So, those who want to play the real-estate cycle, do it though the “safer” proxies like cables, pipes, tiles, and paints.

Since it is hard to value and understand real estate developers, we usually good variation between price and value. How to calculate the value is something we’ll cover in this article.

This article is a walkthrough of how to actually value a developer, using one real company as the worked example the whole way through: AGI Infra Ltd, a fantastic Jalandhar-based developer in Punjab.

By the end, you will have a repeatable framework you can point at any developer.

First - Let’s Understand Why Developers Are Interesting

Three reasons this sector is worth the effort.

1. The accounts hide the business, not the other way around: A developer sells a flat 3 to 4 years before delivering it. However, the revenue only shows up in the P&L at FINAL handover. So today’s reported numbers describe the business of years ago. Meanwhile the customer via his installments ( linked to project completion milestones) is funding the project.

2. The valuation gap is unusually wide. The “safer” ancillaries (cables, pipes, tiles, paints through which people generally play real-estate cycle) have traded at 30-50x earnings for years. Developers, once you measure their earnings correctly, sit at 6-13x while growing pre-sales 25%+ (the correct metric you should be looking at). That is one of the widest gaps between a sector and its own suppliers you will find.

3. You get the specific story. If you want exposure to a localised growth story - Eg: Punjab township boom, or land near a new airport. You can’t get this through a cable company. Only the developer who owns that land does. This is the best exposure to localised growth stories

4. This cycle is structurally different. The 2005-08 cycle was debt-fuelled land hoarding, and it ended in a lost decade of demonetisation, RERA, the NBFC crisis, and COVID. What replaced it is a consolidation cycle. Listed players have repaired their balance sheets. Many players are now effectively net cash.

Why You Cannot Use P/E Here

This is the single most important idea.

• A flat sold today is not revenue today.

• Under Ind AS 115, most developers book revenue only at handover (possession).

• AGI Infra follows this same rule.

So AGI Infra’s reported P/E of ~39x reflects on stale profit. It reflects flats handed over last year, not flats being sold now. So what do we look at instead?

The right tool is NAV, not P/E.

This is a simple way to understand NAV (Similar to liquidation value):

If you unlocked everything the company owns today (land, unsold inventory, cash) and paid off what it owes, what is left for shareholders?

The Formula

This is the whole framework in one line and we’re going to use this on the stock for discussion today

NAV = (GDV × Embedded Margin, discounted) + Rental Book (fair value) − True Net Debt

The terms look difficult, I know. We’ll explain each one - so hang around. We will now build each piece using AGI Infra's numbers

Term 1: GDV × Embedded Margin

This term attempts to capture the “profit” still trapped inside unsold land and inventory. Let’s break it down

1. GDV = total sellable value of everything the developer can still build and sell.

Let me give you an example: If a developer owns land where they can build 100 flats.

Each flat will sell for ₹50 lakh. In this case,

GDV = 100 × ₹50 lakh = ₹50 Cr

GDV is just the total value of everything they can build and sell, priced at today’s rates. It’s the revenue they’d collect if they sold every unit.

2. Embedded margin = the profit % baked into that value.

But ₹50 Cr isn't profit. Building those flats costs money. Say all-in cost = ₹40 Cr.

In this case the Profit = ₹50 Cr − ₹40 Cr = ₹10 Cr.

Embedded margin = ₹10 Cr ÷ ₹50 Cr = 20%

So embedded margin is just: out of every ₹100 of sales, how much ends up as profit.

Putting it together

GDV × Embedded Margin = ₹50 Cr × 20% = ₹10 Cr

That ₹10 Cr is the profit still locked inside the land and unsold flats, waiting to be earned as the project gets built and sold.

Step A: Get the area

AGI Infra’s investor deck gives clean, cross-checked figures:

• Unsold in completed projects: 3.31 lakh sq ft

• Unsold in ongoing projects: 91.57 lakh sq ft

• Upcoming projects (all unsold): 78.73 lakh sq ft

• Total unsold: ~173.61 lakh sq ft

Step B: Get the margin

• AGI’s deck guides to ~25% PAT margin on its pipeline. A rare, direct disclosure.

• We haircut this to 20%, because you never take management’s number at face value.

Step C: Get the price (our biggest assumption)

AGI does not disclose a rupee GDV or a per-sq-ft price. So we have to build it from live listings from AGI’s own projects, project by project, then group by segment.

First, a technical point that confuses a lot of people. The distinction between saleable area and carpet area

Teaching note (carpet vs saleable): “Saleable area” already excludes parks, roads, and open green space. It is only the chargeable floor area sold to buyers. So we correctly value all 173.6 lakh sq ft.

But, there’s another term you must’ve heard -Carpet Area. AGI’s own site shows a Jalandhar Heights III 2BHK as 1,000 sq ft saleable vs 753 sq ft carpet. This comes to a a ~33% loading.

If you apply a carpet-quoted market price to a saleable area, you overvalue by a third. So a key thing to watch is to match the basis. AGI’s pipeline is disclosed in “saleable” area, so we use saleable-basis prices.

Now we will estimate prices from various AGI project listings:

• AGI Sky Villas, Ludhiana (luxury): ~₹6,200/sqft (developer rate); ₹1.85 Cr for a ~2,600 sqft unit

• Jalandhar Heights III, 3BHK (premium): ₹90L for 1,720 sqft → ~₹5,200

• Jalandhar Heights, 4BHK (premium): ₹1.22 Cr for 2,800 sqft → ~₹4,360

• JH-III 2BHK (affordable): ₹44.5-47L for 1,000 sqft → ~₹4,600

Grouping the unsold pipeline by segment at these sourced rates:

Step D: Apply the embedded margin

• AGI guides to ~25% PAT. Its actual delivered margin has run 18-21%.

• We anchor to delivery, not the guidance, and use 18%.

• Profit pool = ₹9,030 Cr × 18% = ~₹1,625 Cr

Note: AGI’s audited FY25 net margin was 20.52%, so our 18% embedded-margin assumption is deliberately a touch conservative.

Step E: Discount for time and execution (properly)

As we discussed, the profit does not come in today. Most of it arrives 5 to 7 years out, as projects finish and flats are handed over. Our DCF brains tell us that money that far away is worth less now. So we shrink it in two steps.

1. Time discount.

• We keep sale prices at today’s levels (conservative, no betting on price rises).

• Because prices are held flat, we discount at a real rate of ~10% a year.

• Rule of thumb: profit arriving on average ~6 years out is worth roughly 60-65% of face value today.

• ₹1,625 Cr × ~0.63 = ~₹1,024 Cr

2. Execution haircut: 15%.

• Analysts apply 20-40% to developer NAV for delay and cost-overrun risk.

• AGI’s clean delivery record (all 10 completed projects fully occupied) argues lower. But ~44% of value sits in unlaunched projects in unproven new markets, which argues higher. 15% splits it.

• ₹1,024 Cr × 0.85 = ~₹880 Cr

Term 2: Rental Book (Fair Value)

This is another nuance in valuing a real-estate business. Some developers own rent-earning assets (malls, offices).

These have to be valued separately, like a REIT: annual rent × a multiple.

Why rental assets matter: they are the sector’s downturn insurance. When the property cycle turns and new sales dry up, rent keeps flowing in regardless. That steady cash lets a developer finish its under-construction towers while pure-play rivals are stuck begging lenders for money. It is the difference between surviving a downturn and being forced into distress.

AGI Infra basically does not have this.

• Its “Investment Property” line = ₹191 Cr.

• But almost all of it is land held for future development, not rented assets.

• Actual annual rent = just ₹2.67 Cr.

So we carry it at book value (₹191 Cr) and move on.

Important note: There is no disclosed fair value for this land, unlike Oberoi (which discloses ~3.6x book). AGI’s land is probably worth more than book, but we will not invent a number. We knowingly understate here.

Term 3: True Net Debt

Never trust the reported debt number in real estate. I’ll give you 3 ways to do this. Run three checks.

Reported net debt: ₹109.85 Cr [Net debt = Total borrowings − Cash & bank balances
=> ₹137.42 Cr − ₹27.57 Cr = ₹109.85 Cr]

1. Add JV debt. Why: Projects held in joint ventures keep their loans off the parent’s balance sheet, they show up as one net “investment” line, with the debt invisible. You have to add back your economic share ( Debt * holding %).

For AGI: Nothing to add. Its only subsidiary is dormant, with no debt of consequence. No hidden JV leverage. A good sign!

2. Add quasi-debt. Why: Deferred land payments, dues owed to landowners, and soft instruments (like payable-when-able NCDs) are real claims that don’t sit in the debt line. (Basically, the builder tells pre-pay land-owner directly, they defer payments). They behave like debt, so treat them like debt.

For AGI: None found. No deferred-land structures, no landowner obligations, no soft instruments. Genuinely clean on this test.

3. Run the interest test. Why: Interest is treated differently. Developers are allowed to capitalise interest costs. You have to find Interest paid (cash flow) minus interest expensed (P&L). This number reveals how much interest a developer is burying in inventory instead of running through profit. A big gap means reported margins are flattered.

The concept. When a developer borrows money to build a project, it pays interest on that loan. Accounting rules (Ind AS 23) let developers do something specific with this interest: instead of showing it as an expense in the P&L, they add it to the value of the inventory, the under-construction project itself. This is completely normal and allowed. The logic is that the interest is a genuine cost of building the asset, so it sits with the asset until the flats are sold. It only hits the P&L later, as part of cost of sales, when those flats are handed over.

Now to AGI. It shows only ₹12.66 Cr of finance cost despite having ₹137 Cr of borrowings. That gap tells you most of its interest isn’t being expensed, it’s tucked inside the ₹818 Cr inventory instead. This is standard practice, not a warning sign. The one limitation: AGI doesn’t disclose exactly how much interest was capitalised, so we can’t measure precisely how much this flatters current margins. A minor transparency flag, nothing more.

Note: Strictly, the interest test one is about margins, not debt, so it doesn’t change our ₹109.85 Cr figure. We run it here because it’s the same kind of detective work: things hiding in the notes that the headline numbers don’t reveal.

The summary: the two big debt-inflators, JV loans and quasi-debt, are genuinely absent. Interest is being capitalised (normal for the sector), which flatters margins but doesn't change net debt. On the debt line itself, this is a clean balance sheet.

Hence, True Net Debt: ₹109.85 Cr (unchanged from reported)

Putting It Together

• NAV per share: ~₹77 (on 12.5 Cr shares)

• Current price: ~₹298 (market cap ~₹3,672 Cr)

• Price to NAV: ~3.8x

But Static NAV Misses One Thing: Growth

Let me explain this with an analogy.

Think of NAV as counting the fruit currently on a tree’s branches. That’s the 173.6 lakh sq ft, projects already named, disclosed, and owned.

But a good developer doesn’t stop there. It takes the profit from this year’s harvest and buys new land for next year’s projects, land that doesn’t exist as a named project yet, so it can’t appear in our area table. NAV has no way to price fruit that hasn’t grown yet, even if the tree looks healthy and is likely to keep bearing.

That is fine for a developer winding down. It understates one that keeps buying land and launching new projects.

AGI Infra genuinely shows signs of being the second kind, and this deserves real credit:

• All 10 completed projects are at 100% occupancy. In a sector where delays are routine, this is a rare, verifiable execution record.

• Recent PAT margins (21%, 29%, 28%) already run close to, and sometimes above, its own long-term margin guidance.

• Land reserves beyond the disclosed pipeline: the company holds a whopping ~6.25 lakh sq ft of additional FAR across several Punjab parcels, separate from the 173.6 lakh sq ft already counted. This presents Real optionality for the next cycle of projects.

• Something called as MIVAN construction technology: an aluminium formwork system, has meaningfully sped up build times, and gives AGI a genuine edge over local, less mechanised peers.

• New geographies (Ludhiana, New Chandigarh) carry higher price points than home-market Jalandhar, real upside if the expansion lands.

The Discipline: Sanity-Check the Growth Story Against Its Own Timeline

This is the most useful, transferable lesson in the whole exercise.

A growth story is a hypothesis. Test it against the company’s own numbers, not just the narrative.

AGI’s own investor deck sets a horizon: 172 lakh sq ft of saleable area by 2032. That is roughly the entire unsold pipeline we valued, ~₹9,030 Cr of GDV, expected to convert into revenue over the next six to seven years.

Do the arithmetic on what that requires. Converting ₹9,030 Cr of GDV into revenue by 2032 needs an average of well over ₹1,000 Cr a year in the closing years, since delivery is back-loaded toward 2030-2032.

Check that against what has actually shown up: FY24 revenue was ₹292 Cr. FY26 revenue was ₹352.5 Cr, growth of under 10% a year.

That is not, on its own, damning. Most of this pipeline is genuinely not due to convert for several more years, so a slow start is expected, not alarming. But it is the checkpoint to watch: the pace needs to accelerate sharply and soon for the 2032 target to be realistic

Second Lens: Economic PAT

Quick cross-check.

Economic PAT = current sales × embedded margin.

The term exists because accounting PAT is misleading; remember, a developer books profit only at handover, years after the actual sale. Meanwhile the customer is paying installments based on completion milestones

Economic PAT asks what the company is really earning on what it is selling right now.

The right way to test this is backlog and not margins. A company’s reported margin can look perfectly fine even while a huge pile of already-sold value sits unrecognised on the balance sheet. The margin tells you about the quality of profit on whatever got recognised this year. It tells you nothing about how much is still waiting in line. For that, look at contract liabilities (customer advances for flats not yet handed over) against annual revenue. That ratio tells you how many years of already-sold value is sitting in the pipeline, unrecognised.

AGI Infra:

• Contract liabilities (FY25): ₹731 Cr

• FY25 revenue: ₹325 Cr

• Ratio: ~2.25x, roughly two and a quarter years of already-booked sales still waiting to convert to reported revenue.

Compared to the big Mumbai names this framework was built around:

• Kolte-Patil: ~5.9x (₹4,312 Cr advances / ₹730 Cr revenue)

• Godrej Properties: ~7.0x (₹36,000 Cr / ₹5,131 Cr)

Takeaway: AGI’s backlog is real, roughly 2.25 years, but it is a fraction of Kolte-Patil’s or Godrej’s. That means AGI’s reporting lag is genuinely shorter, and correspondingly, less of its true earnings power is sitting hidden, waiting years to surface. This is a real, useful finding. It is a softer one than “there’s little hidden value here”: 2.25 years is still a meaningful backlog, just a much smaller one than a big metro developer carries.

Third Lens: Management Character

Read the accounting policy as a character statement.

• Conservative revenue recognition (handover-based, meaning revenue is recognised at handover, not at the time of customer paying installments). Good sign of being super conservative.

• Unquantified interest capitalisation. We discussed this already.

• Two stock splits + a QIP in 12 months. Not alarming, but adjust every per-share number before comparing.

• All 10 completed projects at 100% occupation. Genuinely strong, verifiable delivery record.

• Promoter holding ~71%. Strong alignment, but thin free float means the price can swing hard on low volume.

• Zero promoter pledge. 71% promoter holding with 0% pledged (declared under SEBI SAST). Removes the most common small-cap real estate red flag. A genuine positive.

• A related-party web worth watching. AGI has a ₹325 Cr construction contract with AGI Construction Pvt Ltd (promoter-linked), and arrangements with AGI Majestic (where AGI Infra collects buyer payments in its own name, then passes them on) and AGI Hospitalities (a ₹18 Cr corporate guarantee, plus project land still held in the related party’s name). All disclosed and stated to be at arm’s length. Not evidence of wrongdoing, but the kind of structure that makes “whose margin is this really” harder to answer.

What This Case Study Teaches

Valuing a developer is two steps

Step 1: Build the static NAV.

• Estimate GDV, apply a defendable margin, discount for time and risk.

• Add a real rental book only if one exists.

• Compute debt yourself, do not trust the reported line.

Step 2: Price growth separately.

• Ask if there is a genuine engine for new projects (land, tech, new markets).

• Then stress-test that growth against delivered results before paying for it.

The thread tying both together: never take a number on faith. Not the reported P/E, not management’s own long-term target, not a static NAV assumed to be complete.

AGI Infra is a clean worked example of exactly why that discipline matters. It has a genuinely good, well-run business: a clean delivery record, consistent margins, and real growth optionality. Whether today’s price is fully earned by that quality, or is running ahead of what has been delivered so far, is the open question this framework is built to help you ask, not answer for you.

Run this framework on the next developer you look at, and the whole sector stops being scary and starts being readable.

Comment and tell me what you liked and hated!

Leave a comment

PS: I learnt a lot from this article: Link mentioned below. Read Nine One Capital’s article.

Link

Disclaimer: This article is for educational and informational purposes only. I am not a SEBI registered investment advisor. I might hold positions in the securities discussed here. Nothing here is a buy, sell, or hold recommendation, and none of it is investment advice. Investing involves risk, including the loss of principal. Do your own work. Consult a qualified advisor before acting on anything you read.

A Rs 1400 Cr company in Vadodara is chasing Rs 2,500 Cr topline. Here’s the catch.

It makes washers. The little little steel rings that go under a bolt.

I’m sure you’re going to throw a fit about this being another commodity business, but hold on - there’s more to it than meets the eye.

This company one is India's number one producer and exporter of washers. Fortune 500 factories across 31 countries buy from it.

Before we dive into the main course, here are some basic screener numbers for starters:

• Market cap Rs 1,470 Cr, on sales of Rs 1,069 Cr.

• Five-year profit CAGR of roughly 54%.

• Five-year average ROCE of ~16%. The god metric, and it’s good.

• Trades at about 22x earnings. For that kind of growth, not expensive at all.

• CMP ₹206 ( up 54% in the last 1 year)

So where does the Rs 2,500 crore actually come from? Straight from the horse’s mouth. Here’s the MD on the last concall:

“Stainless business and existing business will be growing with the 25% CAGR, and in next 3 years, we can reach to ₹1,800 crores from the current business line. And equally, that CCL business can be established in the next 2 years. That can give ₹750 crores as a top line. So altogether, in a consolidated leader, ₹1,800 crores plus ₹750, that can be reached to ₹2,500 crores of the top line.”

We’ll come to business lines in a bit, but let’s decode this first - Split that into two bets.

• The boring, “proven” core compounds at 25% and reaches Rs 1,800 crore in three years.

• The new unproven bet - The copper plant gets built and adds on another Rs 750 crore. That’s roughly 40% more topline, at nearly double the margins ( We’ll have to see how much of this pans out by FY28; in either case, the story is interesting)

Here's the value investor's version of the same thing. If the copper bet flops, you still own a core growing ~25% a year, bought at 22x.

It doesn’t take a genius to see the setup. If the Rs 2,500 crore lands, we’re looking at a 2x in the stock with good downside protection. If it works, you get that 40% optionality on top, and the stock likely doubles conservatively.

Our job in this article would be to find out whether the management can do what they have committed to. So the core question is..

How does a washer maker pull that off?

Two answers. One is a quietly brilliant playbook it has run for 25 years. We’ll come to it in a minute; it’s the best part.

The other is that second bet. And it’s a strange one.

Before we get to the clever bit, you need the map. It takes five minutes. Let’s take a visit to the hardware store.

Hardware Store 101. What They Actually Make.

Ratnaveer (that’s the company under investigation today) runs four product lines today, across five facilities in Gujarat. All four are stainless steel.

There’s a fifth line coming. Copper. That one breaks every rule, so we’re parking it for now.

PS: I’ve attached videos of products for better understanding. But they’re not really necessary

Here are the four:

1. Washers and Fasteners (circlips, retaining rings, e-clips, nuts, bolts) - The crown jewel.

• Purpose: Their function is to keep 2 things together.

• Who uses: automotive and industrial OEMs, Fortune 500 names, 31 export countries

• Revenue contribution: ~20%

• Special mention: India’s #1 producer and exporter of SS washers. 2,500+ washer types, 1.5 bn units a year. This is the crown jewel and also the moat

2. Finishing Sheets (satin, hairline, mirror, scotch-bright)

• Purpose: polished stainless sheet for surfaces you see every day

• Who uses: elevators, appliances, metro rail, airports, construction

• Revenue: ~52%, the single biggest line

• Special mention: basically a polish-and-sell business. Thin margins, no moat, but it’s the volume engine and cash cow that funds everything else.

3. Tubes + Pipes (welded, cold-drawn, electropolished, duplex)

• Purpose: stainless tubing for industrial and hygienic use

• Who uses: oil and gas, pharma, power, nuclear, architecture

• Revenue: ~17% (FY24)

• Special mention: the weakest, smallest line. Plain tube is commodity. The electropolished/duplex sliver is where they’re trying to climb.

4. Solar Mounting Hooks (25 variants)

• Purpose: precision steel clamps that fix solar panels to roofs

• Who uses: solar installers in India and Europe

• Revenue: sits inside sheet-metal components, ~8% (FY24)

• Special mention: This was an opportunistic bet. Promoters saw the steel sheet raw material and married it with a brand-new demand pool (solar roofs). They saw the solar ganga and washed their hands in it.

The takeaway - All these are pieces to a larger puzzle :

Washers are the moat. Sheets are the cash. Tubes are the filler. Hooks are the bet.

Confused? Now look at that list again. Washers, sheets, tubes, hooks. It reads like a hardware store that lost the plot. And they’re about to bolt copper laminate for circuit boards onto it.

Looks like a mess, right? It isn’t.

There’s one thread running through all of it. Once you see it, the whole company clicks. That thread is the playbook.

Let’s get into it. The playbook

The Playbook. Why the Mess Is Actually a Machine.

Ratnaveer doesn’t sell random stainless steel products. It runs the sharp playbook.

There are actually two playbooks stacked on top of each other. A small one for how they build each business. A big one for where they hunt.

Playbook 1: how they build.

1. Niche: Pick a niche so small the big players never bother entering it. (Washers.)

2. Integrate backward: Make your own raw material, so you can build any variety on demand.

3. Roll out a commodity: Add a plain, high-volume product on the same setup. Thin margins, but it keeps the factory full and the cash coming.

4. Hit adjecencies: Use that scale and cash to keep finding higher-value niches that run on the same set-up.

5. Deepen the niche you own: Sell your existing customers more. Become their one-stop shop, not just their washer guy.

That’s the whole thing. Simple to write. Took them 25 years to build. Hard to copy.

Now let’s take it apart, one step at a time.

Step 1: Own a niche that the big players aren’t interested in.

The big mills make SS sheets and aren’t interested in making washers. A washer is a nuisance to them, too small, too many varieties. Ratnaveer built its whole company around that nuisance.

Think about BKT - It doesn’t fight MRF on car tyres. It makes thousands of odd-sized off-highway tyres in small batches the giants won’t run. That variety is the moat.

Washers are the same. 2,500+ types, every order small. One washer is a commodity. 2,500+ SKUs, made to spec and delivered on time, is not exactly a commodity.

Long tail capture + Customisation + Small batch size + reliability is the differentiation

Step 2: Make your own steel.

To serve 2,500 varieties, you can’t wait three weeks on a supplier. So Ratnaveer melts and rolls its own sheet. It’s not a big cost edge. What it buys is speed: any size, any grade, on demand. Without this, the catalogue breadth of 2500+ collapses.

Step 3: Add a bulk product for cash.

Finishing sheets are the biggest line by revenue, most commoditised and easy: buy a stainless coil, polish it to a mirror or matte finish, sell it. No moats. and margins are thin. But why?

The boring product quietly pays for all the interesting ones.

Polishing sheet in bulk keeps the factory running near full capacity. This spreads the fixed costs thin. It makes Ratnaveer a large steel buyer, so it gets better prices on the raw material that feeds every other product.

And it throws off steady cash, month after month.

Step 4: Climb the value ladder on the same machines.

Take the cash and the steel skill. Keep adding products that earn more per kilo: circlips, electropolished tubes, solar hooks. Same lines.

Watch the margins:

• FY24: ~8% → FY25: ~10% → FY26: ~11%+

As higher-value products grow, the blended margin starts inching up. So essentially, the commodity gives the platform, the niches lift the margin.

Step 5: Go deeper into customers you already have.

We spoke of big clients and Fortune 500 buyers. Getting approved by them is the hard part. Months of audits and trials. But once you’re in, you’re in.

So Ratnaveer cross-sells and grows its “wallet share.” It has now become the end-to-end fastening solution: washers, circlips, nuts, bolts, one vendor for the whole list.

I really like this move. Operating leverage at the sales channel.

Land with one product, build trust, then use the same setup and the same network to keep pushing adjacent products through a pipe you’ve already paid for.

Pidilite did exactly this. Got in with Fevicol, then sold the same shops M-Seal, Fevikwik, Dr. Fixit, fifty products riding one relationship.

Zooming out.

So that’s the machine. Pick a niche, make your own raw material, fund it with a boring bulk product, climb into higher-value stuff, then sell your customers more. Run it for 25 years, and you get today’s Ratnaveer.

Now management wants to run the same machine one more time. Same idea: find something big, boring and mostly imported, make it in India, sell it cheaper. Except this time the “something” isn’t steel.

It’s copper. And that changes everything.

Because look at what carries over, and what doesn’t.

• The hunting instinct carries over. Spot an import-heavy market, go localise it. That’s pure Ratnaveer.

• The steel skill does not. Copper laminate has nothing to do with washers or sheets. New material, new process, new customers, new competitors.

For 25 years, every new product sat on the same steel base. Copper is the first time they’ve stepped off it. So the real question in this whole article is simple.

Does the playbook still work when you take away the one thing that made it work?

Let’s find out.

The Copper Bet

First, let’s understand what we’re even talking about. CCL is the product they’re going after.

What is CCL, in one breath

Open any gadget you own, and you’ll find a green board inside (remember motherboard?). That’s a printed circuit board (PCB). And every PCB starts life as a plain sheet called copper clad laminate (CCL).

India imports over 90% of it. A ₹5,000 crore market, almost entirely handed to China and Taiwan. That’s a big, boring, import-heavy niche.

Why management thinks they can win

Their case is genuinely decent. It’s a stack of small edges that add up:

Here, local competes well compared to the export players before you even talk price.

Bringing CCL into India means freight, duty, weeks of shipping, and cash stuck in inventory the whole time. For the PCB makers, reliability and those shorter lead times matter. Hence, they would prefer a local partner even if China were a bit cheaper. And here, China won’t even be cheaper.

How?

• A 2% cash subsidy from the government on local supply.

• Ratnaveer claims a ~10% pricing edge over imports anyway, once you add up the wait times, the trapped inventory, the freight and the logistics.

Why now: The political tailwinds also seem to be in their favour. The government wants CCL to be made at home. Obviously it wants to cut off its dependence on a hostile China for such a critical part of the electronics supply chain.

So the support behind local CCL is a hard push, with subsidies attached.

The prize: ₹750 crore of revenue at full scale, at 20% margins, nearly double what washers earn.

How They Actually Plan to Pull Copper Off

A reasonable question would be: if they’ve never made copper laminate, how do they expect to just start? The answer is they’re not building it alone.

• They hired a Chinese expert to hand them the playbook. Ratnaveer partnered with a seasoned Chinese CCL technology vendor, the people who’ve done this a hundred times.

• The partner delivers almost everything, not just machines. The vendor supplies the equipment, installs and commissions it, writes the operating procedures, trains the workforce, and sets up quality control. Ratnaveer is basically buying a turnkey factory.

• The partner sticks around after switch-on. Support is committed for 18 to 24 months post-commissioning, specifically to help Ratnaveer clear customer approvals.

• They’re de-risking the rollout, one line at a time. Line 1 commissions first (late 2026). The remaining lines come only after Line 1 proves itself. So they’re not betting the whole plant at once, they’re testing before they scale.

The honest read: this is a sensible way to enter a business you don’t know.

Buy the expertise → automate the plant → keep the expert around until the experiment works out → scale only after proof.

But notice the flip side, which we’ve flagged before. The capability is rented, not owned. The tech, the process, the training, all of it comes from someone else. It works. It just isn’t a skill that belongs to Ratnaveer yet. The first real test of whether the teaching stuck is Line 1 switching on, on time, and producing laminate customers actually accept.

Risks and What to Watch

No thesis is complete without the other side of the ledger. Here’s what can go wrong, roughly in order of how much it should bother you.

The core business risks

1. Margins are tied to the steel cycle. Most of their raw material is bought at market. And there’s no clean price pass-through to customers. When scrap and nickel prices move, margins move with them.

Eg: FY24 is the proof. Operating margin fell to ~8% when steel prices dropped and inventory got written down.

2. Working capital is stretching. This is the one I’d watch closest.

• Receivables ballooned from ₹65 crore to ₹175 crore.

• Operating cash flow turned negative in FY26.

Profit is growing faster than cash. Forgivable since the company is in a growth spurt, but it’s something that needs to be tracked.

3. They’re raising equity. A ₹330 crore rights issue is on the way. And here’s the detail worth noticing: ₹255 crore of it is for working capital, not the shiny copper plant. So they’re raising money partly to plug the cash the core business is consuming. Watch it.

4. Promoters are trimming, not adding. Promoter holding is down roughly 10% over three years, with a further dip recently around the warrant and share issuance. This has been a point of contention. But as I see it, they’ve now committed to fully subscribing the rights issue, which is reassuring.

5. The copper-specific risks - You’ve heard these already, so just the quick list:

• No domain experience: First time in copper, electronics, laminates. The Chinese partner covers the gap for now.

• No protective wall: No anti-dumping duty on CCL imports YET, unlike washers. That said, given how badly the government wants this made at home, some protection arriving later isn’t unthinkable.

• A crowd: Wipro and a dozen-plus others are building too.

• Raw materials dependency: Copper foil, resin, glass fabric, mostly from abroad.

The timeline keeps moving: Management leads with the ambitious date. The honest ramp is slower.

Final thoughts

Lets strip away the noise and make the whole thing simple.

You’re buying the core business. Copper is a more of an optionality - a lottery ticket stapled on top.

The core is a genuinely good business. It owns a niche nobody big wants to fight for. It has run the same smart playbook for 25 years. Margins are climbing as the mix gets better, and management thinks the existing business can keep growing at ~25% a year. None of that depends on copper, it compounds on its own.

Then there’s copper sitting on top of it.

If it works, we do very well. A whole new business at fat margins, nearly doubling the company, and the stock likely re-rates hard. If it flops, or shows up years late, we still own a good little exporter growing at a healthy clip, bought at a fair 22x.

Heads, we win big. Tails, we do just fine. The downside is protected, because we were never paying for the copper anyway. That’s the whole point. Real upside, with a floor under us.

Here’s a fun fact before we start. Companies that generate more cash than they can even use, that grow without touching debt, and that compound quietly while nobody’s watching, they’re rare. When you find one, riding a wave that lasts decades, not quarters, that’s when Charlie Munger tells you to “go into greed mode.”

I found one. Let me show you the numbers first, company later.

• Revenue up 40 % last year.

• PAT up ~80%. ROCE went from 14% to 36% in 4 years.

• EBITDA margins expanded 600+ basis points.

• Debt? Practically zero.

• All capex funded from internal cash.

• Management is already guiding for bumper revenue growth next year, and then again the year after.

• The PE is around 28x. Against a 5-year PAT CAGR of roughly 50%. Do that division. You get a PEG under 1. That’s the “growth at a reasonable price” zone that value investors dream about and rarely find.

So what does this company make?

Springs for Railways. Wait, wait. Don’t go off. Be with me for a bit, I’ll make it worth your time. Here’s the punchline.

India is spending more on railways than at any point in its history. Rs 2.92 lakh crore this year alone. That money has to go somewhere ( ideally in our pockets).

Most investors will chase the obvious plays. But somewhere in that chain sits a company nobody talks about, that sits underneath every single train on this network, and has quietly built one of the tightest moats in Indian manufacturing.

This article is about finding that company. Not through a screener. By understanding how government money becomes demand, how railways actually work, and where pricing power hides in the supply chain.

Fair warning. This is long (and boring to most). If you want a target price and an Excel model, close this tab immediately and run for your life.

But if you want to actually understand a business before buying it, stick around. This is your chance to awaken Sheldon Cooper - the train enthusiast inside you. Bogies, suspension systems, RDSO approvals( jibberish we’ll understand in a bit). I knew none of this a few weeks ago. If I learned it, you can too.

One thing before we get onboard (pun intended). This stock rides four value migration trends simultaneously. All pushing the same direction. All compounding into one tiny company. That’s rare enough to be worth your weekend.

Print this article. Block distractions. This rewards a slow read, not a scroll.

Ready? This train is about to leave the platform. It’ll be a bumpy ride in places. But by the end, we’ll find a way to smooth it out.

(The answer is better suspension. I’ll see myself out.)

Disclaimer: This article is for educational and informational purposes only. I am not a SEBI registered investment advisor. I hold positions in securities discussed here. Investing involves risk, including loss of principal. Do your own work. Consult a qualified advisor before acting on anything you read.

The Bets I’m Making

Before we get into railways and suspension systems and all the nerdy stuff, let me be upfront about the macro bets embedded in this thesis. If these don’t hold, the thesis crumbles. You should know what you’re signing up for.

Bet 1: BJP stays in power and continues the infra push. Railways critical for this government. Rs 54,000 crore in FY14 to Rs 2.92 lakh crore in FY27. 5x in a decade! I’m betting this direction holds.

Bet 2: Ashwini Vaishnaw keeps delivering. IIT Kanpur. IAS topper. Now Railway Minister. This guy is genuinely cracked. Vande Bharat rollout, 99%+ electrification, record wagon production, all under his watch. I bet my money on him.

Bet 3: Logistics cost reduction stays a national priority. India’s logistics cost as a percentage of GDP is still among the highest globally. We want to shift more and more of logistics from road to rail.

If even 2 of these 3 hold, the demand environment for railway component manufacturers stays strong through 2030 and likely well beyond.

