The risk ratings, as specified above, are collective readings on the pre-specified scale and reflect the underlying credit-risk for a prospective exposure. The CRF could be separate for relatively peculiar businesses like banking, finance companies, real-estate developers, etc. For all industries (manufacturing sector), a common CRF may be used. The peculiarity of a particular industry can be captured by assigning different weights to aspects like entry barriers, access to technology, ability of new entrants to access raw materials, etc. The following step-wise activities outline the indicative process for arriving at risk-ratings.

1. Step I: Identify all the principal business and financial risk elements

2. Step II: Allocate weights to principal risk components

3. Step III: Compare with weights given in similar sectors and check for consistency

4. Step IV: Establish the key parameters (sub-components of the principal risk elements)

5. Step V: Assign weights to each of the key parameters

6. Step VI:Rank the key parameters on the specified scale

7. Step VII: Arrive at the credit-risk rating on the CRF

8. Step VIII: Compare with previous risk-ratings of similar exposures and check for consistency

9. Step IX: Conclude the credit-risk calibration on the CRF

The risk-rating process would represent collective decision making principles and as indicated above, would involve some in-built arrangements for ensuring the consistency of the output. The rankings would be largely comparative. As a bank's perception of the exposure improves/changes during the course of the appraisal, it may be necessary to adjust the weights and the rankings given to specific risk-parameters in the CRF. Such changes would be deliberated and the arguments for substantiating these adjustments would be clearly communicated in the appraisal documents. 2.5.5 Standardisation and Benchmarks for Risk-Ratings

In a lending environment dominated by industrial and corporate credits, the assignors of risk-ratings utilise benchmarks or pre-specified standards for assessing the risk profile of a potential borrower. These standards usually consist of financial ratios and credit-migration statistics, which capture the financial risks faced by the potential borrower (e.g. operating and financial leverage, profitability, liquidity, debt-servicing ability, etc.). The business risks associated with an exposure (e.g. cyclicality of industry, threats of product or technology substitution etc.) are also addressed in the CRF. The output of the credit-appraisal process, specifically the financial ratios, is directly compared with the specified benchmarks for a particular risk category. In these cases, the risk rating is fairly standardised and CRF allocates a grade or a numeric value for the overall risk profile of the proposed exposure.

Illustration

The CRF may specify that for the risk-rating exercise:

1. If Gross Revenues are between Rs.800 to Rs.1000 crore - assign a score of 2

2. If Operating Margin is 20% or more - assign a score of 2

3. If Return on Capital Employed (ROCE) is 25% or more - assign a score of 1

4. If Debt : Equity is between 0.60 and 0.80 - assign a score of 2

5. If interest cover is 3.50 or more - assign a score of 1

6. If Debt Service Coverage Ratio (DSCR) is 1.80 or more - assign a score of 1

The next step would be to assign weights to these risk-parameters. In an industrial credit environment, the CRF may place higher weights on size (as captured in gross revenues), profitability of operations (operating margins), financial leverage (debt: equity) and debt-servicing ability (interest cover). Assume that the CRF assigns a 20% weightage to each of these four parameters and the ROCE and DSCR are given a 10% weightage each. The weighted-average score for the financial risk of the proposed exposure is 1.40, which would correspond with the extremely low risk/highest safety level-category of the CRF (category 1). Similarly, the business and the management risk of the proposed exposure are assessed and an overall/ comprehensive risk rating is assigned.

The industrial credit environment permits a significantly higher level of benchmarking and standardisation, specifically in reference to calibration of financial risks associated with credit exposures. For all prominent industry-categories, any lender can compile profitability, leverage and debt-servicing details and utilise these to develop internal benchmarks for the CRF. As evident, developing such benchmarks and risk-standards for a portfolio of project finance exposures, as in the case of the bank, would be an altogether diverse exercise.

The CRF may also use qualitative/ subjective factors in the credit decisions. Such factors are both internal and external to the company. Internal factors could include integrity and quality of management of the borrower, quality of inventories/ receivables and the ability of borrowers to raise finance from other sources. External factors would include views on the economy and industry such as growth prospects, technological change and options.

