Collection efficiency / collection rate

Collection efficiency (also called the collection rate or CE) measures the proportion of scheduled loan repayments that a lender actually collected during a period. It is expressed as a percentage of the amount due.

Collection efficiency = Amount collected / Amount due × 100

It is a flow metric: it describes repayment behaviour during a specific window, in contrast to stock metrics such as portfolio at risk, which describe the condition of the book at a point in time. That distinction is the source of most of its usefulness and nearly all of its misuse.

Collection efficiency is the standard operational metric for field-collections lending — microfinance, group lending, asset finance and consumer credit — because it responds immediately to what is happening on the ground, while PAR takes weeks to move.

The Formula Variants

There is no single agreed definition, and the same portfolio can honestly report figures ten percentage points apart depending on which one is used. The main variants:

1. Current collection efficiency (CCE)

Also called billing efficiency or demand collection efficiency.

CCE = Collections against the current period's demand / Current period demand × 100

Only instalments falling due in the period count, and only payments applied to those instalments. Arrears recovered from earlier periods and prepayments are excluded. CCE cannot exceed 100%. This is the cleanest measure of whether current obligations are being met on time.

2. Total (or gross) collection efficiency

Total CE = All collections received in the period / Current period demand × 100

The numerator includes recovery of past arrears, prepayments, early settlements and foreclosure proceeds. Total CE can and frequently does exceed 100%, which makes it a flattering headline number and a poor standalone indicator.

3. Cumulative collection efficiency

Cumulative CE = Total collected since origination / Total demand raised since origination × 100

Applied to a cohort or vintage rather than a period. Useful for assessing how a specific origination month has performed over its life, but slow-moving and insensitive to recent change.

4. Bucket-wise collection efficiency

Collection efficiency calculated separately for each delinquency bucket — current, 1–30 days past due, 31–60, and so on. Sometimes reported as resolution or roll-back rates: the proportion of accounts in a bucket that returned to current during the period.

This is the most diagnostically useful variant, because a single blended figure tells you nothing about where collections are failing.

Worked Example

A lender's loan book for a given month:

• Scheduled instalments falling due this month (demand): 1,000,000
• Collected against this month's demand: 920,000
• Collected against arrears from previous months: 60,000
• Prepayments and early settlements received: 40,000
• Total cash collected: 1,020,000

Current collection efficiency = 920,000 / 1,000,000 × 100 = 92%
Total collection efficiency = 1,020,000 / 1,000,000 × 100 = 102%

The headline figure of 102% reads as excellent. In fact 8% of the month's obligations went unpaid, and the portfolio added 80,000 to arrears. Only the exclusion of arrear recovery and prepayments reveals it.

Practical rule: report CCE and total CE side by side, and never publish a figure above 100% without stating what is in the numerator.

The Growth Masking Problem

This is the most important limitation of the metric, and the one most often missed.

Newly disbursed loans almost always pay their first few instalments on time. In a rapidly growing portfolio, recent originations make up a large share of current demand, so their near-perfect early performance dominates the blended figure — while older, deteriorating vintages are diluted into invisibility.

The consequence: collection efficiency can hold steady or improve during a period of worsening credit quality, purely because the book is growing. The metric is most reassuring at exactly the point where growth is outrunning underwriting quality.

Two defences:

• Vintage or cohort view. Calculate collection efficiency separately by origination month, so each cohort's performance is visible on its own terms rather than blended with newer lending.
• Exclude first-instalment accounts from the blended figure, or report them separately. A first-payment default rate is itself a high-value underwriting signal.

Collection Efficiency vs Other Credit Metrics

• Collection efficiency (Flow): Measures the share of amounts due that were collected in the period. Responds immediately.
• Portfolio at risk / PAR (Stock): Measures the share of the outstanding portfolio with overdue balances. Responds with a lag.
• Roll rate (Flow): Measures the movement of accounts between delinquency buckets. Responds immediately.
• Write-off ratio (Flow): Measures losses formally accepted in the period. Responds at policy trigger.
• On-time repayment rate (Flow): Measures the share of instalments paid by the due date. Responds immediately.

