Sector-Specific Sub-Score Adjustments

First published 23 Aug 2026 · Last verified 29 Aug 2026

In Naya Bazaar, there is a tailor named Hari Kaji who has cut cloth for forty years. He keeps one master pattern pinned to his wall — a basic shirt block, with a collar, two sleeves, a yoke, and a hem. Every shirt that leaves his shop starts from that same block. But Hari Kaji never cuts a farmer's work shirt the same way he cuts a bank manager's dress shirt. The farmer needs wider armholes, because he lifts and swings all day. He needs a shorter hem, so it does not catch on a plough. He needs thicker cotton, because it will be washed in a stream and dried on a rock. The bank manager needs a fitted waist, a stiff collar for a necktie, and cloth thin enough to stay cool under a suit jacket. Same block. Same starting measurements for the neck and the shoulder. Completely different final shirt.

If Hari Kaji used the farmer's cut for the bank manager, the shirt would hang like a sack. If he used the bank manager's fitted cut for the farmer, the seams would tear apart in a single planting season. The pattern is not wrong. It is simply generic. It has to be adjusted for the body wearing it and the work that body will do.

The Canon Score works the same way. Chapter 64 gave you one master pattern: seven dimensions, a 0-to-100 scale, and a common language for grading any company on NEPSE. That pattern is correct, and you should not throw it away. But a hydropower company under construction has no revenue at all, so a dimension that measures "profitability" cannot be read the same way it is read for a bank that has been profitable every year since it opened its doors. A microfinance institution lends tiny amounts to thousands of rural borrowers with no collateral, so a governance red flag that matters enormously for microfinance barely registers for a cement manufacturer. An insurance company holds liabilities that will not come due for thirty years, so its "financial strength" cannot be judged the way you would judge a trading company that turns over its inventory every ninety days.

This chapter is Hari Kaji's fitting room. You already have the master block from Chapter 64. Now you will learn how to cut it differently for four of NEPSE's most important sectors: banks and financial institutions, hydropower, microfinance, and insurance. By the end, you will be able to look at any NEPSE-listed company, recognise which sector's "cut" it needs, and adjust the generic seven-dimension score so that it tells you the truth about that specific company — not a generic truth that happens to be wrong for the business in front of you.

Lesson 65.1 — Why One Scoring Framework Cannot Fit Every NEPSE Sector

Start with a simple question. What does "financial strength" mean?

For a trading company, it might mean: does it have enough cash to pay its suppliers, and is its debt small compared to its equity? For a commercial bank, that question is almost meaningless. A bank's entire business model is to hold enormous amounts of other people's money (deposits) and lend most of it out. A bank with "low debt" is not a strong bank — it is a bank that is barely functioning as a bank. Debt, for a bank, is the raw material of the business, not a warning sign.

This is the core problem this chapter solves. The Canon Score's seven dimensions — Financial Strength & Profitability, Governance & Promoter Behaviour, Liquidity & Tradability, Valuation Reasonableness, Sector & Business Model Durability, Growth Trajectory, and Dividend & Capital Return Discipline — are the right categories to think about. But the specific ratios and thresholds that fill each dimension must change depending on what kind of company you are scoring.

Think of it like a doctor's checkup. Every patient gets checked for blood pressure, weight, heart rate, and general fitness — that is the "generic framework" of a physical exam. But the healthy blood pressure range for a seventy-year-old is not the same as for a twenty-year-old athlete. A doctor who used one single number for "healthy blood pressure" across every patient, regardless of age, would misdiagnose half the people who walk through the door. Sector adjustment in the Canon Score is exactly this: recognising that a "healthy" capital adequacy ratio for a bank and a "healthy" debt ratio for a hydropower company under construction are different numbers, measuring different things, on different scales — even though both questions are really asking the same underlying thing: "is this company's financial foundation sound?"

There are three reasons NEPSE sectors need different treatment.

