Skip to content

alirezarezvani/claude-skills / finance/skills/stock-analysis

stock-analysis

Unverified

Analyze, research or value a listed stock or company (US, global, NSE or BSE) with verified data, ratios, scores and valuation. Use whenever asked if a stock is a buy, what it is worth, or how a company looks.

FileKindSize
SKILL.mdskill md32.7 KB
evals/evals.jsonconfig13.2 KB
evals/fixtures/synthetic-ar-excerpt.mdmarkdown5.7 KB
evals/fixtures/synthetic-drhp-excerpt.mdmarkdown3.4 KB
examples/README.mdmarkdown1.9 KB
examples/forensic-analysis-example.mdmarkdown8.8 KB
examples/standard-analysis-example.mdmarkdown15.6 KB
references/01-data-sourcing.mdmarkdown38.5 KB
references/02-core-factors.mdmarkdown49.3 KB
references/03-earnings-quality.mdmarkdown40.1 KB
references/04-balance-sheet-and-cashflow.mdmarkdown56.6 KB
references/05-returns-and-dupont.mdmarkdown49.8 KB
references/06-valuation.mdmarkdown53.7 KB
references/07-forensic-red-flags.mdmarkdown57.5 KB
references/08-governance.mdmarkdown67.0 KB
references/09-risk-and-macro.mdmarkdown52.7 KB
references/10-peer-set.mdmarkdown45.5 KB
references/11-scoring-rubric.mdmarkdown36.8 KB
references/12-report-template.mdmarkdown26.6 KB
references/13-situations.mdmarkdown60.6 KB
references/14-accounting-comparability.mdmarkdown48.6 KB
references/15-document-diligence.mdmarkdown76.1 KB
references/16-market-mechanics-and-tax.mdmarkdown48.6 KB
references/17-process-and-epistemics.mdmarkdown59.6 KB
references/18-forensic-mode.mdmarkdown16.5 KB
references/19-ipo-mode.mdmarkdown15.6 KB
references/20-challenge-pass.mdmarkdown14.1 KB
references/21-data-integrity-tools.mdmarkdown7.2 KB
references/sectors/_index.mdmarkdown67.8 KB
references/sectors/auto.mdmarkdown47.5 KB
references/sectors/aviation-hotels.mdmarkdown49.3 KB
references/sectors/banks.mdmarkdown38.6 KB
references/sectors/biotech-clinical.mdmarkdown47.8 KB
references/sectors/chemicals-cement.mdmarkdown46.2 KB
references/sectors/exchanges-payments.mdmarkdown41.2 KB
references/sectors/fmcg-consumer.mdmarkdown35.7 KB
references/sectors/holdco-assetmgr.mdmarkdown41.4 KB
references/sectors/infra-capitalgoods.mdmarkdown43.0 KB
references/sectors/insurance-brokers-services.mdmarkdown49.9 KB
references/sectors/insurance.mdmarkdown38.0 KB
references/sectors/it-saas.mdmarkdown44.3 KB
references/sectors/metals-mining.mdmarkdown49.4 KB
references/sectors/mortgage-reit-specialty-finance.mdmarkdown43.5 KB
references/sectors/nbfc.mdmarkdown40.3 KB
references/sectors/oil-gas.mdmarkdown43.5 KB
references/sectors/people-businesses.mdmarkdown46.3 KB
references/sectors/pharma-healthcare.mdmarkdown40.6 KB
references/sectors/rail-freight.mdmarkdown66.2 KB
references/sectors/realestate-reit.mdmarkdown44.3 KB
references/sectors/retail-ecommerce.mdmarkdown48.3 KB
references/sectors/semiconductors.mdmarkdown43.2 KB
references/sectors/shipping-logistics.mdmarkdown47.2 KB
references/sectors/telecom-media.mdmarkdown43.1 KB
references/sectors/utilities-power.mdmarkdown44.9 KB
references/sectors/waste-environmental.mdmarkdown46.5 KB
scripts/benchmarks.jsonconfig200.5 KB
scripts/lint_report.pyscript47.8 KB
scripts/ratios.pyscript127.4 KB
scripts/score.pyscript96.8 KB
scripts/valuation.pyscript37.9 KB
scripts/verify_data.pyscript72.5 KB

references/sectors/chemicals-cement.md

View on GitHub
# Specialty chemicals, agrochemicals, fertilisers and cement — sector playbook

Use this when: the company converts bulk feedstock into a physical product sold by the tonne or the kilogram — specialty and fine chemicals, CDMO/CRAMS, commodity and petrochemical intermediates, dyes and pigments, agrochemical technicals and formulations, urea and P&K fertilisers, and cement, clinker and ready-mix concrete.

This is not one sector. It is at least four businesses stapled together by an index label, and the generic ratio checklist breaks in a *different* way in each. Specialty chemicals are pass-through-priced, so margin and revenue growth are both accounting artefacts of input costs. Fertilisers are administratively priced, so ROCE is set by a government notification rather than by competitive advantage. Cement is a regional logistics cyclical where P/E is actively inverted and book value bears no relation to replacement cost. Agrochemicals sit between the two, with a registration-and-patent-cliff structure closer to pharma than to chemicals. Your job is to identify which sub-sector you are actually in before you quote a single ratio, then replace contaminated financial ratios with unit economics (EBITDA per tonne, gross profit per kg), physicals (utilisation, clinker factor, energy per tonne against norm) and capital-allocation tests (incremental ROCE on the last capex cycle).

## Contents

1. [Why the generic ratio set fails here](#why-the-generic-ratio-set-fails-here)
2. [The metrics that actually matter](#the-metrics-that-actually-matter)
3. [How to value companies in this sector](#how-to-value-companies-in-this-sector)
4. [Peer set construction](#peer-set-construction)
5. [Sector-specific red flags](#sector-specific-red-flags)
6. [Cycle and structural context](#cycle-and-structural-context)
7. [India vs global notes](#india-vs-global-notes)
8. [Checklist](#checklist)

---

## Why the generic ratio set fails here

Work through these before quoting any standard ratio. In most cases the right action is to suppress the metric or replace it with the sub-sector version, not to caveat it.

