Working Paper 2020-1

NAFINDEX: Measure of Financial Inclusion based on

NABARD All India Rural Financial Inclusion Survey
(NAFIS) Data

Dr. K.J.S. Satyasai, CGM

Dr. Ashutosh Kumar, DGM

राष्ट्रीय कृ षि और ग्रामीण षिकास बैंक

National Bank for Agriculture and Rural Development
आर्थिक षिश्लेिण और अनुसंधान षिभाग, प्रधान कायािलय, मुंबई
Department of Economic Analysis and Research
Head Office, Mumbai
July 2020
The views expressed in this paper are authors’ only. This is work in progress and any
use of material may be done keeping this in mind.

Suggestions/comments may be sent to the e-mail : [email protected]

2 K J S SATYASAI AND ASHUTOSH KUMAR 2020

NAFINDEX: Measure of Financial Inclusion based on NABARD All

India Rural Financial Inclusion Survey (NAFIS) Data

Dr. K J S Satyasai and Dr. Ashutosh Kumar*

National Bank for Agriculture and Rural Development, Mumbai

Abstract
Financial inclusion (FI) is a multi-dimensional phenomenon unlike its
pre-cursor concepts of access to credit or access to savings bank account which
define financial inclusion in a narrow sense. Hence, measuring financial inclusion
is complicated and requires developing a suitable index. Several scholars
developed FI index mostly following methodology of Human Development Index.
Sharma (2008), Mehrotra (2009), Ambarkhane et al (2012), Gupte et al (2012),
Goel and Sharma (2017) are a few of them. CRISIL’s Inclusix is an index at
district level. Department of Financial Services (DFS), Ministry of Finance,
Government of India also is constructing an index of financial inclusion to help
monitoring over the years. These indices covered different dimensions. All these
indices are constructed using data from secondary sources and measure supply
side access. That is, they mainly represent the access an individual can have.
Actual use of a financial service by an individual or household is not reflected in
these indices. World Bank’s Findex is one index developed based on survey data
of individuals. We recommend that a FI index should manifest the actual usage of
financial services in terms of breadth, intensity and extent of digital penetration.
We, therefore, propose NAFINDEX, based on state-wise household level access to
financial services based on data from NABARD All India Rural Financial
Inclusion survey (NAFIS). Based on the field level data collected through NAFIS
2016-17, NAFINDEX has been constructed for different states of India. Three
dimensions, traditional banking products, modern banking products, and

CGM and DGM, DEAR respectively. Corresponding Author: K. J. S. Satyasai

Email: [email protected]
Views expressed are those of authors alone. Usual disclaimers apply.
2020 MEASURING FINANCIAL INCLUSION 3

payment systems, are considered for constructing the index. The average value of
index at all India is 0.337. There are variations across states in the value of
NAFINDEX and dimension indices. Interestingly, many states which saw lower
penetration of traditional banking products as reflected in the respective
dimension index, the modern banking products and payment mechanisms
showed higher values. This underlines the direction for the future banking
expansion in hither to unreached states.

Keywords: Financial Inclusion, Index, NAFINDEX

-----------------------------------------------------------------------------------------------
4 K J S SATYASAI AND ASHUTOSH KUMAR 2020

1 Introduction
Financial inclusion is increasingly being recognized world over as a key
driver of economic growth and poverty alleviation. Apart from these benefits,
financial inclusion (FI) imparts formal identity, provides access to the payments
system and to savings safety net like deposit insurance, and enables the poor to
receive direct benefit transferred in a leak-proof manner. At a macro level,
greater FI is considered crucial for sustainable and inclusive socio-economic
growth for all. However, the FI is not an end in itself as it is only a means to
reach higher levels of development. The potential for development in the various
sectors of the economy such as primary sector (agriculture and allied sectors) and
Micro, Small and Medium Enterprises (MSME) sector is enormous. However, the
limited access to affordable financial services such as savings, loan, remittance
and insurance services by the vast majority of the population in the rural areas
and unorganised sector is believed to be acting as a major constraint to the
growth impetus in these sectors. It is widely believed that access to affordable
financial services - especially credit and insurance - enlarges livelihood
opportunities and empowers the poor to take charge of their lives. Such
empowerment also adds to social and political stability in the economy.

2 What is Financial Inclusion?

With an objective to extend such financial services to a sizeable majority of
population particularly who continue to remain excluded from the opportunities
and services provided by the financial sector, a Committee on Financial Inclusion
(CFI) was set up by the Govt. of India under the Chairmanship of Dr. C.
Rangarajan in 2006. This Committee on Financial Inclusion (Rangarajan, 2008)
defined Financial Inclusion as:
“process of ensuring access to financial services and timely and adequate
credit where needed by vulnerable groups such as weaker sections and
low income groups at affordable costs.”
2020 MEASURING FINANCIAL INCLUSION 5

The report identified demand and supply sides of financial services and
emphasised on improving human and physical resource endowments.
Subsequently, Planning Commission, Govt. of India (2009) in a Report of the
Committee on Financial Sector Reforms mentioned:
“Financial Inclusion is not only about credit but involves a wide range of
Financial Services including savings accounts, insurance and remittance
products. Moreover, credit provision without adequate measures to
create livelihood opportunities and enhance credit absorption amongst
poor will not yield desired results.”