Now, the moment you’ve been waiting for - the company of discussion ( in brief)

Frontier Springs Limited. Kanpur. Incorporated 1981. Has three products (we’ll explain the products in a bit, just know the names for now):

1. Hot-coiled compression springs (the cash-cow),

2. Air spring suspension systems (the growth engine), and

3. Forgings (which we’re going to mostly ignore, the way our political class ignores air pollution).

We’ll get to why those first two matter so much. But first, you need to understand how a train actually works. Humor me for a few minutes. Let’s become rail enthusiasts.

Rail 101. Become a Train Nerd in 5 Minutes.

You need four terms and two concepts. That’s it.

The four things that run on a railway track:

Wagon. They carry goods (Maal-gaadi if you will), not people. Eg: Coal, containers, ore, oil. They have no seats, no AC, no complaints about pantry food.

Coach: These carry passengers. They’re the box you sit in when you take a Rajdhani or a Shatabdi. They come in two designs, and the difference between them is central to this entire thesis. We’ll get there in a second.

Locomotive - The engine. It pulls wagons or coaches. Doesn’t carry passengers or freight itself. Think of it as the horse, everything else is the cart.

Trainset. A self-propelled train where motors are distributed under the coaches themselves. No separate locomotive needed. For example, Vande Bharat is a trainset. This distinction matters. Remember it.

Now the two coach designs you need to know:

ICF (Integral Coach Factory design). The old dabbas. Steel body, designed in the 1950s, max speed around 110 kmph. If you’ve taken a Sleeper class train that rattled your bones and kept you awake, that was probably ICF. These are being retired for LHBs.

LHB (Linke-Hofmann-Busch design). This is the replacement for older dabbas. They’re German-licensed. Made of stainless steel. They’re anti-telescopic (coaches don’t climb over each other in derailments. Now this is a big deal for safety), max speed 160 kmph, significantly quieter.

Indian Railways has committed to converting the entire passenger fleet to LHB or better by 2030. Nearly all new coach production is already LHB.

And Vande Bharat? Not an LHB coach. A different, more advanced thing altogether. LHB coaches need a locomotive to pull them. Vande Bharat is a self-propelled trainset. Here motors are built into the coaches. Think of LHB as the upgrade from a Maruti 800. Vande Bharat is the jump to a sedan.

Both LHB and Vande Bharat use the same suspension architecture though. And that’s where the money is. So let’s talk suspension. we’re getting closer to the main idea. We’re almost there.

How a train stays on its tracks without shaking you to death:

Every coach or wagon sits on two bogies. Each bogie has two suspension stages as shown below.

Primary suspension: It absorbs track-level shocks. It uses coil springs - Wagons, coaches, locos, trainsets. Coil springs are universal here.

Secondary suspension: It controls ride quality, the swaying and bouncing you feel as a passenger.

This is where it gets interesting.

On old ICF coaches: secondary suspension was also coil springs. Cheap, simple, adequate at 110 kmph. But, but, but…

On modern LHB coaches and Vande Bharat: secondary suspension uses air springs.

Why you ask? At 130+ kmph, coil springs can't maintain stable ride height across varying loads (empty train vs. packed train sit at very different heights on coil, vibrate differently). Hence air springs are needed.

TL;DR - No air springs, no 130+ kmph clearance from RDSO( government regulatory and standards authority that must test and approve every component before it can be used by Indian Railways. They’re the gatekeeper). Given the massive progress, upgradation to high speed rails, the probability of going back to primary coils is near zero.

This trend is now structural

The per-coach math:

For 1 LHB coach there are : 16 coil springs + 4 air springs ; One old ICF coach: coil springs only, both stages (primary and secondary).

But, Air springs cost roughly 4x per unit compared to a coil spring. So when an ICF gets replaced by LHB, you’re adding a higher-value component (air) on top of the existing coil demand.

Suspension content per coach goes from ~Rs 2.5 lakh to ~Rs 5 lakh. Same train, same passenger, double the suspension spend.

Warm-up done. Now we get to the real stuff.

Value Migration. The Only Framework That Matters.

This is something I learnt from Raamdeo Agarwal. I personally love this mental model.

Value migration is when profit pools shift from one part of an industry to another because of a structural change. Here are few examples

• Infosys rode the migration of IT services from in-house to outsourced.

• HDFC Bank rode the migration from public sector to private sector banking.

• Bajaj Finance rode the migration from bank lending to NBFC consumer credit.

The pattern is always the same: a structural trend moves money from Point A to Point B, and the company standing at Point B compounds for years

The question is never “is this company good?” Lots of companies are good. The question is: “is there a structural shift moving money toward this company, and will it last?”

Indian Railways right now has four such shifts running simultaneously. All four benefit one tiny company. Let’s walk through them.

Four Migrations, One Company.

Migration 1: Road to Rail.

The equation is simple: More Freight on Tracks => More Wagons => More Springs.

India moves most of its freight by road, only 27% is by rail. That's embarrassingly low. In comparison, the government (through the National Rail Plan) targets 45%.

This is absolutely mandatory for India to compete in manufacturing. Data has shown massive political push from the government in this direction ( Can ignore this: The logistics cost as % of GDP has reduced from 14% to 9% now. We’re rapidly moving towardw 7% which is the global average. For this, goods have to move through rail and not roads)

The infrastructure is being laid - Two Dedicated Freight Corridors (basically tracks/ corridors for freight trains dedicatedly) are largely operational. A third was announced this budget.

Hence, this push will continue. For us, this means certainty ( aka future demand)

Now the number that matters - Wagon production. Just see the massive difference.

For FY14-24 average: we made ~16,000/year. Whereas for FY25, we produced a whopping 41,929 wagnons. A 2.6x jump! Orders placed for ~90,000 wagons over three years. Every single wagon runs entirely on coil springs. Now to that, add replacement demand from an installed fleet of 300,000+ wagons (springs replaced every 5-6 years), and this is the single largest volume driver for coil spring demand in India.

Frontier holds ~20-22% share here. This is the volume play.

Migration 2: ICF to LHB. Suspension Value Doubles.

This is the big one. Old ICF coaches are being retired and replaced by modern LHB coaches. Indian Railways has mandated full conversion by 2030. You already know what this means. This decision doubles the suspension content per coach from ~Rs 2.5 lakh to ~Rs 5 lakh.

The actual production ramp also testifies to progress on ground ( 2004-14, India built roughly 2,300 LHB coaches total. The current run rate is 7,000+ per year. Twice every year of what we produced in a decade. We’re gunning for 10,000 per year). And then there’s the replacement cycle on the existing LHB fleet itself. Springs on over 42,600 LHB coaches that have been built cumulatively need replacing every 5-6 years. That alone generates recurring demand roughly matching new-build volumes each year. Now for the punchline.

Frontier holds ~50% share in passenger and locomotive coil springs.

There are only 2-3 approved suppliers for this. Also, in air springs, they hold ~40% market share (Nov 2025 concall) among just 3 approved suppliers nationally. We’ll get into why so few companies can play here when we discuss moats.

Migration 3: Imported to Indigenous.

The government mandated local content requirements for public procurement in 2017. Suppliers need 50%+ domestic content to qualify as Class-I, and in categories with sufficient domestic capacity, only Class-I can bid. Railways follow this.

The implication for springs is massive.

It narrows competition to domestic manufacturers and eliminates the Chinese threat altogether ( No Chinese competition, phew!). No import-heavy player can even submit a tender. Whereas Frontier manufactures ~70% of its air springs in-house, putting them comfortably in Class-I territory

Result: there are exactly three RDSO-approved air spring suppliers in India today. RDSO approvals are not easy to get. This makes for a government-mandated moat. Those are the best kind.

Migration 4: Conventional to Vande Bharat.

Now, something we all know about Vande Bharat is the fanfare. The PM flags off trains personally. The amount of political capital it has generated implies that this is a program that the government will only double down on. Let’s see the numbers on the same.

Current fleet: is ~80-150 trainsets ( sets of coaches). The target is a whopping 800 by 2030, 4,500 by 2047. Now even if you take conservative numbers and shoddy execution, you land at 500, which is a 6x jump! On the ground, multiple manufacturing contracts are already signed and in production.

Each trainset needs coil springs and air springs (same architecture as LHB that we’ve discussed). Frontier is a confirmed, named Vande Bharat supplier with zero failures across 15,000+ air springs since 2022 (only 3 RDSO certified suppliers here. Competition working with Chinese imports have seen failure rates and hence the RDSO is averse to them). To ensure the quality of Vande Bharat, the tenders don’t go to the lowest cost bidders but to those who ensure high quality. Frontier wins here due to its proven track record across decades.

Migration 5: Track to Trains.

The one nobody’s watching, and the one that answers “what happens after 2030?”

Please read this carefully.

Right now, ~66% of railway capex goes to track infrastructure, and ~29% to rolling stock ( coaches, wagons, etc.). But, post-2031, this equation flips. The Rolling stock capex share rises to ~39%.

The logic is simple. You build infrastructure to 2050-scale capacity now (because it takes years), then spend two decades filling it with trains. So train capex lags infra capex. After 2030, train capex increases. These Trains need springs. Springs that wear out every 5-8 years.

Even if the overall railway budget flattens (it hasn't, +10% this year), the rolling stock slice within it is structurally growing. The pie that matters for a spring company is getting bigger inside a pie that's itself still expanding. Five migrations - All simultaneous.

• Wagon volume are growing.

• Suspension content per coach is doubling.

• Import competition is locked out by policy.

• Vande Bharat adding politically bulletproof demand.

• And the capex mix is shifting from track to trains for two decades.

All compounding into the same handful of approved suppliers. Now, let’s talk about what makes Frontier specifically hard to replace.

The Rs 1,000 Cr Question. What's Next After Air Springs

From what I understand, the game here is to keep finding smaller niches of demand from Indian Railways and become a market leader. The management has a proven track record of doing this.

Here’s what I think is actually going on. The game isn’t “sell springs to Indian Railways.” It keeps finding smaller, technically hard niches within the railways’ demand, and becomes the leader in each one before anyone else shows up.

Management has been explicit about this. Their own words: keep finding products that are technically hard to make, where competition is thin, and margins are fat. Reminds me of Peter Thiel’s philosophy of “Competition is for losers”

I love this. They’re selective. No low-margin products. Combined with zero appetite for leveraged growth, this is a high-quality management signal.

And they’ve actually done it. Leaf springs to coil springs to forgings to air springs. Each jump was a real technology and qualification barrier, not just added capacity. Air springs is the proof point. They went from 0% to ~40% market share against a 20-year incumbent.

The next bet is FIBA (Failure Indication and Brake Application). FIBA is a safety system that auto-triggers braking if an air spring fails mid-journey. For this, the RDSO approval is targeted mid-year (could be any time now); once this happens, the stock will re-rate as analysts change the terminal value of the business) This is a Rs 100 crore market. And the competition is limited to very few players, all currently importing. Frontier’s built it entirely in-house and filed for a patent. If this plays out as air springs did, we get another free lunch. Though this is very, very likely to play out, my thesis is built without this optionality. We’ll add this to our margin of safety!

A second optionality sits quietly alongside it: metro projects. Indirect supply linkages to global OEMs like Alstom, riding the same suspension technology into an entirely different rolling-stock category. Though this is Small today. Again, we’ll ignore it and add it to our margin of safety ( Which by now is really, really fat)

The Moats. What Makes This Hard to Replicate.

Let's get one thing straight. Frontier makes springs. That sounds like a commodity. A lot of it is. But there are five specific barriers stacked on top of each other that make this business genuinely hard to enter. Not one barrier. Five. And they're sequential.

Barrier 1: RDSO Approval. The 4-5 Year Clock.

You cannot sell a single spring to Indian Railways without RDSO qualification. No exceptions. The process involves design validation, prototype testing, field trials on actual running trains, and periodic audits. This takes 4-5 years minimum. There is no way to accelerate it. No amount of capital, connections, or technical capability lets you skip the clock.

And the approval is spec-level, not category-level. Being approved for wagon springs doesn't mean you can supply coach springs. Being approved for 120 kN air springs doesn't mean you can supply 160 kN ( essentially a different variety of air springs). Each product variant needs its own multi-year qualification. Frontier has been RDSO-approved across multiple specs for over 30 years.

Our train has reached a stop. Everything above is free, and I think it’s more useful than most paid reports out there. What follows (moats, process barriers, air spring deep-dive, business quality, governance, and the full thesis) is behind the paywall. If you found value, a subscription costs less than a bad stock tip and funds hundreds of hours of research on companies nobody else covers.

Read more

• Gravita India: up >1200% in five years

• Pondy Oxides: up >600%

• Tinna Rubber And Infrastructure Ltd : >3000%

This is telling of an important story. That the market beginning to price in something structural: recycling is no longer a niche ESG story. It is becoming one of the most durable business themes of this decade.

And most retail investors have barely scratched the surface of how to analyse these companies.

I spent weeks going deep on this. I read through Indian players like Gravita India, Pondy Oxides, and Tinna Rubber. I studied international counterparts like Genan (Denmark), Liberty Tire (US), and Black Bear Carbon (Netherlands) to understand what world-class looks like in this space.

By the end of this piece, you will know exactly what makes a recycling business great, what metrics actually matter, and how to spot one before the market does. No jargon. No fluff. Let’s go. But first, lets look at the tailwinds.

Why This Theme is Inevitable

Four forces are converging right now, all pointing in the same direction:

1. The waste problem is getting un-manageable: The world now creates 2.1 billion tonnes of municipal waste every year. Let that number sink in! We have accumulated waste of centuries and now we’re adding to it at an unprecendented pace ( 260 kg for every human being on Earth per year). When even the great Gautam Gambhir couldn’t solve ghazipur garbage mountain problem (at it’s peak it was taller than Qutub Minar) - the fate of humanity is really bleak.

2. The raw materials to make new things are running out - New mines are harder to find and more expensive to build. And we also know that rare earth metals are effectively a Chinese monopoly.

   Now in such cases, resilliance for supply chains becomes super critical. At some point, the most reliable mine is not on the other side of the world. It is the waste pile at home.

3. Governments are mandating it, everywhere: There’s a recycling framework known as EPR - Extended Producer Responsibility. It require manufacturers to take back and recycle what they produce. Europe has it. The US has it. China has it and now vishwaguru too. This is providing filip to recycling ( More on this later)

4. Deglobalisation is making domestic waste a strategic asset: Supply chains are becoming more and more politicised. They’re being used as an intrument of geo-politics ( politicisation of trade if you will). Every nation is wanting to secure supply chains at home. When you don’t have mines, you turn to heaps of waste and start digging for the gold in it ( quite literally too).

Now, I want to talk about some important mental models.

Recycling is the Opposite of Manufacturing ( Read this carefully, this is foundation of what we’ll discuss later)

In manufacturing, you take multiple raw materials and combine them into one product. For an example, think of a car. Steel, rubber, glass, electronics all go in, one car comes out.

Recycling is the exact ulta:

One raw material, say a used tyre, goes in. And Multiple products come out: crumb rubber, steel wire, carbon black, pyrolysis oil, Micronised Rubber Powder ( don’t fret over the names). The entire analytical framework flips.

This inversion changes what you look for in a recycling business completely.

At the very Core, the recycling business is: Procure → Refine → Sell. The earn the spread between what they pay for waste and what they realise from processed outputs. Wider spread the spread, better business. Simple.

But each of those three steps has hidden complexity that separates great recyclers from mediocre ones. Let’s go through them, very throughly. This is where we’ll understand what separates the average recyclers from the very great ones.

I’ll also keep contrasting recycling business with a manufacturing business. So that we don’t apply manufacturing frameworks blindly while analysing a recycler. Lets not do that mistake.

Step 1: Procurement - Where the Game is Won or Lost

Here’s what makes recycling procurement fundamentally different from normal manufacturing.

If you’re a steel plant, you call a supplier, get a quote, place an order. The material arrives at exact specification you need. Done. The market is 1. liquid, 2. prices are discovered, 3. logistics are sorted. These are signs of a mature market.

In recycling, none of that applies. Markets are not mature. Lets understand with the example of used tyres ( AKA end of life tyres - ELT). The market has the following characteristics:

• Fragmented: Waste is collected from thousands of dealers, fleet operators, garages across the country. Not a huge vendor that just supplies whatever you need.

• Lumpy : supply peaks and troughs with scrappage cycles. It’s not a steady tap you can turn on. After hanging out in groups of recycler folks (Yes, you read that right), I understood this is their biggest pain-point. Un-reliable and lumpy supply of scrap.

• Heterogeneous : a truck tyre and a car tyre have completely different rubber content, steel content, and compound formulations.

So how does a good recycler win on procurement? The Moats:

• Diversification of procurement: Tinna has six processing facilities across India: Panipat, Wada, Haldia, Gummidipoondi, and others. Each plant is a procurement catchment. Why? If supply dries up in one region, other plants compensate(if one engine shuts, others start firing). The benefits are 3 fold:

  • This ensures high capacity utilisation ( which is not as easy to increase as in the case of manufacturing where you can just put capex, order raw materials and boom. Here the procurement constraint is very very real).

  • Anti-fragility is baked in: (I was super psyched when I realised this and I hope you’ll be psyched too). We were in an era of extreme efficiency till now. Now the tides have turned. In an era of extreme uncertainity (Thanks to Mr Trump and Xi and Putin and a few others). Its the business that’s more resiliant and anti-fragile to shocks that will win. Businesses will increasingly have to increase focus on resilliance by even compromising on efficiency. This is baked into a good recycler by design.

  • Scale locks Import relationships: In india, waste collection (as discussed) is fragmented and unorganised. Hence, a bigger player that has setup connects can import “good quality waste” reliably ( ironic isn’t it?) . So they naturally import. Exporters aren’t accessible to everyone. Here scale compounds and improves the moat.

• Sometimes you’re paid to procure. In some markets and categories, waste generators pay you to take waste off their hands. That’s negative procurement cost - you’re essentially being paid before you’ve processed anything.

• Volume and relationship power. When you’re buying thousands of tonnes a month, dealers come to you. You negotiate better terms. In comparison, Small players struggle for feedstock ( hence can’t match capacity utilisation levels which creates real competetive disadvantages).

• Strategic plant location. “Plants near ports” allow import of Waste eg: ELT (End of life Tyres if you’ve forgotten already) from markets where supply is surplus.

  • In case of Tinna, they went one step ahead. They created a plant in Oman. Nows the waste undergoes one layer of processessing. The impact is this - Logistics cost is optimised since the amount of waste carried is reduced due to 1 layer of processing.

Step 2: Refining - The Learning Curve Nobody Can Shortcut

Here we answer the million dollar question - if recycling is such a good business, why doesn’t a well-capitalised new entrant just come in and disrupt it?

1. Because unlike regular manufacturing, here, the process knowledge can’t be bought off a shelf.

In normal manufacturing, your raw material is standardised. You build a plant to those specs, run SOP, done.

In recycling, your feedstock is never standardised. Going back to our tyre example, every batch of tyres is different. Different age, different brand, different compound, different steel content.

Your process has to adapt constantly. The knowledge of how to handle that variation ( what temperature to run, how to calibrate the magnetic separators, what mesh output you’ll get from truck tyres vs car tyres ) is learned over years of actual processing.

This is the moat nobody talks about. Tinna has been doing this since 1977. That’s nearly 50 years of process learning. A new entrant with ten times the capital cannot replicate that on day one.

Here also procurement brilliance matters. Good procurement ensures CONSISTENT HIGH-QUALITY feedstock, which ensures good downstream product. This is very very hard to crack.

2. Recovery Rate - the single most important operational metric.

Recovery rate = what percentage of value in your input waste you actually capture as saleable output.

From 100 tonnes of end-of-life tyres, Tinna recovers roughly:

• 45 tonnes of crumb rubber

• 15 tonnes of steel wire

• Small amounts of textile fibre

Every additional tonne and every additional material recovered from the same input is pure margin.

Think about it, procurement cost is already sunk. There's no additional buying, Overtime, you start recovering everyt single think from the waste material. In tyres you start with Rubber and then progress to even the fibres that go into it.

Every incremental material recovered is a by-product sold and it contributes to the margins. So time and scale compounds, and these are difficult barriers to breach once a player is established.

Here’s the flywheel: Better process capability → higher recovery rate → better margins on identical feedstock → business gets better without growing.

3. Output consistency is mandatory for sticky customers.

In recycling, since input is inconsistent, there’s a risk of output also being inconsistent too. Here, a company buying raw material from re-cyclers cannot tolerate batch-to-batch variation. Eg: If MRF procures from Tinna - If the rubber size varies, their formulations go for a toss, and tyre performance suffers. Consistent output requires tight feedstock control, robust SOPs, and process discipline - all of which compound over time.

4. The VAP (Value Added Product )journey: moving up the value chain.

Early stage recyclers sell whatever comes out of the machine. They start with commodity grade, low realisation, fully price-exposed output. But, as process capability matures, the same input waste starts yielding higher-grade outputs.

For example - A lead recycler moves from scrap lead to refined lead to speciality alloys. A tyre recycler moves from coarse crumb rubber to micronised powder to recovered carbon black. Same waste in, completely different economics out.

Each step up means higher realisation on the same tonne of input. This value-add trajectory takes years, sometimes decades, to walk. But when a recycler crosses a quality threshold that unlocks a new customer segment, the business re-rates sharply ( Yes, P/E re-rates sharlpy and we’ve seen examples of it)

My Article on Shilchar tech

Step 3: Selling - Why Stickiness Builds Naturally

1. Like precision engineering companies - Certification creates moat.

Tyre recycling outputs like Crumb Rubber and Micronised Rubber Powder (MRP) have certification barriers. Crumb rubber need a BIS certification. For MRP - A supplier to a tyre OEM goes through 18-36 months of compound trials before getting qualified. Once you’re in, the customer doesn’t switch

2. Diversified end-markets provide resilience - Anti-cyclicality is in the nature.

Since there are many by-products, end customers are automatically diversified. Eg: Tinna sells into roads, tyres, sports surfaces, auto parts, abrasives. When government road spending slows, tyre demand holds. When construction is weak, export demand might compensate. No single industry concentration means no single point of cyclical failure.

3. Relationships are hard to replicate.

B2B rubber compounding is a relationship business. Procurement managers who’ve worked with a reliable supplier for years don’t switch for a 5% price advantage. They switch when there’s a failure. Consistent delivery is the best sales strategy in this industry.

A Few Things Worth Knowing Before You Analyse Any Recycler

1. EPR credits are a new, high-quality revenue line.

   India’s EPR framework allows recyclers to earn tradeable certificates for processing ELT (Meaning if the manufacturer of goods can’t recycle themselves, they purchase certificates from those who do). Tinna earns ₹20-30 Cr in EPR credits in at near-100% margin. This grows automatically as tyre production volumes grow and compliance tightens.

2. Don’t use EBITDA margin as your primary metric.

   Raw material prices - waste tyre procurement costs - fluctuate significantly. An EBITDA margin of 15% in one year and 18% the next might just reflect input price movement, not operational improvement.

   The right metric is EBITDA per tonne processed. This strips out input price noise and shows you whether the business is actually getting operationally better.

3. Scale deepens the moat - this is a business that gets better as it gets bigger.

   More volume → better procurement terms → more capital for process improvement → higher recovery rates → better output quality → stickier customers → more volume. The flywheel is real. but it takes years to get spinning fast enough to be self-sustaining.

4. Hedging and pass-through matter.

   Check whether the company can pass through input cost increases to customers or whether they absorb them.

Read more

GCC is one of the megatrends in India and I’ve wanted to play the GCC theme for a while.

The most obvious way was coworking spaces. So, I took a position in one of the promising players. Now, I’m almost Buffett with one small difference. The stock that I purchased goes down 30% instead of up, just that small difference. The stock that I purchased had an ambitious promoter. More than I anticipated. No marks for guessing the name.

So I kept looking.

Eventually, I found a proxy.

Simple business. Easy to understand. Good growth. High ROCE. Reasonable valuation.

And this time the promoter doesn’t appear to be a chor.

Before asking what will drive the stock price of this company, let’s see what the company drives. Employees! Yes, it drives your employees to work. Literally.

First, what does the company actually do?

The company is ECOS India Mobility. Most descriptions make it sound far more complicated than it is.

ECOS does two things:

1. Chauffeured Car Rentals (CCR): Following are the use cases

   1. Executive visiting Mumbai for the day? Needs a car.

   2. Foreign delegates arriving at the airport? Need cars.

   3. Company conference requiring 20 vehicles for three days? Need cars.

ECOS arranges them. Fixed tariffs. Trained chauffeur. No surge pricing. Available across 130+ cities

2. Employee Transportation Services (ETS)

   This is the more interesting business.

   The cab that picks up your IT employees and dozens of other corporate slaves every morning; The vehicle that drops a night-shift employee home at 3am; The daily movement of thousands of employees between home and office, every single day. That’s ETS.

The interesting thing is that ECOS doesn’t own most of the cars. It aggregates thousands of vendor-owned vehicles, adds technology, compliance, and operational oversight on top, and takes a cut. Uber model if you will.

Asset-light. No large fleet on the balance sheet. No capital constantly being poured into buying vehicles.

Just demand aggregation and execution.

Simple enough.

But to understand why the business exists, you need to understand who they’re solving for.

Put Yourself in the HR Head's Chair

You’re the HR head of a 2,000-person GCC in Bangalore supporting US operations. Half your workforce is on night shifts. Many are women.

Your problems:

• Karnataka law mandates cab service for women after 8pm. Non-negotiable.

• Global HQ wants quarterly vendor governance reports

• Drivers need background verification (you obviously don’t want drivers with crimilar records driving your employees daily), GPS tracking (for compliance and risk management), and documented SLAs

• Employees are spread across the entire city and you have zero bandwidth to manage 15 local operators each having their own area of operations.

One bad incident and it’s not just a compliance failure. It’s a headline that damages the company’s reputation. No HR wants to take this career risk just to save a few rupees ( which anyway they’re not paying from the pocket)

So you make one call. One contract. One invoice. One throat to choke.

That’s ECOS.

Why Is Growth Happening Now?

India has had IT companies for decades. Corporate transportation isn’t new. So why is the business suddenly interesting?

Three things happening simultaneously.

1. GCCs are exploding.

India had around 1,000 GCCs in 2020. Today that number is closer to 1,500. Government projections suggest roughly 2,400 by 2030.

Every new GCC needs employee transportation from day one. A GCC supporting US operations cannot function without night-shift logistics. As GCCs grow, transportation demand grows alongside them. Automatically.

2. Regulations help.

Many GCC-heavy states require safe transportation arrangements for women employees working late-night shifts. Karnataka, Maharashtra, Telangana.

For companies operating round the clock, transportation isn’t a a good to have. It’s a must have. And once you start providing commute options for women, you can’t discriminate against men, so you’ve to provide commute for all

This increases demand

3. The organized market is taking share.

Most of the industry remains fragmented. Local operators. Small fleet owners. Regional players with 20 cars and a WhatsApp group.

That works until a large multinational starts asking questions.

Who verified the driver? Where’s the GPS record? Where’s the audit trail? Where’s the incident log?

The larger companies get, the less tolerance they have for informal operators. That naturally pushes business toward organized platforms.

So company growth becomes: GCC growth + shift from unorganized to organized. Both engines running at the same time.

But none of that is the actual thesis. The thesis starts here.

Why This Industry Wants to Consolidate

When I first looked at this business I thought it was transportation. The more I studied it, the more it started resembling a route-density game. And route-density businesses often become winner-take-most markets (Read this twice. Think of Food delivery, e-commerce).

Here’s why.

Coverage wins contracts: Imagine a company has employees spread across Bangalore. Whitefield. Hebbal. Koramangala. Bannerghatta Road. Electronic City. Everywhere.

A local operator with 20 cars in Whitefield cannot serve that company. Even if he’s excellent. His network simply isn’t broad enough. The company gravitates toward whoever can serve the entire city. The player with the widest network wins. So the sector naturally consolidates over time - by design!

Density creates better economics.

Now the interesting part.

Suppose ECOS has 10 corporate clients in Bangalore. Employees from all 10 companies spread across the city. That gives ECOS more pickup points. More routes. More optimisation opportunities. More employees packed into fewer kilometres.

The economics improve. The flywheel starts spinning.

Why new entrants struggle.

Now imagine you’re a new competitor entering Bangalore. You build the same technology. You recruit vendors. You hire operations teams. You do everything right.

One problem remains. You don’t have enough pickup density. Your routes are thinner. Your costs are higher. Your economics are worse.

The leader isn’t winning because they have better software. The leader is winning because they got there first and density compounds. Every day that passes makes the gap harder to close.

Why small operators become suppliers, not competitors.

Large enterprises care about GPS tracking, audit trails, compliance reporting, driver verification, SLA monitoring (All things we’ve already talked about).

All of this costs money. Technology costs money. Operations teams cost money. Processes cost money. All of these are fixed costs that only make on scale for the unit economics to make sense. Most local operators don’t and can’t invest here. Instead, many become suppliers to aggregators. They provide vehicles. The aggregator owns the customer relationship.

The local operator isn’t really competing with ECOS. He’s supplying ECOS. And ECOS dictates the price.

Free marketing baked into the model.

Close your eyes. Picture a Swiggy logo. You almost certainly imagined the delivery bag on the bike. Not the app. Not the website. The physical presence on the road.

ECOS has the same thing. Every branded ECOS car driving through Whitefield at 7am is a free impression on every HR head in every building it passes. The more cars in a city, the more visibility, the more inbound interest.

Scale creates marketing. Marketing creates more scale. Zero incremental spend. Most companies pay to acquire customers. ECOS’s product is on every street corner.

Why Clients Don’t Switch

The obvious question. If transportation is largely outsourced anyway, why doesn’t a company simply change vendors every year?

Because the decision isn’t purely about price. Imagine being the HR head. You save 8% on transportation costs. Six months later a serious safety incident occurs.

Now explain to the board why changing vendors was a good idea. Nobody wants that conversation. Nobody wants that career risk. The downside is massive. The upside is tiny.

So companies stick with providers they trust. The numbers support this. More than half of ECOS’s revenue comes from clients that have been with the company for over five years. Active client base grew 34% year on year in the most recent quarter. No meaningful client has been lost in recent memory.

The Part the Market May Be Missing

Margins have been under pressure. Revenue growth has been healthy. Client additions have been healthy. Trip growth has been healthy. Yet profitability growth hasn’t followed proportionately.

At first glance that looks concerning. There’s another way to view it.

When ECOS wins a large client, costs arrive first. New routes. New drivers. New vendors. New operations teams. Everything gets set up before the network reaches efficient utilisation.

Expenses hit immediately. Benefits arrive gradually: Economically it behaves like growth capex. Accounting treats it as an expense. Generally capex are capitalised, but not here.

As route density improves and wallet share expands inside the client, operating leverage starts showing up. The market can already see the revenue growth. The more important question is whether margins recover as recent client cohorts mature.

Management guides 13-15% EBITDA at steady state. At Rs.1,000-1,200 crore revenue, roughly 18-24 months away at current growth rates, they believe operating leverage kicks in meaningfully. PAT could approach Rs.100-110 crore from Rs.56 crore today.

The margin recovery is the thesis. Everything else is already visible.

The Virus Strategy

One thing I particularly like. ECOS rarely wins everything on day one.

It gets a foot in the door. One office. Then another city. Then executive travel. Then airport transfers. Then annual offsites. Then a national mandate.

The relationship spreads through the organization. Almost like an virus.

Many large companies still use 10-20 transport vendors. Multiple contracts, multiple SLAs, multiple headaches. One vendor. One contract. One dashboard. One point of accountability.

The direction of travel is obvious. ECOS already has 8 clients where they replaced all other vendors and became the sole provider. That number is growing.

And It Survived COVID

Revenue collapsed from Rs.402 crore in FY20 to Rs.100 crore in FY21.

Asset-light model meant no stranded fleet. No fixed lease obligations on thousands of vehicles. Vendor relationships paused rather than terminated. Cost structure compressed with revenue.

By FY23 back to Rs.423 crore. By FY25 Rs.654 crore. It didn’t just recover. It came back stronger. Every unorganized operator that disappeared during COVID made ECOS look more reliable by comparison.

Adversity accelerated the exact shift that drives their growth. A crisis that should have hurt them ended up strengthening their position.

Antifragile by design. I really like seeing how companies have survived shocks. The more shocks the company survives the more mechanisms it has created in its DNA to survive further shocks ( Learnt this from the book - what I learnt about investing from Darwin)

The Investor Angle

The market already knows GCCs are growing. That’s not where the edge is.

The more interesting question is whether route density eventually creates a stronger business than investors currently appreciate.

If GCC growth continues, if the organized market keeps taking share, and if recent client cohorts mature as expected, earnings can grow materially faster than revenue.

The route-density flywheel. The compliance layer. The sticky contracts. The amoeba expansion. The free marketing on every street corner.

That’s the real bet.

Share vaibhav tambade

Disclaimer

This article is for informational and educational purposes only. Nothing written here constitutes investment advice, a buy or sell recommendation, or an offer to transact in any security.

I may or may not hold positions in securities mentioned. Do your own research before making any investment decision.

I am not a SEBI registered investment advisor. This is not research under SEBI (Research Analysts) Regulations, 2014.

Risks Forensic Analysis Report

Read more

Generally, I see analysts take one glance at the P/E ratio of these new-age companies and reject them. This article is about adding nuance to the analysis.

You’re currently using an industrial-age ruler to measure digital-age economics. The P&L is fiction. The balance sheet hides the real assets. And your DCF measures the wrong things entirely.

Internet businesses are built differently. They build moats that don’t show on balance sheets. They create value through compounding loops that linear models can’t capture. And founders deceive you with vanity metrics that look impressive but mean nothing.

