The Borrower Who Never Gets a Knock on the Door

You live four kilometers off the main road. Your plot is small, your income seasonal, and you have never borrowed formally in your life. A microfinance institution operates in your district. Its field officers pass through the nearest market town every fortnight. And yet, in three years, not one of them has shown up at your door.

This is not an accident. It is policy. Just not written policy.

The mechanics of exclusion inside a microfinance institution rarely appear in any loan manual. They live instead in the performance targets set for field officers, the portfolio quality ratios that branch managers report upward, the incentive structures that determine whether a loan officer gets promoted or quietly reassigned. When those internal governance levers are calibrated in one direction, certain borrowers become, in the institution's practical logic, invisible. Understanding exactly how that happens is worth more than any amount of hand-wringing about financial inclusion in the abstract.

Portfolio at Risk and the Math That Drives a Field Officer's Morning

Start with a single number: Portfolio at Risk, almost universally abbreviated to PAR30, meaning the share of an institution's loan book where repayments are more than thirty days late. Most MFIs report this figure to donors, regulators, and rating agencies. A PAR30 above five percent is uncomfortable. Above ten, it triggers alarm. That number, small enough to fit on a dashboard, carries more weight than any mission statement the institution has ever printed.

Now put yourself in the shoes of a field officer named Amara, covering a cluster of villages in a semi-arid region. Amara has forty active borrowers and a monthly disbursement target of fifteen new loans. Her supervisor reviews her PAR30 every week. Her end-of-year bonus, and her chance of becoming a senior officer, depends on it staying below three percent.

Amara knows, from experience and from watching colleagues, that borrowers in the two villages closest to the tarmac road repay reliably. They grow irrigated vegetables, sell at a predictable market, and have family members in town who can cover a gap. The households beyond the ridge are different. Primarily rain-fed farmers. Their income arrives in two pulses per year, and a drought, a pest, a delayed harvest means a weekly repayment schedule becomes impossible for six weeks running. Not because they are dishonest. Because the cash literally isn't there yet.

Amara doesn't hate those households. She just doesn't visit them. The math of her personal incentive structure, set by governance decisions made in a head office she has never seen, tells her not to.

The Governance Decisions That Live Three Levels Above the Village

Trace Amara's incentive back to its source and you find a chain of deliberate choices.

First: the board's risk appetite statement. Most MFI boards set a maximum acceptable PAR ratio and build that figure into management scorecards. A board dominated by commercial investors tends to set a tighter ceiling than one with a development-finance mandate, and that ceiling flows down into branch targets, which flow into field officer appraisals. By the time the number reaches Amara, it has the force of gravity.

Second: the loan product design approved by the credit committee. Weekly or fortnightly repayment schedules, which became something close to orthodoxy in the Grameen-influenced model, work well for small traders with daily cash flow. They are structurally hostile to smallholder farmers whose income is lumpy and seasonal. An MFI that offers only one repayment cadence is pre-selecting its clientele through product design long before any field officer makes a routing decision. This is not a side effect. It is the outcome.

Third: the cost-per-loan metric that finance departments report to the board. Processing a loan of fifty dollars costs nearly as much in staff time as processing one of five hundred. Institutions under pressure to demonstrate operational sustainability quietly raise their minimum loan sizes, or require collateral documentation that landless laborers cannot produce. Neither policy says "do not serve the poorest." Both achieve that outcome, reliably, quarter after quarter.

Fourth, and perhaps the least discussed: the geographic incentive structure. Many MFIs pay field officers a flat travel allowance regardless of how far they ride. Amara gets the same reimbursement whether she visits a borrower two kilometers away or twelve. The rational response is to cluster her portfolio as tightly as geography allows. The remote hamlet loses, not because anyone decided against it, but because nobody decided for it.

What Happens Inside a Lending Group When Governance Pressure Mounts

Group lending was supposed to solve the exclusion problem through social collateral. The theory: neighbors who guarantee each other's loans will screen out bad risks themselves, and the field officer's job is simply to facilitate. In practice, governance pressure warps even this mechanism, bending it into something its designers would not recognize.

Consider a village solidarity group of twelve women. Two of them are among the poorest households in the area: thin land holdings, no male wage earner, irregular income. When the MFI's field officer is under pressure to maintain a clean portfolio, she signals, sometimes explicitly, sometimes just through which applications she processes quickly, that the group's renewal loan depends on uniform repayment. The existing members draw the obvious conclusion. They stop inviting the two poorest women to meetings. They suggest, gently, that those women "aren't ready yet."

The field officer's PAR30 stays clean. The group cohesion survives. And the two most vulnerable potential borrowers are screened out by their own community, because the institution's governance structure created the incentive that made their neighbors do it. Think of it as a filter shaped like a friendship: soft enough to feel voluntary, hard enough to be nearly impenetrable.

Researchers in development finance sometimes call this the elite-capture problem at the micro level. It isn't corruption. It is rational behavior by everyone in the chain, all the way up to the board, each responding sensibly to the incentive they face.

The Honest Tension at the Center of This

None of the above means that MFIs are villains, and anyone who argues otherwise hasn't read a set of audited accounts lately. The pressure to maintain portfolio quality is not invented by callous executives. An institution with a collapsing loan book cannot pay its staff, cannot fund new disbursements, and eventually closes. When an MFI closes, all of its borrowers lose access, not just the marginal ones. The governance mechanisms that exclude the hardest-to-reach borrowers are, in many cases, the same mechanisms keeping the institution solvent enough to serve anyone at all.

So ask yourself: who should bear the cost of that tension?

A tightly commercial MFI, answerable to equity investors expecting twelve-percent returns, will calibrate its governance toward portfolio quality and operational efficiency. The rural poor become externalities, priced out by process rather than prejudice. A development-finance-backed institution with a patient capital structure can afford a higher PAR tolerance and a seasonal loan product that a commercial MFI would never approve. The difference in outcomes for borrowers like the household four kilometers off the main road is not marginal. It can be the difference between access and permanent exclusion.

Some institutions have tried to resolve this through tiered product design: a standard weekly-repayment product for traders, a separate agricultural loan with bullet or semi-annual repayment for farmers, each with its own PAR tracked separately so that a bad harvest doesn't contaminate the whole book. The Grameen Bank's own evolution over decades reflects exactly this kind of product differentiation. But tiering requires governance commitment at the board level to absorb the higher operational complexity and the higher short-term risk. Without that commitment, it dies in committee, usually sometime between the second slide and the lunch break.

Found in the Fine Print: What Governance Documents Actually Signal

If you want to know which borrowers an MFI will systematically bypass, read three documents in order: the staff performance appraisal rubric, the credit policy's minimum loan size and eligible collateral list, and the board's stated PAR tolerance. Those three documents, taken together, describe the institution's actual client with far more precision than any mission statement about serving the unbanked.

Field officers are not the problem. They are the final expression of decisions made much higher up. Amara's routing choices are entirely predictable once you know her bonus structure. Change the bonus structure and you change the routing. Change the PAR tolerance and you change who gets a product designed for their actual income pattern. Change the board's composition and mandate and you change the tolerance. Each link in the chain is mechanical, not moral.

Any institution reporting above eighty-five percent portfolio quality on a weekly-repayment product in a rain-fed farming region almost certainly did not get there by serving the most vulnerable households. It got there by not serving them. That figure deserves scrutiny, not applause.

The borrower four kilometers off the road isn't waiting for a better field officer. She's waiting for a board that counts her in the denominator.