Our goal with The Daily Brief is to simplify the biggest stories in the Indian markets and help you understand what they mean. We won’t just tell you what happened; we’ll tell you why and how too. We do this show in both formats: video and audio. This piece curates the stories that we talk about.
You can listen to the podcast on Spotify, Apple Podcasts, or wherever you get your podcasts and watch the videos on YouTube. You can also watch The Daily Brief in Hindi.
In today’s edition of The Daily Brief:
1. Why is India paying more for floating solar?
Floating solar costs more than conventional solar, but India is betting that avoiding land constraints, reusing existing grid infrastructure, and conserving water could justify the premium at carefully chosen sites.
2. What’s behind India’s first tokenized bond?
India’s first tokenized corporate bonds are testing whether securities and money can move together on a shared digital ledger, potentially simplifying settlement, but questions around liquidity, regulation, and interoperability remain unresolved.
India has become remarkably good at building solar power.
As of August 2026, the country had 168 GW of solar capacity. Nearly 124 GW came from utility-scale projects, along with ~33 GW of rooftop solar. Costs have fallen, developers know how to build these projects, and solar parks have gone from ambitious experiments to a fairly standard part of India’s power system.
Now, the government now wants to put solar panels on water.
In July, the Union Cabinet approved the Pradhan Mantri Surya Sarovar Yojana (PM-SSY). It will spend ₹5,070 crore to support 5 GW of floating solar projects, each accompanied by at least two hours of energy storage. Prime Minister Modi even shared a carousel about it on Instagram.
But this is not an obvious bet.
Floating solar is harder to build, trickier to maintain and generally more expensive than solar on land. India currently has only around 0.7 GW of it. The new scheme could increase that eightfold. And according to the National Institute of Solar Energy (NISE), the country has around 102 GW of technical floating-solar potential.
Those numbers look attractive, but the economics suggest something rather less straightforward.
At first glance, floating solar may not look particularly revolutionary. After all, the solar cells do the same job as those in any other solar plant. They convert sunlight into direct-current electricity. Inverters turn that into alternating current, and transformers feed it into the grid.
The difference, though, lies in everything holding the panels up.
See, a ground-mounted project fixes its modules onto steel piles or concrete foundations. But a floating project places them on interconnected plastic floats. According to a World Bank study, 90% of floating solar projects use high-density polyethylene (HDPE), a buoyant and durable plastic commodity which is also found in pipes and shampoo bottles.
These floating islands must then be anchored to the reservoir bed or shore so they do not drift away. The anchoring system must handle waves, currents and changing water levels. Cables must be able to withstand moisture and constant movement. The panels are also kept at angles below 15°, compared to 15-35° for land solar. This reduces the wind load on the floating structure.
Construction happens partly onshore, after which the assembled sections are pushed into the water. Maintenance may require floating walkways, boats and even human divers.
NTPC’s 100 MW Ramagundam project, for instance, was divided into 40 floating blocks. Each block contained 11,200 modules, along with an inverter, transformer and high-voltage breaker mounted on a floating platform.
Floating solar places a familiar power plant on a much more difficult surface. And the complications can be serious, ranging across humidity, corrosion, electrical hazards, and the constant movement of the floating structure that makes the above much harder than usual.
So, why bother putting solar on water at all? Well, that requires thinking of a standard solar project as a land acquisition problem first.
A large solar project needs a large, uninterrupted parcel of land. Acquiring it can take time, particularly when ownership is fragmented or the land supports agriculture or local communities.
Theoretically, India is not literally running out of land for renewable energy. A CEEW assessment found that land constraints remain relatively manageable until India crosses roughly 1,500 GW of renewable capacity.
But that is not the same as easy access. CEEW also found that only 27% of India’s solar potential lies in areas with fewer than 250 people per square kilometre. Only 41% is in areas free from historical land conflicts, while just 22% is located in places that combine low land prices with low climate risk.
Beyond sidestepping land issues, floating solar’s stronger advantage may be access to the grid.
A solar park needs more than just panels and land. It also needs substations and transmission lines to carry the electricity elsewhere. Building this infrastructure can require more land and take much longer than constructing the solar plant itself.
We are already seeing the consequences of not having built this transmission infrastructure. According to Ember, India curtailed around 300 million units of renewable electricity because of transmission constraints in the first quarter of 2026. That accounted for nearly two-thirds of all electricity curtailed during the period.
This is where certain waterbodies become useful. Hydropower reservoirs, power-station reservoirs and industrial cooling ponds may already sit beside substations, transmission lines and trained staff.
