Things that happened, and what I think they actually did
to us. The opinions are the point — you can get the dates anywhere.
1985World
Plaza Accord
Five governments agreed to talk the dollar down — and it worked, falling ~40% against the yen in two years. The lesson wasn’t about currencies. It was that coordinated intervention can move a market further than fundamentals suggest, for longer than you can stay solvent.
1987World
Black Monday
The Dow lost 22.6% in a single session with no news event to explain it. Portfolio insurance was supposed to limit losses and instead manufactured the selling. Every risk model since has had to answer for that day, and most still can’t.
1991India
The pledge, and the opening
India had three weeks of import cover and flew 47 tonnes of gold to the Bank of England as collateral. Liberalisation followed in July. The reforms that stuck were the ones nobody had time to argue about — a crisis is the only window when an entrenched system agrees to change.
1992India
Harshad Mehta
The Sensex peaked in April 1992 and the scam unravelled weeks later. The mechanism was mundane: bank receipts recycled as collateral. Fraud in Indian markets has almost never been exotic; it has been ordinary plumbing, used at a scale nobody was checking.
1997World
The Asian crisis
Thailand broke its dollar peg and contagion took Indonesia, Korea and Malaysia within months. Currency mismatch was the common thread — borrow in dollars, earn in local. India was largely spared because it hadn’t opened the capital account. Being late was, that once, protective.
1998India
Pokhran and sanctions
Nuclear tests brought US sanctions and a downgrade. Markets fell, then recovered faster than anyone expected. Geopolitical shocks price in quickly when they don’t change cash flows — the market’s memory for sanctions is far shorter than the diplomacy suggests.
2000World
The dot-com unwind
The Nasdaq lost ~78% peak to trough. India’s own IT names went with it, then Ketan Parekh’s K-10 basket collapsed in 2001. Two bubbles, one structure: a narrative that made valuation feel like a failure of imagination.
2001World
September 11
Markets closed for four days and reopened sharply lower. The durable consequence wasn’t the sell-off, it was the rate-cutting cycle that followed — cheap money for years, and a housing market that took full advantage.
2004India
The election nobody priced
The NDA was expected to win. It didn’t. On 17 May 2004 the Sensex fell ~11% and trading halted twice. The most expensive risks are the ones where consensus is so complete that nobody bothers hedging.
2008World
Lehman
The crisis is told as a housing story. It was a funding story — institutions financing 30-year assets with 30-day money. India’s banks survived on luck as much as prudence, and the regulation that followed is a large part of why 2020 went better.
2009India
Satyam
A promoter confessed to fabricating ₹7,000 Cr of cash that never existed. Auditors signed it for years. This is why I run a Beneish screen before I run a DCF — the second is worthless if the first is lying.
2013World
The taper tantrum
Bernanke suggested the Fed might buy fewer bonds. The rupee fell to 68.8 and India was named a Fragile Five economy. A change in the second derivative of policy repriced an entire asset class — markets trade the direction of change, not the level.
2014India
A mandate, and cheap oil
A decisive election result coincided with crude collapsing from $110 to under $30. The reform narrative got the credit; the terms-of-trade windfall did much of the work. Worth remembering when attributing performance to policy.
2016India
Demonetisation
86% of currency by value was voided overnight. The stated goals mostly failed. The unstated one — pushing household savings into financial assets — worked almost too well, and every asset manager in India is still living off the SIP flows that followed.
2016World
Brexit
A vote that no serious model gave better than a third of a chance. Sterling had its worst day in modern history. The recovery took months, not years — the second lesson in a decade that political shocks decay faster than economic ones.
2017India
GST
One tax replaced seventeen. The transition was messy, working capital tightened, and the unorganised sector took the brunt. The lasting market effect was formalisation — a decade of quiet share transfer to listed players that’s still running.
2018India
IL&FS
An infrastructure lender nobody thought of as systemic defaulted, and the NBFC funding market froze behind it. Liquidity risk is not credit risk, and the two are usually confused until the week they aren’t.
2020World
March 2020
Nifty fell ~38% in about a month, and diversification stopped working for roughly six weeks. Every backtest you will ever read quietly assumes correlations hold. They hold until the moment holding them matters.
2021India
The retail arrival
Demat accounts more than doubled and SIP flows became the market’s ballast. Domestic money structurally changed who sets the price in Indian equities — FII selling no longer does what it did in 2008.
2022World
Ukraine, and the end of free money
War, an energy shock, and the fastest Fed tightening in forty years. Duration was repriced across every asset at once — the clearest reminder in my investing lifetime that the discount rate is not a background assumption.
2023India
Hindenburg
A short-seller’s report erased over $100bn of group market value in weeks. Concentration risk hid inside an index — passive holders owned a governance question they had never underwritten.
2023World
Silicon Valley Bank
A bank failed because it held government bonds — the safest asset, at the wrong duration, funded by depositors who could all leave on the same afternoon. “Risk-free” describes credit, never price.
2024India
4 June 2024
Exit polls said one thing. The count said another. The Nifty fell ~6% intraday and recovered within days. Twenty years after 2004, the same lesson, at the same price: consensus is not a hedge.
2026World
Hormuz
A supply shock that repriced everything, and then mostly didn’t happen. I initiated ONGC and Oil India as a hedge and wrote it up at the time. What it cost to be wrong was far smaller than what it would have cost to be unhedged.