The Microfinance Sector Playbook

First published 23 Aug 2026 · Last verified 29 Aug 2026

In the spring of 2018, a branch manager in Jhapa district was proud to report that his laghubitta had disbursed loans to nine hundred new borrowers that quarter, doubling the branch's portfolio in under a year. His regional office cheered the number. Head office, listed on NEPSE and paying a lucrative bonus share, cheered louder. What none of the cheering acknowledged was that roughly three hundred of those nine hundred "new" borrowers were already carrying loans from two other microfinance institutions operating in the same ward, borrowing from one lender to service the instalment on another. Four years later, when the merry-go-round stopped, that branch's portfolio at risk crossed thirty percent, its parent company's share price had been cut in half, and Nepal Rastra Bank was writing the strictest microfinance directives the sector had ever seen. This is the pattern every serious NEPSE investor needs to understand before touching a single laghubitta share: the business model, when it works, is one of the most powerful poverty-reduction and profit-generation engines Nepal has ever built. When it is mismanaged, it fails in a very specific and very predictable way, and that failure shows up in the numbers months before it shows up in the share price.

Deepa Sharma, a research analyst at a mid-sized Kathmandu brokerage, has spent the better part of a decade watching this cycle repeat in miniature. She was in university during the first branch-expansion boom, watched several classmates' parents in the Tarai take out group loans they could not really afford, and later built her career partly on the conviction that microfinance stocks are among the most misunderstood counters on the exchange — priced by retail investors as if they were fast-growing consumer banks, when in fact they are highly levered, socially exposed, regulation-dependent institutions that require an entirely different underwriting lens. Her approach to the sector, refined across several boom-bust cycles, forms the spine of this chapter. This is the microfinance sector playbook: how the model works, why it keeps breaking, how Nepal Rastra Bank has responded, and exactly what to check before you buy or hold a single microfinance share on NEPSE.

Lesson 73.1 — The Grameen Model, Nepali Style

Nepal's microfinance institutions did not invent group lending; they imported and adapted it. The Grameen Bank model pioneered in Bangladesh rests on a simple insight: the rural poor, especially women, are creditworthy but collateral-less, and peer accountability inside a small group can substitute for the collateral a commercial bank would otherwise require. Nepal's laghubitta bittiya sanstha sector, built up from the 1990s onward with donor support, cooperative roots, and later NRB licensing, took that model and localized it into what practitioners call the joint liability group, or JLG, structure.

In practice, a field officer organises borrowers — overwhelmingly women — into groups of five to seven, and several groups into a "centre" of twenty-five to forty members that meets weekly or biweekly, usually in a member's courtyard or a rented room. Loans are small, typically ranging from a few thousand rupees for a first cycle to several hundred thousand rupees for a mature, repeat borrower running a small enterprise. Repayment is collected in cash at the centre meeting, in front of the group, and group members are jointly and socially — though rarely legally in the strict sense — on the hook if one member defaults; social pressure, not a lien on property, is the enforcement mechanism. Interest is charged on a declining balance, disbursement is preceded by a compulsory savings requirement, and the field officer's own compensation and career progression is tied heavily to portfolio growth and, in principle, to portfolio quality.

Nepal's listed microfinance sector splits into two functional layers that investors frequently conflate. The first is the retail layer: the deprived-sector lenders that actually run centre meetings and disburse to end borrowers — names like Chhimek, Nirdhan Utthan, Swabalamban, Mithila, Naya Nepal, Vijaya, RSDC, Manushi, Sworojgar, and dozens of smaller district and regional players, most of them listed on NEPSE and together forming the microfinance sub-index. The second is the wholesale layer: institutions such as Sana Kisan Bikas Bank and the erstwhile RMDC, which historically lent to retail MFIs and cooperatives rather than to end borrowers directly, functioning more like a refinancing conduit than a Grameen-style lender. An investor who treats a wholesale lender's credit risk profile as identical to a retail centre-meeting operator's is making a basic category error; their exposure to over-indebtedness, branch cost structure, and regulatory caps differs meaningfully even though both trade under the "microfinance" label.

The deprived-sector lending mandate is the other structural fact every investor needs internalized. Nepal Rastra Bank requires commercial banks, development banks, and finance companies to channel a fixed percentage of their loan portfolio to the "deprived sector" — broadly, low-income and marginalized borrowers — and banks satisfy much of this obligation by wholesale-lending to, or investing in, microfinance institutions rather than building their own branch networks into remote wards. This wholesale funding line is a double-edged structural feature. On one hand it has historically supplied MFIs with relatively cheap, policy-driven capital, supporting rapid balance sheet growth. On the other hand, it means MFI funding costs and availability are partly a function of commercial bank credit policy and NRB's deprived-sector quota rules, not purely of the MFI's own creditworthiness — a variable that can tighten or loosen the sector's cost of funds independent of anything the MFI's management does.