Ramagundam is a good example. Its 100 MW floating project was built on NTPC’s own reservoir and connected to the plant’s existing switchyard. NTPC followed a similar playbook elsewhere. By August 2022, it had commissioned 242 MW of floating solar across reservoirs attached to its power stations. These sites do not make floating solar cheap, but they allow NTPC to reuse land, reservoirs and grid infrastructure it already controls.
Floating solar has a few other advantages, although each comes with a rather large asterisk.
The first is efficiency. Solar panels become less efficient as they heat up, and the water beneath them can keep them cooler. NISE estimated a ~7% generation gain when floating and ground-mounted panels were placed at the same shallow angle.
But once floating panels were compared with ground panels installed at their optimal angle, the gain fell to just 2.35%.
Even that is not guaranteed. The IEA found that some floating systems run cooler, while others show little improvement or can become hotter when closely packed floats block airflow beneath the panels. Cooling helps, but it is a site-specific bonus and not a reason by itself to pay more for floating solar.
The second advantage is water conservation. Floating panels shade the reservoir and reduce the wind passing over its surface, which can slow evaporation. That water may be valuable for industry, irrigation or hydropower.
Project owners have made some dramatic claims. For instance, Tata Power estimates that its 126 MW project at Omkareshwar will save 32.5 million cubic metres every year.
But these are still estimates. The actual result depends on the share of the reservoir covered, local weather and how the reservoir is operated. The value also differs by location; water saved in a drought-prone industrial reservoir may matter far more than in one that regularly spills during the monsoon.
Hydropower reservoirs offer one more possibility. Solar can generate during the day, allowing the dam operator to conserve water and run the turbines later, when solar output falls. The two plants may also share a substation or transmission connection.
But simply placing solar panels beside a hydro plant does not turn the reservoir into a battery. The operator must be able to change when it releases water, which may not be possible when irrigation, drinking water or flood control dictates the schedule.
All this can make it sound as though floating solar puts otherwise idle water to work. Except the water is rarely idle. Reservoirs can support fishing, irrigation, drinking-water supply and local ecosystems. Floating solar may sidestep land, but it does not avoid competing claims over water.
The environmental and social assessment for Omkareshwar identified livelihood effects on 312 fishermen across six villages. An on-ground report placed the wider affected community at 1,877 people. Many of these families had already lost land when the dam was built and had moved into fishing afterwards.
The ecological effects are harder to measure. Floating panels block sunlight and wind, which can change water temperature, oxygen levels, aquatic plants and fish habitat. A USGS study covering 11 reservoirs found no single ecological outcome. Floating solar consistently cooled the surface, but its effects on oxygen and fish habitat differed between reservoirs.
Even before accounting for those harder-to-measure costs, floating solar has one straightforward problem: it costs more to build.
NISE estimates that a floating project can cost around 25% more than ground-mounted solar. Floats, anchors and mooring systems accounted for much of the difference. The small efficiency gain from placing panels over water usually cannot make up for this extra cost.
But even the 25% premium is only a rough guide. There is no standard floating-solar site, and therefore no single cost that represents the industry. The price changes with the depth of the reservoir, wind and wave conditions, water-level fluctuations, anchoring requirements and access to existing grid infrastructure. A calm reservoir beside a power station can look very different from a large hydropower reservoir exposed to severe weather.
If floating solar is judged only by the cost of producing electricity, it usually loses.
The standard measure used for this comparison is the levelised cost of electricity (LCOE). It adds up the cost of building, financing and operating a power plant, and divides that by the electricity it will generate over its lifetime. LCOE works well when comparing one solar project with another, but it does not capture everything that floating solar may offer. It may save water, help a hydro operator generate more electricity in the evening or make use of infrastructure that already exists.
Some of these benefits reduce the project’s costs directly, while others go to somebody else. The irrigation department may benefit from the saved water. The hydro operator may gain more flexibility. The grid may get a smoother supply of electricity. But the company paying for the project cannot earn money from these benefits unless its contract rewards them.
In short, the floating project may be expensive even when it makes the wider power system cheaper.
The economics also change from one waterbody to another. Existing reservoirs at power plants and industrial sites are usually the easiest places to start. Hydropower reservoirs may offer more benefits, but they are also harder to build on and serve many other users. Natural lakes are the weakest option. They carry greater ecological and social risks while offering few of the advantages that can justify floating solar’s higher cost.
There are indeed merits to India building floating solar, but they have to be chosen carefully.
