The Tale of Indian PSBs — Non-Performing Assets and Capital Erosion
In the recent past, several PSBs have been severely hit with a high level of capital erosion as a consequence of bad debts and defaults. In 2018–19, the gross amount of Non-Performing Assets (NPAs) as a proportion of advances was at 11.6% for PSBs, while the number was just 5.3% for the new private sector banks (Source: RBI — DBIE).
The collapse of a PSB as a consequence of the NPA crisis could have devastating economic ramifications. The government is hence committed to addressing the issue — palliatively at the least. Capital Infusion has been one of the primary (palliative) solutions that the government has implemented to address this situation, wherein the government invests additional capital in the PSBs to bolster equity.
In most cases, the money for investing in equity does not directly come from the government’s coffers, but instead comes from the banks themselves. To achieve this, PSBs use their deposits to subscribe to non-transferable, HTM (held to maturity) recapitalisation bonds floated by the government. The capital raised from these bonds is in turn invested in the banks as equity, by the government. This enables the PSBs to move some of their deposits (liabilities) to low-risk assets (technically government bonds) and equity. The increase in equity implies that banks can expand their lending activities while remaining within the RBI stipulated capital requirement of 8% Capital to Risk Assets Ratio (CRAR). Ideally, the increased lending activities of PSBs should result in higher dividends and yields to the government, which would eventually be used to pay off the bonds.
In the 2021–22 budget speech, the Finance Minister proposed an additional ₹200 billion to be infused into the PSBs to ensure that they meet the capital requirements (8% CRAR), and for enabling them to expand their lending activities.
Measurement — Keeping a Check on Moral Hazard
The continual monitoring of PSBs is vital to ensure that there is no moral hazard induced due to capital infusion. In the absence of monitoring protocols, PSBs might have no tangible incentive to improve their asset quality or to streamline operations. This results in taxpayers’ money being treated as a free source of capital to cover up for the bad loans that the PSBs had extended. To avoid such a situation, it is important to develop simple yet comprehensive measures to holistically assess the profitability of PSBs and to gauge their asset quality.
To this end, I constructed a non-parametric asset quality index and profitability index using Principal Component Analysis (PCA), to rank PSBs along these two dimensions. The PCA based indices are constructed using data of indicators reported as of 30th September 2020, which are provided by the RBI in their ‘Database of Indian Economy’ portal. I restrict the indicators to only reflect domestic operations as not all PSBs have a comparable global presence. The list of indicators used for profitability and asset quality are listed in Table 1. (See the appendix for more details on the PCA methodology).
The Results — Signs of Clustering
The following graph plots PSBs on the index of asset quality and profitability.
The results clearly indicate that SBI is the best performing bank in terms of both asset quality and profitability.
Punjab National Bank, Union Bank of India, and Bank of Baroda are closely clustered at a distant second place. Indian Bank, Bank of India, and Canara Bank are clustered at third place.
Indian Overseas Bank, Bank of Maharashtra, Central Bank of India, and UCO Bank are clustered at fourth place. Punjab and Sindh Bank is a distant last in the ranking.
It is important to note that despite the recent bank amalgamations, PNB, BoB, and Union Bank are still in the second position.
Which cluster(s) of PSBs truly deserve further capital infusion, have some failed to demonstrate improvement?
How does the Ranking System help?
Capital infusion is absolutely necessary for the continuance of PSBs in cases where the capital requirements stipulated by the RBI (i.e. 8% CRAR) are not met. Hence in these situations, a performance-based infusion strategy is out of the question. However, in other cases where there is scope for performance-based infusions, the PCA based ranking provides evaluators with a data-driven (i.e. non-parametric) estimate of the relative performance of the PSBs. The ranking could be an important source of evidence to determine who can be rewarded with further capital infusion to expand their businesses. Moreover, such quantitative approaches would bring a certain degree of consistency to performance-based capital infusion policies.
From a management perspective, the understanding of relative positions of the PSBs could enable the RBI and the government to understand best practices of the more successful banks, which could then be replicated by other banks.
The Road Ahead
The government cannot perennially provide PSBs with the necessary capital. Guaranteed government support would only create a moral hazard where PSBs continue to accumulate bad debts, without any course correction. The sustenance of PSBs should therefore depend on their ability to use the new capital injected to improve profitability and asset quality. Eventually, they should prove to be viable businesses which can attract non-government investors, and meet their capital needs by raising investments from the market at large. Hence, the consistent monitoring of PSBs, coupled with a mix of performance-based carrots and sticks would definitely enable these banks to become viable investment opportunities in the market at large.
Author: Akshay Natteri Mangadu
Originally Published on February 2, 2021, on LinkedIn.
Appendix — PCA Methodology
I run two PCAs, one for the correlation matrix of the profitability variables and the other for the correlation matrix of asset quality variables. I project the vector of profitability variables of each bank onto the space spanned by the first principal component of the profitability-correlation matrix. Similarly, the vector of asset quality variables of each bank is projected onto the space spanned by the first principal component of the asset quality-correlation matrix.