Emphasizing the importance of those financial products, the report

recommended that access to safe and remunerative methods of savings,
remittances, insurance and pension need to be expanded. They suggested crop
insurance for farmers and health insurance for the poor as vulnerability reducing
instruments.

The recent developments in banking technology have transformed banking

from the traditional brick-and-mortar infrastructure like staffed branches to a
system supplemented by other channels like automated teller machines (ATM),
credit/debit cards, internet banking, online money transfers, mobile money, UPI,
etc. The moot point, however, is that access to such technology is restricted only
to certain segments of the society. Indeed, some trends, such as increasingly
sophisticated customer segmentation technology – allowing, for example, more
accurate targeting of certain sections of the market – have led to restricted access
to financial services for some groups. There has been a growing divide, with an
increased range of personal finance options for a segment of high and upper
middle-income population on one hand and a significantly large section of the
population who lack access to even the most basic banking services on the other.
This is termed “financial exclusion”. These people, particularly, those living on
low incomes, cannot access mainstream financial products such as bank
accounts, credit, remittances and payment services, financial advisory services,
insurance facilities, etc. The essence of financial inclusion is in trying to ensure
6 K J S SATYASAI AND ASHUTOSH KUMAR 2020

that a range of appropriate financial services is available to every individual and

enabling them to understand and access those services. Total financial inclusion
or “Sampoorn Viteeyea Samaveshan” (SVS) envisaged to cover six broad areas 1)
Ensuring every district with 1,000-5,000 households had access to banking
services within 5 km by March 2016, 2) Provide financial literacy, 3) Provide
basic banking for all beneficiaries of government schemes by March 2016, 4) An
overdraft of Rs. 5,000, 5) Micro insurance and 6) Pension scheme for the
unorganized sector (Mehta and Shah, 2014). Now the question arises, how do we
get to know the level of financial inclusion of a population in a geography? This
necessitates measurement of level of financial inclusion through an objective tool
say financial inclusion index.

3 Why Financial Inclusion Should be Measured?

Financial inclusion is a key policy area and the central banks world over
has an interest in it. Greater financial inclusion is essential for sustained
economic welfare and for reducing poverty. It also supports economic, monetary
and financial stability, by making saving and investment decisions more efficient,
enhancing the effectiveness of monetary policy instruments, and facilitating the
functioning of the economy (IFC Bulletin No. 38, Bank for International
Settlements). In turn, economic stability helps to develop and strengthen a
smoothly functioning financial system that can support financial inclusion.
Therefore, it is very essential to measure financial inclusion objectively.

4 How to Measure Financial Inclusion?

Now, the question arises, how we measure financial inclusion. Financial
inclusion is a multi-dimensional phenomenon and hence, its measurement
remains inadequate if crucial dimensions are not included. Further, data on
various indicators of financial inclusion raise important issues. Well-founded
data frameworks are essential while developing financial services for the poor, in
both formal and informal markets. Appropriate indicators in adequate number
are a precondition for good financial inclusion measurement. They ensure that
financial inclusion is properly assessed and that policies aimed at it are
2020 MEASURING FINANCIAL INCLUSION 7

adequately implemented, monitored, and adjusted as required. Good statistics

can also help to strike a fine balance between encouraging innovation and the
growth of financial services on the one hand, and ensuring that financial stability
is preserved, on the other.

5 Developing Indices of Financial Inclusion

Measurement of financial inclusion could be done through developing a
suitable financial inclusion index (FII). A composite financial inclusion index,
provides scope for multiple dimensions of financial inclusion to be reduced to a
single one, making it simpler for analysts and policymakers alike. In general,
such indices have no units and are constructed by making all the measured
dimensions comparable. Such an index can be a valuable instrument to diagnose
the financial inclusion situation for a specific geographic location, and to
facilitate spatial and temporal comparisons. In turn, the index based on a set of
identified key performance indicators can be established as a benchmark and
used to identify best practices. Nevertheless, FII cannot be considered as a
universal or exclusive policy tool. In fact, developing composite index is not a
goal in itself. The quality of underlying data, however, is crucial.

