Thank you dear subscribers, we are overwhelmed with your response.
Your Turn is a unique section from ThePrint featuring points of view from its subscribers. If you are a subscriber, have a point of view, please send it to us. If not, do subscribe here: https://theprint.in/subscribe/
“The only function of economic forecasting is to make astrology look respectable” – Erza Solomon
NPAs forecasting
A Financial Stability Report (FSR) is a bi-annual survey by a nation’s central bank to highlight the vulnerability and risks in the existing financial system. This document is prepared from inputs from various regulatory bodies to gauge the intertwined risks across different sectors and asset classes. Predictions are a significant part of these stability reports. Banks’ bad assets have far-reaching implications for the economy; thus, forecasting non-performing assets (NPA) is a prerequisite for determining the financial stability of financial institutions. The Gross NPAs (GNPAs) data are given in different economic scenarios- good, bad, and ugly- in layperson’s terms. In the financial environment, they are termed as “baseline,” “medium,” and “severe” as per the macro-economic environment predictions. The baseline is the regular course in which the economy will grow, the medium is a little deviation from the ordinary course, and a severe stress level is an extraordinary circumstance. Table 1 gives the compiled FSR predictions under all three stress levels of the macroeconomy across India’s different scheduled commercial banks. The actual figures for GNPAs for the same segment of banks are provided in Table 2.
Table 1: FSR predictions of NPAs (%)
| All Banks | Public Sector Banks | Private Banks | Foreign Banks | |||||||||||
| FSR Year | FSR No | Projection Year | Baseline | Medium | Severe | Baseline | Medium | Severe | Baseline | Medium | Severe | Baseline | Medium | Severe |
| Dec-23 | 28 | Sep-24 | 3.1 | 3.6 | 4.4 | 4.1 | 4.5 | 5.1 | 2 | 2.6 | 3.6 | 1.4 | 1.5 | 1.8 |
| Jun-23 | 27 | Mar-24 | 3.6 | 4.1 | 5.1 | 4.8 | 5.3 | 6.1 | 2.1 | 2.7 | 3.8 | 2 | 2.1 | 2.6 |
| Dec-22 | 26 | Sep-23 | 4.9 | 5.8 | 7.8 | 6.4 | 7.4 | 9.4 | 3.1 | 3.8 | 5.8 | 2.2 | 2.7 | 4.1 |
| Jun-22 | 25 | Mar-23 | 5.3 | 6.2 | 8.3 | 7.1 | 8.4 | 10.5 | 3 | 3.7 | 5.7 | 2 | 2.5 | 4 |
| Dec-21 | 24 | Sep-22 | 8.1 | 8.7 | 9.5 | 10.5 | 11 | 11.9 | 5.2 | 5.4 | 5.9 | 3.9 | 4.4 | 5.1 |
| Jul-21 | 23 | Mar-22 | 9.8 | 10.3 | 11.2 | 12.5 | 13.0 | 13.9 | 5.8 | 6.0 | 6.4 | 4.9 | 5.3 | 5.9 |
| Jan-21 | 22 | Sep-21 | 13.5 | 14.1 | 14.8 | 16.2 | 16.8 | 17.6 | 7.9 | 8.2 | 8.8 | 5.4 | 6 | 6.5 |
| Jul-20 | 21 | Mar-21 | 12.5 | 13.5 | 14.2 | 15.2 | 15.5 | 15.9 | 7.3 | 7.7 | 8.3 | 3.9 | 4.5 | 5.1 |
| Dec-19 | 20 | Sep-20 | 9.9 | 10.2 | 10.5 | 13.2 | 13.4 | 13.5 | 4.2 | 4.8 | 5.4 | 3.1 | 3.6 | 4.2 |
| Jun-19 | 19 | Mar-20 | 9 | 9.2 | 9.6 | 12 | 12.1 | 12.2 | 3.2 | 3.8 | 4.4 | 2.9 | 3.5 | 4.1 |
| Dec-18 | 18 | Sep-19 | 10.2 | 10.5 | 10.8 | 14.3 | 14.4 | 14.5 | 3.3 | 3.8 | 4.5 | 3 | 3.5 | 4.1 |
| Jun-18 | 17 | Mar-19 | 12.2 | 12.7 | 13.3 | 16.3 | 16.7 | 17.3 | 4.2 | 4.7 | 5.3 | 3.8 | 4.3 | 4.8 |
| Dec-17 | 16 | Sep-18 | 11.1 | 11.5 | 12.1 | 15.3 | 15.7 | 16.2 | 3.7 | 4.2 | 4.8 | 3.4 | 3.7 | 4.2 |
| Jun-17 | 15 | Mar-18 | 10.2 | 10.7 | 11.2 | 13.2 | 13.7 | 14.2 | 3.7 | 4.2 | 4.8 | 4.3 | 4.7 | 5.2 |
Table 2: Gross NPAs (%)
| Year | All | PSBs | Pvt | Foreign |
| 2016-17 | 9.3 | 11.7 | 4.1 | 4 |
| 2017-18 | 11.2 | 14.6 | 4.7 | 3.8 |
| 2018-19 | 9.1 | 11.6 | 5.3 | 3 |
| 2019-20 | 8.2 | 10.3 | 5.5 | 2.3 |
| 2020-21 | 7.3 | 9.1 | 4.8 | 3.6 |
| 2021-22 | 5.8 | 7.3 | 3.8 | 2.9 |
| 2022-23 | 3.9 | 5.0 | 2.3 | 1.9 |
Why the divergence?
