Lending Pain Map

Collection efficiency for NBFCs: three formulas, and why 105% can be a bug

Current, monthly and cumulative collection efficiency divide different things. How each is calculated, why one can pass 100%, and which figures to report.

Collection efficiency for NBFCs: three formulas, and why 105% can be a bug: cover art
Quick answer

Collection efficiency is collections divided by billing, but lenders use at least three versions. Current collection efficiency counts only what was collected against this month’s instalments. Monthly collection efficiency adds overdue collections to the numerator and can pass 100%. Cumulative collection efficiency runs from a start date, sometimes with opening overdues in the denominator. Report which one you use, and never compare figures built on different formulas.

If you report collection efficiency to a board, a lender or a rating agency, a careful reader’s first question is which formula you used. Two NBFCs can both report 98% and be collecting very differently.

“Collections” and “billing” each have more than one reasonable meaning, and a monthly figure of 105% looks like a strong month whether or not it is one. The fix is to pick a formula, say which one, and produce it from the same loan records every month.

Key takeaways
  1. Three formulas share one name. Current, monthly and cumulative collection efficiency divide different things.
  2. Above 100% is a formula effect, or a bug. Arrears in the numerator can push past 100 in a catch-up month. With prepayments kept out, if arrears collected do not cover the excess, the data is wrong.
  3. Write-offs and prepayments move the number. Neither says anything about this month’s paying behaviour.
  4. Report two figures. Current collection efficiency beside the total tells the reader how much came from catching up.
  5. Every receipt needs a target. The loan system has to record which demand each rupee cleared.

What is collection efficiency?

It is the share of what borrowers owed in a period that the lender actually collected. The definitions differ in what counts as owed, what counts as collected, and over what period. ICRA prints the formula as a footnote in its securitisation rating reports, and a lender’s own reports can do the same.

Before it rates a pool, an agency also reads the loan-level data in the lender’s loan tape.

Which three formulas do lenders mix up?

Current, monthly and cumulative. The definitions below follow ICRA’s June 2025 securitisation methodology and its rating reports. ICRA lists current collection efficiency beside a monthly measure that explicitly adds the overdues collected, so current counts collections against the month’s billing.

MeasureNumeratorDenominatorCan it pass 100%?
Current collection efficiencyCollections against this month’s instalmentsThis month’s billingNo
Monthly collection efficiencyAll collections in the month, current and overdueThis month’s billingYes
Cumulative collection efficiencyAll collections since a start dateBilling since that date, with or without opening overduesOnly if arrears from before the start date are collected and opening overdues are left out of the denominator

A fourth measure sits beside them. The 2019 edition of ICRA’s securitisation methodology described arrears collection efficiency: collections of overdue amounts in a month divided by the overdue amount at the start of that month. It answers a different question, which is how much of the backlog the team cleared.

Even the cumulative measure has two versions in current use. Some ICRA reports divide by cumulative billing alone. Others add the overdues that were already on the pool at the start of the transaction. The second version is stricter, and the two numbers for the same pool will differ.

How can collection efficiency pass 100%?

When overdue collections go into the numerator and only this month’s billing goes into the denominator. Take an illustrative book. The figures are made up and rounded so the arithmetic is easy to check.

The month

₹10 crore of instalments fall due in October. Borrowers pay ₹9.2 crore of that during October. Borrowers already behind pay ₹1.3 crore of older arrears. ₹4 crore was overdue at the start of the month.

Current and monthly

Current collection efficiency is 9.2 ÷ 10, or 92%. Monthly collection efficiency is (9.2 + 1.3) ÷ 10, or 105%. Same month, same book.

Arrears

Arrears collection efficiency is 1.3 ÷ 4, or 32.5%. ₹2.7 crore of the ₹4 crore backlog is still there, which neither of the other two figures shows.

A 105% month can be good news: borrowers who fell behind are catching up. It can also mean the book carries a large pile of arrears that some borrowers are paying down, while 8% of this month’s billing went unpaid. The monthly figure alone cannot tell the two apart.

Can a figure above 100% be a bug?

Yes, and the formula itself gives the check. If monthly collection efficiency is above 100% and prepayments are kept out of it, overdue collections have to cover at least the excess. At 105% on ₹10 crore of billing, at least ₹0.5 crore must have come from arrears. A report that shows more than 100% alongside little or nothing collected from overdue accounts does not add up.

We met exactly that on BACO, the synthetic NBFC we run to test Lokta. A day-close briefing reported month-to-date collections of ₹8.27 crore against demand of ₹7.81 crore, or 105.9%, while showing nothing recovered from overdue and no account changing bucket. The numbers could not all be true. The data source had been briefly unreachable and the feature had fetched dummy data. The digital twin post tells the story. The lesson for a reporting team is to publish the overdue collections beside any figure above 100%, so the check is visible.

