Methodologies - RBI - Reserve Bank of India
Methodologies
2.1 Scheduled Commercial Banks (a) Banking stability map and indicator The banking stability map and indicator present an overall assessment of changes in underlying conditions and risk factors that have a bearing on the stability of the banking sector during a period. The six composite indices represent risk in six dimensions - soundness, asset quality, profitability, liquidity, efficiency and sensitivity to market risk. Each composite index is a relative measure of risk during the sample period used for its construction, where a higher value would mean higher risk in that dimension. The financial ratios used for constructing each composite index are given in Table 1. Each financial ratio is first normalised for the sample period using the following formula: ![]() where Xt is the value of the ratio at time t. If a variable is negatively related to risk, then normalisation is done using 1-Yt . Composite index of each dimension is then calculated as a simple average of the normalised ratios in that dimension. Finally, the banking stability indicator is constructed as a simple average of these six composite indices. Thus, each composite index and the overall banking stability indicator takes values between zero and one. (b) Single factor sensitivity analysis ─ Stress testing As part of quarterly surveillance, stress tests are conducted covering credit risk, interest rate risk, liquidity risk etc. and the resilience of commercial banks in response to these shocks is studied. The analysis is done on individual SCBs as well as on the system level. I. Credit risk (includes concentration risk) To ascertain the resilience of banks, the credit portfolio was given a shock by increasing GNPA ratio for the entire portfolio. For testing the credit concentration risk, default of the top individual borrower(s) and the largest group borrower(s) was assumed. The assumed increase in GNPAs was distributed across sub-standard, doubtful and loss categories in the same proportion as prevailing in the existing stock of NPAs. However, for credit concentration risk (exposure based), the additional GNPAs under the assumed shocks were considered to fall into sub-standard category only and for credit concentration risk (based on stressed advances), stressed advances were considered to fall into loss category. The provisioning requirements were taken as 25 per cent, 75 per cent and 100 per cent for sub-standard, doubtful and loss advances respectively. These norms were applied on additional GNPAs calculated under a stress scenario. As a result of the assumed increase in GNPAs, loss of income on the additional GNPAs for one quarter was also included in total losses, in addition to the incremental provisioning requirements. The estimated provisioning requirements so derived were deducted from banks’ capital and stressed capital adequacy ratios were computed. II. Sectoral credit risk To ascertain the sectoral credit risk of individual banks, the credit portfolios of a particular sector was given a shock by increasing GNPA ratio for the sector. Sector specific shocks based on standard deviation (SD) of GNPA ratios of a sector are used to study the impact on individual banks. The additional GNPAs under the assumed shocks were considered to fall into sub-standard category only. As a result of the assumed increase in GNPAs, loss of income on the additional GNPAs for one quarter was also included in total losses, in addition to the incremental provisioning requirements. The estimated provisioning requirements so derived were deducted from banks’ capital and stressed capital adequacy ratios were computed. III. Interest rate risk Under assumed shocks of shift in the INR yield curve, there could be losses on account of the fall in value of the portfolio or decline in income. For interest rate risk in the investment portfolio: AFS, FVTPL (including HFT book) and HTM categories, a duration analysis approach was considered for computing the valuation impact (portfolio losses). The portfolio losses on these investments were calculated for each time bucket of AFS, FVTPL (including HFT book) and HTM categories based on the applied shocks. These estimated losses were reduced from banks’ capital and market risk weighted losses from RWA to arrive at stressed capital ratios. Interest rate risk of banks refers to the risk to a bank’s capital and earnings arising from adverse movements in interest rates that affect bank’s books. The impact on earnings is measured using the traditional gap analysis (TGA) and the capital impact is measured by duration gap analysis (DGA). The focus of TGA is to measure the level of a bank’s exposure to interest rate risk in terms of the sensitivity of its net interest income (NII) to interest rate movements over one-year horizon. It involves bucketing of all rate-sensitive assets (RSA), rate-sensitive liabilities (RSL), and off-balance sheet items as per residual maturity/ re-pricing date, in various time bands and computing