kennedy School.
Lavanya, R., & Ramachandran, S. (2014). A Study on the Higher Education Loan in India.
Indian Journal of Applied Research .
Lochner, L., & Monge-Naranjo, A. (2004). Education and Default Incentives with
Government Student Loan Programs. Ontario: University of Western Ontario.
Matthew, F. R. (2009). Leadership and Organizational Strategy. The Innovation Journal .
Ministry of Human Resource Development, Government of India. (2013). National Higher
Education Mission.
NABARD. (2013). Status of Micro Finance in India.Mumbai: NABARD.
Narayana, M. R. (2005). Student Loan by Commercial Banks: A Way to Reduce State
Government Financial Support to Higher Education in India. The Journal of Developing
Areas , 171-187.
NSSO. (2003). NSSO 59th Round Schedule 33 Situation Assessment survey of Farmers 2003.
Ministry of Statistics and PI.
Office of the Registrar General and Census Commissioner. Census of India’s 2011
Provisional. Ministry of Home Affairs.
Pant Joshi, D. (2011). Financial Inclusion and Literacy. OECD Seminar.

119
Pascarella, E. T., & Terenzini, P. T. (2005). How College Affects Students. (K. A. Feldman,
Ed.) San Francisco: Jossey-Bass.
Planning Commission. Eleventh Five Year Plan. Government of India.
Podgursky, M., Ehlert, M., Ryan, M., Donald Watson, & John, W. (2002). Student Loan
Defaults and Enrollment Persistence. Journal of Student Financial Aid , 27-42.
Psacharopoulus, G., Tan, J.-P., & Jimenez, E. (1985). Financing Edcuation in Developing
Countries: An Exploration of Policy Options. World Bank .
Puttaswamaiah, S. (2010). Financing Higher Education : A Study of Educational Loans.
Centre for Multi-Disciplinary Development Research.
Rajan, R. (2009). A Hundred Small Steps, Committee on Financial Sector Reforms.
Government of India. Planning Commission.
Rangarajan, C. (2008). The Committee on Financial Inclusion. Ministry of Finance.
Government of India.
Reserve Bank of India. (2006). Financial Inclusion by Extension of Banking Services - Use of
Business Facilitators and Correspondents. Circular.
Reserve Bank of India. (2014). Reserve Bank of India Annual Report 2013-14.
Rothstein, J., & Rouse, C. E. (2011). Constrained After College : Student Loans and Early
Career Occupational Choices. Journal of Public Economics , 149-163.
Srinivasan, R., & Das, D. (2011). Analysis of Education Loan ; A Case study of NC Territory
of Delhi. International Journal of Management and Strategy .
Steiner, M., & Teszler, N. (2005). Multivariate Analysis of Student Loan Defaulters at Texas
A&M University. Texas Guaranteed Student Loan Corporation.
Steiner, M., & Tym, C. (2005). Multivariate Analysis of Student Loan Defaulters at the
University of South Florida. Florida: TG Research and Analytical Services.
The World Bank. (n.d.). World Data Bank. Retrieved January 17, 2015 from
www.worldbank.org: Development Research Group
Tilak, J. B. (1993). Financing Higher Education in India; Principles, Practice and Policy
Issues. Higher Education .
University Grants Commission, Government of India. (2014). Annual Report 2012-13. Vijaya
Bank. (2015 14-Febraury). Retrieved 2015 14-Febraury
Volkwein, J. F., & Szelest, B. P. (1995). Individual and Campus Characteristics associated
with Student Loan Defaults. Research in Higher Education , 41-72.

120
W.Schultz, T. (1961 March). Investment in Human Capital. The American Economic Review,
1-17.
Walsh, W. B. (1973). Theories of Person-Environment Interaction : Implications for the
College Student. Iowa: American College Testing Program.
Webster, J., & Watson, R. T. (2002 June). Analysing the Past to Prepare for the Future:
Writing a Literature Review. MIS Quarterly .
Wilms, W. W., Moore, R. W., & Bolus, R. E. (1987). Whose Fault is Default? A Study of the
Impact of Student Characteristics and Instituitional Practices on Guaranteed Student
Loan Default Rates in California. Educational Evaluation and Policy Analyis , 41-54.
Woodhall, M. (1983). Student Loans as a Means of Financing Higher Education: Lessons
from International Experience. World Bank .
World Bank. (2008). World Development Report 2009 : Reshaping Economic Geography.
Washington: World Bank.
Ziderman, A., & Shen, H. (2009 Mar). Student Loans Repayment and Recovery:
International Comparisons. Higher Education , 315-333.

121