Lesson 73.2 — Reading the Vital Signs: PAR, OSS, and the Metrics That Matter

Deepa's first rule for any microfinance counter is that the headline numbers a company chooses to publish in its glossy annual report — net profit, EPS, return on equity — are the last numbers she looks at, not the first. Profit in a lending business is an opinion; it depends entirely on how aggressively or conservatively the company recognises bad loans and provisions against them. The numbers that tell the truth about a microfinance book sit further down the disclosure, in portfolio quality and efficiency metrics that most NEPSE retail investors never open the annual report far enough to find.

Portfolio at risk, or PAR, is the single most important credit quality metric in microfinance and the one most different from how a commercial bank investor is trained to think. PAR30 measures the percentage of the outstanding loan portfolio that has at least one instalment overdue by thirty days or more; PAR90 does the same at ninety days. Crucially, PAR is not calculated as overdue amount divided by total portfolio — it is calculated as the entire outstanding balance of any loan with an overdue instalment divided by total portfolio. A single missed instalment on a large loan puts the whole loan's balance into the PAR numerator, which is why PAR ratios can move sharply even when actual missed cash amounts are still small; it is a leading indicator built deliberately to be sensitive.

Operating self-sufficiency, or OSS, measures whether an MFI's operating income covers its operating and financial costs without subsidy — an OSS above 100 percent means the institution earns enough from its lending operations to sustain itself commercially; below 100 percent, it is structurally dependent on grants, subsidized wholesale funds, or capital infusions to keep operating, a red flag for any institution now expected to behave like a fully commercial, dividend-paying, NEPSE-listed financial company. Cost per borrower and average loan size, read together, tell you where an MFI sits on the mission-drift spectrum: a rising average loan size alongside a falling number of active borrowers per branch usually signals the institution has quietly pivoted from serving genuinely poor, first-time borrowers toward larger, less labor-intensive loans to already-banked small entrepreneurs — a shift that can improve near-term margins while eroding the very deprived-sector mandate and social mission that gives the institution its regulatory license and funding advantages in the first place.

The provisioning line deserves its own emphasis because it is where reported profit and real economic profit diverge most in this sector. NRB's directives require graduated provisioning against loans classified as pass, watchlist, substandard, doubtful, and loss, with provisioning rates rising sharply as a loan ages further into delinquency — but the classification itself involves judgment, and a management team under pressure to protect the dividend has every incentive to restructure or reschedule a stressed loan just before it crosses a classification threshold, resetting its aging clock without actually improving the borrower's ability to repay. Deepa's habit, therefore, is to read at least three consecutive years of an MFI's loan loss provision line alongside its rescheduled and restructured loan disclosure, because a pattern of provisions that stay suspiciously flat while restructured loans quietly climb is a company managing the appearance of asset quality rather than the underlying reality of it.

Lesson 73.3 — Boom, Bust, and the Lessons of 2015-2022

To understand why Nepal's regulators now treat microfinance with the wariness usually reserved for cooperatives, an investor needs the historical arc. Following the 2015 earthquake and the subsequent economic recovery, Nepal's microfinance sector entered a period of extraordinarily rapid branch expansion. Licensed laghubitta institutions, flush with wholesale funding from banks eager to meet their deprived-sector quotas and encouraged by a listing boom that rewarded portfolio growth with rich valuations on NEPSE, raced to open branches in the same districts, sometimes the same wards, as competitors. It was not unusual, by the late 2010s, for a single rural settlement in the eastern or central Tarai to have field officers from four or five different microfinance institutions cycling through on different days of the week, each running their own centre meetings, each unaware of — or willfully ignoring — how many other institutions' loans a given borrower was already carrying.

This is the phenomenon practitioners call multiple lending or loan stacking, and it is the single most reliable precursor to a microfinance sector crisis anywhere in the world, from Andhra Pradesh in 2010 to Cambodia in the late 2010s to Nepal itself in the early 2020s. A borrower who takes a second loan to service the instalment on her first is not expanding her enterprise; she is running a personal Ponzi scheme that only the continued availability of new credit keeps from collapsing. As long as MFIs kept expanding and disbursing, the stacking was invisible in the portfolio quality numbers — repayment rates stayed high because borrowers had fresh loan proceeds to make old payments with. The moment growth slowed, even slightly, the whole structure was exposed at once, because there was no new borrowing left to paper over instalments that borrowers could never actually afford from their own cash flow.

The uncomfortable truth Deepa emphasises to newer analysts is that the crisis was not primarily a story of external shock; it was a business model problem that the pandemic merely revealed earlier than it otherwise would have. Branch density had outrun genuine market depth in dozens of districts. Field officer incentive structures rewarded disbursement volume over portfolio quality. Head offices under pressure from NEPSE shareholders to sustain the growth rates that justified rich price-to-book multiples kept pushing branches to open in already-saturated areas rather than genuinely underserved ones. And crucially, no credit information sharing mechanism existed that would let one MFI's loan officer see that a prospective borrower was already carrying loans from three other institutions — the single structural fix that, once regulators forced it, did more than any interest rate cap to slow the stacking problem.

