The next big transition in global steelmaking | The Daily Brief #381

ELI5/TLDR

The world is slowly switching from the old, coal-fired way of making steel to a cleaner, electricity-fired way. That new method runs on giant rods of graphite, and there are only a handful of companies on earth that can make them — two of the biggest sit in India. The industry just went through a brutal shakeout, and the survivors might be standing at the bottom of a cycle that once turned a tiny loss into a 30x profit in two years. The second story: when India quietly raised how much of your bank deposit the government insures, people who hoard “safe” money responded by dumping their PSU stocks — quietly nudging those share prices down by 5%.

The Full Story

Two ways to make steel, and why the old one is on the clock

We’ve been making steel at industrial scale for almost two centuries. You’d think the recipe is settled. It mostly is — and that’s the problem.

The classic method, the Bessemer process from ~180 years ago, happens in a blast furnace. You blast iron ore and coking coal with superheated air to get a molten, carbon-rich iron slurry, then blast it again with hot oxygen to burn off the impurities. What’s left is steel. It works, it’s cheap, it built the modern world. It also has two ugly traits. First, it’s a carbon machine — two to three tons of CO₂ for every ton of steel. Second, you can’t turn it off.

It stays on for years on end, like a man-made volcano. Let it cool and everything inside solidifies into a rock-like mass that’s stuck to your furnace.

So blast furnaces are welded to fossil fuels by design.

The alternative is the electric arc furnace. Picture a pile of steel scrap with giant graphite rods — twice as tall as a person — dangling above it. You push an absurd electric current down those rods, 100,000 to 150,000 amperes. At that intensity the current doesn’t bother with the metal touching; it leaps through the air, turning the air itself into plasma and heating it to 3,000–4,000°C. That heat melts the iron.

The payoff: as little as one-tenth the carbon of a blast furnace, and you can switch it off whenever you like. As electricity gets cheaper, it gets cheaper too. It used to be limited to recycling scrap, but we’re learning to run it on fresh ore as well. Today it makes ~30% of the world’s steel; the research house MK reckons that hits 40% by 2030.

The chokepoint nobody thinks about: graphite

Here’s the catch. Electric arc furnaces only work because of one material: graphite. It’s the one thing that can carry thousands of amperes while surviving 3,000°C — and, oddly, it gets stronger as it heats, where most metals would melt. MK calls it “steel’s gatekeeper.”

Making these electrodes is a near-impossible business to enter. A mid-sized plant costs roughly a quarter of a billion dollars. The manufacturing is fiddly, and a badly made rod at those temperatures can blow up your furnace — so customers put you through a two-to-three-year qualification process before they’ll buy a single rod. That combination (huge capex, deep know-how, long trust-building) means the industry constantly drifts toward a few big survivors.

India punches well above its weight here. Two of the world’s largest makers — HEG and Graphite India — are based here. HEG runs the single largest graphite electrode site on the planet. Between them they make ~200,000 tons a year, about 40% of global capacity outside China.

The Darwinian reset

The past few years have been miserable for the industry. Demand was tight, and any plant that was high-cost or financially weak simply died — roughly 150,000 tons of capacity left the market. MK’s word for this is a “Darwinian reset.” The silver lining is who’s left: only the lean, disciplined operators survived.

The deeper problem has been raw materials. The best electrodes need needle coke — an ultra-pure carbon where the atoms line up in tiny parallel needles (hence the name). That orderly structure means it barely expands when heated, so it can cycle from room temperature to 3,000°C without cracking. Needle coke is 60–70% of the cost of an electrode. And it’s made by only a few specialists, so supply is rigid.

Then the world went electric. Lithium-ion batteries also crave high-grade graphite — and the best of it is made from needle coke. So suddenly EV battery makers were competing for the same scarce material. By some projections, half the world’s needle coke could be feeding batteries by the late 2020s. The result is a tiered market: whoever can secure cheap needle coke wins. India’s two giants have historically imported it, though Indian Oil recently opened a domestic plant, easing the squeeze a little.

Has this movie played before? Yes — and it was spectacular

A decade ago the industry was in an identical hole. China had flooded the world with blast-furnace steel, killing electric-arc demand and shutting electrode plants. Then China launched its “blue sky war” against pollution, ruthlessly closing dirty factories — including its own electrode and steel plants — and pushed many of its steelmakers toward electric arc furnaces.

For the survivors, this was a windfall. Competition vanished while demand exploded.

In the space of a single year, they went from $2,500 per ton to approximately $15,000 per ton.

The numbers at HEG tell it: a ₹44 crore loss in FY2017 became a ₹1,099 crore profit in one year, then nearly tripled to ₹3,026 crore by FY2019. Things normalized after that, but the industry came out structurally stronger.

The open question is whether 2017 can repeat. The setup rhymes: 110 million tons of new electric steel capacity is slated to come online; Europe’s carbon border tax (CBAM) now penalizes dirty blast-furnace steel; and countries are walling off Chinese imports — Japan slapped a 95.2% duty on Chinese electrodes. More demand, fewer competitors, weak players already gone. MK thinks the cycle may be turning. That’s the bet.