First, some sectors are regulated by an entirely different authority with entirely different rules. Banks, development banks, and finance companies answer to Nepal Rastra Bank (NRB), the central bank. Insurance companies answer to the Nepal Insurance Authority (NIA, formerly called Beema Samiti). These regulators impose their own capital rules, provisioning rules, and disclosure rules — rules that do not exist for a hotel or a manufacturing company. A serious investor has to read those regulatory filings, not just the standard annual report, to score these companies properly.

Second, some sectors have a business model that unfolds in distinct phases with almost nothing in common between the phases. A hydropower company spends years — sometimes seven, eight, or ten years — building a dam, tunnels, and a powerhouse before it generates a single unit of electricity. During that construction period it has essentially no revenue but very large debt. Once it reaches what the industry calls COD — Commercial Operation Date, the day the plant is legally allowed to sell electricity into the grid — its entire financial character flips. Scoring a pre-COD hydropower company on "profitability" is like scoring a pregnant woman on "number of children currently in her arms." The right number is coming, but it is not there yet, and that absence is not itself a red flag.

Third, some sectors carry risks that are invisible in the standard financial statements unless you know to look for them. A microfinance institution's real risk often lives in how loan officers behave in the field — whether they are pressuring borrowers, whether loans are being recycled to hide default, whether the same household is borrowing from five different microfinance companies at once. None of that shows up as a single ratio on the balance sheet. You have to know the sector's specific danger zones.

Part VI of this Canon (Chapters 29 through 34) already walked you through how to read the financial statements of banks, development banks, microfinance institutions, hydropower companies, insurance companies, and manufacturing, trading, and hotel businesses, each in their own accounting logic. This chapter does not repeat that accounting detail. It tells you how to translate what you learned in Part VI into adjustments on the seven-dimension Canon Score you learned in Chapter 64. Think of Part VI as learning to read the individual measurements on a body, and this chapter as learning which of those measurements Hari Kaji actually uses when he is cutting a shirt for that particular body.

One more foundational point before we go sector by sector. Sector adjustment does not mean lowering your standards for a sector you like, or raising them for a sector you are suspicious of. It means using the right ruler. A well-run microfinance institution can still score well on Governance if it meets the sector-appropriate governance bar — even though that bar looks completely different from a bank's governance bar. A poorly built hydropower project can still score badly on Financial Strength even after adjusting for construction-phase norms, if its debt is excessive even by hydropower standards. The adjustment changes what you measure and how you interpret it. It never changes your commitment to being honest about what the numbers say.

Lesson 65.2 — Adjusting the Score for Banks, Development Banks, and Finance Companies

Nepal's banking system has three tiers of deposit-taking institutions licensed by Nepal Rastra Bank: "A" class commercial banks, "B" class development banks, and "C" class finance companies. All three take deposits from the public and lend that money out, and all three are supervised under NRB's capital and provisioning framework, though the exact thresholds differ by class. If you are scoring any of these, four of the seven Canon dimensions need real surgery.

Financial Strength & Profitability. Throw out generic ratios like current ratio or debt-to-equity — they simply do not describe a deposit-taking institution correctly. Replace them with three bank-specific measures.

The first is the Capital Adequacy Ratio (CAR) — a measure of how much of a bank's own money (equity and reserves) it holds as a cushion against its risk-weighted loans, expressed as a percentage. NRB requires Nepali commercial banks to maintain a minimum total capital fund of around 11% of risk-weighted assets under the Basel III framework it has adopted, with a minimum core capital (Tier 1, the highest-quality capital) requirement of roughly 6%, plus an additional buffer that NRB can require in good times to be drawn down in stress. A bank sitting only slightly above the regulatory minimum has very little room for absorbing a bad year of loan losses. A bank sitting comfortably above minimum — say two to four percentage points higher — has real shock-absorbing capacity built in.

