Trimming the Chinese Wall

0:00In today's episode, we'll break down two important stories. First, we'll talk about India easing its border investment wall, and then we'll talk about brains being worth more than shares in defence. 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. If you're listening to this on your commute, on a walk, or at the gym, you can also find the Daily Brief as an audio podcast on Spotify, Apple Podcasts or wherever you listen to your podcasts. I'm your host Akshara, and today is Monday 16th March. Coming to the first story. So last week, the union cabinet approved its first big relaxation to the storied press note 3 of 2020. The sweeping restrictions India imposed back in April 2020, just as the COVID-19 crisis had hit us on foreign investment that was coming in from any of our bordering countries. In practice, press note 3 was targeted at a single country.

1:01It was India's China investment wall. It forced any investment with even a trace of Chinese beneficial ownership to seek government approval, no matter its size, sector or intent. And with our relationship with China soon turning rocky, this is how things remain over the next six years. Until this amendment came in, that is. Only immediately after the relaxation, the government had to clarify, this easing was not meant for Chinese firms. After all, the media had already rushed out stories on how this marked a thought between India and China, while politicians painted pictures of India capitulating to external pressure. Reality was hardly as spicy. This is actually a story about how well-intentioned policies drawn clumsily can have consequences that draft us never imagined. You might want to stop opportunities from taking advantage of a temporary weakness, but you could accidentally lock your own economy out of some of the world's most liquid pools of money or some of its most advanced technologies. Until April 2020, Chinese investors could put their money into India freely, just like

2:05American or Japanese ones could. The only investments we scrutinized were those from Pakistan and Bangladesh. But then, COVID fell upon us. The markets tanked and they stayed down for more than a month. It felt like an apocalypse had descended on us. Everyone rushed to withdraw their money, but to some, this looked like a once in a lifetime opportunity. On April 12, 2020, a regulatory disclosure revealed that the People's Bank of China, China's Central Bank hiked its stake in HDFC from 0.8% to 1.01%. This felt off. HDFC controlled a sprawling empire, India's largest private bank, a massive AMC and an insurance firm. It was a gateway to India's financial plumbing. So why was China's Central Bank putting so much money here, even as the broader market had crashed 40%, surely they weren't just seeking returns. Right now, critical assets were selling at distressed prices and the Chinese government was on a buying spree.

3:06By all accounts, in fact, China was planning to do more. At that very moment, Chinese state-linked entities were gathering over $600 million to buy stakes in India's financial services. China, in fact, had already accumulated a remarkable amount of control over India's new wage businesses. And at that very moment, two Chinese companies, Alibaba and Tencent, backed nearly half of India's 31 unicorns between them. This included massive stakes in some of our biggest startups from BTM to Swiggy to Flipkart. Most of these investments were practically invisible and they had come in from Singapore, Mauritius and Hong Kong through shell entities without the government being in the know. China had the power to reshape India's startup ecosystem on a whim and we hardly knew how. And so, within five days of HDFC's disclosure, press node three was in place. To that point, any investments that came from Bangladesh and Pakistan would have to go through the approval route. That is, it was only if the government gave its thumbs up to the investment, could it

4:07enter India? If you were from anywhere else, though, there were many sectors in which you could invest freely. Press node three expanded that list rather vaguely to countries sharing a land border with India. Nobody had any doubts on who we were targeting, though. Our focus was clearly not on investments from Nepal and Bhutan and it was squarely on China. But it wasn't enough to stop investments from China. Formerly, China just had $2.51 billion in Indian investments. But in reality, there was almost twice as much coming in every year. Basically, it was passing through other countries and so we introduced a beneficial ownership provision. That is, it didn't matter where an investment itself came from. If it was ultimately for the benefit of someone from a neighboring country, it would be caught in the net. Now this made sense, but it also had a critical design flaw. It didn't actually specify what a beneficial owner was and it didn't define how far up you would look for one. It didn't define how much ownership was a problem and other Indian laws that talked about

5:09beneficial ownership clearly didn't so. Under the company's act, for instance, you would only start worrying about beneficial ownership once someone indirectly owed more than 10% of an entity. Press Note 3, on the other hand, specified nothing. If you wanted to read this conservatively, even if a Chinese person owned an eligible number of shares somewhere up the ownership chain of an investor, its investment would get stuck. Now Press Note 3 was meant to stop opportunistic COVID era attempts at taking over a major part of our industry for cheap, but in practice, it applied to every investment. Was that what we wanted? That isn't clear, but two months later, Indian and Chinese forces clashed at the Galwan border and 20 Indian soldiers were killed. The relationships between our two countries hardened instantly. Galwan ensured that any changes to the Press Note would be politically radioactive for years and the policy calcified. But it did create some difficulty for Indian companies. On the ground, it was banks that actually processed incoming investments from abroad.