Once we construct an index and measure financial inclusion, we can strive

to achieve beyond its benchmark level. However, how do we measure financial
inclusion depends on how we define it. In India, there have been several attempts
to measure financial inclusion based on proportion of adult population having
access to formal banking system, proportion of adults having bank account, bank
accounts per 1000 adult population, ATMs per 1000 sq. km, population being
serviced per branch, etc. However, all these are supply-side factors determining
the status of financial inclusion. Similarly, there are demand-side factors such as
income level, credit absorption capacity, financial awareness and literacy level of
people, availability of livelihood opportunities in the area, etc. which determine
the level of financial inclusion of a particular geography. Further, the quality of
financial services being supplied and availed is another dimension determining
the quality of financial inclusion. Thus, financial inclusion is a multi-dimensional
8 K J S SATYASAI AND ASHUTOSH KUMAR 2020

phenomenon that can be better summarised by a composite index. The financial

inclusion index (FII) should be such that: (i) it represents the true situation as far
as possible, (ii) it is simple and easy to compute so that it is amenable for
comparison, (iii) it should have meaningful bounds (say 0 and 1), and, (iv) it
should have monotonicity (higher values indicating higher level of financial
inclusion)

There are 4 Steps to be followed in constructing FII. Develop a clear

theoretical framework, to begin with, to have a sound basis for selecting the
individual indicators of interest. Second, define precisely the data content,
analysis, weighting and aggregation scheme for the selected indicators. Third,
conduct sensitivity and robustness analysis to ensure quality. For instance, the
indicator should not change dramatically if one of the individual components is
excluded, or if a different scheme of weights is used. Lastly, create a framework
for representing and communicating information provided by an FII, especially
when making cross-country comparisons on the overall performance of the index,
and the contribution of the various indicators to it.

6 Various Approaches to Construction of Index of FI

FI index is constructed using multi-dimensional framework representing
demand and supply factors. Beck, Kunt and Peria (2007) make a clear distinction
between (1) access and the possibility of use, and, (2) the actual use of financial
services. Honohan (2005) included contribution of financial access to household
wellbeing and firm productivity on demand side, while product/service design
(usefulness for the poor), cost and information barriers to access on supply side.
And, they used following financial access indicators:
1) Payments: Inland and international remittances –crucial for the
families dependent on migrant income.
2) Savings mobilization (deposit services).
3) Monitoring of users of funds (mechanisms for building credit
worthiness)
2020 MEASURING FINANCIAL INCLUSION 9

4) Transforming Risk (Insurance etc.)

Sarma (2010) considered 3 dimensions: penetration (number of bank

accounts per adult population), availability (number of banking outlets (branches
and ATMs) per 1000 population), and usage (volume of credit and deposit as
proportion of GDP). Arora (2010) considered outreach, ease, and cost. Outreach
is measured by branch and ATM penetration per area and population. Ease is
measured by a) minimum amount to open saving account, b) minimum amount
to maintain saving account, and, c) number of documents required to open bank
account. Cost includes fees for different services offered by the bank. Here again,
all the dimensions are related with banks only, and other financial services are
left out in the process. Gupte, Venkataramani and Gupta (2012) considered four
dimensions, namely, outreach, usage, ease, and cost of transaction, which are
combined taking geometric mean. Kunt and Klapper (2012) also measured
financial inclusion using four indicators, namely, 1) formal accounts, 2) savings
behaviour, 3) sources of borrowing, purposes of borrowing, and use of credit
cards, and 4) use of insurance products. Rahman (2013) considered four
indicators, namely, convenient accessibility, take up rate, responsible usage, and
satisfaction level. All are assigned equal weights adding to unity. Yorulmz (2013)
followed the method suggested by Sarma (2008) and used multi-dimensional
approach. Normalized inverse Euclidean distance from the ideal point for three
dimensions, access, availability, and usage are considered. A summary of various
Researcher/Social Scientists and variables used by them are presented below
(Table 1)

Table 1: Summary of various Researcher/Social Scientists and

variables used
Researchers Variables used
Beck, Kunt and Peria 1. Access and possibility of use and 2. Actual use
(2007)
Honohan (2005) 1. Payments, 2. Savings mobilization, 3. Monitoring
10 K J S SATYASAI AND ASHUTOSH KUMAR 2020

of users of funds and 4. Transforming Risk

The Consultative Group 1. Savings, 2. Payments, 3. Credit and 4. Delivery
to Assist the Poor (2009)
Sarma (2010) 1. Penetration, 2. Availability and 3. Usage
Arora (2010) 1. Outreach, 2. Ease and 3. Cost
Rahman (2013) 1. Convenient Accessibility, 2. Take Up Rate, 3.
Responsible Usage and 4. Satisfaction level.
Gupte, Venkataramani 1. Penetration, 2. Availability, 3. Usage 4. Ease and
and Gupta (2012) 5. Cost.
Kunt, Klapper (2012) 1. Formal accounts (a. the mechanics of the use, b.
purpose, c. barriers, d. alternatives to formal
accounts, e. penetration and f. receipt of payments),
2. Savings behavior (a. use of accounts, b. use of
community-based savings methods and c. the
prevalence of savings goals),
3. Sources of borrowing, purposes of borrowing, and
use of credit cards and
4. Use of insurance products
Yorulmz (2013) 1. Access, 2. Availability and 3. Usage
Credit Rating and 1. Branch penetration, 2. Credit
Information System of
Indian Ltd. (CRISIL)
(2013)
Amberkhane et al (2014) Drag factors besides demand, supply and
infrastructure