The second is the Non-Performing Loan (NPL) ratio — the percentage of a bank's total loan book that is not being repaid on schedule (generally loans overdue by more than 90 days). Nepal's banking sector NPL ratio has been rising in recent years and has been reported around the mid-5% range across commercial banks in 2025, though individual banks range far more widely, with some reporting NPL ratios above 10% and even above 15%. A bank scoring well on this sub-dimension should sit meaningfully below the sector average NPL ratio, and — just as importantly — should show that ratio holding steady or improving over several quarters, not just in one favourable snapshot.

The third is Net Interest Margin (NIM) — the difference between what a bank earns on its loans and investments and what it pays on deposits and borrowings, expressed as a percentage of its earning assets. NIM is the bank equivalent of a manufacturer's gross margin: it tells you how profitable the bank's core lending business is before overhead and provisioning. A bank with a NIM compressing quarter after quarter, even while reporting flat profit, is often a bank leaning on one-off income (like trading gains or fee waivers reversing) to hide a weakening core business.

Governance & Promoter Behaviour. For banks, the specific governance red flags to watch are: large loans to promoter-linked companies (related-party lending), loan concentration in a small number of large borrowers, and frequent related-party transactions disclosed in the notes to accounts. NRB limits how much a bank can lend to a single borrower or group as a percentage of its capital fund precisely because concentrated lending to insiders has caused failures in Nepal's banking history before. A bank whose related-party disclosures are thin, late, or vague should be marked down here even if its headline profit looks fine.

Liquidity & Tradability. The generic Canon liquidity dimension (share trading volume, free float) still applies to bank shares on NEPSE largely unchanged, since most listed banks are reasonably liquid. What is added for banks specifically is a second layer: balance-sheet liquidity, meaning the bank's own ability to meet deposit withdrawals and short-term obligations, tracked through NRB's Credit-to-Deposit (CD) ratio ceiling and statutory liquidity requirements. A bank running close to the maximum CD ratio has less room to grow lending without either raising fresh deposits or curbing loan growth — a real constraint on future Growth Trajectory scoring too.

Growth Trajectory. Loan book growth and deposit growth replace generic revenue growth. But growth in loans is not automatically good — the reader should check whether deposit growth is keeping pace with loan growth (funding it prudently) or whether the bank is stretching its CD ratio to fund loan growth beyond what deposits support.

Dividend & Capital Return Discipline and Valuation Reasonableness are adjusted more lightly for banks: valuation still uses Price-to-Book Value heavily (more heavily than for other sectors, since a bank's assets are mostly financial and closer to book value than a factory's physical assets are), and dividend discipline should account for NRB's capital adequacy rules, which can restrict how much a bank is allowed to distribute in a given year regardless of what management might otherwise prefer.

For the full accounting mechanics behind CAR, NPL classification, and provisioning, revisit Chapter 29 (commercial bank accounting) and Chapter 30 (development bank and finance company accounting) in Part VI before applying these adjustments.

Lesson 65.3 — Adjusting the Score for Hydropower Companies (Pre- and Post-COD)

No sector on NEPSE needs a more dramatic rewrite of the Canon Score than hydropower — because a single hydropower company can require two completely different scorecards at two different points in its own life, without the underlying quality of the project changing at all.

Recall from Part VIII (the project finance and hydropower modelling chapters) the concept of COD — Commercial Operation Date. This is the date on which a hydropower plant is licensed and connected to begin selling electricity, usually to the Nepal Electricity Authority (NEA) under a Power Purchase Agreement (PPA), a long-term contract fixing the price NEA will pay for the plant's electricity. Before COD, the company is in the construction phase: pouring concrete, boring tunnels, installing turbines, funded mostly by loans and shareholder capital, generating no revenue. After COD, the company enters the operating phase: it sells electricity and earns real cash flow.