6:11And with such little clarity on what the Press Note was trying to catch, many chose to be as conservative as they could. Different banks applied different thresholds, they demanded different promises from any company receiving an investment, and the same transaction would be seen differently depending on which bank processed them. As this happened, many investors were routinely held up. For one, many global PE and VC funds, those run by BlackRock, Carlisle, Sequoia and the like, typically have Chinese limited partners. So these were investors that gave their money for the fund to manage. They passively held single digit percentages of these funds with little control over where the money went. This was clearly money that India wanted, and yet these investments could get entangled in Press Note 3's requirements. Similarly, there were Chinese investments that could have been good for India, bringing in world-class technology. China, after all, was the factory of the world, and Indian manufacturers would benefit from Chinese technical partners who could give us equipment and know-how.

7:12The standard we have doing this was to set up joint ventures where the Chinese entity would get a stake in exchange for what it brought to the partnership. Only, these joint ventures came under the Press Note scanner, and approvals could drag on endlessly. For instance, when Dixon Technologies tried to set up a smartphone JB with long cheer, the partnership was only approved after huge delays, even though Dixon owned three-fourths of the partnership. BYD's billion dollar investment plans into India, similarly, practically died because the necessary approvals weren't forthcoming. We needed a middle ground. Even if it was important to take a hard look at Chinese investments, these want-on rejections hurt India itself. The capital and knowledge we needed to become a manufacturing powerhouse had been locked away, and according to media reports of the 526 proposals filed until April 2024, just 124 went through, or less than a quarter. More than 200 were rejected outright, while another 200 was stuck in the system.

8:15Decisions typically took six to ten months to come through, and in some cases remained pending for years. So the cabinet's recent decision now tries to ease the load into distinct ways. Number one, it clarifies what a beneficial owner even is, tying it to how the term is seen under money laundering laws, and it cordons off where the scrutiny will go. We look at the investor entity instead of an endless chase through every upstream layer. Having pinned this definition in place, it opens two channels of relief. So an investment into India is safe, as long as someone from a neighboring country owns less than 10% of the investor and can't control it, then standard investment laws apply. So if there's an American VC fund, 6% of the capital of which comes from a Chinese entity, the VC can still put money into India. It wouldn't have to face the threat of a protracted delay or outright rejection. Now this channel does nothing for investors incorporated in China or Hong Kong, so if you're a fund from a neighboring country, things are exactly where they were before.

9:16It's simply for anyone that became collateral damage because of a minor Chinese stake. But there's another channel of relief. Let's say an Indian company genuinely needs Chinese support and plans to set up a joint venture. The investment will still need approvals of course, but now the government is creating a new channel for these applications. So as long as the Indian entity controls the joint venture, finally, those approvals will be expedited within 60 days. For now, this route is limited to just seven sectors, advanced battery components, rare earth processing, electronic components and the like. These are the spaces where India most needs Chinese equipment and process technology to build domestic manufacturing capacity. We are trying to find a middle ground here, one where we're trying to maintain some scrutiny, even as we invite Chinese technical partnerships. Everything else though remains exactly as before. Unlike so many bombastic claims in public, to us, this seems like a sensible compromise. This removes the blockers on a highly liquid pool of blocked capital, global institutional

10:18money that never forced security concern. It also opens the gates for Chinese technology across seven sectors where we need it the most. These are broadly sensible ideas. India spent the last five years learning just how blunt, simple, broad FDI screening laws are. Capital flows far too easily for one to get one's policies exactly right. You can either catch too much as we did in the last five years or too little as we did before. New amendments are a serious attempt to calibrate them better. They try bringing in more capital and technology for India's businesses while holding off Chinese strategic interests. This is a necessary compromise, not a big thought. If you prefer reading the daily brief instead of watching the video, check out the link to the newsletter in the description. Coming to the second story, we have written about India's defence sector a few times now. The history of how India buys weapons, the evolving procurement framework, this state

11:18of the industry as it stands. Those pieces spent a lot of time on the macro picture, rising budgets, geopolitical tensions, the Atmanarbar Bharat push, all of that still holds true. Perhaps more so now with a war in West Asia and the resulting global security disorder. But we are not here to talk about that today. Instead, we want to zoom into the internal economics of India's defence industry where that money actually goes, who captures it and what kinds of businesses are emerging inside the sector. A recent HDFC security is the magic report on the sector is a useful starting point to get a sense of the industry level architecture. Now the short version of the story is this, India's defence industry is gradually becoming a business of sensors, software, integration and lifetime support economics far more than a business of just building big physical platforms. We'll get to explaining what it means. So when you think of defence manufacturing, the image that comes to mind is probably a factory floor producing fighter jet air frames, warship hulls or tank chassis.

