One is sold in London. The other is sold in India.

Same orange logo. Same company. Same global brand.

But the drink inside is very different.

According to Coca-Cola’s own product pages, British Fanta contains 4.5 grams of sugar per 100ml. Indian Fanta contains 13.7 grams.

That means a 330ml British can contains roughly 15 grams of sugar. The Indian can contains more than 45 grams. That’s nearly three times as much.

Put differently, the British can has roughly four teaspoons of sugar. The Indian version has around eleven. Coca-Cola’s Indian nutrition table implies that one 330ml can delivers roughly 90% of the recommended daily limit for added sugar.

The British version also contains fruit juice concentrates and gets part of its sweetness from artificial sweeteners. The Indian ingredient list contains sugar, flavouring and Colour 110, better known as Sunset Yellow FCF.

That colour is permitted in India and Europe. But if it is used in covered European food products, the packaging must warn that it may affect children’s activity and attention.

In India, it appears as “Colour 110” in the ingredient list.

Same Fanta. Very different conversation.

The tax wrote the recipe

It would be easy to turn this into another “Big Food is bad” lecture.

But the more interesting question is this:

Why would the same company sell two radically different versions of the same drink?

Part of the answer is localisation.

Recipes change across countries because tastes, ingredient prices, supply chains and purchasing power differ. A product designed for a ₹20 price point cannot always use the same formulation as one designed for a British supermarket.

That is normal.

But there is another ingredient in the recipe that never appears on the packet: regulation.

Britain introduced its Soft Drinks Industry Levy in 2018. Drinks containing less than five grams of sugar per 100ml pay no levy. Drinks above that threshold are taxed, with the highest rate applying once sugar crosses eight grams.

British Fanta currently contains 4.5 grams.

Conveniently, just below the tax line.

If the Indian 13.7-gram recipe were sold in Britain, it would fall comfortably inside the highest tax bracket.

The tax did not ask Coca-Cola to become health-conscious. It simply made a lower-sugar recipe the financially smarter option.

And companies responded.

Between 2015 and 2020, the average sugar content of British soft drinks covered by the levy fell by 46%. Around 65% of drinks that originally exceeded the threshold were reformulated to escape the tax.

Britain is now lowering the threshold again from 2028. The current British Fanta recipe may have to change yet again.

Move the regulatory line, and the recipe follows.

India’s tax system creates a different incentive. Applicable aerated, sweetened and flavoured drinks sit inside a flat GST category. Reducing sugar from 13.7 grams to 4.5 grams does not produce the same direct tax saving.

One country taxes the category.

The other taxes the sugar.

Guess which one gets the lower-sugar Fanta.

Subscribe now

The warning label that never arrived

India has debated stronger front-of-pack food labels since 2017.

The idea was simple. If a product contained excessive sugar, salt or fat, the front of the packet would say so clearly.

Not in microscopic writing.

Not inside a table that requires a calculator.

Right on the front, where you cannot miss it.

But according to a Reuters investigation, food companies pushed back strongly during consultations with the Food Safety and Standards Authority of India.

Industry representatives argued that warning labels oversimplified nutrition, could confuse consumers and failed to account for Indian tastes and portion sizes. They also warned that naturally sugary products such as coconut water could get caught by a poorly designed system.

Those are not completely ridiculous concerns. Nutrition is more complicated than a red symbol.

But the commercial stakes are also enormous.

India’s packaged-food and beverage market is worth more than $100 billion. Industry representatives estimated that nearly 80% of its products could be classified as high in sugar, salt or fat under a colour-coded system.

In other words, large parts of the supermarket could suddenly turn red.

After the industry consultation, FSSAI moved towards a numerical black-and-white table rather than an interpretive warning. The Supreme Court is now scrutinising that approach, and no final warning-label system has been adopted yet.

The fight is not really about whether nutritional information should exist.

The information is already somewhere on the packet.

The fight is about whether consumers should notice it.

Why a tiny red label terrifies Big Food

The front of a packet is some of the most valuable real estate inside a supermarket.

That is where companies place:

“High protein.”

“Made with real fruit.”

“Fortified with vitamins.”

“Zero cholesterol.”

The front sells you the dream. The back handles the paperwork.

A prominent “HIGH SUGAR” warning changes that arrangement.

It turns every packet into a compulsory negative advertisement against itself, displayed at the exact moment you are deciding whether to buy it.

