FMCG business model

4.2.1 What the business actually is

An FMCG company owns almost nothing you can photograph. Its plants are modest and often partly outsourced. The trucks belong to transporters and the shops belong to shopkeepers. What it does own is two intangibles, and each one makes the other stronger:

• The brand is a promise installed in millions of heads by decades of advertising. It does two jobs. It lets the company charge ₹80 for ₹50 of materials, which is pricing power. And it makes the shopkeeper need to stock it, because customers walk in asking for it by name. That's shelf pull.
• The distribution system is the physical ability to put that brand within arm's reach of desire. In India that means reaching a universe of roughly 12 to 13 million kirana outlets. Hindustan Unilever's products reach on the order of 9 million outlets, and around 3 million of them are covered directly by company or distributor salesmen (illustrative, modelled on recent disclosures, so pull the latest annual report). A challenger can copy the formula in a lab in a month. Replicating nine million relationships takes decades. In the US, distribution runs through concentrated organised retail. Walmart, Procter & Gamble's single largest customer, is roughly 15% of its sales (illustrative, check in the 10-K's customer-concentration note). So the KPI shifts from outlet count to shelf-space share and to ACV-weighted distribution, the % of all-commodity retail volume that flows through stores stocking you.

You track the machine with three distribution dials. Numeric reach is how many outlets stock you. Weighted reach is the share of the category's sales flowing through those outlets. The two can be far apart, since 60% numeric reach in tiny shops can be 30% weighted reach. Direct coverage counts the outlets your salesforce bills directly. Those outlets get better fill rates, new-launch placement, and merchandising, and widening direct coverage is how incumbents squeeze challengers.

The product is bought weekly and priced in tens of rupees, so demand holds steady across cycles. Staples volumes barely notice recessions. Put that stability together with pricing power and the capital structure below, and you can see why the market treats large FMCG franchises as quasi-bonds with a growth kicker. It prices them that way too, as the valuation section shows.

4.2.2 Why ROCE is enormous: computed

Two design features make FMCG returns on capital look like typos.

Feature 1: asset-light. A staples company with ₹10,000 crore of revenue might carry only ~₹2,000 crore of net fixed assets. Its plants are simple and long since depreciated, and part of the output is contract-manufactured. That gives fixed-asset turns of 4 to 6×, against ~1× for a steel plant.

Feature 2: negative working capital. The trade pays fast. Distributors typically pay within days, often through channel financing, so receivable days run 10 to 15. Meanwhile the company pays its input and media suppliers slow, at payable days of 90+ computed on COGS. Payables also cover advertising and service vendors, which inflates a day count built on COGS. Note the mechanics, then move on. Inventory covers ~50 to 60 days. Add it up and suppliers and distributors finance the business's operations, and then some.

How is an FMCG company's ROCE calculated?

Build it for Saral Consumer Products (synthetic, ₹ crore):

• Pre-tax ROCE = EBIT / capital employed = 2,100 / 1,500 = 140%
• Post-tax ROIC = EBIT × (1 − 25%) / 1,500 = 1,575 / 1,500 = 105%

Every rupee invested in the operating machine pays itself back, after tax, in under a year. That has two consequences.

1. Growth self-funds. Growing sales 10% needs ~₹220 cr of new PP&E, minus ~₹70 cr of extra free financing from working capital. That nets to ≈ ₹150 cr, against ₹158 cr of incremental NOPAT. Reinvestment rate = g / ROIC = 10% / 105% ≈ 9.5% of profits. The other 90% is free to distribute. That's how big FMCG names pay out most of their earnings and still grow.
2. Reported ratios understate/overstate depending on ballast. Let the same company sit on ₹3,000 cr of accumulated cash and it reports capital employed of ₹4,500 cr and a ROCE of ~47%. Acquisitions add goodwill too. HUL's reported ROE fell to ~20% after the GSK Consumer merger loaded goodwill and intangibles onto the balance sheet, while the operating machine underneath still earns triple-digit returns on core capital. So always compute core ROCE (ex-cash, ex-acquisition intangibles) next to the reported number, and say which one you're quoting.

Now link this back to moats. The numbers prove the intangible-asset moat. A 105% ROIC sustained for decades, with stable market share, is the brand-and-distribution barrier turning up in the returns, right where Mauboussin said it must.

4.2.3 The KPI panel

These thresholds are analysts' rules of thumb for large-cap staples, so calibrate them to sub-category, geography, and cycle. A food company with commodity content runs a structurally lower gross margin than a beauty company. That gap is mix, and it doesn't signal weakness.