Amberkhane et al. (2014) considered drag factors besides demand, supply,

infrastructure dimensions to construct FI index. On demand and supply sides
the indicators are related banks, NBFCs and insurance with 50% weight to banks
and 25% weight for each of the other two. On supply side, the indicators are
2020 MEASURING FINANCIAL INCLUSION 11

about spread of branches or outlets. On demand side, the indicators are related to
deposits, loans, remittances, density of SHGs, insurance penetration, etc.
Infrastructure indicators are on irrigation, transport, power, literacy, and health.
Drag factors considered are population growth, law and order situation, and
corruption. The values of all indicators are normalized to converts values of
indicators between 0 and 1 using formula below:

Ai -- m
di = ------------ (1)
M -- m
.
where, for ith State
di is the normalized value of indicator.
Ai is the actual value of indicator.
M is maximum value of indicator.
m is the minimum value of indicator.

Then FI index is the Euclidean distance measured by using displaced ideal (D.I.)
method. Financial inclusion index for rth state was obtained by inverse
normalized distance from the ideal as given below:

√{ (1-sr)2 + ( 1 – dr)2 + ( 1 – ir)2 }

1 − ------------------------------------------ (2)
√3

The term √{ (1-sr)2 + ( 1 – dr)2 + ( 1 – ir)2 } in (2) above is the Euclidean

distance of the point (sr,,dr,,ir) i.e. position of rth State from the ideal (1 1 1) (This
is the distance between the points Sr and P in Figure 1) and dividing it by √3
normalizes it, in three dimensional space. Further subtracting this from 1 (i.e.
normalized distance of ideal point from the origin) gives inverse normalized
distance, which is the index. This Index satisfies all intuitive properties of an
index suggested by Nathan, Mishra, and Reddy (2008); namely Normalization,
12 K J S SATYASAI AND ASHUTOSH KUMAR 2020

Anonymity, Monotony, Proximity, Uniformity, and Signalling.

Figure 1: Diagrammatic representation of Euclidean distance method

Source: Nathan, Mishra and Reddy (2008)

The final Financial Inclusion Index is derived after applying the drag
factors. Suppose Dr is the drag index for rth state, the impact factor is taken as
1/(1+ Dr) (3)
Comprehensive Final Inclusion Index is arrived at by multiplying impact factor as
above with Financial Inclusion Index obtained earlier.

Points of difference with other methodologies:

1) Method is like UNDP approach following multidimensional approach
2) This method uses Linear Averaging method for calculation of indices of
Demand, Supply and Infrastructure dimensions, whereas Displaced Average
Method is used for combining three indices.
3) Signalling characteristics of D.I. Method is considered more suitable to
proposed index as it indicates unique optimal path to reach higher value.
Moreover, the signalling characteristic implies that an improvement in a
dimension that has lower value is more important than an equivalent
2020 MEASURING FINANCIAL INCLUSION 13

improvement in a dimension that has a higher value. Methodology suggested is

flexible as any other relevant factor or indicator identified can be added to any of
the dimensions or in drag.

However, it is questionable if an index to measure a phenomenon shall

include explanatory factors such as infrastructure related or drag factors. For
instance, higher corruption may inhibit inclusion because of which the financial
inclusion index may be lower in a state. We consider it inappropriate to multiply
with drag factor. Even if inclusion of infrastructure dimension is justified, some
indicators therein are about physical aspects of infrastructure. Certain others are
proxies. For instance, female literacy cannot be taken as proxy for educational
infrastructure. This is a methodological issue.

Goel and Sharma (2017) have used following parameters for constructing index:
 Banking Penetration (D1) - demographic branch penetration i.e., no. of
accounts (deposits and loans) per 1,000 populations with different
financial institutions (d1).
 Availability (D2) of banking services – number of ATMs per 1,00,000
population (d2),
 Number of bank branches per 1,00,000 population (d3),
 Number of ATMs per 1,000 sq. km (d4)
 Number of scheduled commercial banks per 1,000 sq. km (d5).
 Access to Insurance (D3) – number of life insurance (LIC) offices (d6).

The indicators are normalised, and indices are constructed using weights.
The FII is measured as the simple average of two indices, X1 and X2, measured,
respectively, based on distance from zero, the ideal point, and, ideal point, W, for
each indicator.
Ad -md
dd = Wd∗ --------------- (4)
Md-md
14 K J S SATYASAI AND ASHUTOSH KUMAR 2020

Where,
wd = Weight attached to the dimension d, 1≥ wd ≥ 0;
Ad = Actual value of dimension d;
md = Minimum value of dimension d;
Md = Maximum value of dimension d;
dd = Dimensions of financial inclusion d.

