alirezarezvani/claude-skills / finance/skills/stock-analysis
stock-analysis
UnverifiedAnalyze, 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.
references/sectors/waste-environmental.md
View on GitHub# Waste management, environmental and water treatment services — sector playbook
Use this when: the company collects, transfers, processes or disposes of solid, hazardous, medical or industrial waste; owns or operates landfills, transfer stations, incinerators/waste-to-energy plants or material recovery facilities; recycles packaging, plastics, e-waste or metals under producer-responsibility mandates; or operates water and wastewater treatment assets under municipal contracts or concessions. Also route here for environmental remediation and industrial-cleaning services.
This is a route-density and permitted-asset business, not a manufacturing business. Two things determine almost all of the economics: how many stops a truck can service in a day within a given geography, and whether the tonnes that truck collects go into a hole in the ground the company owns. Everything else — price escalators, recycling commodity prices, fleet efficiency — is second order. The moat is a permit that a competitor cannot obtain at any price, and the offsetting cost is a liability that outlives the asset by thirty years and sits mostly outside reported debt.
The single most common analytical failure here is aggregation. A national EBITDA margin and one consolidated EV/EBITDA multiple describe a company that does not exist: these businesses are won or lost market by market, and a group average blends a dense metro franchise earning 40% margins with sub-scale rural routes losing money. Never accept a company-level margin without asking which markets produce it.
Two adjacent business models must be routed elsewhere. **Pollution-control and water-treatment equipment makers** (scrubbers, ESPs, membranes, pumps, filtration skids, ZLD systems) are industrial capital-goods businesses — order book, execution, working capital, customer capex cycle — and belong in `references/sectors/infra-capitalgoods.md`, not here; the fact that the end use is environmental changes nothing about their economics. **Regulated water utilities** with a rate base and an allowed return belong in `references/sectors/utilities-power.md`. **Waste-to-energy plants selling power under a PPA** are IPPs first and waste companies second — value the power contract with `utilities-power.md` and use this file only for the tipping-fee side.
## Contents
- [Why the generic ratio set fails here](#why-the-generic-ratio-set-fails-here)
- [The metrics that actually matter](#the-metrics-that-actually-matter)
- [How to value companies in this sector](#how-to-value-companies-in-this-sector)
- [Peer set construction](#peer-set-construction)
- [Sector-specific red flags](#sector-specific-red-flags)
- [Cycle and structural context](#cycle-and-structural-context)
- [India vs global notes](#india-vs-global-notes)
- [Checklist](#checklist)
## Why the generic ratio set fails here
Most generic ratios are computable here — that is the trap. They produce numbers that look sensible and mean the wrong thing.
**Consolidated OPM / EBITDA margin — an average of structurally different businesses, and of structurally different geographies.** Collection, transfer, disposal and recycling have margins that differ by 20–40 percentage points by design. A company that shifts mix toward disposal shows "margin expansion" with no operational improvement; a company that wins a large low-margin municipal collection contract shows "margin compression" while adding value. Worse, the same line of business earns completely different margins in different markets: a landfill with 60% of the local disposal market and a two-hour haul radius with no competing permitted site is a different asset from an identical landfill with three competitors. Demand the line-of-business split and, where disclosed, the market-level or region-level commentary. If neither exists, say the consolidated margin is uninterpretable rather than reporting it.
**EV/EBITDA as normally computed — systematically understates the price paid.** Closure, post-closure and environmental remediation obligations are contractually mandated, non-discretionary, cash-settled and long-dated. They are debt. They sit in provisions, not borrowings, so standard screeners exclude them from enterprise value. Add them. On a landfill-heavy operator this adjustment can move EV by a mid-to-high single-digit percentage, and it moves it most for exactly the companies with the oldest, most depleted asset bases — the ones a naive screen flags as cheap.
**D/E and net debt/EBITDA — mis-scaled in both directions.** High leverage is normal and appropriate: contracted, recession-resistant cash flows support it. But reported net debt omits AROs, remediation reserves, and any unfunded portion of financial assurance obligations. A company at 2.8x reported net debt/EBITDA can be at 3.4x on an honest measure. Compute both and use the adjusted number.
**FCF and FCF yield — routinely overstated, and the overstatement is deliberate in some disclosures.** Landfill cell development capex is not growth capex. It is the cost of producing the airspace you are about to sell — economically it is cost of goods sold, paid years in advance and capitalised. A company that classifies development capex as "growth" and reports FCF after maintenance capex only is reporting a number that can never be realised, because the spend recurs for as long as the landfill operates. Reconstruct FCF as operating cash flow minus fleet/container/equipment replacement capex minus landfill development capex; only genuinely new sites, new markets and new processing plants are growth.