The 5 GW floating installation planned under PM-SSY is small beside India’s 168 GW solar fleet. That makes sense: floating solar carries too many costs to become the default alternative to solar parks. The government should begin with the obvious sites: reservoirs attached to power stations, industrial ponds, mine pits and hydropower projects where land, grid infrastructure or flexible generation already exists.
PM-SSY should be treated as a test rather than a race for megawatts. If the higher cost does not buy clear benefits elsewhere in the system, India is merely subsidising expensive solar.
On September 7, REC, the state-owned power financier, raised ₹500 crore through a bond paying 7.3% a year. An ordinary deal, except that REC calls it India’s first tokenised corporate bond. Two days later, L&T became the first private Indian firm to raise another ₹500 crore the same way.
Beyond the label of being tokenised, little about these deals is public. Tokenisation itself is still very early. About $8 billion of tokenised bonds had been issued worldwide by mid-2025, a drop in the bucket next to $80 trillion of government bonds. Nobody knows what shape it will take even two years from now.
Why write about it, then? Because regulators, central banks and researchers keep talking about it, and SEBI and the RBI are running pilots. Finance has had plenty of innovations that turned out to be gimmicks, and this may be one. But it is very much a new way to design, trade and settle securities.
To make sense of it, we’ve leaned on research from the Bank for International Settlements (BIS), often called the central bank for central banks, and the IMF.
Let’s start with the obvious objection: isn’t a bond already digital? Yes, but in a particular way.
Tokenisation leaves the bond alone and changes the plumbing around it. Even SEBI’s chairman calls it “not a new asset class”.
Today, a bond is an entry in a depository’s database. When you sell, brokers, depositories, banks and clearing corporations each update their own records, then check they match. Every trade is a chain of messages and reconciliations, and everyone has to trust whoever runs the database.
A token folds the record and the rules together. It carries information about what the asset is, who owns it, and the rules for when it can move.
Think of it as cash across a counter. Hand over a ₹500 note for your groceries, and paying and settling are the same act. With bonds, the security and the money sit in separate systems and settle one after the other. A token puts both on one ledger, so they swap in a single step, or not at all. That’s called atomic settlement.
There’s also a difference between a tokenised asset and a digital one like bitcoin. Bitcoin exists only on its ledger. A token stands for something outside it, like a bond or a bar of gold, so it’s only as good as the link between the two.
If the old system works, why bother with tokenization?
Well, in many markets, settlement can take up to two business days. India is usually quicker, but every trade still passes through several hands. Tokenization promises to cut that down.
With one shared record, nobody has to check their books against anyone else’s. Because the bond and the money move together, failed settlements go away. Interest can pay itself, and markets could run round the clock. Collateral can move almost instantly, something a US pilot has already tested. And finer slicing means smaller minimum tickets.
The boldest idea is what the BIS calls composability, which is snapping transactions together like Lego so they run as one. Sell a bond, use the cash to repay a loan, and get back the shares you’d pledged for it, all in a single step.
Early numbers are encouraging, if thin. Tokenised bonds had bid-ask spreads — or the gap between what buyers offer and sellers ask — of under 0.2 percentage points, against 0.3 for ordinary bonds from the same issuers. Issuance costs were no different, and the researchers call the evidence suggestive.
Small gains per trade add up in a busy market. India’s corporate bond market isn’t one. Of nearly 33,000 bonds outstanding, only 400 to 500 trade on a typical day.
The gains are real, if still unproven at scale. But the speed behind them creates new problems.
The catch
One such problem is cash. Say Bank A owes Bank B ₹100 crore, and Bank B owes Bank A ₹80 crore. Settle at the end of the day, and only ₹20 crore moves. Settle every trade instantly and in full, and each bank needs its whole amount ready. That’s why token systems tend to need money parked upfront, which is expensive.
Those delays also work as shock absorbers, giving banks time to raise cash and regulators time to step in. Without them, an automatic margin call — a demand for extra security when prices fall — can force a sale that moves prices and triggers the next call, quicker than regulators can respond.
The law hasn’t caught up either. In many countries, a token may be just a claim on whoever created it, not ownership of the asset behind it. If that entity goes bust, you’re one more creditor in the queue.
And the middlemen don’t vanish. Someone still has to vet borrowers, and code can’t do that alone. Tokenisation moves where we place our trust. It doesn’t remove the need for it.
What would it take to make it work?
None of this rules tokenisation out. But more than a few things have to go right.
First, the token has to be the real thing. In a native issue, the only record of the bond is on the new ledger. The easier route is a digital twin of an ordinary bond, but then two records must always agree — the very reconciliation tokenisation was meant to remove.