**OPM is not comparable — not even within the sector.** Structural margins differ by design: specialty chemicals 18–25% (CDMO/CRAMS 25–35%), commodity and bulk chemicals 8–14%, cement 15–25% (collapsing below 10% at cycle troughs), urea 5–8%, agrochemical distribution and traded fertiliser 1–4%. A fertiliser company at 6% OPM can earn a better ROE than a chemicals company at 20%, because the capital intensity and the working-capital funding are completely different. **Ranking on OPM inside this sector produces nonsense.** Always compare margin to the company's own history and to its sub-sector, never across the group.

**Chemicals margin and revenue growth are both pass-through artefacts.** Most specialty contracts are formula-priced: raw material cost plus a conversion fee. When crude, benzene, acetone, phenol or Chinese intermediate prices fall, revenue falls while gross profit per kg is unchanged — so OPM *mechanically expands* while "revenue growth" collapses. Both signals are false, and both invert when input prices rise. Strip the pass-through by working in **gross margin and gross profit / EBITDA per kg**, and decompose revenue into volume × realisation before commenting on growth.

**Fertiliser "revenue" is not revenue and "margin" is not a margin.** For Indian urea, 60–75% of realisation is government subsidy under NPS-III / the modified scheme; for P&K it is NBS. The return is administratively fixed (broadly a ~12% post-tax return-on-net-worth construct for urea, plus whatever the plant earns by beating its energy norm). A high ROCE therefore means the norms are generous or the plant is efficient against them — **not that a moat exists**. Margins can also jump for a quarter purely on recognition of prior-year subsidy arrears.

**D/E and working capital are structurally distorted for fertilisers.** Delayed subsidy disbursal, plus special banking arrangement borrowings, means large short-term debt exists solely to fund a sovereign receivable. Headline D/E of 1.5–2.0x can be economically ~0.3x. Compute debt **ex-subsidy-backed working-capital debt**, and receivable days **ex-subsidy**, or the comparison to any other company is meaningless.

**P/E is actively dangerous for cement, and inverted.** Cement earnings peak with price and utilisation, so the stock screens cheapest at exactly the top and most expensive (or loss-making) at the bottom. The correct reading is the opposite of the generic one. The same trap applies to commodity chemicals and to any specialty name that caught a one-off pricing windfall — the 2021–22 refrigerant-gas, agrochem-intermediate and China-shutdown spikes are the standing example of 35% margins reverting to 15% within six quarters.

**FCF is structurally negative through the build phase, and that is not a defect.** Cement greenfield takes 3–4 years at roughly ₹6,500–8,500 crore per 10 mtpa in India (~$100–120/t globally); chemical multipurpose plants take 18–30 months to build and 2–3 years to ramp. Screening on positive FCF systematically rejects the companies that are compounding capacity. Split **maintenance capex** (cement roughly ₹150–250/t of capacity per year; chemicals ~3–4% of gross block) from growth capex, then judge the return on the growth capex.

**ROCE is depressed by CWIP and flattered by old assets.** A company mid-build carries large capital work-in-progress earning nothing, so reported ROCE understates the operating business — compute it **ex-CWIP**. Conversely, a 30-year-old cement plant carried at historic cost shows spectacular ROCE and a misleadingly low P/B, because replacement cost is 3–5x book. **P/B is close to meaningless for cement for exactly this reason**, and "low P/B" is not a value signal.

**Inventory turnover cannot be compared across the three.** Cement has a ~3-month shelf life and cannot be stockpiled, so inventory days are structurally 15–30. Chemicals run 60–110 days. Fertilisers and agrochemicals are violently seasonal (Kharif/Rabi in India, the Northern-Hemisphere spring season globally) and can legitimately hold 90–150 days. Cross-comparison, or year-on-year comparison struck at the wrong quarter-end, is pure noise.

**Asset turnover looks "bad" by design.** Cement runs ~0.5–0.8x gross-block turns, chemical plants ~1.0–1.5x. Measured against an FMCG-calibrated 3x+ benchmark, every company here fails a screen it was never in scope for.

**EV/Sales and revenue-based screens misfire on traders.** Fertiliser importers and chemical distributors book large traded volumes at 1–3% margin purely to inflate topline. A distributor at 4% margin and a manufacturer at 20% margin cannot share a revenue-based screen.

**Current ratio and interest cover mislead in opposite directions.** A fertiliser company's current ratio is inflated by an uncollected sovereign receivable; a mid-build cement or chemicals company's interest cover is flattered because interest is being **capitalised into CWIP** rather than expensed. Always read the capitalised-interest disclosure before quoting cover.

---

## The metrics that actually matter

Ranges below are **indicative only**. They vary by market, sub-sector, cycle position, energy prices and reporting period, and several are quoted in nominal currency that drifts with inflation. A company's own 5–10 year history and its *direct sub-sector* peers at the *same point in the cycle* always override any absolute band quoted here.