earnings-at-risk (EAR) i.e., loss of income under different interest rate scenarios over a time horizon of one year. Advances, investments, swaps/ forex swaps and reverse repos are the major contributors to RSA whereas deposits, swaps /forex swaps and repos are the main elements under RSL. The DGA involves bucketing of all RSA and RSL as per residual maturity/ re-pricing dates in various time bands and computing the modified duration gap (MDG) to estimate the impact on the market value of equity. MDG is calculated with the following formula: MDG = [MDA - MDL * (RSL / RSA)], where MDA and MDL are the weighted averages of the modified duration (MD) of items of RSA and RSL, respectively. Thereafter, change in market value of equity (MVE) is computed as ΔE/ E = ‐[MDG]*RSA* Δi/ E, where Δi is the change in interest rate and E is equity (i.e. net worth). IV. Equity price risk Under the equity price risk, the impact of the shock of a fall in the equity price index, by certain percentage points, on bank capital was examined. The loss due to the fall in the value of the portfolio on account of change in equity prices is deducted from the bank’s capital to arrive at the stressed capital. (c) Bottom-up stress testing: Derivatives portfolios of select banks Stress tests on derivatives portfolio (in terms of notional value) were carried out by a sample of 25 banks, constituting the major active authorised dealers and interest rate swap counterparties. Each bank in the sample was asked to assess the impact of stress conditions on their respective derivatives portfolio. In case of domestic banks, the derivatives portfolio of both domestic and overseas operations was included. In case of foreign banks, only the domestic (Indian) position was considered for the exercise. Derivatives trades where hedge effectiveness was established were exempted from the stress tests, while all other trades were included. The stress scenarios incorporated four shocks consisting of the spot USD-INR rate and domestic interest rates as parameters (Table 2). 2.2 Primary (Urban) Co-operative Banks Single factor sensitivity analysis ─ Stress testing Stress testing of UCBs was conducted with reference to the reported position as of September 2024. The banks were subjected to baseline, medium and severe stress scenarios in the areas of credit risk, market risk and liquidity risk as follows: I. Credit Default Risk
II. Credit Concentration Risk
III. Interest Rate Risk in Trading Book
IV. Interest Rate Risk in Banking Book
V. Liquidity risk
2.3 Non-Banking Financial Companies (NBFCs) Single factor sensitivity analysis- Stress Testing Credit and liquidity risk stress tests for NBFCs have been performed under baseline, medium and high risk scenarios. I. Credit risk Major items of the balance sheet of NBFCs over one year horizon were projected by applying moving average and smoothing techniques. Assets, advances to total assets ratio, earnings before profit and tax (EBPT) to total assets ratio, risk-weight density and slippage ratio were projected over the next one year; and thereafter, based on these projections – new slippages, provisions, EBPT, risk-weighted assets and capital were calculated for the baseline scenario. For the medium and high-risk scenarios, GNPA ratios under baseline scenario were increased by 1 SD and 2 SD and accordingly new capital and CRAR were calculated. II. Liquidity Risk Stressed cash flows and mismatch in liquidity position were calculated by assigning predefined stress percentage to the overall cash inflows and outflows in different time buckets over the next one year. Projected outflows and inflows as on September 2024 over the next one year were considered for calculating the liquidity mismatch under the baseline scenario. Outflows and inflows of the sample NBFCs were applied a shock of 5 per cent and 10 per cent for time buckets over the next one year for the medium and high-risk scenarios, respectively. Cumulative liquidity mismatch due to such shocks was calculated as per cent of cumulative outflows and, NBFCs with negative cumulative mismatch were identified. 2.4 Stress Testing Methodology of Mutual Funds The SEBI has mandated all open-ended debt schemes (except overnight schemes) to conduct stress testing. Accordingly, Association of Mutual Funds in India (AMFI) prescribed the “Best Practice Guidelines on Stress Testing by Debt Schemes of Mutual Funds”. The stress testing is carried out internally by all Asset Management Companies (AMCs) on a monthly basis and when the market conditions require so. A uniform methodology is being followed across the industry for stress testing with a common outcome, i.e., impact on NAV as a result of the stress testing. Stress testing parameters The stress testing is conducted on the three risk parameters, viz., interest rate risk, credit risk and liquidity risk. A. Interest Rate Risk Parameter For interest rate risk parameter, AMCs subject the schemes at portfolio level to the following scenarios of interest rate movements and assess the impact on NAV.