Here are the mental models you actually need! Let’s get into it. But 1st I’d like to make an appeal. It takes a lot to write these thoroughly researched articl…

Read more

Generally, I see analysts take one glance at the P/E ratio of these new-age companies and reject them. This article is about adding nuance to the analysis.

You're currently using an industrial-age ruler to measure digital-age economics. The P&L is fiction. The balance sheet hides the real assets. And your DCF measures the wrong things entirely.

Internet businesses are built differently. They build moats that don’t show on balance sheets. They create value through compounding loops that linear models can’t capture. And founders deceive you with vanity metrics that look impressive but mean nothing.

Here are the mental models you actually need! Let’s get into it. But 1st I’d like to make an appeal. It takes a lot to write these thoroughly researched articles. If you derive any value from it (only if), please share, restack the article, and consider subscribing or moving to a paid version. Back to the frameworks….

Framework 1: CAC/LTV Must Exceed 3x (And Most Calculate It Wrong)

Every internet business lives or dies on a metric called CAC to LTV ratio. It is CAC (Customer Acquisition Cost: cost to get one paying customer) divided by LTV (Lifetime Value: total profit that customer will generate over their entire relationship with you).

Thumb rule: A ratio below 3x, you’re destroying value. Above 5x, you’re building a compounding machine.

But most analysts calculate both incorrectly.

LTV is not revenue per customer. It’s present value of all future gross profit after discounting for churn. A Swiggy user ordering twice weekly for three years generates different economics than a BigBasket user ordering monthly. Same category, different LTV.

CAC is not marketing spend divided by users. It’s fully loaded, cost per incremental paying customer: ads, discounts, referrals, sales salaries, and platform fees. For lending fintechs, include rejected loan applications. For B2B SaaS (software as a service: subscription software sold to businesses), include pre-sales engineering.

Framework 2: The P&L is Theater (Front-Loading Trap)

Accounting was designed for manufacturing in 1920. Buy steel, make cars, sell them in the same quarter. Costs and revenues match.

For internet businesses, it’s fiction.

Quick commerce example: Company spends to acquire customers today (expense hits P&L immediately). Those customers order for years generating cumulative gross profit (recognized slowly over time). P&L shows losses today. Economic reality is building an asset. But this asset is not found in the balance sheet!

When contribution margins (revenue minus all “variable” costs: delivery, discounts, payment processing) turn positive on existing customers while new customer acquisition continues, aggregate P&L still shows losses even though newer cohorts are profitable.

Product development trap: Tech companies employ thousands of engineers building apps, infrastructure, and algorithms. That entire cost hits P&L as employee expense. But they’re creating assets. If this were a factory, you’d capitalize it (record as an asset on the balance sheet) and depreciate over 20 years.

A company could cut engineering headcount dramatically. App keeps working. Revenue doesn’t drop proportionally. P&L magically improves.

Framework 3: Digital CAC vs Rent (One-Time vs Forever)

Lets do a quick retain vs E-commerce comparison.

Retail store: Pays monthly rent for visibility and foot traffic. Next month, rent bill again. Forever. Rent doubles as marketing cost as rent in high-visibility areas ensures customers see the shop ( marketing). This is recurring cost to acquire a customer.

Digital business: Spends on customer acquisition Year 1. Same customers transact for years. Zero incremental CAC after Year 1.

Over 10 years: Retail pays rent continuously. Digital pays the acquisition cost once, then only variable costs.

Retail rents customer access forever. Digital buys it once. The P&L treats both as operating expenses. Your valuation model shouldn’t.

Framework 4: Winner-Take-All (Second Place Means Nothing)

Most internet markets don’t end with five players. They end with one or two dominant players and maybe one challenger. Everyone else dies ( or sells their start-up to someone else and starts another one OR becomes a Linkedin Guru).

Three forces drive this:

1. Unit economics only work at scale. Eg: Ride-hailing needs rider-driver density per micro-geography; If its, below threshold, wait times spike and utilization crashes.

2. VCs fund #1 and #2, never #3 (since there is no clear path to winning, and capital requirement too high).

3. Users keep 1-2 apps per category (third gets deleted). Everyone has memory issues

The niche exception: Rapido didn’t compete head-on with Uber/Ola. Found their blind spot: bikes/autos at lower prices in tier 2/3 cities. Different customers, different economics. By the time they had scale, they weren’t “third player in ride-hailing,” they were “leader in two-wheeler taxi.” Different market.

In winner-take-all markets, #1 captures 50-70% of value, #2 gets 20-30%, everyone else splits 0-10%.

Framework 5: Network Effects (The Only Moat)

Physical businesses have moats from assets, brands, patents. Internet businesses have network effects: the product gets more valuable to each user as more people use it. And unlike physical assets, this strengthens over time instead of depreciating.

Three types: Direct (WhatsApp: more users = more people to message). Indirect (Swiggy: more restaurants attract more customers; more customers attract more restaurants). Data (Google Search: more queries improve results; better results attract more users; competitors can’t replicate decades of accumulated data).

The cold start problem: You raise capital for a social network competitor with better features. You tell users, “Join our network.” They ask, “Are my friends here?” You say “not yet.” They say, “I’ll wait.” Everyone waits forever. Stuck in equilibrium where no one moves first.

Network effects create binary outcomes. Either you have them or you don’t. When you find real ones, “insane” valuations often prove conservative.

Framework 6: Marketplace Flywheels (Hardest to Build, Hardest to Kill)

Marketplaces need supply and demand simultaneously. Too much supply without demand, suppliers leave. Too much demand without supply, customers leave.

But once the flywheel (a self-reinforcing cycle where each turn makes the next turn easier) spins, it’s nearly impossible to stop.

The mechanic: Onboard initial supply. Customers transact. Supply sees orders, tells others. More supply joins. Better selection attracts more customers. Eventually hit liquidity (enough supply and demand that transactions happen easily without the platform actively pushing). Self-sustaining.

Competitor enters. Supply: “Why list? You have no customers.” Customers: “Why use? You have limited supply.” Flywheel works in reverse.

Marketplaces are binary. Either flywheel spins (valuable) or doesn’t (vulnerable).

Leave a comment

Framework 7: Platform Optionality (Free Call Options)

Internet platforms can test multiple products at near-zero marginal cost. Each new product is like a free call option on a large market.

Traditional business launching new product: Hundreds of crores, multiple years, low success rate. Platform launching new product: Small team, few months, test multiple, kill failures fast.

With large engaged user base, high trust, and existing infrastructure, each product launch is low-cost test with high payoff if it works.

Amazon’s AWS (Amazon Web Services: cloud computing infrastructure) story: Built compute infrastructure for e-commerce. Had idle capacity. Engineers asked “can we rent this?” Zero marginal cost to test. Small initial investment. Today AWS generates tens of billions in revenue with industry-leading margins, worth more than retail business. Emerged from experiment.

Valuation: For platforms with large engaged bases, high trust domains, high frequency, optionality value can exceed core business by 2-3x. Analysts valuing only current revenue miss the portfolio of options on adjacent massive markets.

Framework 8: Some Metrics That Actually Matter (One Line Each)

GMV (Gross Merchandise Value): Total transaction value through platform before discounts/returns. Vanity metric, ignore it.

NMV (Net Merchandise Value): GMV minus discounts, returns, cancellations. What customers actually paid for and kept. Better, but not revenue.

Revenue: What the company actually earns (take rate(commission) × NMV for marketplaces, or selling price for 1P). This is what hits P&L.

Take Rate: Revenue as % of NMV. Commission percentage platform charges. Higher = more pricing power.

Contribution Margin: Revenue minus variable costs (COGS + delivery + payment gateway + discounts). Positive = unit economics work. Negative = model broken.

Gross Margin: Revenue minus COGS (cost of goods sold: direct costs of making or sourcing the product) only. Target 20%+ for inventory models, 15%+ take rate for marketplaces.

AOV (Average Order Value): Revenue per transaction. Higher = better unit economics (fixed costs spread over a larger base).

Retention Rate: % of customers active this period who were active last period. Target 60%+ consumer, 90%+ B2B SaaS. Low retention kills everything.

Churn Rate: % of customers who leave each period. Inverse of retention. Target under 5% monthly consumer, under 2% B2B SaaS.

CAC (Customer Acquisition Cost): Fully loaded cost per incremental paying customer, including discounts, referrals, sales salaries.

LTV (Lifetime Value): Present value of all future gross profit from a customer after discounting for churn. Must be 3x+ CAC.

Payback Period: Months to recover CAC from contribution margin. Target under 12 months.

Attachment Rate: Average products per customer. More products per customer = higher revenue without new acquisition.

MAU (Monthly Active Users): Users who opened app this month. Vanity metric unless you track conversion to paying users.

Cohort Retention Curve: % of users from each acquisition cohort still active over time. Should flatten after 12-18 months (means churn is stabilizing).

Framework 9: The Metrics Founders Use to Fool You

Founders will dazzle you with big numbers that mean nothing. Here’s what to watch.

GMV vs NMV vs Revenue:

GMV is what founders quote. Sounds huge.

But subtract discounts, returns, cancellations. NMV is lower.

But the platform doesn’t keep NMV. They take commission (take rate). Revenue is a fraction of GMV.

Then apply gross margin. After costs, actual gross profit can be 10-20x smaller than the GMV number founders quoted.

Founders quote GMV because it’s often 20-30x bigger than what matters.

The discount trap:

Platform offers heavy discounts. Orders spike. Founders claim “massive demand, huge TAM (total addressable market: maximum revenue opportunity if you captured 100% of a market).”

Cut discounts. Orders drop sharply. Real TAM just revealed itself.

Most TAMs are discount-inflated fiction. When subsidies go, demand shrinks, real market size appears. Ask: what’s order volume at zero discounts? That’s your actual TAM.

The funnel deception:

Founders quote MAU (Monthly Active Users: people who opened the app at least once this month). “We have 50 million MAU!” Sounds impressive.

Build the funnel (the step-by-step path from awareness to purchase):

• MAU: 50 million (opened app)

• Product page visited: 20 million (40% drop-off)

• Checkout initiated: 5 million (75% drop-off from original)

• Transaction completed: 2 million (96% drop-off from original)

Real transacting users: 2 million, not 50 million.

Now check retention (percentage of customers who come back next period). If 60% churn (stop using the product), next month you have 800K retained users. That’s 1.6% of “50 million MAU.”

Founders sell you top of funnel. You need bottom of funnel + retention.

Attachment rate matters more than you think:

Attachment rate drives revenue per user without new customer acquisition.

E-commerce platform: Customers buying only from one category have low annual spend. Add second category: spending doubles. Add third category: spending triples or more. Same customer, multiple times the revenue, purely from cross-sell (selling additional products to existing customers).

This is why platforms obsess over cross-sell. Acquiring customers is expensive. Selling them more products is nearly free.

Upgrade to Read the Full Analysis

This is where the free version ends.

The frameworks above give you the foundation. But the real edge comes from the 6 advanced frameworks that follow (Will be shared over email):

Subscribe now

The Bottom Line

Traditional tools were built for manufacturing companies in 1920. They measured factories, inventory, and quarterly margins.

Internet businesses don’t have factories. Their assets are customer relationships that never touch the balance sheet. Their costs are front-loaded investments accounting treats as expenses. Their moats are network effects that compound exponentially. And founders deceive with vanity metrics that look impressive but mean nothing.

Using old tools on new businesses doesn’t just give wrong answers. It gives backwards answers.

The wealth-creating businesses over the next decade won’t have the best P&L today. They’ll have the best unit economics, strongest network effects, deepest moats hidden in places your DCF doesn’t measure. And the discipline to ignore GMV while obsessing over contribution margins and cohort retention curves.

Traditional analysts keep saying these are “overvalued” while missing the compounding value underneath. The market structure is winner-take-all. Unit economics are positive and improving. Retention curves are flattening. Flywheels are spinning. Optionality value is multiple times current revenue.

Your spreadsheet just can’t see it.

This is why some internet businesses trade at high multiples despite current losses. Why second place means nothing in winner-take-all markets. Why the path to profitability shows in unit economics years before the P&L turns positive.

They’re not crazy. They’re using the right ruler.

India went from 1.5% ethanol blending to 20% in a single decade. No other country has moved this fast at this scale.

And most investors haven’t noticed.

Ethanol blending is simple in concept. You take ethanol, a biofuel made from sugarcane or surplus grain, and mix it with petrol before it reaches the pump. E20 means 20% ethanol, 80% petrol. No engine change for the consumer, no new pump infrastructure, no behavior change. The fuel just arrives blended. The entire machinery that makes this happen sits upstream, mostly in sugar mills across UP and Maharashtra.

Here is what that machinery has been doing:

• Ethanol supply to OMCs (Oil Marketing Companies) grew from 38 crore litres in FY14 to over 700 crore litres by FY24. Nearly 20x in a decade

• Blending crossed 20% in 2025, five years ahead of the original 2030 deadline

• India’s crude import bill was ₹14 lakh crore in FY25. Over 88% of consumption is imported. Modiji’s recent push to reduce dollar outflows makes this a national priority, not just energy policy

• ₹1.44 lakh crore in forex already saved since the program began

• Gadkari ji has said it plainly: conventional fuel vehicles have no long-term future. EVs aren’t ready for mass transition. Ethanol is the bridge

• E20 is not the ceiling. The government is already drafting E25/E27/E30 roadmaps. This is mid-innings

• Sustaining E20 alone needs 10.16 billion litres annually. Moving to E30 means roughly 50% more volume demand from here

• From April 2025, all new petrol vehicles must be E20-compatible. Flex-fuel vehicles running on E85 are arriving from Maruti, Tata, and TVS through 2026

• India is now the world’s third-largest ethanol producer, and exports almost none of it

The EV story needs consumer behavior to change, capital at scale, and a grid that can handle the load. The ethanol story needs none of that. It needs sugarcane, policy, and distilleries. Two of those India has in abundance. The third has been unusually consistent.

The question worth asking: who in this value chain is actually capturing the value? And has the market figured that out yet?

Leave a comment

The Value Chain Nobody Is Drawing

Before picking stocks, you need to understand how ethanol actually moves from a sugarcane field to your fuel tank. Most coverage stops at “sugar companies benefit.” That framing misses half the story.

Six distinct layers sit between the farm and the pump. They don’t all benefit equally. Some are pass-throughs. Some are commoditized. One has genuine pricing power.

Layer 1: Feedstock Farmers growing sugarcane, maize, and other grain crops. The government raises something called as the Fair and Remunerative Price (FRP) for sugarcane roughly 4-5% annually. Farmers benefit, but there is no listed play here. Fragmented, price-taking, move on.

Layer 2: Distillery Technology and EPC These are the companies that design and build the distilleries themselves. When a sugar mill decides to add ethanol capacity, someone has to engineer the plant, supply the equipment, and commission it. This layer is asset-light, IP-driven, and not exposed to feedstock volatility. It is also where the only real moat in this chain sits. More on this shortly.

Layer 3: Integrated Sugar Mills with Distilleries. The bulk of listed plays. These companies crush sugarcane, produce sugar, and divert molasses ( to be read as cane juice) into distilleries to make ethanol. They sell directly to OMCs at “government-administered prices”(when you read terms like these, you stay away) under long-term offtake agreements. Revenue is recurring, buyer is guaranteed, price floor is set. The risk, as we will see, sits in the gap between input costs and administered output prices.

Layer 4: OMCs IOCL, BPCL, HPCL. They procure ethanol from producers, blend it at their depots, and sell it at the pump. Mandatory buyers by law. The ethanol program barely moves the needle on their overall margins. Think of them as infrastructure, not beneficiaries.

Layer 5: Logistics and Storage Ethanol is classified as a Class-A flammable liquid. It needs dedicated storage tanks, separate tanker fleets, and eventually dedicated pipelines. Today this is largely unorganized. The government is planning Ethanol Blending Corridors under the National Logistics Policy 2022 (I recommend you read this sometime). No clean listed play exists yet, but this is a layer worth watching as volumes scale toward E30.

Layer 6: Vehicle OEMs E20 compliance is a cost for automakers, not a moat. Flex-fuel variants (E85-capable) carry a modest 5-7% price premium. Maruti, Tata, TVS, Hero are all moving on this through 2026. But OEM valuations are driven by ten other factors before ethanol enters the conversation. Not the way to play this theme.

Simple chain. But the value does not distribute evenly across it. Two layers are worth serious attention. One is already well-owned by the market. The other is not.

Where The Money Actually Is

Farmers don’t list. OMCs don’t move on ethanol. Logistics has no clean play yet. OEMs have fifteen other things driving their stock price before ethanol enters the picture.

That leaves two layers worth your time.

The Compounder: Distillery Technology

Praj Industries is the only company in India that engineers and builds biorefinery plants at scale. When any sugar mill anywhere in the country decides to add ethanol capacity, the odds are high that Praj is involved. They have over 1,000 customer references across 100+ countries. More than 40% of revenue comes from repeat clients.

This is not a commodity business dressed up as technology. Praj has a dedicated R&D center in Pune with 100+ technologists and 20 PhDs, organized around five Centers of Excellence. They hold proprietary process technology across fermentation, distillation, and multi-feedstock handling. That IP is why a sugar mill calls Praj and not a generic EPC contractor.

And boy do we love good numbers….

The business is asset-light. Praj does not produce ethanol, does not own distilleries, and carries no feedstock risk. When the ethanol capex cycle runs, Praj books orders. When it pauses, order inflow slows.

Which is exactly what happened in FY24-25. India hit E20 ahead of schedule, which caused a temporary pause in new capacity additions. Sugar mills stopped ordering new plants. Praj’s bioenergy segment saw muted performance for roughly six quarters (We enjoy watching paint dry, we can be patient!)

That air pocket is now clearing. The E25/E30 roadmap means a fresh capex cycle is coming. And Praj has been using the pause well, building order books in Sustainable Aviation Fuel (SAF), Compressed Biogas (CBG), and bioplastics. Management has guided for revenues to triple by 2030, from ~₹3,400 crore today to ₹10,000 crore. Export share is targeted to reach 50%.

The risk is order lumpiness. This is a project-based business and revenue can be uneven quarter to quarter. Look at annual numbers, not quarterly.

The Cyclicals: Integrated Sugar Mills

These are not pure ethanol companies. They are sugar businesses that are structurally shifting their revenue mix toward distillery. Five years ago distillery was 10-15% of revenue for most of them. Today it is 25-30% and rising(Value mix changing is a very good mental model to identify earnings growth). The direction is clear. The pace depends on two things: ethanol procurement prices and feedstock costs.

Here is the central tension in this layer. The government raises the FRP(aka price) for sugarcane every year, typically 4-5%. Ethanol procurement prices, set by OMCs under government guidance, have not kept pace. That gap compressed margins sharply in FY25 across the board. This is the single biggest risk in the sugar-ethanol space and it is policy-driven (Already told you to stay away from it), which means it can correct as quickly as it appeared.

Multi-feedstock flexibility is the key differentiator here. Mills that can switch between B-heavy molasses, cane juice, and maize depending on price economics are structurally better positioned than single-feedstock operators.

Balrampur and Triveni are the quality anchors. Balrampur in particular has been the most consistent operator in the sector, with distillery revenue growing from ₹566 crore in FY20 to ₹1,689 crore in FY24, almost entirely through ethanol. Triveni has a useful cushion in its engineering and gearbox business which buffers sugar volatility.

Shree Renuka has the largest distillery capacity in India at 1,250 KLPD but carries significant leverage and weather-related earnings risk. High volume, lower quality.

One line on the smaller names: Bannari Amman is net cash, profitable, underfollowed, and South India gives it geographic diversification away from the UP-Maharashtra concentration. Worth a closer look if you want something off the beaten path.

What Can Go Wrong

Every structural theme has a ceiling. Here are the real ones.

The FRP trap is the most immediate risk. The government raised sugarcane FRP to ₹355 per quintal for 2025-26. Ethanol procurement prices did not move proportionately. Mills get squeezed from both ends and have no pricing power to push back. This is not a market problem, it is a policy problem, which means it resolves only when the government decides it resolves. The good news is the government has strong incentive to keep distilleries profitable, otherwise the entire blending program stalls.

Overcapacity is already here. Installed ethanol production capacity in India is now ~20 billion litres. Annual demand at E20 is ~11 billion litres. Utilization sits around 55%. The E25/E30 roadmap absorbs this, but the timeline matters. If the government moves slowly on higher blend mandates, the capacity overhang suppresses returns for longer than the market expects.

The monsoon is always in the room. A bad sugarcane crop compresses feedstock availability, cuts ethanol volumes, and hits earnings hard. Dwarikesh Sugar’s red-rot disease episode is a live example of how quickly a crop event can derail a season. Multi-feedstock mills that can pivot to maize are better insulated but not immune.

EV acceleration is a long-dated tail risk. If EV penetration in India moves faster than current projections through 2030-35, petrol demand peaks earlier. Ethanol demand follows. This is not an immediate concern given where EV infrastructure stands today, but it is worth keeping in one corner of your mind on a 7-10 year horizon.

Praj specifically carries order lumpiness risk. The pause in FY24-25 showed how quickly bioenergy order inflows can slow when domestic capex takes a breather. The SAF and CBG diversification helps, but Praj is still project-dependent. Quarters will be uneven. Oh, and did I mention the P/E ratio?

None of these risks invalidate the theme. The government has too much at stake, having already committed its credibility to a decade of aggressive targets, to let this stall. But they do define where position sizing and stock selection matter.

How To Think About This As An Investor

This is not one trade. It is three different bets with three different risk profiles. Be clear about which one you are making.

Bet 1: The Compounder (Praj Industries)

If you want quality, this is your answer. Asset-light, debt-free, IP-driven, globally diversified, and now expanding into SAF, CBG, and bioplastics as the next growth leg. The business compounds well because it does not carry feedstock risk, does not depend on government procurement pricing, and earns on every new capex cycle regardless of which sugar mill wins.

The catch is valuation. Praj trades at a significant premium to the sector, and the market is already pricing in the E25/E30 cycle and the SAF/CBG opportunity to some degree. At current levels, you are paying for a lot of the future today (I’m not comfortable at these valuations, but I’ll monitor). The entry point matters here more than it does for a commodity cyclical.

Bet 2: The Cyclical Mix-Shift (Balrampur Chini / Triveni Engineering)

This bet is on a structural revenue mix shift that is already happening. Sugar was 85-90% of revenue five years ago. Distillery is now 25-30% and heading higher. As that mix shifts, earnings quality improves, cyclicality reduces, and the market should rerate these businesses over time.

Balrampur is the cleaner of the two. Best-in-class operations, 1,050 KLPD capacity, multi-feedstock flexibility, and a balance sheet that is not stretched. Triveni has the added cushion of its engineering and gearbox business, but carries more debt after recent expansions (we have debt! except for lending businesses - which we love - Check out the post on Northern Arc Capital).

The near-term risk is the FRP trap. If procurement prices don’t get revised upward in the next Ethanol Supply Year, margins stay compressed. Watch for the annual ethanol price revision announcement from the government. That is the single biggest near-term catalyst for this segment.

Valuations here are more reasonable than Praj but not cheap. Both stocks have re-rated meaningfully over the last two years as the market has woken up to the distillery mix shift. You are not buying ignorance. You are betting on execution and policy normalization.

The Decade Ahead Is More Interesting Than The Decade Behind

E20 was chapter one. The government is already writing the next one.

E25, E27, E30 roadmaps are being drafted. Flex-fuel vehicles arrive in showrooms this year. Ethanol for gensets and industrial applications is being actively piloted. The demand pool is expanding well beyond transport blending.

What separates this from most policy-driven sector stories is the execution track record. Every blending target India set, it hit ahead of schedule. That consistency is itself an investment input.

The value chain is still being built. The logistics layer has no listed play yet. Praj’s SAF and CBG ambitions are 2027-28 stories. Second-generation ethanol is early stage. There is more runway here than the current valuations suggest, but only if you are patient enough to let it play out.

The fuel in your tank is already 20% sugarcane.

Most investors still haven’t processed what that means for their portfolio.

Subscribe now

Shout out to folks at SEBI!

This article is for informational and educational purposes only. It does not constitute investment advice, a recommendation, or a solicitation to buy or sell any securities. The views expressed are personal and based on publicly available information. Investments in securities markets are subject to market risks. Please read all related documents carefully and consult a SEBI-registered investment advisor before making any investment decisions. The author may or may not hold positions in the securities mentioned.

“The first step in an investment program is to see how money has been most successfully made in the past.” - Phil Fisher

Between 2008 and 2025, India added its first trillion dollars to GDP. During that stretch, 35 companies turned Rs 1 lakh into Rs 17 lakh or more. Bajaj Finance delivered 39% CAGR. Eicher Motors, 37%. SRF, 34%. There are some patterns that these “Common” stocks followed. These need to be decoded by anyone interested in the next set of multi-baggers, but we’ll get there in a minute ( or more). Going more top down for now…

India now stands at $4 trillion GDP, projected to hit $16 trillion by 2042. The playbook that worked for the last trillion holds clues for the next.

Why This Time Is Different ( I can’t believe I said that, it’s never different)

India went from $1 trillion to $4 trillion in 17 years. The next 17 should take it from $4 trillion to $16 trillion. The question you should ask is what changes when a country quadruples its GDP from this base. Here’s what changes:

• Spending change: At $2,500 per capita income, most money goes to roti-kapda-makaan. Food, clothing, shelter. The basics. But once those are covered, the next dollar goes somewhere else. A two-wheeler. An air conditioner. A mutual fund SIP. Health insurance..

Economists call this high marginal propensity to consume. I call it the point where things get interesting. For that we need to understand a concept

wealth effect: As asset prices rise (stocks, real estate, gold), consumers feel richer and spend more. That spending flows into corporate profits. Those profits lift stock prices. The cycle feeds itself ( Imagine yourself having found a multibagger after reading my articles, you’re going to be spending that money. At least some of it on fancy stuff, maybe even a paid subscription to this newsletter. Who knows! )

Where the money flows:

• Consumption: ACs, cars, two-wheelers explode

• Government finances: More tax collections, higher capex

• Corporate profits: Financial services, telecom, engineering, cement, steel all benefit

I’m hearing you say, “All this is fine, but show me the money”. Alright, I heard you. Let’s talk about the compunders that follow…

Five Laws Every Compounder Obeyed

I pulled the top 35 wealth creators from 2008 to 2025. Companies that delivered 16%+ price CAGR over the period. These weren’t random winners. They followed patterns.

Here are the five laws.

Law 1: Sector Tailwinds Matter Most

70% of compounders came from sectors with structural tailwinds.

The sectors that worked (2008-2025):

• Agro chemicals.

• Auto ancillaries

• two-wheelers.

• Banks, capital markets, and NBFCs.

• Cables and cement.

• Others: IT. Mining. Paints. Pharma. Retail. Healthcare.

Some of these make obvious sense. Banks and NBFCs benefit when credit penetration rises. Two-wheelers win when incomes cross a threshold. Paints ride the housing cycle. Others are more specific to that period. Agro chemicals had their moment. Cables benefited from infra build-out.

Portfolio implication: The point isn’t to replicate this exact list for 2025-2042. The point is to know that - Sector selection matters more than stock selection within a bad sector. Repeat after me - “Sector selection matters, tailwinds matter”. One more time…

Law 2: Consumer-Facing Businesses Win

60% of the Compounders were B2C businesses.

Why does B2C beat B2B and B2G?

• Sticky behavior. Consumer habits change slowly. Once you’ve had “Pahado wali Maggie” in your house, you’ll probably have it again and not some random ramen. Brand loyalty persists.

• Pricing power. Individual consumers can’t negotiate. They pay the sticker price. Large B2B customers can squeeze suppliers. Government buyers run tenders where the lowest bidder wins, but not in B2C.

• Higher returns on capital. B2C companies command premium returns. B2B businesses often operate in commodity-like conditions with thin margins.

• Less cyclical. B2B and B2G are prone to business cycles and technology disruption. Consumer businesses move slower.

Look at the list: Bajaj Finance, Cholamandalam (consumer finance). Eicher, TVS (consumer vehicles). Titan (consumer durables). Asian Paints, Berger Paints (consumer-facing building materials). Abbott, Torrent Pharma (consumer healthcare).

Portfolio implication: As far as possible, prefer B2C over B2B and B2G for long-horizon portfolios.

Law 3: Market Leadership in the Base Year

75% of the compounders were market leaders

Why do leaders keep winning?

• They enjoy a larger share of the current and future profit pool.

• They have the scale to withstand macroeconomic shocks.

• They have the financial power to invest through downturns.

• They have network effects and brand moats that compound over time

Bajaj Finance was already a leader in consumer finance by 2008. Asian Paints was number one in decorative paints. HDFC Bank was a top-tier private bank.

Portfolio implication: Don’t bet on turnarounds or the number four player hoping to become number one. Back existing leaders with room to run.

Law 4: High Return on Equity in the Base Year

80% of the compounders enjoyed 15%+ RoE in 2008 ( start of the period of this study)

In a free market, competition is like gravity - it eventually pulls high profits back down to earth. Any business that consistently makes returns higher than its cost of capital (RoE > CoE) is essentially defying economic gravity. It’s a rare anomaly that proves the company has a real 'moat' and a management team that actually knows how to defend it.

Some examples:

• Bajaj Finance: 2% RoE (2008) to 17% RoE (2025).

• Titan Company: 35% RoE (2008) to 29% RoE (2025).

• Abbott India: 29% RoE (2008) to 33% RoE (2025).

• TCS: 41% RoE (2008) to 52% RoE (2025).

The outliers with low starting RoE either restructured (Bajaj Finance, TVS Motor) or operated in industries where RoE naturally expands with scale (capital markets, some infrastructure plays).

Portfolio implication: Ensure most companies in your portfolio enjoy RoE meaningfully higher than 12% (the approximate cost of equity in India).

Law 5: Reasonable Entry Valuations

54% of compounders had a 2008 P/E lower than the Nifty P/E of 17x.

Stock returns are driven by two engines(Or shall I say double engine :) ):

• Earnings growth and

• valuation re-rating.

Most compounders enjoyed both. Lets have some examples shall we?

• SRF: 5x P/E (2008) to 70x P/E (2025). That’s a 14x re-rating on top of 13% PAT CAGR.

• JK Cements: 4x P/E (2008) to 48x P/E (2025). A 12x re-rating plus 7% PAT CAGR.

Coforge: 5x P/E (2008) to 68x P/E (2025). A 13.6x re-rating plus 11% PAT CAGR.

Contrast with high P/E starters:

• Godrej Consumer: 20x P/E (2008) to 63x P/E (2025). Only a 3x re-rating.

• CRISIL: 33x P/E (2008) to 74x P/E (2025). Only a 2.2x re-rating.

Both delivered strong earnings growth, but the valuation engine contributed far less.

Portfolio implication: A reasonable purchase price is non-negotiable. Lower P/E equals higher potential for the re-rating engine to fire. So repear after me “Bhav kam, return zyada (P/E in the floor, returns out the door.)"

The China Analog: What Happens at $2,500 Per Capita

India’s per capita GDP today sits around $2,500. China hit that level in 2007. What happened next in China offers a blueprint.

Between 2007 and 2011, China’s per capita GDP doubled to $5,652. Here’s what exploded:

Meanwhile, necessities flatlined:

Moral of the story: Discretionary items grew 15-21% annually. Necessities went nowhere.

Once per capita income crosses $2,500, consumption shifts from staples to durables and services. China saw it. Thailand saw it. Indonesia saw it. India is next.

India’s Consumption Math: Where the Money Goes

Private Final Consumption Expenditure (PFCE) is the total spending by Indian households. It’s the single best proxy for where consumer money flows.

PFCE grows at the same rate as GDP, but its share of GDP rises. This shows the wealth effect in action. As households feel richer, consumption as a share of GDP increases.

Now let’s break down where PFCE growth will concentrate.

Durables: The 22x Explosion

Durables include cars, two-wheelers, white goods (ACs, refrigerators, washing machines).

Durables’ share of PFCE is expected to rise from 3% to 10%. That would put India at current US levels of durable ownership.

Why 20% CAGR? Penetration is abysmal today. Rising incomes hit the tipping point for durable purchases. The aspiration shift from necessities to lifestyle products is just beginning. This is the single most explosive category in India’s consumption story over the next 17 years.

Services: The Steady Compounder

Services will continue their uptrend even on a rising base. Not as explosive as durables, but more consistent. The shift from goods to services is a universal pattern as economies mature.

Non-Durables: Slower but Still Massive

Non-durables include food, clothing, and FMCG.

The share of PFCE declines as households spend more on durables and services. But it still grows in absolute terms due to the base effect. A smaller slice of a much bigger pie.

All this is just fancy talk, till I know where to invest my money. That’s what we’re going to talk about.

The Sector Shortlist: Where to Hunt

The universe is classified into 8 groups and narrowed to those with structural tailwinds. Here's the framework

Consideration Set (The Hunt):

Non-Consideration Set (Avoid):

Now let’s go deeper on the top five sectors.

1. Financials: The Prime Mover

Financials were the top wealth creators from 2008 to 2025. Expect them to sustain that position through 2042.

Seven Tactical Takeaways from the Study

1. Forget Markets, Think Stocks

Between 2020 and 2025, market benchmark indices delivered muted returns. Nifty did 5-15% CAGR depending on the period you measure. But the top wealth creators still delivered 10-15% outperformance.

Lesson: Even in flat markets, compounders compound. Stock selection beats market timing.

2. Frothy P/Es Are Unmissable Selling Opportunities

Compounding is a function of earnings growth plus P/E expansion. But occasionally, stocks rally only on P/E expansion without earnings growth. These are distribution opportunities, not accumulation zones.

