District-level Assessment of Credit Depth in Uttar Pradesh

By Nishanth K, IFMR Finance Foundation

In this post we assess the state of bank credit depth for the state of Uttar Pradesh (UP) and the variation in credit depth across the 75 districts of UP during the 2004-12 period. We define ‘credit-depth’ as the ratio of total bank credit outstanding to gross domestic product (CGDP) of a particular sub-economy. This is a commonly used measure for the adequacy of credit relative to GDP.

At an all-India level, CGDP went from 38% in 2004-05 to 88% in 2011-12. During this same period, CGDP in UP went from 37% to 68%. It will be observed that despite a similar starting point, UP grew below the national average during this seven year period and ended with a significantly lower level of credit depth.

In addition, this growth in credit depth seems to be largely driven by a substantial increase in concentration of bank credit within a few districts of the state. This is evident from the graph below: among the districts1 in UP considered, the top seven districts ranked on the basis of credit depth in 2004-05 witnessed a significant increase in credit depth (the median CGDP of these seven districts increased from 40% in 2004-05 to 72% in 2011-12). In comparison, although the median credit depth of bottom seven districts also increases significantly, it is still a mere 24% in 2011-12. This is low when compared to both the median of district level credit depth (68%) and the median credit depth of the top seven districts (72%).

The highest district CGDP in 2004-05 was only 54% whereas in 2011-12 this had risen dramatically to almost 200% for Lucknow. In comparison, districts such as Kaushambi (5% in 2004-05 and 9% in 2011-12) and Auraiyya (7% in 2004-05 and 10% in 2011-12) have seen their credit depth levels remain low in this time period. In some districts such as Chitrakoot, Sant Ravi Das Nagar and Etah, there has worryingly been a decrease in credit depth of about 2-4%.

In 2004-05, the bank credit outstanding in the district of Lucknow alone accounted for about 11% of total outstanding in the state. As of 2011-12, bank credit in Lucknow accounted for 19% of the state’s total outstanding credit. Therefore, although there has been increase in the overall credit depth of the state, it is evident that this increase is skewed by the increase in credit supplied to certain districts. This potentially has significant growth consequences for the state.

In the next post, we provide a more detailed discussion of the working paper that evaluates the Tamil Nadu data in detail with respect to the growth dynamics of credit depth.

1 – For the sake of continuity we have omitted the districts that were created post 2004-05. This does understate values for parent districts following the separation of the new district. We have also omitted districts for which GDP data was not available. Hence, our analysis considers only 69 districts.


Determining Optimal Credit Allocation at a District Level

As part of IFMR Finance Foundation’s Working Paper series, Nishanth K & Irene Baby have authored the latest research paper on Determining Optimal Credit Allocation at a District Level.

Paper Brief:

A fundamental role of the banking and financial sectors is to allocate capital to its most productive use. While India has a deep history of policy making aimed at providing and improving access to formal financial services, large swaths of India still remains unbanked – specifically with respect to delivery of formal credit to rural and excluded regions. Although considerable progress has been made, the extent of exclusion is still vast, and even in sectors, segments and regions that contribute significantly to the growth of the economy.

It is in this context that we contend that the design of credit allocations and targets must account for the potential of economic growth at sub-national level. In this paper, we advocate a class of methods that looks at disaggregated growth responsiveness of districts specifically within the state of Tamil Nadu. Such an approach to credit allocation using elasticity based weighting scale to apportion credit to different districts would help direct credit to the most productive regions and sectors. This could enable policy-making bodies to identify districts with excessive and deficient levels of credit depth and can also inform dynamic, district-level interventions resulting in a more holistic growth.

Click here to read the paper.


Exploratory Analysis of Credit and GDP Growth Rates for Tamil Nadu

By Nishanth K & Irene Baby, IFMR Finance Foundation

This post is the next in the credit depth series. While the previous post covered trends in Credit-to-GDP variable for the districts of Tamil Nadu, this post will elucidate trends related to Compounded Annual Growth Rate (CAGR, henceforth referred to as growth rate)1 of Credit, Gross State Domestic Product (GSDP) and Credit-to-GDP variables.