**EBIT, D&A and therefore P/E — estimate-driven.** Landfill airspace is amortised on units-of-consumption: cost per tonne = capitalised plus estimated future development cost divided by estimated total remaining airspace. Raising the airspace estimate lowers the amortisation rate and lifts EBIT with no change in cash or operations. Add roll-up amortisation of acquired customer relationships, periodic remediation charges, ARO revisions and impairments, and reported EPS becomes one of the least comparable numbers in the sector. EBITDA is more comparable than EBIT here — but only because it hides the estimate, not because the estimate does not matter.
**ROCE / ROIC — broken by vintage and by goodwill.** A landfill permitted in 1985 and long since depreciated shows a spectacular return on a book value that bears no relation to what the permit is worth. The identical asset acquired last year, carried at fair value plus goodwill, shows a mediocre one. In a serial acquirer, group ROCE mostly measures acquisition vintage. Use ROIC only pre-goodwill and only against the company's own history, or replace it with cash return on the replacement value of the asset base.
**P/B — meaningless.** The core asset is a permit. Permits obtained organically carry essentially no book value; permits acquired carry an arbitrary purchase-accounting value. Book value is the sum of a depreciated fleet, historical-cost land and acquisition goodwill.
**Current ratio and working capital — inverted.** Subscription collection is billed in advance, so negative working capital is the healthy state. A rising current ratio usually means receivables are stretching — in India, that it is not being paid by a municipal body.
**Headline revenue growth — uninformative until decomposed.** Revenue growth is the sum of core price, fuel and environmental surcharges, volume, recycled-commodity price, recycled-commodity volume, acquisitions, divestitures and FX. These have entirely different quality and persistence. Eight percent growth from core price is a franchise; eight percent from a cardboard price spike reverses.
**Order book / order-to-revenue ratios — do not apply** to services operators, however common they are in the sector's own presentations for the EPC and equipment arms. If order book is the main disclosure, you are looking at a capital-goods business and should re-route.
## The metrics that actually matter
Ranges are **indicative only**. They vary by market density, sub-sector, regulatory regime, commodity cycle and period, and any of them can be wrong for a specific company in a specific year. Comparison against a tight peer set and against the company's own 5–10 year history overrides every absolute band below.
| Metric | Definition / how to compute | Indicative healthy range | Why it matters |
|---|---|---|---|
| **Route density** | Serviced stops (or lifts) per route-day; stops per hour on route; revenue per route-day; miles driven per stop; population or commercial-establishment density in the served postcodes. Where not disclosed, proxy with revenue per truck and revenue per employee by region. | Residential 700–1,200 stops per route-day in dense suburban geography; commercial 15–25 lifts per hour; miles per stop falling year over year. | The core economic driver of the entire sector. Truck, driver, fuel and depot cost are fixed for the day; the incremental stop on an existing route drops through at close to 100% margin. This is why a 30% local share earns nothing and a 55% local share earns 40% margins on the same price. It is also why national margin is meaningless: density is a local variable. Two operators with identical prices and identical fleets earn completely different returns if one has twice the stops per mile. |
| **Internalisation rate** | Tonnes collected that are disposed into the company's own landfills or routed through its own transfer stations, as a % of total tonnes collected. Report separately by region. | Integrated operators 60–80%; above 80% is very strong; below 40% means the largest single cost is set by a competitor. | Disposal is where the margin lives. If a third party owns the only permitted landfill in the haul radius, it captures the spread and can raise your gate rate at will — your collection business is then a low-return logistics operation with a supplier that is also your competitor. Vertical integration, not scale, is what converts route density into profit. Falling internalisation after an acquisition means the acquired volume was bought without the disposal to serve it. |
| **Remaining permitted airspace and remaining life** | Remaining permitted airspace (cubic yards or tonnes) plus, separately, "probable expansion" airspace not yet permitted. Divide by annual disposal volume for years of remaining life, by site and weighted for the group. | 20–40 years of weighted average life is strong; under 10 years without a filed and progressing expansion application is a value cliff. Probable-expansion airspace above ~20–25% of reported total deserves scrutiny. | Airspace is depleting inventory that cannot be restocked without a permit that may take 5–10 years and may simply be refused. A landfill with 8 years left is a wasting asset whose cash flows terminate and then invert into closure spending; a perpetuity growth assumption on it is nonsense. Track remaining life every year: if it falls faster than one year per year, the company is consuming its franchise. |
| **Permitting position and barrier to entry** | Number of permitted sites in each served market and who owns them; status of pending expansion and new-site applications; time since the last greenfield MSW landfill was permitted in that jurisdiction; local opposition, litigation and appeals. | Qualitative. The strongest position is a market where no new permit has been granted in 20+ years and the company owns the only site inside the economic haul radius. | This is the actual moat, and it is political rather than economic. Capital cannot replicate it — a competitor with unlimited money still cannot site a landfill next to yours. It is also the moat's fragility: the same political process can deny an expansion, impose host-community fees, or cap daily tonnage. Read the permit and litigation disclosures, not just the financials. |
