Unveiling Asset Quality Metrics of Banks: Exploring the Myth of Net NPA Ratio and Dispelling Provisioning Misconceptions

Conceptual Framework

\'Asset quality\' is a feature of the credit portfolio of banking companies. It entails examining banks\' loans to assess the level and size of credit risk associated with their loans. Loans are considered as high-quality assets if they can be converted into cash easily and immediately with no or little loss of value. Usually, loans carry a \'normal to higher\' degree of risk. Even the collaterals given may possess certain risks. The apex bank of the country (Reserve Bank of India, RBI) directs the banking companies to assess the actual level of bad loans and to classify a loan account as NPA if interest/principal is not paid in 90 days from the due date.

The loans provided by banks are commonly referred to as their \'assets\' as they generate a stream of interest income. However, in addition to \'loans and advances\', the term \'assets\' includes items like equipment, intangibles, etc. Despite this distinction, this paper uses the term \'assets\' to specifically refer to \'loans\' disbursed by banks. Therefore, \'asset quality\' in this context refers to the quality of loans, i.e., the timeliness with which borrowers pay interest and repay borrowed sums as per the credit terms.

When a loan asset ceases to generate interest income for the lender-banker, it is classified as a non-performing asset (NPA). Operationally, when a borrower fails to make the scheduled payment of interest or principal for a period of 90 days (from the due date), the account is classified as NPA. The NPAs are classified into three categories: (i) sub-standard assets (if a loan account is in NPA for 12 months or less), (ii) doubtful assets (if it remained NPA for more than 12 months), and (iii) loss assets (if the loan is considered/identified \'uncollectible\').

There are several reasons why asset quality deteriorates and transitions into an NPA. Natural calamities like COVID-19, economic recession frequent strikes impacting production, faulty projections, wilful defaults, target-driven pressures, etc., exert a substantial influence on the prevalence of bad loans (Sanjeev, 2007). Research spanning 76 countries indicates that \'corruption\' exacerbates bad loans in the banking sector, leading to the misallocation of banks\' funds from sound projects to risky ventures, further deteriorating asset quality (Park, 2012). Additionally, diverting borrowed funds for unauthorized purposes also contributes to the rising NPAs (Richard, 2011).

However, the RBI requires the banks to set aside money to cover potential losses on these NPAs at the specified rates. The money so set aside is called, \'Provisioning\' - a proactive measure to mitigate credit risk and ensure financial resilience. Adequate Provisioning strengthens the balance sheets of banks and maintains the investor/ depositor confidence. Although it is construed as a sign of mismanagement, it is fundamental to risk management and regulatory compliance reflecting prudent financial stewardship.

The rate of Provisioning depends on the category of loan/NPA - 0.25% on \'standard assets\' to 100% on \'loss assets\'. Both NPAs and Provisioning indicate a deterioration in asset quality, manifested by a decrease in the proportion of standard assets in gross advances and a consequent rise in the proportion of NPAs in gross advances, necessitating increased Provisioning.

Further, the amount of Provisions held is subtracted from the amount of Gross NPAs (GNPAs i.e., aggregate of sub-standard assets, doubtful assets, and loss assets) to arrive at the amount of Net NPAs (NNPAs). Based on the amounts of NPAs and Advances, two NPA Ratios are computed:

GNPA Ratio = [GNPAs / Gross Advances] * 100 ...(1)

NNPA Ratio = [NNPAs / Net Advances] * 100 ...(2)

where, Net Advances = (Gross Advances - Repayments of Principal)

Magnitude of NPAs - An Overview

The amount of GNPAs of scheduled commercial banks (SCBs, public sector banks, private sector banks, and branches of foreign banks are considered here) increased from Rs. 593.73 billion up to 31 March 2005 to Rs. 10,387.86 billion by 31 March 2018, representing an increase of Rs. 9,794.13 billion or 16.50 times. However, it declined to Rs. 7,354.87 billion by 31 March 2022 due to recovery, write-offs, etc., which is a commendable achievement. Similarly, the GNPA Ratio rose from 4.90% at the end of 2004-05 to 11.20% by the end of 2017-18 but decreased to 5.80% by 31 March 2022.

Similarly, the amount of NNPAs increased from Rs. 217.55 billion at the end of 2004-05 to Rs. 5,204.01 billion by 31 March 2018, representing an increase of Rs. 4,986.46 billion or 22.92 times. However, it declined to Rs. 2,015.01 billion by the end of 2021-22. Even the NNPA Ratio increased from 0.76% in 2007-08 to 5.62% in 2016-17 but declined to 2.70% in 2018-19. This clearly shows that initially, the NPA Ratios registered a continuous increase due to deterioration in asset quality, followed by a continuous reduction.

However, the deterioration in asset quality (i.e., a decline in the Standard Assets Ratio) has multifarious ill effects on the performance of banks (Figure 1).

