Case Study 4 — A Manufacturing Company

First published 26 Aug 2026 · Last verified 29 Aug 2026

Kabita has now sat through three full Canon Score workups in this Part — a commercial bank in Chapter 83, a hydropower developer in Chapter 84, a microfinance institution in Chapter 85 — and something has changed in how she works. The first case study took her most of a weekend, with a printed copy of the 100-point framework next to her laptop and sticky notes marking which dimension she was on. By the third case study, the microfinance one, she could do a rough pass in an evening and knew, almost from memory, which questions the framework wanted answered under each of its seven dimensions. She arrives at this fourth case study, a manufacturing company, expecting the same rhythm: profile the business, walk the seven dimensions, apply the sector adjustment from Chapter 65, tally the score, decide. What she does not expect is how much of that confidence the manufacturing sector will take back from her within the first hour of research.

The company is Arghakhanchi Cement Limited, traded on NEPSE under the symbol ARGCL, one of Nepal's cement manufacturers and — as she will discover — one of the harder companies in this book to pin down with a clean set of numbers. Chapter 75 warned her this was coming: manufacturing is NEPSE's smallest, thinnest-covered sector cluster, without the swarm of brokerage notes and comparison sheets that follow banks, and without even the specialist attention hydropower and microfinance get from their own dedicated NRB circulars and IPO cycles. She had briefly considered Bottlers Nepal, the Coca-Cola bottler chapter 75 uses as its FMCG example, but decided a cement name would put her through the sharper version of the sector's core tension — the import-substitution story chapter 75 introduced, playing out against a real domestic capacity race that is happening as she writes.

Lesson 86.1 — Setting up the case: a cement company in a construction cycle

Arghakhanchi Cement Limited runs its production plant in Siyari Rural Municipality, Rupandehi district, in Nepal's southern plains near the Indian border, with a corporate office in Thapathali, Kathmandu. It makes ordinary Portland cement and Portland pozzolana cement — OPC and PPC in industry shorthand, the two most common cement grades used in Nepali construction, differing mainly in the proportion of clinker (the kiln-fired intermediate product that gives cement its strength) blended with other materials. The plant's clinkerization unit runs at roughly one million metric tons of annual capacity, with a grinding unit capacity of about 800,000 metric tons a year, and it uses Danish vertical roller mill technology along with a waste heat recovery system generating about 2.5 megawatts, which lets the plant claw back some of the enormous energy cost that a cement kiln otherwise burns straight into the atmosphere.

The board is chaired by Pashupati Murarka, a former president of the Federation of Nepalese Chambers of Commerce and Industry, alongside directors from the Siddhartha Group and the Kedia Organisation — three separate, well-established Nepali business houses sitting on one board. That detail matters more than it might first appear, and Kabita flags it for herself early, because it will resurface directly in the governance dimension.

Why does a manufacturing company need its own distinct reading of the Canon Score, on top of the general framework she has now applied three times? Three reasons, and Chapter 75 had already named all three before she opened a single filing.

The first is cyclicality of a specific and identifiable kind. A bank's fortunes track the whole economy in a diffuse way; a hydropower company's fortunes track rainfall and a fixed power purchase agreement; a cement company's fortunes track the construction cycle almost directly — how many buildings are going up, how much road and bridge work the government is funding, how confident households feel about starting a home addition. When construction slows, cement demand slows with it, often before the slowdown shows up anywhere else in the economy. An FMCG manufacturer like a bottler faces a gentler version of the same idea, tracking consumer demand and festival-season spending rather than construction, but the principle is the same: manufacturing revenue rises and falls with a cycle that is visible and, to some degree, predictable, rather than moving in a straight line.

The second is import-substitution economics. Nepal imports enormous quantities of finished goods from India, and for decades a lot of its cement did come across the border too. Domestic cement manufacturers have grown mainly by displacing those imports — hence import substitution, the general economic pattern of a country producing at home what it used to buy abroad. That displacement is protected less by tariffs than by simple geography: cement is heavy, and heavy things are expensive to move long distances. A tonne of cement carried three hundred kilometres by truck picks up a transport cost that a tonne made forty kilometres from the construction site does not. That transport-cost cushion is a real competitive advantage for Nepali producers close to the markets they serve, but it is not a permanent, unbreakable moat — a change in Indian export incentives, a new road, or a large new domestic competitor can erode it, which is exactly the sector-specific risk Lesson 86.4 will dig into.

The third reason is the one that will occupy much of Kabita's time on this case: thin coverage. For the bank, she had NRB circulars, quarterly disclosures in a standardised format, and half a dozen brokerage comparison sheets. For the hydropower company, she had the power purchase agreement itself, a public document she could read clause by clause. For the microfinance institution, NRB's microfinance-specific disclosure regime gave her portfolio-quality data down to the district level. For Arghakhanchi Cement, several of the numbers she wants simply are not sitting in one clean place. Dividend history, an item she pulled up in thirty seconds for the bank, took real digging here and still came back incomplete.