| Metric | Definition / how to compute | Indicative healthy range | Why it matters |
|---|---|---|---|
| **EBITDA per tonne (cement)** | Cement EBITDA ÷ sales volume in tonnes, decomposed into realisation/t minus cost/t. Strip out state incentive/SGST-deferral income and non-cement segments first, and disclose whether you have. | India: ₹800–1,200/t is a healthy mid-cycle band; >₹1,400/t is peak (treat as unsustainable); <₹600/t is trough. Developed markets: US$25–40/t, with US leaders running $40–55/t in tight cycles. | The single unit-economics number in cement, and the basis on which every management guides. It separates an earnings jump caused by **price** (cyclical, mean-reverting) from one caused by **cost or efficiency** (durable), and it is the input to the only sensible cement valuation: mid-cycle EBITDA/t × capacity. |
| **Capacity utilisation, and clinker vs grinding balance** | Volume sold ÷ installed grinding capacity, read alongside the clinker-to-grinding ratio and the split-grinding-unit footprint. Track *industry/regional* utilisation, not just the company's. | India: 70–85% is normal; the industry historically gets pricing power only above ~80–85% system utilisation. An efficient single player should run 5–15 points above industry. Mature global markets 75–85%. | Cement prices are set by regional supply-demand, not by any single company. Utilisation is the leading indicator of whether the pricing cycle is turning. A company adding capacity into a sub-70%-utilisation region destroys value for itself and everyone else in that region. |
| **Cost per tonne build-up: power & fuel, freight, lead distance** | Cost/t split into power & fuel (~25–30%), freight & forwarding (~20–25%), raw material and limestone, employee, other. Lead distance = average km from plant to market. Track fuel mix (petcoke vs imported/domestic coal vs AFR) and rail vs road share. | Lead distance <350–400 km (best-in-class 200–300 km). Power <70–75 kWh/t of cement; thermal energy <720–750 kcal/kg of clinker. Waste-heat recovery plus renewables >30% of power (leaders target 50–60%). Alternative-fuel/thermal substitution 5–15% India, 30–60% Europe. | Cement is a logistics business with a kiln attached — nobody ships cement profitably beyond ~500 km. Lead distance and fuel mix are the only genuinely defensible cost moats, and they separate a structurally low-cost operator from one that merely caught a good price cycle. Petcoke and coal moves pass straight into EBITDA/t, so the fuel mix tells you how next quarter moves. |
| **Clinker factor / blend ratio and CO₂ intensity** | Tonnes of cement per tonne of clinker consumed (i.e. how much fly ash, slag or limestone filler substitutes for clinker), and net CO₂ per tonne of cementitious material. | Clinker factor 0.62–0.70 (blend ratio ~1.45–1.60) is good in India where PPC/PSC dominate; developed markets historically 0.75–0.80 and falling. CO₂ <550–600 kg/t cementitious is competitive; European leaders target <500. | Every point of clinker substitution is a direct cost saving *and* effectively free capacity, because clinker is the energy-intensive, CO₂-emitting, capital-intensive part. It is also the main determinant of carbon-cost exposure: EU ETS plus CBAM (cement is in scope from 2026) turns CO₂/t into a hard P&L line for European producers and for anyone exporting into the EU. |
| **Trade vs non-trade mix and premium product share (cement)** | Split of volume between trade (retail/dealer, individual home builder) and non-trade (bulk, institutional, infra contractors, RMC), plus branded premium SKUs as a share of trade volume. | Trade share >65–70% is a strong sign; premium products >12–20% of trade volume for leading brands. | Trade sales carry roughly ₹200–400/t higher realisation and are where brand equity actually exists. Management reporting flat realisation while quietly shifting to non-trade is losing pricing power and hiding it in mix. A rising premium mix is one of the few genuine margin levers that is not the cycle. |
| **Gross margin and gross profit per kg (specialty chemicals)** | Revenue minus raw materials and consumables, expressed both as % and in absolute currency per kg sold. Track per kg through a full input-price cycle. | Genuine specialty / CDMO: 45–60% gross margin. Fine chemicals and agrochem intermediates: 35–45%. Below ~30% you are looking at a commodity chemical company describing itself as specialty. Global peers (Croda, IFF, Novonesis) run 35–55%. | The only chemicals margin not contaminated by input pass-through, inventory gains and mix accounting. A company whose OPM rises while gross profit per kg is flat has had a raw-material tailwind, not an improvement. **Stable gross profit per kg through an input-price cycle is the actual proof of a contracted, sticky, specialty business.** |
| **R&D intensity and product vintage** | R&D as % of sales; number of active molecules; pipeline by commercialisation stage; share of revenue from products launched in the last 5 years. Check patents/DMFs filed and pilot-plant/kilo-lab capacity. | Specialty/CDMO: 2–5% of sales in India, 3–6% globally. <1% means commodity. New-product revenue share of 15–30% indicates a functioning innovation engine. | Specialty multiples are paid for the ability to keep replacing molecules as they commoditise — every specialty product decays into a commodity in 5–10 years. Sub-1% R&D alongside 25% margins is a warning that the margin is a temporary supply dislocation, not a capability. |
| **Customer, product and geography concentration; contracted order book** | Revenue share of top 5 customers and of the largest molecule; share under multi-year contracts or LTSAs; CDMO/CRAMS committed order book; single-source dependence on Chinese intermediates. | Top 5 customers <40–50%; largest single product <20–25%. CDMO order book >1.0–1.5x annual revenue with named innovator customers. Anything where one molecule is >35% of EBITDA should be valued as a single-product company. | Indian specialty chemical blowups are almost always single-molecule or single-customer events: the innovator loses exclusivity, in-sources, dual-sources, or a Chinese competitor restarts a shut line. Concentration is the dominant risk factor and is **invisible in every financial ratio**. |