That can affect demand. It can damage a product’s healthy image. It can push parents towards another brand. Most importantly, it can force the manufacturer to spend money reformulating the product without ruining its taste, price or shelf life.

A nutrition table makes the shopper interpret the numbers.

A warning label does the interpretation for them.

That difference may look tiny on a packet. On a billion-dollar portfolio, it is anything but tiny. There is also an uncomfortable contradiction here.

Coca-Cola and its bottling partners voluntarily use traffic-light-style labels in several European markets. Nestlé has used interpretive traffic-light labels in Britain since 2013.

Apparently, European consumers deserve information that is easy to understand. Indian consumers are expected to bring a calculator.

Fanta is only the beginning

Reuters found similar differences across other global brands.

Standard KitKat sold in India contains 4.5% cocoa solids, while Australian milk chocolate contains at least 22%. Indian Maggi uses palm oil, while several British variants use the more expensive sunflower oil. Some Maggi packets sold in Britain even carry prominent red warnings about their salt content.

Different does not automatically mean illegal or unsafe.

But a global brand creates an expectation of consistency. When the formulation changes while the logo remains identical, the small print becomes extremely important.

That is why creators such as Revant Himatsingka, better known as Food Pharmer, have built audiences of millions by simply placing two ingredient labels next to each other.

When the regulator does not put the warning on the packet, the internet puts it in the algorithm.

That creates its own form of taxation: a reputation tax.

Nestlé, for instance, announced a sugar-free Cerelac variant in India after years of criticism over why Indian consumers were being sold versions with added sugar when sugar-free alternatives existed elsewhere.

Consumer outrage changed the spreadsheet.

So, where is the money?

If India eventually adopts clear front-of-pack warnings, the winners will not just be consumers.

Lower-sugar products suddenly become more competitive. Cleaner-label challenger brands gain an opening. Ingredient companies that help manufacturers preserve taste while reducing sugar, salt or fat see more demand.

The losers will be companies forced to choose between three uncomfortable options:

Reformulate the product.

Increase the price.

Or print a warning on the front.

And that is the real lesson hidden inside this orange can.

The Indian Fanta did not become sweeter because chemistry works differently in Mumbai.

It became sweeter because the incentives were different. The logo is global. The accountability is local. And perhaps the most revealing part is not that the Indian can contains more sugar.

It is that the front of the can tells you less.

If this story made you look twice at something inside your refrigerator, subscribe to Show Me the Money.

Subscribe now

Every week, we follow the hidden incentives behind an ordinary product, policy or company, because the small print is usually where the money is.

And if you know someone who assumes a global brand means a global recipe, send this to them.

Nothing unusual there.

Except it is not heading south towards Singapore, across the Indian Ocean and through the Suez Canal.

It is going north.

The PanStar Acro will sail past Russia’s Kamchatka Peninsula, cross the Bering Sea and travel along Russia’s Arctic coast before stopping at Felixstowe, Rotterdam and Gdansk.

This is South Korea’s first commercial container voyage through the Northern Sea Route. The government wants to know whether melting ice has created a viable new trade corridor between Asia and Europe.

Because, on a map, the route looks almost absurdly attractive.

South Korea estimates the distance from Busan to Rotterdam at roughly 13,000 kilometres. Depending on the ports involved, travelling through the Arctic can remove around 4,800 kilometres and 11 days from the conventional journey through the Strait of Malacca and Suez. Korean authorities believe voyage time could fall by about 35%.

That is not a minor efficiency improvement.

That is the maritime equivalent of discovering a new motorway between two continents.

A canal nobody had to dig

The Suez Canal matters because it allows ships travelling between Asia and Europe to avoid the much longer journey around Africa.

The Northern Sea Route could create another shortcut, except nobody is excavating it.

The planet is.

At its annual September minimum, Arctic sea-ice extent has declined by roughly 13% per decade since satellite monitoring began. Less ice does not make the Arctic easy to navigate. But it can make sections of the route usable for longer each summer.

Shipping is already responding. Vessels completed 103 Northern Sea Route transits in 2025, up from just 43 in 2022.

Now consider what a shorter voyage does economically.

The ship burns less fuel. Crew and charter costs are paid for fewer days. Goods reach customers sooner. More importantly, companies have less money trapped inside products floating around at sea.