Figure 1 plots growth rate of bank credit outstanding for various districts on the x-axis and growth rate of Gross District Domestic Product (GDDP) for these districts on the y-axis (both in constant prices). The colour of each bubble (as given by the scale on the right hand side) indicates the district’s share in total credit outstanding in Tamil Nadu in 2011-12. The size of each bubble indicates the proportion of each district’s Gross District Domestic Product (GDDP) to Tamil Nadu’s Gross State Domestic Product (GSDP) in 2011-12.

Figure 1: Credit Growth Rate versus GDP Growth Rate of Districts at Constant Prices (2004-05 to 2011-12)

  • Observation 1: There is a catch-up effect (or convergence)2 in credit growth rate, relative to Chennai for districts like Kancheepuram, Vellore and Tiruvallur. However, because of the wide disparities at the base level, the time period for the eventual ‘catch-up’ can be substantial. To elucidate, Chennai with credit growth rate of 23% and GDDP growth rate of 7.5% constitutes more than 50% of total credit outstanding in Tamil Nadu in 2011-12 (indicated by its colour- yellow). Kancheepuram, with credit growth rate of 26%, and GDDP growth rate of 15%- almost twice as Chennai’s- still only constitutes roughly 2% to the total credit outstanding in Tamil Nadu in 2011-123.
  • Observation 2: 12 of the 32 districts, (including Perambalur, Ariyalur, Nilgiris and Dharmapuri) in Tamil Nadu contributed less than 1% each to the state’s total bank credit outstanding. However, it is also to bear in mind that the outliers in the plot: Krishnagiri (on the extreme right, top corner), Tirupur (on the extreme left, top corner) and Ariyalur (on the extreme left, bottom corner) are recently formed districts (as explained in the previous post).

Figure 2 plots the credit-to-GDP ratio for each district in Tamil Nadu in 2004-05 on the x-axis and the same variable for 2011-12 on the y-axis. The colour of the bubbles (as given by the scale on the right hand side) indicates credit-to-GDP in 2011-12.

Figure 2: District Wise Credit-to-GDP Ratio in Tamil Nadu at Current Prices (2004-05 and 2011-12)

  • Observation 3: As discussed in the previous blog post, the credit-to-GDP ratio for Tamil Nadu in 2011-12 was approximately 71.4% (current prices). As evidenced from Figure 2, this high credit depth was primarily driven by Chennai (561% in 2011-12, current prices) and Coimbatore (131% in 2011-12, current prices).

Figure 3 shows a zoomed-in version of the same figure to better understand the trends in districts besides Chennai and Coimbatore.

  • Observation 4: Districts such as Tiruvallur and Vellore had low levels of credit depth with 15% and 18% respectively (current prices). Hence, there was a wide variation in the levels of credit depth in Tamil Nadu as of 2011-12.

Figure 4: Growth in Credit-to-GDP Ratio for Chennai and Coimbatore (2004-05 to 2011-12)

Figure 5: Growth in Credit-to-GDP Ratio for Select Districts (2004-05 to 2011-12)

Figures 4 and 5 show distinct patterns for different districts in their growth of Credit-to-GDP ratio from 2004-05 to 2011-12.

  • Observation 5: Districts such as Chennai (Figure 4) and Theni (Figure 5) have been growing relatively steadily whereas districts such as Thiruvallur and Dharmapuri have relatively stagnated.

Thus, the exploratory analysis undertaken clearly establishes the wide variation in distribution and depth of credit among districts in Tamil Nadu. Our future work will focus on better understanding the factors that lead to this difference, both from demand and supply sides.

(Note: Please note that the figures in this post are interactive files, you can hover or zoom in on these for more details.)

1 – CAGR = [(Amount / Principal) ^ (1/ Number of years)]-1
2 – http://www.econ.nyu.edu/user/debraj/Courses/Readings/BarroGrowth.pdf
3 – Credit-to-GDP of Kancheepuram has increased from 15% in 2004-05 t0 20% in 2011-12.