| **Agrochemical registration portfolio, patent-cliff pipeline and channel inventory** | Count and geography of product registrations (CIB&RC Section 9(3) vs 9(4) in India; EPA in the US; EU Reg. 1107/2009 Annex I), share of revenue from post-patent molecules, the molecules coming off patent in the next 3–5 years, technical vs formulation mix, and distributor channel inventory / days of stock in the trade. | Technicals (backward-integrated) should be >40–50% of agrochem revenue for a manufacturer, not a formulator. A registration pipeline that replaces >10% of revenue over 5 years. Channel inventory at or below normal season-end days. | Agrochemicals are a registration business, not a chemistry business: the moat is the data package and the country registration, which take 3–7 years and real money to build. Revenue is recognised on sell-*in* to distributors, so a season of channel stuffing shows as growth and then as a collapse — the 2023–24 global destocking cycle halved earnings at companies whose end-demand never changed. Always reconcile sell-in to sell-out. |
| **Energy consumption per tonne of urea vs the notified norm (Gcal/MT)** | Specific energy consumption of each urea plant against the government's cut-off/target energy norm under NPS-III / the modified scheme. Consumption below the norm is retained as profit; above it is an uncompensated loss. | Gas-based post-2015-policy plants ~5.0–5.5 Gcal/MT; notified target norms cluster near 5.5, with older plants at 6.0–6.5 and being squeezed. Above ~6.5 Gcal/MT is a stranded-asset candidate. | In a regulated cost-plus business this is the **only** operating lever that converts into profit. It also tells you who dies at the next norm revision: the government periodically tightens cut-off norms, and a plant running above the new norm loses its entire efficiency margin overnight. This is the fertiliser equivalent of cost per tonne. |
| **Non-subsidy EBITDA share (fertilisers)** | EBITDA from businesses outside the administered-price regime: complex/P&K with pricing freedom, crop protection, specialty nutrients and water-solubles, industrial chemicals (melamine, TAN, methanol), seeds, retail agri-inputs. | Re-rating candidates are those pushing non-subsidy EBITDA above 30–50%. Pure regulated urea players sit near 0–10%. | Regulated urea earnings are bond-like and deserve a utility multiple; non-subsidy agri-input earnings are market-priced and deserve a chemicals multiple. This ratio drives the sum-of-the-parts valuation and is the main structural story in Indian fertiliser equities. |
| **Working-capital cycle calibrated by sub-sector, including subsidy receivable days** | Inventory + receivable − payable days, benchmarked to the *right* sub-sector norm. For fertilisers, compute subsidy receivable days separately (outstanding subsidy ÷ daily revenue) and subsidy as a % of revenue. For agrochemicals, isolate Latin American receivables and barter/crop-payment terms. | Cement 15–35 days total (near-cash, dealer-financed). Specialty chemicals 80–130 days. Agrochemicals 120–200 days, worse in Latin America. India fertiliser subsidy receivable 45–90 days normal; >120–150 days signals a budget-shortfall year. Subsidy is typically 60–75% of urea revenue, 30–50% of P&K. | The single largest source of false comparisons in this sector. A sovereign receivable dictates fertiliser borrowings, interest cost and reported D/E — none of which are commercial decisions, and a spike is usually a fiscal event rather than a company event. Cement's near-zero cycle is why it can carry leverage that would kill an agrochemical company. |
| **ROCE excluding CWIP and cash, plus incremental ROCE on the last 3 years' capex** | EBIT ÷ (net fixed assets *in operation* + working capital), deliberately excluding CWIP and surplus cash. Then: change in EBIT over 3 years ÷ capital deployed over the same period. | Specialty chemicals 18–25% ex-CWIP is good, >25% excellent. Cement 12–18% mid-cycle (an efficient operator can print 20%+ at peak — do not extrapolate). Fertilisers 12–16%, effectively capped by regulation. Incremental ROCE must clear WACC: ~11–13% India, 7–9% developed markets. | This whole sector is a capital-allocation game — **capacity is the product**. Reported ROCE punishes companies mid-build and rewards those with fully depreciated assets, so it says nothing about management skill. Incremental ROCE on recent capex is the only number that tells you whether the next ₹5,000 crore of expansion creates or destroys value. |
| **Capex per unit of installed capacity vs peers, and project execution record** | Announced/actual project cost ÷ capacity created, benchmarked to comparable recent projects; plus the record of on-time, on-budget commissioning and time to full ramp. | Cement greenfield India ~₹6,500–8,500 crore per 10 mtpa (≈$100–120/t globally); brownfield and standalone grinding units far cheaper at ₹2,500–3,500/t. Acquisitions have transacted at $60–160/t. Chemicals: judge asset turn at full ramp — a plant should deliver 1.0–1.5x sales on gross block. | The highest-signal forensic ratio in the sector. Capex per tonne materially above the peer benchmark is the classic mechanism for extracting cash through inflated EPC contracts to related parties; materially below can mean corners cut or capacity overstated. It also *sets* the future ROCE — you cannot earn a good return on an overpriced plant. |
| **Net debt / EBITDA and interest cover, adjusted for subsidy-backed and construction debt** | Net debt/EBITDA computed on the *operating* business: exclude fertiliser subsidy-financing borrowings, and separately quantify debt attached to assets not yet generating EBITDA. Note how much interest is capitalised rather than expensed. | Cement <1.5x comfortable, <2.5x acceptable mid-build, >3x through a downturn is dangerous. Specialty chemicals <2.0x. Fertilisers ex-subsidy debt <1.5x. Interest cover >5x. | Fixed-cost plants with cyclical prices mean EBITDA can halve in four quarters; leverage taken against peak EBITDA is how cement companies end up being sold at the bottom of the cycle. The adjustments matter because unadjusted leverage makes fertiliser companies look distressed and makes mid-construction companies look fine right up until the ramp slips. |
| **Environmental compliance and licence-to-operate status** | Consent-to-operate validity; pollution control board notices and closure orders; zero-liquid-discharge status; effluent and hazardous-waste handling; EHS incident and fatality record; limestone reserve life and mining-lease tenure (cement); gas allocation and pipeline connectivity (fertiliser). | Cement: 40–50+ years of limestone reserves at planned capacity, leases secured. Chemicals: ZLD achieved at all sites, no pending closure orders, no repeat consent violations. Zero fatalities. | This is an existential operating KPI, not an ESG nicety. Indian pollution boards shut chemical plants for weeks with no notice in the Gujarat, Telangana and Tamil Nadu clusters; a cement company without secured limestone has no business regardless of its P&L; a fertiliser plant without gas allocation cannot run. A single site closure at a concentrated manufacturer removes a year of earnings and appears in no financial ratio. |