Figure 1: Implications of Asset Quality Deterioration

• Decrease in Standard Assets Ratio,
  • Increases: GNPAs, GNPA Ratio, Provisioning, NNPAs, NNPA Ratio
  • Reduces: Credit Recycling Capacity, Profit, Profitability, etc

There is a significant negative relationship between NPAs and the profitability of banking companies - lower asset quality, higher the NPAs (and provisioning), leading to lower profit, return on assets (RoA), and return on equity (RoE), and vice-versa (Kadioglu et al., 2017). Additions to loan-loss reserves are charged to the income statements of banks, crushing their earning power (Eavis, 2008). Even the different sectors of the economy are affected by NPAs as the smooth flow of credit is disrupted when the cycle of lending-repaying-borrowing is broken.

Objectives and Methodology

The important objectives are, (i) to point out the deficiencies in the NNPA Ratio recommended by regulators for assessing the asset quality of banking companies, and propose a new practical tool for evaluating bank\'s asset quality, and (ii) to examine the adverse implications of provisioning on the financial results of banks, which are not currently highlighted.

The second objective of the study is also addressed with a null hypothesis, namely H0: There exists no significant relationship between the amounts of provisioning made by the SCBs and their profits. Besides ratios, a \'correlation\' test is carried out to test the hypothesis. Data essential for this study is gathered from RBI reports and other secondary sources.

Shortcomings of NNPA Ratio - An Evaluation

For assessing the asset quality of banks, regulatory authorities have recommended the use of the NNPA Ratio. The Padmanabhan Committee endorsed the NNPA Ratio for this purpose (Reserve Bank of India, 1995). Additionally, the central bank of the country incorporates the NNPA Ratio as a key measure in constructing the Banking Stability Map and Indicator (Reserve Bank of India, 2020). In its 2021 \'Prompt Corrective Action Framework,\' the RBI specifically identifies the NNPA Ratio as the primary parameter for evaluating asset quality (Reserve Bank of India, 2021). Moreover, many researchers have employed the NNPA Ratio as a tool for assessing asset quality (Gowda, Inchara P. M. 2019). From the above, it is evident that more emphasis is placed on the NNPA Ratio as the tool for measuring and evaluating the asset quality of banking companies.

However, this Ratio seems to be an inappropriate tool for evaluating asset quality as analyzed below.

As previously explained, the NNPA Ratio is calculated using the amount of NNPAs in the numerator of Formula-2. This amount of NNPAs is affected by various factors, including the amount of Provisioning made - in the opposite direction - a higher amount of Provisioning leads to a lower amount of NNPAs (assuming other factors remain constant), and vice versa. Therefore, a decrease in the amount of NNPAs (caused by an increase in Provisioning) may result in a decrease in the NNPA Ratio, falsely suggesting an improvement in asset quality. True improvement in asset quality is observed only when the reduction in NNPAs and NNPA Ratio is a result of the actual recovery of amounts due.

Further, the reduction in the amount of GNPAs should stem from recovery rather than write-offs. This is crucial because, one of the key factors behind the decrease in NPAs of SCBs in India is the rise in NPAs being written off, not an enhancement in the recovery performance (Kalyanasundaram, 2020). For instance, in 2021-22, PSBs wrote off Rs. 1,197.13 billion of loans, while the reduction (presumably from recovery) was only Rs. 946.34 billion. This indicates that the decrease in NPAs by the end of 2021-22 (in comparison to the beginning of the same period) is predominantly due to write-offs rather than recoveries. The details of loans written off and the reduction amounts (assumed to be from recoveries) are illustrated in Figure 2 below, showcasing that the decline in NPAs from 2017-18 onwards is primarily a result of write-offs rather than recoveries.

Figure 2: Loan - Recovery and Write-off (Rs. billion)

[Graph showing Loan Written-off (INR Bn) and Reduction/Recover (INR Bn) from 2012-13 to 2021-22. Key data points: Loan Written-off peaked at 1,831.68 in 2018-19; Reduction/Recover peaked at 1,278.35 in 2018-19.]

Although NPA write-offs reduce the amounts of Gross and NNPAs, and their relative shares, they do not contribute to the improvement in the asset quality of banking companies. It is clear that (i) authorities use the NNPA Ratio as a key measure for assessing the asset quality of banks, and (ii) the NNPA Ratio fails to properly evaluate the asset quality of banks. This is because the decrease in the NNPA Ratio also arises from subtracting Provision from the GNPAs to determine the amount of NNPAs, which influences the NNPA Ratio. Interpreting this reduction in the NNPA Ratio as an enhancement in the asset quality of banks is inaccurate. Therefore, an Adjusted GNPA Ratio is suggested.

Adjusted GNPA Ratio - Suggested

In light of the loophole in the NNPA Ratio, we suggest making a few adjustments to the GNPA Ratio. We propose naming this adjusted ratio as the \'Adjusted GNPA Ratio\' (GNPA RatioAdj), where two modifications are incorporated into the GNPA Ratio to derive the GNPA RatioAdj as outlined below:

(a) The amount of Provisioning may be added back to the amount of NNPAs (as in Formula-2) to arrive at the Adjusted amount of GNPAs.