| **Line-of-business mix and EBITDA margin by line** | Revenue and EBITDA split across collection (residential / commercial / industrial-roll-off), transfer, disposal, recycling/MRF, and other (special waste, remediation, energy). Compute margin on each and track mix shift. | Indicative: collection 25–32%; transfer 25–35%; landfill/disposal 35–50%+; recycling anywhere from negative to 25% depending on commodity and contract form; residential collection typically the weakest line. | The only way to know whether group margin moved because operations improved or because mix shifted. It also tells you where the earnings are exposed: a company with 25% of EBITDA from recycling has a commodity book inside a utility-like wrapper. Mix explains most cross-company margin differences before any judgement about management quality is warranted. |
| **Core price vs volume, and price above cost inflation** | Core price = same-store yield excluding fuel and environmental surcharges, excluding recycled-commodity price, excluding acquisitions. Report alongside volume change and against the company's own cash cost inflation (labour, fuel, third-party disposal, maintenance). | Core price 3–6% in a normal environment, higher in inflationary periods. The test that matters: core price minus cash cost inflation positive and stable. Volume flat to modestly negative is acceptable if price is holding. | This sector compounds on price, not volume. Underlying tonnage grows roughly with population and GDP at low amplitude, so the entire value creation is the ability to raise price faster than cost every year without losing customers — which is only possible where the permit and the density make switching expensive. A company reporting good revenue with decelerating core price is renting growth from surcharges, commodity or M&A. |
| **Contract structure: franchise / exclusive / contracted vs open market** | % of revenue under exclusive municipal franchise, non-exclusive municipal contract, long-term commercial contract, or open-market (subscription/spot). Weighted average remaining contract life; escalator basis (CPI, a trash-and-water-specific CPI sub-index, fixed %, cost pass-through); renewal and retention rates; re-bid calendar. | Long-dated contracted/franchise revenue 40–70% of the book is a strong, defensible position. Weighted average remaining life 5+ years. Municipal contract retention above 90% at re-bid. | Determines both the durability and the ceiling of pricing. Exclusive franchises give guaranteed density — the single best margin structure in the sector — but cap price to a formula and expose the whole block to a single re-bid date. Open-market commercial gives pricing freedom and higher margins but churns. Check what index the escalator uses: a contract escalating on headline CPI while the cost base is 55% driver wages compresses in a tight labour market, and the mismatch is invisible until it has run for three years. |
| **Customer churn and retention (open-market commercial)** | Net churn = lost revenue from cancellations and price-downs, less recovered, as % of opening recurring revenue. Split defection to competitors vs business closures. | Net churn below 8–10% a year on a commercial book. Rising churn concurrent with above-market price increases is the classic warning that price has run ahead of the moat. | Price and churn are a single decision, not two metrics. Any company can post strong core price for two years by pushing rates; the question is what happens to the route density that made the margin possible. Churn is the check on whether pricing power is real or borrowed. |
| **Closure, post-closure and remediation liabilities** | Recorded ARO / provision for closure and post-closure; the **undiscounted** total; the discount rate and inflation rate used; the accretion charge; remediation reserves separately; Superfund/CERCLA PRP site count and allocation share; expressed also per remaining tonne of airspace. Sensitivity to a 100 bp change in discount rate. | Judge the disclosure quality first. Recorded liability materially below the undiscounted figure is arithmetically normal; a rising discount rate assumption or a shrinking liability against a growing site base is not. | This is the long-dated debt that funds the asset. It is discretionary in accounting but not in cash: closure spending is mandated, post-closure monitoring typically runs 30 years or more after a site stops accepting waste, and remediation is joint and several under CERCLA-type regimes. Management sets the discount rate, the inflation rate, the assumed closure date and the assumed post-closure period — four levers, all of which move the liability without moving reality. Always add the liability to EV and check the assumption trail year over year. |
| **Capex intensity and its split** | Total capex / revenue; then split into (a) maintenance — fleet replacement, containers, plant overhaul, (b) landfill cell development, cost per cubic yard of airspace developed, (c) growth — new sites, new markets, new processing capacity. Also fleet age and replacement cadence. | Total capex 9–13% of revenue for an integrated operator. Maintenance alone typically 5–8%. Development capex should track disposal volume; a sharp drop is deferral, not efficiency. | Determines whether the reported cash flow is real. Development capex is the cost of the tonnes you will sell and must be deducted before any FCF claim. Maintenance capex is deferrable for two or three years, which is exactly how a company facing a bad quarter manufactures cash flow — and the deferral shows up later as a rising average fleet age and a maintenance-cost spike. |
| **Free cash flow conversion (honest basis)** | (Operating cash flow − maintenance capex − landfill development capex) / EBITDA, and the same over revenue. Reconcile to the company's own FCF definition and state the difference. | Integrated solid waste: FCF/EBITDA 40–55%, FCF/revenue 10–15%. Concession-heavy and build-phase businesses will be negative and that is structural, not a defect. | The sector's real claim to quality is cash conversion, and it is the number most often overstated. If the company's definition and yours differ by more than a couple of points of revenue, find out which capex line is being reclassified and report both. |