B. Credit Risk Parameter For credit risk parameter, AMCs may subject the securities held by the scheme to the following:
C. Liquidity Risk Parameter For liquidity risk parameter, the following analysis is being undertaken:
AMCs additionally consider extreme stress scenarios of time bound liquidation (viz 5 days, 3 days and 1 day) of full portfolios and its impact on NAV by applying suitable haircuts. 2.5 Methodology for Stress Testing Analysis at Clearing Corporations SEBI has specified the granular norms related to core settlement guarantee fund (SGF); stress testing and default procedures to create a core fund (called core SGF) within the SGF against which no exposure is given and which is readily and unconditionally available to meet settlement obligations of clearing corporation in case of clearing member(s) failing to honour settlement obligation; align stress testing practices of clearing corporations with Principles for Financial Market Infrastructures (norms for stress testing for credit risk, stress testing for liquidity risk and reverse stress testing including frequency and scenarios); capture the risk due to possible default in institutional trades in stress testing; harmonise default waterfalls across clearing corporations; limit the liability of non-defaulting members in view of the Basel capital adequacy requirements for exposure towards central counterparties (CCPs); ring-fence each segment of clearing corporation from defaults in other segments; and bring in uniformity in the stress testing and the risk management practices of different clearing corporations especially with regard to the default of members. Stress testing is carried out at clearing corporations (CCs) to determine the minimum required corpus (MRC), which needs to be contributed by clearing members (CMs) to the core SGF. The MRC is determined separately for each segment (viz. cash market, equity derivatives, currency derivatives, commodity derivatives, debt and tri-party repo segment) every month based on stress testing subject to the following:
For determining the MRC for cash, equity derivatives and currency derivatives segment, CCs calculate the credit exposure arising out of a presumed simultaneous default of top two CMs. The credit exposure for each CM is determined by assessing the close-out loss arising out of closing open positions (under stress testing scenarios) and the net pay-in/ pay-out requirement of the CM against the required margins and other mandatory deposits of the CM. The MRC or average stress test loss of the month is determined as the average of all daily worst case loss scenarios of the month. The actual MRC for any given month is determined as at least the higher of the average stress test loss of the month or the MRC arrived at any time in the past. For the debt segment, the trading volume is minimal, and hence the MRC for the core SGF is calculated as higher of ₹4 crore or aggregate losses of top two CMs, assuming close out of obligations at a loss of four per cent less required margins. The tri-party repo segment and commodity derivatives segment also follow the same stress testing guiding principles as prescribed for equity cash, equity derivatives and currency derivatives segments. For commodity derivatives segment, however, MRC is computed as the maximum of either credit exposure on account of the default of top two CMs or 50 per cent of credit exposure due to simultaneous default of all CMs. Further, the minimum threshold value of MRC for commodity derivatives segment of any stock exchange is ₹10 crore. CCs carry out daily stress testing for credit risk using at least the standardized stress testing methodology prescribed by SEBI for each segment. Apart from the stress scenarios prescribed for cash market and derivatives market segments, CCs also develop their own scenarios for a variety of ‘extreme but plausible market conditions’ (in terms of both defaulters’ positions and possible price changes in liquidation periods, including the risk that liquidating such positions could have an impact on the market) and carry out stress testing using self-developed scenarios. Such scenarios include relevant peak historic price volatilities, shifts in other market factors such as price determinants and yield curves, multiple defaults over various time horizons and a spectrum of forward-looking stress scenarios in a variety of extreme but plausible market conditions. Also, for products for which specific stress testing methodology has not been prescribed, CCs develop extreme but plausible market scenarios (both hypothetical and historical) and carry out stress tests based on such scenarios and enhance the corpus of SGF, as required by the results of such stress tests. 2.6 Interconnectedness ─ Network analysis Matrix algebra is at the core of the network analysis, which uses the bilateral exposures between entities in the financial sector. Each institution’s lendings to and borrowings from all other institutions in the system are plotted in a square matrix and are then mapped in a network graph. The network model uses various statistical measures to gauge the level of interconnectedness in the system. Some of the important measures are given below: ![]() II. Cluster coefficient: Clustering in networks measures how interconnected each node is. Specifically, there should be an increased probability that two of a node’s neighbours (banks’ counterparties in case of a financial network) are neighbours to each other also. A high clustering coefficient for the network corresponds with high local interconnectedness prevailing in the system. For each bank with ki neighbours the total number of all possible directed links between them is given by ki (ki-1). Let Ei denote the actual number of links between bank i’s ki neighbours. The clustering coefficient Ci for bank i is given by the identity: ![]() The clustering coefficient (C) of the network as a whole is the average of all Ci’s: ![]() III. Tiered network structures: Typically, financial networks tend to exhibit a tiered structure. A tiered structure is one where different institutions