(5)

(6)

(7)

Depending on the value of FII, the time period under study has been categorized
as
1. 0 ≤ FII ≤ 0.4; indicates low financial inclusion, LFI;
2. 0.4 < FII ≤ 0.6; indicates medium financial inclusion, MFI
3. 0.6 < FII ≤ 1; indicates high financial inclusion, HFI

From the computation of FII across a time period of twelve years, India
can be categorized under low financial inclusion during 2005 to 2012. During this
time period, the value of FII ranged between 0 - 0.4. During 2013, condition of
financial inclusion improved, and India fell under medium financial inclusion
with FII from 0.4 to 0.6. The objective of inclusive growth was achieved further
during 2014-2015 and India fall under high financial inclusion range in this time
2020 MEASURING FINANCIAL INCLUSION 15

period. The value of FII ranged from 0.6 to 1. Unlike earlier studies where only
indices such as banking penetration, availability of banking services and usage of
banking system were used, Goel and Sharma (2017), included indicators such as
access to savings and access to insurance also. Also, FII is constructed for a
longer period of twelve years.

Sriram and Sundaram (2015) measured FII using 3 dimensions, viz,.

access, availability, and usage. These dimensions are assigned weights of 1 for
access and0.5 each for the remaining two. Access is measured through number of
bank accounts in the area, availability, through number of access points
(branches, ATMs, banking correspondents) in the area, and, usage, through
number of accounts (savings, deposits, loan and credit) held by respondents.
These dimensions are combined to compute FII as the difference of square root
of Euclidean distance with reference to the ideal (i.e. weight) from unity. The
formula is as below:

FII = 1 –sqrt{[(1 – Pi) ^2 + (0.5 – Ai) ^2 + (0.5 – Ui) ^2] / 1.5} (8)

Where,
FII = Financial Inclusion Index
Pi = Access
Ai = Availability
Ui = Usage

Sarma (2008) has computed the values of IFI for 54 countries using the
three basic dimensions of financial inclusion–accessibility, availability and usage
of banking services. Accessibility has been measured by the penetration of the
banking system proxied by the number of bank A/C per 1000 population.
Availability has been measured by the number of bank branches and number of
ATMs per 100,000 people. The proxy used for the usage dimension is the volume
of credit plus deposit relative to the GDP. Gupte et al. (2012) considered 4
dimensions, outreach, usage, ease and cost. Outreach has two sub-dimensions,
16 K J S SATYASAI AND ASHUTOSH KUMAR 2020

penetration, and availability. Ease too has two, directly related, and inversely
related. Total 5 indicators were included, and no indicators were considered for
ease and cost dimensions.

Demirguc-Kunt & Klapper (2012) delineated the methodology and

computing Global Findex which measured the use of financial services (demand)
as opposed to the access (supply) to them. Several indicators were used for
computing Findex. The first set of indicators focuses on formal accounts; the
mechanics of the use of these accounts (frequency of use, mode of access); the
purpose of these accounts (personal or business, receipt of payments from work,
government, or family); barriers to account use; and alternatives to formal
accounts (mobile money). The account penetration indicator measures individual
or joint ownership of formal accounts—accounts at a formal financial institution
such as a bank, credit union, co-operative, post office, or microfinance
institution. It includes those who report having a debit or ATM card tied to an
account. Indicators relating to the receipt of payments measure the use of formal
accounts to receive wages (payments for work or from selling goods), payments
or money from the government, and family remittances (money from family
members living elsewhere). The second set of indicators focuses on savings
behaviour. This relates to the use of accounts, as people often save at formal
financial institutions. Other indicators explore the use of community-based
savings methods and the prevalence of savings goals. The third set focuses on
sources of borrowing (formal and informal); purposes of borrowing (mortgage,
emergency or health purposes, and the like); and use of credit cards. The fourth
focuses on use of insurance products for health care and agriculture.

Most of the above are at country level and one or two are there for states.
Credit Rating and Information Services of Indian Ltd (CRISIL) (2013) calculated
index, Inclusix, available at district level. It considered three dimensions, namely
branch penetration, credit penetration and deposit penetration. However, it has
serious limitations in terms of coverage of dimensions and indicators. It is only
2020 MEASURING FINANCIAL INCLUSION 17

in terms of number of accounts and not amount. Further, it covers scheduled

commercial banks data only.

Mehrotra et al. (2009) also built up an index for financial inclusion using
similar kind of aggregate indicators like number of rural offices, number of rural
deposit accounts, volume of rural deposit and credit from banking data for
sixteen major states of India. Here also, Financial Inclusion Index is estimated at
the district level in India.