What else moves the number?

Four things: write-offs, prepayments, the date a receipt is counted on, and the order a part-payment is applied in. The first two say nothing about how borrowers paid this month.

  • Once an account is written off, its instalments leave the billing. ICRA’s securitisation methodology makes the point directly: a lender with a stricter write-off policy tends to show better collection efficiency than one that writes off less often.
  • If prepayments and foreclosures count as collections, a month in which a block of loans closes early looks like a strong collections month.
  • A payment made on the 30th but posted on the 1st counts in one month or the next, depending on whether the report uses value date or posting date.
  • When a borrower pays less than the full amount due, the lender’s appropriation order decides whether the money clears current demand or older arrears, and so which formula it improves. The payment appropriation post works through how allocation rules change the result.

Which formula should you report?

Two figures side by side, each with its formula written out: current collection efficiency, and monthly or cumulative collection efficiency as your lenders or the transaction documents define it. Add arrears collection efficiency when overdues are material. State whether prepayments and written-off accounts are included.

For the collections team, the same measures cut by bucket are more useful than the book-wide figure. The X bucket, accounts that were current when the month began, is where a first miss shows up. Definitions vary between lenders, so print yours beside the number. The bucket-by-bucket collection strategy post sets out what each bucket needs. Roll rate and cure rate show whether accounts in each bucket slide into a worse one or come back to current, and the collections KPIs post gives the formulas.

What does the loan system need to produce it?

Four things, recorded as events rather than rebuilt at month end:

  • each instalment’s demand and due date, as it stood when it fell due
  • each receipt, with its value date and the rail it came through
  • the demand each receipt cleared, split into current and overdue, under the product’s appropriation order
  • prepayments, foreclosures and write-offs, each dated and flagged

With those four, every version of the formula comes from the same records, and a changed definition can be run back over earlier months. Without them, a finance team rebuilds the split in a spreadsheet, and two teams working from the same book can report different numbers.

Which path fits your finance team?

If you want to be the finance head whose collection efficiency figure survives a rating analyst’s first question, there are three ways to report it.

  1. Report one monthly figure, as the collections dashboard calculates it. It is easy. Nobody outside the team can tell how much of 105% came from this month’s instalments and how much from arrears, and the formula may change when the dashboard does.
  2. Report current and monthly collection efficiency together, formulas written out, produced from receipt-level allocation in the loan system. Setting up the split takes effort once. After that the numbers reconcile to the ledger and survive a change of definition.
  3. Rebuild the figure in a spreadsheet for each lender or investor, in the format each asks for. Each report matches its reader. None of them traces back to the ledger, and the board sees yet another number.

The second path depends on the loan system recording which demand each receipt cleared. If that split is missing today, the first months of reporting it will take manual work to backfill.

Lokta’s loan management system keeps what each borrower owes and when, what was paid and how it was appropriated, on a double-entry, event-sourced ledger, with receipts applied in the order each product sets. The ledger comes from the team that wrote Apache Fineract.

Frequently asked questions

What is the collection efficiency formula?

There is more than one. Current collection efficiency divides what was collected against this month's instalments by this month's billing. Monthly collection efficiency, as ICRA defines it, divides all collections in the month, including overdue amounts, by this month's billing. Cumulative collection efficiency divides collections since a start date by billing since that date, sometimes with opening overdues added to the denominator.

Why is collection efficiency above 100%?

Because the formula allows it. When collections of older arrears are added to the numerator but the denominator is only this month's billing, a month in which borrowers catch up on past dues can show more than 100%. That can be a good month, but it can also mean the book is carrying arrears from earlier months. Report current collection efficiency alongside it.

What is X bucket collection efficiency?

It measures how much of the billing due from accounts that were current when the month began was collected. Lenders use it to watch borrowers who were up to date when the month began, where a first miss is the earliest sign of stress. The same care about numerator and denominator applies: say whether it counts only same-month collections against the same month's billing.

Do prepayments count in collection efficiency?

That is a policy choice, and it should be stated with the number. Including foreclosures and prepayments raises total collections without any change in how borrowers are paying their instalments. Lenders who report to rating agencies or investors should use the definition in the transaction documents and keep prepayments separate where the definition allows.


Sources:

Bring us your live book

See what agents can do after approval.

Talk to us
Founder-led adoption

Adopt the agentic loan servicing platform.

Lokta is built for enterprise deployment, VPC or single-tenant cloud, with an audit trail in every state change. We work with a select group of institutions through a founder-led model: deep adoption, deliberate scope, a delivery window the team commits to in writing.