Explainers  /  17 of 20

Extended vs untreated

Comparing extended loans with similar loans that were not extended, and what that comparison can and cannot prove.

Extended vs untreated: loans that got a payment extension in a month, compared with loans that were just as late that month but did not get one, by where each group stands 1, 3 and 6 months later.

The comparison in plain words

  • Take every loan that was 30-59 days past due in one month.
  • Extended: those that got an extension the next month. Untreated: the rest, same month, same lateness.
  • For each group, count where the loans stand 1, 3 and 6 months on: paid off, current, a few days late, 30+ days late, or charged off. Only loans whose status that month is reported count.
  • The deck shows the latest month for which all three checks are in.

Worked example: ten loans

Ten loans, all 30-59 days late in February. Four got an extension in March; six did not.

Loan March April (1 month on) September (6 months on)
A extended Current Current
B extended Current Current
C extended Current Current
D extended Current Charged off
E not extended 30+ days late Current
F not extended Current Current
G not extended 30+ days late Paid off
H not extended 30+ days late 30+ days late
I not extended 30+ days late Charged off
J not extended 30+ days late 30+ days late

Share 30+ days late or charged off:

1 month on 6 months on
Extended 0 of 4 = 0% 1 of 4 = 25.0%
Not extended 5 of 6 = 83.3% 3 of 6 = 50.0%

The real numbers

Loans 30-59 days late in March 2026, the latest month with all three checks reported. Prime is FICO 660 and up at origination, subprime under 660. Share 30+ days late or charged off:

1 month on 3 months on 6 months on Loans with a six-month status
Subprime, extended 18.9% 51.1% 53.6% 14,325
Subprime, not extended 65.4% 61.0% 60.5% 177,799
Prime, extended 16.5% 41.2% 44.1% 2,449
Prime, not extended 55.3% 50.7% 49.8% 42,042

One month on, the gap is 46.6 points for subprime, mostly because the extension itself resets the loan. By six months it has narrowed to 6.9 points (53.6% against 60.5%). Six months on, extended subprime loans were also less often charged off (11.0% against 18.0%) and less often paid off (2.6% against 5.1%).

The comparison says extended loans did somewhat better than loans that were just as late in the same month. It does not say how much of that the extension caused.

Common pitfalls

  1. Reading month one as a result. An extension moves the missed payment to the end of the loan, so the loan reports current straight away. The one-month gap is mostly bookkeeping. Look at three and six months.
  2. Reading the gap as cause and effect. Servicers choose who to extend. The comparison matches on two things only, the month and how late the loan was; it does not see why the borrower fell behind, their income, or why the servicer said yes. If extensions go to borrowers who were likely to recover anyway, or to those in deeper trouble, the gap reflects that choice as much as the extension. The comparison points a direction; it does not prove the extension caused the difference.
  3. Forgetting who drops out. Loans whose status is not reported at the check month are left out of both groups, so each share is over the loans we can see.

Where to see it

Source: LoanTape, SEC Form ABS-EE loan-level data, through September 2026.