Skip to content

FRM Part I · FRM Exam Part I

External and Internal Credit Ratings for FRM Part I

Credit ratings rank borrowers by creditworthiness. External ratings come from agencies such as S&P, Moody's and Fitch. Internal ratings are built by banks. You must read rating scales, use transition matrices to get default and migration probabilities, separate through-the-cycle from point-in-time ratings, and know how systems are validated and backtested.

What this chapter covers

This chapter covers how credit quality is measured and ranked. It starts with external ratings from agencies, the scales they use and the line between investment grade and speculative grade. It then moves to the numbers behind ratings: default rates, transition matrices and how to turn one-year probabilities into multi-year ones.

The second half is about judgment and design. You compare through-the-cycle (TTC) and point-in-time (PIT) ratings, study the criticisms of agency ratings, and see how banks build internal rating systems and test them. Validation covers discriminatory power, calibration and stability.

The chapter links to other parts of the paper. Default probabilities feed expected loss (EL = PD × LGD × EAD) and credit risk models. Transition matrices use matrix multiplication and probability from Quantitative Analysis. Rating philosophy matters for capital requirements and procyclicality in the bank regulation material. Backtesting echoes the VaR backtesting you see elsewhere in Valuation and Risk Models.

Rating questions are usually concept-heavy with a small calculation, so they are good value for the time you spend. A typical question asks you to read a transition matrix, compound a probability over two years, or decide whether a rating is TTC or PIT. The ideas also support credit risk, capital and regulation questions, so a solid grasp here pays off across several topics. Because all 100 questions carry equal weight and the exam is 4 hours long, quick, accurate points from this chapter help you save time for harder quantitative items.

External and Internal Credit Ratings: topics in the order to study them

  1. 1Credit Rating Agencies and Rating ScalesStart with the vocabulary: agencies, letter grades and the investment-grade boundary. Everything else uses these terms.
  2. 2Rating Transition Matrices and Default ProbabilitiesThis is the main calculation topic. Learn it early so you can practise it repeatedly.
  3. 3Through-the-Cycle vs Point-in-Time RatingsOnce you know how migration and default rates behave, you can see why the two philosophies give different patterns.
  4. 4Limitations and Criticisms of External RatingsBuilds on TTC behaviour: slow adjustment, rating stickiness and conflicts of interest make more sense after the last topic.
  5. 5Internal Rating Systems and Rating PhilosophyNow apply the ideas to how banks design their own ratings, including data, structure and philosophy choice.
  6. 6Validation and Backtesting of Rating SystemsFinish with testing, since you need to know what a system is meant to do before you judge whether it works.

How to prepare External and Internal Credit Ratings

Split your time between a few formulas and many concepts. Do the calculations by hand until they are automatic, then use concept drills to sharpen distinctions.

  1. Learn the rating scales of S&P/Fitch and Moody's side by side, and fix where investment grade ends.
  2. Work through transition matrices: read a row as a probability distribution that sums to 100%, and treat default as an absorbing state.
  3. Practise multi-year default probabilities by multiplying matrices, and by survival logic: P(survive 2 years) = P(survive year 1) × P(survive year 2 | survived year 1).
  4. Write a short comparison of TTC and PIT: what inputs each uses, how stable each is, and how each affects capital over the cycle.
  5. List the criticisms of agency ratings and the matching internal-rating fixes, so you can answer why banks still build their own systems.
  6. Study validation in three parts: discriminatory power, calibration and stability. Be able to say what each test tells you.
  7. Finish with mixed practice questions under time pressure, aiming for about two to three minutes per question.

Common mistakes in External and Internal Credit Ratings

  • Treating a transition matrix row as a column or failing to check that rows sum to 100%.

    Fix: Rows are the starting rating and columns are the ending rating. Check that the row you use sums to 100% before calculating.

  • Adding yearly default probabilities to get a two-year probability.

    Fix: Use cumulative logic: two-year PD = PD1 + (1 − PD1) × PD2 where PD2 is the conditional default probability, or multiply matrices.

  • Mixing up TTC and PIT ratings.

    Fix: Remember that TTC ignores short-term swings and is stable, while PIT updates with current conditions and is volatile.

  • Assuming agency ratings are always wrong or always reliable.

    Fix: Frame ratings as useful but limited: they rank risk reasonably well, but adjust slowly and have incentive problems.

  • Confusing discriminatory power with calibration.

    Fix: Discrimination is about ranking borrowers correctly. Calibration is about getting the PD levels right. A system can do one well and the other badly.

  • Ignoring the conditions behind the matrix method.

    Fix: Remember that matrix powers assume the same one-year matrix applies each year and that migrations depend only on the current rating. Real ratings can show momentum and cycle effects.

Last-day revision: External and Internal Credit Ratings

  • Investment grade means BBB-/Baa3 or higher; below that is speculative grade.
  • Each row of a transition matrix sums to 100%.
  • Default is an absorbing state: once there, a borrower stays there.
  • Multi-year transitions come from multiplying the one-year matrix by itself, assuming it is stable over time.
  • Expected loss = PD × LGD × EAD.
  • TTC ratings look past the cycle, so they are stable and change slowly.
  • PIT ratings reflect current conditions, so they move with the cycle and give more procyclical capital.
  • Agency ratings are criticised for lag, cliff effects, conflicts of interest and reliance on issuer-paid models.
  • Internal ratings can use bank-specific data and allow a rating for every borrower, including unrated ones.
  • Discriminatory power asks whether a system separates defaulters from non-defaulters.
  • Calibration asks whether predicted PDs match observed default rates.
  • Backtest by comparing predicted PDs with realised default frequencies over time.

External and Internal Credit Ratings practice questions

External and Internal Credit Ratings in other exams

The same ground in other exams, if you are preparing for more than one or want another angle on it.

External and Internal Credit Ratings: frequently asked questions

How do I calculate a two-year default probability from a transition matrix?

Multiply the one-year matrix by itself and read the default column for the starting rating. For a single rating you can sum, over each possible rating after year one, the probability of moving there times that rating's one-year default probability, then add the chance of defaulting in year one.

What is the difference between TTC and PIT ratings?

A through-the-cycle rating assesses a borrower across a full economic cycle and changes little. A point-in-time rating uses current conditions and moves more with the cycle. PIT ratings make capital more procyclical.

Why do banks build internal ratings when agency ratings exist?

Most borrowers have no agency rating, and banks hold data that agencies do not. Internal ratings can be tailored to the portfolio and updated more often. They must still be validated to show they work.

What does validation of a rating system involve?

It checks discriminatory power, calibration and stability. You test whether the system ranks borrowers correctly, whether predicted default rates match outcomes, and whether ratings are consistent over time. Backtesting compares predictions with realised defaults.