8 Issues/Limitations with Existing Measures

Whatever existing measures were used so far, they had some issue or the
other to be resolved/improved upon. Some of them are as under:
• Mostly based on secondary and administrative data though ‘demand side’
and ‘usage’ based indicators are incorporated. Exceptions like Findex exist
• Large scale survey-based data not used. Again, Findex is an exception.
There too, the sample for any individual country is limited and
representative at country level only.
• Mix up between indicators and their drivers is an issue. The question is
should the financial inclusion index should combine indicators of access
and/or use as well as factors influencing these indicators, together.
• Many dimensions and large number of indicators are available. Should
there be any standardization and consensus on indicators?

9 Index Based on NAFIS data

To take care of some of the limitations of the existing measures, we have
tried to build an index, NAFINDEX, based on NABARD Rural Financial Inclusion
Survey (NAFIS) 2016-17 data. NAFIS was undertaken by NABARD pan-India
during 2016-17 covering both financial and livelihood aspects of 40000 sample
households across 29 states. The survey covered all aspects of financial inclusion
from a household perspective, viz., savings, borrowing, investment, remittances
& payments, and insurance. Besides, the survey also covered financial literacy
and experience of households with payment mechanisms.
18 K J S SATYASAI AND ASHUTOSH KUMAR 2020

The index is generated at all India and state-level based on the field level
data collected from households. For constructing NAFINDEX, we covered three
dimensions – traditional banking products (T), modern banking services (M),
and payment mechanisms (P). Traditional banking products covered savings,
investments, loans, and others (insurance & pension); modern banking services
included usage level of ATMs, internet banking, and mobile banking; and,
payment mechanisms covered usage of cheque and credit/debit card as well as
ease of using them. The indicators used and weights assigned for this Index are
given in Table 2.

Table 2: Indicators used for constructing NAFINDEX

Dimension Service/sub- Indicator Symbol of Weight
dimension normalised
indicator
Traditional Savings % households that made any T11 0.125
Banking saving in the last 1 yr
Products mean savings (with all agencies) T12 0.125
per household in the last 1 year
[base: saver household who
reported their saving amount]
Investment % households that made any T21 0.125
investment in the last one year
mean investment in all assets for T22 0.125
household reporting any
investment in the last one year
Loans incidence of indebtedness T31 0.125
average outstanding debt per T32 0.125
indebted household (rs.)
Others % households with at least one T41 0.125
member having any insurance
2020 MEASURING FINANCIAL INCLUSION 19

% households having pension T42 0.125

Modern usage % ATM users M11 0.167
Banking % internet banking users M12 0.167
services % mobile banking users M13 0.167
Ease in % users having ease of using ATM M21 0.167
using
% users having ease of using M22 0.167
internet banking
% users having ease of using M23 0.167
mobile banking
Payment usage % users of cheque P11 0.25
Mechanisms % users of debit/credit card P12 0.25
Ease in % users having ease in using P21 0.25
using cheque
% users having ease in using P22 0.25
debit/credit card

The indicators are combined to form dimension indices which are in turn
combined into NAFINDEX. The values of all indicators are normalized to scale
down values of indicators between 0 and 1 using formula at (1).

Individual dimension indices are computed as below,

Tn = ∑(Wij*Tij)
Mn = ∑(Wij*Mij)
Pn = ∑(Wij*Pij)

Where,
Tn is the dimension index for traditional banking products for nth state;
Mn is the dimension index for modern banking services; and,
Pn is the dimension index for payment mechanisms.
Subscripts i and j stand for sub-dimension and indicator, respectively.
20 K J S SATYASAI AND ASHUTOSH KUMAR 2020

NAFINDEX = 3√(Tn*Mn*Pn)
We have fitted a linear regression model to understand the explanatory factors
for variation of NAFINDEX across states.
Dependent variable: NAFINDEX = Financial Inclusion Index
Independent variables:
Mf-membership = index of per cent HH having membership with
microfinance institutions
% trained = proportion of HH received training
income index = index of HH income
% institutional loan = share of institutional loan in total
The regression is worked for agricultural households, non-agricultural
households and all rural households.

10 State-wise NAFINDEX values

The state wise Index of FI calculated based on NAFIS data are given in
Table 3. The NAFINDEX for all India is 0.337 in a scale of 0 to 1. The value of
the index for banking products dimension is 0.307. The value for the payment
mechanisms dimension is the highest at 0.370 followed by 0.345 for banking
services. Punjab, Kerala, and Karnataka ranked top three states in banking
products dimension while Bihar, Chhattisgarhi, and Madhya Pradesh are at the
last three positions. Goa, Manipur, and Nagaland are at the top for banking
services dimension and Jharkhand, Madhya Pradesh, and Meghalaya are at the
bottom. For the payment mechanisms dimension, top ranking states are Goa,
Assam, Manipur, and Tripura while Uttarakhand, Rajasthan, and Chhattisgarhi
are at the bottom.
Table 3: NAFINDEX values for different states and all India
State Banking Rank Banking Rank Payment Rank NAFIND Rank
products Services mechanism EX
Goa 0.472 5 0.946 1 0.761 1 0.600 1
Punjab 0.617 1 0.473 12 0.383 19 0.486 2
Karnataka 0.533 3 0.430 14 0.438 13 0.483 3
Telangana 0.482 4 0.563 8 0.478 8 0.480 4
2020 MEASURING FINANCIAL INCLUSION 21