**Sourcing notes.** India: volumes, realisation, EBITDA/t, lead distance, trade mix, blend ratio, energy norms and subsidy receivables live in the **quarterly investor presentation and concall transcript**, not the financials. Segment notes give sub-sector splits; the CARO report and related-party note carry the governance signals; contingent liabilities carry CCI penalties and tax/environment disputes. Global: 10-K MD&A and segment reporting, sustainability reports for CO₂/t and clinker factor, and industry data (CMA/DIPP dispatch data in India; USGS, Cembureau, GCCA globally; Phillips McDougall/AgbioInvestor for agrochemicals).

---

## How to value companies in this sector

**Value each sub-sector on its own basis. A single multiple across the group is wrong.** For a diversified group, SOTP is mandatory.

### Cement — asset-based and mid-cycle, never trailing P/E

**Primary: EV per tonne of installed capacity, benchmarked to greenfield replacement cost.** India has transacted in the ₹4,500–9,000/t range, with listed leaders trading at ₹9,000–15,000/t against a replacement cost of roughly ₹6,500–8,500/t. The global analogue is $/t against a ~$100–150/t replacement cost. Below replacement cost is the classic cyclical entry; above it you are paying for brand, regional market position and pricing power — which the large Indian players in the south and west genuinely have. State explicitly which you are paying for.

**Secondary: EV/EBITDA on mid-cycle EBITDA/t.** Build it bottom-up: capacity × utilisation × normalised EBITDA/t. India mid-cycle 10–14x for leaders, 6–9x for regional players; developed markets 6–9x. The multiple must *contract* as you move up the cycle — applying a constant multiple to a rising EBITDA estimate is a momentum model, not a valuation.

**Also useful:** replacement-cost DCF, and the greenfield-vs-acquisition cost-per-tonne arbitrage (which is the actual decision management faces).

**Explicitly avoid P/E and P/B.** P/E inverts across the cycle; book value is historic cost far below replacement cost. Cement is bought at an optically high P/E on trough earnings and sold at an optically low P/E on peak earnings.

### Fertilisers — sum-of-the-parts, with utility logic for the regulated leg

The **regulated urea business is a quasi-utility**: value it on P/B against the regulated return on net worth (broadly a ~12% post-tax RoNW construct), i.e. roughly 1.0–1.5x book depending on how far energy efficiency runs ahead of the norm, or on a low 5–8x EV/EBITDA. It does not deserve a growth multiple because both volumes and returns are capped.

The **non-subsidy businesses** — P&K with pricing freedom, crop protection, specialty nutrients, industrial chemicals, seeds — get market multiples (roughly 12–20x P/E, 8–14x EV/EBITDA in India). The re-rating thesis is almost always a rising share of the unregulated leg.

**Normalise earnings first**: strip prior-period subsidy arrears, subsidy-related interest, and inventory gains/losses from NBS rate changes. Prefer free cash flow to reported PAT, because subsidy timing distorts PAT in both directions.

In developed markets (Nutrien, Mosaic, Yara, CF) there is no subsidy regime at all — these are pure commodity cyclicals valued on mid-cycle EV/EBITDA (5–8x) and on gas-cost-curve position. **Indian fertiliser conventions do not transfer to them, or vice versa.**

### Specialty chemicals and agrochemicals — growth and cash-return based, with a capex-ramp DCF

**Primary: P/E and EV/EBITDA on forward earnings that reflect the ramp of announced capex.** India has paid 30–50x P/E and 18–30x EV/EBITDA for high-ROCE, high-visibility specialty and CDMO names; developed-market peers (Croda, Lanxess, Arkema, Evonik) trade at 8–15x EV/EBITDA and 12–20x P/E. That premium is a real convention driven by China+1, higher structural growth and better ROCE — but it also means Indian valuations **de-rate violently when growth stalls**, and a de-rating from 45x to 25x overwhelms any earnings growth you were underwriting.

**Build an explicit capex-ramp DCF.** For each announced plant, model commissioning date, asset turn at full utilisation (1.0–1.5x), steady-state EBITDA margin, and the resulting incremental ROCE. Because these companies are in continuous capex, near-term FCF is negative and only a DCF captures value being created.

**Use EV/EBITDA rather than P/E when mid-build** (depreciation and capitalised interest distort EPS). Cross-check with EV/Sales (2–5x specialty, 5–8x high-margin CDMO) **only** for pre-profit ramping assets. PEG is used in India as a sanity check on the growth premium.

For **generic agrochemicals**, value closer to generic pharma: the registration portfolio and technical backward-integration are the assets, earnings are lumpy with the destocking cycle, and mid-cycle EV/EBITDA of 8–12x is more appropriate than a specialty multiple. For **commodity chemicals** inside the group, revert to cement logic: mid-cycle EV/EBITDA and replacement cost, never trailing P/E.

### Cross-cutting

- **SOTP is mandatory for diversified groups** (Indian conglomerates commonly hold cement plus chemicals plus fertilisers), with a 10–20% holding-company discount where segments are unrelated, and more where a listed subsidiary is the main asset.
- **Sanity-check every valuation against replacement cost of the asset base as a floor**, and against incremental ROCE on recent capex as the test of whether growth is worth paying for.
- **What NOT to use:** trailing P/E for cement or any commodity chemical; P/B for cement; EV/Sales for distributors and traders; a group-wide multiple; DCF with a perpetual-growth terminal value applied to a single-molecule specialty business whose product will commoditise inside a decade.

---

## Peer set construction

A valid comparable shares **sub-sector, pricing regime, region and integration level** — not the label "chemicals". If you cannot state each peer's sub-sector, gross margin per unit, regulatory regime and cycle position, you have a list of tickers, not a peer set.