An automaker can replenish European warehouses faster. An electronics company can respond to demand without ordering quite as far in advance. Importers may need less buffer stock.

The real prize is not merely cheaper shipping.

It is a faster supply chain with less working capital tied up inside it.

The Arctic economy

Busan wants to become the Asian gateway to this route. Northern European ports could receive goods sooner. Russia gains strategic importance because most of the passage hugs its coastline.

But the less obvious winners may sit around the ship itself.

Regular Arctic commerce would require ice-class vessels, stronger hulls, polar navigation systems, satellite communications, real-time ice forecasting, trained crews, emergency-response infrastructure and specialised marine insurance.

South Korea is already treating this voyage as a data-gathering exercise. Officials will study its fuel use, travel time, operating cost and demand from cargo owners. The officers received polar training, an experienced Arctic navigator is aboard, and the ship will be monitored around the clock.

This is not simply a possible new shipping lane.

It is the possible birth of an Arctic logistics industry.

But this is not Suez 2.0. Not yet.

Ice conditions remain unpredictable and the route is still seasonal. Ships may need expensive reinforcement or icebreaker assistance. Ports, repair yards and rescue facilities are sparse. Insurance can be costly.

The route is also most useful for direct trade between northeast Asia and northern Europe. It is less helpful for ships that collect and unload cargo across Southeast Asia, the Middle East and the Mediterranean.

Then there is Russia.

Using Suez creates dependence on Egypt and the security of the Red Sea. Going north replaces some of that dependence with Moscow.

Rosatom oversees navigation permits and icebreaker services along the Northern Sea Route. South Korea also consulted Russia about possible assistance if the PanStar Acro became trapped in ice.

So the route does not eliminate geopolitical risk.

It moves it north.

The most brutal feedback loop

A shorter journey may reduce fuel use per trip. But ships also emit black carbon, or soot, which can settle on bright snow and ice.

The darker surface reflects less sunlight and absorbs more heat.

Melting ice opens shipping. More shipping can contribute to further melting.

The International Maritime Organization banned most use and carriage of heavy fuel oil in Arctic waters from July 2024, although exemptions and temporary waivers remain possible until 2029. An oil spill would also be exceptionally difficult to contain in cold, remote and ice-covered waters.

One trial voyage will not dethrone Suez.

But it shows that climate change is beginning to redraw the physical map on which global trade was built.

Remote ports may become strategic. Niche ship designs may become valuable. Russia could gain leverage over a new chokepoint. Companies may eventually choose between the politically unstable waters around Suez and the climatically unstable waters of the Arctic.

The original Suez Canal was built by engineers.

The next one may be built by a warming planet.

If you enjoy stories like this, where climate, money, business and geopolitics collide in ways that sound almost unbelievable, subscribe to Show Me the Money.

Share

Subscribe now

Nobody is worried that Nvidia cannot sell enough chips.

The worry is stranger: what happens when its customers run out of money before Nvidia runs out of GPUs?

The AI boom has a brutal maths problem. Everybody wants more computing power, but serious AI infrastructure costs billions. And the next phase may require so much capital that corporate balance sheets alone cannot comfortably fund it.

Nvidia’s solution is audacious. Don’t wait for customers to become richer. Make the chips financeable.

And Jensen Huang thinks that, in the process, AI infrastructure could become an entirely new asset class.

The half-trillion-dollar plan

On 10 August, Nvidia announced plans to work with Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs and KKR on financing platforms designed to mobilise more than $500 billion of third-party capital.

The basic idea is fairly simple.

An AI company needs an enormous amount of computing capacity.

Outside investors finance the infrastructure upfront. The AI company leases the capacity and pays for it over time. The investors collect a return.

Nvidia gets to sell the hardware today.

That opens AI infrastructure to pools of money that already finance airports, power plants, aircraft and commercial property.

It also solves a growing problem for Nvidia.

Its customers may want considerably more compute than they can currently afford.

But what is a five-year-old GPU worth?

This is where things get complicated.

Infrastructure lenders care enormously about what their collateral will be worth in the future.

A commercial aircraft can remain useful for decades. A building can survive generations of tenants.

Computing hardware ages much faster.

A GPU bought today could be competing with dramatically more powerful and energy-efficient hardware a few years from now.

If an AI company fails, the financier needs someone else willing to pay for the infrastructure.

Nvidia thinks its ecosystem makes that possible.