Exploratory Analysis of Credit-to-GDP Variable for Tamil Nadu

By Nishanth K & Irene Baby, IFMR Finance Foundation

In the previous blog posts of this series, we had outlined that the various aspects of financial development like depth (Credit to Gross Domestic Product (GDP) ratios), and access (per cent of population with bank accounts in urban and rural areas, and distribution of payment access points per 10,000 eligible persons) vary not only between Indian states, but also considerably between different districts. In this context, in order to assess the depth of credit for each district, it is important to study and understand the trends in district level credit depth defined as the ratio of quantum of credit outstanding in district to district GDP.

This post will focus on understanding the trend in credit-to-GDP variable for Tamil Nadu as a state, and for the 32 districts in Tamil Nadu for the time period 2004-05 to 2011-12. We also present evidence for increasing median credit depth for Tamil Nadu.

Figure 1: Trend in Credit-to-GDP (Current Prices) for Tamil Nadu from 2004-05 to 2011-12

Observation 1: In 2011-12, the Credit-to-GDP ratio for Tamil Nadu was approximately 71.4% (Tamil Nadu’s credit-to-GDP [at constant prices] was 110% in 2011-12). However, this high value of overall credit depth in Tamil Nadu was mostly driven by the very high levels of credit depth in urban areas like Chennai (561%) and Coimbatore (131%). Conversely, districts such as Thiruvallur and Vellore had low levels of credit depth with 15% and 18% respectively. Hence, there was a wide variation in the levels of credit depth within the districts of Tamil Nadu as of 2011-12.

The mapping visualisation below (Figure 2) captures this variation in credit depth for all districts of Tamil Nadu between 2004-05 and 2011-12. It is important to note that Tirupur district was part of Coimbatore district from 2004-05 to 2009-10. Ariyalur was under Perambalur from 2004-05 to 2006-07. For these districts, the values of parent districts have been applied for time periods in which they did not exist.

Figure 2: Credit-to-GDP Ratio for Districts of Tamil Nadu between 2004-05 and 2011-12 (In Percentage)

Please select the year to load the corresponding data map for that particular year.

Figure 3 below is a interactive box plot which captures the variations in credit-to-GDP (current prices) ratio from 2004-05 to 2011-12 (with Chennai and Coimbatore being outlier districts have been left out for lucidity in constructing this box plot).

Figure 3: Box Plot of Credit-to-GDP (Current Prices) for Districts of Tamil Nadu between 2004-05 and 2011-12

Observation 2: As is evidenced from the plot above, the median credit depth (shown by the horizontal line inside the box for each year) in the state is increasing (with an anomaly being the year 2010-11 where credit-to-GDP ratio marginally reduced to 0.26 from 0.27 in 2009-10). Simply put, this means that there is a movement towards higher credit depth in the economy of Tamil Nadu from 2004-05 to 2011-12. However, this upward trend could also be primarily caused by the exponential credit growth in urban centres such as Chennai and Coimbatore (as discussed earlier).

Observation 3: Another important observation is that the lowest quartile of credit-to-GDP is increasing from 2004-05 (0.1) to 2011-12 (0.15). While this implies a higher minimum credit depth in 2011-12 than in 2004-05, we cannot, however, ascertain the distribution effects of this increased depth.

The next blog post in this series will discuss the trends in growth of credit-to-GDP variable and highlight important district specific trends.


Measures of Financial Depth and their Limitations

By Nishanth K & Irene Baby, IFMR Finance Foundation

In the previous post of the blog series on financial depth, we attempted to understand the nature of the relationship between financial depth and economic growth. The natural question that follows is how to measure financial depth. This post answers that question by:

  1. Introducing various measures of financial depth
  2. Discussing the limitations of these conventionally used measures

Measures of Financial Depth

Early literature used the ratio of total bank credit to Gross Domestic Product (henceforth GDP) as the indicator of financial depth[1]. However, private firms are more likely to stimulate growth through their risk evaluation and corporate control capacities than credit to the government or state-owned enterprises[2]. Hence, ratio of domestic credit lent to the private sector to GDP has now emerged as the most commonly used indicator of financial depth of an economy. The private credit, therefore, excludes credit issued to governments, government agencies, and public enterprises. It also excludes credit issued by central banks[3].