Lesson 86.2 — Hemisphere 1: Liquidity, Governance, and Durability

Kabita followed the same method the bank, hydropower, and microfinance case studies had taught her: split Chapter 64's seven dimensions into Hemisphere 1 (Liquidity & Tradability, Governance & Promoter Behaviour, and Sector & Business Model Durability, 40 points total) and Hemisphere 2 (the four dimensions built from the financial statements). Manufacturing is not one of the four sectors Chapter 65 covers directly — banking, hydropower, microfinance, and insurance — so for the dimensions that need a sector-specific ruler, Kabita reached for Chapter 34's manufacturing, trading, and hotel accounting instead, exactly as Chapter 65's own closing lesson instructs for any company that does not fit its four named sectors.

Liquidity & Tradability (10 points). This is where the sector's thin coverage bites first. For the bank, third-party sources gave Kabita a free-float percentage to two decimal places; for Arghakhanchi Cement, no verified, current free-float or average-daily-volume figure was available from a source she trusted enough to cite with confidence. What she could observe directly is the ownership structure: three named business houses — the Murarka family, the Siddhartha Group, and the Kedia Organisation — hold board seats, which in a small-cap Nepali industrial company is a strong indirect signal of a concentrated promoter block, but not a substitute for an actual float percentage. Chapter 64 asks for a real number here, not an inference, and manufacturing the number would be worse than admitting she does not have it. Kabita scored both sub-components conservatively: volume at 2 out of 5 and free float at 2 out of 5, explicitly because the data was unavailable rather than because she had evidence the stock was illiquid. Liquidity & Tradability: 2 + 2 = 4 out of 10.

Governance & Promoter Behaviour (15 points). Chapter 34's related-party framework, built around NAS 24 disclosure requirements, is the right lens here, since nearly every Nepali manufacturing name sits inside a larger private family conglomerate. Promoter shareholding stability and pledging (6 points): the three-family board structure suggests a stable, long-standing promoter base with no evidence of pledging, but without a verified shareholding percentage or CDSC record to check against, Kabita scored this 4 out of 6 rather than a full mark. Related-party transactions and audit opinion (5 points): the Siddhartha Group's footprint extends well beyond cement into banking and other sectors, and Chapter 34 is explicit that a manufacturer buying from or selling through related entities can shift margin between the listed company and its private affiliates through transfer pricing — not necessarily improperly, but in a way that deserves scrutiny. Kabita could not verify the granular related-party transaction detail with the confidence NRB's disclosure regime gave her for the bank, so she scored this 3 out of 5. Disclosure timeliness and board independence (4 points): ICRA Nepal's LBB+/A4+ credit rating implies at least some external scrutiny of the company's financials, a mild positive, but board independence was not separately verified, so she scored this 3 out of 4. Governance & Promoter Behaviour: 4 + 3 + 3 = 10 out of 15.

Sector & Business Model Durability (15 points). The moat sub-component (8 points) has two real layers: the outer layer is transport-cost protection against Indian imports, described in Lesson 86.1, and the inner layer is company-specific — a genuine, disclosed export recovery to India, rebounding to about 12 percent of total trade from around 5 percent the year before, plus a real cost-position advantage from the Danish vertical roller mill and a waste-heat recovery system generating roughly 2.5 megawatts. This is a real but not license-grade moat — commodity-adjacent, resting on transport economics rather than a regulatory barrier — so Kabita scored it 6 out of 8. Revenue concentration and dependency risk (7 points), read here as sector-capacity risk rather than customer concentration, is where a genuine structural threat sits: Shivam Cements' joint venture, Hongshi Shivam, has been expanding its own plant from roughly 6,000 tonnes of daily capacity toward a doubled 12,000 tonnes — a real, disclosed, named competitive threat that adds supply to the whole Nepali cement market regardless of how the construction cycle is doing. Kabita scored this 4 out of 7. Sector & Business Model Durability: 6 + 4 = 10 out of 15.

Lesson 86.3 — Hemisphere 2: What the Numbers Actually Say

With Hemisphere 1 assessed, Kabita turned to the quantitative half of the Canon Score: Financial Strength & Profitability (20 points), Valuation Reasonableness (15 points), Growth Trajectory (15 points), and Dividend & Capital Return Discipline (10 points).