12:23But the most valuable part of modern weapons systems is no longer the metal and machinery. In the defence industry, these systems are often called platforms, which simply means large equipment like fighter jets, warships or missile batteries. So what really matters now are the electronics inside them. The report estimates that these electronics already account for about 40% of a platform's total value and that share keeps rising. Now this isn't an India-specific phenomenon. Back in 2006, electronic systems already made up 31% of the cost of the American F-35 fighter jet and that number has only gone up since while the cost of building the airframe has actually come down. The same pattern is playing out across Navy's and Army's globally. The physical shell is becoming a smaller share of the bill while the digital brain is becoming a larger one. In essence, this shifts the centre of gravity in the industry from metal fabricators to firms

13:23that own technology, radar design, avionics, electronic warfare capability and system integration expertise. It also changes what gets sold and how often. So when a platform's value sits primarily in its physical structure, you sell it once and that's that, a 30-year asset. But when the value sits in electronics and software, you sell the initial platform and then keep selling updates like software refreshes and radar replacements every few years across the platform's entire 30-year life. And the industry starts to look less like traditional heavy manufacturing and more like enterprise software with recurring revenue from an installed base. The industry is also changing how it builds things. So traditionally, defense electronics would bespoke as every program got its components designed from scratch. That's expensive and slow. Increasingly, companies are shifting to what's called COTS or commercial of the shelf architecture. Here you build a library of standard reusable electronic building blocks and then assemble them in different configurations for different uses, just like Jenga blocks or Lego bricks.

14:29The R&D is paid for once and every reuse after that is essentially free engineering. That's why this approach lets companies bid lower on contracts while actually earning higher margins when the expensive work was already done. BEL for instance does this with its sonar systems. The same module core can be used for a frigate, a destroyer or a coastal petrol vessel, all which are different types of ships. Data patterns meanwhile does it with its circuit boards. The same military grade computer board goes into a missile tester, a bomb guidance checker and a radar system. If electronics are where the value sits within a platform, the tail end of the life cycle of a platform is where the profit sits within the business. See, the initial sale of a defense platform, delivering a new fighter jet or warship, grabs the headlines. But after that first delivery, the platform needs space, periodic repair, major overhalls and eventually a mid-life upgrade where you rip out aging systems and install modern ones. This long tail of life cycle services runs for decades and it carries structurally higher

15:31margins than the initial hardware delivery. That's because it doesn't require huge capital investment, has more predictable demand and also creates secure customer relationships. In some ways, this isn't all that different from auto-encileries where after sales and service earn higher margins than the core parts themselves. For instance, when we talk about HL, you'll probably imagine a company that makes fighter jets. And of course, you should. But for HL, repairing, maintaining and supplying space for those jets have together accounted for around half of its revenue today. Upgrading and aging aircraft costs roughly 35% of the price of a brand new platform. India is currently retrofitting 30 to 40-year-old jets and tanks with advanced radars, glass cockpits and night vision fire control systems. Steady, high margin work that doesn't depend on new orders coming in. And this gives the signals to understand who exactly captures this long-term revenue. Now, under the old import-heavy model, Indian companies mostly worked as low margin assembly

16:32hubs for foreign design equipment. The more profitable part, space, software updates and long maintenance contracts usually went back to the foreign companies that owned the design. But now, with the design and intellectual property being developed locally, Indian firms keep much of this revenue for themselves. Owning the design gives you near-exclusive control over decades of support work that follow. However, while real, this localization is still incomplete, as India is still one of the world's largest arms importers. Now the defense sector is what economists call a monopsony or a market with essentially one buyer. In India, this is the Ministry of Defense. Development cycles are 12 to 18 months long, sometimes even longer, and then they're followed by extensive field trials and certification processes before a single unit can be delivered. All of this traps capital in inventory and work in progress for a very long time. The cash conversion cycle, which is the time between spending money on inputs and receiving payment for outputs, commonly runs to 400 to 600 days in this industry.