A recent research paper by the IMF outlines the various measures of financial depth[4]:


However, in the context of developing countries, it is reasonable to assume that pension fund, mutual fund and insurance markets are not adequately developed to be used as measures of financial depth. Therefore, private-sector credit-to-GDP has remained the most widely used measure of financial depth.

Limitations of a Credit-to-GDP Ratio as a measure of Financial Depth

The above discussed measures are not without their limitations. This section discusses why these measures are inadequate to understand excessive provision of credit within an economy. Given below are the major limitations[5]:

1) Ignores heterogeneity in credit demand across countries: These measures ignore that the demand for credit across countries varies with their levels of economic, financial and institutional development. The main drivers of this heterogeneity in demand across countries are differences in financial depth, access to financial services, use of capital markets, efficiency and funding of domestic banks, central bank independence, the degree of supervisory integration, and experience of a financial crisis.


Figure 1: Bank Credit Can be a Misleading Measure of Financial Depth and Access – Korea and Vietnam have similar values of banking depth—a private credit to GDP ratio. However, they differ, by a large amount, in terms of access. The use of banking accounts is virtually universal in Korea, but in Vietnam only one-quarter of adults have a bank account.[6]

2) Assumes unit-elastic relationship between credit demand and income (as proxied by GDP) and price (as proxied by GDP Deflator): Depth indicators assume that a one-percent increase in GDP or the GDP Deflator results in a corresponding one-percent increase in the demand for credit. This assumption is violated, particularly for developing countries because of the varying usage of credit in economic transactions due to different levels of development.

3) Assumes that financial depth variables are stationary: A stationary variable is one whose statistical properties such as mean, variance and autocorrelation are constant over time. Credit-to-GDP ratio is empirically shown to be a trending variable, which means that credit grows considerably faster than the nominal GDP.

4) Fails to capture the effect of financial deepening at the sub-national levels: As conventionally used measures are constructed using readily available macroeconomic data, they are reflective of macroeconomic outcomes in financial deepening. In other words, they do not necessarily capture how well finance accomplishes various functions at various levels of the economy.

Consider the case of measuring financial depth in India where depth of financial services varies significantly amongst different regions. Ideally, the value of INR 1 of credit in a credit starved district should be greater than its value in a district which is relatively well-penetrated by credit. Inclusix published by CRISIL, which is an objectively measured index of financial inclusion provides a score for every district (out of 100), with 100 indicating a fully financially included district.[7] Scores in the index range from 96.2 for Pattanamthitta district in Kerala to a low of 5.5 for Kurung Kumey district in Arunachal Pradesh. This suggests that there is an inherent need for accounting for sub-national variations for one to appropriately measure the level of financial depth in an economy[8].

One measure of financial depth that does away with the limitations of conventional measures discussed above is ‘Equilibrium Credit’[9]. It is an important concept because it helps identify both excessive and deficient credit provision at sub-national levels of the economy. In other words, equilibrium credit helps to identify benchmarks against which the credit provision in various levels of the economy is judged to be excess or not. The next post in the series will discuss the idea of equilibrium credit in detail.

[1] King and Levine 1993. Finance and Growth: Schumpeter Might Be Right

[2] Ibid

[3] Global Financial Development Report: Key Terms Explained. 2012. http://go.worldbank.org/E9RQKCP9J0

[4] Sahay et al. 2015. Rethinking Financial Deepening: Stability and Growth in Emerging Markets

[5] Buncic and Melecky. 2014. Equilibrium Credit: The Reference Point for Macroprudential Supervisors

[6] Sahay et al. 2015. Rethinking Financial Deepening: Stability and Growth in Emerging Markets

[7]Report, Committee on Comprehensive Financial Services for Small Businesses and Low Income Households. 2014, submitted to Reserve Bank of India. https://rbidocs.rbi.org.in/rdocs/PublicationReport/Pdfs/CFS070114RFL.pdf

[8] ibid

[9] Same as footnote 3