Financial Strength & Profitability. Chapter 64's ROE sub-component (8 points) was the first place the thin-coverage problem hit hard: Kabita could not find a disclosed net profit or equity figure precise enough to compute a real return on equity for Arghakhanchi Cement, only revenue and margin data. Rather than estimate a number she could not defend, she scored this conservatively at 4 out of 8, informed loosely by the margin-expansion trend and by peer earnings context (Shivam's historically strong EPS, Ghorahi's current loss) rather than a direct calculation. Leverage and interest-coverage discipline (7 points) is where real, specific data exists: the gearing ratio (total debt divided by equity) sits at about 0.72 times, comfortably under Chapter 64's 1.0x top-band threshold for non-financial companies, and the debt service coverage ratio has improved from about 2.05 times to about 2.95 times alongside a real deleveraging trend (debt-to-operating-profit falling from 2.85x to 2.13x). This is not quite the literal "interest coverage over 5x" the top band asks for, but the direction and the underlying cushion are both genuinely strong, so Kabita scored this 6 out of 7. Earnings quality and consistency (5 points): revenue moved from Rs 5.77 billion to Rs 6.06 billion to Rs 5.87 billion over three fiscal years — a dip-and-rebound rather than a smooth climb — but margin expanded a real 22 percent to 25 percent over the same window with no reported loss year, so Kabita scored this 4 out of 5. Financial Strength & Profitability: 4 + 6 + 4 = 14 out of 20.

Valuation Reasonableness. This is where the thin-coverage problem hits hardest of all. For the bank and the hydropower company, Kabita had a current P/E and P/B pulled straight from a data provider. For Arghakhanchi Cement, the data providers she checked returned incomplete or unpopulated fields for current price, EPS, and book value. What she could build instead was context, not a number: Ghorahi Cement, a close peer, is currently loss-making — trailing EPS around minus Rs 6.13 — yet still trades around Rs 338, more than twice its book value of about Rs 165.73, telling her the market is pricing a temporarily unprofitable cement producer as though its earnings power will return. Shivam Cements posted EPS in the high twenties to high thirties of rupees in its strongest recent years. None of this substitutes for Arghakhanchi Cement's own multiple, and Kabita was careful not to borrow a peer's number and present it as the company's own. She scored both sub-components at the conservative end explicitly because of the data gap: P/E at 2 out of 8 and P/B at 2 out of 7. Valuation Reasonableness: 2 + 2 = 4 out of 15.

Growth Trajectory. Revenue CAGR (8 points): the three-year revenue path — Rs 5.77 billion, Rs 6.06 billion, Rs 5.87 billion, and a nine-month run-rate implying roughly Rs 6.0 billion for the current year — works out to a multi-year CAGR in the low single digits despite a strong recent nine-month figure of 14 percent year-on-year growth. Chapter 64's band reads the trailing multi-year picture, not the most recent quarter alone, so Kabita scored this 3 out of 8, while flagging the recent acceleration as a real, watchable signal the trailing average does not yet capture. Earnings consistency (7 points): with no verified multi-year EPS series available for Arghakhanchi Cement itself, Kabita scored this conservatively at 2 out of 7, again a data gap rather than a negative finding. Growth Trajectory: 3 + 2 = 5 out of 15.

Dividend & Capital Return Discipline. This is the dimension the original research kept circling back to without ever resolving: dividend history, which took thirty seconds to pull for the bank, remained genuinely unconfirmed for Arghakhanchi Cement across the public dividend trackers Kabita checked. Chapter 64 does not allow a dimension to simply go unscored because the data is hard to find — an unscored dimension is a silent zero dressed up as an omission, which is worse than an honest, low, clearly-labelled score. Consistency of payout (6 points): scored 1 out of 6, explicitly for lack of verifiable data, not evidence the company skips dividends. Sustainability of payout (4 points): scored 1 out of 4, same reasoning. Dividend & Capital Return Discipline: 1 + 1 = 2 out of 10 — Arghakhanchi Cement's weakest dimension, and the clearest example in this whole case study of a score that is low because the analyst could not verify, not because the business failed a test.

Lesson 86.4 — When a Data Gap Is the Score

This is the lesson where Kabita had to think hardest about what her own honesty was actually doing to the final number. Four of the seven dimensions she just scored — Liquidity & Tradability, part of Financial Strength & Profitability, Valuation Reasonableness, and part of Growth Trajectory, plus the entire Dividend & Capital Return Discipline dimension — carry a real, disclosed data gap rather than a negative finding about the business itself. That is a genuinely different situation from the bank, hydropower, and microfinance case studies, where every dimension rested on a verifiable current number.

The temptation, faced with that many gaps, is to either skip the affected dimensions entirely (which silently inflates the total by removing points from the denominator) or to guess at plausible-looking numbers to avoid an uncomfortable string of low scores (which manufactures false precision). Kabita rejected both. Chapter 64's seven dimensions and 100-point structure are not optional extras to be dropped when a company is inconvenient to research — they are the whole discipline of the framework, and a company that cannot be verified across a third of the rubric should score as though that uncertainty is real, because it is.