Rule: Just like low P/Es are buying opportunities, frothy P/Es (40x+, 60x+) are highly profitable selling opportunities. Don’t hold through euphoria.

3. Speed with Consistency

Seven of the top 10 consistent wealth creators traded at single-digit P/Es in March 2020 after the COVID crash. Five of those seven had 15%+ RoE in FY20.

Lesson: Sharp market corrections offer even healthy companies at attractive valuations. This is when you get speed (high RoE) plus consistency (proven track record) at value prices. Buy quality in crashes (Glories to Lord Buffett ).

4. Sensex Is a Weak Earnings Machine

Sensex earnings growth has been muted for several years. Over the long term, G = R. Growth in earnings equals return on the stock. A portfolio with even a slightly superior earnings profile versus Sensex should consistently outperform.

Lesson: Don’t benchmark against Sensex composition. Build a portfolio with superior earnings growth potential. The Sensex is not the enemy. It’s just not the right yardstick.

5. Banking & Finance Will Sustain Top Spot

India’s GDP is expected to double from $4 trillion to $8 trillion by 2033. Banking and finance is the lubricant of economic growth. Add to that the capital markets boom driven by retail participation.

Prediction: Banking and finance will sustain the number one spot in wealth creation for the foreseeable future.

6. PEG < 1x Is a Formula for Superior Returns

PEG stands for price-to-earnings-to-growth ratio. It’s calculated as (P/E) divided by (5-year earnings CAGR).

Using perfect foresight of 5-year earnings, we calculated PEG for every stock in March 2020. If a stock’s P/E was 20x in 2020 and its PAT CAGR from 2020-2025 turned out to be 25%, its PEG was 0.8x (20 ÷ 25).

Result:

Key insight: Stocks with PEG less than 1x delivered the highest returns. 40% wealth creation CAGR. 28% PAT CAGR.

Lower PEG equals higher return. Clear inverse relationship.

What This Means for Your Portfolio

Let’s bring it all together.

Sector allocation:

• Highest weight: Financials (banks, NBFCs, capital markets, insurance). 30-40% of portfolio.

• High weight: Autos (two-wheelers, passenger vehicles). 15-20%.

• Medium weight: Consumer durables, healthcare, real estate. 10-15% each.

• Low weight: FMCG, IT, pharma. 5-10% each. These will grow but not explosively.

• Avoid: Commodities, utilities, agri-based businesses.

Quality filters:

• RoE > 15% in the current year

• Market leader or strong number two in their category

• B2C preferred over B2B and B2G

• Sector with visible tailwinds for the next decade

Valuation discipline:

• Target PEG < 1x where possible

• Avoid P/E > 30x unless growth is exceptional and sustained

• Use corrections to buy quality at single-digit P/Es

Market cap strategy:

• Focus hunting efforts on mid caps ranked 101-250

• Add large caps for stability, but expect lower returns

• Avoid small caps unless conviction is very high

Holding period:

• Minimum 5 years

• Ideal 10-15 years

• Compounding takes time

Rebalancing:

• Sell when P/E exceeds 40-50x without commensurate earnings growth

• Trim positions that become more than 15% of portfolio

• Redeploy into under-owned sectors or new opportunities

Credits & Disclaimer

This article is a summary and interpretation of Motilal Oswal Financial Services’ 30th Annual Wealth Creation Study (2020-2025). The underlying data, exhibits, and sector framework are drawn from their exhaustive research report.

Mohnish Pabrai once said: “The best ideas are not original. I’m a shameless cloner. I steal ideas from people all the time.”

You don’t need to do every study yourself. You need to shamelessly pick the best from others and build on it. Read the original study if this sparked your interest. The full report has far more detail than I could cover here.

Special shoutout to bros at SEBI

The information provided in this article is for educational purposes only and should not be construed as investment advice. Investing in equities involves risk, including the risk of loss of principal. Past performance is not indicative of future results. Readers are advised to consult with a qualified financial advisor before making any investment decisions. The author may hold positions in securities mentioned.

If you found this useful:

• Share with fellow investors

  Share

• Comment: Which sector are you most bullish on for India’s next trillion?

  Leave a comment

• Subscribe for more deep dives into Indian equities, macro, and investing frameworks

  Subscribe now

• Revenue growing at 24% year on year

• EBITDA margins at 35% and still expanding

• Return on equity at 30-33%, consistently

• Net debt falling from ₹850 crores to ₹322 crores in 18 months

• Dollar-denominated revenue, recession-proof sector, rupee depreciation works in your favour

You would be interested, right? Oh, and one more thing - it’s an Indian IT company — a sector investors have spent two years punishing over AI disruption fears. But for this one, AI is a tailwind, not a threat ( more on this shortly).

The company is Inventurus Knowledge Solutions — IKS Health — and its clients are American doctors.

To understand why this is interesting, you need to understand what is broken about American medicine. And to understand that, you need to understand how US healthcare actually works. From scratch ( This next part is painful, but please bear with me, you really need to understand this to understand the company well, but that pain will be relieved when we talk about the moats and the thesis)

The Crisis: American Doctors Are Drowning in Paperwork

Picture a physician in Seattle. A fully American physician named Ms. Patel. She sees patients from 8 am to 5 pm. Then she works until 9 pm. Not on medicine, but on paperwork. Forms, codes, approvals, and appeals. Understandably, she is frustrated. She’s not an anomaly; other fully American doctors like Dr Singh and Dr Shah are also frustrated.

Enough jokes, here’s the real deal. American physicians now spend roughly 50% of their working time on administrative tasks. The most expensive man-hours in the building are doing data entry.

The downstream chain is ugly:

The EHR initiative (Electronic Health Record - the digital system every hospital uses to store patient data) was supposed to be the great moderniser. Companies like Epic, Cerner, and NextGen operate here. Guess what happened? The EHR digitised the record, but it did not reduce the work. Physicians now spend more time clicking through these systems than they spent on paper. Dr Shah was still unhappy.

What nobody built was the layer that actually does the work and makes Dr Shah happy. That gap is IKS.

Before IKS Makes Sense, You Need to Understand the Plumbing

In India, patients pay out of pocket, but in the US, Insurance pays almost everything. Insurance can either be government-based ( Eg, Medicare for over-65s, Medicaid for low-income populations. Mentioned this to give you a sense of the US market). or private commercial insurers (Can be employer plans or individual plans).

The problem: There are hundreds of private insurers, each with its own rules, rates, and approval processes. When Dr Shah treats the patient. He must convince the insurer to pay for it. That process is where the administrative burden lives.

Why? Every single patient encounter must be documented, coded, and submitted in formats that the insurer accepts. If the documentation is off even by a mark, the claim is denied (think of your last visit to SBI)!

How does the documentation work? Every hospital and clinic in the US runs an Electronic Health Record (the digital database that stores patient data). The EHR records that Dr. Smith saw a 58-year-old with hypertension and ordered a stress test. It does not know what billing codes (standardised codes that tell the insurance company about the diagnosis and prescription for them to bill) to attach, whether the insurer requires pre-approval for that test, whether the patient’s coverage is active, or what to do when the insurer rejects the claim.

That gap between the EHR recording what happened and the insurer paying for it, is where IKS operates.

The clearing house is the pipe in the middle: Between the physician and the insurer sits a clearinghouse ( Companies like Change Healthcare and Availity). It acts as the Universal travel adapter. It’s a connector between the provider’s EHR and hundreds of insurance companies. It translates raw clinical data into the specific electronic formats required by each unique insurer to ensure claims are accepted and processed.

Before IKS, the doctor is the bottleneck, manually translating, coding, and formatting at every step.

After IKS - the doctor speaks, IKS handles the rest (records and transcribes). The burden zone collapses to a single automated step.

Coding: The Translation Layer

As discussed, before a claim hits the clearinghouse, every clinical note must be converted into standardized numeric codes.

The precision required is absolute:

• Wrong Code: Immediate denial.

• Over-coding: Risk of federal fraud audits.

• Under-coding: Revenue left on the table.

This is where “revenue leakage” happens.

Why this is the “IKS Moat”: Well, you can’t just “software” your way out of this. Because insurers constantly change their “rules,” a pure AI play often fails the edge cases. IKS wins because they combine Agentic AI with a Global Human-in-the-loop (13,000 experts) to ensure the translation is always compliant.

Oh, by the way, this was just one example I elaborated to show you the pain. Such admin work is dispresed across the entire journey of the patient. See this flow diagram

I know you skipped it. Please go through it the entire flow, read all the text once more to really understand all that IKS does.

Why does it all matter? Here’s why - Every step is administrative. Every step is either the physician's time or a staff member's time. Collectively, this work represents about 15% of total physician revenue annually - roughly $225 billion (with a “B”) in the US physician market alone. That $225 billion, my fellow investors, is the TAM.

The 16 Chores: Three More Examples ( You can skip this if you want or read it later)

IKS in total does 16 such interventions. You have already seen prior auth and coding. Here are three more to complete the picture.

Scribble: IKS's version of Notetaker - an ambient AI listens to the patient-physician conversation and writes the clinical note in real time ( Earlier, our Dr Shah would have to type it all).

Denial Management: Insurers are notorious for denying claims (~a whopping 10% denial rate). A large physician group billing $500 million can see $50 million in revenue vanish annually. Most groups simply write off a huge portion. Why? because the manual cost of “chasing” the insurer exceeds the value of the claim itself. IKS flips this math using the AI triages denials based on their probability of recovery (as it’s trained on decades of proprietary data), automatically resolving simple rejections. It also automates the “chase” using AI. That’s how IKS captures millions in lost revenue that was previously considered unrecoverable.

Last one, this is interesting…

Value-Based Care: In the traditional US model, doctors are paid for the volume of tests they run; in this new model, called Value-Based Care (VBC), they are paid for the quality of the outcome. If you were an insurer, your goal would be to reduce claims by incentivizing clinics to prevent patients from getting sick in the first place. IKS provides the essential infrastructure for this transition: their AI filters the patient base to identify high-risk individuals with “care gaps,” extracts the quality reports insurers demand for higher payouts, and initiates proactive AI-driven outreach to nudge patients into the office for preventative exams. Again by doing so, its putting more money in the clinic’s hand without human intervention.

All these chores are categorized into 3 types as follows:

The TAM: How Big Is the Opportunity

Though we’ve already discussed it once, I want to touch upon it once more very quickly, to understand the market segments and where exactly IKS operates.

• The TAM = $260B is the total US admin spend

  • From it, only ~13% has been outsourced so far (growing 12-14% annually )

    • From it, IKS revenue: ~$330M ↓ IKS market share: ~1%

The US healthcare admin market splits into two worlds.

1. Outpatient - physician clinics, specialist groups, primary care. This is where IKS dominates.

2. Hospital/acute - inpatient care, health systems, emergency. IKS entered this via an acquisition called AQuity in 2023. This added further $35B to the TAM. Together, they add up to ~$260B in total admin spend annually.

Where IKS stands today:

• ~$330M revenue,

• ~650 client organisations,

• ~150,000 physicians,

• 95%+ recurring revenue, near-zero churn.

Here is the floor case: Even if IKS just grows with the outsourcing market at 12%, you get 12% constant currency growth, in dollars, in a sector with zero correlation to Indian macro, with rupee depreciation padding every quarter. No market share gains assumed. That is the floor. Now here is where it gets interesting.

Why the business is interesting to look at now? The Answer lies in business evolution

The evolution: from billing outsourcer to outcome partner

IKS 1.0 - the back-office operator (2007–2012): IKS started doing what dozens of Indian IT companies were doing: processing medical bills for US hospitals. You get paid per claim processed. You grow by hiring more people. Margins follow headcount.

IKS 2.0 - the technology platform (2013–2022): Billing was just one of 16 admin chores. IKS decided to own all sixteen. For that, they transitioned into software, not just people. The pricing model followed the logic. Instead of charging per transaction, IKS took a cut of client revenue. If the physician org grows, IKS grows.

IKS 3.0 - outcome ownership (2023–present) [This is super important]

This is where it gets interesting. IKS stopped being a vendor and started acting like a partner with skin in the game. It began co-investing in select practices it serves through Joint ventures (and also something called MSOs but don’t fret about it). Taking equity stakes. Putting its own capital at risk alongside its clients.

Here’s how the JVs work:

The doctor owns the medicine. IKS owns the machine.

In a JV, the doctor retains 100% clinical control. IKS takes over everything else: technology, staffing, billing, compliance. Then it writes a cheque into the practice. Not a service fee. Actual capital at risk. From that point, they share a P&L. If IKS makes the practice run better, both sides make more money. If IKS fails to deliver, it eats the loss.

The revenue flows like this

We see this in numbers too: Between Q1 FY25 and Q3 FY26, IKS’s revenue grew 24% year-on-year. Headcount grew 0.5%. A traditional services company growing 24% would have hired 20–22% more staff. IKS hired almost nobody.

You guessed the reason - The technology is doing the incremental work. Operating leverage!

The Thesis: Why This Is Hard to Attack

If you’ve read so far, here’s the reward. Thank you for your patience. This section is where I have differentiated insights compared to the other brokerage reports you’ll read. See for yourself. My experience in B2B enterprise software taught me these, here goes

[A] In B2B enterprise, the moat is never the technology. It is the sales.

A health system CFO does not browse the internet for billing software. They call a peer and ask who they use. That answer is driven entirely by who already has the marquee names in the segment.

This is the Big Logo strategy. Land the market leader in a segment. Every smaller player in that segment then assumes three things: the software is battle-tested at scale, the edge cases are already solved, and the risk of being an early adopter is gone. The big logo does the selling for you.

Its also why some tech bro in a hoodie can’t replace them: No CFO replaces a $50 million outcome-guaranteed contract because a startup shipped a better model last Tuesday. Sales motion and reference architecture are the real barriers. Not the code.

[B] Platform vs Point Solutions: Who Wins Where

The market has hundreds of point-solution vendors (vendors who solve just one problem - eg: Denial claims redressal). One for coding. One for prior auth. One for documentation. Each is easier to buy individually - smaller contract, faster sign-off, lower perceived risk.

But when you run eight of them and the math flips. Eight contracts, eight integrations, eight renewal cycles, eight compliance audits. And then there’s the biggest of all issues: these systems need to be stitched together. Data from one system needs to flow into another smoothly and the whole plumbing should work.

Think of it like hiring multiple cooks for your kitchen. One chops vegetables. One sautées. One kneads the dough. One roasts the chapatis. Each is excellent at their job. But getting them to coordinate a meal on time? That is a huge headache. The CFO eventually adds up the total cost of managing this stack, plus the value leaked through the gaps, and asks: is there a better way?

That question is IKS’s sales pitch. No competitor has all sixteen features. So when a top-tier physician group CEO decides to consolidate ( and consolidation is increasingly the decision, not the exception ) IKS is the only call. Here’s the feature wise list of competetion.

[C] The Land and Expand Strategy

Every strategy has its ceiling. In large health systems above $5 billion in revenue, the platform pitch does not land. These are not physician group CEOs making outcome-driven calls. These are massive institutions where the CIO runs procurement, decisions go through multi-layered committees, and budgets are siloed by department. Nobody is signing a sixteen-feature consolidation deal. They are approving one tool at a time.

So IKS uses the Land and Expand strategy (enter a large organisation with a single product, prove value, then systematically expand wallet share over time). Enter with coding or clinical documentation, prove ROI over 12 months, get the next meeting, add the next feature. Repeat until you own the whole stack.

Slower, yes. But the wallet at these organisations is enormous. And critically, IKS did not have this sales muscle before AQuity. Hospital procurement is a different animal from physician group sales. AQuity brought 40 years of CIO-level relationships in exactly this market. The land-and-expand playbook in large health systems is, in large part, an AQuity inheritance. One of the top five health systems in the US is currently in this expansion motion. That is the proof point.

[D] Can EHRs Just Do This Themselves?

It is the first question every investor asks. EHRs are already the backbone of every hospital. They store the data, they sit inside the workflow, they have the physician relationship. Why can they not just build upward and replace IKS?

Three reasons. Different buyer, different incentives, different organisational DNA.

The buyer is different.

EHRs sell to CIOs. IKS sells to CFOs. Same hospital, completely different conversation. The CIO wants stability, uptime, and compliance. They are evaluated on systems not going down. The CFO wants EBITDA improvement. They are evaluated on margins. IKS’s pitch, “We will improve your margins by 850 to 1,200 basis points after paying for our platform,” is a CFO conversation. The CIO does not make that call and frankly does not care.

The incentive structure is different.

EHRs have spent decades building an ecosystem, an open platform that hundreds of third-party vendors build on top of. Think Microsoft Windows. Microsoft did not try to own every application built on Windows. They built the platform and let others build on top. EHRs operate the same way. IKS is actually on Epic’s App Marketplace. Epic benefits from IKS solving the billing problem. They have low incentive to compete with a partner that strengthens their ecosystem (might change with advent of AI though).

The organisational DNA is different.

Delivering on an EBITDA guarantee requires a 13,000-person global delivery organisation trained in US healthcare administration. It requires driving physician adoption, managing workflow transitions, and writing $16.5 million guarantee cheques to hospital CFOs. EHR vendors are software businesses. Lean, high-margin, built to scale code, not operations. They are not building a global human-in-the-loop delivery org. It contradicts everything about their business model.

The backbone and the engine room are two different jobs. EHRs own the backbone. IKS owns the engine room. And they are, quietly, partners.

[E] Once IKS Is In, Getting Them Out Is Not a Decision. It Is a Crisis.

Think of those overhead electricity wire bundles you see in Delhi. Hundreds of wires, intertwined over decades, each one connected to something critical. Nobody knows exactly which wire does what anymore. You need to fix one connection, but pulling the wrong wire takes down three buildings. So nobody touches it.

That is what replacing IKS looks like from the inside.

The exit friction stacks up fast:

• Workflow disruption. Every staff member, every physician, every administrative touchpoint has been trained on IKS’s systems. Retraining hundreds of people on a new platform, while keeping a live healthcare operation running, is a months-long nightmare.

• Data migration. Years of claims history, denial patterns, coding decisions, and patient-level billing data live inside IKS’s platform. Moving that cleanly to a new vendor is not straightforward.

• Contractual lock-in. Palomar signed a 15-year contract with no exit for convenience. WWMG signed a 30-year perpetual MSO contract. There is no clause that says “if you find something cheaper, good luck.”

• Equity entanglement. In MSO structures, IKS owns 48% of the administrative company. You cannot fire a partner who owns half your back office. There is no clean mechanism to do it.

A cheaper competitor can show up with a better demo. It does not matter. The cost of switching, in disruption, retraining, data migration, and contractual exposure, almost always exceeds whatever savings are on the table.

[F] The regulatory tail winds

Two regulatory acts have quietly become the strongest tailwind in IKS’s favour.

The 21st Century Cures Act (2016) and CMS-0057-F together mandated that all US healthcare data must flow via FHIR APIs (standardised pipes for health data exchange). Information blocking (due to data not flowing across point solutions) is now illegal, penalties up to $1 million per violation, and the final deadline is January 2027. For years these sat in IT departments as compliance to-do items. Around 2022, enforcement began in earnest. The slow crawl became a sprint.

Every fragmented vendor must now independently achieve FHIR compliance. Every integration must be audited. The CFO’s legal team is telling them that managing 10 non-integrated vendors is a board-level liability. The pitch writes itself: consolidate to one FHIR-compliant platform. IKS has all 16 features, already integrated, already compliant.

[F] NEVA: The Part That Makes This Truly Different

You already understand the JV structure. IKS co-invests, shares the P&L, skin in the game. NEVA takes that logic one step further.

NEVA (Net Economic Value Add) is the measurable dollar improvement IKS’s full platform generates for a client. At full deployment, 850 to 1,200 basis points of EBITDA improvement for a business running at 6-8%. Potentially a doubling. Instead of asking clients to trust that projection, IKS advances a portion of it as cash upfront, before generating a single dollar of value.

Traditional vendor: client pays vendor. NEVA: IKS pays the client. The deal with Palomar Health (a newly acquired client) is the proof point.

Here’s how the deal works out if you’re interested ( else skip it)

No healthcare IT vendor writes a $16.5 million guarantee cheque to a client. Doing so requires three things at once: deep confidence in your outcomes, a clean balance sheet, and full platform breadth to actually generate the NEVA. IKS has all three. Nobody else does.

The result is two pools of economics: traditional platform fees on the P&L, and equity and NEVA upside from the practices IKS serves. At scale, the second pool could exceed the first. Nothing on Indian exchanges looks like this.

[G] The AQuity acquisition: 1+1 = 5

Most analysts read this as a TAM expansion play. They were right, but they missed the more important parts.

AQuity served hospitals and health systems. A buyer IKS had never penetrated, and crucially, a completely different persona. IKS’s customer is the physician group CFO. AQuity’s was the hospital CIO. Same industry, different conversation entirely. IKS could not have walked into those rooms cold.

What the acquisition added was a second sales arm. Not just contacts. A tenured team with years of earned trust in a market IKS had no access to. Those relationships will compound. Every hospital in the base is a reference for the next deal. Every cross-sell adds to the network.

Four things money cannot buy quickly:

• Second sales arm. A tenured team with warm pipelines in a market IKS couldn’t enter organically at speed.

• Buyer credibility. Hospital CIOs trusted AQuity. That trust transferred to IKS on day one.

• New segment entirely. Hospitals are bigger, more complex, and buy differently from physician groups. IKS had none of this before 2022.

• Cost structure. AQuity’s operations absorbed into India-based delivery. Same work, lower cost.

The opportunity sitting on top: AQuity’s customers only ever bought one thing. IKS now has sixteen features to sell into that base. That pipeline did not exist before the deal closed.

Then the proof. Margins fell to 24% post-acquisition. Management guided eighteen to twenty-four months to recover. They did it in six to eight quarters, ahead of their own targets. Bought messy. Integrated clean.

[H] Earnings quality: predictable, sticky, and accelerating

Long-duration contracts. Revenue tied to client performance, which means it grows as clients grow. Net revenue retention above 100%. Churn in the low single digits.

IKs is not your typical SaaS business with monthly subscriptions that can be cancelled with a click. These are multi-year operational relationships where the cost of switching is real and measurable. The earnings profile looks more like a utility than a software company: predictable, growing, and structurally protected.

The margin trajectory adds a layer. Post-AQuity integration, IKS reached 35% EBITDA margins, while simultaneously tripling R&D investment as a share of revenue. The technology is doing more of the work. Operating leverage!

That’s the thesis, folks!

If you’ve read this far, you now understand IKS better than most institutional investors who have looked at it. The business model, the evolution, the market structure, the moat, the AQuity logic, the NEVA deal mechanics, and the risks. That is a complete picture for any serious retail investor.

Writing this took a significant amount of time. Primary research, earnings call transcripts, brokerage reports, filings, and a lot of thinking. If this added to your understanding of how to analyse a business like this, that genuinely means a lot. If you liked it, do leave a comment.

Leave a comment

For paid subscribers, I’ve put three additional sections behind the paywall (Will be shared in a separate new letter):

• The AI tailwind: Why IKS is structurally different from every other Indian IT company, and why the same force that is an existential threat to Infosys is a margin expansion engine for IKS.

• The risks in detail: A forensic pass through the six things that could go wrong, written to be readable, not intimidating.

• The forensic analysis: Is there earnings manipulation, or fictitious revenue? or is the prompter dishonest? We’ll find out

These are not filler. They are the parts of the analysis that take the most time to do properly, and the parts that most retail investors never see.

The information provided herein is for informational and educational purposes only and does not constitute financial, investment, or trading advice. The author is not registered with the Securities and Exchange Board of India (SEBI) as an investment advisor. All investments are subject to market risks, including the possible loss of principal. Past performance is not indicative of future results. Readers are advised to do their own due diligence and consult a SEBI-registered investment advisor before making any investment decisions. The author disclaims any liability for actions taken based on this content.

That business is CAMS. This is the full investment case.

A quick word before we begin. Writing this takes weeks of research, verification, and editing. I get approached regularly to write favourably about companies in exchange for fees. I have turned all of them down and I will continue to. If you find this work valuable, please consider pledging your support. It means more than you know.

Subscribe now

RTAs are the caterers of the mutual fund industry.

A mutual fund's (AMCs) job is to attract investors, manage money, and make investment decisions. Everything operationally behind that ( processing SIPs, handling redemptions (withdrawals), updating folios (like accounts), reconciling transactions (matching records) ) is handed off to an RTA.

Like a wedding family that plans the guest list, curates the menu, and hosts the evening, but lets the caterer run the kitchen. The guests eat well. The family never touches a ladle.

That’s the AMC-RTA relationship. AMC owns the experience. The RTA executes it.

How about the competetion?

India has exactly two of them. CAMS and KFin Technologies. SEBI hands out the licenses for RTAs. No new significant player has entered this space in over 20 years. Read that again. Twenty years. Zero new competition. No new entrant to disturb your sleep.

Whos’s the winner: CAMS

• 68% market share by mutual fund AUM - Rs 55 lakh crore of the industry’s Rs 82 lakh crore (December 2025)

• 4 of the top 5 AMCs as clients: SBI MF, HDFC AMC, ICICI Prudential, and Kotak MF

• 10 of the top 15 AMCs on its platform

• 82% market share in new fund offerings (NFOs) in FY25

KFin has more AMC clients by count - 25 versus CAMS’s 21. But most of KFin’s clients are smaller, newer AMCs. CAMS has the large ones.

In the RTA business, this distinction matters completely. Revenue comes from 1. Folio count 2. Transaction volume (Please make a mental note of this, its important for later). Big AMCs have them both. That’s where the money comes from, CAMS rules here.

CAMS: Business Lines

1. Mutual Fund RTA: The Core Engine (~86% of revenue)

Everything that happens after you click “invest” is handled by CAMS. Specifically:

• You start a SIP → CAMS registers the instruction ( handles the execution)

• Your monthly SIP instalment runs → CAMS processes it, reconciles the payment, and allocates units

• You submit a redemption → CAMS validates it, initiates the payment back to your bank, and updates your folio

• You change your nominee or bank account → CAMS updates the record

• You request a statement → CAMS generates it

• A company distributes a dividend → CAMS processes and credits it to every eligible folio

The AMC doesn’t do any of this. They outsource the entire investor services and record-keeping function to CAMS.

Who uses it: The AMC contracts with CAMS. The end-user is every one of that AMC’s investors. CAMS currently maintains records for over 4 crore unique investors and processes 5.7 crore live SIPs as of Q4 FY25.

How CAMS gets paid:

Two revenue streams, both recurring:

• Per-folio annual maintenance fee: CAMS charges the AMC a fee for every folio (investor account) it maintains. A folio sitting dormant for a year still earns CAMS a maintenance fee. This is the structural, non-cyclical part of the revenue.

• Per-transaction fees: Every SIP processed, every redemption, every switch between funds.

The structural tailwind inside the model: India's MF industry is shifting toward equity. Equity AUM as a percentage of total AUM has been rising every year. Higher equity mix means higher fee yield per rupee of AUM serviced.

Growth levers and metrics to track:

2. CAMSPay - The Payment Infrastructure Arm

CAMSPay is a payment infrastructure business serving the BFSI sector. It handles the plumbing behind recurring financial transactions: UPI AutoPay mandates for SIPs, NACH for loan EMIs and insurance premiums, card-based gateways for lump-sum investments, and instalment collections for NBFCs and banks.

Its client base goes well beyond mutual funds. NBFCs like ICICI Home Loans, banks including Canara Bank, Indian Bank, and HSBC, and insurance companies all use CAMSPay for payment collections. 25 new deals were signed in Q2 FY26 alone. The non-MF expansion here is real and accelerating.

Growth levers and metrics to track:

• CAMSPay revenue growth: 41% YoY in 9M FY26. The pace tells you whether the non-MF expansion thesis is working.

• UPI AutoPay transaction volumes: 96% YoY growth in 9M FY26. India’s fastest-growing payment rail is also CAMS’s fastest-growing product.

• New non-MF client additions: Each new NBFC or bank is a recurring revenue stream outside the MF base. Tracks how well CAMS is reducing concentration in its core industry.

3. CAMSRep - Insurance Repository

Let me set some context first:

Most people understand that shares are held digitally in a demat account ( in a depository like CDSL or NSDL ) rather than as paper certificates. The same concept is now being applied to insurance policies in India. IRDAI, the regulator for all insurance in India) has been pushing to change paper heavy nature of the insurance industry.

What CAMS is doing here:

CAMS operates CAMSRep, one of India’s four IRDAI-licensed insurance repositories. When an insurer dematerialises a policy, it becomes an eIA (e-Insurance Account). The flow works like this:

Bima Central is the platform through which policyholders can view, update, and service all their dematerialised policies in one place.

Who uses it: Insurance companies (as the issuers who dematerialise policies) and their policyholders.

Current numbers:

• 40%+ market share in e-insurance accounts in India

• 13 million policies digitised and maintained

• 4 major insurers integrated, including LIC of India (the largest insurer in the country - and the last major holdout to join)

How CAMS gets paid: Annual maintenance fee per eIA account maintained, roughly similar to how it earns per folio in the MF business.

The regulatory tailwind: IRDAI is pushing mandatory e-policy issuance. When this becomes compulsory:

Growth levers and metrics to track:

• Policies onboarded per quarter: Leading indicator of future recurring revenue. As mandatory e-policy rules deepen, this number accelerates.

• New insurer integrations: Each new insurer brings its entire policy book to CAMSRep. LIC’s integration alone was a step-change because LIC holds hundreds of millions of policies.

• IRDAI regulatory progress on mandatory e-policies: The policy trigger for this business to go from good to exceptional.

• CAMSRep revenue growth: 19% YoY in 9M FY26. Still early-stage but accelerating. Revenue impact is proportional to policies maintained - every new policy onboarded creates a permanent annual fee.

4. CAMS KRA - KYC Registration Agency

We all have done this. In india for everything, you need to complete a KYC (Know Your Customer) process. This involves verifying their identity, address, PAN card, and bank details.

Now rather than everyone doing it themselves, we’ve centralised it via KRAs. Do your KYC once with a SEBI-registered KRA, and every other financial institution can fetch that verified record instead of re-running the same check.

Note: CAMS recently acquired the NSE Dotex KRA business, significantly expanding its KYC infrastructure and client reach

How CAMS gets paid: Per-registration fee for new KYC records created. Per-verification fee when an existing record is fetched. Growing investor base = growing number of KYC events = growing revenue.

Growth levers and metrics to track:

• Number of PAN-linked KYC records maintained

• New investor onboarding volumes

• Non-MF KRA clients: More brokerages = more KYC events = more revenue, specifically from outside the MF ecosystem.

The MF business is the cash cow. The non-MF businesses are the growth engine. Both are compounding simultaneously.

Why You Need Both: Growth and Moat in the Same Name

Why You Need Both: Growth and Moat in the Same Name

One fundamental truth: stock price growth is a slave to earnings growth. No earnings, no compounding. So you need a business that grows.

But growth alone isn’t enough. Fast-growing businesses attract competition. Competitors undercut on price. Margins compress. Revenue goes up, profits don’t follow. You own a treadmill, not a compounder.

This is why the moat matters. It’s what stops competition from eating the margin. Growth without a moat is just market share waiting to be stolen.

CAMS has both. Growth that compounds. Margins that hold. Both in one name.

The Moat: Why Nobody Is Taking This Business

1. Regulatory protection: SEBI licences RTAs. The bar to get one is extremely high: capital, proven infrastructure at scale, regulatory relationships, and a compliance track record. No new significant RTA has entered in over 20 years. More importantly, SEBI doesn't want new players here. . The regulator wants large, stable, proven incumbents. Its instinct is to protect them, not disrupt them.

2. Switching costs: Let’s go back to the wedding reception example. 500 guests seated, food mid-service. The host decides to switch caterers right now, while guests are eating.

Switching an RTA mid-operation is harder.

An AMC moving away from CAMS must:

• Migrate crores of investor folios with full transactional history

• Rebuild SIP mandates for millions of investors, touching their bank accounts

• Reconfigure distribution systems for thousands of distributors

• Re-file regulatory documentation with SEBI

• Manage investor communication throughout

• Accept full liability for every error during migration

My profound question to you is: Kaun karega bhai itna matha-pacchi? ( translated: too difficult to do). One error touches a real investor's money. AMC executives have careers on the line. CAMS has never lost a major AMC to KFin. The lines are drawn and they stay drawn.

3. Gravity of scale: When a new AMC launches, they almost always go to CAMS. Not because CAMS is cheaper, but because handing a mission-critical function to a smaller, less-proven operator carries career risk. CAMS retained 82% of NFO market share in FY25. The more investors already sit on CAMS’s platform, the easier it is for a new AMC to reach them. Network effect layered on top of switching cost moat.

This is not theoretical. I’ve worked as a product manager handling technology integrations. This is what actually happens.

4. Systems integration creates cross-sell gravity: Once CAMS is embedded in an AMC’s stack for RTA, every adjacent need, payments, KYC, alternatives servicing, defaults back to CAMS.

Say a product manager wants UPI AutoPay for SIPs. They’ve 2 options:

• Option A: call CAMS. Legal agreements done. Folio data already there. Live in weeks.

• Option B: new vendor. IT security clearances, data-sharing agreements, infosec audits, integration testing, stakeholder approvals across IT, legal, compliance ( Sounds like breathless 3.0). This really is months of work. It also has real risk of errors mid-operation.

