IndexDecoded @IndexDecoded

Buffett's biggest positions are Apple, Bank of America, and Coca Cola. None of them benefited from the AI trade in the last 12 months. The S&P 500 was carried by Nvidia, Microsoft, and Meta. Berkshire lagged because it owns the economy that surrounds AI, not the companies building it

@KobeissiLetter Let's see

@ekwufinance Commodity sector market cap reaching 20-30% again would require oil majors, miners, and agriculture companies to add $7-10 trillion in market cap

@MBAeconomics1 Oil ran on OPEC supply cuts :a real, sustained supply shock

3/3 This is why the error correction section of a methodology document matters. Most investors skip straight to the weighting rules. The correction policy tells you how much discretion the index committee has when something goes wrong, and who bears the cost. Usually it's the fund. Which means it's you.

2/3 The provider issues a correction notice. If the error is caught before the rebalance goes live, they fix the methodology application and republish the constituent list. Funds haven't traded yet. Damage is limited. If it's caught after funds have already bought and sold based on the wrong data, it gets harder. The index gets restated. Funds that tracked it now have a portfolio that doesn't match the corrected index. They have to trade again to catch up, paying transaction costs twice. No regulator steps in. The methodology document governs everything. Whatever it says about error correction is what happens.

1/3 Index providers make mistakes. Prices get wrong data. Corporate actions get missed. A stock gets added that shouldn't have been. Most investors don't know this happens. Fewer know what happens next.

@ShashankUdupa1 Insightful

@RenukaJain6 @SamsungIndia What happened

@amitisinvesting Let's see

@dmuthuk What's the reason

@dmuthuk @grok verify

@Akshat_World Data source?

@WeekendInvestng Long run blended costs same

@Akshat_World @abhinavenigma The indices are not the same asset class.

India wants to build chips. The government has committed ₹76,000 crore to make it happen. Most of the coverage treats this like a done deal. It isn't. Here's what the headlines skip. ──────────────────── Building a fab is not like building a factory. A chip fabrication plant needs ultra-pure water, uninterrupted power, and air cleaner than a hospital operating room. One power cut during a production run can destroy an entire batch of wafers. Taiwan and South Korea took decades to get this infrastructure right. India is trying to compress that timeline into years. ──────────────────── The talent gap is real. Semiconductor design and fabrication need a specific kind of engineer. India has plenty of software talent. It has very few people who've actually run a fab line. Most of the early hires for India's new fabs come from Taiwan, the US, or Indian engineers who left for TSMC and Intel years ago and are now being asked to return. ──────────────────── What India is actually good at right now. Chip design, not chip manufacturing. A large share of the world's chip design happens in Indian offices of companies like Qualcomm, Intel, and AMD. India already has deep talent here. The country exports design work, then imports the finished chips made elsewhere. The government's fab push is trying to close that gap. It's a much harder problem than design. ──────────────────── The first fabs won't make cutting-edge chips. Micron's Sanand plant and the Tata-PSMC fab in Gujarat are targeting older process nodes, the kind used in cars, appliances, and industrial equipment. Not the advanced chips in your phone or laptop. That's the right starting point. Bleeding-edge fabs cost tens of billions of dollars and take a decade to become reliable. Older nodes are where India can realistically compete first. ──────────────────── For investors, here's the index angle. A few thematic semiconductor indices in India already exist. Pull up the methodology before buying one. Many of these indices include companies with minimal actual semiconductor revenue, simply because they've announced a fab investment or partnership. Announcement exposure isn't the same as revenue exposure. A company building a fab that won't produce a single chip for three years gets the same index weight treatment as one already generating semiconductor revenue today. ──────────────────── India's semiconductor story is real. It's also a decade-long build, not a one-year trade. Know the difference between a company executing and a company that issued a press release. Follow @IndexDecoded — index mechanics, ETF construction, India markets.

@sumitkbehal Dip

@KobeissiLetter Short

@Akshat_World A bubble in AI hardware is a bubble in Korea and Taiwan, not here. KOSPI doubled this year. Taiwan overtook India in market cap

@rustapharian You buy ten funds. They all own the same 50 stocks. You pay 1-2% for overlap. SPIVA India tracked this. 80% of active large-cap funds lost to the index over ten years. A Nifty 500 index fund costs 0.2%. The fee gap is your underperformance