| **Recycling commodity exposure and contract form** | Tonnes processed; revenue per tonne by commodity (OCC, mixed paper, aluminium, PET, HDPE); % of inbound tonnes on fee-for-service / processing-fee contracts vs revenue-share or floor-price contracts; contamination and residual rate; the disclosed EBITDA sensitivity to a fixed move in the commodity basket. | Fee-for-service share above 70–80% of processed tonnes is a materially de-risked model. Residual/contamination rate below 15–20%. | Contract structure decides who owns the commodity cycle. Under a processing-fee model the customer pays a fixed rate per tonne and takes the commodity risk; under a revenue-share model the operator does. The same physical recycling business is a stable service annuity or a leveraged commodity bet depending purely on paperwork. Companies rarely re-paper contracts toward fee-for-service while prices are high — the incentive runs the wrong way, which is why the exposure surfaces at the bottom of the cycle. |
| **Landfill gas, renewable energy and environmental credits** | Installed gas-to-energy or RNG capacity; volumes sold; revenue and EBITDA from RINs, LCFS credits, RECs, carbon credits or Indian equivalents; % of group EBITDA; the price assumption embedded in guidance. | Treat any line above ~5–8% of EBITDA as a separately valued, policy-dependent stream, not core. | These are genuine cash flows on assets the company already owns, and the incremental margin is very high — but the price is set by regulation, not by a market with supply and demand fundamentals you can forecast. Credit regimes have been repriced by administrative decisions repeatedly. Value at a lower multiple than collection and disposal, and never let a credit-price spike be presented as operating improvement. |
| **Special waste and event-driven volumes** | Share of landfill tonnes and revenue from contaminated soil, remediation projects, industrial clean-outs, disaster debris, one-off construction and demolition. | Ideally identified and quantified. Sustained above ~10–15% of disposal volume needs a separate persistence judgement. | The highest-margin tonnes and the least repeatable. A hurricane, a large remediation project or a single plant demolition can add several points to disposal volume for four quarters and then vanish. Companies book it in the ordinary run rate; the next year's "volume decline" is then blamed on the economy. Strip it before extrapolating. |
| **Safety and regulatory compliance record** | TRIR / lost-time injury frequency rate, DART rate, preventable vehicle accidents per million miles; notices of violation, consent decrees, penalties, permit suspensions; odour, leachate and groundwater complaints; in India, CPCB/SPCB directions, closure notices and NGT orders and environmental compensation levied. | TRIR trending down and below the sector average; zero material consent decrees; no repeat violations at the same site. | This is a licence-to-operate metric, not an ESG decoration. A serious incident or a pattern of violations can cost a permit renewal or an expansion approval, which is the whole asset. It is also the cleanest available proxy for operational discipline in a business with thousands of vehicle movements a day — safety records and maintenance discipline degrade together, and both degrade before the financials do. Driver turnover belongs alongside: above ~25–30% a year it drives accidents, overtime and route inefficiency simultaneously. |
| **Water/wastewater operating KPIs (where applicable)** | Contracted treatment capacity (MLD or MGD); plants under O&M and under concession; plant availability/uptime; effluent quality compliance against consent conditions; non-revenue water reduction where contracted; receivable days from municipal counterparties; concession residual life. | Availability above 95–98% and compliance at or near 100% of tested parameters — penalties and annuity deductions are tied to these. Municipal receivable days are the number to watch, not margin. | For contract operators the technical KPIs determine whether the annuity or O&M fee is paid in full, since deductions are formulaic. But the binding risk is counterparty: an operator can hit every technical target and still fail because the urban local body cannot pay. Model collection, not just billing. |
## How to value companies in this sector
**Primary — EV/EBITDA with environmental liabilities added to enterprise value.** Use EBITDA rather than earnings because D&A is estimate-driven (airspace amortisation rates, acquired-intangible amortisation) and capital structures differ. But the standard EV is wrong here. Construct:
> **Adjusted EV = market cap + net debt + closure/post-closure liability + remediation reserves + unfunded financial assurance + pension deficit + minorities − equity-accounted stakes at value**
Use the recorded liability as the minimum; where the discount rate looks aggressive relative to peers or to the market rate on comparable-duration debt, re-discount the undiscounted disclosure yourself and use that. State which you used. Indicative multiples: high-quality integrated solid waste in developed markets 11–15x adjusted EV/EBITDA; non-integrated collection-only 6–9x; hazardous and specialty waste 8–12x; recycling-led lower and cycle-dependent; Indian concession-based operators 6–10x. Indicative only, and the spread within each band is explained mostly by internalisation rate and contracted-revenue share.
**Co-anchor — FCF yield on honest maintenance capex.** Because these are cash-compounding businesses with modest growth, FCF yield is often more decision-useful than the multiple. Compute FCF after maintenance *and* landfill development capex, as defined above, and compare to the company's own definition. A gap between the two definitions of more than a couple of points of revenue is itself a finding. Cross-check FCF yield against the ten-year government bond in the same currency; the sector's claim to a premium multiple is a bond-like, inflation-linked cash stream, so that spread is the real question.