**Splits that must not be mixed:**

- **Specialty vs commodity vs CDMO.** A CDMO with contracted innovator volumes at 30% EBITDA margin, a specialty formulator at 20%, and a bulk intermediate producer at 10% have different multiples, different cyclicality and different failure modes. Gross margin below ~30% disqualifies a "specialty" label regardless of how the company describes itself.
- **Manufacturer vs formulator vs distributor/trader.** A backward-integrated technical manufacturer, an agrochemical formulator buying technicals, and a distributor at 1–4% margin cannot share revenue-based or margin-based screens.
- **Regulated urea vs NBS-linked P&K vs unregulated agri-inputs.** These are three different pricing regimes inside one company. Split them before comparing anything, including to the same company's own history if the mix has changed.
- **Cement by region, not by country.** Indian cement pricing is regional: south, west, north, east and central markets have independently moving prices, utilisation and consolidation levels. A south-focused player and an east-focused player are not comparables in the same quarter. The same is true of US regions and of European country markets.
- **Integrated vs grinding-only cement.** A grinding unit buying clinker has different capital intensity, different cost structure and a different EV/t benchmark than an integrated plant with captive limestone. Never compare their EV/t without stating clinker self-sufficiency.
- **Innovator vs generic agrochemicals.** An innovator with proprietary molecules under patent, and a post-patent generic manufacturer competing with Chinese supply, have opposite margin trajectories and should never share a multiple.
- **Capex phase.** A company mid-build with 40% of capital in CWIP and a company with a fully ramped asset base will show ROCE 8–10 points apart with identical asset quality. Compare ex-CWIP, and state each peer's CWIP-to-gross-block ratio.
- **India vs China vs developed markets.** Chinese producers operate with different environmental cost structures and state support; developed-market peers carry carbon costs and higher energy prices. Cost-curve logic transfers; multiples do not.
- **India-specific:** PSU fertiliser and cement entities carry administered pricing, disinvestment overhang and social obligations; private peers do not. Companies with large state SGST/VAT incentive accruals have an EBITDA component that expires on a known date — normalise it out before comparing EBITDA/t.

---

## Sector-specific red flags

### Cement

- **Capacity or acquisition announcements without secured limestone reserves and mining leases.** Reserve life below ~25 years at planned capacity, or leases won at aggressive auction premiums, means the announced capacity may never operate, or will operate at a permanent cost disadvantage.
- **Paying materially above replacement cost** (roughly ₹8,500/t or $120/t) for an acquisition, or greenfield capex per tonne well above the peer benchmark. Inflated project cost routed through EPC contracts to promoter-linked entities is the standard cash-extraction mechanism in this sector.
- **Quarter-end channel stuffing.** Cement has a ~3-month shelf life and dealers hold limited stock, so a jump in dispatches with rising receivable days and falling realisation the following quarter means volume was pushed into the trade rather than sold to end users. Watch the trade/non-trade mix quietly shifting to low-margin institutional volume while management reports "stable" realisation.
- **Leverage sized on peak-cycle EBITDA.** Debt at 2.5x peak EBITDA becomes 5x+ when prices normalise, because operating leverage in a fixed-cost kiln business is brutal. Also watch capitalised interest and pre-operative expenses parked in CWIP for years without commissioning.
- **CCI cartelisation penalties and regional "price discipline"** sitting in contingent liabilities. Coordination inflates current EBITDA/t and is simultaneously a legal liability and a signal that current margins are not competitively earned.
- **State incentive income (SGST/VAT deferral, capital subsidies) presented inside operating EBITDA** without disclosure of expiry. It can be ₹100–300/t at incentive-heavy plants and it disappears on a scheduled date.

### Fertilisers

- **Recognising subsidy income before it is notified or accrued**, or booking prior-year subsidy arrears into the current quarter without separate disclosure. This creates phantom margin expansion in a business whose real margins are fixed by regulation.
- **Ballooning subsidy receivable days alongside rising short-term debt and interest cost while EBITDA looks stable.** Reported profit holds up while cash generation collapses, and the interest carry on the receivable is often not fully compensated.
- **Plants running above the notified energy norm, or a company lobbying loudly about norm revisions.** When cut-off norms tighten, sub-scale and older gas/naphtha plants lose their entire efficiency margin at once. Also watch dependence on a single gas allocation or pipeline connection.
- **Low-margin traded/imported fertiliser volume presented as "growth"** at 1–3% margins. Separately, any exposure to diversion of subsidised urea to industrial use is a live regulatory and reputational risk in India.
- **Imported DAP/MOP inventory carried through an NBS rate change.** A rate cut strands inventory bought at the old economics; the loss lands in one quarter and is often described as "one-off".

### Specialty chemicals and agrochemicals

- **A one-off pricing windfall in a single molecule presented as structural.** The 2021–22 refrigerant-gas, agrochem-intermediate and China-shutdown spikes produced companies at 35% margins that reverted to 15% within six quarters. Test: did *gross profit per kg* rise, or only price per kg — and did volume grow at all?
- **Margin expansion attributed to "better product mix" with flat or falling volumes and no rise in gross profit per kg.** Usually this is inventory gain from a raw-material move, or a falling input cost passed through to a formula-priced customer with a lag. Both reverse.
- **R&D capitalised into intangibles rather than expensed**, or R&D below 1% of sales while marketing the company as specialty. Check product vintage: if no meaningful revenue comes from products launched in the last five years, the current margin is decaying.
- **Concentration disclosed only in fine print.** One molecule >35% of EBITDA, one innovator customer >25% of revenue, or a single unhedged Chinese source for a key intermediate. Loss of innovator exclusivity, in-sourcing or a Chinese restart removes the earnings base with no warning.
- **Sell-in growth without sell-out confirmation in agrochemicals.** Rising distributor receivable days plus rising channel inventory ahead of a season is the destocking cycle loading. Watch for extended credit terms and for revenue concentrated in the last month of a quarter.
- **Environmental shutdowns and consent violations.** Repeat pollution-board notices, missing ZLD, or a single-site manufacturer in a cluster subject to blanket closure orders. EHS incidents and fatalities are a leading indicator of maintenance underinvestment — reactor incidents shut sites for months.