Pre-COD scoring. During construction, Financial Strength & Profitability cannot be scored on profitability at all — there is none, and that is expected, not a defect. What replaces it is project financial discipline: is the project being built within its original budget, or has cost overrun eaten into the equity cushion? Is construction progressing against the original timeline, or has it slipped by years (each year of delay usually means more interest capitalised onto the loan, silently inflating the eventual debt load)? Is the debt-to-equity ratio of the project in line with what was originally financed (commonly financed with a large share of debt, often 70-80% of project cost, and the remainder equity), or has additional debt been layered on to cover overruns? Growth Trajectory during this phase should be scored on construction progress and the credibility of the COD date — has the company met its own prior guidance on timeline, or does it keep pushing the date back?

Post-COD scoring. Once a plant is operating, Financial Strength & Profitability can finally be scored using real revenue and margins — but even here, a generic reading is misleading, because of Nepal's wet and dry seasons. Nepali rivers carry far more water during the monsoon (roughly Ashad through Ashwin, June to October) than during the dry winter and spring months. Most Nepali hydropower plants are "run-of-river" plants (they use the river's natural flow rather than storing large volumes behind a big dam), so their electricity output — and therefore their revenue — swings sharply between a strong wet-season quarter and a weak dry-season quarter. A single quarter's profit figure, read without adjusting for season, can make a perfectly healthy hydropower company look like it is collapsing (comparing a dry-season quarter to the prior wet-season quarter) or look euphorically strong (comparing wet season to wet season, ignoring debt building up). The Canon Score reader should always compare a hydropower quarter to the same quarter one year earlier, never to the immediately preceding quarter, and should look at trailing twelve-month figures for a fair full-cycle picture.

The Liquidity & Tradability dimension for hydropower connects directly back to Chapter 58's discussion of circuit traps — the daily price-movement limit NEPSE imposes on individual stocks. Hydropower counters, especially smaller ones with thin free float, are disproportionately prone to circuit-trap behaviour: a wave of retail enthusiasm (often around monsoon season, or around a COD announcement) can send a small hydropower stock to its daily circuit limit for several consecutive sessions, with almost no actual sell-side volume clearing at that price. This is not genuine liquidity — it is the appearance of demand with very little real two-way trading underneath it. A Canon Score reader scoring Liquidity for a small hydropower counter should check the depth of the order book and recent traded volume in Rupee terms, not just whether the stock "moved."

For the underlying project finance mechanics — debt sizing, PPA tariff structures, cost overrun analysis, and how to model a hydropower company's cash flows across a full monsoon-to-dry-season cycle — revisit the hydropower modelling chapters in Part VIII, and Chapter 33 in Part VI for hydropower-specific accounting treatment (including how construction-period interest gets capitalised onto the balance sheet rather than expensed).

Lesson 65.4 — Adjusting the Score for Microfinance Institutions

Microfinance institutions (MFIs) lend small amounts, typically without traditional collateral, to low-income and rural borrowers, often organised into borrower groups that provide informal social pressure to repay. Nepal has a large listed microfinance sector on NEPSE, and it needs some of the sharpest governance and liquidity adjustments of any sector in this chapter.

Governance & Promoter Behaviour needs the heaviest rewrite here. The generic Canon governance checklist (board independence, related-party transactions, promoter share pledging) still applies, but for microfinance it must be supplemented with sector-specific red flags that live in how the institution treats its borrowers, not just its shareholders. Two practices in particular deserve attention. The first is loan recycling — issuing a new loan to a struggling borrower specifically to repay an old loan that is about to become overdue, which keeps the reported NPL ratio artificially low while the underlying borrower's debt burden actually worsens. The second is multiple borrowing, sometimes called overlapping — the same borrower or household taking loans from several different microfinance institutions simultaneously, often without any single lender knowing about the others, which leaves the borrower over-indebted and eventually unable to repay any of them. Nepal Rastra Bank has pushed the sector toward credit information sharing specifically to curb this problem, and NRB introduced a mandated base-rate-linked interest rate ceiling for microfinance lending (capping the effective lending rate, reported around 15% under the framework introduced in 2025) partly in response to concerns about aggressive rural lending practices. A microfinance institution whose growth has consistently outpaced the sector's average loan growth, without a clear explanation of how it avoids overlap with other lenders in the same geography, should be marked down on Governance even if its reported NPL ratio looks clean — because a clean NPL ratio built on recycled loans is not really clean.