**Asset-based — value per unit of airspace and per tonne of annual capacity.** Permits cannot be reproduced, so private-market transaction evidence in currency per remaining cubic yard (or tonne) of permitted airspace, and per tonne of annual disposal or collection volume, gives a replacement-value floor that is genuinely informative — unusual for a services business. Useful for cross-checking a roll-up's acquisition prices against public trading multiples, and for judging whether an old, nearly depleted asset base is being valued as though it were perpetual.
**DCF — the correct primary method for a single landfill or a concession, and a useful discipline for the group.** Model a landfill explicitly as a depleting asset: volume over remaining permitted life, gate rate escalating at contracted or historical rates, then the closure spend and 30-plus years of post-closure monitoring as explicit negative cash flows after the site stops accepting waste. Never apply a perpetuity growth rate to a landfill with a finite airspace estimate — that single error is responsible for a large share of the sector's mis-valuations. For concessions, model to the end of the concession term with a terminal value equal to the contractual residual or transfer value, which is frequently zero.
**Concession and annuity structures — value the contract, not the company.** For Indian MSW, waste-to-energy and hybrid-annuity wastewater projects, build a project-level equity cash-flow model over the concession term and solve for equity IRR against cost of equity, then sum SPVs and deduct holdco debt and costs. The logic is identical to a HAM road; cross-reference `references/sectors/infra-capitalgoods.md`. The critical input is not tariff but collection probability from the municipal counterparty.
**Sum-of-the-parts — the default for anything mixed.** A group with solid waste services, a hazardous-waste business, an environmental-remediation consultancy, a waste-to-energy IPP and an equipment-manufacturing arm must be valued piece by piece on each piece's own convention. A single blended multiple across those is arithmetic without meaning.
**Do not use:** unadjusted EV/EBITDA (understates by excluding closure and remediation obligations); P/B (permits carry no meaningful book value; goodwill dominates in roll-ups); standalone P/E (distorted by amortisation policy, ARO revisions, remediation charges and divestiture gains); EV/Sales across companies with different line-of-business mix or different gross-vs-net revenue recognition; DDM; and any single national multiple applied to a company whose market positions are heterogeneous — decompose or state the limitation.
## Peer set construction
A valid comparable shares *degree of vertical integration, waste stream, contract structure and regulatory regime*. Label similarity is worthless: "waste management" spans a landfill monopoly and an asset-light broker.
**Splits that must never be mixed:**
- **Integrated (owns disposal) vs collection-only vs pure disposal vs brokers.** Integrated operators earn the disposal margin; collection-only companies pay it away; pure landfill/transfer companies have higher margins and shorter effective asset lives; brokers gross up revenue for third-party disposal they never touch and show low single-digit margins on inflated revenue. Their EBITDA margins are not on the same scale and should never appear in one table.
- **Municipal solid waste vs hazardous/industrial (RCRA Subtitle C or Indian HOWM Rules) vs medical/biomedical vs E&P and radioactive waste.** Different permitting difficulty, different pricing power, and completely different liability tails. Hazardous waste commands higher margins precisely because the permit and the liability are heavier.
- **Waste services vs waste-to-energy IPPs vs equipment manufacturers.** WtE profit is largely a power tariff or PPA (`utilities-power.md`); equipment is order-book capital goods (`infra-capitalgoods.md`). Both are routinely bundled into "environmental services" indices and both destroy a peer set.
- **Landfill-led markets vs incineration-led markets.** In much of northern Europe and Japan, landfilling is effectively prohibited or heavily taxed; airspace is irrelevant and the moat is incinerator permits plus district-heating offtake. Comparing a US or Indian landfill operator to a European WtE operator on EV/EBITDA compares two different assets.
- **Franchise/contract-heavy vs open-market commercial-heavy.** Different growth ceilings, different pricing mechanics, different re-bid risk profiles.
- **Recycling-exposed on revenue-share vs fee-for-service.** One is a commodity business; the other is a service annuity.
- **Water: regulated utilities vs contract O&M operators vs EPC builders vs technology/product companies.** Four different playbooks behind one word.
- **India-specific: concession/BOT SPV operators vs asset-owning private operators vs EPC contractors to municipal bodies.** Under Ind-AS service concession accounting the first group's reported revenue and margin are an artefact of the accounting model chosen, not of operations — see the India section.
**Also align:** market density (a metro-focused operator and a rural one are not comparable regardless of size); accounting regime (ARO discount rates, airspace estimation practice and remediation-reserve conventions differ enough between US GAAP, IFRS and Ind-AS to make margins and multiples non-comparable without adjustment); fiscal year; the share of revenue from acquisitions in the last three years; and revenue recognition gross vs net of third-party disposal.
Aim for 4–8 peers. State the basis in the report, and benchmark every metric twice — against peers and against the company's own 5–10 year record.
