Reliances Eight Perfect Minutes And A Sixteen Year Old Sebi Case The Daily Brief 478

TITLE: Reliance’s eight perfect minutes and a sixteen year old SEBI case | The Daily Brief #478 CHANNEL: Markets by Zerodha DATE: 2026-06-03

---TRANSCRIPT---

In today’s episode, we’ll break down two important stories. First, we’ll talk about an 8-minute fraud and then we’ll talk about whether the ATMs in India are dead. Welcome back to the daily brief by Zeroda where we cut through the noise to help you understand what’s actually happening in the most important stories from business and markets. I am your host Axara and today is Wednesday 3rd June.

Coming to the first story. So it was a late November afternoon back in 2007. 8 minutes were left on the clock. It was the last day the November futures on Reliance Petroleum would trade. Everything now depended on how much the shares sold for during the day’s last half hour. Once the market shut, those futures contracts would all cash out at the average price for the period. 3/4s of Reliance Petroleum was owned by its storied parent company Reliance Industries. But that afternoon, the parent had taken a large bet against its share price. Using 12 different entities as a front, it has built up massive short positions in the company. And all of these would pay off if Reliance Petroleum’s shares fell. The bigger the fall, the more they would profit. And then Reliance Industry seemed to cause the very fall it would profit from. Over those last 8 minutes of trading, it sold 1.95 cr Reliance Petroleum shares into the market and prices tanked. The shares average price for that closing half hour crashed. All the futures Reliance Industries bought suddenly proved to be spoton. By the time those 8 minutes ran out, the futures had generated gains of rupees 513 cr. Now to Sebie, it all seemed like a conspiracy. But was it? That question is the subject of a recent order from the Supreme Court.

So let’s go back a year. Back in 2006, Reliance Petroleum had listed at about rupees 60 a share. But then its price kept rising until by October 2007, it traded at nearly rupes 250. By now, analysts uniformly considered it overvalued. With the market at euphoric highs, Reliance Industries decided to pair its stake down by 5% or roughly 22.5 cr shares. But that was a problem. It was hard to sell so many shares without wrecking the price. Now, they couldn’t sell them all at once. No company could survive that. Their best option was to sell it over the course of weeks, slice by slice. While they sold though, they were vulnerable. So if the shares price was to fall halfway through that series of sales as seemed imminent at the time, every sale afterwards would yield a pittance. There was only one way to solve the problem though. They could hedge as they sold. They could also take a side bet which would pay out if share prices fell. That would effectively let them have their cake and eat it too. If share prices stayed steady, you could profit by selling them. If they fell, you could profit from the side bet. And so to protect its sale of 22.5 cr shares, Reliance Industries took short positions on 9.92 cr shares, less than half of what it was selling. Only Reliance Industries wasn’t forthright about any of this. And the rules didn’t allow them to dominate the derivatives market. They capped how large a position a single client could take. And so instead of taking those positions itself, it chose to do so through 12 other entities. Each of them signed the same agreement. They would trade only on Reliance Industries instructions and every rupee of profit or loss would belong to Reliance Industries. These looked like free agents to the market, but they all led back to Reliance Industries. According to Sebie, collectively 93.63% of the open bets on Reliance Petroleum’s November futures featured one of these 12 entities as a party. If that were true, they were effectively the entire market. So to Sebie, when you took all this together, the multiple fronts, the cornered markets, the simultaneous positions across share and futures markets, the last minute selling, it looked almost like Reliance Industries was trying to defraud the market. And so it charged the company with market manipulation. The company was ordered to surrender what to sebi were its illgotten gains nearly rupees 450 cr along with interest and this was later upheld by the securities appellet tribunal by a 2:1 majority.