Story two: how deposit insurance quietly moves the stock market

The second story is a neat piece of behavioral finance. Every Indian bank deposit is insured by a government body (DICGC) up to a limit — ₹1 lakh until February 2020, then raised to ₹5 lakh.

A paper by Balakrishnan, Limodio and Vats found that this limit acts as a psychological kink. If you have ₹15 lakh and only ₹5 lakh is insured, two-thirds of your money is technically exposed. Some people respond by keeping exactly the insured amount in the bank — the authors call them “bunchers.” A high share of bunchers means a lot of customers quietly assume their bank could fail.

The twist: bunchers aren’t risk-lovers. They’re the opposite. They want more safe assets than the insurance limit allows, so they’re forced to park the excess in stocks, mutual funds and PPF — and they favor liquid ones they can sell fast.

When the limit jumped to ₹5 lakh in 2020, it created a natural experiment. Bunchers — specifically the ones who already traded — piled the freed-up money back into bank deposits (mostly liquid savings accounts). And over 70% of that money came from selling shares, not from cutting spending or borrowing.

Which shares? Overwhelmingly PSU stocks — NTPC, ONGC, Coal India. Before the change, bunchers were unusually heavy in PSUs. Nothing else about the stocks mattered — not sector, not volatility, not dividends. Only state ownership. The reason: a government-owned company feels like a safe asset, a homemade substitute for the insurance they couldn’t get. Once real insurance arrived, they didn’t need the proxy, so they sold.

The selling was big enough to push PSU prices down ~5% — pure flow, nothing to do with the businesses — before other buyers absorbed it within a month. The lesson for regulators: deposit insurance policy doesn’t stay inside banks. It ripples into equity markets.

Key Takeaways

Blast furnace vs electric arc furnace is the central divide in steel: the former is fossil-anchored, can’t be switched off, emits 2–3 tons CO₂/ton; the latter runs on electricity, emits ~1/10 the carbon, and can be turned on and off.
Electric arc furnaces make ~30% of global steel today, projected ~40% by 2030, driven by carbon regulation (Europe’s CBAM) and cheaper electricity.
Graphite is the irreplaceable chokepoint — the only material that conducts huge current at 3,000°C and gets stronger as it heats rather than melting.
The electrode business is a fortress: ~$250M per plant, deep know-how, and a 2–3 year customer qualification process — which is why it consolidates into a few survivors.
HEG and Graphite India together hold ~40% of global electrode capacity outside China; HEG runs the world’s largest single site.
Needle coke is 60–70% of an electrode’s cost, made by a few specialists, and now contested by EV battery demand — possibly half of global supply going to batteries by late 2020s.
The 2017–19 cycle: China’s “blue sky war” wiped out competition while demand spiked; electrode prices went $2,500 → $15,000/ton; HEG swung from a ₹44 cr loss to ₹3,026 cr profit in two years.
Bunchers = depositors who keep exactly the insured limit in the bank. A rising share of them signals declining trust in banks — a potential early-warning metric for bank-run risk.
Risk-averse Indians built a homemade safe asset out of PSU stocks (implicit government backstop). When deposit insurance rose to ₹5 lakh, they sold PSUs and moved cash to banks.
A purely policy-driven flow (no business change) pushed PSU prices down ~5%, recovering within a month — deposit-insurance changes spill directly into equity markets.
The buncher base is middle/upper-middle class (inverted-U vs income) and concentrated where trust in government is high or trust in banks is low.

Claude’s Take

The steel story is the stronger of the two, and it’s a genuinely good piece of explanatory journalism — it takes an industry most people have never heard of (graphite electrodes) and lands the “this is a real chokepoint” point cleanly. The chemistry is handled honestly. Where you should keep your guard up is the investment subtext. The whole segment leans on a single research house (MK), and the entire bull case rests on “history rhymes.” It might. But the 2017 super-spike was a freak event triggered by China’s specific, one-off pollution crackdown. CBAM and anti-dumping duties are real tailwinds, but they’re gradual policy grinds, not a sudden supply shock — so “another 2017” is a possibility being dressed up as a base case. The needle-coke / EV-battery competition is the more durable, more interesting structural story, and arguably more important than the cycle-timing question the video fixates on.

The deposit-insurance story is the sleeper. It’s a clever academic finding made digestible: people who look like stock investors were actually frustrated savers in disguise, and a banking policy lever had a measurable, if temporary, effect on PSU share prices. The “bunchers as a bank-run early-warning signal” idea is the kind of thing that’s genuinely useful and rarely discussed. My only caution is the usual one with this kind of natural-experiment paper — a 5% move that fully reverses within a month is real but modest, and it’s easy to over-narrate. The video mostly resists that temptation.

Seven out of ten. Clear, well-structured, no hype in the delivery even if the framing tilts optimistic on steel. Two unrelated stories stapled together is the format’s nature, and both earn their place.