### All three

- **CWIP sitting on the balance sheet for multiple years without commissioning**, growing interest capitalisation, and repeatedly slipping commissioning guidance. Either the project is impaired or the spend is not real. Check announced capacity against actual production once commissioned.
- **Large advances to suppliers, loans and guarantees to unlisted group entities, and rising related-party transactions alongside a capex cycle.** Combined with high promoter share pledge, auditor resignation or a mid-year auditor change, this is the standard pre-blowup pattern in Indian mid-cap chemicals.
- **Maintenance capex reported implausibly low** — cement below roughly ₹150/t of capacity per year, chemicals below ~2–3% of gross block. Deferred maintenance flatters near-term FCF and returns later as unplanned shutdowns and falling utilisation.
- **Unhedged foreign-currency debt or import exposure**, and for European or EU-exporting producers, unprovided carbon costs. EU ETS plus CBAM from 2026 turns CO₂ per tonne into a cash cost; a high-clinker-factor producer with no decarbonisation plan is carrying an undisclosed liability.

---

## Cycle and structural context

**Say where you think the cycle is before you quote a ratio, and say what evidence you used.** For cement: regional utilisation, price trends by region, announced industry capacity additions over the next 3 years, and the housing/infrastructure spend cycle. For chemicals: Chinese capacity restarts and export prices, crude and naphtha, customer inventory levels, and China+1 order-book commentary. For fertilisers: the budget subsidy allocation, gas and ammonia/phosphoric acid contract prices, and monsoon/reservoir data. For agrochemicals: channel inventory, farm-gate crop prices, and the position in the global destocking cycle.

**Cement is a capacity cycle, not a demand cycle.** Demand grows steadily with construction; earnings are destroyed by supply. Announcements cluster at the top (when EBITDA/t is high and financing is easy) and arrive 3–4 years later into a weaker market. Track *industry* announced capacity, not the company's, and downgrade any region where announced additions exceed 3–4 years of demand growth. Consolidation cuts the other way: regions where the top 3–4 players hold most capacity sustain higher utilisation and prices.

**Cement is regional and seasonal.** Prices are set within a ~300–500 km radius, so national averages hide everything. In India, Q1 (April–June) is the strongest quarter and Q2 (monsoon) the weakest; comparing sequential quarters without adjusting for this manufactures trends that do not exist.

**Specialty chemicals ride the China+1 relocation, which is real but not uniform.** Chinese environmental enforcement and cost inflation genuinely shifted intermediate manufacturing toward India, but Chinese capacity restarts and aggressive export pricing have repeatedly compressed margins in exactly the molecules where Indian players had the biggest windfalls. Treat "China+1" as a reason to check whether the *specific molecule* is defensible, not as a thesis on its own.

**Every specialty product commoditises.** The 5–10 year decay from specialty to commodity is the central economic fact of the sub-sector, which is why R&D intensity and product vintage carry more weight than current margin. Anti-dumping duties (India imposes them frequently) can pause the decay but are a policy asset, not a moat — check expiry dates.

**Agrochemicals run a global inventory cycle on top of a patent cycle.** Post-patent molecules face Chinese generic entry and price collapse; the industry periodically over-stocks the channel and then spends 4–6 quarters destocking, during which reported revenue is unrelated to farm demand. Patent expiries over the next 3–5 years are the generic manufacturers' opportunity set and the innovators' cliff.

**Fertilisers are policy assets.** The economics are set by the Department of Fertilizers, not by the market: NPS/modified-NPS energy norms for urea, annually notified NBS rates for P&K, gas pooling, urea import canalisation, and DBT — under which subsidy is released only on POS sale to the farmer, which permanently changed the working-capital profile. Structural themes to track: new gas-based urea capacity displacing imports and older plants, nano-urea and coated-urea policy, One Nation One Fertilizer branding (which erodes brand equity in P&K), and the annual budget subsidy allocation, which sets receivable days for the whole industry.

**Carbon regulation is now a P&L line, not a disclosure.** Cement is one of the largest industrial CO₂ sources; EU ETS plus CBAM (phasing in from 2026) prices embedded carbon on imports into Europe. Clinker factor, alternative fuel rate and CO₂/t therefore become cost variables for European producers and for exporters into Europe. India's CCTS/PAT carbon-credit scheme is developing along similar lines. Chemicals face REACH and PFAS restrictions in Europe, which can obsolete entire product lines.

**Energy is the swing input everywhere.** Petcoke and coal for cement, natural gas for urea and for European chemicals, crude derivatives for specialty feedstock. An energy price move of 20% is a first-order earnings event in all three sub-sectors, and the lag with which it passes through to customers determines whether it shows up as margin expansion or contraction first.