Collection efficiency and PAR answer different questions and can diverge legitimately. A portfolio can post 98% CCE — strong current collections — while carrying high PAR from a legacy of older delinquency that is no longer generating current demand. The reverse is also possible: low PAR and falling CCE indicate deterioration that has not yet aged into the PAR buckets.

Neither metric is reliable alone. Read them together, with roll rates between them.

Common Measurement Pitfalls

Defining "demand" inconsistently. Demand should mean instalments contractually falling due in the period. Some institutions include carried-forward arrears in demand, which lowers the ratio; others exclude them, which raises it. Both are defensible; mixing them across periods is not.

Payment allocation effects. Where a partial payment is received, how it is allocated across penalties, fees, interest and principal under the waterfall rules determines what counts as "collected against demand." A waterfall that clears fees first can leave the principal instalment technically unpaid despite cash having been received. Allocation logic must be understood before the ratio can be interpreted.

Cash versus accrual cut-off. Payments made on the last day of the month but posted on the first day of the next shift the ratio between periods. Digital collections received outside business hours are a frequent source of this.

Counting non-borrower sources as collections. Amounts applied from compulsory savings, group funds, security deposits or guarantor payments are recoveries from collateral, not evidence of borrower repayment capacity. Including them inflates the ratio and hides deterioration.

Restructuring. Rescheduling a distressed loan reduces or resets its demand, which mechanically improves collection efficiency without any improvement in the borrower's position. Restructured volume should be reported alongside CE.

Not annualising or not fixing the period. Weekly, fortnightly and monthly collection efficiency are not comparable, and short-tenor products with weekly instalments produce very different figures from monthly-repayment portfolios.

Blending products. A group loan portfolio with weekly meetings and an individual monthly-repayment portfolio have structurally different collection dynamics and should not share a single CE figure.

What Drives Collection Efficiency

Repayment channel. Cash collection at meetings depends entirely on staff presence and attendance. Digital channels, standing instructions and mobile money debits remove that dependency but introduce failure modes of their own — insufficient balance, expired mandates, network outages.

Instalment frequency and timing. Matching due dates to the borrower's income cycle — market days, harvest, payday — measurably improves on-time payment.

Reminder discipline. Pre-due reminders are consistently cheaper and more effective than post-due follow-up.

Field staff capacity and turnover. Caseload per officer and officer tenure are both strong predictors. Collection performance often degrades sharply in the weeks after an officer leaves a portfolio.

Group meeting attendance, where the methodology depends on it.

Seasonality. Agricultural cycles, school fee periods and festival seasons produce predictable dips. Comparing month-on-month without a seasonal baseline generates false alarms and false comfort in equal measure.

Incentive design. Staff incentives tied to collection efficiency can produce perverse behaviour: pressuring borrowers into informal borrowing to make an instalment, advancing funds personally, or misallocating payments across accounts. Incentives should reference vintage performance and portfolio quality, not the headline ratio alone.

How to Use It Well

• Report CCE as the primary figure, with total CE alongside and clearly labelled.
• Cut by vintage first. Origination-month cohorts remove the growth masking effect and are the only view that reliably attributes performance to the lending decision that caused it.
• Add bucket-wise resolution rates. Knowing that 92% of current demand was collected is far less actionable than knowing that current-bucket collection is 98% while the 31–60 day bucket resolves at 12%.
• Track first-payment default separately. It is an underwriting metric masquerading as a collections metric, and it is one of the earliest available signals of a screening problem.
• Segment by branch, officer, product, sector and loan cycle. Internal variance is usually wider and more addressable than any period-on-period movement.
• Pair with PAR, roll rates, restructured volume and the write-off ratio. Any one of these can be improved in isolation without a real change in credit quality; all five together are difficult to distort.
• Compare against the same month last year, not the previous month, in seasonal portfolios.