State Banking Rank Banking Rank Payment Rank NAFIND Rank

products Services mechanism EX
Andhra
0.424 7 0.703 4 0.529 5 0.473 5
Pradesh
Kerala 0.609 2 0.446 13 0.362 21 0.470 6
Manipur 0.385 12 0.791 2 0.558 3 0.464 7
Tripura 0.366 14 0.523 10 0.558 3 0.452 8
Jammu &
0.420 8 0.427 15 0.450 12 0.435 9
Kashmir
Odisha 0.379 13 0.381 24 0.477 9 0.425 10
Haryana 0.409 10 0.328 26 0.423 14 0.416 11
Mizoram 0.322 16 0.580 6 0.476 10 0.392 12
Assam 0.237 21 0.482 11 0.625 2 0.385 13
Himachal
0.460 6 0.565 7 0.310 23 0.377 14
Pradesh
Meghalaya 0.318 17 0.240 29 0.403 17 0.358 15
Arunachal
0.337 15 0.353 25 0.374 20 0.355 16
Pradesh
Sikkim 0.253 20 0.678 5 0.486 7 0.351 17
Nagaland 0.318 17 0.734 3 0.325 22 0.322 18
West Bengal 0.202 25 0.419 16 0.507 6 0.320 19
Maharashtra 0.224 22 0.416 18 0.416 16 0.305 20
Jharkhand 0.200 26 0.321 27 0.451 11 0.301 21
Gujarat 0.215 24 0.531 9 0.420 15 0.300 22
Uttar
0.217 23 0.417 17 0.397 18 0.294 23
Pradesh
Tamil Nadu 0.387 11 0.404 20 0.208 25 0.284 24
Uttarakhand 0.420 8 0.401 21 0.189 27 0.281 25
Bihar 0.198 27 0.387 23 0.264 24 0.229 26
Rajasthan 0.276 19 0.398 22 0.178 28 0.222 27
Madhya
0.141 29 0.266 28 0.195 26 0.166 28
Pradesh
Chhattisgarh 0.160 28 0.411 19 0.055 29 0.094 29
All India 0.307 0.345 0.370 0.337

Table 4 gives results of linear regression model estimated to explain the

variation in NAFINDEX. Of the four variables included in the model two
variables Mf-membership and income index are significant for agricultural, non-
agricultural and overall rural households. Proportion of households trained has
significant effect on NAFINDEX. That is, states where the penetration of
22 K J S SATYASAI AND ASHUTOSH KUMAR 2020

microfinancing institution is higher and where households reported higher

income, the financial inclusion index is also higher. The NAFINDEX among non-
agricultural households is higher in states with higher proportion of households
with trained households. The explanatory of power the regression is 48 to 55 per
cent and is statistically significant.
Table 4: Factors explaining variation in NAFINDEX
Variable/description Particular Agri HH Non-Ag HH Rural HH
Constant Coefficient 0.252247 *** 0.140434 0.211313 ***
***
std error 0.0518705 0.0486343 0.046158
p -value <0.0001 0.0081 0.0001
Mf-membership Coefficient 0.204269 *** 0.232292 0.219316 ***
(index of per cent ***
HH having std error 0.0573099 0.0708419 0.0646751
membership with p -value 0.0016 0.0032 0.0024
microfinance
institutions)
% trained Coefficient −0.00254901 0.186124 0.0588672
(proportion of HH ***
received training) std error 0.0683906 0.0720409 0.0693977
p -value 0.9706 0.0163 0.4047
income index Coefficient 0.243671 *** 0.165032 0.294322
(index of HH ** ***
income) std error 0.0768699 0.0841741 0.0831651
p -value 0.0041 0.0616 0.0017
% institutional loan Coefficient −0.0352766 0.102054 −0.0385928
(share of std error 0.0826278 0.073567 0.0792173
institutional loan in p -value 0.6732 0.1781 0.6306
total)
Note: ***, ** significant at 1% and 5%, respectively
Mean dependent var 0.346811 0.365558 0.362676
Sum squared residual 0.136775 0.208367 0.168062
R-squared 0.478384 0.546915 0.499992
F(4, 24) 5.502705 7.242553 5.999799
Log-likelihood 36.52308 30.41914 33.5362
2020 MEASURING FINANCIAL INCLUSION 23

11 Conclusion
Based on the field level data collected through NAFIS 2016-17,
NAFINDEX has been constructed for different states of India. Three dimensions,
traditional banking products, modern banking products, and payment systems,
are considered for constructing the index. The average value of index at all India
is 0.337. There are variations across states in the value of NAFINDEX and
dimension indices. Interestingly, many states which saw lower penetration of
traditional banking products as reflected in the respective dimension index, the
modern banking products and payment mechanisms showed higher values. This
underlines the direction for the future banking expansion in hither to unreached
states.