What will a PM choose? He asks himself the same profound questions (Kaun karega bhai itna matha-pacchi?) Option A wins every time. Not because CAMS is better at payments. Because Option B is painful and no product manager volunteers for it. CAMSPay growing at 41% is not a coincidence. It is a direct consequence of this dynamic (This is also called client-mining by the way)

CAMS’s cost base is people doing repetitive, rule-based work: transaction processing, query handling, compliance reporting, reconciliation. These are textbook AI automation targets. CAMS is already running a multi-year Google Cloud migration and has acquired Think360.ai for exactly this. As AI handles more volume per rupee of headcount, costs grow slower than revenue, operating leverage kicks in, and margins expand. For IT sector, AI compresses revenue. For CAMS, it compresses cost per transaction while the revenue line holds. Different business, different outcome.

Growth: The Numbers and the Runway

The natural floor. CAMS’s MF revenue grows with mutual fund AUM. Not one-to-one, because SEBI’s telescopic pricing compresses fee yields slightly as AUM scales. But as India’s MF AUM expands from Rs 82 lakh crore today toward Rs 200 lakh crore by 2030 (Bain Research projection), CAMS’s fee base grows materially even without a single new client win.

New AMCs keep arriving. Six new AMCs went live on CAMS in the last 8 months. JioBlackRock is now active. Every new AMC launch defaults to CAMS for the reasons above. Each adds permanently to the folio and transaction base.

Non-MF at 26% CAGR. Non-MF revenue has grown from 9.8% of total five years ago to 14.5% today. Management targets 20% by FY27-28. Three separately growing businesses feeding into a single P&L.

Management guidance. 5-year CAGRs of 17% for revenue and 18% for EBITDA and PAT from the FY21-26 base. Non-MF specifically guided at 20%+ annual revenue growth with EBITDA margins improving to ~25% within two years.

Before moving on, I wanted to give you something that I discovered and that truly amazed me. A lot of readers also wrote to me asking if brokers and excahanges were better bets on capital markets in India. Here’s what I think…

Cyclicality: Why This Matters and Why RTAs Are the Least Affected

Indian capital markets are cyclical. Every investor needs to understand it before making a valuation call.

The history is instructive:

Now, how does each infrastructure segment behave through these cycles?

• How each segment earns, and why it matters in a downturn:

  • Brokers earn on every trade placed. No trading, no revenue. When markets fall and volumes dry up, their top line falls with it. SEBI’s F&O crackdown made this worse: Zerodha’s brokerage revenues dropped 40% in a single year.

  • Exchanges earn on every transaction that flows through their platform. Slightly more insulated because they also earn from data and other services, but transaction volumes are still their biggest revenue driver. When markets go quiet, they feel it too.

  • Depositories earn two ways: a fixed annual charge per demat account that comes in regardless of market conditions, and a transaction fee every time shares move. The annual charge holds in a downturn. The transaction fee shrinks. Partial protection, not full.

Verdict: All three feel the cycle. The degree varies, but none are immune.

RTAs are the least cyclical of all. Here’s why:

• Folio base doesn’t move. Per-folio annual maintenance fee accrues whether the investor transacts or not. 4 crore investor accounts don’t disappear when Nifty corrects 16%. This is the floor.

• SIPs don’t life. History shows this consistently: when sentiment turns, investors stop writing fresh cheques but keep their standing instructions running. Lump-sum flows dry up. SIPs hold. That standing instruction is CAMS’s revenue.

• Bull markets overcompensate. When markets recover, lump-sum flows surge, new AMC launches accelerate, new investor onboarding spikes, and CAMS’s transactional revenue runs well ahead of the base. Downside shocked. Upside rocked ( terrible meme reference).

The net result: Of all the equity infrastructure segments, RTAs have the most defensive revenue profile in a downturn and the most leveraged profile in an upturn.

Leave a comment

Valuation: The Right Framework

Why simple PE misleads here

CAMS trades at approximately 40-42x TTM earnings. That looks expensive. But these are trough earnings during a cyclical slowdown. The question is not what the PE is today. It’s what normalised earnings look like and what multiple this quality of business deserves. Let’s find frameworks to evaluate this..

The spawner framework: why the market will rerate this stock

Here is the valuation argument most analysts miss.

In a standard DCF, terminal value ( what the business is worth beyond the forecast period ) accounts for 60-80% of total valuation. Near-term cash flows matter, but the lion’s share of value comes from perpetuity assumptions.

When analysts model CAMS today, the base case has the MF RTA business compounding at 15-17% per year. That gives you a fair price around Rs 3,500-4,000.

But when the non-MF businesses start contributing 20-25% of revenue at 26%+ growth, the analyst has to rebuild the model. They’re no longer pricing a single-segment MF infrastructure company. They’re pricing a financial infrastructure platform with multiple high-growth legs. The terminal growth rate assumption in the DCF moves up. Even a 1-2 percentage point increase in terminal growth assumptions inflates present value dramatically. The multiple expands. The stock reprices.

This mechanism has a name: the spawner rerating.

Bajaj Finance is the cleanest Indian example. It started as a two-wheeler lender. A decent business at a modest multiple. Then it added consumer durables financing, then personal loans, then deposits, then wealth management. At each stage, analysts upgraded their terminal value assumptions because the addressable market kept expanding. We all know what the stock did. Earnings grew, yes. But the multiple also expanded significantly as the market accepted this was no longer just a two-wheeler lender.

CAMS is at a similar inflection. The market prices it today primarily as an MF RTA. CAMSPay is growing at 41% and contributes 19% of revenue in the non-MF bucket. CAMSRep has LIC on board and 13 million policies. None of these are small anymore. None are fully priced into the current terminal value.

The moment one of these reaches Rs 400-500 crore in annual revenue with clear visibility to Rs 1,000+ crore, analyst models change permanently. You don’t need all three to work. You need one to get large enough that the market can no longer ignore it. That window is currently open.

Risks worth watching:

• TER compression: SEBI keeps pushing AMC fees lower. Lower fees mean AMCs earn less per rupee of AUM, and eventually renegotiate RTA fees downward. CAMS’s revenue per unit of AUM has been compressing gradually. Volume growth has offset this so far. Watch it every quarter.

• AMC concentration: Top 5 AMC clients contribute roughly 65-67% of CAMS’s MF revenue. An AMC merger reduces the client count and gives the surviving entity more negotiating leverage on fees.

• Near-term margin pressure: CAMS is investing in non-MF businesses while absorbing annual salary increases. EBITDA margins dipped in recent quarters. Temporary, but the stock reacts to every miss.

• Regulatory risk on RTAs: SEBI could mandate changes to the RTA fee structure, as it has done with TERs for AMCs. Low probability, but not zero.

• Non-MF execution risk: The spawner thesis only works if at least one of CAMSPay, CAMSRep, or Alternatives scales to meaningful size. These are early-stage businesses. Most adjacent bets by incumbents fail. CAMS has better odds than most given its distribution, but it is not guaranteed.

Gaurav is a fantastic writer and has an uncanny ability to find hidden microcaps

Check out

Conclusion:

CAMS is a toll booth on India’s savings migration. It earns per folio, per SIP, per transaction, without taking a rupee of market risk. The moat is unbroken in 20 years. The market is growing for the next 20. Three new businesses are scaling on the same base, and at least one of them will shift how analysts value this company. Current earnings are trough earnings. The PE you see today is not the PE you’re actually paying.

SEBI Disclaimer

This newsletter is for informational and educational purposes only. Nothing here constitutes investment advice, a buy/sell recommendation, or a solicitation to invest in any security mentioned. I am not a SEBI-registered research analyst or investment adviser. All views are personal. Do your own research before making any investment decision. Past performance of any stock or sector is not indicative of future returns. Equity investments are subject to market risks. Read all scheme-related documents carefully before investing.

Every decade or so, a shift in how an entire country allocates its savings creates a multi-decade wealth transfer. India is in the middle of one right now. The question isn’t whether to participate. It’s where in the food chain to stand.

Follow the Tailwind, Not the Stock

The fastest way to make serious money in markets isn’t better stock picking. It’s being in the right industry at the right time.

Look at what happened to investors who got into structural tailwinds early:

• Indian IT, late 1990s: Mediocre management, average product, irrelevant whether you picked Infosys or a second-tier player. Y2K + global offshoring = decade-long compounding machine.

• Private sector banks, post-2003: Public sector banks were sleepy. Private banks were building distribution, going retail, and riding India’s credit growth. Owning the sector mattered more than picking the best bank.

• Telecom infrastructure, Jio era: The moment Reliance announced Jio in 2016, anyone holding tower companies or fibre infrastructure was sitting on a structural tailwind that lasted years.

And, full disclosure: yours truly identified the defence sector theme early, made a bomb on it, and was on the verge of changing his last name to Buffett. (The companies I picked were terrible. The tailwind carried them anyway. But shhh, we don’t talk about that part.)

The pattern is simple:

Identify value migration → Find businesses that sit in its path → Size up → Wait

The tailwind does the heavy lifting. Your job is to not get off the train early.

India has one of these shifts running right now. Multi-decade. Structural. Already underway but still early. It’s called the financialisation of savings.

Mom, Dad, and That One Uncle

For generations, the Indian household playbook looked like this:

• Mom: Gold. Every occasion, every windfall. Gold. ( After this year’s rally: Mom - 1, Beta - 0)

• Dad: Fixed deposit and land. “Safe, steady, tangible.”

• That one uncle or mama: Invested in the stock market. Lost everything. Was right about some stock “in 1992.” Still brings it up at weddings.

That playbook is changing. Partly because markets have delivered real wealth to enough people. Partly because smartphones made investing frictionless. And partly because a few bull runs have done serious rehabilitation work for those uncles. They’re no longer the cautionary tale at family dinners. Some of them are now the ones people are asking for tips.

The data confirms the shift is real and it’s huuuuuge:

• Demat accounts: 4 crore in FY20. 21 crore by late 2025. Fivefold in five years. (NSDL/CDSL)

• SIP monthly inflows: Under Rs 8,000 crore in FY18. Rs 31,000 crore in December 2025, an all-time record. (AMFI)

• Mutual fund AUM: Rs 24.5 lakh crore in March 2019. Rs 82 lakh crore by end-2025. Roughly 3.3x in six years.

• Equity fund inflows: 58 consecutive months of positive flows as of December 2025. Not a single month of outflows for nearly five years.

Now the part that makes this a long-duration structural bet and not a bull-market story:

India’s mutual fund AUM to GDP ratio is 16%. Global average: 63%. US: 140%. China: 21%. Only 7% of India’s population is currently in the securities market (CDSL Q3 FY25 earnings call).

You don’t need a model to tell you what happens when that 7% moves toward 20% or 30% over the next two decades.

The investor’s question: which businesses are permanently enriched every time another Indian household opens a demat account, starts a SIP, or moves savings from a fixed deposit into equity? Those businesses exist. They’re listed. And some of them are still reasonably priced.

The Plumbing Nobody Talks About

Before picking winners, you need to understand the landscape. Every rupee that moves from a retail investor’s bank account into the market passes through several hands. Each takes a cut. Some cuts are large and growing. Some are being squeezed by competition. Some are structurally protected by regulation and scale

.

Here’s how the chain works:

Investor → Broker → Exchange → Depository → AMC → RTA

Each node in this chain is a different business with a different moat and a different relationship to the tailwind. Let’s break them down properly.

1. Brokers

What they do: The front door. Zerodha, Groww, Angel One, ICICI Direct. They sit between you and the exchange, onboard you, execute your orders, and hold your hand (or don’t, if you’re on Zerodha). Think of them as the shop floor where investing actually happens for the retail investor.

How they earn:

• Brokerage commissions on trades

• Distribution fees on mutual fund sales

• Margin lending (MTF)

• Subscription fees.

Examples: Zerodha, Groww, Angel One, ICICI Direct, Upstox.

Nature of competition: Brutal. Zerodha invented zero brokerage on delivery. Groww copied it. Everyone else followed. SEBI then killed the exchange rebate model (October 2024) that was quietly subsidising this zero-brokerage offer. The industry’s F&O revenue, which was funding all the free stuff, fell off a cliff. Zerodha’s brokerage revenues dropped ~40% YoY in Q1 FY26. Angel One saw ~25% revenue decline. Groww held better but still felt it.

The competitive dynamics are the definition of a race to the bottom. Every time someone finds a revenue line, regulation or competition compresses it.

2. Exchanges

What they do: The marketplace. NSE and BSE are where buyers and sellers get matched. Every trade in the Indian market flows through one of them. Think of them as the stadium where the game is played: whether the home team wins or the visitors do is irrelevant, the stadium still earns.

How they earn:

• Transaction fees per trade (the primary line)

• Listing fees from IPOs

• Data licensing

• Index licensing (NSE licenses Nifty data globally)

• Newer platform revenues like BSE StAR MF (mutual fund transaction platform).

Examples: NSE (private, not listed), BSE (listed, BSE Ltd).

Nature of competition: NSE dominates. 93% of cash equity, 97% of equity options, nearly 100% of equity futures, all of it routes through NSE (Opensourceinvestor Substack, 2025). BSE has been staging a recovery, gaining options market share, and has a genuinely interesting moat in its mutual fund platform. But NSE is the giant and it’s not listed. The investable listed play here is BSE.

3. Depositories

What they do: The vault. Not for money, for securities. When you own shares, they don’t physically sit anywhere. They live as electronic records in a depository. CDSL or NSDL holds your shares the way a bank holds your money. Every demat account in India lives inside one of these two.

Who pays them: Issuers (companies that list), brokers (who are Depository Participants, or DPs), and investors indirectly through annual maintenance charges. Every new demat account generates recurring AMC revenue. Every IPO generates dematerialisation charges. Every KYC verification (through CDSL’s subsidiary CVL) generates fees.

How they earn:

• Annual maintenance charges per demat account

• Transaction settlement fees

• Issuer charges for IPO dematerialisation

• KYC/data services through subsidiaries.

Examples: CDSL (listed), NSDL (listed August 2025 via IPO).

Nature of competition: Duopoly. Legally and structurally. SEBI governs entry into this market. No new depository has been permitted in nearly three decades. CDSL has ~80% of all demat accounts (October 2025: 167.7 million of 210 million total). NSDL dominates in value of assets under custody (Rs 500+ lakh crore vs CDSL’s Rs 42 lakh crore), because it custodies institutional and HNI accounts.

They don’t really compete in any normal sense. Brokers choose one or the other when they become a DP, and investors follow their broker. Switching is possible but rare.

4. AMCs (Asset Management Companies)

What they do: Pool money from investors and manage it. An AMC runs mutual funds across equity, debt, and hybrid categories. It decides what to buy, when to buy, and charges you for the privilege.

How they earn: TER (Total Expense Ratio, the annual management fee expressed as a percentage of AUM). More AUM = more fee, automatically, every day, whether markets go up or down (fund managers chant do AUM chanting daily). Equity TERs run upto 2.25%. Passive/index funds run much lower. The fee accrues daily. It’s a royalty model on a growing pile.

Examples: HDFC AMC, SBI MF, ICICI Prudential AMC, Nippon India AMC. Top 5 AMCs control over 55% of industry AUM.

Nature of competition: High barriers to entry (SEBI licensing, brand trust, distribution network). But the competitive dynamics within the sector are intensifying. SEBI has been compressing TERs over the years. Passive investing is growing globally and starting to gain ground in India. Any AMC that relies purely on active fund performance faces existential questions over 10-15 years as more investors realise most active funds underperform their benchmark post fees.

5. RTAs (Registrar and Transfer Agents)

What they do: The invisible engine room of the mutual fund industry. Every SIP processed, every redemption handled, every folio created, every dividend paid out - that’s the RTA’s job. The AMC is the restaurant that decides the menu and serves the guests. The RTA is the kitchen: cooking the dishes, managing inventory, counting covers, processing payments, handling complaints, and keeping everything running without the guest ever knowing it exists.

Spoiler alert: We’ll go much deeper on RTAs in Part 2 since this is where the most interesting investment logic sits. For now, the essentials:

How they earn:

• Per-folio annual maintenance fees

• Per-transaction processing fees (every SIP instalment, redemption, switch). Higher equity AUM = higher per-folio fee. More transactions = more revenue. Pure volume play with near-zero marginal cost at scale.

Examples: CAMS (listed, ~68% market share by AUM), KFin Technologies (listed, ~32% market share).

Nature of competition: Duopoly, even more concentrated than depositories. Each AMC has an exclusive RTA contract. Once an AMC is integrated into an RTA’s systems, switching is operationally brutal. Data migration, system reconfiguration, investor communication, regulatory filings, the whole thing. It doesn’t happen. CAMS services 4 of the top 5 AMCs. Their ROCE is north of 50% (Foreshadowing of what is coming).

A Note on Misc Players

The chain also includes vendors who service the chain itself: KYC platforms (Digio, CAMS KRA), payment processors, data analytics firms, and compliance tech companies. These are mostly private or embedded inside listed entities. Not our focus here, but worth knowing they exist as part of the ecosystem.

I missed THE MOST IMPORTAT SEGMENT: Substack writers - They put a lot of effort but make no money ( To be read as a desparate call to subscribe or upgrade to paid)

Industry Grows. But Who Keeps the Value?

Here’s a key framing for what comes next.

Industries decide growth. They don’t decide who keeps the profits. Competitive structure does.

India’s equity infrastructure will grow for decades. That’s close to certain. But an industry growing doesn’t mean every player in it compounds wealth for shareholders. Airlines grow. Airlines destroy capital. Telecom grew. Most telcos were terrible businesses. The difference is always the same thing: moat.

So the right question isn’t “which segment benefits from financialisation?” All of them do. The question is: which segment has a moat strong enough to keep competitors from taking the profit away, and a market large enough to keep growing into for the next 15 years?

For that, we use Porter’s Five Forces, applied to each segment. Not as an academic exercise. As a tool to eliminate the weak and find the durable. Elimination works because Munger said:

“All I want to know is where I’m going to die. I’ll never go there”

Porter’s Five Forces: Segment by Segment ( Don’t worry if you dont know who Porter is or think its an app to deliver goods on your phone)

Brokers

Verdict: Weak structural position. High growth industry, but profits keep getting competed away or regulated away. Zerodha’s brokerage revenues dropped 40% in a single year. Angel One down 25%. This is not a business where you can sustainably earn high ROCE over a decade. Eliminated.

Exchanges (BSE, since NSE is private)

Verdict: Strong structural position but the crown jewel (NSE) is unlisted. BSE is interesting, especially its mutual fund platform (BSE StAR MF handles a large share of MF transactions nationally). But BSE’s revenue is exposed to trading volumes, which are cyclical and partly dependent on derivatives activity SEBI is currently trying to suppress. Interesting, but conditionally.)

Depositories (CDSL, NSDL)

Verdict: Exceptional structural position. Legally mandated usage, zero substitutes, regulated duopoly, structural volume growth as India adds 2-4 crore demat accounts annually. CDSL’s ROCE is ~30% (three-year average per Screener), profit CAGR of 37.8% over five years. Revenue has two parts: transactional (volatile, tied to market activity) and structural (annual maintenance per account, grows even in bear markets as the account base accumulates).

The structural revenue piece is what makes CDSL a genuinely different animal from brokers or exchanges. Survives the cut.

AMCs

Verdict: Good but not clean. AMCs are a royalty model on growing AUM and that’s a beautiful business. But TER compression is real and structural. SEBI has been pushing fees down for years and that trend continues. Large AMCs like HDFC and ICICI Prudential are better positioned than smaller ones because scale offsets compression. The passive fund shift, still nascent in India, is the 10-year risk.

AMC stocks trade at 35-45x earnings. For the right price, they compound well. But the moat has a leak. Survives the cut, with a watch on TER trajectory.

RTAs (CAMS, KFin)

Verdict: The strongest structural position of any listed equity infrastructure player. CAMS’s ROCE is 50%+. Five-year revenue CAGR of ~15%. Profit CAGR of 23%. EBITDA margins of 44-46%, consistent. Debt-free balance sheet. The revenue model is: more mutual fund investors = more folios = more transactions = more fee, automatically. No market risk. No directional bet. Every SIP that gets processed, CAMS earns. Every folio that sits dormant, CAMS earns an annual maintenance fee.

The tailwind is not just the market going up. It’s the sheer volume of India’s savings migrating into mutual funds, regardless of returns. CAMS doesn’t care if Nifty is at 24,000 or 18,000. It cares about folio count and transaction volume. Both of those only go one direction over a 15-year horizon.

Strongest candidate. Going deep in Part 2.

The Elimination Round

So where does that leave us?

Three segments survived the cut.

Part 2 goes deep on one. You pick which. The segment with the most votes gets the full treatment: ROCE history, growth runway, risk flags, and a valuation view. Vote below.

Part 2 is out:

Part 2: CAMS the monopoly

If this was useful, you know what to do.

Writing takes time. Researching takes more. A subscribe costs nothing and tells me there are real humans on the other end.

Subscribe now

A restack gets this in front of someone who might actually need it. A like at least tells the Substack algorithm I exist.

Share vaibhav tambade

I am not above begging. Clearly.

Leave a comment

Also Megatrend investor writes incredible stuff, do check it out

Megatrend Investor

SEBI Disclaimer

This newsletter is for informational and educational purposes only. Nothing here constitutes investment advice, a buy/sell recommendation, or a solicitation to invest in any security mentioned. I am not a SEBI-registered research analyst or investment adviser. All views are personal. Do your own research before making any investment decision. Past performance of any stock or sector is not indicative of future returns. Equity investments are subject to market risks. Read all scheme-related documents carefully.

Subscribe now

This week, I want to connect some dots that I think most people are missing.

Because here’s the thing, most of what you’re watching on the news right now looks like separate stories. Russia. Venezuela. Greenland. Iran. They feel disconnected. They’re not.

There’s one thread running through all of it. And once you see it, you can’t unsee it.

The Frame: It’s a Two-Horse Race

The world today is being shaped by one overarching rivalry: the US vs China race for global dominance. And the primary battleground isn’t military. It’s not even trade. It’s AI.

Look at the American stock market. The entire rally over the last few years has essentially been carried by 7 companies. The Magnificent Seven. AI is literally underwriting the US stock market’s valuations. China is a few steps behind, but closing the gap fast. DeepSeek rattled the world for a reason.

Now here’s the question that unlocks everything else: what does AI run on?

Energy.

Not just in the abstract sense. Training large models, running inference at scale, powering the data centres - it is a colossal, growing, ungodly amount of energy. The leader in AI tomorrow will be the nation that can supply the cheapest, most abundant energy to power it. And the US knows this.

So when you see Trump doing something that looks random or chaotic on the foreign policy front, run it through this filter first: is this about energy?

Nine times out of ten, the answer is yes.

The Playbook: Five Moves, One Strategy

Let’s go through them one by one.

Move 1: End the Russia-Ukraine War

Why is Trump willing to sideline NATO, risk alienating European allies (and commenting on President Macron’s wife’s jab on his face, yikes!), and eat political heat at home just to end this conflict quickly?

Nobel Peace Prize, sure. But the second reason is more important:

End conflict → Russian gas re-enters global markets → Global energy prices fall

Russia was the biggest gas supplier to Europe before the war. Sanctioned, cut off, its energy locked out of Western markets. The US wants those supply taps reopened because more global supply means lower prices. Lower energy prices = cheaper AI training = US wins the AI race on cost.

Move 2: Venezuela

Venezuela sits on some of the largest proven oil reserves in the world. Problem: it was sanctioned, Maduro was selling to China at discounted rates, and that oil never hit open markets.

With the political takedown of Maduro, the US effectively gets to redirect that supply. One move, a massive amount of oil shifts from China’s discounted basket to global market pricing.

Move 3: Greenland (and the Arctic, the real story)

This one gets written off as Trump being eccentric. It isn’t.

Here’s what’s actually happening in the Arctic:

• Global warming is melting glaciers and opening up the Arctic Ocean

• New trade routes between the western and eastern hemisphere are becoming viable, routes that simply didn’t exist before

• Sitting under all that ice? Roughly 20% of the world’s undiscovered oil and gas reserves

• US is not technically an Arctic nation

Greenland changes that. If the US gets Greenland (and also nudges Canada makes Canada its 51st state), it suddenly has skin in the Arctic game: access to the shipping lanes, proximity to those reserves, and a geopolitical footprint in a region that’s going to matter enormously in 30 years. You can’t keep Uncle Sam far from oil for long.

Move 4: The Petrodollar OPEC Layer

OPEC countries were already in the US orbit via the petrodollar arrangement. Gulf states price oil in dollars, recycle those dollars into US Treasuries, everyone is happy. That system has held for decades. To understand this better, read my article on the petrodollar.

The Petro-Dollar

But here’s what happened with the sanctions policy on Russia, Iran, and Venezuela: the US inadvertently handed China a massive gift. China, with its economic weight, stepped in and bought cheap discounted oil from all three sanctioned nations. Instead of isolating them, the US actually funded China’s energy security at below-market rates. Classic own goal.

Now the US wants to unwind that. End the conflicts, bring those supplies into open markets, end China’s access to the discount bin.

Move 5: Iran

Iran is the most combustible piece of this puzzle, so it deserves its own section (Which we’ll come to in a minute ).

The Own Goal That Changed Everything

Take a step back. OPEC countries (already in the US orbit through the petrodollar ) were not the problem.

The problem was the next tier of big producers: Russia, Iran, Venezuela. All three were sanctioned by the US. All three were locked out of global markets.

Now here’s the question I want you to sit with for a second: if these countries couldn’t sell to the West... who did they sell to?

China.

China, with its economic muscle, walked in and bought discounted oil and gas from all three sanctioned nations. Iran. Russia. Venezuela. The very countries the US was trying to isolate became China’s private discount energy aisle. The sanctions didn’t hurt China. They handed China a “Trump” card. Cheap energy, at scale, for years.

Think about what that means. The US, trying to squeeze its adversaries, inadvertently subsidised China’s AI ambitions. Every discounted barrel of Iranian crude going into a Chinese refinery was essentially the US paying for its own defeat in the AI race.

That is the sting in the tail. And fixing this blunder is exactly what’s driving the next move.

The Iran Chapter: Why This One is Different

The US attacked Iran’s oil infrastructure. The stated reasons are many. The real ones are two:

1. Iran’s oil and gas infrastructure represents a large, non-US-controlled energy asset

2. The Strait of Hormuz - the narrow chokepoint through which 7% of global energy supply flows every day - was at risk of Chinese influence

China had been making heavy inroads into the Middle East. A $400 billion, 25-year cooperation treaty with Iran. Military port access in the region. This was strategic encirclement of US energy dominance.

Uncle Sam couldn’t let that stand. So they attacked. And now here’s where it gets interesting..

Few characteristics of this war that you have to keep in mind:

1. It’s not about who has more firepower. It’s about who can absorb more damage.

Iran is fighting for its survival — not just for this year, but for the next few decades. Survival of the government. Security guarantees. The whole thing. When your existence is on the line, you can justify an enormous amount of pain. The US, on the other hand, is a democracy. Trump is already on thin ice for even being in this war. Public tolerance for casualties, cost, and chaos is limited. Iran can keep going a lot longer than Washington’s political capital allows.

2. Asymmetric warfare - you heard this a million times already, but let me repeat it one last time.

Iran’s cheap $1,000 missiles keep flying as long as someone presses the button. The cost of the defensive systems the US and its allies deploy ( THAAD, Iron Dome ) is 100x higher per intercept. Iran is literally bleeding the US defense budget one cheap drone at a time.

3. Here’s the one that most people miss: the US is in a war with Iran, but Iran is in a war with the American market.

Iran doesn’t need to destroy a carrier group. They just need to keep pounding until the market starts to panic. Market panic brings Trump to his knees. Imagine the phone call: “Sir, please give me a deal.” ( Trump reference if you didn’t get it) That’s the play. Every 2% drop in the Dow is a small win for Tehran.

4. They’re also bombing the daylights out of US allies in the region.

This isn’t for fancy military objectives, it’s financial. Every ally that gets hit calls Washington and applies pressure to get a deal done. The petrodollar system depends on Gulf stability ( A lot of dollars go back into US bonds and the Mag 7). Destabilise the Gulf, you destabilise the entire arrangement.

5. Israel has no incentive to stop.

This is Netanyahu’s dream scenario. An American president willing to fight Iran, after all these decades of waiting. A wounded Iran that survives is actually more dangerous than one that’s finished (and establish Akhand Israel - greater Israel). Israel wants to finish the job while Uncle Sam is still in the mood. They are not going to pump the brakes.

6. China is watching, not acting. For now.

China is deeply tied to global trade. Much of its growth is export-oriented. A full global recession is bad news for Beijing. But right now? They’re enjoying the show. The US is exhausting itself, fracturing NATO, alienating allies. China is letting it play out to maximum humiliation. At some point though, when it risks tipping into actual global depression, China will quietly apply pressure on Iran to step back. Not yet. But that moment is coming.

7. Russia is cashing out.

Sanctions being eased. Oil that was locked up is flowing. Every extra month of this conflict is money in the bank for Moscow. They are not in a hurry for peace.

8. When does the war end?

No one knows. The market thinks it ends every week, catches a 2% rally, then realises it hasn’t - and hands back 4% the next day. ( The poor retail investor can be heard crying).

Here’s the actual logic: war is just politics by other means. It’s what happens when parties can’t agree at the negotiating table, so they fight until the stakes shift enough that a deal becomes possible. War ends when both leaders can go home and call it a win.

Right now, those conditions don’t exist. Iran wants US military bases gone, Israel contained, and its security guaranteed for decades. Not just a ceasefire. A permanent arrangement. Israel wants Iran dismantled. The gap between those two positions is enormous.

My read: the US gets forced to the table first. Markets are already feeling it. If the bombing continues, Iran calls in a favour with their homies - the Houthis to resume attacks on Red Sea shipping ( Boy have we seen that movie already happened). If that restarts at full scale, boy are we in trouble - even the athiests will pray.

And here’s what the market keeps getting wrong: this war will not end quickly because Iran doesn’t want it to end quickly. Iran wants to extract maximum pain to ensure their survival not just this year but for decades. They will not stop until deterrence is established. The conditions for a deal aren’t there yet. And when eventually the big daddies ( Daddy Trump and Daddy Xi ) decide enough is enough and apply pressure on their respective jockeys, that’s when it winds down. Watch for that signal.

The Hormuz Math: Why This Is Not a Drill

Let me give you the numbers clearly.

7% of global energy flows through the Strait of Hormuz. That’s the combination of oil (20% of global oil supply) and gas (25% of global LNG).

If the Strait is disrupted for an extended period:

• Even substituting 2% from alternate routes, you still have 5% of global energy supply wiped out

• 5% of energy supply = roughly 5% of global GDP gone, directly

• And that’s before the second-order effects

How the Pain Travels

Oil and gas are not just fuels. They are raw materials and inputs for a staggering range of industries. A disruption here doesn’t stay in the energy sector. It radiates outward, wave by wave, with a lag. Let me walk you through the chains.

Chain 1: The Inflation Chain

Oil price rises → transport costs rise (every truck, ship, flight) → logistics cost rises for every product in every supply chain → manufacturing input costs rise (plastics, chemicals, fertilisers are all oil derivatives) → companies either absorb margin hits or pass costs to consumers → consumer prices rise → CPI inflation spikes → central banks face pressure to raise rates → credit becomes expensive → investment slows → GDP growth falls

Numbers to pin down: per $10 increase in crude, India’s CPI rises 35-60 bps. We’ve moved roughly $40-45 above the comfortable baseline. Run that math yourself.

Chain 2: The Currency Chain

Oil priced in USD → India pays more dollars for the same oil → more dollar outflow → rupee weakens → weakened rupee makes all imports more expensive (not just oil, everything) → inflation gets a second round from the currency side → RBI has to choose: raise rates and kill growth, or let inflation run → either path is painful

Numbers: $10 increase in crude = $12-15 billion added to India’s annual import bill. CAD widens 30-40 bps per $10 increase. At $100/barrel sustained, CAD goes from 0.7% of GDP to 1.9-2.2% of GDP (ICRA). In a pessimistic scenario (Bernstein), rupee tests 110 to the dollar.

Chain 3: The Fiscal Chain

Government tells OMCs to not raise pump prices → OMCs absorb losses of $1.4 billion/month at $100/barrel → government backstops OMCs → fertiliser subsidies rise by Rs 200 billion → total government expenditure rises by Rs 3.6 trillion in FY27 → fiscal deficit target of 4.3% becomes very hard → government has to cut infrastructure capex to compensate → growth slows structurally

Keep this number: $110/barrel is the threshold where the government’s fiscal buffers run out and retail price hikes become inevitable. Below that, they can manage. Above it, it becomes a consumer-felt crisis.

Chain 4: The Demand Destruction Chain

Higher fuel and food prices → household disposable income compresses → consumer spending falls → discretionary sectors — retail, durables, autos, hospitality — see volume drops → companies cut capex → employment growth slows → consumption contracts further → GDP growth takes the hit

SBI Research estimate: $100/barrel sustained for a year drops India’s GDP growth from the 7%+ trajectory to ~6.6%.

Chain 5: The Investment Freeze Chain

The Lag Problem: Why Peace Tomorrow Doesn't Mean Recovery Tomorrow

Here’s what the market keeps getting wrong every time it catches a ceasefire rally.

Even if the war ends tomorrow, the disruption doesn’t end tomorrow. Supply chain shocks travel forward in time like waves. The damage was done upstream. The pain shows up downstream, later.

Three specific lags to understand:

1. Fertiliser to food (3-6 month lag) Fertiliser supply has been disrupted. That affects what farmers sow right now. The shortage hits grocery shelves at harvest time — months from now. Your vegetable price is the last domino to fall, not the first.

2. Sea mines (45 day minimum) Laying a mine takes minutes. Clearing a minefield in a 34km-wide strait takes a minimum of 45 days, according to naval demining experts - and that’s if the work starts immediately and no new mines are laid. Until the Strait is fully cleared, insurance premiums for vessels on those routes stay elevated. Logistics remain choked even in “peace.”