## Sector-specific red flags
- **Core price decelerating while headline revenue holds up on surcharges, commodity or acquisitions.** The franchise is the price spread; everything else is borrowed. Decompose the revenue bridge every quarter and judge on core price minus cash cost inflation.
- **A change in the total airspace estimate that lowers the amortisation rate.** Margin improves, cash does not. Any restatement of estimated total airspace, assumed final elevation or assumed compaction density should be found and quantified before crediting a margin gain.
- **Rising share of "probable expansion" airspace in reported remaining life**, or a weighted average remaining life that falls faster than one year per year. The reported life is being propped up by permits not yet granted.
- **An expansion permit denied, appealed, or conditioned with a daily-tonnage cap or host-community fee.** This is a direct hit to the asset's cash flows and often to its remaining life; it rarely appears in guidance until the following year.
- **ARO assumption drift.** A rising discount rate, a lowered inflation assumption, a pushed-out assumed closure date, or a shortened post-closure period — each shrinks the recorded liability with no change in obligation. Track the four assumptions in a table across five years. A liability that falls while the site base grows needs a written explanation.
- **Remediation reserves recorded at the low end of a range, and reserve releases running through income.** Where the accounting permits the low end of a range when no point estimate is better supported, the low end is what gets booked. A company reporting environmental "gains" from reserve releases is recognising a benefit it has not banked.
- **Superfund / PRP exposure growing quietly.** New site designations, a rising allocation share, or co-PRPs becoming insolvent — joint and several liability means the solvent party pays. Read the legal proceedings and environmental matters notes in full, every year.
- **Roll-up arithmetic.** Revenue growth carried by acquisitions while same-store volume is flat or negative, goodwill and acquired intangibles rising faster than EBITDA, and pre-goodwill ROIC drifting down. Tuck-ins bought at 5–8x by a company trading at 13x create value; the question is whether the pipeline of dense, integrable targets still exists or whether the company has started buying volume it cannot internalise.
- **Falling internalisation rate after acquisitions.** Volume bought without the disposal to serve it. The acquired revenue arrives at a fraction of the assumed margin, and the shortfall is usually described as "integration timing".
- **FCF flattered by capex reclassification or deferral.** Development capex labelled growth; maintenance capex postponed; a rising average fleet age and rising per-truck maintenance cost are the tells that follow.
- **Special waste, disaster-debris or large one-off remediation volumes carried into the run rate.** Quantify and strip before extrapolating.
- **Recycling contracts left on revenue-share while commodity prices are high**, or a company describing a commodity-price windfall as an operational turnaround. Ask what the same book earns at trough commodity prices.
- **Environmental-credit income (RINs, LCFS, RECs, carbon) growing as a share of EBITDA and valued at the group multiple.** Policy-set prices are not operating income of the same quality.
- **Municipal contract losses at re-bid, or a large block of franchise revenue re-bidding in one year.** Check the re-bid calendar. Also check for contracts won at aggressive prices to build density that never arrived — those show up as a market with persistent sub-scale margins.
- **Deteriorating safety and compliance metrics.** Rising TRIR, repeat notices of violation at one site, a consent decree, a spike in odour or groundwater complaints, or driver turnover above ~30%. These precede permit trouble and margin trouble in that order.
- **PFAS and emerging-contaminant exposure not addressed.** Leachate treatment costs, potential retrospective liability at landfills and remediation sites, and insurance exclusions. A company that has not quantified its position is deferring a known and growing cost.
- **India — receivables from urban local bodies stretching, or a tipping-fee revision pending with the authority for more than a year.** Revenue recognised is not revenue collected. Also: SPV-level debt invisible in standalone accounts, promoter-affiliated EPC contracts inside a concession, and concession disputes or termination notices disclosed only in the notes.
- **Financial assurance backed by parent guarantees or self-insurance rather than bonds, letters of credit or funded trusts.** If closure and post-closure obligations are self-assured, the obligation and the credit risk sit in the same place.
## Cycle and structural context
**The volume cycle is real but shallow; the price cycle is what matters.** Municipal solid waste tonnage falls only modestly in a recession — households and offices keep generating waste — which is the basis of the sector's defensive reputation. But the cyclical components are much sharper: construction and demolition debris, industrial and manufacturing waste, roll-off pulls per container, and special waste all track construction and industrial activity closely and can fall by double digits. A company with heavy C&D and industrial exposure is not the defensive business the sector average implies. Judge cyclicality by mix, not by label.
**Cost cycles pass through with a lag, and the lag is the earnings volatility.** Diesel surcharges reset monthly or quarterly against an index, so a fuel spike compresses margin for a quarter or two and then normalises — do not extrapolate either direction. Labour is the harder problem: driver and technician shortages are structural in developed markets, wage inflation has repeatedly outrun the CPI indices embedded in municipal contracts, and turnover carries a compounding cost in accidents, overtime and lost route efficiency.