Financial Strength & Profitability for microfinance should weight loan-loss provisioning coverage (how much the institution has set aside against expected defaults, relative to its loan book) more heavily than it would for a commercial bank, precisely because unsecured group lending carries structurally higher default risk than a bank's typically collateralized book. A healthy microfinance institution should be building provisioning steadily as it grows its loan book, not lagging behind loan growth.

Liquidity & Tradability differs for microfinance in two ways. First, many listed MFIs on NEPSE are relatively small-cap with thin free float, so — similar to smaller hydropower counters — the Canon reader should check real traded volume rather than headline price movement. Second, on the funding side (distinct from share liquidity), microfinance institutions often rely heavily on wholesale borrowing from commercial banks rather than retail deposits, which makes their own funding liquidity more sensitive to changes in bank lending appetite and interest rates than a deposit-funded commercial bank's would be.

Sector & Business Model Durability for microfinance should account for regulatory concentration risk: because the entire sector's interest rates, provisioning rules, and lending practices are tightly directed by NRB policy (including periodic rate caps and consolidation pushes), a microfinance institution's durability is unusually exposed to regulatory shifts compared to, say, a manufacturing company operating in a less directly regulated market.

For the accounting detail behind microfinance loan classification, group-lending structures, and provisioning rules, revisit Chapter 31 in Part VI.

Lesson 65.5 — Adjusting the Score for Insurance Companies

Insurance companies — both life insurance and non-life (general) insurance — sell a promise: pay a premium now, and the company promises to pay a much larger sum later if a specified event happens. That promise is the entire business, and it changes how nearly every Canon dimension should be read.

Financial Strength & Profitability for insurance is unrecognisable next to the generic version. The headline metric to replace generic profitability ratios is the claims ratio (also called the loss ratio) — the percentage of premium income that the company pays out in claims. A non-life insurer (motor, fire, marine, and similar general insurance) with a persistently high claims ratio is pricing its policies too cheaply relative to the risk it is taking on, or is facing a genuinely bad run of claims (a bad monsoon flood season, for instance) — either way, a rising claims ratio over several quarters deserves real scrutiny. A life insurer's claims pattern is read differently again, since life insurance claims (death benefits, maturity payouts) unfold over decades rather than within a policy year, so a life insurer's Financial Strength should weight actuarial reserve adequacy — whether the company has set aside enough today to cover promised payouts decades from now — alongside the claims ratio.

Alongside the claims ratio, investment portfolio quality matters enormously for insurance, because insurers invest the premiums they collect (the "float") before claims come due, and that investment income is often a larger share of total profit than the core underwriting business itself. A Canon reader should check where an insurer's investment book sits — government securities and fixed deposits are the safest, followed by rated corporate debt, with direct equity holdings carrying the most risk and volatility. An insurer with a large share of its investment book in speculative equity positions has effectively layered stock-market risk on top of its insurance risk, and that combination deserves a lower Financial Strength score than an insurer holding a more conservative, bond-heavy portfolio, even if both report similar profit in a good year.

The specific regulatory yardstick to use is the solvency margin (or solvency ratio) — a measure of how much surplus capital an insurer holds above what its risk-weighted liabilities require, similar in spirit to a bank's CAR but calculated very differently. The Nepal Insurance Authority introduced a Risk-Based Capital and Solvency Directive in 2024/2025 that requires insurers to maintain solvency ratios above a defined regulatory minimum, replacing the older, simpler capital-based rules. Nepal's non-life insurance sector has recently reported average solvency ratios comfortably above the regulatory floor (reported around the high-2x range across the industry in early 2025), which gives a Canon reader a rough sector benchmark: an insurer sitting only marginally above the regulatory minimum solvency ratio has a thinner capital cushion than the sector average and should be scored more cautiously than one running well above it. The Nepal Insurance Authority also raised minimum paid-up capital requirements sharply in recent years — to roughly NPR 5 arba (5,000 million) for life insurers and NPR 2.5 arba (2,500 million) for non-life insurers — and an insurer that took years past the original deadline to meet that threshold, or needed repeated deadline extensions, is worth noting under Governance.