(b) The amount of Gross Advances (i.e., denominator of GNPA Ratio, Formula-1) may be adjusted to account for the amount of repayments of principal received. The resulting figure would be the \'net advances\' as used in the denominator of the NNPA Ratio (Formula-2).

Adjusted GNPA Ratio may be calculated by dividing the aggregate amount of \'NNPAs and Provisioning\' by the \'Net Advances\' as shown below:

GNPA RatioAdj = [(NNPAs + Provisioning) / Net Advances] * 100 ...(1)

The above formula seems to offer a more comprehensive assessment of asset quality for banks than Gross and NNPA Ratios. In this context, the Gross and NNPA Ratios (currently used as per Formulas 1 and 2) and the outcomes derived from the GNPA RatioAdj for PSBs of over 10 years are computed and tabulated below (Table 1).

It is evident from Table 1 that the order of ratios is as follows: \'GNPA Ratio > GNPA RatioAdj > NNPA Ratio\'. The NNPA Ratio appears lower due to the deduction of Provisioning from the GNPAs, leading to a potential underestimation. This misrepresentation could give stakeholders a false sense of improved asset quality. Conversely, employing the proposed GNPA RatioAdj provides a more accurate assessment of asset quality.

Table 1: Gross and NNPA Ratios, and Adjusted GNPA Ratio of PSBs

Source: (Reserve Bank of India, 2023)

Implications of Provisioning - An Analysis of Misconception

There is a misconception about the implications of NPAs and Provisioning on the profit and profitability of banking companies. As stated earlier, Provisioning is a charge to the Statement of Profit and Loss causing a reduction in the amount of profit for that year, which, in turn, lowers the profitability of banks. Of course, through Provisioning, banks build a buffer to take care of write-offs, haircuts (i.e., Loan Amount Due - Loan Amount Recovered), etc. When banks write off their lost assets, these are charged against the accumulated Provisioning created. It is important to note that, in the case of default,

(a) GNPAs do not amount to the actual loss for the banks as the Provisioning on the unpaid debts is not subtracted.

(b) NNPAs amount to the actual loss for the banks as the Provisioning on the unpaid debts is subtracted.

However, this line of argument appears to be based on a misconception about the Provisioning accumulated over the years. The Provisioning buffer is built gradually by charging it against income annually, resulting in a decrease in yearly profits. In the scenario where there are no GNPAs (i.e., when GNPA Ratio = 0), the Provisioning amount would have been significantly lower, given the nominal rate of Provisioning on Standard Assets (e.g., at 0.25%), thereby minimizing the negative impact on reported profits. This could have allowed banking companies to either report higher profits or lower losses (Table 2).

After absorbing the amounts of Provisions, PSBs reported profits for five years (2012-13 to 2014-15 and 2020-21 to 2021-22), and losses for another five years (2015-16 to 2019-20). If they had reduced the NPAs to zero, they would have reported profit for all 10 years as the amount of provision charged is higher than the amount of loss/profit reported. Most importantly, Provisioning is not an item of appropriation of profit but a mandatory charge to the Profit and Loss.

Table 2: Impact of Provisioning on the Profit of PSBs

\'r\' value (relationship between Provision and Profit after Provision) = -0.92603

Source: (Reserve Bank of India, 2023)

For testing the null hypothesis, H0. There exists no significant relationship between the amounts of provisioning made by the SCBs and their profits, a \'correlation\' test is carried out for the type of relationship between the amounts of Provision and profit. The critical value associated with the degree of freedom (df) = 8 (n=10) is ±0.632 (i.e., -0.632 to 0.632). In this case, the calculated \'r\' value of (-) 0.92603 (Provision Vs Profit) falls outside the acceptance region of ±0.632. Therefore, a strong negative correlation exists between provision and profit.

Conclusion

It is obvious from the above that (i) the NNPA Ratio fails to evaluate the asset quality of banking companies, and (ii) the GNPA RatioAdj is a better alternative for evaluating the asset quality of banking companies. It is hoped that the regulatory authority, the RBI, takes note of this suggestion and gives effect to the same by modifying its PCA Framework of 2021 appropriately.

Furthermore, to keep stakeholders informed about the adverse implications of Provisioning on profits, it is recommended that banking companies disclose in the \'Notes to Accounts\' the extent to which profit has decreased due to Provisioning. Alternatively, the central bank may direct banking companies to make this disclosure.

Banks are required to adhere to regulated provisioning requirements, which involve accurately assessing and setting aside funds for potential losses. This transparency in financial reporting and investment activities demonstrates the banks\' commitment to responsible practices, building trust among stakeholders and enhancing the credibility of banks as reliable financial intermediaries.