So here’s the thing there’s little doubt that Reliance had broken the rules through those 12 entities. It had indirectly taken a bigger position than it was directly allowed to under the law. But what did this imply? So, a breach of the law is a very different thing from outright fraud. Think of it like this. Imagine a parent that tells their teenage child to make sure they’re home from an evening out with friends by 9:00 p.m. That’s a rule. If they break it, they might be punished. There are some things though that could make it much, much worse. If the child wasn’t home because they were drinking alcohol, for instance, that would be a much worse problem. And the fact that they were home late would be secondary. This would be a monumental breach of their parents’ trust. and the consequences would be far more severe. So what was it? Did Reliance Industry simply violate a rule or did they defraud the markets outright? Was it simply trying to take a larger position than it could? Or was it trying to rig the entire market for profit? The former might involve a penalty. The latter would bring severe action. So Sebi asked the court to look at everything together and treat it as fraud. The overly large futures positions Reliance had taken to the regulator were part of a larger conspiracy. It was doing so in order to capture the market, sell its shares at the last moment, and make obscene profits out of the resulting price drop. But to do so, Sebi had to establish three claims. That Reliance built its position in order to manipulate the market, that it cornered enough of the market to pull it off, and that those last 8 minutes were the smoking gun, the moment when Reliance Industries went for the kill. Now to begin with, Sebi argued that Reliance Industry’s futures position wasn’t a hedge at all. And by the time November came to a close, Reliance had already sold most of what it meant to. So why then was it holding on to its futures? But the Supreme Court didn’t buy this. The short positions Reliance’s web of agents had taken total to less than half of the shares they wanted to sell. Sure, as they kept selling shares through the month, their future positions remained, but they were under no compulsion to unwind their positions with every share sale. At the very least, it was hard to call this manipulation. Reliance Petroleum share price was falling exactly as they feared and it was just sensible to hedge against that drop. But it was worth asking anyway, what if Reliance was the entire short end of the market? Traders on the other side may have thought they were in a fair market with traders just like themselves expressing their views on where the market would go. Only in reality, every single trader that thought prices would tend down was actually Reliance in disguise. That was practically the case with Reliance Petroleum’s November futures. 93.63% of short positions in the market belong to one of Reliance’s 12 agents. Wasn’t that suspicious? It wasn’t so to the Supreme Court. For one, their share of November futures had fluctuated through the month and only peaked at above 93%. When most others were no longer willing to sustain such a bet, there were also other future contracts to trade on for subsequent months. And in total, Reliance’s web of agents were on the short side of 40% of all open future contracts. Now, this wasn’t fair. Reliance had absolutely taken far more bets than it was legally allowed to. It was in breach of a rule and would have to pay for it. But was it positioning itself for fraud? It was big enough to pull one off if it wanted to. But did it actually try to do so? That was the big question. Now, it wasn’t enough for Sebie to just show that Reliance made money off those trades. It had to draw out the whole mechanism. How Reliance wanted to push the price down, that it actually did so, and that it actually benefited from that drop. For one, when you zoomed out, Reliance wasn’t just trying to crater the price of Reliance Petroleum. If it really wanted to do so, it would wait till the market hit a bottom and then sell heavily to precipitate a route. But it didn’t. Instead, Reliance seemed to have a price floor in mind. It wanted to sell if each share fetched at least rupees 210. But when prices fell below that, it stopped selling. In fact, just a few days ago, prices had gone down to rupees 190 and Reliance held off. Then it only sold its shares finally when the price recovered to rupees 224. This wasn’t the behavior of someone trying to manufacture a bad price, but someone trying to take advantage of a good one. Two, it wasn’t just Reliance’s agents that were trying to unload Reliance Petroleum shares in that last half hour the November futures were live. Others were selling heavily too, dumping over a crow shares in the market, more than half of what Reliance sold. ZBI hadn’t looked at why there was such heavy selling and it didn’t examine what it did to the shares price. If so many others were selling too, was Reliance alone responsible for crushing the market? More importantly, if Reliance Petroleum’s share fell drastically, nobody would be affected as badly as Reliance itself. It owned more than 2/3 of the company after all. Its shorts might earn a temporary profit, but it would lose many times that amount in all the remaining shares it held. So, would Reliance deliberately destroy the value of one of its companies for what was comparatively pocket change? To the Supreme Court, this only looked like manipulation until you saw the facts closely. Then the case fell apart.

Now, to be fair, the Supreme Court didn’t exonerate Reliance Industries. It absolutely wasn’t supposed to corner so much of its market. Its secret web of 12 entities was not all right. It was an attempt to bypass a legal restriction by doing something indirectly. Reliance broke a rule, and this was something it should be penalized for. And so the Supreme Court let Sebie’s penalty stay. The court did find, however, that Seby’s fraud case was a stretch, at least for the evidence it had put together. Reliance may have taken massive positions through an opaque structure, and it may have timed its trades a little too well. But none of this by itself demonstrated that Reliance Industries had manipulated the market. Sebi couldn’t ask for its rupees 500 plus cr gains to be disorgged and it would have to return the rupees 250 cr alliance industries had deposited with it. Ultimately it boiled down to evidence. Sebi can’t just point to an overly perfect trade as evidence of wrongdoing. After all, most profitable trades can be made to look sinister in hindsight. Seby’s job wasn’t to poke at oddities, but to prove that there was manipulation. On this day, there wasn’t enough proof. Without it, the charges simply wouldn’t stick. For all the sources mentioned in this video, don’t forget to check out our newsletter. The link is in the description.