This is also the right place to correct something Kabita noticed while researching this case: Chapter 65, the sector-adjustment chapter this book has leaned on for the bank, hydropower, and microfinance case studies, does not cover manufacturing at all — its four sectors are banking, hydropower, microfinance, and insurance. Chapter 65's own closing lesson says so directly, and points readers to Chapter 34's manufacturing, trading, and hotel accounting instead for exactly this situation. The real, useful sector-specific reasoning for a cement company — reading capacity utilisation and operating leverage, distinguishing a sector-wide construction slowdown from a structural capacity threat, and treating related-party disclosures under NAS 24 with real scrutiny — comes from Chapter 34, not Chapter 65, and this chapter now cites it correctly rather than borrowing an authority it does not have.

With that correction made, the underlying investment reasoning still holds: Ghorahi Cement's current loss looks, at a glance, like a company in real trouble, but construction activity across Nepal has been soft — a cyclical condition hitting the whole sector at once — and the market's own pricing of Ghorahi above twice book value despite the loss suggests other participants read it the same way. Chapter 34's capacity-utilisation lens is the right tool for telling this apart from something structural: a plant running below capacity because customers aren't buying is cyclical and likely to recover, while a plant running below capacity because more competing plants are splitting the same demand is structural, and Hongshi Shivam's capacity doubling is exactly that kind of structural fact, unrelated to where the construction cycle currently sits.

Lesson 86.5 — The Full Worked Canon Score

Kabita assembled her tally using exactly the seven dimensions and point weights Chapter 64 defines.

Summed, Arghakhanchi Cement's Canon Score comes to 49 out of 100 — inside Chapter 64's Weak/Avoid band. No dimension triggers the governance override on its own (Governance & Promoter Behaviour scored 10 out of 15, above the 5-point floor), but the total lands in the same band that override would produce anyway.

Kabita sat with this result longer than any of her first three scores, because it demanded an honest answer to an uncomfortable question: is 49 telling her Arghakhanchi Cement is a weak business, or that it is an unverifiable one? Her honest answer was "mostly the second, but that is still a real reason for caution." Four of the seven dimensions carry a genuine data gap rather than a proven weakness — if Liquidity, the ROE sub-component, Valuation, the earnings-consistency sub-component, and Dividend were all filled in with real, merely-average numbers instead of conservative placeholders, the total could plausibly sit fifteen to twenty points higher, comfortably in the Adequate band. But Chapter 64's discipline does not let an analyst pre-award those points on the assumption that the missing data would turn out fine. A company this hard to verify earns a cautious score precisely because it is this hard to verify, and an investor has to decide whether they are comfortable holding a position sized to that uncertainty, not to a more generous number they have not actually earned the right to assign.

Lesson 86.6 — The Decision, and What Would Move It

A score of 49 out of 100 sits inside Chapter 64's Weak/Avoid band — needing an unusually strong specific justification to hold, per the book's own definition of that band. For Kabita, the honest reading is not "avoid this company forever," but "this is not a position to take on the strength of currently available public information." The real financial signals she could verify — gearing, deleveraging, DSCR, margin expansion, a genuine export recovery — are all quietly encouraging. The dimensions she could not verify are numerous enough, and concentrated enough in the areas (liquidity, valuation, dividend discipline) that most directly affect whether a position can be sized and exited sensibly, that she chose to treat this as a name to keep researching rather than a name to buy or write off.

Three things would most plausibly move this score. First, and most directly actionable: better disclosure. If Arghakhanchi Cement's own reporting, or third-party coverage of it, improves enough to fill in a real free-float figure, a current P/E and P/B, a computable ROE, and a confirmed dividend record, four of the seven dimensions could move meaningfully in either direction once the real numbers are known — this is not a bet on the business improving, just on the fog clearing. Second, the pace at which Hongshi Shivam's new capacity comes online and gets absorbed by demand — if construction activity recovers fast enough to absorb the new supply without a margin hit, the durability score should hold or improve; if the new capacity lands into continued softness, it should fall further, and this is a specific, checkable fact rather than a mood. Third, any change in Indian trade or export-incentive policy toward cement and clinker, which could widen or narrow the transport-cost protection the whole domestic industry leans on.

Kabita closed her notebook on this fourth case study having learned something the first three had not fully prepared her for: that the Canon Score's honesty sometimes has to point inward, at the analyst's own inability to verify, rather than only outward at the company's fundamentals — and that a low score built substantially from disclosed data gaps is still the right score to report, not a reason to quietly round it up.