3. Oil infrastructure restart (weeks to months) This one people really underestimate. The bombed refineries, damaged processing plants, and shut-in oil wells don’t just flick back on. Think of it like a restaurant kitchen that’s been shut down for months. It’s not just about unlocking the door — the pipes have corroded, the equipment has seized, the pressure systems need recertification, the staff has dispersed. An oil well shut in under adverse conditions needs pressure testing, equipment integrity checks, chemical treatments to clear blockages, and sometimes full workover operations before it flows again.

Rystad Energy has confirmed wells in the Gulf need weeks to months to restart. Qatar’s LNG facilities could take weeks just to get partially operational. South Pars, which has taken direct hits, is months away from normal output even in a best-case scenario.

The market prices the headline. It doesn’t price the lag. That’s the mismatch to exploit. Now we’ll see sector wise impact.

Sector Impacts: What History Tells Us, Applied to Today

We don’t have to guess what a major oil shock does to sectors. We’ve lived through this twice. The 1973 OPEC embargo quadrupled oil prices in months. The 1979 Iranian Revolution shock more than doubled them again. Both times, the pattern of winners and losers was remarkably consistent. Here’s what history says, mapped to today. Here’s a sectoral breakdown

Most Hurt

• Aviation : fuel is 25-26% of total operating costs (IATA 2025); at $100+ crude losses compound fast with nowhere to pass the full increase;

• OMCs (HPCL, BPCL, IOC) : government-directed to hold pump prices, absorbing $1.4 billion/month in losses at $100/barrel; accumulated buffer depletes in 3-9 months before price hikes become inevitable

• Fertilisers : natural gas is the primary feedstock; in 1973-74, fertiliser shortages fed directly into food crises across developing nations; same chain is activating now; subsidies balloon, margins collapse

• Paints and chemicals : crude derivatives are the core raw material; historically among the first margin casualties in every oil shock; no different now

• Cement and construction : energy-intensive kilns, diesel logistics, project deferrals; in 1979-80, US construction activity fell sharply as input costs spiked and credit tightened simultaneously

• Autos and consumer durables : demand destruction as household budgets compress; 1973 was the moment Detroit (USA’s erstwhile auto capital) started dying; today the demand hit is immediate and the input cost pressure is simultaneous

• Retail and discretionary : consumer spending the first behavioural casualty when fuel and grocery bills rise; happened in 1974, happened in 1980, happening now

• Shipping and logistics : insurance premiums for Gulf routes have spiked; Rystad confirms fields are shutting in because storage is full with nowhere to ship; working capital gets trapped in transit

Moderately Hurt

• FMCG : packaging (raw materials are petroleum derivatives) and logistics costs spike but demand is relatively inelastic; companies with pricing power absorb and pass through; took 2-3 quarters to show up in margins in past cycles

• Real estate : construction cost inflation delays projects and squeezes developer margins; luxury holds better than affordable as always in inflationary periods

• Banks and NBFCs : indirect hit through slowing credit growth; watch for NPA stress in MSME and agri lending if rural income compresses; in stagflationary environments rate hikes simultaneously hurt borrowers and slow credit book growth

Relatively Insulated or Benefiting

• Value retail and discount formats : this is a consistent, historically proven winner in every inflationary downturn; when household budgets compress, consumers trade down from premium to value. people don't stop buying, they just buy cheaper

• IT exporters : revenue in USD, costs in INR; rupee depreciation expands margins; genuinely insulated from the physical supply shock; true in every prior oil crisis for India’s export-oriented sectors ( But be careful of AI led crisis which is another beast in itself)

• Pharma exporters : same USD hedge; inelastic demand for medicine globally; no oil shock in history has ever crushed pharmaceutical demand

• Domestic upstream oil (ONGC, Oil India) : higher realizations hit the P&L positively and directly; analysts estimate Rs 57,800 crore combined PBT upside at $100-105/barrel; upstream producers were the single biggest wealth creators of the 1973-1980 period globally

• Renewable energy and EVs : both oil shocks of the 1970s permanently accelerated the shift away from fossil fuels; Japan pivoted to nuclear and efficiency tech post-1973 and never fully returned to oil intensity; the same structural acceleration is happening now, just with solar, wind and EVs as the destination

• Defence : geopolitical tension sustains government defence spending for years, not months; the 1970s crises kicked off a decade of elevated global defence budgets

• Gold : textbook flight to safety in every oil crisis; 1973 to 1980, gold went from $100 to $800 per ounce; context is different today but the direction is the same

• T&D infrastructure : the structural re-rating story I’ve written about separately; the energy war thesis I’ve laid out today is exactly the macro backdrop for that call; this is the sector that gets permanently repriced upward from this shock, just as nuclear and efficiency tech did post-1973

  Article on Shilchar Tech

  What Do You Do With All of This as an Investor?

  No stock tips here. What I want to give you is a way to think.

• First: ask who can’t cut their demand no matter what.

  In a supply-driven inflationary shock, the companies that survive best are the ones whose customers have no choice but to keep buying. No substitutes. Essential goods. Businesses with pricing power that can pass cost increases straight through to the customer without losing volume.

  • Avoid the other side of that trade: Companies with high import content in their costs, thin margins where a 2% inflation surprise wipes out the year, and long working capital cycles where cash gets trapped.

• Second: separate the trades from the structural re-ratings.

  Every big shock permanently reprices certain sectors upward. COVID did it for pharma, digital, home consumption. The sectors that get the permanent re-rating from this shock are energy infrastructure and T&D. I have written about why T&D shifted from cyclical to structural : the entire energy war thesis I laid out today is exactly the macro backdrop for that call. These are not 6-month trades. These are 3-5 year re-ratings. The investor who buys them in the chaos and holds will do a lot better than the one trying to time the news cycle.

• Third: stress test every holding against a simple scenario.

  Rupee at 95-100. Crude at $100 for 12 months. What happens to earnings?

  • Revenue in USD (IT exporters, pharma exporters): partially insulated, actually benefits from rupee depreciation ( but they’re facing other complications like AI so be careful)

  • Input costs in USD, revenues in rupees (importers, energy-intensive manufacturers): hurt, meaningfully

  • Government as a major customer: be careful - the government is under serious fiscal pressure, receivables can stretch

  Run every holding through that filter. The ones that fail it, size down.

The Mental Model to Keep

Here’s what I want you to walk away with.

The next time a geopolitical headline crosses your feed, before you do anything with it, run it through this question: who does this help or hurt in the global energy chessboard?

That single filter will make you right more often than not about what’s actually happening and why.

The US is running an energy dominance strategy to win the AI race. Every move( Russia, Venezuela, Greenland, Iran) is a piece on that board. China is playing defense and running its own energy independence game (they added 100 GW of coal power in a single year while simultaneously leading in solar and wind buildout, your American grid, by contrast, is 50 years old and hasn’t had a serious upgrade).

The war for tomorrow is being fought today. And it’s not being fought with tanks.

It’s being fought with pipelines, shipping lanes, data centres, and the ungodly amounts of electricity needed to power them.

The investor who understands that doesn’t just understand geopolitics. They understand the next decade of markets.

If you found this useful, please share it. One tap gets this in front of someone who needs to read it.

Share vaibhav tambade

Drop a comment below. What did you take away from this? What should I cover next? I read every one.

Leave a comment

And if this was worth your time, do say hello!

Disclaimer: This newsletter is for educational and informational purposes only. Nothing here constitutes investment advice or a recommendation to buy or sell any security. I am not a SEBI registered investment advisor. Please do your own research or consult a qualified financial advisor before making any investment decisions. Investments in securities are subject to market risks. Past performance is not indicative of future results.

If you read Part 1(Don’t worry if you haven’t. You can understand this directly without the part 1), you now have something most retail investors don’t: a working framework for evaluating any lending business. You know what NIM means, why GNPA without NNPA tells you half the story, and how leverage amplifies the RoA tree in both directions. You also know why the credit cycle is turning and why that’s a tailwind for every lender in India right now.

Haven’t read Part 1? No worries, I got your back. I’ve written this article in such a way that you can read this part independently also. But would prefer if you go read 1st part first. This will make significantly more sense.

Read the Part 1 - Here

For those continuing, we take everything we built and point it at one company. Northern Arc Capital. Business line by line, RoA tree filled with real numbers from actual filings, peer comparison, and the valuation case for why I think this atleast doubles from here.

Same promise as Part 1: by the end you’ll understand not just Northern Arc but how to evaluate any NBFC from scratch. No projections. No price targets pulled from thin air. Just the business, the numbers, and the logic.

I’m invested in Northern Arc. Skin in the game. Not a SEBI registered advisor. This is not investment advice. Do your own research - Please.

Now let’s get into it.

Three Engines, One Company

Northern Arc is not a typical NBFC. Most lenders do one thing: take deposits or borrow money, lend it out, earn the spread. Northern Arc runs three distinct businesses simultaneously, each with different economics, different risk profiles, and different valuation implications. Understanding all three is essential before you look at a single number.

The three businesses are:

1. Lending (balance sheet, NIM-driven)

2. Placements (asset-light, fee-driven)

3. Fund Management (asset-light, fee-driven)

We’ll go into them one by one

Business 1: Lending

Borrow at one rate, lend at a higher rate, pocket the NIM. We’ve decoded this already. What makes Northern Arc different from a plain vanilla NBFC is where it sits in the chain. It runs two distinct lending models simultaneously.

[A] Wholesale Lending (Intermediate Retail)

The problem it solves

The problem: India has ~9,000 registered NBFCs. Banks will comfortably fund maybe the top 50 (loan books above ₹5,000 crore, strong parentage, established track records). The next 250-300 are legitimate businesses with credible promoters and real borrower demand, but banks won’t underwrite them individually and bond markets are out of reach at their scale. They’re capital-starved.

Solution: Northern Arc fills this gap. It evaluates these smaller NBFCs and MFIs (Micro Finance Institutions) on their books, governance, and underwriting quality, then lends directly from its own balance sheet into theirs. These partners on-lend to end borrowers across MSME, microfinance, vehicle finance, and affordable housing.Why this is a genuine moat

• Network (effect) built over 15+ years, 350+ originator partners. This wasn’t built overnight. Each relationship was earned through individual credit assessments, deal execution across multiple cycles, and repeat performance. A new entrant can’t replicate this in 2-3 years.

• Hard to replicate - Almost no one else does this at scale. Across India’s listed NBFC universe, MAS Financial Services is the closest peer with a similar B2B lending model, but with a narrower partner base and less sector diversity. NAC is arguably the only player that does this across six sectors (MSME, MFI, consumer, vehicle, housing, agri) simultaneously.

• First-mover in fintech NBFC relationships. NAC identified the fintech lending wave early (around 2016) and built funding relationships with companies like KreditBee, EarlySalary, MoneyView, and Navi when they were still small. Many of these relationships started as wholesale funding and later matured into co-lending partnerships. That progression from funder to partner is extremely hard to displace.

• This is also the feeder pipeline for Businesses 2 and 3 (placements and fund management, which we’ll cover later). A small NBFC that starts by borrowing from NAC’s balance sheet often graduates to having its loans placed with institutional investors through NAC’s platform, or packaged into NAC-managed credit funds. One relationship, three revenue streams. Hence creating a spectacular flywheel

The economics and where the book stands today

Now let’s walk through the numbers of wholesale lending. There’s a clear trade-off in this business.

The tradeoff: The credit risk is lower (probabilty of default), yes. NAC is lending to an NBFC entity, not to an individual borrower. And the loan is typically secured through hypothecation (the NBFC pledges its own underlying loan pool as collateral to NAC, so if the NBFC defaults, NAC has a claim on those retail assets). This structural seniority is why credit costs (losses when borrowers don’t repay loans) on the wholesale book have stayed below 0.4% for 15 consecutive years.

But the NIM is also lower. NAC earns a spread of roughly 3.4-3.5% on wholesale lending. Compare that to 9.5% on D2C lending, nearly 3x higher. Lower risk, lower reward. That’s the trade-off.

This (the low spread) is exactly why management is deliberately shifting the AUM mix toward D2C. D2C is a higher-yield but higher-credit-cost game. The bet is that NAC’s underwriting and tech can keep credit costs manageable while earning meaningfully fatter spreads.

Current position:

• Wholesale AUM: ₹6,629 crore (Dec’25), grown at 12% CAGR over FY21-25

• Share of total AUM: 44% (Dec’25), down from ~81% in FY21

• Growth rate: a deliberate 10-12% annually. It’s a conscious capital allocation decision. NAC is choosing to deploy incremental equity toward D2C where each rupee of capital earns a higher return

• Estimates 10% wholesale AUM CAGR through FY26-28, supported by expanding NBFC sector demand

The wholesale book won’t be the growth engine. It’s the low-volatility foundation that lets NAC take calculated bets on D2C without putting the entire business at risk.

Now let’s get into D2C. This is where the real magic happens, and where you need to understand the business deeply to appreciate why NAC is genuinely differentiated. Stay with me here. By the end of this section you’ll understand co-lending, direct lending, FLDG, and why this model is so scalable. There’s a lot to unpack, but it’ll be worth it ( I’ll use this opportunity to teach about the nature of the business, that’s my secret agenda really). Digging right in…

Retail Lending (D2C: Direct to Customer)

The faster-growing, higher-yielding half of the book. Here, NAC lends directly to end borrowers, not to NBFCs. This happens through two distinct models, and understanding both is critical.

Before we get into the flow, three quick definitions you’ll need:

• Origination: finding and acquiring the borrower. The customer discovery and application process.

• Disbursement: actually transferring the loan money to the borrower’s account.

• Collections: recovering the EMI payments from the borrower every month, and chasing overdue accounts (Collection efficiency is a performance variable that is tracked).

Model 1: Partnership / Co-Lending

In traditional lending, one institution does all three. In partnership lending, these tasks are split between NAC and its fintech partners. Here’s how:

Why this model is genuinely clever

The fintech partners are exceptional at things traditional banks and NBFCs are frankly bad at: customer acquisition, digital marketing, building smooth app experiences that convert browsers into borrowers, and retaining users. If you’ve used a banking app and modern fintechs, you know the UX gap (ICICI Imobile be like - bhai mujhe kyu toda). These apps know how to acquire customers cheaply at scale.

NAC doesn’t try to compete on that front. Instead, it provides the financial rails. The fintech does the customer-facing work (origination, servicing, collections). NAC provides the balance sheet (capital), the credit decisioning layer (NuScore), and the integration platform (nPOS). Hence the heavy ops, customer acquisiton cost sits with the partner. NAC’s cost is primarily technology and credit assessment.

This makes NAC extraordinarily scalable. Without building a single consumer branch, it can plug into 50+ fintech partners and access millions of borrowers through their apps. Each new partnership is essentially a new distribution channel with near-zero marginal cost to NAC.

The underwriting is not passive

NAC is not blindly funding whatever fintechs send. Every application passes through NuScore, NAC’s proprietary ML credit engine trained on 47 million+ historical loan data points. The fintech originates. NAC decides independently.

The data richness here is what sets this apart from traditional lending. Modern fintech underwriting goes far beyond a bureau score. These apps (with borrower permission) scan SMS records to detect existing loan obligations and real-money gaming activity, analyse spending behaviour, pull bank statements via Account Aggregator for cash flow verification, and flag device-level signals like multiple lending apps. A borrower who looks clean on paper can get rejected on behavioural signals that traditional banks don’t capture. NAC gets all of this through its fintech integrations, layered on top of its own NuScore assessment.

As part of my research for this piece, I validated 20+ individual loans disbursed through these partnerships, tracing them from origination through to repayment status. I also spoke with channel partners, credit managers, and engineers at fintechs who built the technical integrations with NAC. The picture that emerged: NAC’s tech stack is robust and scalable, their diligence is granular (down to individual cohort-level monitoring), and the feedback loop between field data and model recalibration is fast. This is a microcap, so this level of ground-truthing matters.

FLDG: the critical risk protection

FLDG (First Loss Default Guarantee) is the key risk protection in this model. The fintech partner deposits a percentage of the loan pool value (typically 5%) with NAC upfront. If borrowers default, this deposit absorbs the loss first, before a single rupee hits NAC’s balance sheet. The partner who sourced the borrower bleeds first. This creates a strong incentive for fintechs to keep origination quality high, because bad loans eat directly into their own money. The proof that it works: 90+ days past due (Note: X+ days DPD is an imp metric to understand loans going bad early ) on partnership-originated books was just 0.4% as of December 2024 (source: ICRA Rating Rationale, 2025).

Model 2: Own-Branch Direct Lending

NAC also lends directly through its own physical infrastructure, with no fintech intermediation:

• MSME / LAP (Loan Against Property): 84 branches across 7 states. These are secured business loans, meaning the borrower pledges a property as collateral. If they default, NAC has a legal claim on that property - this is the concept of Collateralisation. It is the oldest and most reliable form of credit protection. Ticket sizes are ₹10-12 lakh, yields run 17-20%.

• Microfinance / Rural: Operated through subsidiary Pragati Finserv across 285 branches. Group lending model targeting underserved rural borrowers.

This is the purest form of D2C. NAC owns origination, underwriting, disbursement, and collections end to end. Higher operational intensity, but also the highest margin and the most direct data feedback loop on borrower behaviour.

D2C in aggregate: AUM compounded at 64% CAGR from FY21 to FY25, reaching ₹7,064 Cr. Share of total AUM: 19% (FY21) → 56% (Dec 25). This is a deliberate strategic pivot executed at serious pace.

• Within D2C (₹8,492 Cr), consumer finance is the largest segment at ₹4,266 Cr (~50%), followed by MSME at ₹3,292 Cr and rural at ₹934 Cr

The takeaway is simple but powerful: no single sector dominates. When MFI cratered in FY25, the MSME and consumer books kept growing. This diversification is the reason NAC held GNPA at 1.36% with 19% MFI exposure during the worst microfinance cycle in a decade. Most pure-play MFI lenders saw GNPAs blow past 5-10%. NAC didn’t.

• Note: What’s equally important is how risk varies across these segments. Credit costs are not uniform:

  • Consumer (~6.5%) and rural (~5.3%) are meaningfully higher, while wholesale (intermediate retail) remains extremely low at ~0.4%. This dispersion highlights the core trade-off in NAC’s strategy: shifting toward higher-yield D2C segments while retaining a low-risk wholesale base as a stabiliser.

The Numbers: RoA Tree and Key Metrics ( This is heavy stuff, go slow here, take your time)

Edit: The credit rating is actually AA-(Stable)

The table makes the mispricing hard to ignore. A lender with 2.7% RoA, 23% AUM growth, and 1.36% GNPA trades at 1.1-1.3x book. Chola, with a lower RoA of 2.5%, trades at 3-4x. MAS Financial, with higher GNPA, trades at 2-3x. Something doesn’t add up. Even a valuation correction to 2x book, gives us a doubler.

Also note:

• Leverage stands at ~3.0x, providing headroom for further balance sheet growth

• Cost of funds has declined to ~8.5%, improving spreads & supporting profitability

Business 2: Placements (Debt Syndication and Structuring)

This is the business most people overlook, and it completely changes how you should value Northern Arc.

What it is, in plain English ( Read carefully, this might be new for some)

In the lending business, NAC holds loans on its own balance sheet. That ties up equity capital, which is expensive and finite. In the placements business, NAC does something different: it takes loan exposures it has originated or evaluated and places them with third-party investors (banks, mutual funds, insurance companies, AIFs, family offices, DFIs). The investor buys the credit exposure. NAC earns a placement fee (typically 20-25 basis points of the deal value) and frees up its own capital for the next transaction.

NAC is essentially a credit marketplace. Gross transaction volume reached ₹10,401 Cr in Q3FY26, reflecting strong platform throughput beyond balance sheet lending. It knows which NBFCs are creditworthy (15+ years of underwriting them) and which investors want Indian credit exposure (1,400+ investor partners). It sits between the two and facilitates the match.

Why this business is special:

• Capital-light, high-RoA. Earns fees without holding the loan. Return on capital deployed is multiples higher than lending.

• Low credit risk. NAC earns the fee for arranging. The credit risk transfers to the investor.

• Pure monetisation of relationships. The 350+ originator relationships and 1,400+ investor relationships that NAC built over 15 years are the product. The fee income is what those relationships yield when you connect them.

• Scalable without limits. No balance sheet constraint. More deal flow from originators + more investor appetite = more fees. No equity capital needed to grow.

• Counter-cyclical hedge. When credit stress rises and NAC’s own lending P&L takes a hit from higher provisions, placement volumes can still grow because investor demand for structured credit doesn’t disappear. Different rhythm from lending.

Now lets talk about the different ways this placement is enabled. The products NAC structures and places include:

• Securitisation: packaging a pool of retail loans (similar to how equities are pooled into Mutual funds) into a tradable debt instrument that investors can buy. NAC pioneered the Multi-Originator Securitisation (MOSEC) structure in India.

  

• Direct assignment: selling a portfolio of loans from one entity’s balance sheet to another’s ( No pooling)

  

• Bond placements: arranging NCD (Non-Convertible Debenture - basically fixed-interest loan instruments (Bonds) issued by NBFCs to raise money from investors) issuances for partner NBFCs and placing them with institutional investors.

• Syndicated loans: structuring a loan that multiple lenders co-fund.

Why this business matters

Three reasons, and they’re all significant:

1. Near-zero capital usage. Fee earned for arranging, not holding. Return on capital is multiples higher than lending.

2. Recurring and growing with the sector. Cumulative facilitated financing: ₹2.3 lakh crore+ since 2009 across 200+ investor partners. Small/mid NBFC borrowings have compounded at 22% CAGR over FY18-25 (Ambit Capital). As the sector grows, placement demand grows with it.

3. Flywheel that deepens the moat. Every deal strengthens both sides: the originator got funded, the investor got a performing asset. More originators bring more deal flow. More deal flow attracts more investors. More investors enable larger placements. Genuine network effect, rare in financial services.

The numbers

• Placement volumes: ₹12,393 crore in FY25. Q3 FY26 volumes of ₹3,669 crore grew 73% YoY (source: ICICI Direct).

• 9M FY26 placement volumes: ₹9,068 crore, up 13% YoY.

• Fee income: NAC earns a net fee of 20-25 basis points per transaction. On ₹12,000+ crore of annual volumes, that’s meaningful income with near-zero capital requirement.

• Core fee income grew 24% YoY in FY25, even while balance sheet expansion was only 12%. Management explicitly flagged this: fee income is growing faster than the lending book. That’s the strategy working.

• Ambit Capital projects 12-15% placement volume CAGR through FY26-28, with 15% fee income CAGR.

How to think about this as an investor

Most NBFCs have one revenue stream: net interest income (NII). Northern Arc has two: NII from lending, plus fee income from placements (and fund management, which we’ll cover next). In FY25, non-interest income from placements and fund management contributed roughly 36% of pre-tax profit. At most comparable NBFCs, fee income contributes 7-28%.

This fee income is less sensitive to credit cycles. When credit costs spike (as they did in FY25), fee income keeps flowing because the placement business is driven by transaction volumes, not by the quality of NAC’s own loan book. This creates a natural hedge within the P&L. The lending business has cyclicality. The placement business dampens it ( well, to some extent).

Business 3: Fund Management (NAIM)

This is the third leg, and the most capital-light of all.

What it is

It manages debt-focused Alternative Investment Funds (AIFs) and Portfolio Management Services (PMS) strategies. In simple terms: they pool money from institutional investors and HNIs, invests that pooled capital into debt instruments of NBFCs and mid-market corporates (the same universe NAC knows deeply from its lending and placement businesses), and earns a management fee for doing so.

If lending is NAC using its own money, and placements is NAC arranging deals for others, fund management is NAC investing other people’s money and earning a fee for the expertise.

The numbers

• Fund AUM: ₹3,198 crore across 9 active AIFs/PMS strategies as of September 2025 (source: ICRA, Dec 2025). Up from ₹2,796 crore in December 2024.

• Cumulative commitments raised: ₹6,500+ crore since inception across all AIFs and PMS strategies.

• Cumulative deployment: ₹12,000+ crore across sectors (source: NAIM Annual Report FY25).

• Track record: 6 AIFs and 1 PMS strategy fully exited. Zero capital loss to any underlying investor across all exited funds. No fund extensions. 100% on-time distributions over the last decade.

• Management fee: 100-110 basis points of fund AUM annually (source: Loggical Investor, Jan 2026).

• Growth: Fund AUM has been growing at 12-15% CAGR. Ambit Capital estimates 13% fund management AUM CAGR through FY27-28.

• NAC’s own exposure: 12% of fund AUM as of September 2025 (up from 5% in March 2024). This is “skin in the game” that aligns NAC’s interests with fund investors.

Why this matters

1. Extremely capital-light, recurring income. 100-110 bps management fee on ₹3,200 crore of AUM translates to roughly ₹32-35 crore of annual fee income. This requires almost no incremental capital. It’s pure operating leverage. As AUM grows, fees grow proportionally without requiring more equity.

2. The knowledge advantage is built-in. The credit underwriting capability, the 15-year data on NBFC performance, the 350+ originator relationships, the proprietary NuScore analytics: all of this is shared infrastructure. Hence this business doesn’t need to build a separate research team. It piggybacks on the parent’s institutional knowledge. One ecosystem, three monetisation channels.

3. It’s an investor acquisition funnel. Investors who come in through these funds (700+ investors across corporates, family offices, HNIs, global funds, DFIs) become part of NAC’s broader investor network. Many of these same investors later participate in placement transactions or co-invest alongside NAC’s balance sheet. The fund management business feeds the placement business, which feeds the lending business. The flywheel again.

AltiFi: the retail extension

Worth a quick mention. AltiFi is NAC’s retail-facing bond investment platform (operated through subsidiary Northern Arc Securities). It lets individual investors and small corporates invest in fixed-income instruments (bonds, commercial paper, securitised notes, sovereign gold bonds) starting from as low as ₹10,000.

As of June 2025: ~59,000 registered users, ₹1,220 crore+ in investments facilitated. Still small, but directionally interesting. It opens up debt instruments (bonds, securitised notes, commercial paper) that were previously institutional-only to retail investors and small corporates. Could evolve into a distribution channel for NAC's own issuances and placement-structured instruments. We'll wait and watch.

Why all three businesses together matter more than any one alone

Here’s the key insight that most people miss when they look at Northern Arc.

This is not a company with a lending business and two small side-shows. The three businesses form an interconnected ecosystem where each one feeds the other

One originator relationship generates lending income, placement fee income, and fund management fee income. One investor relationship participates across all three channels. The data from all three improves the underwriting for all three. This is what a platform business looks like in financial services.

And here’s the financial impact: in FY25, non-interest fee income (placements + fund management) contributed ~36% of PBT. For most comparable NBFCs, this number is 7-28%. That means roughly a third of NAC’s pre-tax profit comes from businesses that are capital-light, recurring, and largely insulated from credit cycle stress.

The market prices NAC at 1.1-1.3x book as if it were a pure lending NBFC. It is not. The placement and fund management businesses carry a fundamentally different risk-return profile and deserve a different valuation treatment. We’ll come back to this when we discuss valuation.

Key Risks to Monitor

No thesis is complete without an honest accounting of what could go wrong. These are the things I’m watching:

• MFI asset quality normalisation is not guaranteed. Stress has peaked and early signs of recovery are visible, but Karnataka remains a problem state and rural borrower overleveraging doesn’t reverse overnight. If MFI credit costs stay elevated beyond H1 FY27, the earnings recovery gets delayed.

• D2C execution risk is real. The pivot from wholesale to retail is the heart of the thesis. But retail lending is operationally harder: more borrowers, smaller tickets, higher collection intensity. Partnership-based D2C depends on fintech partners maintaining origination quality. Own-branch MSME is still only 84 branches across 7 states. Scaling this without letting underwriting standards slip is not trivial.

• Consumer digital book has shown cracks. NAC took a one-time ECL adjustment of ₹23.4 crore on older consumer digital cohorts where actual credit outcomes were worse than NuScore predicted. The model was recalibrated, but it’s a reminder that ML-based underwriting on new-to-credit borrowers is still being battle-tested.

• Concentration in the NBFC ecosystem. NAC’s entire business (lending, placements, fund management) is built around the Indian NBFC/fintech ecosystem. A systemic shock to smaller NBFCs (regulatory tightening, a liquidity event like IL&FS 2.0) would hit all three revenue streams simultaneously.

• Small-cap liquidity. Market cap is ~₹4,000 crore. Institutional holding is still low at ~16%. This means the stock can be volatile on both sides, and large block trades can move the price meaningfully.

• PCR at 52% is below comfort. Provision coverage is lower than the 60-70% I’d ideally want. At current GNPA levels it’s manageable, but it leaves less buffer if a fresh wave of stress hits.

• Regulatory risk. RBI imposed a ₹2.7 lakh penalty on NAC in January 2026 for KYC/AML control gaps. Cosmetic, not structural. But as a growing NBFC with complex partnership-based lending, regulatory scrutiny will only increase. Compliance infrastructure needs to keep pace.

Parting words

Northern Arc is not a simple NBFC. It is a credit platform that operates across lending, debt placements, and fund management, built on top of a 15-year-old network of 350+ originator partners, 1,400+ investor partners, and a proprietary technology stack that ties all three together.

The lending business is mid-transition: shifting from lower-margin wholesale toward higher-margin D2C retail, with NIM already at 9.9% and rising. The placement and fund management businesses contribute ~36% of pre-tax profit while using almost no balance sheet capital. The credit cycle is turning in NAC’s favour: RBI is cutting rates, MFI stress is peaking, and cost of funds is headed down.

At 1.1-1.3x book value and ~12-13x earnings, the market is pricing this as a mediocre NBFC with a 10% ROE. It’s not pricing in the D2C-driven NIM expansion, the normalisation of credit costs from 2.9% to an expected 1.5-1.8%, the operating leverage from a ~3x-to-5x leverage ramp, or the fee income optionality. If even half of this plays out, ROE moves toward 15%+ and the stock re-rates meaningfully.

This is a business I own and continue to track actively. I could be wrong. The MFI book could surprise negatively. The D2C transition could stumble. The market could stay irrational longer than expected.

That’s why I want to hear from you.

I spent a lot of time researching, writing, and refining this for you all. Would be super grateful if you could subscribe

If this piece helped you understand lending businesses better, or if it changed how you think about Northern Arc, please share it with someone who’d find it useful.

Share

If you think I’ve got something wrong, I genuinely want to know. Write to me, leave a comment, challenge the thesis.

Leave a comment

The best investment writing gets better through debate, not agreement.

And if you’d like more deep-dives like this, please subscribe and restack. It helps more than you’d think.

Let me start with a confession.

For the longest time, I avoided lending businesses. Not because I thought they were bad. I knew they were great. I’d seen Bajaj Finance do things that seemed almost illegal in a free market. But every time I tried to go deeper, read a concall, open an annual report, I’d hit a wall of terminology. GNPAs. NNPAs. DPDs. NIMs. Credit costs. Risk weights. I’d close the tab and go back to something comfortable.

If that’s you, this article is for you.

I’m not writing this as an expert handing down wisdom. I’m writing it as someone who was exactly where you might be and eventually decided to push through. What helped wasn’t a textbook. It was working at Mahindra Finance, one of India’s largest NBFCs, and later building lending products myself. That gave me mental models I wish I’d had earlier. This article is an attempt to pass those on.

Two things in one piece: Lending Business 101, and enough of a glimpse into Northern Arc Capital to make you want to dig deeper. I’m invested in the company. I’m biased. You’ve been warned.

The Greatest Wealth-Creating Sector Most Retail Investors Ignore

Look at the biggest, most consistent wealth creators in Indian equity markets over the last two decades. Not the flashy ones, not the venture-backed moonshots, but businesses that quietly compounded year after year. A disproportionate number are lenders.

Bajaj Finance grew its loan book (total amount of money a bank or lender has given out as loans that borrowers are still repaying - You strated learning already!) from roughly ₹17,000 crore in FY15 to over ₹4.1 lakh crore by FY25. Around 24x in a decade, with book value per share compounding at over 25% annually. Chola Finance, Muthoot, HDFC Bank, Kotak, Gruh Finance before its merger. Patient holders of these businesses made life-changing returns, not because of luck, but because the business model, when run with discipline, is structurally engineered for compounding.

Most retail investors still stay away, largely because the language feels impenetrable. By the end of this piece, you’ll be able to open any NBFC concall and follow along. That’s the promise.

Why the Business Model Is So Powerful

A lending business does one thing: borrow money at one rate, lend it at a higher rate, and keep the spread (the difference between what you earn on loans and what you pay to borrow, this is the engine of the entire business). The raw material is money, and unlike cotton or iron ore, money never goes out of fashion.

India is structurally money-scarce. Credit penetration, the ratio of total credit to GDP, sits at roughly 57% in India versus 170%+ in developed markets. Hundreds of millions of people have aspirations that consistently outpace their savings. A farmer needing seeds before the harvest, a small business owner bridging a payment gap, a family wanting a home before they’ve fully saved for one. This demand isn’t a product of good times, it’s a product of human nature. Human ambition doesn’t pause for Trump’s tariff tantrums or Iran’s posturing in the Strait of Hormuz. As long as people aspire to more than they currently have, the demand for credit persists.

What makes this compound so beautifully is the flywheel. Earn interest, retain earnings, redeploy into the loan book, earn more interest on a larger base, repeat. In most industries, scaling erodes returns because growth invites competition and compresses margins. In lending, if you’re underwriting well, growth creates returns. Bajaj Finance compounded its book value per share at over 25% annually for a decade. Chola, Muthoot, Gruh Finance before its merger, all delivered similar multi-decade compounding to patient holders. The runway here is the unmet credit demand of 1.4 billion people, and that runway isn’t shrinking.

