Average loan size

Average loan size is the mean value of the loans in a portfolio. It is one of the most widely quoted metrics in lending, and one of the most inconsistently defined, because "average loan size" can mean four different numbers depending on two choices: whether you measure amounts disbursed or balances outstanding, and whether you divide by loans or by borrowers.

Stated precisely, the metric answers the question: how large is a typical exposure in this portfolio? That single figure carries a lot of information. It signals which market segment the lender serves, it largely determines whether the unit economics work, and in microfinance it is treated as a proxy for how poor the clients are.

The four variants

Average loan disbursed = Total value disbursed in a period ÷ Number of loans disbursed in that period

Measures the size of new lending. Useful for tracking what the lender is currently writing and for comparing across products or officers.

Average loan balance outstanding = Gross loan portfolio ÷ Number of active loans

Measures the size of current exposures. Always lower than average disbursed for an amortising portfolio, because loans are partly repaid.

Average loan balance per borrower = Gross loan portfolio ÷ Number of active borrowers

Higher than the per-loan figure wherever borrowers hold more than one loan. This is the variant used for outreach analysis, since it reflects total exposure to a person rather than to an account.

Median loan size — the midpoint rather than the mean. Often more informative than the average, for reasons covered below.

Worked example

A lender with a gross portfolio of 2,400,000 outstanding across 800 active loans held by 640 active borrowers, having disbursed 1,200 loans totalling 4,800,000 over the past year:

• Average loan disbursed: 4,800,000 ÷ 1,200 = 4,000
• Average loan balance outstanding: 2,400,000 ÷ 800 = 3,000
• Average loan balance per borrower: 2,400,000 ÷ 640 = 3,750

Three defensible answers from one portfolio. Any external comparison that does not state which one is being used is not a comparison.

Average loan size as a depth-of-outreach indicator

In microfinance the standard practice is to express average loan balance per borrower as a percentage of gross national income per capita:

Depth of outreach = Average loan balance per borrower ÷ GNI per capita

The logic is that poorer clients borrow smaller amounts relative to national income, so a low ratio suggests the institution reaches deeper into the market. A widely used rule of thumb treats a ratio below about twenty percent as an indicator of serving very poor clients, with higher ratios indicating progressively less poor segments.

Dividing by GNI per capita is what makes the figure comparable across countries — an average balance of 400 means something very different in a low-income economy than in a middle-income one.

The indicator has well-known weaknesses and should be read as a signal rather than a measurement:

• Loan size reflects what a client can service, which is influenced by product design, competition and the lender's own ceilings, not only by poverty.
• National GNI per capita says little about the income distribution the lender actually operates in — a lender serving the poorest quintile of a middle-income country will show a high ratio.
• The measure captures the client at a point in the loan cycle, so a portfolio weighted toward first-cycle borrowers looks deeper than the same institution measured a year later.
• It says nothing about whether clients are better off.

Why average loan size matters

Unit economics

A large part of the cost of a loan is fixed per account: assessment, documentation, disbursement, monitoring, collection, statements. Those costs barely change with the amount lent. So the smaller the average loan, the higher the cost per unit of portfolio, and the higher the interest rate required to cover it.

This is the central economic fact about small-balance lending, and it explains most of the rate differential between microfinance and mainstream credit. It also means average loan size is the single strongest determinant of what an institution's operating expense ratio can plausibly be.

Officer productivity

Caseload for a loan officer tracks inversely with loan size. Officers handling small group loans manage far more borrowers than officers assessing individual SME exposures. Comparing officer productivity across institutions without adjusting for average loan size produces meaningless conclusions.

Concentration risk

A rising average, particularly a rising maximum alongside it, means a growing share of the portfolio depends on a small number of borrowers. Single-obligor limits under prudential guidelines exist precisely to constrain this.

Segment identification

Average loan size is the quickest indicator of what a lender actually does, regardless of how it describes itself. An institution calling itself a microlender with a five-figure average balance is doing SME lending.

What drives average loan size

Product mix. Emergency loans, school fees advances and salary advances pull the average down. Asset finance, business and agricultural loans pull it up.

Client tenure. Progressive lending means each successful cycle unlocks a larger loan, so an ageing client base raises the average without any change in policy.

Methodology. Village banking and solidarity group lending operate at small individual amounts by design. Individual lending starts higher and graduates further.

Competition. Where several lenders pursue the same clients, loan sizes tend to rise as institutions compete on amount — which is also how over-indebtedness builds in a market.

Deliberate strategy. Moving upmarket raises the average quickly, because a small number of large loans shifts the mean substantially.

Inflation and currency. In a high-inflation economy the nominal average rises every year with no change in real terms at all. Comparing nominal averages across years in such a market is misleading.

Average loan size and mission drift

Rising average loan size is the most commonly cited evidence of mission drift — an institution founded to serve poor clients gradually shifting toward better-off ones, because larger loans are cheaper to administer per unit lent and carry lower loss rates.

The concern is legitimate, but a rising average is weak evidence on its own. At least four benign explanations produce the same signal:

• Client graduation. Existing borrowers succeeding and taking larger loans is the outcome the model is designed to produce.
• Inflation. Nominal growth with no real change.
• Portfolio maturity. A younger portfolio has proportionally more first-cycle loans.
• Product expansion. Adding an asset finance or SME product raises the blended average while the original product is unchanged.

Distinguishing drift from these requires looking behind the average: the distribution of loan sizes rather than the mean, the average size of first-cycle loans specifically, the number of clients below a defined threshold, and whether the small-loan segment is shrinking in absolute terms or only in share.

Common errors

Comparing incomparable variants. Average disbursed against average outstanding, or per-loan against per-borrower. The differences are large and systematic.

Reporting the mean where the distribution is skewed. A portfolio of 199 loans of 1,000 and one loan of 500,000 has an average of 3,495 and a median of 1,000. The mean describes no actual borrower. Where a few large exposures sit alongside many small ones — which is most portfolios that have added an SME product — the median and the distribution are more honest than the average.

Point-in-time rather than period-average balances. A single month-end snapshot is distorted by disbursement and repayment timing. Averaging monthly balances across the period is more stable.

Inconsistent denominators. Including or excluding written-off loans, loans in arrears, dormant accounts and undisbursed approvals changes the count. The rule should be documented and applied identically each period.

Confusing borrowers with accounts. Where one borrower holds three loans, dividing by loans understates exposure per person by two thirds.

Ignoring group structure. In group lending the loan may be booked to the group. Whether the borrower count is groups or members changes the figure by an order of magnitude.

Mixing currencies. A multi-currency portfolio needs a stated conversion basis and date, or the average moves with the exchange rate rather than with lending.

Nominal comparison across years in inflationary markets. Without deflating, the metric measures inflation.

How to compute it defensibly

• State which of the four variants you are reporting, every time.
• Use period-average balances rather than a single snapshot.
• Define the active loan and active borrower counts explicitly, and hold the definition constant.
• Report the median and the distribution alongside the mean.
• Segment by product, branch, methodology and client cycle before drawing conclusions from a blended figure.
• Present real as well as nominal figures where inflation is material.
• For outreach analysis, use average balance per borrower over GNI per capita, and treat the result as directional.