What Is Credit Analysis? The 5 Cs, Key Ratios and How Banks Decide Loans

Credit Analysis at a Glance

Credit analysis is the work of judging, before money is lent, whether a borrower can repay — and whether they will. Every loan a bank sanctions, every bond an investor buys, passes through this judgment first.

Bank examiners state the test plainly. The FDIC's examination manual is the handbook US bank supervisors work from. It says: "The willingness and ability of a debtor to perform as agreed remains the primary measure of a loan's risk." Ability is arithmetic: cash flows against dues. Willingness is track record and character.

The timing for learning this skill in India is unusually good. Bank loan books are the cleanest in decades — gross NPAs (non-performing assets, loans gone bad) fell to a multi-decadal low of 1.8% by March 2026. That is per RBI's Financial Stability Report of June 2026. And RBI's new expected-credit-loss rules, effective 1 April 2027, are pushing every large lender to hire people who understand credit deeply.

This guide walks one illustrative loan through the whole judgment — a ₹50 lakh, 5-year term loan to a small food company. We cover the 5 Cs, the ratios, the decision, and the rating language around it.

What Exactly Is Credit Analysis?

You already do informal credit analysis. Before lending ₹500 to a friend, you check three things in your head. Do they earn enough to return it? Have they repaid before? And what happens if they don't? A credit analyst asks the same three questions about companies — with financial statements instead of gut feel.

The formal version has a precise shape. The OCC is the US regulator that supervises national banks. It "expects all rating systems to address both the ability and willingness of the obligor to repay and the support provided by structure and collateral." Obligor is simply the borrower. Structure means the loan's terms. Collateral is the security pledged against it.

The word every analyst works backwards from is default — in the OCC's plain definition, "the failure to make a required payment in full and on time." India adds an operational trigger to that idea. Under RBI's asset-classification rules, a term loan turns into an NPA when interest or principal stays overdue for more than 90 days. (Working-capital accounts like cash credit use a related "out of order" test instead — the 90-day rule is the term-loan version.)

So credit analysis, precisely: estimating how likely a borrower is to default, and how much the lender would suffer if they did — then pricing and structuring the loan so the risk is worth taking.

What Are the 5 Cs of Credit?

Bankers traditionally compress the whole borrower assessment into five words starting with C. The list is industry folklore in the best sense — not owned by any regulator, but taught on every credit desk. Here is what each C means, applied to our illustrative borrower: a small packaged-food company asking for a ₹50 lakh term loan.

Plain takeaway: capacity decides whether the loan works on paper, and character decides whether it works in real life. The other three Cs decide how badly things go if you are wrong.

Notice that only one of the five is pure arithmetic. That is why credit analysis is called a judgment skill: the numbers are necessary, never sufficient.

Which Ratios Do Credit Analysts Actually Use?

Rating agencies publish exactly which numbers they check, so we do not have to guess. Crisil Ratings' public criteria say it "considers eight crucial financial parameters while evaluating a company's credit quality: capital structure, interest coverage ratio, debt service coverage, networth, profitability, return on capital employed (RoCE), net cash accrual to total debt (NCATD) ratio, and current ratio."

Four of these carry most of the weight in day-to-day bank credit work. Let us compute them for our borrower. The setup, all illustrative: the company earns PBDIT of ₹20 lakh a year (profit before depreciation, interest and tax — the operating cash engine, close to what analysts call EBITDA). It wants ₹50 lakh for 5 years at roughly 11% interest. In year one that means about ₹5.5 lakh of interest and ₹10 lakh of principal repayment. Its profit after tax is ₹9 lakh, depreciation ₹3 lakh, net worth ₹40 lakh, and total debt after this loan ₹60 lakh.

Plain takeaway: interest cover looks strong, but DSCR — the ratio that includes principal — is thin at 1.13×; that is where this loan's real risk shows.

Why does DSCR matter most? Because it is the full test. Crisil's criteria spell out the reading: a ratio "greater than 1 time implies that an entity would be able to repay its debt in a particular year from cash accrual generated during that period", while an entity below 1 "may have insufficient accrual to meet all debt obligation, and hence has a higher probability of default." (Crisil itself uses a refined "cash DSCR" that also sets aside part of working-capital growth — the textbook version above is the fresher-friendly starting point.)

Two more terms you will hear on Indian credit desks. TOL/TNW — total outside liabilities to tangible net worth — is banker shorthand for what Crisil calls the total indebtedness ratio. It measures all outside claims, not just borrowings, against the owners' real stake. And drawing power is the amount a working-capital borrower may actually draw, recalculated from current stock and receivables.

How Does a Bank Decide a Loan in Practice?

Ratios feed a document, and the document feeds a committee. The analyst's core output is the credit appraisal note (often called a credit memo) — the file on which sanction is approved or refused. A typical note covers six things, in order:

• Borrower and management profile — who they are, and their record.
• Business and industry view — how the company earns, and what could disturb it.
• Financial analysis — the ratios from the previous section, with trends.
• Security and collateral — what protects the bank, and its realistic value.
• Terms and covenants — covenants are promises the borrower must keep, like maintaining a minimum DSCR.
• The recommendation — lend, refuse, or lend on tighter terms.