Coming to the second story, when was the last time you took cash out of an ATM? The existence of UPI has at least minimized, if not outright, removed the use of ATMs for many people in Indian cities. But believe it or not, India is sitting on more physical cash today than it has at any point in its history. According to the RBI’s currency in circulation data, the total value of bank notes in circulation in India was 41 lakh cr. That’s actually more than double the level just before demonetization in November 2016 when bank notes in circulation stood at approximately rupees 18 lakh cr. The currency that denomination was supposed to remove has more than come back. Now UPI didn’t kill cash. It took over the small fast daily transactions. the local tea or coffee shop, the auto, the kirana shop, and so on. But large parts of the country still run on cash for reasons that have little to do with UPI’s technical capabilities. In much of rural India, where the only bank branch is ours away, cash is just what works. Many of the people transacting in cash like agricultural laborers, daily wage construction workers, and small traders in rural and semi-urban India don’t even have working bank accounts or have lost trust in them. A lot of transactions in real estate and the unorganized economy are settled in cash precisely to stay outside the formal banking train. And then there’s the question of what kind of cash this is. Of the 41 lakh cr rupees and banknotes, roughly 35.2 lakh cr rupees is in 500 rupee notes alone. Now the 500 rupee note isn’t the one you use for everyday spending. It sits in wallets, drawers, and almiras. In a 2023 research paper, the RBI concluded that digital payments are now substituting cash for everyday transactions, but the store of value motive for holding cash remains intact. Cash in India has grown not because people are transacting in it more, but because they’re holding more of it. Now, the demand for physical cash is very much alive. What UPI has changed is how people get hold of it and because of it, the economics of running ATMs. After all, several ATM operators, including one of the two largest companies that ran the country’s ATM network, have collapsed in the last few years. But the chain of events isn’t as straightforward as you’d think. And that’s what we’ll be getting into in this story.

So, let’s start with the fundamentals of how ATM operators make money. Say you have an HTFC bank account. You walk up to an SPI ATM and withdraw rupees 10,000. Behind the scenes, HTFC Bank, which is your bank, pays SBI a small fee for letting you use their machine. This fee is called the interchange fee. Now, the interchange fee is the source of margins for an offsite ATM, which is an ATM not attached to a bank branch. An ATM exists because somebody runs them and earns 19 rupees every time you use one. Sometimes the owner is a bank itself, paying a third party to manage the machine on a per transaction basis. Sometimes the owner is a separate non-bank company called a white label operator like Indicash or Hitachi Money Spot whose whole living comes from interchange fees. The model works as long as enough people use each machine repeatedly. But the fee has barely moved. It was 15 rupees from 2012 raised to 17 rupees in August 2021 and then to 19 rupees in May 2025. Two hikes in 13 years since UPI took over small daily transactions. An ATM that used to do 200 transactions a day now does a fraction of that. And at the same time, the costs of running an ATM machine, from the rent on the ATM cubicle to the diesel for the cash van that comes to refill the machine, have all gone up. Perhaps the biggest cost increase license wages. The new labor codes which standardize minimum wages and benefits across India are pushing up the cost of ground level workforce that runs this industry like drivers, security guards, and cash custodians. The Currency Cycle Association, which is the industry body for cash operators, told Business Standard in May 2026 that the industry is now in a force majour situation. That’s a term for when a business can no longer operate as designed because of forces it can no longer control.

So the first major domino that fell victim to a force majour situation was AGS Transact Technologies. AGS was the second largest ATM managed services provider in India. At its August 2021 peak, it had installed, maintained or managed around 72,000 ATMs and cash recyclers across the country. It went public in January 2022 at rupees 175 a share. But by mid 2025, the company found itself in freefall. In early 2025, nearly 38,000 ATMs serviced by AGS went dark, including 14,000 SBI ATMs, India Post ATMs, and Yes Bank ATMs. On August 25th, 2025, the National Company Law Tribunal in Mumbai admitted AGS to corporate insolveny and by February 2026 admitted creditor claims against the company had reached Rs 13,171 cr. The stock was at 5 rupees, down 97% from its IPO price. AGS was the most dramatic failure, but not the only one. White label operators have been hit just as hard. Since they only earn from Interchange and run no other line of business, a 19 rupee fee that no longer covers cost is an existential problem rather than a margin question. Data Communications, for instance, divested its white label network India to Find India in March 2026, exiting the business it had pioneered in 2013.