4.2.4 Volume vs value: the decomposition (worked)

Revenue growth is the product of its parts. You multiply them:

(1 + value growth) = (1 + volume growth) × (1 + price/mix growth)

Take Saral. It reports revenue up 9.2% and volume up 4.0%.

price/mix = 1.092 / 1.040 − 1 = 5.0%

That's healthy. Volume is growing at mid-single digits and the pricing is sensible. For small numbers, 4 + 5 ≈ 9 works as a quick check, but use the exact division in anything you publish.

Why does volume growth matter more than price?

Now try an inflationary year. A peer reports revenue +15% and volume −1%, so price/mix = 1.15 / 0.99 − 1 = +16.2%. That's fifteen percent "growth" with fewer units leaving the factory. It's inflation pass-through, and it tells you nothing about franchise strength. When input costs deflate, the price plank vanishes and reported growth falls back to the volume truth underneath. The India staples cycle of the FY23 era looked like this. So did P&G's FY23 globally, with organic sales up ~7% on pricing of ~+9% and volume roughly −2 to −3% (approximate, modelled on FY23 disclosures). Pull the 10-K's net-sales bridge, which P&G helpfully splits into volume/price/mix/FX every year.

So why do you want volume growing at mid-single digits? Because volume growth means more households, more occasions, more usage. It's the franchise physically getting bigger. Price growth can be borrowed from inflation, cosmetic when it's really mix, or stolen from the future by pushing price until share leaks to regional brands. Three refinements:

• Volume ≠ units. Indian FMCG "volume" is usually tonnage. Grammage cuts, where the ₹10 pack shrinks from 80g to 72g, show up as negative tonnage but flat unit sales. Companies call this "price-pack architecture." Read the volume definition in the results footnotes. HUL reports UVG (underlying volume growth), and Ind AS revenue is net of GST and most trade discounts.
• Mix hides inside price. "Price/mix +5%" could be +5% of like-for-like pricing, which is strong. It could be zero pricing with customers trading up to premium SKUs, which is also strong but works another way. Or it could be +8% pricing masking −3% down-trading, which is weak. Management commentary and the rural/urban split usually let you tell them apart.
• Watch the price-volume elasticity play out over 2 to 3 years. Aggressive pricing loses share to regional players. That forces price rollbacks and LUP launches, and the margin makes a round trip. Run the decomposition as a time series and it shows you the whole movie.

4.2.5 A&P: the P&L line that is really capex

Advertising & promotion runs at 8 to 12%+ of sales for a healthy franchise. P&G spends roughly $8 to 9bn a year, ~10 to 11% of sales (illustrative, modelled on ~FY24 scale). It's expensed through the P&L, but it works as maintenance-plus-growth capex on the brand asset. Accounting treats the brand as free, and the P&L carries its upkeep. That creates the sector's most seductive earnings lever.

What happens when a company cuts A&P to lift margin?

Illustration: the margin that ate the brand (Saral, year 2). Sales grow 6% to ₹10,600 cr. Gross margin holds at 50%, or ₹5,300 cr, and other opex grows 6% to ₹1,802 cr. Then management cuts A&P from 10.0% to 8.2%. That's ₹869 cr, against the ₹1,060 cr a constant ratio would have cost.

• EBITDA = 5,300 − 869 − 1,802 = ₹2,629 cr, a 24.8% margin (vs 23.0%)
• Had A&P held at 10%, EBITDA = 5,300 − 1,060 − 1,802 = ₹2,438 cr, or 23.0%. Flat.
• So all 180bps of "margin expansion" (₹191 cr, 1.8% of sales) came from the A&P cut. Headline EBITDA grew +14.3%. The underlying business grew +6%.

The street cheers for two quarters. Then the shelf fights back. Regional rivals and D2C brands outspend you where it hurts, trial of your new launches slows, and volume growth fades. The lag runs 12 to 24 months, long enough for everyone to pretend the two facts are unrelated. So build the expert's reflex: any quarter where EBITDA margin rises while A&P % falls gets a bridge. Ask how much of the expansion is A&P. Expansion driven by gross margin is real, and it comes from input costs or premiumisation. Expansion driven by A&P is borrowed.