17:36On one level, that looks like a devastating cash crunch for defense firms. But interestingly, none of this means defense companies are actually start for cash. See, companies negotiate milestone-based payments receiving massive customer advances when contracts are signed and when development benchmarks are hit. Well, before the final product is delivered, these advances don't make the inventory move faster or the trials shorter and the operational cycle stays brutal. But importantly, they give companies the cash to survive while that long cycle plays out, without needing to borrow. HL, for instance, had Rs.52,219 crore in customer advances as of March 2025, and it's virtually debt-free. Now, this advance cash is often parked in banks to earn non-business income in the form of interest. For instance, HL's other income grew from approximately 7.5% of profit before tax or PBT in FY20 to approximately 24% in FY25. Their customer advances grew at a 14.5% CAGR,

18:37strongly outpacing revenue growth of 7.6%. Meanwhile, for Mazagorn dog shipbuilders, other income at one point constituted 77% of total PBT. Most of their profit was just interest on government money. So if there's a takeaway here, it's that two firms with identical operating capability can show very different reported profitability depending on their advance payment structures, interest rates, and cash discipline. On the flip side, when advance growth slows or interest rates fall, profit margins can normalize even if the operating business is doing fine. Now, India's defense exports hit a record Rs.23,622 crore in FY20,425. Roughly 34 times the level from a decade ago. That sounds like an unqualified success story, but the headline number hides a crucial distinction about how much value is really captured in exports. So take Astra microwave products which makes radar and electronic warfare subsystems. Its domestic defense orders built on in-house developed technology carry 40% to 45% gross margins.

19:42But its export orders, which are mostly offset and built to print work, where you manufacture someone else's design under license, carry gross margins of just 8% to 10%. That's a nearly five-fold margin gap. Recognizing this, Astra made a hard turn towards the domestic market with the share of exports in its order book falling from 59% in FY20 to just 9% in FY25. Now, why is the gap so large? So intellectual property is approximately 70% of the cost of a defense product. You only capture the line share of the value if you own the design and the architecture. But if you're assembling someone else's blueprints, you're only capturing commoditized margins because many others can do the same assembly. This means India's export growth can be misleading if the composition isn't examined. If most exports are assembly for export, the value capture is thin. So growing exports is only strategically meaningful when you're exporting your own intellectual property. Now to be fair, India's procurement framework does try to protect Indigenous players.

20:44And as we've covered before, the defense acquisition procedure ranks procurement categories in a strict hierarchy. Now at the top sits IDDM. Products that are indigenously designed, developed, and manufactured in India, where only Indian companies can bid and at least 50% of the contract value comes from domestic content. Below that is by Indian, where the product is made in India, but not necessarily designed here with a higher 60% domestic content bar. The government is required to exhaust the higher categories before opening up to lower ones. Now the problem is what happens after you're through the gate. So once two companies both cross the 50% IDDM threshold, the lowest bidder wins. A company at 51% Indigenous content competes on the same footing as one has 80%. There's no reward for going deeper. And the firm that invested more in Indigenous R&D has higher costs to recover. The one that barely cleared the bar doesn't. Now a recent analysis of the draft DAP 2026 flagged a further risk. That companies which purchased foreign design licenses could potentially qualify as IDDM,

21:49diluting the category altogether. The framework rewards Indigenous capability up to a point. But past that point price takes over. So India's defense sector is becoming a brain business, not a brown business. The companies that define the next decade will be the ones that own the electronics, the integration authority, and the life cycle tail. But the question worth watching is whether the procurement machinery can keep pace with the industry's own evolution. If the framework keeps rewarding lowest cost assembly over Indigenous R&D, the economic transformation might be real. But the strategic one could stall. Now coming to the tidbits. India's retail inflation rose to 3.21% in February 2026 up from 2.74% in January, driven by price hikes and food, personal care, and precious metals. Despite the uptick, inflation has now stayed below the RPS medium-term target of 4%. However, it will be under strict watch as companies have already begun raising gas prices

22:52in the wake of the Australian R&D conflict. Coming to the next tidbit. India's packaged water industry is facing rising costs as the R&D disrupts global supply chains and making bottle caps, labels, anything that uses plastics more expensive. Around 2000 smaller bottled water makers have already hiked prices for resellers by roughly 1 rupee per bottle, or 5% increase, with a further 10% rise expected in the coming days while larger companies are absorbing costs for now. Coming to the final tidbit. Over half of Maharashtra's sugar mills may not operate in the next crushing season due to a decline in sugar cane availability and mounting financial pressure on mill operators. The FRP or Fair and Renumerative Price for Sugarcane has been raised six times in the past six years, but the MSP for Sugar has remained frozen at Rs 31 per K recent 2019, squeezing mill margins. So to break even on administrative costs, mills need to run for at least 150 days, but most are currently operating for only about 90 days.

23:53That's all the news I have for you. Thank you so much for watching and see you in the next one.