**The recycling commodity cycle repriced the industry once and can again.** The 2018 Chinese import restrictions collapsed mixed-paper and mixed-plastic values and pushed the industry from revenue-share toward fee-for-service processing contracts — a structural improvement in earnings quality that is only as durable as the discipline to keep repapering contracts through the next price upswing. Track the fee-for-service share; it drifts.
**Regulation is the growth engine and the cost driver simultaneously.** Landfill diversion targets, methane control rules on landfill gas, extended producer responsibility shifting packaging and e-waste costs onto producers, landfill taxes, and carbon pricing all create revenue pools for processors while raising compliance costs for disposal. PFAS is the live structural risk in developed markets: leachate treatment obligations, possible retrospective liability, and insurance market withdrawal. Score what is in flight in the specific jurisdiction before extrapolating margins.
**Consolidation is the sector's default corporate strategy, because permits are the moat and permits are acquired rather than built.** The arithmetic of buying tuck-ins at mid-single-digit multiples into a business trading at low-double-digits is genuinely value-creating while dense, adjacent targets remain. It exhausts. The late stage looks like acquisitions further from existing routes, lower internalisation, rising goodwill and flat organic margins.
**India is at a different point on the same curve.** The business is transitioning from informal, largely uncontracted collection toward formal, contracted municipal systems under the Solid Waste Management Rules, with a very long runway and a genuine structural growth story — but the sector's economics are dominated by concession terms and by the credit quality of urban local bodies rather than by route density and landfill ownership, because sanitary landfill airspace ownership by private operators is comparatively rare and processing/WtE/EPR-driven models are more common. Read the concession, not the industry.
## India vs global notes
| Dimension | India | US / global |
|---|---|---|
| Regulators and rules | CPCB and state pollution control boards; consent to establish / consent to operate under the Water Act 1974 and Air Act 1981; Solid Waste Management Rules 2016; Hazardous and Other Wastes (Management and Transboundary Movement) Rules 2016; Plastic Waste Management Rules and E-Waste Management Rules 2022 and Battery Waste Rules 2022 with EPR targets and tradable EPR certificates; Bio-Medical Waste Rules 2016; the National Green Tribunal as an active adjudicator imposing environmental compensation | US: EPA under RCRA Subtitle D (MSW) and Subtitle C (hazardous), CERCLA/Superfund for legacy contamination, Clean Air Act NSPS/EG rules on landfill gas, state environmental agencies issuing the actual permits. EU: Landfill Directive, Waste Framework Directive and the waste hierarchy, national EPR schemes |
| Typical business structure | Concession-heavy. A 15–30 year concession from an urban local body or state agency, held in an SPV, for collection and transportation, processing, waste-to-energy, or an integrated MSW facility. Wastewater increasingly on hybrid annuity (part construction payment, part performance-linked annuity, e.g. under Namami Gange), plus large EPC pipelines under AMRUT and Jal Jeevan Mission | Asset ownership. Companies own landfills, transfer stations and MRFs outright and hold perpetual-until-revoked operating permits; municipal work is a contract layered on owned disposal, not the asset itself |
| Revenue mechanics | Tipping fee per tonne escalated by a contractual formula (often WPI/CPI-linked), plus power sale under a PPA for WtE, plus RDF/compost/recyclate sales, plus viability gap funding or capital grants. Counterparty is a municipal body with weak finances — collection risk, not demand risk, is the central issue | Gate rate per tonne set commercially, collection priced per lift or per subscription, municipal franchises priced by tender with index escalators. Counterparty risk is minimal; pricing power is the variable |
| Environmental liability accounting | Ind-AS 37 provisions for closure and post-closure; disclosure is far thinner than the US ARO rollforward, and the discount rate, inflation assumption and post-closure period are often not separately stated. Expect to estimate. There is no Superfund analogue; legacy dumpsite remediation is largely government-funded (a revenue opportunity for operators) rather than a retrospective corporate liability, with "polluter pays" applied case-by-case through NGT environmental compensation | US: ASC 410-20 asset retirement obligations discounted at a credit-adjusted risk-free rate with accretion charged to operating cost, plus ASC 410-30 environmental remediation liabilities — commonly recorded at the low end of an estimated range. A full ARO rollforward, remaining airspace, remaining life and site counts are standard 10-K disclosure. IFRS uses IAS 37 with a current market discount rate, so recorded amounts are not directly comparable to US GAAP |
| Accounting trap to check first | **Service concession accounting (Ind-AS 115 / Appendix D).** Whether a concession is accounted for under the financial-asset model (an annuity receivable) or the intangible-asset model (a right to charge users) changes reported revenue, EBITDA margin and ROCE completely. Under the intangible model, construction revenue is recognised during the build phase with a thin margin, inflating revenue and depressing margin, then collapsing when construction ends — the identical distortion seen in HAM road developers. Identify the model before comparing any margin | Straightforward operating-company accounting. The main comparability issues are airspace estimation, ARO assumptions and acquired-intangible amortisation policy |