It can be seen from both tables that there is a deviation between the predicted and actual values of NPAs. The deviation is drastic, especially for the public sector banks. The main reason for the NPA differential can be pointed out to the loan write-offs done by the government amounting to almost Rs 14 lakh crores from 2015 to 2023 (Table 3). A loan written off is where a bank removes the non-performing loan (NPL) from its assets by declaring its loss. Even if there is a collection on this loan, the profits generated later are the “additional” income of the bank. Loan written-offs impact the bank in two ways: the return on equity (ROE) and capital adequacy ratio (CAR).
Table 3: Loan written off over years
| Year | Written Off |
| 2022-23 | 2,09,000 |
| 2021-22 | 1,74,966 |
| 2020-21 | 2,02,781 |
| 2019-20 | 2,34,170 |
| 2018-19 | 2,36,265 |
| 2017-18 | 1,61,328 |
Table 4: Loan written offs by banks
| Sr. No | Bank Name | (Rs crores) | No of wilful defaulter accounts |
| 1 | SBI | 79,271 | 1881 |
| 2 | PNB | 41,353 | 2143 |
| 3 | Union Bank | 35,623 | 1747 |
| 4 | BOB | 22,754 | 2203 |
| 5 | IDBI | 24,192 | 335 |
| 6 | All private | 30,809 | 1822 |
| TOTAL | 203,193 |
It can be deliberated whether the PSB’s internal performance has increased due to enhanced due diligence in credit risk management viz the external factors such as bad assets being unburdened from its balance sheet. Astonishingly, brushing off bad loans under the carpet has made the investors exuberant. Corporate loans are given latitude, especially in the gems & jewellery segment. The author suggests that there should be a discussion between RBI and the Government of India (GOI), and the tentative amount to be written off can be restricted collectively at a certain level. Even if it is an implausible thought, the higher the corporate loan segment waiver amount should have implications for GOI’s dividend payout by the banks and RBI.
Estimations and stability
In 2017, a time before COVID-19, the GNPAs were predicted to be between 11% and 12% for severe stress levels. 2023, they are expected to be between 4% and 5%. The current global environment is volatile due to the aftereffects of the pandemic, the Russia-Ukraine war, the Israel-Gaza war, China’s slowdown, and the repercussions of inflation and interest rates. Is our financial structure becoming resilient inspite of the global environment? Or are we failing to see the signs of the worldwide turmoil and capture the risk appropriately?
With due respect to the experts drafting the FSR, there is a need to look at the FSR sensitivity and scenario inclusions of shock parameters. FSRs lack a forward-looking perspective, and capturing systemic risk interconnectivity is missing (Muñoz et al., 2012). Secondly, though this is not my area of expertise, there can be a relook at the model estimators, and non-linear models like a random forest, which give 76% accuracy, can be used (Abdullah et al., 2023).
The off-loading of bad assets has only saved banks from temporarily building higher capital (CAR). It does not necessarily imply bank resilience and sets poor governance precedence without taking responsibility for the inferior assets. As per the World Bank, financial instability can severely shake the confidence in the financial and economic system as transactions in the real economy are made through the financial system. A CEPR report 2023 pointed out that bank management mistakes do not remain idiosyncratic but have an external impact. This highlights the need for better supervisory insights to arrive at banks’ perceived risk and make the system genuinely robust.
These pieces are being published as they have been received – they have not been edited/fact-checked by ThePrint.