have different degrees or levels of connectivity with others in the network. In the present analysis, the most connected banks are in the innermost core. Banks are then placed in the mid-core, outer core and the periphery (the respective concentric circles around the centre in the diagram), based on their level of relative connectivity. The range of connectivity of the banks is defined as a ratio of each bank’s in-degree and out-degree divided by that of the most connected bank. Banks that are ranked in the top 10 percentile of this ratio constitute the inner core. This is followed by a mid-core of banks ranked between 90 and 70 percentile and a 3rd tier of banks ranked between the 40 and 70 percentile. Banks with a connectivity ratio of less than 40 per cent are categorised in the periphery. IV. Colour code of the network chart: The blue balls and the red balls represent net lender and net borrower banks respectively in the network chart. The colour coding of the links in the tiered network diagram represents the borrowing from different tiers in the network (for example, the green links represent borrowings from the banks in the inner core). (a) Solvency contagion analysis The contagion analysis is in the nature of a stress test where the gross loss to the banking system owing to a domino effect of one or more banks failing is ascertained. We follow the round by round or sequential algorithm for simulating contagion that is now well known from Furfine (2003). Starting with a trigger bank i that fails at time 0, we denote the set of banks that go into distress at each round or iteration by Dq, q= 1,2, …For this analysis, a bank is considered to be in distress when its Tier I capital ratio goes below 7 per cent. The net receivables have been considered as loss for the receiving bank. (b) Liquidity contagion analysis While the solvency contagion analysis assesses potential loss to the system owing to failure of a net borrower, liquidity contagion estimates potential loss to the system due to the failure of a net lender. The analysis is conducted on gross exposures between banks comprising both fund based ones and derivatives. The basic assumption for the analysis is that a bank will initially dip into its liquidity reserves or buffers to tide over a liquidity stress caused by the failure of a large net lender. The items considered under liquidity reserves are: (a) excess CRR balance; (b) excess SLR balance; and (c) 18 per cent of NDTL. If a bank is able to meet the stress with liquidity buffers alone, then there is no further contagion. However, if the liquidity buffers alone are not sufficient, then a bank will call in all loans that are ‘callable’, resulting in a contagion. For the analysis only short-term assets like money lent in the call market and other very short-term loans are taken as callable. Following this, a bank may survive or may be liquidated. In this case there might be instances where a bank may survive by calling in loans, but in turn might propagate a further contagion causing other banks to come under duress. The second assumption used is that when a bank is liquidated, the funds lent by the bank are called in on a gross basis (referred to as primary liquidation), whereas when a bank calls in a short-term loan without being liquidated, the loan is called in on a net basis (on the assumption that the counterparty is likely to first reduce its short-term lending against the same counterparty. This is referred to as secondary liquidation). (c) Joint solvency-liquidity contagion analysis A bank typically has both positive net lending positions against some banks while against some other banks it might have a negative net lending position. In the event of failure of such a bank, both solvency and liquidity contagion will happen concurrently. This mechanism is explained by the following flowchart: ![]() The trigger bank is assumed to have failed for some endogenous reason, i.e., it becomes insolvent and thus impacts all its creditor banks. At the same time it starts to liquidate its assets to meet as much of its obligations as possible. This process of liquidation generates a liquidity contagion as the trigger bank starts to call back its loans. Since equity and long-term loans may not crystallise in the form of liquidity outflows for the counterparties of failed entities, they are not considered as callable in case of primary liquidation. Also, as the RBI guideline dated March 30, 2021 permits the bilateral netting of the MTM values in case of derivatives at counterparty level, exposures pertaining to derivative markets are considered to be callable on net basis in case of primary liquidation. The lender/creditor banks that are well capitalised will survive the shock and will generate no further contagion. On the other hand, those lender banks whose capital falls below the threshold will trigger a fresh contagion. Similarly, the borrowers whose liquidity buffers are sufficient will be able to tide over the stress without causing further contagion. But some banks may be able to address the liquidity stress only by calling in short term assets. This process of calling in short term assets will again propagate a contagion. The contagion from both the solvency and liquidity side will stop/stabilise when the loss/shocks are fully absorbed by the system with no further failures. 2.7 Financial System Stress Indicator (FSSI) FSSI is compiled using risk factors spread across five financial market segments (equity, forex, money, government debt and corporate debt), three financial intermediary segments (banks, NBFCs and AMC-MFs) as well as real sector (Table 3). FSSI lies between zero and unity, with higher value indicating more stress. For its construction, the risk factors pertaining to each component segment are first normalised using min-max method and thereafter aggregated based on simple average into a sub-indicator ‘yi‘ representing the ith market/ sector. Finally, the composite FSSI is obtained as, ![]() where the weight ‘wi‘ of each sub-indicator ‘yi‘ is determined from its sample standard deviation ‘si‘, as, ![]() |