CUDA is used across an enormous range of AI workloads, developers and companies. Nvidia argues that the same infrastructure can therefore be shifted between customers and applications. It also says software improvements can increase the productivity of hardware already installed.

That makes an Nvidia cluster potentially easier to lease again if its original customer disappears.

And Nvidia is now willing to put serious money behind that argument.

The $105 billion experiment

On 17 August, Nvidia disclosed a remarkable arrangement involving OpenAI and a giant data-centre campus being developed by SB Energy in Ohio.

OpenAI will be the tenant for leases covering roughly 4.25 gigawatts of IT load.

Nvidia has agreed to provide residual-value guarantees with a cumulative initial cap of $105 billion.

The guarantee covers a specific risk.

If OpenAI becomes insolvent or fails to make lease payments, Nvidia may have to cover a shortfall between an agreed minimum value and what SB Energy ultimately recovers by leasing the premises to someone else or selling them.

Nvidia also has several options if that happens. It can assume a lease, help find another tenant, initiate a sale or temporarily support certain project costs.

OpenAI has agreed to reimburse Nvidia for payments made under these guarantees.

The arrangement can run for as long as 20 years, although Nvidia’s exposure can end earlier under several conditions.

This pulls Nvidia much deeper into what happens after a GPU leaves the factory.

The future value of AI infrastructure now matters directly to Nvidia’s own balance sheet.

Why would Nvidia do this?

Because the upside could be enormous.

If investors become comfortable lending against Nvidia-powered AI infrastructure, more companies can afford to build it.

More infrastructure means more Nvidia systems.

More Nvidia systems expand CUDA’s installed base.

A larger installed base creates more potential users for those systems if they ever need to be redeployed.

That makes the infrastructure easier for lenders to value.

And easier financing funds another round of AI infrastructure.

There is an important competitive advantage buried inside this.

Imagine two AI companies want to build similar data centres.

One can borrow against hardware that financiers already understand, have historical utilisation data for and know can be deployed across thousands of workloads.

The other cannot.

Even if the hardware performs similarly, their financing costs may be different.

Nvidia’s moat could eventually extend beyond the chip and the software ecosystem.

Its customers could simply find it easier to raise money.

And Nvidia does not have to provide hundreds of billions of dollars itself. The proposed financing platforms are designed primarily around outside capital.

Nvidia’s contribution can be concentrated in places where its support makes financiers comfortable committing much larger sums.

We have seen versions of this before

There is some uncomfortable history here.

During the telecom boom around 2000, equipment manufacturers provided financing to customers buying their networking equipment.

Credit helped customers build faster. That produced more orders for equipment manufacturers.

Then telecom companies started failing.

Nortel disclosed in 2002 that it had increased provisions against customer financing because some customers were no longer expected to meet their repayment obligations.

Nvidia’s arrangement today is considerably different. Independent financial institutions are underwriting the proposed platforms, the customer base for AI infrastructure is broad and Nvidia expects third-party investors to provide most of the capital.

Still, the old telecom episode leaves behind a useful question.

How much infrastructure are customers buying because their businesses already generate enough money to support it?

And how much are they buying because the financial system expects those revenues to arrive later?

That distinction matters enormously if AI growth disappoints.

AI is becoming a credit story

Imagine an AI company that has built enormous computing capacity.

Its revenue comes in below expectations.

Lease payments become difficult.

At roughly the same time, a newer generation of chips reaches the market and makes the older equipment less attractive.

Now the borrower is struggling while the collateral behind the financing is also becoming less valuable.

This risk is beginning to spread through the financial system.

The Bank of England’s July 2026 Financial Stability Report cites an OECD estimate that private credit’s share of AI-investment financing jumped from 9% in 2024 to 34% in 2025.

AI companies are also tapping bonds, structured finance and other forms of debt.

The AI boom is becoming a Wall Street story.

And that may be Jensen Huang’s bigger bet.

If AI factories really become productive infrastructure with long economic lives, Nvidia could help create a massive new market for financing compute.

That would remove one of the biggest constraints on its own growth.

It would also mean that the health of the AI boom increasingly depends on credit remaining available and the assets behind those loans retaining their value.

For years, everyone wondered whether Nvidia could manufacture enough GPUs to satisfy the AI boom.

Soon, the more important question might be whether Wall Street can finance enough of them.