Second, the money has to be on the same ledger, because a bond can’t swap in one step with cash sitting elsewhere. That’s where central bank digital currency comes in. It is rupees issued by the RBI itself, just in token form, unlike crypto, which has no central bank behind it.
India has two kinds: the retail e-rupee for people and shops, and the wholesale e-rupee, used in these pilots, for institutions settling large trades. Neither is the same as bank reserves, the balances banks keep at the RBI, which most investors can’t hold. Pilot investors used a CBDC wallet linked to their bank account.
The BIS wants tokenised central bank money, bank deposits and government bonds on one shared platform, and Hong Kong is already testing it.
Third, the platforms have to talk to each other. Every operator is tempted to build its own, and a dominant one could pocket the savings, while too many rivals could leave money stranded. Fragmentation can be as dangerous as concentration.
How do REC’s bonds measure up?
Honestly, we can’t fully say. We know the bond and the money moved together, and that pay-in, allotment and listing happened on the same day. Holdings sit on a new SEBI ledger called Demat 2.0, in place of the usual depository ledger. But how the ledger works, how the roles of brokers, custodians and depositories changed, or which record counts legally, all are unclear.
Much else is untested. There’s no way to trade these bonds yet, and secondary trading is reportedly expected by December. The pilot isn’t open to retail investors. And the tokenization didn’t make much material difference to REC’s bond price.
Why bother, then? Not for faster settlement, which India mostly has. A plain bond sits at the easy end of what the BIS calls the tokenisation continuum, where gains are modest because the market already works.
The bet is on what comes next. Will Demat 2.0 entry become the legal record of ownership, as Hong Kong made it for its first tokenised bond? Will the RBI’s own Unified Markets Interface, which will also tokenise assets, connect with Demat 2.0, or will India end up with two systems that don’t talk to each other? Those are just some of the questions tokenization needs to prove a foothold in.
– This edition of the newsletter was written by Kashish & Mridula
[1] PhonePe secures first international licence in UAE
PhonePe has received its first international licence from the Central Bank of the UAE, paving the way for local operations in its first overseas market. The licence will allow it to serve merchants directly and offer payment and cross-border transaction services.
Source: Financial Express
[2] US suspends fresh green-card sponsorship filings by Cognizant
The US Department of Labor has suspended new Permanent Labor Certification, or PERM, filings by Cognizant amid an investigation into alleged immigration fraud. The move blocks a key first step in employment-based green-card sponsorship while the probe continues.
Source: The Hindu
[3] L&T secures ₹2,500–5,000 crore offshore order from ONGC
L&T’s hydrocarbon offshore business has won a ₹2,500–5,000 crore order from ONGC for projects off India’s west coast. The work includes three wellhead platforms, a riser platform, subsea pipelines and modifications to existing offshore facilities.
Source: The Economic Times
[4] Cement firms may invest up to ₹13,000 crore to expand green power capacity
Major Indian cement manufacturers are projected to invest ₹12,000–13,000 crore to expand their green power capacity to 5.8–6.0 GW by March 2028, according to rating agency ICRA. The shift is expected to yield annual cost savings of over ₹6,200 crore, with a 25% green power replacement capable of expanding operating margins by up to 160 basis points.
Source: The Economic Times
[5] Australia flags India’s sugarcane support at WTO
Australia has raised concerns at the WTO over India’s sugarcane market-price support between FY19 and FY25, claiming it far exceeded permitted limits. It estimates support at 90–95% of production value, compared with the WTO’s 10% threshold for product-specific support.
Source: Financial Express
Hi everyone, I’m Mridula, and welcome to the third edition of Weekly Tidbits.
There’s always more happening at Markets by Zerodha.
The Chatter: How is HUL driving volume growth via structural premiumization and a Horlicks nutrition pivot? How is Shiprocket leveraging commerce data across 42 couriers to scale D2C brands? Plus, updates from Arcil, Fortis, and Sandhar.
Aftermarket Report: Why did Nifty remain under pressure despite a sharp CAS recovery? How are Brent crude at $105 per barrel and USD/INR at 95.44 adding to India’s inflation worries? Plus, record ₹32,297 Cr SIP flows, SEBI’s tokenised bond pilot, and Xi Jinping’s India visit.
Join us on WhatsApp, where we share interesting soundbites from concalls, articles, and everything else we come across throughout the day. You’ll also get notified the moment a new video or article drops so that you can read or watch it right away.
Join us
Thank you for reading. Do share this with your friends and make them as smart as you are 😉
No posts

source