Loan vintage tells a buyer when each loan was originated or settled; seasoning is the time it has remained on the book. These fields matter because trail income is commonly linked to the outstanding, eligible loan balance under the applicable commission terms—not merely to the number of client files—and that balance can change through scheduled principal repayment, extra repayment, refinance, discharge, redraw or product changes.

Neither a young nor an old book is automatically better. A young cohort may carry larger balances but has less observed history. An older cohort has survived for longer, yet may have lower balances or be closer to refinance, sale or maturity. The useful question is: how have loans from each vintage actually behaved, and what does that evidence imply for future net receipts?

Define vintage before comparing cohorts

Specify which date creates the vintage: application, approval, origination, settlement, first drawdown or the date the loan entered the seller’s book. For trail analysis, settlement or funding date will often be more useful than application date, but the contractual and system definitions must govern. A transferred or refinanced loan may also acquire a new system identifier even though the client relationship is older.

The Reserve Bank of Australia’s securitisation reporting framework treats origination or settlement date, maturity date, remaining term and “seasoning” as separate loan-level fields (RBA). Do not substitute one field for another; retain the source date beside any calculated age.

Create fixed cohorts—such as settlement year or quarter—and keep them consistent across every monthly snapshot. APRA’s residential mortgage guidance identifies loan vintage as one dimension for assessing portfolio composition and quality, alongside product, distribution channel, LVR, arrears and other characteristics (APRA). APRA’s risk framework does not value a broker’s trail book; it shows why age belongs with other portfolio attributes.

Why age changes the income profile

For a principal-and-interest loan, regular repayments reduce principal over the loan term. For an interest-only loan, principal is not reduced during the interest-only period (Moneysmart). Two loans settled on the same day can therefore season differently if their products, repayment behaviour, redraws and balance increases differ.

Age affects a trail book through several mechanisms:

  • Amortisation and extra repayments: scheduled and voluntary principal reductions can reduce the balance on which trail is calculated. The RBA distinguishes scheduled principal repayments, partial prepayments and full prepayments in explaining housing-credit movement (RBA).
  • Refinance and discharge: a loan can leave the book when the borrower changes lender, sells or repays it. Moneysmart describes refinancing costs and borrower incentives to seek a lower rate, showing why retention cannot be inferred from contractual maturity alone (Moneysmart).
  • Product milestones: fixed-rate expiry, interest-only expiry and maturity may create decision points, but their effect must be observed rather than assumed.
  • Survivor selection: an old cohort contains only loans still present. Looking at today’s survivors without the original cohort can hide the loans already discharged.
  • Client relationship: seasoned clients may offer servicing or refinancing opportunities, but they also require compliant, client-focused advice. ASIC’s best-interests guidance makes clear that a broker must investigate the consumer’s circumstances and available products; acquisition assumptions should not treat refinancing as an automatic revenue event (ASIC).

Evidence to request by loan and month

Ask for de-identified monthly loan-level snapshots that preserve a stable loan ID and, where legitimately available, a stable household or client ID. At minimum, request:

  • settlement or funding date, original loan amount and original term;
  • current balance, scheduled balance and remaining term;
  • lender, product type, fixed or variable status and relevant expiry dates;
  • principal-and-interest or interest-only status;
  • gross trail, deductions and net trail by month;
  • discharge, refinance, closure, arrears, hardship or non-paying flags and effective dates;
  • redraw, top-up or loan-split events, with a bridge between old and new identifiers; and
  • the data dictionary, missing-value codes and any changes in reporting method.

Reconcile snapshot totals to aggregator statements and bank receipts before relying on cohort results. A file containing only active loans at today’s date cannot measure historic survival because its denominator has already lost the exits.

Build a vintage matrix rather than one average

For each settlement cohort, calculate the original loan count and balance, current survivors and balance, months observed, and trail received. Then compare:

  1. Loan survival: active loans divided by the cohort’s original loan count.
  2. Balance survival: current eligible balance divided by the cohort’s original eligible balance, adjusted transparently for top-ups or splits.
  3. Receipt yield: actual gross and net trail relative to the relevant balance, with timing and deductions explained.
  4. Exit pattern: discharges and refinances by month since settlement, lender, product and responsible broker.
  5. Data completeness: the proportion with valid dates, stable IDs and explained status changes.

Use both calendar time and loan age. Calendar-time analysis exposes market-wide events affecting several cohorts at once; “months since settlement” shows how cohorts behave at comparable stages. Weighted average age can supplement this work, but it should not replace the distribution. The same average could describe a balanced five-year book or a mix of very new and very old loans with different risks.

Hypothetical example: a book has two $20 million original-balance cohorts. After 24 months, Cohort A retains 84% of its original eligible balance; Cohort B retains 68%. The gap might reflect different lenders, products, borrower profiles, refinancing conditions or data treatment. It is a prompt for investigation, not proof that Cohort A will retain 84% in future. Reconcile receipts, examine exits and test whether top-ups or identifier changes distort either figure. These numbers illustrate the method only; they are not benchmarks.

Questions that keep the analysis honest

  • Does “loan age” run from settlement, first drawdown or another date?
  • Are refinances within the same lender treated as continuations, exits or new loans?
  • Do splits and top-ups duplicate balances or reset vintage?
  • Which cohorts experienced unusual fixed-rate expiries, campaigns or lender policy changes?
  • Are recent cohorts large because the seller grew, acquired another book or refinanced existing clients?
  • Is apparent survival caused by missing historical exits?
  • Do older cohorts produce lower receipts because of amortisation, non-payment, fee changes or unexplained deductions?
  • Does the forecast apply one run-off assumption to every cohort despite different evidence?

Turn findings into acquisition decisions

Use cohort results to challenge the forecast, not to manufacture a precise lifespan. Model at least a base case and downside case using documented assumptions for balance run-off, discharge, net deductions and missing data. Keep any new lending or refinancing upside separate from the acquired trail; it depends on future client needs, compliant advice and operational execution.

Ask an accountant or valuation specialist to review the model, a lawyer to review commission and transfer rights, and credit-compliance advisers to assess servicing obligations. The seller and aggregator should explain data lineage, identifier changes and exceptional cohorts. Where evidence remains uncertain, discuss transaction-specific protections with advisers.

Frequently asked questions

Is an older trail book safer than a newer one?

Not necessarily. Older loans provide more observed history, but survivors may have smaller balances or approaching decision points. Compare historic survival, balance movement, receipts and data quality by cohort.

What is a “seasoned” loan?

There is no single transaction-wide threshold. Define seasoning as elapsed time from a specified event—usually settlement or funding—and disclose that definition. Analyse the full distribution rather than labelling the whole book seasoned.

Can average loan age support a valuation?

It is one descriptor, not a valuation method. A buyer still needs loan-level cash reconciliation, cohort run-off, contractual rights, deductions and scenario analysis.

Sources

Disclaimer

This article provides general information only. It does not constitute legal, tax, accounting, financial, credit, valuation or other professional advice. Circumstances differ. Obtain advice from appropriately qualified advisers before acting or entering a transaction.