The catch is bad loans and the cyclicality of the sector. Because lenders run on leverage (borrowing several rupees for every rupee of their own equity to then lend out, which amplifies both gains and losses), a wave of defaults can cascade fast, as IL&FS, DHFL and Yes Bank showed us. Demand being infinite doesn’t protect a lender with sloppy underwriting (the process of checking how risky it is to give someone a loan before approving it.). Which is precisely why picking the right lender matters as much as picking the right sector.

How a Lending Business Works

Don’t get overwhelmed by this. We’ll deconstruct this line item by line item.

You borrow money at an interest rate (cost of funds ), lend it out at a higher interest rate(called yield on assets). The difference is your NIM, Net Interest Margin. This is the starting point, and the single biggest driver of profitability. Add that in. Now subtract your operating expenses (salaries, technology, branch costs, the everyday cost of running the business). What remains is PPOP, Pre-Provision Operating Profit (your earnings before setting aside any buffer for bad loans).

Now, you don’t wait for a borrower to formally default before acting. Based on early signals like missed EMIs or deteriorating financials, prudent lenders proactively set aside a buffer called provisions (money charged to the P&L today, anticipating loans that may go bad tomorrow). The annualised rate at which a lender provisions, expressed as a percentage of the total loan book, is called the credit cost (think of it as the price the business pays for the risk it has taken on). Subtract provisions from PPOP, account for tax, and what remains is PAT, Profit After Tax.

Divide PAT by total assets and you get the single most important metric in evaluating any lender: RoA, Return on Assets (how much profit the business generates for every rupee deployed, the purest measure of operational quality). A well-run NBFC typically delivers 2%+ RoA. Bajaj Finance has consistently delivered 4%+ RoA, which is why the market has rewarded it so richly.

But lenders don't just use their own money. They borrow aggressively to lend more. This is leverage. Multiply RoA by leverage and you get RoE, Return on Equity (the actual return shareholders earn on their invested capital). A lender with 2.5% RoA running at 5x leverage generates 12.5% RoE. Push leverage to 7x and that becomes 17.5% RoE. This is the power and the risk, simultaneously.

Each line of this tree has specific metrics you should track to evaluate whether a lender is doing a good or poor job at that layer. That’s exactly what we decode next, with real benchmarks from Indian NBFCs so you have a reference point and not just abstract definitions.

What Makes a Good Lender: Three Things That Actually Matter

Now that we have the RoA tree, let’s populate it with the metrics that tell you whether a lender is healthy or not. Everything comes down to three questions. Is it earning well (Earnings Quality)? Is the book clean (Asset Quality)? And can it survive a bad cycle(Structural Resilience)? Let’s take each in turn.

1. Earnings Quality

Risk-adjusted yield (yield on assets minus credit cost) is the metric most investors miss. Raw yield is vanity, risk-adjusted yield is sanity, hope this brings clarity ( I’ll pause my attempt at rap for the sake of humanity) A microfinance lender charging 24% but running 9% credit costs has a risk-adjusted yield of 15%. An MSME lender charging 16% with 1.5% credit costs lands at 14.5%. Similar, but the second business is far less volatile and far easier to scale.

NIM, Net Interest Margin (yield on assets minus cost of funds) is the core spread after paying for borrowed money. For a healthy NBFC, 7-9% is solid. Drop below 5% and there’s very little room for error when credit costs rise.

Cost of funds is where two identical loan books produce dramatically different profitability. An AAA-rated NBFC borrows at 7.5-8.5%. A BBB-rated one pays 11-12%, a brutal structural disadvantage compounding every quarter. Equally important is liability diversity (the range of sources a lender borrows from: banks, mutual funds, insurance companies, bond markets, securitisation). An NBFC dependent on two PSU banks for 70% of funding is one relationship away from a liquidity crisis. Ask IL&FS creditors how that feels (doesn’t matter if you don’t know).

Cost-to-Income ratio (operating expenses as a percentage of total income). For a direct retail lender, 45-55% is reasonable. A declining ratio as the book scales is the signature of a business model that gets better as it gets bigger.

2. Asset Quality

All these metrics answer one question: how much of the loan book is quietly planning to not come back (and going to fly to England)?

GNPA or Gross Non Performing Assets (percentage of the loan book that is atleast 90+ overdue, officially classified as bad debt aka NPA). Below 2% is healthy. Above 4% and the business is running a charity with extra steps.

But GNPA alone can be flattering. Always read it alongside 2 metrics:

• NNPA (GNPA minus provisions (the buffer, remember?) already held. This is the residual risk actually sitting on the balance sheet)

• PCR, Provision Coverage Ratio (provisions divided by gross NPAs. Meaning what fraction of bad loans are already absorbed). A lender with 3% GNPA and 0.8% NNPA has been prudent. One with 3% GNPA and 2.7% NNPA has been hoping for the best.

• Some more: 90 days is a lagging indicator, though for bad loan observation. Hence we create something known as DPD buckets (Days Past Due: how many days a borrower is behind. This is bucketed into 30, 60, 90 DPD). Under Ind-AS, these map to

  • Stage 1 (performing),

  • Stage 2 (30-89 days overdue = the watchlist),

  • Stage 3 (90+ days, formally impaired).

  • Note: A swelling Stage 2 pool is tomorrow’s GNPA problem, showing up before the headline number moves.

3. Structural Resilience

This is about whether the lender survives when things go wrong. And in lending, things always go wrong eventually.

Think of CRAR, Capital Adequacy Ratio as the lender’s personal savings buffer. Imagine you lend out money and some of it doesn’t come back. CRAR is the cushion of your own money sitting behind the business to absorb those losses without the whole thing collapsing. RBI mandates NBFCs maintain at least 15% of their risky assets covered by this buffer. Well-run lenders stay well above this, giving them room to absorb a bad cycle and keep growing without panicking.

We’ve already covered leverage. The short version: 4-7x is the healthy operating range. Below 3x and capital is sitting idle. Above 8x and you’re praying the cycle doesn’t turn.

Finally, granularity rarely gets discussed but quietly matters the most. A ₹10,000 crore book with 10 borrowers is catastrophically fragile. One bad corporate relationship and 10% of your book is in trouble overnight. The same book spread across 500,000 small borrowers is far more resilient. No single default moves the needle

The Credit Cycle: Why Timing Matters in Lending

I said earlier that demand for credit is effectively infinite. So if demand never disappears, why does the lending sector go through painful boom-bust cycles? The answer lies entirely on the supply side.

Why Lending Is Cyclical

Credit supply is governed by three things:

1. The availability of capital

2. The appetite for risk

3. Regulatory guardrails.

All three move together, and when they move, they overshoot in both directions.

Here is how a typical cycle plays out.

When the economy is growing lenders are flush with capital, competition intensifies, underwriting standards quietly loosen. Everyone is lending to everyone. Borrowers take on more than they can service. At some point, defaults start rising. Lenders pull back, tighten standards, sometimes simultaneously. Credit supply contracts sharply, even though demand hasn’t gone anywhere. This is the bust. The irony of credit cycles is that supply withdrawal happens precisely when borrowers need money most.

What amplifies this in India specifically is the leverage inherent in the business. A lender running 6x leverage doesn’t need a 30% default rate to get into trouble. A 3-4% credit cost spike can wipe out a full year of earnings. This makes lenders extremely sensitive to sentiment shifts, and sentiment in credit moves in herds.

RBI adds another layer. It doesn’t just watch, it intervenes. Sometimes preemptively, sometimes reactively. When it tightens, the entire sector feels it simultaneously, which is exactly what happened in late 2023 when RBI raised risk weights (the regulatory multiplier applied to certain loan categories that determines how much capital a lender must hold against them; higher risk weights mean less room to lend) on unsecured lending, deliberately engineering a slowdown.

Cycles Are Not Uniform Across Segments ( Can skip/ read later)

Different lending segments have very different cycle lengths and volatility profiles. Worth knowing briefly.

Microfinance is the most volatile. Cycle length of 3-5 years, driven by borrower overleveraging, political loan waivers and collection disruptions. FY25 was a textbook MFI bust: multiple lenders to the same borrower, rising delinquencies, sharp credit cost spikes across the sector.

Unsecured consumer lending cycles over 3-4 years, closely correlated with employment and urban income trends.

Vehicle finance is tied to commercial vehicle demand cycles, typically 4-6 years, closely tracking infrastructure and freight activity.

MSME lending cycles over 4-7 years, sensitive to GST compliance, working capital stress and broader economic activity.

Secured lending (home loans, LAP) is the most stable, cycles of 7-10 years, cushioned by collateral. Rarely blows up catastrophically.

The key takeaway: a lender concentrated in one segment carries the full force of that segment’s cycle. A diversified lender across segments dampens the volatility, which is one reason diversification in a loan book is not just a marketing talking point.

Where Are We in the Cycle Right Now?

Before we get into the signals, a quick picture of how money actually moves through the Indian financial system. Most people know RBI exists but don’t fully appreciate how directly its decisions affect every lender and borrower in the country.

The repo rate (the rate at which RBI lends money to banks overnight, essentially the price of money at the top of the system) is RBI’s primary lever. When it cuts this rate, cheaper money cascades down the entire chain. Banks borrow cheaper from RBI, lend cheaper to NBFCs, NBFCs pass it on to businesses and individuals. The entire cost of credit in the economy drops. When RBI cuts rates, it is explicitly signalling: we want more money flowing through this system. That is a tailwind for every lender in the country.

My read is that we are at or very close to the trough of this cycle, with meaningful evidence that the turn has already begun. Five signals, all pointing the same direction.

RBI has shifted decisively to easing. The repo rate has been cut 50 basis points in 2025 so far, with more widely expected. For every 50 bps reduction in cost of funds on a ₹10,000 crore book, that’s ₹50 crore of additional annual earnings flowing straight to the bottom line, without the lender doing anything differently.

RBI is reversing the risk weight tightening. The elevated risk weights introduced in November 2023 are being walked back, directly freeing capital headroom and allowing lenders to grow without raising fresh equity.

System-level asset quality is at multi-year lows. The RBI’s own Financial Stability Report flagged improving bank-level NPA ratios. MFI stress, which dominated FY25 headlines, is showing early signs of stabilisation.

Fiscal policy is explicitly targeting domestic demand. With global trade uncertainty from tariff disputes dampening export prospects, the government has pivoted hard toward consumption. Personal income tax relief in Budget 2025, GST rate rationalisation, redirected spending all point the same way: more money in borrowers’ hands, more demand for credit.

MFI credit costs remain elevated but are stabilising. The one amber signal. Worth watching for 2-3 more quarters before declaring the all-clear.

Put it together: cost of capital falling, regulatory headroom expanding, borrower stress peaking, fiscal stimulus arriving. All five levers governing credit supply moving in the same direction simultaneously. This is the setup that precedes a lending upcycle. The best returns in lending stocks are made not when the cycle is obviously good and everyone is piling in, but when data suggests the turn is happening and the market hasn’t fully priced it yet.

Which brings us to Northern Arc Capital.

What’s Coming in Part 2

Northern Arc Capital is an NBFC most retail investors have never heard of. It listed in September 2024, trades close to book value, has a GNPA of 1.36% through one of the worst credit cycles in recent memory, and is sitting on a capital adequacy ratio of 26% at a time when the lending cycle is just beginning to turn. A lender with fortress balance sheet metrics, trading at the price of a mediocre one.

Here is why I think it doubles from here.

The company has been quietly transitioning its loan book from wholesale to retail lending. The market is still pricing it like a wholesale lender, at roughly 1.1-1.3x book. As that retail mix crosses 55-60%, the re-rating that follows is almost mechanical. Retail lending franchises in India don’t trade at 1.1x book. They never have. Add to that a post-IPO capital adequacy of 26% against a regulatory minimum of 15%, meaning the company has an enormous amount of dry powder sitting idle, waiting to be deployed into a cycle that is just now turning in its favour. As that capital gets deployed, RoE expands. As RoE expands, the multiple re-rates. That is the double.

In part 2 we’ll go deep on all of it:

What Northern Arc actually does. The business model is genuinely unusual. They sit in a part of the lending ecosystem most investors don’t think about, and that obscurity is a big reason the mispricing exists.

The numbers versus peers. We’ll run the full RoA tree on their actual financials and compare them against Chola Finance, MAS Financial and others using every metric we just decoded together.

The loan book transition thesis in full. This is the heart of the valuation argument and where the numbers get really interesting.

Risks. Because no thesis is complete without an honest look at what can go wrong. There are real ones here and we’ll not shy away from them.

A valuation framework for lending businesses. What the right multiple should be, what normalised RoE looks like as capital gets deployed, and what the stock needs to do to get there.

If this piece gave you the language to read a lending business, part 2 will give you a specific reason to act. Subscribe so you don’t miss it when it drops.

Subscribe now

Disclaimer: I am invested in Northern Arc Capital. This is not investment advice. Please do your own research before making any investment decisions. I am not a SEBI registered investment advisor.

If this piece was useful, restack it, drop a comment, hit the like button. It genuinely means a lot. I spend a significant chunk of my weekends on these, and knowing it’s landing with people is honestly the only reason I keep sacrificing my Sundays to financial statements. Your 10 seconds of engagement buys me another weekend of this. Fair trade I think.

Leave a comment

See you in Part 2.

Fair warning: I am no expert in macroeconomics or geopolitics. But I genuinely believe what follows will benefit you. Print this one. Read it slowly. Keep it at the back of your mind when making investment decisions.

Here is the advice everyone gets today. Buy the dip. SIP karte raho (continue your investments). Long term mein paisa banta hi hai ( Markets always recover).

This has been true for 40 years. And that is precisely why it may be dangerous now.

The conditions that made it true are dissolving. This article is not about whether you should do your SIP or not. It is simply about awareness. Awareness that we may be entering a world with different rules. Rules unseen in our investing lifetimes. That awareness might strengthen your conviction in your current approach, or it might change it. That is entirely up to you.

Over the past few weeks, I have been reading obsessively on this topic. What follows is my synthesis of other people’s ideas. I have no original ideas of my own here. I am a borrower and synthesiser - and I hope a useful one.

Chapter I — How the present economic world was built (and who built it)

1.1 — Bretton Woods and the dollar’s coronation

World War II ended in 1945, leaving Europe and Japan in ruins. One major economy had not only survived but emerged stronger - the Uncle Sam ( United States). It held nearly two-thirds of the world’s gold reserves and was the only functioning industrial power standing.

In 1944, even before the war ended, 44 nations gathered in Bretton Woods, and struck a deal. Every currency would be pegged to the US dollar ( conversion rate fixed to the dollar). The dollar alone would be convertible to gold at $35 per ounce.

The mechanism was elegant. The US supplied the world with dollars. The world used those dollars to buy American goods, American bonds, and to settle trade amongst themselves. Every central bank needed dollar reserves. Every trading nation needed dollar access to function.

But there was a contradiction buried inside from day one. For the world to have enough dollars to trade, the US had to constantly run deficits ( spend more than it earned), print more than it held in gold. The system carried its own undoing within it ( we’ll come back to it later)

1.2 — President Nixon’s shock and the birth of the fiat era

Vietnam War and welfare programs had flooded the world with dollars. European nations, led by a suspicious France, said waaaait a minute, the math ain’t mathing. They started demanding gold for their dollars. US gold reserves were draining fast. Then came a history-defining moment. On August 15, 1971, Nixon went on television and announced that the US would no longer convert dollars to gold. Unilaterally. The architecture the world had built its financial order on for 27 years, was dismantled in a single evening broadcast ( like it is now with Trump’s tweets or Truths).

What followed was arguably the greatest financial sleight of hand in history. The dollar (now backed by nothing except trust in the United States) did not collapse. It remained the world’s reserve currency. Every other nation was too dependent on it to walk away. The US had effectively written itself a blank cheque.

1.3 — The petrodollar: genius or extortion?

But soon, President Nixon had a problem. He had just untethered the dollar from gold. A currency backed by nothing should, in theory, collapse. The world needed a reason to keep using dollars. So he went looking for a new anchor - and found one in the most essential commodity on earth ( any guesses?)

In 1973–74, the US struck a deal with Saudi Arabia. Simple and elegant: Saudi Arabia would price all its oil exclusively in US dollars. Every other OPEC nation would follow. In return, the US would provide military protection to the Saudi royal family, guarantee the security of the Gulf region, and offer Saudi Arabia something equally valuable - a place to park its suddenly enormous oil revenues. US Treasury bonds (and also markets): safe, liquid, dollar-denominated, and backed by the most powerful nation on earth. For a monarchy sitting on a desert with more oil money than it could spend, this was an attractive arrangement.

The loop this created was beautiful, Uncle Sam loved it.

Every country on earth needs oil to run. To buy oil, every country needs dollars. So every nation - Japan, Germany, India, China - must constantly earn, borrow, or hold dollars just to keep its economy functioning. This created a permanent, structural global demand for the dollar that had nothing to do with gold, nothing to do with US fiscal discipline, and everything to do with the fact that you simply could not keep the lights on without it.

The second half of the loop was even more elegant. Saudi Arabia and OPEC were now accumulating more dollars than they could spend domestically. Those surplus dollars flowed back into US Treasury bonds ( and also the markets). Which meant the US could borrow cheaply (as the US bond became safe havens for money, rates remained stable and low), run deficits freely, and finance its consumption and military dominance. Who funded it? Rest of the world. The French called this America’s “exorbitant privilege.” They were not wrong.

This one deal underwrote 50 years of American power. It is only now, for the first time, showing serious cracks. We will get to that.

Chapter II — The decades of the cycle (Chokhani's framework)

2.1 — The alternating machine: hard assets vs financial assets - A short detour - the framework you need before we go further

Before we continue the history, let me introduce an idea from Manish Chokhani that will make everything else click.

Removing the gold standard removed all discipline from money creation. Every decade since 1971 has produced a new bubble - alternating between financial assets and hard assets - as cheap money finds a new home each cycle.

• 1970s: oil, gold, commodities surge as the dollar weakens fresh off the gold standard ( when Nixen removed it few sentences ago - remember?).

• 1980s–90s: the Fed crushes inflation by raising rates aggressively → Dollar strengthens. Equities and bonds enter a historic bull run.

• 2000s: China’s rise powers a commodity supercycle. BRICs become the story.

• 2010s: Quantitative Easing (Low interest rates) floods the world with cheap money, all of it finding its way into US tech. The Magnificent Seven were children of this era.

• 2020s: Chokhani believes the cycle is turning again - back to hard assets, gold (we have already seen a trailer of this), real infrastructure. Most investors haven’t noticed yet.

This cycle is the spine of everything that follows. With that detour done, back to our history lesson.

2.2 — The Plaza Accord: when America forcibly crashed its own currency

Here is a story most investors don’t know - and one that looks uncomfortably familiar right now.

The US REER ( Real Effective Exchange Rate, which is simply a measure of how expensive the dollar is relative to all other currencies after adjusting for inflation) is today at the same extreme level it was in 1985. The last two times it was this high, the dollar fell 30 to 38%.

What happened in 1985?

The Fed had raised rates aggressively to kill 1970s inflation - nearly 20% at peak. It worked, but created a dollar so strong it hollowed out American manufacturing. The mechanism is simple:

Strong dollar → other currencies weaken ( you could sell the same goods for much less dollar value) → their exports flood US markets → American factories can’t compete → jobs disappear.

Japanese cars, German machinery, and Korean electronics poured in at prices American manufacturers couldn’t match. The Rust Belt hollowed out. Trade deficits exploded.

In September 1985, finance ministers of the five largest economies gathered at the Plaza Hotel in New York and agreed to jointly sell dollars to deliberately weaken the currency.

It worked. The dollar fell 35% in two years. The yen doubled, from 240 to 120 per dollar.

Here is where it gets dark. Japan’s exports were suddenly twice as expensive overnight. Japan slashed rates to cushion the blow. Cheap money flooded stocks and real estate. The Nikkei went from 12,000 to 39,000 by 1989. Then it burst. Two lost decades followed. A nation that looked set to overtake the US was frozen for thirty years.

Why does this matter today? Same extreme REER. Same trade imbalances, now with China. Same political pressure, now expressed as tariffs. But China is not a compliant ally. There is no Plaza Hotel moment available. If the dollar corrects this time, it will not be managed. It will be messy.

Chapter III — The fractures (why the old system is breaking)

3.1 — The hidden cost of the petrodollar: deindustrialisation and the wage trap

The petrodollar had a price paid not by Wall Street but by the American worker.

A permanently strong dollar made American exports permanently expensive and American factories permanently uncompetitive. Over 50 years, the US hollowed out the industrial base that won World War II — the factories, shipyards, and steel mills - offshored to wherever labour was cheaper.

Real wages for the median American worker have barely grown since 1971 in inflation-adjusted terms. Asset prices boomed. Wages didn’t.

The political consequence was inevitable. The working class that lost manufacturing jobs turned to Trump - twice. Tariffs are the expression of that anger. But tariffs are the symptom, not the cure. You cannot undo 30 years of offshoring in a presidential term. Rebuilding a semiconductor fab or a naval shipyard takes decades of industrial policy. The problem is real. The solution being applied is blunt and largely theatrical.

3.2 — De-dollarisation: slow, structural, and now accelerating

The dollar’s share of global reserves has fallen from 73% in 2001 to 58% today. Slow - but the direction is unambiguous. Five things are moving simultaneously.

1. Russia. The US froze $300 billion of Russian reserves post-Ukraine. Every central bank drew the same lesson: dollar reserves aren’t truly yours. Over 75% of Russia-China trade now settles in non-dollar currencies.

2. Saudi Arabia. The petrodollar covenant - all oil priced only in dollars - is cracking. Saudi Arabia has begun accepting yuan for Chinese oil.

3. BRICS expansion. The bloc now covers 45%+ of the global population and a dominant energy share. Reducing dollar dependence is an explicit policy, not a mere fringe position.

4. Central banks buying gold. Record purchases three years running. Gold has no counterparty risk - no government can freeze or sanction it.

5. Payment plumbing. China’s CIPS (assume it to be UPI for inter-country payments. Its a replacement for American swift system which was a monopoly till now ) now has 100+ participating nations. The mBridge project settles transactions between central banks without touching the dollar system entirely.

Yet there is a nuance: this is not a dollar collapse. All currencies have a problem and we don’t have a clear winner yet. Yuan is not freely convertible (massive govt interventions here), euro is fragmented, gold cannot scale into a payments system. But a fall from 58% to 40% of global reserves means higher US borrowing costs, a weaker dollar, and structurally higher inflation. You don’t need a collapse for it to matter.

3.3 — The return of the military-industrial complex ( Read this one very carefully)

The post-Cold War peace dividend is over.

Europe is rearming at a pace not seen since the 1930s. NATO nations are scrambling from 1% toward 3-4% of GDP on defence. Germany has announced its largest rearmament programme since World War II.

Ukraine exposed something uncomfortable: decades of globalisation had optimised the US military for drone strikes and counterinsurgency, not sustained conventional warfare. US artillery shell production couldn’t keep pace with a single conflict ( because it has no manufacturing base left). American shipbuilding is now a fraction of China’s.

Every major power is simultaneously rebuilding defence production capability that was allowed to atrophy. That spending doesn’t turn off after a conflict — it becomes a new baseline.

The implication: a world of fragmented supply chains, competing geopolitical blocs ( as against a unified world order dictated by Uncle Sam), and military rearmament carries a permanently higher premium on self-sufficiency. That is structurally inflationary (read Mehengai). It is not a temporary spike.

Chapter IV — The war that changes everything

4.1 — The Clean Break blueprint: Israel’s 30-year plan

In 1996, a policy document called “A Clean Break” was penned for an incoming Benjamin Netanyahu. The thesis was: Israel cannot survive through peace processes with neighbours who reject its existence. The only path to security is regional hegemony - systematically removing every state-level threat around it.

The sequence laid out: Iraq, then Syria, then Lebanon, then (or now) Iran.

What is remarkable is how methodically it has been executed.

• Iraq 2003 → the US removes Saddam.

• Syria 2011 to 2024 → Assad’s regime falls.

• Lebanon 2024 → Hezbollah’s entire leadership is decapitated.

• Iran 2025-26 → the final chapter, now underway.

Three decades. Three US administrations (who for no real reasons waged Israels wars). The document written for Netanyahu in 1996 reads today less like a policy proposal and more like a checklist with almost every item ticked ( but one, until now).

Netanyahu has described the Iran campaign as a dream realised. Watching the sequence, it is hard to disagree with his assessment of what this moment represents for Israeli strategy.

4.2 — Why this war will not end quickly

All three actors are trapped by their own logic.

Trump needs a quick political victory - a headline, a handshake, something to declare won. But the decapitation of Khamenei neither toppled the regime nor sparked a popular uprising. Regime change requires a sustained campaign, not a news cycle. His timeline and Netanyahu’s war aims are already pulling in opposite directions.

Netanyahu has no reason to stop. Every piece of the Clean Break plan is now in motion. A ceasefire leaves Iran damaged but intact - with the capacity to rebuild and retaliate. The doctrine he has followed for 30 years demands completion, for he wants to go down in history to create Greater Israel ( Aka Akhand Israel).

Iran cannot negotiate. They came to the table in June 2025 and were struck mid-talks. They came again in February 2026 and were struck again - this time with their supreme leader killed. No Iranian leadership can sign anything that looks like surrender after that humiliation. The only path left is to make the cost of future strikes so severe that no one seriously contemplates them again. They are not fighting to win. They are fighting to deter.

Three actors. Three incompatible imperatives. No exit ramp in sight ( Yes read that again, no exit ramp, this war will stretch)

4.3 — What Iran is doing and why

Iran cannot win a conventional war against the US-Israel military machine. So it is doing the next best thing - widening the cost until everyone around it screams for it to stop.

The strikes on GCC nations are not irrational rage. They are calculated. Ras Laffan, Qatar’s LNG hub → shut down. Ras Tanura, Saudi Arabia’s largest refinery → hit. UAE oil facilities → struck. Oil jumped 13%, European natural gas 40%. Brent crossed $100.

The Strait of Hormuz ( through which 20% of global oil transits) -remains closed as what Iran’s new supreme leader has explicitly called “a tool of pressure.” The cascade goes well beyond oil. One third of global fertilizer trade transits the Strait. So does helium for semiconductors, sulfur for copper processing. US Midwest corn and soy planting season is now at risk from fertilizer price spikes alone.

Iran also struck the Palm Jumeirah in Dubai, residential towers in Manama, and an AWS data centre in the UAE - reportedly the first time in history a major cloud facility was hit in warfare.

The message to Gulf states is deliberate and cold: you are collateral in a war you didn’t choose, and the US umbrella did not protect you. Iran is not just fighting America and Israel. It is trying to fracture the coalition of silence that the Gulf’s acquiescence represents. Iran is striking at the foundations of the petro dollar and American hegemony in the region

4.4 — The petrodollar loop under existential stress

Recall the three pillars of the petrodollar:

1. Gulf states price oil in dollars

2. They recycle surpluses into US Treasuries ( and markets)

3. And they get the US to guarantee their security.

All three are cracking simultaneously.

The deep irony is this: the US went to war partly to preserve its Middle East influence. But the war is accelerating the erosion of the very system that underwrites American power. Iranian missiles have now struck Dubai, Doha, and Riyadh. The US umbrella did not stop them. Gulf states are drawing their own conclusions. A security architecture less dependent on Washington is no longer a fringe discussion (well, why the hell would Saudi do a defence deal with Pakistan otherwise).

If GCC states begin reducing their recycling of petrodollars into US Treasuries, American borrowing costs rise, and fiscal pressure intensifies. The exorbitant privilege starts unwinding structurally, not just at the margins.

Combined with everything happening on de-dollarisation, this war may have compressed a decade of dollar erosion into two or three years.

Part V — What this means for you, the investor

5.1 — The regime change in markets

Okay. After all of that, you’re probably asking — yaar, kehna kya chahte ho ( what are you trying to say)?

Here is what I am saying.

Everything that made “buy the dip” work for 40 years rested on four conditions. Falling interest rates - from nearly 20% in 1980 to near-zero in 2020. Expanding globalisation - cheap labour, cheap goods, low inflation. Dollar dominance recycling into US financial assets. And geopolitical stability - the post-Cold War peace dividend that let governments cut defence and spend on growth.

All four are now reversing simultaneously.

This is not an ordinary market correction. The rules that governed investing for four decades were not laws of nature. They were products of a specific world order. That order is being rewritten in real time.

5.2 — What is inevitable

Certain things are not predictions anymore. They are already in motion.

Deglobalisation is happening. Supply chains are being deliberately fragmented along geopolitical lines. Every major power is trying to reshore what it offshored. This is structurally inflationary. Making things closer to home costs more than making them in the cheapest place on earth. That cost doesn’t go away.

Manufacturing reshoring is a multi-decade capital expenditure cycle. Semiconductors, defence equipment, shipbuilding, pharmaceuticals, energy infrastructure - every government is now treating these as strategic, not commercial. The capital flowing into physical production over the next 20 years will dwarf what flowed into software over the last 20.

Hard assets are returning to favour. Gold has already been telling you this - it has been the best-performing major asset class since 2020, largely ignored by the average equity investor. Commodities, real infrastructure, energy - these are the assets that perform when inflation is structural, supply chains are fragmented, and the dollar is weakening.

The military-industrial complex is a structural theme. Europe is rearming. Asia is rearming. The US is scrambling to rebuild production capacity. Defence spending once committed does not get cut — it becomes a floor.

5.3 — What should an investor do

I want to be careful here. I am not telling you to sell your SIPs tomorrow. I am not predicting a crash next month. Markets can stay irrational longer than any of us can stay solvent, as Keynes reminded us.

But here are the questions worth sitting with before your next investment decision.

Does this investment benefit from cheap money that is unlikely to return? US tech, high-multiple growth stocks, and long-duration assets were priced for a zero-rate world. That world is gone.

Does it depend on globalised supply chains that are being deliberately disrupted? Businesses whose entire model rests on frictionless global trade face a structurally different environment.

Is it a financial asset or a claim on something real? In an era of dollar debasement and structural inflation, the distinction between owning a real asset and owning a financial claim on one starts to matter enormously.

Within India specifically, look toward businesses that benefit from supply chain diversification, domestic manufacturing, energy transition infrastructure, and defence.

And gold. If you have none, that is worth reconsidering. Not as a trade, as insurance against a monetary system under genuine stress for the first time in 50 years ( But Buffett has a disdain for gold and by virtue of that, me too).

Conclusion

The dip will come. It always does. The question is not whether to buy it - it is whether what you are buying will look the same on the other side.

The map that guided investors for 40 years was drawn in a specific world. That world is changing - not at the edges, but at the foundations. The dollar, the petrodollar, the globalisation dividend, the peace dividend - these were the invisible infrastructure beneath every investment return you have seen in your lifetime.

Understanding why they worked is the first step to understanding when to trust them - and when to draw a new map.

That is all I wanted to say.

A note before you go

If this changed how you see the world even slightly, please take 30 seconds to do these things.

Subscribe if you haven’t already. I write to make complex things simple and useful. If this landed, there is more where it came from.

Subscribe now

Restack this. You probably know someone - a friend, a colleague, a family member -who is blindly buying the dip right now without understanding the ground beneath their feet. This article took weeks of reading and days of writing to put together. One click from you puts it in front of someone who needs it. That one click costs you nothing and might genuinely change how someone thinks about their money.

Share

Like it. I know it sounds trivial. It isn’t. Every like tells Substack this is worth showing to more people. It is the simplest thing you can do.

Comment and tell me what you learnt. Seriously - what was the one thing that shifted something for you? Was it the petrodollar loop? The Plaza Accord? The Clean Break document? The Iran strikes and what they mean for oil? I read every comment. I respond to most. This piece was a departure from my usual company deep dives and I genuinely want to know if it was worth your time — and mine.

Leave a comment

I put an embarrassing amount of effort into this one. If it was useful, the best thing you can do is tell me — and tell others.

See you in the next one.

I’ve studied this business thoroughly but haven’t invested yet. Still, I’m publishing this because the mental models here are too good to keep to myself. A fascinating industry, a moat I found hard to ignore, and valuations that are hard to argue with. I spent weeks on this, talking to analysts, digging through filings, before finally putting it into words. Let’s get into it.

The mental model that brought me here:

A moated cash cow at the core, not a fast-grower, but one where the cash the business throws off is being deployed into interesting adjacencies, all within the moat. The moat keeps competition out, which buys the business time to get a foothold in these new areas and let the profits from those bets compound.

And all of this comes with meaningful downside protection.

Warren Buffett’s two rules of investing:

1. Never lose money

2. Don’t forget rule number one

The Downside Is Protected (Pun Intended)

Warren Buffett’s two rules of investing:

1. Never lose money

2. Don’t forget rule number one

I like my downside covered. Here’s what makes this interesting on that front:

• Sub-10 TTM P/E, and that’s at depressed earnings

• Dividend yield of roughly 6.8%

• 60% of market cap sitting as cash on the balance sheet

The moat isn’t just a story either, it shows up in the numbers:

• ROCE of 26.2%

• CMP / FCF of 7.98x

Now don’t look at the price chart and close the tab. Hang with me. If you’re a value investor, depressed valuations shouldn’t depress you, they should excite you.

The Elephant in the Room

If the return ratios are that good, why is this available at a sub-10 PE?

Let me borrow from Lord Parai here: because cash is a four-letter word.

This business is in cash logistics. And the moment you say that, the market’s eyes glaze over. UPI has taken over, nobody carries cash anymore, the whole thing is going to zero, blah blah blah.

Here’s what the data actually says.

Cash Isn’t Going Anywhere

Let me speak to the contrarian in you.