Lesson 73.4 — NRB's Regulatory Vice: Caps, Capital, and Consolidation

Nepal Rastra Bank's response to repeated boom-bust cycles in microfinance has been a steadily tightening set of directives that any investor in this sector must track as closely as the companies' own financials, because regulatory action here has moved share prices far more violently than earnings surprises ever have.

The starting point is interest rate control. As far back as 2016, NRB capped the interest rate spread microfinance institutions could charge over their cost of funds at roughly seven percentage points, an explicit acknowledgment that MFIs' funding cost advantage from cheap wholesale deprived-sector credit should not simply be captured as margin at the expense of poor borrowers paying effective rates well above what a commercial bank customer would ever be charged. Over the following years NRB layered on an absolute ceiling as well, eventually capping the maximum lending rate microfinance institutions could charge end borrowers at around fifteen percent, regardless of the spread calculation, a level many MFI managements complained compressed margins to the point of threatening sustainability, particularly for smaller, higher-cost-to-serve institutions operating in remote hill districts.

By 2023, NRB's own internal study had concluded that a flat rate cap was a blunt instrument poorly suited to a sector with such wide variation in operating cost structures between an efficient Tarai-based lender and a high-cost hill-district operator, and recommended moving toward a spread-based, base-rate-linked framework instead. That recommendation became policy through the FY 2081/82 and FY 2082/83 monetary policy cycles: NRB's mid-term review formally shifted microfinance lending rates onto a base rate plus premium structure, similar in spirit to the base-rate framework long used for commercial bank lending rates, with new guidelines taking effect from the start of the Nepali fiscal year in Shrawan. The practical effect for investors is that microfinance net interest margins are no longer governed by a single, easily modelled flat percentage cap; they now move with the same base-rate mechanics that drive commercial bank margins, meaning changes in the banking system's overall cost of funds pass through to MFI margins in ways that require the same base-rate tracking discipline covered elsewhere in this book's banking sector chapters.

Capital adequacy and ownership rules form the second regulatory front. In the wake of the 2021-2022 stress episode, NRB progressively raised minimum paid-up capital requirements for microfinance institutions, forcing an entire tier of small, often single-district laghubittas either to raise fresh capital, merge with a stronger peer, or exit the sector. Deadlines for meeting the higher capital thresholds have repeatedly been extended and then enforced, and companies that missed successive deadlines found themselves under direct NRB pressure to pursue forced mergers rather than continue operating under-capitalised. Layered on top of the capital rules, NRB has separately tightened scrutiny of major shareholders and promoter groups in microfinance institutions, restricting related-party lending and cross-holdings that had previously let a handful of promoter families control multiple MFIs simultaneously — a structure that, when it existed, made loan stacking across "competing" institutions considerably more likely than genuine market competition would suggest, since the institutions were not truly independent in their credit decisioning even though they looked independent on NEPSE's ticker list.

The debt-to-equity and leverage dimension is the third front, running in parallel with capital rules: because microfinance institutions fund their loan books overwhelmingly through wholesale borrowing from banks and other financial institutions rather than public deposits, NRB directives constrain how much an MFI can borrow relative to its own core capital, explicitly to prevent a repeat of the pre-2022 pattern where thinly capitalised institutions built loan books many multiples the size their equity base could safely absorb losses against. An investor evaluating any NEPSE-listed microfinance stock should treat its capital-to-risk-weighted-assets ratio and its core capital fund as being every bit as central to the analysis as an equivalent metric would be for a commercial bank — arguably more so, given microfinance's structurally thinner margin for error against social and credit shocks.

Finally, the merger and consolidation directives themselves warrant a table, because the trend line matters more to an investor's holding-period thesis than any single deadline. NRB has explicitly signalled — through statements requiring stricter scrutiny of merger and acquisition proposals, criteria updates on dividend distribution and nationwide operating status tied to scale, and repeated capital deadline enforcement — that it intends the sector to shrink from the dozens of small, often sub-scale institutions that proliferated during the 2010s branch race into a smaller number of larger, better-capitalised, more professionally governed entities.

Lesson 73.5 — Early Warning Signs and the Over-Indebtedness Checklist

The hardest part of evaluating a microfinance stock is that the earliest warning signs of client over-indebtedness never appear in the profit and loss statement first; they appear in operational disclosures that most retail investors skip entirely. Deepa's early-warning routine focuses on five specific signals, read together rather than in isolation, because any one alone can have an innocent explanation while the combination rarely does.