| Disclosure available | Annual report with segment note under Ind-AS 108, concession-wise disclosure that varies from good to nil, quarterly results and concall. Airspace, remaining life, route counts, internalisation and safety metrics are rarely disclosed — you will often have to reason from capacity in TPD/MLD and from tender documents. CARO qualifications and the related-party note matter, particularly for EPC awards to promoter entities | 10-K with line-of-business revenue and EBITDA, internalisation rate, remaining permitted and expansion airspace, average remaining site life, core price / volume / surcharge / commodity revenue bridge, ARO rollforward, TRIR, and fleet composition. The revenue bridge and internalisation disclosure are the two most valuable items and are essentially standard |
| Units and conventions | ₹ crore and lakh; capacity in TPD (tonnes per day) for waste and MLD (million litres per day) for water; fiscal year April–March; SEBI listing-obligation disclosures; promoter holding and pledge are material | $ millions; tonnes or short tons, cubic yards of airspace, MGD for water; calendar fiscal year typical; SEC filings on EDGAR; widely held ownership |
| Financing and assurance | Project finance at SPV level, often with a lender-mandated escrow over concession receipts; financial assurance for closure is weakly enforced in practice — treat that as risk, not as a saving | Financial assurance for closure and post-closure is mandatory via surety bonds, letters of credit or funded trusts; check the form, the cost and whether any portion is self-assured on the parent balance sheet |
## Checklist
- [ ] Confirm the business model: services operator, concession SPV, WtE IPP, or equipment maker — re-route equipment to `infra-capitalgoods.md`, WtE power economics to `utilities-power.md`, regulated water to `utilities-power.md`.
- [ ] Refuse to report a consolidated EBITDA margin without the line-of-business split; state that a national margin is uninterpretable if the split is unavailable.
- [ ] Build the revenue bridge: core price, surcharges, volume, recycled-commodity price and volume, acquisitions, divestitures, FX. Judge on core price.
- [ ] Compute core price minus the company's own cash cost inflation. That spread is the franchise.
- [ ] Establish route density by region — stops or lifts per route-day, revenue per truck, miles per stop — and say which markets produce the group margin.
- [ ] Compute the internalisation rate and track it against acquisitions; falling internalisation after M&A is volume bought without disposal.
- [ ] Pull remaining permitted airspace, probable-expansion airspace, and weighted average remaining life by site; check whether life is falling faster than one year per year.
- [ ] Check every pending expansion or new-site permit, its status, opposition and conditions. The permit is the moat.
- [ ] Test whether the airspace estimate changed, and quantify the effect on the amortisation rate before crediting any margin gain.
- [ ] Tabulate the ARO assumptions across five years: discount rate, inflation rate, assumed closure date, post-closure period. Compare recorded vs undiscounted.
- [ ] Read the environmental matters and legal proceedings notes in full: remediation reserves, range low-end recording, PRP site count and allocation share, PFAS exposure.
- [ ] Add closure, post-closure, remediation and unfunded assurance obligations to enterprise value before computing any EV/EBITDA.
- [ ] Rebuild FCF after maintenance capex *and* landfill development capex; reconcile to the company's definition and report both.
- [ ] Split capex into maintenance / development / growth; check fleet age and per-truck maintenance cost for deferral.
- [ ] Map contract structure: franchise, contracted and open-market shares, weighted average remaining life, escalator index vs actual cost base, and the re-bid calendar.
- [ ] Check commercial churn against the price increases taken — pricing power that raises churn is borrowed.
- [ ] Quantify recycling exposure and the fee-for-service vs revenue-share split; ask what the book earns at trough commodity prices.
- [ ] Separate landfill gas, RNG and environmental-credit EBITDA and value it at a lower, policy-risk-adjusted multiple.
- [ ] Identify and strip special waste, disaster debris and one-off remediation volumes before extrapolating.
- [ ] Review safety and compliance: TRIR, DART, vehicle accidents, notices of violation, consent decrees, driver turnover; treat these as licence-to-operate metrics.
- [ ] For a landfill or concession, build an explicit finite-life DCF including closure and 30-year post-closure outflows. Never apply perpetuity growth to finite airspace.
- [ ] India: identify the service-concession accounting model (financial asset vs intangible) before comparing any margin or ROCE.
- [ ] India: check municipal receivable days, tipping-fee revision status, SPV-level debt, concession disputes, CPCB/SPCB directions and NGT orders, EPR obligations and certificate revenues.
- [ ] Peer set: same integration level, waste stream, contract structure, market density and accounting regime — stated explicitly, 4–8 names.
- [ ] Benchmark every metric twice: against peers and against the company's own 5–10 year history.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| 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: 115k | Updated | Claude CodeCodex | Verified | Install |
| add-runner-evalpaperclipai/paperclipAdd or extend a Paperclip Runner protocol evaluation definition, roster, assertion, or report fixture with provenance and narrow validation. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| 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: 115k | Updated | Claude CodeCodex | Verified | Install |
| 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: 115k | Updated | Claude CodeCodex | Verified | Install |