---

## India vs global notes

| Dimension | India (NSE/BSE, Ind AS) | US / global (10-K, GAAP/IFRS, EDGAR) |
|---|---|---|
| Fertiliser pricing regime | Administered. Urea MRP fixed by government with subsidy under NPS-III/modified scheme; P&K under NBS with annually notified per-nutrient rates. Subsidy is 60–75% of urea revenue. DBT releases subsidy only on POS retail sale. | No subsidy regime. Nutrien, Mosaic, Yara, CF are pure commodity cyclicals priced off gas cost, ammonia and grain prices. Valuation and margin conventions do **not** transfer in either direction. |
| Cement conventions | Reported in mtpa capacity and ₹/bag (50 kg) or ₹/tonne realisation; EBITDA/t is the universal guided metric. Volumes in crore/lakh tonnes — convert consistently and state the unit. PPC/PSC blends dominate, so clinker factor is structurally lower. | Reported in short tons or metric tonnes and $/t; higher clinker factor historically. Ready-mix and aggregates are usually a much larger share of the group, so segment-level analysis is essential before comparing multiples. |
| Carbon | CCTS/PAT scheme developing; carbon cost is not yet a large P&L line. Exporters into the EU face CBAM from 2026. | EU ETS is a direct, material cash cost for European cement and chemicals; CBAM extends it to imports from 2026. US has no federal carbon price but state schemes and IRA incentives apply. |
| Agrochemical registration | CIB&RC registration; Section 9(3) (full data, longer moat) vs 9(4) (me-too, faster and cheaper). Large export business built on registrations held in Latin America, Africa and South-East Asia. | EPA (US) and Regulation 1107/2009 (EU) — slower, far more expensive, and increasingly restrictive; EU re-approval cycles routinely remove molecules from the market entirely. |
| Environmental enforcement | State pollution control boards can suspend consent-to-operate at short notice; cluster-wide closures (Gujarat, Telangana, Tamil Nadu) hit multiple companies at once. ZLD is effectively mandatory in many clusters. | Enforcement is slower, more litigated and more predictable, but penalties and remediation liabilities are far larger. US filers must disclose environmental proceedings above a threshold in the 10-K. |
| Disclosure venue | Concall transcripts and quarterly investor presentations carry EBITDA/t, lead distance, trade mix, blend ratio, utilisation, energy norms, subsidy receivables and capex phasing — almost none of it in the financials. CARO, related-party notes and contingent liabilities carry the governance signal. | 10-K MD&A and segment notes, sustainability report for clinker factor and CO₂/t, and quarterly earnings decks. Sell-side publishes explicit price decks for commodity chemicals and fertilisers. |
| Ownership and governance | Promoter holding, pledge levels, promoter-linked EPC and logistics contracts, brand/royalty fees to unlisted parents, and inter-corporate deposits are first-order checks. PSU fertiliser and cement entities carry administered pricing and disinvestment overhang. | Widely held; governance issues centre on board independence, incentive design and the treatment of restructuring "one-offs". Controlling-family stakes exist in several European chemicals majors. |
| Valuation practice | Cement: EV/t against replacement cost plus mid-cycle EV/EBITDA. Fertiliser: SOTP with utility logic on the regulated leg. Specialty: forward P/E and EV/EBITDA at a large premium to global peers, cross-checked with a capex-ramp DCF. | Cement: EV/EBITDA and $/t, with aggregates/RMC valued separately. Fertiliser: mid-cycle EV/EBITDA on the gas-cost curve. Chemicals: EV/EBITDA 8–15x, DCF, and segment SOTP for diversified majors. |
| Accounting detail | Ind AS mirrors IFRS. Watch capitalisation of interest and pre-operative expenses into CWIP, government-grant accounting for state incentives (Ind AS 20), and inventory valuation through an NBS rate change. | IFRS filers similar. US GAAP differs on impairment reversal (not permitted), on development-cost capitalisation (generally expensed), and on LIFO inventory — which materially changes reported chemicals margin in a moving input-cost environment. |

---

## Checklist

- [ ] Identify the sub-sector first — specialty, CDMO, commodity chemical, agrochemical technical/formulation/distribution, urea, P&K, cement, grinding-only — and say so before quoting any ratio.
- [ ] For a diversified group, split segments and run SOTP; do not apply a group multiple.
- [ ] Suppress trailing P/E for cement and commodity chemicals; report it only alongside mid-cycle normalised EPS, and never call a peak-earnings P/E cheap.
- [ ] Do not rank companies inside this sector on OPM. Compare margin only to the company's own history and its direct sub-sector.
- [ ] For chemicals, compute gross margin and gross profit per kg through a full input-price cycle before commenting on margin or revenue growth.
- [ ] Decompose revenue into volume × realisation; state whether growth was volume or price.
- [ ] For cement, compute EBITDA/t, realisation/t and cost/t; separate state incentive income and note its expiry.
- [ ] Check regional utilisation and announced *industry* capacity additions for the company's regions, not national averages.
- [ ] Check lead distance, fuel mix, clinker factor, WHRS/renewable share and CO₂/t — these are the durable cost moats.
- [ ] For fertilisers, strip subsidy: compute subsidy % of revenue, subsidy receivable days, debt ex-subsidy-financing, and non-subsidy EBITDA share.
- [ ] Check each urea plant's energy consumption against the notified norm, and flag any plant above it as norm-revision risk.
- [ ] For agrochemicals, reconcile sell-in to sell-out, check channel inventory and receivable days, and list the registration portfolio and near-term patent expiries.
- [ ] Compute ROCE ex-CWIP and ex-cash, then incremental ROCE on the last 3 years' capex against a WACC of ~11–13% (India) or 7–9% (developed markets).
- [ ] Benchmark capex per tonne / per unit capacity against comparable recent projects; treat a large premium as a related-party red flag and a large discount as an overstated-capacity risk.
- [ ] Split maintenance from growth capex before calling FCF good or bad; flag maintenance capex below ₹150/t (cement) or 2–3% of gross block (chemicals).
- [ ] Test leverage against trough EBITDA, and check how much interest is capitalised into CWIP rather than expensed.
- [ ] Check top-5 customer, single-molecule and single-site concentration; value anything with one molecule >35% of EBITDA as a single-product company.
- [ ] Verify licence to operate: limestone reserve life and lease tenure, consent-to-operate status, ZLD, closure orders, gas allocation, EHS and fatality record.
- [ ] Read contingent liabilities for CCI penalties, environmental and tax disputes; read related-party and pledge disclosures alongside any active capex cycle.
- [ ] Cross-check the final valuation against replacement cost of the asset base as a floor, and state where in the cycle you believe the company is and on what evidence.
add-product-e2e-evalpaperclipai/paperclipAdd or extend a Paperclip full-stack runner E2E workflow, fixture, matcher, or report evidence path for local or Daytona execution.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
add-runner-evalpaperclipai/paperclipAdd or extend a Paperclip Runner protocol evaluation definition, roster, assertion, or report fixture with provenance and narrow validation.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
diagnose-why-work-stoppedpaperclipai/paperclipCompatibility entry point for failed, stalled, looping, or over-recovered Paperclip work. Use fix-failed-paperclip-run for the full diagnosis, regression, repair, real verification, and green-PR workflow; honor an explicitly requested diagnosis-only scope.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
fix-failed-paperclip-runpaperclipai/paperclipDiagnose and fix failed, stuck, slow, or incorrectly blocked Paperclip tasks and runs. Use with a task/run URL or failure report to trace the root cause, prove a regression red then green, check second-order effects, verify the real workflow, and deliver a PR with passing checks.Stars: 115kUpdated Claude CodeCodexVerifiedInstall