4.2.6 Premiumization, rural/urban, and the long runway

• Premiumisation means shifting the portfolio up the value ladder: from soap bar to bodywash, from powder to liquid detergent, from mass to masstige beauty. It lifts realisation per tonne and gross margin, and over a decade it's the main India volume-value kicker. Measure it with the premium share of the portfolio, the realisation/tonne trend, and the gross margin trend with input costs neutralised. Then interrogate it. Is this a real mix shift, or a euphemism for price hikes?
• Rural/urban mix. Rural India is roughly 35 to 40% of FMCG sales. It's more price-elastic, and it moves with monsoons, crop prices, and government transfers. Rural stress shows up first as down-trading. The share of LUP (low-unit-price ₹5/₹10 packs) rises and premium mix stalls, all before anything shows in aggregate volume. If growth is urban-led while rural is flat, the franchise is narrower than the headline suggests.
• The runway is the strategic reason India staples trade rich. Per-capita consumption of most categories is a fraction of global levels, and weighted distribution keeps deepening. A US staple grows 3 to 4% nominal. A well-run Indian staple believes in high-single-digit, volume-led growth for a decade. Whether you should pay for that belief is a valuation question, and it's next.

Anchors. India: Hindustan Unilever has revenue of ~₹62,000 cr, gross margin ~50%, EBITDA margin ~23 to 24%, A&P ~9 to 10% of sales and reach into ~9 million outlets. Its core ROCE is north of 100% once you strip out the GSK-merger goodwill, and that goodwill is why reported ROE is ~20%. All of these are illustrative, modelled on ~FY24/FY25 scale, so pull the annual report and the investor-day deck. US: Procter & Gamble has net sales of ~$84bn, gross margin ~51%, core operating margin ~24 to 26% and advertising of ~$8.5bn. Its net-sales bridge publishes the price/volume/mix split every quarter (illustrative, modelled on ~FY24/FY25 scale, so pull the 10-K). It's the same machine in a different register. HUL's game is distribution depth and premiumising a billion consumers. P&G's is brand share and mix in mature markets, where volume growth is ~1% and the fight is for pricing without losing share.

4.2.7 Valuation: P/E, EV/EBITDA, and India's premium-multiple regime

The lens is P/E and EV/EBITDA on smoothed earnings, with A&P at a level nobody has cut to flatter them. Cross-check with FCF yield. FCF ≈ PAT here because reinvestment is tiny, so check CFO/EBITDA ≥ 80 to 90% and capex ≈ depreciation. Nobody argues about the lens. The argument is always about the level. Large India staples routinely trade at 50 to 70× earnings against 20 to 25× for their global parents (as of mid-2026). HUL ~50 to 55×, Nestlé India ~70×, P&G ~24× and Unilever plc ~19× have been representative orders of magnitude.

How do you reverse an FMCG multiple?

Reverse the multiple with the key-value-driver formula from your valuation work, Value = NOPAT × (1 − g/ROIC) / (r − g). The steady-state fair P/E on NOPAT is then (1 − g/ROIC)/(r − g):

• ROIC 100%, growth 10%, cost of equity 11% gives (1 − 0.10) / 0.01 = 90×
• Same machine, growth 8%, gives (1 − 0.08) / 0.03 ≈ 31×
• Growth 6% gives (1 − 0.06) / 0.05 ≈ 19×

These are teaching values. A single-stage formula is being abused for intuition here, with the valuation phase's blessing. Three lessons fall out.

First, the premium isn't madness. At 100% ROIC almost every rupee of growth is free cash, so growth is worth an extraordinary multiple.

Second, the multiple is a growth-duration bet, and the derating maths is a cliff. Growth fading from 10% to 8% cuts fair value by two-thirds. Two points! That's how a great company at 60× delivers zero returns for seven years while earnings compound 10%, because multiple compression eats the growth. Several marquee India staples lived through this from the early 2020s.

Third, the scarcity and flows layer can hold multiples above any DCF-defensible level for years. That layer is the limited float of reliable compounders, plus relentless domestic SIP flows. Respect it, but never underwrite it. The rule from relative valuation still stands: buy the multiple only if you'd happily underwrite the implied growth for a decade-plus. Otherwise the quality is real, but the price has already eaten it.

4.2.8 FMCG red flags

How do you spot D2C eating into an FMCG moat?

One more word on the D2C flag. Distribution moats are channel-specific, and nine million kiranas mean nothing inside a beauty app's search results. So watch the incumbent's online share of its category against its offline share. A big gap means the moat's edge is fraying, and you can see it years before the aggregate share number moves.