Currency in Circulation (CIC) as a percentage of GDP has stayed remarkably stable, even as the economy has grown. In rupee terms, CIC has gone from roughly 18 lakh crore in 2016 to somewhere between 36 and 39 lakh crore in 2025. It has quietly more than doubled in a decade.

Look at that chart. As countries get richer and more digitally advanced, cash as a share of GDP does come down, but it never disappears. And in India, even after Demonetisation, which the market called a death sentence for cash (and frankly, “Demon”etisation was the right nickname for it), CIC bounced back and kept climbing.

The narrative and the data are telling two completely different stories. The market is listening to the narrative.

So How Does Cash Actually Move?

Before we get to the company, let’s understand the industry it operates in.

The lifecycle of cash is actually pretty simple.

The RBI prints money => it moves to bank vaults => from there it flows to ATMs, => and then into the hands of the general public =>You spend it at a kirana store or a mall, => that cash then needs to find its way back to the bank. That return journey is what Reverse cash logistics companies handle. Here’s a small diagram showing the cycle

How the Industry Breaks Down

The cash logistics industry in India sits across three broad segments:

ATM Cash Management (~50%): Keeping ATMs stocked and running. Every time an ATM doesn’t run out of cash, someone in this industry made that happen.

Dedicated Cash Vans or DCV (~25%): Those armoured vans you’ve seen crawling around with armed guards? That’s this. Leased vehicles moving currency between locations, point to point.

Retail Cash Management (~25%): This is the reverse logistics piece. Cash collected from retail stores, malls, petrol pumps, collected, verified, and transported back to bank branches or processing centres.

Meet Radiant Cash Management Services

The company we’re discussing today is Radiant Cash Management Services. And they operate in the retail cash management segment, and boy do they absolutely dominate it. A whopping 40% market share. Biggest player in the segment.

Take a good look at that image. The retail cash management segment, the one Radiant owns, is the fastest-growing piece of the entire cash logistics pie. When I saw this, I sat up straight.

The market is busy worrying about UPI. Meanwhile, the segment Radiant leads is quietly compounding.

What Does Radiant Actually Do?

Radiant operates across multiple business segments. We won’t get into all of them today, we’ll focus on what matters. And the best place to start is the cash cow.

We’ll understand the business using a few simple questions.

The Cash Cow: Retail Cash Management

What Is It?

Radiant collects cash from retail touchpoints and deposits it into the retailer’s bank branch.

Say a Reliance Retail store closes for the day with 5 lakhs sitting in the cash register. They hand it to Radiant. Radiant deposits it into the store’s bank account. Easy peasy.

Who Uses It?

• Large retail chains like Reliance Retail

• E-commerce players (yes, you read that right, they pick up cash from COD deliveries, well someone has to)

• Petrol pumps

Why Is This Not Optional?

Idle cash is dead cash. Cash sitting in a store’s register earns zero interest. The faster it reaches the bank, the better.

Fungibility. Once the cash hits the bank, HQ can deploy it for operations immediately.

Theft risk. Even one theft incident wipes out money equivalent to roughly four years of CMS fees (This is good for the store managers’ BP as well)

Audit trail. A third party handling your cash creates a clean record of every rupee. For large retail chains, this is a compliance necessity.

The kicker. Cash is uninsurable unless handled by a licensed CMS player. If a retailer manages cash in-house and it gets stolen, there is no insurance. This makes organised CMS mandatory.

How Are Customers Won?

Banks offer CMS as a value added service to their retail clients. Banks are doing the customer acquisition for Radiant.

Why Do Banks Love It?

• F(r)ee income without deploying any assets. Free income with high RoA.

• CASA accretion from faster cash cycling and better NDTL management.

The Growth Runway

India has roughly 2 crore retail outlets. Less than 3% currently use organised CMS. In the US that number is 67%. Radiant services around 72,000 outlets today. The runway is long. Value migration from unorganised to modern retail and e-commerce is only accelerating this further.

If you’ve made it this far, you’re exactly the kind of reader I’m writing for. Subscribe so you don’t miss Part 2 where I get into the P&L, the risks, and whether the management deserves the trust the market is withholding.

Thanks for reading! Subscribe for free to receive new posts and support my work.

The Moat

We want the cash cow protected and printing cash till the adjacent businesses start firing. So what keeps competition out?

Regulatory Moat

You can’t just start a cash logistics business. You need a PSARA licence under the Private Security Agencies Regulation Act. On top of that, RBI operating standards mandate a minimum net worth of 100 crore, 300 vans, CCTVs, and dual-armed guards per van. This kills smaller players before they even get started.

Relationship Moat

Banks prefer incumbents with long operating track records. And the switching logic is asymmetric. No bank wants to risk its reputation to save a few bucks by switching vendors. And why would they switch anyway, when the incumbent is already the lowest cost provider with a clean track record? Once you’re in, you’re in.

Industry Structure

The industry is oligopolistic and consolidating. Top 3 players control roughly 70% of the organised market. Radiant alone holds 40% in retail CMS. Its a industry where, the big keep getting bigger.

Supply Side Moat

Radiant is the lowest cost provider in the industry with the least credit cost, meaning the least theft. Both of these reinforce each other and make it very hard for anyone to undercut them on price or on trust.

And Then There’s Route Density

This is the juiciest part. The moat deepens with scale. One concept explains why. Route density. We’re going deep on this next.

Route Density: The Moat That Keeps Getting Stronger

Every day, Radiant sends vehicles on fixed routes. Each vehicle collects cash from retail stores along the route and deposits it at the bank branch. The key operational metric here is : rupees collected per km or number of pick up points per Km.

The routes are largely fixed. So the more pickup points you have on a route, the better your economics. And those better economics get passed on to the customer as lower prices.

Once you’ve built high route density and acquired these customers, no one can dislodge you.

Here’s the math. Say a competitor comes in aggressively and somehow poaches 20% of Radiant’s customers on a route. Radiant still has 80% of the stores on that route versus the competitor’s 20%. Radiant’s cost per pickup is now roughly one fourth of the competitor’s. No retailer is shifting their cash pickup vendor to pay more. The economics just don’t work for the challenger.

Radiant today operates across 14,095 PIN codes with 70,000 pickup points. It took years to build this. Competition cannot replicate it overnight.

And here’s what makes this really interesting. For retail CMS, the incremental cost of adding a new pickup point is essentially zero. Every additional client goes straight to the bottom line. That’s operating leverage at its cleanest.

Moat = Regulation + Relationship + Network

Why Can’t Anyone Break In?

Smaller players can’t clear the 100 crore regulatory hurdle. Large players look at the operational intensity and the current industry size and decide it’s not worth their time. And as with most RBI regulated industries, incumbents are favoured. The industry has consolidated around the top players and it will stay that way.

The Optionality: DBJ Logistics ( Diamon bullion and Jewelry)

How do you further increase your operating leverage? Deploy capital within the moat to earn more using the same capital. Enter DBJ logistics play.

The company has entered the logistics of Diamonds, Bullion and Jewellery. I call it the double-sweat model and it’s less disgusting than my ability to coin terms. Let me explain.

Retail cash pickups happen during bank hours, 9 AM to 3 PM, for same day credit. What do you do with the vehicles and manpower for the rest of the day? You get them to pick up and deliver high-value goods like DBJ. Due to the sensitivity of the cargo, this logistics generally happens at night.

Same armoured van. Same security backbone. Same routes. Two revenue cycles per day. You’re double-sweating your assets, and that is the zero incremental capital moat. Route density increases further.

Why Radiant Has a Clear Right to Win

Low CAC: They already have an installed base of 1,000+ jewellers served for cash pickups. For them, this is a low CAC cross-sell.

Route density synergy: DBJ runs on the same routes as cash pickup, lowering cost per stop further. This not only gives them a cost advantage in DBJ logistics, it also reduces costs in the core business, strengthening the low cost moat further.

And as the logistics formalises, value will migrate from unorganised Angadias (who currntly do this) onto players providing insurance, GPS tracking and audit trails. More on this in a moment.

Where Radiant Plays: Last Mile

DBJ logistics happens in 3 stages.

• 1st mile: Bullion is imported into the country and sits in vaults,

• Middle mile: This moves to centralised hubs,

• Last mile: Then from hub to retail store.

Radiant operates in stage 3, last mile, where competition is thinner.

Radiant is bypassing the high margin, low frequency bullion game to capture the high frequency, operational retail game where route density provides a permanent cost advantage.

If Radiant captures 25 to 30% of organised last mile by dominating Tier 2 and Tier 3 towns, that’s a 100 crore plus annual revenue stream. And that calculation excludes the unorganised market entirely. As PMLA laws force local mom and pop jewellers to use formal couriers, the actual addressable market could expand by another 200 crore over the next 3 years. Capturing 25% of this would essentially double Radiant’s current profitability due to the massive operating leverage of existing cash routes.

Why This Shift is Inevitable: Value Migration

• Regulation: Jewellery and gold movement above 50,000 rupees requires digitally generated e-Way bills with GST invoice. Unorganised couriers historically avoided this. Crackdowns have begun.

• More Regulation: In May 2023, dealers in precious metals were brought under PMLA. End to end trail of goods is now mandatory. Non compliance carries penalties including 3 to 7 years imprisonment.

• Structural consequence: jewellers must now operate like banks, traceable, insured, compliant logistics. This structurally disadvantages informal players.

Regulation is forcing a shift from informal trust-based transport to institutional, auditable logistics. Exactly where Radiant is positioned.

That’s Part 1 done.

If you liked this, drop a like and comment on what you learnt.

Leave a comment

If you really liked it, please restack this so the effort reaches more people.

Share

Enough love and I’ll write Part 2 covering the P&L, risks, and management discussion.

SEBI Disclosure: I am not a SEBI registered investment advisor. This article is purely for educational purposes and is not a buy or sell recommendation for any security. Please consult your financial advisor before making any investment decisions.

The Tsunami: Energy Transition

You’ve heard this before. Everyone’s talking about it. I’ll keep this brief and quicly shut-up.

India’s renewable energy capacity has grown 20-25% annually over the past decade. With a target of 500GW by 2030 (up from roughly 200GW currently), we’re looking at approximately 15-16% CAGR over the next five years. This growth will persist as India’s GDP per capita rises, but there’s a wild card that amplifies the trend:

The AI Factor: As artificial intelligence drives productivity gains globally, nations are competing not just in algorithms but in the infrastructure to power them. Energy is the foundation of this race. Meanwhile, Europe and America face aging grid infrastructure requiring massive overhauls.

Multi-decadal theme = Rising domestic energy demand+ AI datacenter power requirements+ climate-driven transition mandates
+ deteriorating grid infrastructure needing replacement.

This shift transforms the sector from cyclical to secular for the foreseeable future. That’s our hunting ground.

Before we dive deeper, let me preview where this analysis leads. By the end, you’ll understand a company with these characteristics:

• Five-year earnings growth: 151%

• Nearly debt-free balance sheet

• High-integrity management team

• P/E ratio: 23.6

• Exceptional return ratios: ROCE and ROE both exceeding 50% over the past two years (currently inflated, we’ll deconstruct why)

• Positioned in a powerful multi-decade theme

When you find opportunities like this you say “You go into greed mode”

But not so fast. This will be a long journey.

If you’re like me and prefer finding one solid company to ride for years, thorough selection matters. If you want a quick read, turn back now. If you’re hunting for Excel projections or price targets, this isn’t for you. (Projections are useless. As Charlie Munger noted, no one was ever dissuaded from investing after seeing the projections.)

This analysis will take you deep into transformer components, value chains, and the technical fundamentals necessary for investing in this industry. If you’re looking for a quick trade, close this tab. If you’re casually scrolling for surface-level insights, I’d suggest stopping here.

This is for readers who want genuine understanding. I’m not an electrical engineer, yet I’ve understood this deeply. If I managed it, you absolutely can. Still reading? Let’s dig in.

We’ll explore how transformers work, their components, supply chains, where players sit in the value chain, and who captures the profit pool. The technical details matter. When a company consistently generates 50%+ returns on capital, we must determine whether it’s cyclical or structural. That requires understanding the underlying science to assess whether sustainable moats exist. Without grasping these concepts, proper analysis is impossible.

That’s where I’ll focus. For everything else, Screener or Tickertape are your friends.

Now that we’ve established why this matters, let’s understand the product itself.

Transformers: The Basics

What It Does

A transformer steps voltage up or down. Think of water flowing through a pipe. You need pressure difference to make water move. Electricity works similarly: voltage difference between two points makes current flow. Different parts of the power grid need different voltages. Apply too much pressure to a garden hose and it bursts. Same principle applies here.

How It Works

A transformer changes voltage through electromagnetic induction. It contains three core elements (refer to the image below):

• Primary winding (coil of wire)

• Magnetic core (special steel)

• Secondary winding (coil of wire)

When current flows through the primary winding, it creates a changing magnetic field in the core. This field passes through the secondary winding and induces voltage. More turns in the secondary than the primary? Voltage increases (step-up). Fewer turns? Voltage decreases (step-down). The core efficiently transfers the magnetic field between windings.

Key Components

Beyond the core and windings, transformers contain:

Cooling and containment: Oil insulates and removes heat. The tank encloses core, windings, and oil. Radiators release heat. Cooling fans and pumps enhance heat removal.

Operational control: The conservator stores extra oil for thermal expansion. Tap changers adjust winding turns to control output voltage. Bushings are insulated terminals carrying current in and out.

Safety systems: The breather keeps moisture from contaminating oil. Buchholz relays detect internal faults. Temperature sensors monitor heat levels. Pressure relief devices prevent tank explosions.

Let’s toggle science off for a moment and return to commerce. Time to understand the value chain and where profits concentrate.

Supply Chain Deconstruction: Where the Money Lives

The supply chain flows like this:

Raw materials → Component suppliers → Transformer OEMs → EPC contractors → Developer/utility.

Here’s how margins stack up at each stage:

Why Transformer OEMs Capture Outsized Value

Most transformer components are commodities: tanks, oil, radiators, sensors, basic steel and copper. Even some critical elements like standard bushings use globally available materials and designs.

However, specialized value-adding components resist commoditization: core design (especially advanced geometries), winding design (particularly Continuously Transposed Cable or CTC windings), sophisticated tap changers, and the overall thermal and electrical engineering integration.

The differentiation comes from engineering excellence, not parts procurement. The real value lies in optimizing core geometry, winding layouts, cooling systems, and insulation.

Done right, transformers run efficiently, manage heat and electrical stress properly, and meet project-specific requirements. Done wrong, they fail.

A transformer shoudln’t be though as just assembly. It’s the engineering that matters. Performance depends on design intelligence and reliability, not component quality alone. This is why transformer manufacturers earn 15-30% margins while commodity suppliers scrape by at 8-12% and EPC contractors struggle at 7-11%.

We'll take a short detour now. Feel free to skip this section. It breaks my flow, but I'm including it for completeness and to map the competitive landscape.

Types of Transformers: Who Sits Where

Shilchar operates primarily at the renewable generation-to-grid interface, where power from solar, wind, or hydro plants gets stepped up before entering the transmission or distribution network. About 60% of revenue comes from renewable transformers (especially solar inverter-duty and wind generator transformers), with the balance from distribution, power, and furnace transformers.

This positioning is unique. Renewable transformers aren’t commoditized. The game is different here. We’ll probe deeper to understand why margins are higher even within the already high-margin transformer segment. It’s a premium category within a premium category, protected by genuine moats we’ll explore shortly.

To understand why, we need to compare renewable transformers with their thermal power counterparts.

Thermal vs. Renewable Transformers

Both do the same basic job: step up voltage from the power plant before electricity enters the grid. The difference lies in how power is produced.

• Thermal plants produce steady, continuous power

• Solar and wind plants produce intermittent power that fluctuates with weather

This difference fundamentally changes transformer design requirements (I have over-simplified this to remove complications like harmonics, eddy currents etc to make it readable for everyone).

What Changes Inside the Transformer

Renewable energy swings between extremes: zero load during calm nights, high loads during peak sun or wind. Traditional transformers never face this uncertainty. They run at steady, predictable loads.

This creates technical challenges. Fluctuating loads generate eddy currents and harmonics that increase energy losses (ignore the gibberish). To minimize these losses, system design must be fundamentally different. Companies specify strict loss tolerances with financial penalties for breaches. Since renewable projects vary by location, weather patterns, and grid requirements, designs must be customized. Extreme standardization isn’t possible. I repeat

Designs must be customized. Extreme standardization isn’t possible.

Here’s what differs at the component level:

Core: Each core is built from hundreds of thin CRGO steel laminations, individually cut and stacked. The cutting pattern, stacking sequence, and gaps between laminations determine magnetic flux efficiency and loss levels. Getting this right requires precise engineering: wrong cuts or poor stacking create hotspots and efficiency failures. Every project demands different core geometries based on load profiles.

Windings: The challenge isn’t just conductor selection but configuration. Continuously Transposed Cable (CTC) interweaves multiple insulated strands to equalize current distribution and minimize circulating eddy currents. Winding geometry, turn ratios, and layer arrangement must be optimized for each project’s voltage fluctuation patterns and harmonic profile. Poor winding design causes localized heating, insulation breakdown, and premature failure.

Cooling: This gets highly project-specific. A solar farm in Rajasthan faces 48°C ambient temperatures and massive daytime load swings. A wind farm in coastal Tamil Nadu deals with high humidity, salt spray, and variable winds. Cooling systems (oil circulation rates, radiator sizing, fan capacities) must be engineered for each site’s thermal environment and load cycling pattern. Undersized cooling leads to oil degradation and winding failure. Oversized systems waste capital and energy.

Insulation: Built to survive repeated temperature swings without degrading.

Harmonics tolerance: Renewable power creates electrical distortions from inverters. Transformers must withstand these without performance loss.

Bottom Line

Renewable power is intermittent and passes through power electronics. Transformers must handle variable loads, harmonics, and harsh environmental conditions. This creates three competitive barriers:

Project-specific engineering: No two renewable sites are identical. Designs can’t be copy-pasted.

Strict efficiency requirements: Loss guarantees carry financial penalties. Meeting these demands deep engineering expertise and field experience.

Accumulated capabilities: Success requires design know-how, thermal and electromagnetic optimization skills, customer approvals, and certifications. These come from years of field performance, not hiring fresh engineers.

This is why renewable transformers remain specialized, difficult to replicate, and resistant to commoditization. It explains Shilchar’s premium margins.

Even as capex floods the sector and new players enter, technical differentiation sustains pricing power. Market fears of competition-driven margin erosion miss this point entirely.

The Margin Story

Shilchar’s margins stem from two sources: the specialized, custom-build niche they occupy and significant export exposure to international markets. Entry barriers are real. International certifications take years to achieve. Competitors can’t shortcut this process.

The value investor in me believes margins will hold.

Capital Allocation

Anyone who has followed Shilchar and read their concalls will say this unanimously: management is patient, maintains high integrity, and exercises exceptional capital allocation discipline.

They’re expanding capacity to nearly double current levels, funded entirely through internal accruals. Zero debt. They’re not rushing. Patient and disciplined.

Unveiling the magic of the high ROCE

Any company that consistently returns better than the cost of capital is an anomaly in capitalism. We must dig deep and deconstruct the magic they’re doing. Here’s a deconstruction

ROCE = EBIT margin × Asset turns

Let’s break this down properly.

1) EBIT margins

Broadly, this is how the structure looks:

Why their margins are higher

• Around 50% exports, which command better pricing

• Focus on renewable transformers, which are engineering heavy and not fully commoditised

• Custom, project specific solutions

• Performance sensitive applications where failure is costly, so customers pay a premium for quality

They operate in a niche where reliability and loss optimisation matter. That allows pricing discipline.

Now let’s turn to the next metric

2) Asset turns

Asset turns = Revenue ÷ Capital employed

Shilchar generates more revenue per rupee capital employed. Capital employed mainly includes:

• Net fixed assets

• Working capital

  • Inventory

  • Receivables

  • Less payables

Let’s further break each one.

A) Receivables cycle

Shilchar primarily sells to:

• Renewable EPCs

• Export customers

• Limited exposure to state utilities

Typical receivable cycles in the industry:

With roughly 50% exports and strong EPC exposure, Shilchar likely operates closer to the 45–90 day bucket rather than the utility bucket. Hence a clever picking of niche has provided for a better working capital cycle.

Lower receivables directly improve asset turns.

B) Inventory turns

Large power transformer companies manufacture very large units. They have:

• Long production cycles

• High work in progress

• Significant copper and CRGO locked for months

Shilchar focuses more on mid-sized renewable transformers:

• Faster throughput

• Shorter manufacturing cycle

• More units per year

This leads to higher inventory turns and less working capital locked per unit of revenue.

C) Fixed asset turns

Shilchar operates in the mid-voltage, mid-sized transformer segment, which requires lighter plants and moderate testing infrastructure. It does not need ultra-heavy cranes, massive fabrication bays, or very high-voltage test facilities. Capex per MVA of capacity is therefore lower. And Production cycles are shorter and units are smaller.

In contrast:

• Large power transformer OEMs manufacture very high-voltage, very large units.

• They require heavy engineering plants, large test bays, and extensive material handling infrastructure.

• Capex per MVA is significantly higher.

• Longer build cycles and bigger units reduce fixed asset turns.That increases capital intensity.

Shilchar generates more revenue per rupee of plant because it operates in a mid voltage, faster cycle segment.

Putting it together

High margins come from:

• Export mix

• Engineering heavy products

• Performance premium niche

High asset turns come from:

• Better receivable cycle

• Faster inventory rotation

• Lower capex intensity

High margin × High asset turns = High ROCE

This is absolute magic here. It is how smart positioning in terms of a customer and product segment that combines pricing power with capital efficiency creates for a fantastic business.

Putting It All Together

I pass on any business where I can’t see 5-10 years out. Here’s how Shilchar’s future looks:

Growth: High growth supported by energy transition, rising consumption, AI datacenter demand, and aging Western grids requiring replacement.

Return ratios: Sustained by niche positioning and genuine barriers (engineering expertise, multi-year certification requirements, field-proven reputation).

Balance sheet: Management has demonstrated fiscal discipline. Growth funded internally, zero debt.

Price: The market values this high-growth, high-quality business below 25x earnings. Perhaps it’s viewed as a commodity play. Perhaps near-term growth catalysts aren’t obvious. Perhaps fears of Chinese competition (misplaced, given Shilchar’s export certifications and technical moats) are suppressing the multiple.

Whatever the reason, for a value investor, it’s compelling.

If You Made It This Far

You’ve reached the end. I hope you learned something valuable.

If this analysis helped you, consider contributing at yourbro@ybl. Whatever amount feels right. I’ve spent significant time researching this and want to continue producing deep dives like this. Your contributions (even verbal acknowledgments, likes, or reposts) provide the incentive to keep writing.

“Show me the incentive, and I’ll show you the outcome.” — Charlie Munger

Disclaimer

This article is for educational and informational purposes only and should not be construed as investment advice. I am not a SEBI registered investment advisor. The views expressed are my personal opinions based on my own research. Investing in securities involves risk, including potential loss of principal. Please consult with a qualified financial advisor before making any investment decisions. I may hold positions in securities discussed in this article.

Subscribe now

Today, we’re going to talk about a certification business.

Before we get into the specific company, let’s pause and appreciate the incredible nature of certification businesses in general, and why I believe they represent one of the holy grails of structural moats for capitalists.

Why Certification Businesses Are Special

• Credibility creates pricing power: When high-value goods require certification, buyers prioritize credibility over price, enabling certification businesses to charge significant premiums and earn operating margins above 50%, levels that rival or exceed many software companies. Credit rating agencies illustrate this well, in times of stress, issuers want ratings that stand up in court and public scrutiny, not the cheapest option.

• Standards create entry barriers: Once two or three players become industry standards, entry becomes extremely difficult. In high-stakes ecosystems, no individual wants to make a career-limiting decision by backing a new entrant to save a few basis points, and no company wants reputational damage for appearing “cheap.” Once standards are set, ecosystems freeze around incumbents, as seen with Moody’s, S&P, and CRISIL.

• Supply-chain indispensability: Certification acts as the lubricant of the supply chain. Many transactions are simply not possible without a trusted, neutral third party validating quality. When credibility is essential, market participants gravitate toward the strongest certification, not the cheapest.

• Fixed-cost economics with asymmetric upside: Certification is a fixed-cost business where capacity is people, not factories. Incremental certifications cost almost nothing, allowing revenue to flow directly to the bottom line, while downside risk is limited since costs can be reduced quickly through downsizing, without stranded assets or idle plants. Now that we’ve dusted off the concept, let’s talk about the diamond hiding in plain sight.

The Numbers That Don’t Match the Narrative

A business this good shouldn’t be this cheap.

Look at these metrics:

• ROCE of 68%, you can almost smell the moat already

• Operating margins above 60%

• Profit growth of over 30% CAGR

• CFO to PAT ratio of roughly 0.86x

At this point, you’re probably leaning in and asking one question.
“What’s the P/E?”

Surely this must be a three-digit multiple business.

Wrong.

The P/E is, drumroll please, a mere 27.2x on a trailing basis.

So why is a near-monopoly with extraordinary economics in a structurally growing sector trading at such a valuation?

To answer that, we first need to understand what the company actually does.

What IGI Actually Does

The business itself is simple to understand.

IGI certifies cut and polished diamonds, and to a smaller extent, jewellery.
Their experts grade diamonds across dimensions like cut, colour, clarity, carat, and several other parameters.

They operate 31 laboratories across 10 countries, which conveniently makes them relatively insulated from tariffs. We’ll come back to geopolitics and Mr Trump later.

Industry structure matters here.
In natural diamonds, IGI has roughly 35% global market share.
In lab-grown diamonds, it is effectively a monopoly with around 65% global market share. Let that sink in for a moment.

Now comes the most important part of the thesis. The reasons why the stock has been punished and why this diamond is currently covered in dust.

The Bear Case, in Full

First, the biggest concern.
Lab-grown diamond prices have been falling sharply. Around 55% of IGI’s revenue comes from LGD certification. Falling prices lead to pressure on certification fees, and more worryingly, raise the existential question: if diamonds become cheap, will consumers even care about certification?

Second, the business acquired its Western subsidiaries, which reduced the operating margins of the Indian entity. This raised questions around capital allocation and promoter intent.

Third, speaking of promoters, this is a private-equity-owned company. Private equity funds eventually exit. That exit can introduce stock supply overhang, volatility, and uncertainty around timing.

Fourth, with the explosion of LGDs, some argue diamond wearing itself will reduce. Diamonds derived value from exclusivity and status signalling. If everyone has one, the flex disappears.

Finally, the purist argument. Lab-grown diamonds are “not real diamonds,” and this trend will fade.

Now let’s counter these one by one.

Please sit tight. This part is essential to the thesis and, yes, it’s long. Take a toilet break if needed.

Certification Was, Is, and Will Remain

Certification was there, certification is here, and certification will remain.

Here’s why.

To understand this, look at where certification sits in the diamond value chain.
Diamond origination, whether through mining or lab growth, is followed by cutting and polishing. After that comes certification, then wholesalers, retailers, and finally the consumer.

Crucially, certification is not designed for the end consumer. It exists for the entire value chain upstream.

Procurement departments buy diamonds in bulk using detailed specifications. These specifications define cut, shape, carat, colour, quantity, and importantly, the certifying authority itself. Without certificates, procurement teams have no way of knowing whether what they ordered is what they actually received.

Manufacturers cannot self-certify. That is a clear conflict of interest and carries no credibility.
Procurement teams also cannot manually check every diamond. Doing so would require in-house gemologists, is slow, tedious, and operationally impractical. Missing even a single faulty stone can damage reputations.

This is where incentives matter.
Choosing a cheaper certification offers zero upside and infinite downside. Manufacturers pay for certification, not procurement teams. If a cheaper certificate leads to even one dispute or fake stone, the first head to roll is in procurement. No professional takes that kind of risk.

Certification also enables long-distance and cross-border trade. A wholesaler in the US importing diamonds from India cannot physically verify every shipment. Certificates act as trust bridges, enabling faster, credible transactions. Without certification, manufacturers would struggle to sell beyond their immediate geography.

Certificates also lock in facts at the time of trade. This is critical for dispute resolution. If a wholesaler swaps a stone, or alleges a mismatch, certification determines accountability. Neither side is willing to operate in a grey zone.

Beyond trade, certification underpins grading, sorting, SKU standardisation, and inventory management. Procurement teams know exactly what they have ordered, what they hold in inventory, and what needs to be replaced. Without certificates, manufacturers would be left holding diamonds with indeterminate value. Standardisation enables liquidity and pricing across the entire chain.

Regulation adds another layer of inevitability. Lab-grown diamonds must legally be disclosed as such. Selling them as natural diamonds can be criminal. Certification becomes mandatory, not optional.

Insurance further reinforces this. Retailers insure their inventory. No certificate means no insurance. Certificates also include microscopic engravings on diamonds, functioning like unique identifiers. These help with tracking, claims, recovery, and fraud prevention.

Certificates are equally critical for refunds, replacements, and disputes across the chain. In markets like India, where substitution fraud is a real risk, certificates protect manufacturers, wholesalers, retailers, and insurers alike.

Finally, brand risk seals the argument. Retail jewellery chains charge premiums built on trust. No brand will stake decades of goodwill to save a few dollars on certification or downgrade to a weaker certifying authority.

The conclusion is straightforward.
Certificates are indispensable.

Even if consumers stop asking for them, the system will not. Even if supply increases sharply, differentiation by cut, craftsmanship, size, and quality ensures certification remains essential.

On this point, the case is closed.

My View of the Future of Diamond Demand

Humans will continue to broadcast life events. Proposals and marriages already dominate social media. With increasing westernisation, more men will go down on one knee with a ring in hand, and that ring will have a diamond.

We are psychologically wired to signal these moments. Nobody wants to cheap out at the most visible milestone of their life.

Lab-grown diamonds lower the barrier. You cannot tell an LGD from a natural diamond with the naked eye. In the US, over 50% of bridal rings already use lab-grown diamonds. What happens in the US tends to happen in India with a lag. This mental model has worked repeatedly for consumer trends.

Sociologists call this Sanskritisation. After colonialism, Western habits sit at the top of the social pyramid and percolate downward. Jewellery is no exception.

Lab-Grown Diamonds Are Here to Stay

Diamonds are carbon. Lab-grown or mined, they are chemically identical.

Geopolitics makes LGDs inevitable. Russia supplies roughly 35% of the world’s natural rough diamonds. Sanctions have choked supply. Demand for sparkle hasn’t vanished. Ethical sourcing concerns only accelerate LGD adoption. Conflict diamonds are a real issue, immortalised even in popular culture.

Sanction-led supply shocks combined with ethical demand create a one-way adoption path. Once LGDs go mainstream, there is no rewind button. Certification volumes already reflect this, growing at roughly 25% CAGR (25% CAGR for 10 years = 10X).

On Falling LGD Prices, I Disagree

Yes, LGD prices have fallen sharply (80%) here’s the graph.

But price floors are set by the weakest link in the supply chain. When margins compress everywhere, the segment with the lowest ROIC determines the floor.

That is exactly what is happening today.

The cutting and polishing industry in Surat, responsible for over 90% of global diamond processing, has been devastated. Players are shutting down, unemployment is high, and wages have halved from around ₹60,000 to ₹30,000 per month. This segment cannot absorb further price declines (source https://rapaport.com/magazine-article/the-quiet-tragedy-of-surats-diamond-industry/).)

This, in my view, creates a natural floor.

We see this reflected in the commentary of companies like Goldiam and IGI itself, and in industry reports such as those from ICRA.

Goldiam concall snippet

IGI concall snippet

I read many other reports like this one repot by ICRA on the CPD industry. It seems to suggest the same

Even with the fall in prices, the certification prices have not fallen in proportion (~12% as per my calculations, but I maybe wrong, but its still much much lower than the LGD price correction). It was a very small 1 time price correction that IGI have taken in their prices, that was kind of compensated by the volume increase.

The One Risk I Cannot Refute

Private equity ownership.

This is a genuine risk. PE exits can create volatility and uncertainty. There is no clever counter here.

Why India Will Drive the Next Wave

Adoption of LGDs in India will be driven by organised retail.

If you study jewellery economics, one key metric is the studded ratio, the proportion of stone value relative to gold. Gold pricing is transparent and low margin. Stones drive profitability.

The industry is structurally incentivised to push higher studded ratios. LGDs enable this. Management commentary across organised players already reflects this shift.

Putting It All Together

Diamond consumption will continue to rise. We will see a value migration from natural diamonds to lab-grown diamonds. Inflation and constrained discretionary spending will only accelerate the move to cheaper alternatives without sacrificing aspiration.

India is at an inflection point. LGD penetration is under 5%. That number will rise sharply. Diamonds are no longer restricted to bridal jewellery. Fashion jewellery expands volumes dramatically.

And who benefits most from this?

IGI, with virtually no competition.

Its primary competitor, GIA, is a non-profit and has been slow to embrace LGDs, almost ideologically resistant to mainstreaming them. Much like Paytm betting on wallets while UPI reshaped payments, GIA missed the curve.

Even so, IGI looks like a compounding machine with its moats reinforced, ready to surf the massive tail winds of this lab-grown diamond theme. 25% CAGR (current volume rate) for 10 years will give us a 10X. In that, my friends, we have a potential 10 bagger (with a potential to even re-rate once the LGD prices stabilise)

With humility, I acknowledge that markets are wiser than individuals. Perhaps there are risks I have missed.

But in my view, this remains a diamond in the dust.

When the dust settles, the diamond will shine.

Disclosure : I am not SEBI registered.The information provided here is for education purposes only.I will not be responsible for any of your profit/loss with this channel suggestions.Consult your financial advisor before taking any decisions.Tweets neither advice nor endorsement.