Growth Trajectory for insurance should centre on premium growth — specifically, whether growth in gross written premium is coming from genuinely new policies and renewals (healthy) or from aggressive discounting to win volume, which can quietly push up the claims ratio a year or two later as poorly underwritten business comes due. Dividend & Capital Return Discipline should account for the fact that regulatory solvency requirements can restrict payouts in years when an insurer's capital cushion is thin, exactly as CAR rules do for banks.

For the underlying accounting treatment of premium recognition, claims reserving, and actuarial liability valuation, revisit Chapter 32 in Part VI, which covers insurance company accounting specifically.

Lesson 65.6 — A Cross-Sector Comparison — Applying Adjusted Scores Fairly Across Different Sectors

By now you have four different sets of tailoring instructions in your hands. The natural next question is: once I have sector-adjusted scores for a bank, a hydropower company, a microfinance institution, and an insurance company, can I actually compare them to each other?

The honest answer is: carefully, and only at the level of the final composite score — not dimension by dimension.

The whole reason the Canon Score compresses seven very different kinds of analysis into a single 0-to-100 number is so that, once each dimension has been fairly scored using the right sector-specific ruler, the final numbers become comparable again — the same way a doctor's overall "fitness score" lets you compare a seventy-year-old and a twenty-year-old on a common scale, even though the blood pressure ranges used to build that score were different for each. A Canon Score of 72 for a commercial bank and a Canon Score of 72 for a hydropower company should represent a genuinely similar overall quality and risk level, even though almost none of the underlying ratios that produced those two scores are directly comparable to each other.

What you should never do is compare raw sub-scores across sectors as if they used the same ruler. A "Financial Strength: 14/20" for a pre-COD hydropower company and a "Financial Strength: 14/20" for a commercial bank are not telling you the same thing about the same kind of risk — one is measuring construction discipline, the other is measuring capital adequacy and loan quality. They happen to land on the same number because each was separately calibrated to be a fair "14 out of 20" within its own sector's normal range. Comparing them side by side as if 14 meant the same thing in both cases would undo all the careful tailoring this chapter just walked through.

The table below summarises which of the seven Canon dimensions receive the heaviest adjustment for each of the four sectors covered in this chapter, so you have a quick reference before you build a scorecard.

The second table gathers the specific regulatory benchmark numbers referenced across this chapter, so you have them in one place as a working reference. Treat these as points-in-time figures to be refreshed against NRB, Nepal Insurance Authority, and company disclosures as regulations evolve — they are a starting anchor, not a permanent constant.

There is one last discipline worth naming before you move on. Sector adjustment is not a one-time skill you learn and then apply mechanically forever. NEPSE itself is not static — new sectors get added to the exchange, existing sectors get restructured by regulation (as happened with the microfinance rate cap in 2025 and the insurance risk-based capital directive around the same time), and companies sometimes straddle two sectors (a hydropower company that also holds a manufacturing subsidiary, for instance, or a finance company that also runs an insurance arm). When you meet a company that does not fit cleanly into one of the four sectors covered here — a hotel company, a manufacturing company, a trading company — return to Part VI's accounting chapters (particularly Chapter 34, covering manufacturing, trading, and hotel accounting) for the underlying detail, and apply the same discipline this chapter modelled: ask which of the seven dimensions genuinely needs a different ruler for this business, adjust only those, and leave the rest of the generic Chapter 64 framework standing.