The first signal is branch density relative to disclosed operating districts. An MFI's annual report typically discloses the number of districts and the number of branches it operates. When branch count per district climbs faster than the district's rural population or economic activity would plausibly support — particularly in districts already known to host several competing MFI branches — that is the geographic signature of the same saturation dynamic that preceded the 2021-2022 stress episode. The second signal is the trend in average loan size per borrower relative to the trend in active borrower count. A rising average loan size accompanied by a flattening or falling active borrower count can indicate healthy graduation of existing clients to larger enterprise loans — but it can equally indicate that the institution has quietly shifted from genuine deprived-sector outreach toward larger, easier-to-process loans to already-served clients, a pattern regulators increasingly scrutinize as mission drift.

The third signal is the restructured and rescheduled loan disclosure, tracked over at least three years rather than a single snapshot. A company whose restructured loan balance is rising while its reported PAR stays flat or falls is very likely managing its classification rather than genuinely improving asset quality — restructuring resets a delinquent loan's aging clock without necessarily restoring the borrower's actual capacity to repay. The fourth signal is staff turnover and, where disclosed, borrowers per field officer. Centre meeting collection depends on field officers who know their borrowers personally; a branch network expanding faster than it can train and retain experienced field staff is a branch network whose social-collateral enforcement mechanism is quietly weakening even while headline growth numbers look strong. The fifth signal, harder to find but worth the effort, is any qualitative disclosure or news flow about borrower protests, loan write-off campaigns, or local government intervention in a company's specific operating districts — these local flashpoints reliably precede sector-wide stress by one to two reporting cycles.

Deepa also keeps a simpler heuristic close at hand for a quick first pass on any microfinance stock she has not previously covered: compare the institution's disclosed average loan size and its operating district list against the National Statistics Office's district-level poverty and remittance-dependency data, and against how many other listed MFIs disclose branches in the same districts. A district with five overlapping MFI branch networks, high remittance dependency, and an average loan size that has more than doubled over three years is, in her experience, one of the more reliable geographic fingerprints of a book quietly accumulating stacking risk — even when every individual company's own disclosed PAR still looks comfortable.

Lesson 73.6 — The Microfinance Stock Evaluation Checklist

Bringing the whole playbook together, here is the sequence Deepa walks through, in order, for any NEPSE-listed microfinance stock before she will recommend a position to a client — retail or institutional.

She starts with regulatory compliance status: has the institution met NRB's current minimum paid-up capital threshold, and if not, what is its stated timeline and mechanism — rights issue, merger, or promoter capital infusion — for closing the gap? A company still short of the threshold with no credible near-term plan is not a stock to hold through the deadline; it is a stock to wait out from the sidelines until the capital or merger question resolves, because the dilution or swap-ratio outcome will dominate any operational thesis in the interim.

Second, she checks the interest rate and margin picture against the current base-rate-plus-premium framework rather than against any older flat-cap assumption, confirming the company's disclosed net interest margin is consistent with what the current regulatory regime allows and sustainable given its funding cost structure — a hill-district operator with structurally higher branch operating costs needs a correspondingly higher margin to sustain OSS above 100 percent, and if regulatory caps are compressing that margin below sustainable levels, that is a structural earnings risk no amount of operational excellence can fully offset.

Third, she pulls at least three to five years of PAR30, PAR90, provisioning coverage, and restructured loan trends together, looking specifically for the divergence pattern described in Lesson 73.5 — rising restructuring alongside flat or improving headline PAR is treated as a disqualifying red flag regardless of how attractive the reported profit trend looks.

Fourth, she maps the company's disclosed operating districts and branch counts against known competitor overlap and against remittance-dependency and poverty data, screening specifically for the geographic over-indebtedness fingerprint — heavy branch overlap combined with rapidly rising average loan size in the same districts.

Fifth, she examines ownership and governance: who are the major shareholders and promoter group, do they hold stakes in other microfinance institutions that could create related-party lending or cross-subsidization risk, and has the company been named in any NRB related-party or major-shareholder tightening action.

Sixth, she checks OSS, cost per borrower, and borrowers per field officer trends together as an efficiency and monitoring-capacity check, looking for cost discipline that has kept pace with branch expansion rather than growth that has outrun the institution's actual capacity to know its borrowers.

Seventh and last, only once every prior step has cleared, does she look at valuation — price to book relative to sector peers, dividend history and sustainability given the OSS and provisioning picture, and any pending merger or capital-raise dilution that would change the per-share economics before the position could be expected to pay off. A microfinance stock that clears the first six steps and trades at a discount to sector peers on price to book is, in her experience, one of the more reliable value opportunities on NEPSE precisely because the sector's headline reputation — still coloured by the 2021-2022 stress episode — keeps many retail investors away from names that have since genuinely cleaned up their books and met the new capital and governance bar.