References
Ambarkhane, D., Singh, Ardhendu Shekhar, Venkataramani, Bhama (2014).
Developing a comprehensive financial inclusion index. Symbiosis School of
Banking and Finance, Symbiosis International University. Retrieved
from http://ssrn.com/abstract=2485774 or http://dx.doi.org/10.2139/ssrn
.24857749 (accessed on 15 June 2014).
Ambarkhane, D., Singh, A.S., & Venkitaraman, B. (2016). Measuring
comprehensive financial inclusion index of Indian states, Indian Journal of
Rural Management, IRMA, Anand.
Arora, R. (2010). Measuring Financial Access. Discussion Paper - Economics.
Griffith University.
Beck, Thorsten, Kunt, Asli Demirguc and Peria, Maria Soledad Martinez (2007).
Banking services for everyone? Barriers to bank access and use around the
world. Working Paper Series, Policy Research Working Paper, World Bank.
Credit Rating and Information System of India Ltd.(2013). CRISIL Inclusix.
Mumbai.
Goel, S., & Sharma, R. (2017). Developing a financial inclusion index for
India. Procedia computer science, 122, 949-956.
Govt. of India (2009). Report of the Committee on Financial Sector Reforms.
Planning Commission, New Delhi.
24 K J S SATYASAI AND ASHUTOSH KUMAR 2020

Gupte, R., Venkataramani, B. & Gupta, D. (2012). Computation of financial

inclusion index for India. International Journal of Procedia - Social and
Behavioral Sciences, 37, 133 – 149.
Honohan, P. (2005). Measuring microfinance access: Building on existing
cross-country data. The World Bank.
Kunt, Asi Demirguc, and Klapper, L. (2012). Measuring financial inclusion: the
global findex database. Policy Research Working Paper 6025, World Bank,
Washington, DC.
Mehrotra, N., Puhazhendhi,V., Nair Gopakumaran G., and Sahoo, B.B. (2009),
Financial Inclusion: an overview. Occasional Paper No.48, NABARD,
Mumbai.
Mehta, S., & Shah, P. (2014). Modi government woos India’s unbanked
population, to offer OD bonanza for financial inclusion. The Economic
Times, Jul, 9, 2014.
NABARD (2018). NABARD All India Financial Inclusion Survey (NAFIS) 2016-
17.
Nathan, Hippu Salke Kristle, Mishra, Srijit, Reddy, Sudhakar (2008). An
alternative approach to measure HDI. Working Paper, IGIDR, Mumbai.
Rahman, Z. A. (2013). Developing a financial inclusion index. Central Bank.
Retrieved from www.Centralbanking.com/digital_assets/6715/CB23.
4_May2013_ Financial_inclusion_Rahman_New.pdf (accessed on 10
June 2014).
Rangarajan, C. (2008). Report of the committee on financial inclusion. Ministry
of Finance, Government of India.
Sarma, Mandira (2008). Index of financial inclusion. Working Paper 215. Indian
Council for Research on International Economic Relations (ICRIER), New
Delhi.
Sarma, Mandira & Pais, Jesim (2008). Financial inclusion and development: a
cross country analysis. ICRIER, New Delhi.
2020 MEASURING FINANCIAL INCLUSION 25

Sarma, Mandira (2010). Index of financial inclusion. Discussion Paper 10-05,

Centre for International Trade and Development, School of International
Studies, Jawahar Lal University, New Delhi.
Satyasai K.J.S (2011). Innovation for financial inclusion of rural poor & women
in India, Pal Suresh (Ed.), Agriculture for Inclusive Growth, IARI, New
Delhi (p. 164-180)
Singh, Nirvikar (2017). Financial inclusion: Concepts, issues and policies for
India. Issues and Policies for India (July 4, 2017).
Sriram, M., & Sundaram, N. (2015). Financial inclusion index: a customized
regional model with reference to economically most backward districts of
Tamil Nadu, India. Mediterranean Journal of Social Sciences, 6(6), 209.
The Consultative Group To Assist the Poor (2009). Measuring access to financial
services around the world, World Bank.
The Global Findex Database (2017). Measuring Financial Inclusion and the
Fintech Revolution. The World Bank Group, Washington.
Yorulmz, R. (2013). Construction of a regional financial inclusion index in
Turkey. University of Sheffield, Department of Economics.