So, what does this mean for the ATM? The machine itself is changing. The old ATM was a cash dispenser. The new machine is a cash recycler, which still dispenses cash, but also accepts deposits, so small business owners can deposit their daily takings without queuing at a branch. The same machine can also do video KYC for a new account, issue a fresh debit card, run a cross-ell on a personal loan, or update a passbook. In effect, the new ATM sounds a lot like a partial bank branch with no teller. And that’s not coincidental either. Banks now often use the ATM as a way to offload work from their branches. Anything you used to do at the teller’s counter, except wealth management and lending, you can now do at an ATM. On the latest earnings call of CMS Infosystems, the largest cash management company in India, Chief Business Officer Anush Rakavan explained what this looks like in practice for one of its clients, ICICI Bank. 2 years back, the ICICI Bank had, I think, about 20% cash recyclers and 80% cash dispensers. 2 years since now, since we’ve been working with them, it has gone the other way. We are 80% cash recyclers and 20% dispensers. The contracts under which ATMs are being built are changing too. In the old model, the operator was paid a fee per transaction. So when transactions dropped, operator revenue dropped. Now the new model is a fixed monthly fee per machine. That inherently means the bank takes some of the risk of the transaction volumes being low and the operator gets paid more predictably. In fact, CMS CEO Rajiv Call said on the same earnings call that for all practical purposes, the transaction fee model is dead in the ATM business. CMS itself walked away from a rupees 700 cr contract at 19 rupee per transaction equivalent to 75 cr rupees of annual revenue because the unit economics over the 7-year contract life didn’t work. In FY26, CMS signed long-term fixed fee contracts with SBI, ICICICI Bank, and HDFC Bank that together secured 85% of its FY27 revenue target. With ages gone and smaller white label operators struggling, CMS now holds a whopping 46% market share in India’s ATM cash management.

So, with recyclers taking over dispensers and fixed fees replacing per transaction fees, what does the broader picture look like? See, a cash recycler costs roughly 6 lak rupees to install versus 3.5 lak rupees for a pure dispenser. The economics of these expensive multi-function machines only work if enough people use them for enough things, which means installing them where banks already have customers and where the recycler can act as a branch substitute. In practice, this means the new ATMs are going into urban and tier 1 to tier three locations. Now consider the machines that are being shut down. As per the RBI’s report on trend and progress of banking in India between FY24 and FY25, on-site ATMs grew by roughly 5,000 while off-site ATMs declined by nearly 8,000. So off-site is where you find ATMs not attached to a bank branch like a patrol pump or a small town market street. These are also the ATMs that white label operators who run most of their machines in seburban and rural areas have always depended on. According to an icer consultation report, just 21.6% of India’s ATMs are in rural areas and India has 21.4 ATMs per 1 lakh adults compared to a global average of approximately 39.5. Most of the rural cash access machines that do exist depend on the old outdated interchange fee model. But banks haven’t abandoned rural cash access. They’re just far less reliant on ATMs. For one, there are 22,000 regional rural bank branches or RRBs across 730 districts. Secondly, business correspondents who are individuals authorized by banks to provide basic banking services through their own shops now operate the majority of India’s banking outlets in branchless mode. And then there are microATMs which are handheld biometric or card-based devices through which a Kirana shopkeeper can dispense small amounts of cash from their own dill. These grew to nearly 1.5 million units as of 2024. Now, these substitutes work, but they’re different products. A microATM caps your withdrawal at around rupees 10,000 and depends on the shopkeeper having that much cash on hand, which they often don’t. A business correspondent operates during their own business hours and is dependent on one person’s reliability and connectivity. An RRB branch has fixed hours and cues. None of them is the equivalent of a 24/7 ATM. So, the new infrastructure looks more suited to the areas they’re in. But it’ll be a long while before we get to know whether the substitutes are as good as the off-site ATMs being lost. But what is clearer than ever is that the Indian ATM is no longer the humble lumpy cash machine we have known it to be for so long.

Now coming to the tidbits. JSW group has finalized plans to set up a 30 gawatt hour lithium ion cell manufacturing facility with a joint venture partner investing over $1.3 billion across two phases. The company is evaluating Southeast Asian countries for the plan to leverage India’s free trade agreements and will import sales from China in the interim as it aims to deliver its first new energy vehicles by late 2026 or early January 2027. Coming to the next tidbit, on June 5th, Marauti Suzuki will launch India’s first E00 vehicle. A car that runs on nearly pure ethanol, marking a step beyond the country’s existing E20 blending program. While industry executives estimate E00 could cut per kilometer fuel costs by 25 to 35%, the bigger challenges remain. Limited refueling infrastructure, lower energy density than petrol, and consumer unfamiliarity with flex fuel technology. Coming to the final tidbit, realy firm Anantraj has signed an MOU with the Hana government to invest rups 20,000 cr in large sales data center infrastructure across the state expected to create around 6,000 jobs. This is over and above its existing 307 megawatt expansion pipeline across Manisur, Panchula and Ry and follows a rupees 4,500 cr commitment in Andhra Pradesh 8 months ago. That’s all the news I have for you. Thank you so much for watching and see you in the next one. Disclaimer. This content is forformational purposes only. None of the stocks, brands, or products mentioned are recommendations or endorsements.