Show Me the Money is where I unpack the most exciting and pressing business and money stories hiding beneath the headline. Subscribe if you’d like the next one in your inbox.

Subscribe now

The Sensex has grown 41,000% since it was set up in 1979. It has also crashed 35% or more 7 times since then.

And that’s where the pattern becomes striking.

What is this pattern?

Every 35%+ drop in the Sensex has been followed by a big thrust up — a period when stock prices rise rapidly. This is the huge buying that takes place from the lows. Usually, this buying has lasted an average of 15 months.

Historically, a post-crash thrust up/rally is followed by a downward drift in stock prices. This period has typically lasted 34 months (19 months if the largest sample is not considered).

After the downward drift, the market has risen again. The average price increase in this phase has been about 200%.

The Stock Market Cycle

Share

Typically, the stock market has moved in four stages: 1) The Crash —> 2) The Big Thrust Up —> 3) Downward Drift —> 4) Rise Again —> and then The Crash again

What is the significance of this right now?

Because we are currently in the Downward Drift phase. The Big Thrust Up since the 2020 Covid Crash lasted ~18 months (from April 2020 to October 2021).

So, what should your investment plan be now?

History, as they say, does not repeat but rhymes.

Of course, there’s always a possibility that the same pattern may not repeat. But if it does, the best option is to continue investing more money each month.

Why?

Because the market has traditionally surged by 200% following the Downward Drift. Monthly investments will ensure that you are purchasing at lower prices right now to enjoy future gains.

Share

Most people I speak to have medical insurance.

But almost every one of them is covered for a small amount — and it’s nowhere close to how much cover they should have.

Share

The 3 Factors That Matter While Deciding the Cover Amount

The Major Medical Events

You want your medical insurance to cover serious illnesses and accidents. Medical bills can run up to ₹25-50 lakhs for a major hospitalization/treatment. That's as true for a 25-year-old as it is for a 60-year-old. You will have to pay out of pocket or borrow if you do not have a large cover.

Medical Cost Inflation

Medical inflation in India was 14% in 2021. Generally, 2021 was an outlier in terms of costs. But medical inflation in India has hovered between 8-10% for the past 10 years.

At that rate, the value of money halves every 7 years. So, a ₹20 lakh cover may seem sufficient right now but would be worth ₹10 lakhs in 7 years and ₹5 lakhs in 14 years.

Your medical insurance cover should be able to tackle inflation.

Future-Proofing Yourself

You never want to be in a situation where you need a medical insurance policy but can’t get it. The most common reason for being denied new insurance is the prior existence of a major medical condition. Also, getting good new insurance after the age of 65 is difficult.

As a rule of thumb, it’s always a good idea to get as large a cover as possible while you’re still healthy and under 65. Also, check the insurance cover that your parents have. Bump it up, if needed.

P.S.- The policy is issued based on your health at the time you get a policy for the first time. Your existing medical insurance will continue as long as you keep renewing it — it does not stop on the diagnosis of a serious medical condition or when you cross the age of 65.

What’s the Right Medical Cover Amount?

My mother had once told me this many years ago: 'Investing is like exercising—do as much as you can but even a little is better than nothing.'

That's also true for medical insurance covers. Get as much as you can right now. Your end goal should be to get a cover of 75 lakhs to 1 crore.

For that, you can look at a super useful tool called a 'Super Top-Up'. It’s one of the most well-guarded secrets of the insurance world. You can use it to get a very high cover of ₹50+ lakhs at a low cost.

How low? Here's a real-life example

A few weeks back, I renewed a ‘Super Top-Up’ policy that covers me and my mom for ₹75 lakhs. I paid less than ₹10,000/year for it. I will write about Super Top-Ups in the coming weeks.

Subscribe now

Share

In those 5 years, the Sensex rose by 640%. That’s a growth of 6.4 times. ₹10 lakhs invested in April 2003 would have been ₹64 lakhs in early 2008.

But here's an untold fact about that fantastic mid-2000s bull run.

In that same 5-year period the market dropped:

- More than 30% two times

- More than 10% five times

Your portfolio would have grown 6.4 times but only if you would have had the stomach to stay invested during all those big down moves.

The stock market has a mind of its own. No one really knows what’s going to happen tomorrow or in the next year.

But here’s what we know: the market trends up over the long run.

So, in volatile times like the last 6 months, staying invested becomes vital to enjoy the long-term benefits of the market

Subscribe now