The two critical aspects of the formality of procedure in the risk-rating process are

1. the process-flow through which a credit-transaction would flow across various units and

2. the written communication on the risk-ratings assigned to a particular proposal.

The process-flow required for the credit appraisal exercise, may be explicitly drafted and communicated. It may clearly identify the transactions and linkages between various operating units of the bank.

The above discussion broadly presents some of the essential dimensions in the design of a CRF by banks. These details are indicative of the scope of work required for the CRF. Banks may make appropriate modifications to suit their requirements.

CRFs and aggregation of Credit-Risk

Analysing exposures using the CRF technique would highlight the spread (or frequency-distribution) of the credit-risk in the bank's asset portfolio and would give an indication for its future asset build-up efforts. This section briefly covers some aspects of portfolio credit-risk management, a process which would possibly be facilitated by implementing the CRF.

The conventional internal MIS of a bank would identify the problem-loans in the asset portfolio, as per the guidelines given by the regulator (i.e. the asset-classification guidelines of RBI). These details, however, represent only a component of the credit-risks accumulated in the asset portfolio of a bank. The CRF can be used for informing the top management on the frequency distribution of assets across risk-rating scale, the extent of migration in the past (e.g. movement of exposures from higher to lower risk-categories or vice-versa) and the anticipated developments in the aggregated credit portfolio. The senior management may benefit from such outputs in terms of steering the organisation through various risk-cycles (e.g. initial low-risk low-return phase to consolidation and further to an incremental rise in relatively high-risk high-return exposures).

Adequate Levels of Provisioning for Credit Events

The spread of the asset portfolio across the risk-rating scale and the trends in rating migration would allow the bank management to determine the level of provisioning required, in addition to the regulatory minimum, to absorb unanticipated erosions in the credit quality of the assets. In most cases, provisions for loan-losses are based on the prevailing regulatory and accounting directives. However, the management may find merit in certain prudent level of "over-provisioning". This exercise may add stability and resilience to the capital adequacy and profitability of the bank.

The extent of provisioning required could be estimated from the Expected Loss on Default (which is a product of the Probability of Default (PD) and Loss Given Default (LGD). Since these probabilities can be assigned only after significant empirical details are available, an alternative would be to adopt a policy of allocating/ provisioning an amount which may be a proportion of the aggregate exposures in the risk-rating scale which reflect the likelihood of the assets slipping into the NPA category.

As discussed earlier, a clear analysis of the prevailing risk-posture of the bank, facilitated by the CRF, would give strong recommendations for future asset build-up and business development activities. The extent of provisioning would be based on actual and anticipated erosion in credit quality and would define the "cost" of maintaining an exposure in the bank's credit portfolio. A similar analysis could be undertaken for a specific credit-product and the risk-adjusted return can be assessed. This will involve an analysis of the pricing-decisions, provisioning requirements, loss on default and the incremental impact on bank's profitability.

The benefits of such a CRF system, in addition to those described above, could include a more amenable interaction with rating agencies and regulatory bodies. As regards investment ratings the parameters laid down in para 4.1 of the Risk Management Guidelines issued by RBI in October, 1999 may be followed, i.e., the proposals for investments should also be subjected to the same degree of credit risk analysis, as any loan proposals. The proposals should be subjected to detailed appraisal and rating framework that factors in financial and non-financial parameters of issuers, sensitivity to external developments, etc. The maximum exposure to a customer should be bank-wide and include all exposures assumed by the Credit and Treasury Departments. The coupon on non-sovereign papers should be commensurate with their risk profile. The banks should exercise due caution, particularly in investment proposals, which are not rated and should ensure comprehensive risk evaluation. There should be greater interaction between Credit and Treasury Departments and the portfolio analysis should also cover the total exposures, including investments. The rating migration of the issuers and the consequent diminution in the portfolio quality should also be tracked at periodic intervals.