For smaller working-capital limits, RBI has even standardised the arithmetic. Under RBI's turnover method (as summarised by IIBF, the Indian Institute of Banking & Finance), the requirement is "computed at 25% of the projected gross annual turnover". The bank finances at least 20% of turnover; the borrower brings a 5% margin. For larger limits, the same guidance allows "either the turnover method or cash budgeting method or any other method as considered necessary" — method chosen to fit the borrower, judgment always attached.

And the job does not end at sanction. The same IIBF-summarised guidance requires banks to ensure "proper end use of funds" and to run "periodical review of accounts on regular basis". So the analyst's cycle is: appraise → sanction → monitor end-use → review → re-rate. Credit analysis is a relationship with a loan, not a one-time exam of it.

What Do Credit Ratings Mean?

When the borrower is a large company issuing bonds, the credit judgment becomes public — as a rating. India's scales are deliberately uniform. SEBI standardised rating symbols and definitions across agencies in June 2011 (circular CIR/MIRSD/4/2011). That is why a Crisil, ICRA or CARE long-term scale reads the same way.

The ladder runs from AAA to D. In Crisil's published definitions, AAA securities carry "the highest degree of safety regarding timely servicing of financial obligations"; AA means high safety; A adequate; BBB moderate safety. Below that, the language flips from safety to risk. BB means moderate risk of default, B high risk, C very high risk. And D means the security "is in default or expected to be in default soon." Agencies may attach + or − modifiers from AA down to C.

Market convention treats BBB− and above as "investment grade" — the zone most institutional money is allowed to hold. That line is convention, not a Crisil definition, but it is why a one-notch downgrade near the boundary moves a company's borrowing cost sharply.

Credit Analysis vs Credit Risk Modeling: What Is the Difference?

Everything so far judges one borrower at a time. But a bank holds lakhs of loans. Nobody can hand-write an appraisal note for every credit card in the book — so the same judgment gets rebuilt as statistics. That is credit risk modeling, and it is the natural career upgrade from credit analysis.

The Basel Committee — the global standard-setter for bank regulation — defines the three numbers every credit model estimates. In its own words: "probability of default (PD) per rating grade, which gives the average percentage of obligors that default in this rating grade in the course of one year"; "exposure at default (EAD), which gives an estimate of the amount outstanding … in case the borrower defaults"; and "loss given default (LGD), which gives the percentage of exposure the bank might lose in case the borrower defaults." Multiply them and you get expected loss: EL = PD × EAD × LGD.

India has just made this shift mandatory. RBI issued its final expected-credit-loss directions on 27 April 2026 (the Reserve Bank of India (Commercial Banks – Asset Classification, Provisioning and Income Recognition) Directions, 2026). From 1 April 2027, commercial banks must provision using an ECL approach aligned with the global IFRS-9 style. (Small finance banks, payments banks and local area banks are excluded.) Every loan gets sorted into Stage 1, 2 or 3 — 12-month expected loss while healthy, lifetime expected loss once credit risk jumps.

For your career, the two skills stack rather than compete. Analysis teaches you why borrowers default; modeling teaches you to measure it across a portfolio in Python or SAS. Our credit risk modeling guide covers the modeling half in the same plain language as this post.

Where Do Credit Analysts Work in India?

Four employer families hire this skill, each with a different flavour of the job:

• Banks. The classic home — corporate credit, SME credit, and now ECL implementation teams. The system they join is healthy: RBI's June 2026 Financial Stability Report describes a resilient, well-capitalised banking sector with gross NPAs at that 1.8% multi-decadal low.
• NBFCs (non-banking financial companies). A loan book of roughly ₹48 lakh crore and growing (RBI's Report on Trend and Progress of Banking in India 2024–25) — vehicle finance, gold loans, SME lending. Faster credit decisions, heavier reliance on sharp analysts.
• Rating agencies. Crisil, ICRA and CARE publish the criteria this guide quoted — their analysts apply exactly these frameworks to rated companies. The training is considered among the best in the market.
• Consulting and global teams. Big-4 risk practices and global banks' India centres take credit skills into model-building, validation and ECL projects — the modeling bridge in action.

Pay varies widely by employer family and by whether you carry modeling skills on top of analysis. We keep the sourced numbers in one place — the credit risk analyst salary in India guide — rather than repeating them here.

How Do You Become a Credit Analyst?

There is no licence gate — the path is skills plus proof. The sequence that works for freshers:

• Step 1 — Ratio fluency. Be able to compute and, more importantly, interpret coverage, DSCR, gearing and liquidity from a real annual report. This whole guide is that syllabus in miniature.
• Step 2 — Write one real credit note. Pick a listed company, pull its annual report, and write a two-page appraisal: business, ratios, security, verdict. This single artefact beats a line on a CV in every interview.
• Step 3 — Learn the regulatory vocabulary. NPA and the 90-day rule, ECL and its three stages, SICR (significant increase in credit risk — the Stage-2 trigger). Interviewers now assume ECL literacy.
• Step 4 — Add the modeling layer. Statistics, Python and a PD-model project move you from writing notes to building the systems that read the whole book — the modeling guide maps this jump.
• Step 5 — Certify if you want the global signal. The FRM covers credit risk measurement in depth and is widely recognised on risk desks.

Want the compressed version with faculty support? QuintEdge's Credit Risk Modeling course teaches the analysis-to-model path on realistic lending data — the exact progression this post describes. It pairs naturally with FRM coaching for the credential on top.