Fat tails: why huge crashes aren’t as rare as the maths says

Think about the heights of all the children in a big school. Most of them are somewhere in the middle - not very tall, not very short. Only a few are really tall, and only a few are really short. If you drew a picture of how many children were at each height, you would get a shape like a small hill: high in the middle, and sloping down low on both sides. Maths people call this hill a bell curve, because it looks a bit like a bell.

The bell curve is one of the most useful ideas in maths. It says: ordinary things cluster near the middle, and the further you go from the middle, the rarer things become. A child twice as tall as everyone else? That almost never happens. So the two ends of the bell - the far left and the far right - are called the tails, and in a normal bell curve those tails are very thin. Extreme things are so rare there that you can almost forget them.

Benoit Mandelbrot was a mathematician who spent his life looking at shapes and patterns. And he gave the world a warning: markets are not shaped like a school full of children. The share market has much fatter tails than the bell curve says. Huge, scary events - enormous crashes, giant jumps - happen far, far more often in markets than the neat bell-curve maths predicts. This study is about that one idea, because getting it wrong is how careful-looking people get badly hurt.

Here is the picture to hold in your head. Draw the bell curve - the calm hill where almost everything sits near the middle and the ends fade away to almost nothing. Now draw the real shape of market days on top of it. In the middle they look almost the same. But out at the ends, where the crashes and the giant jumps live, the real curve does not fade away. It stays fat. There is much more weight out there than the bell allows.

Why does this matter so much? Because the whole point of a tail is what lives there: the rare, huge days. A bell curve tells you a giant one-day crash should happen maybe once in a thousand years. But real markets have giant crashes every few decades - sometimes twice in one lifetime. That is the gap. The maths said "almost never," and reality said "quite often, actually."

The reading skill is simple to say and hard to remember: when someone tells you a big market disaster is "almost impossible," ask what shape they assumed. If they quietly used the calm bell curve, their "almost impossible" is worth very little, because the real tails are fat. The rare and scary is not as rare as the neat maths promises. A wise reader treats big crashes not as freak once-in-forever events, but as things that will come around again, even if nobody can say when.

illustrative Suppose Arjun studies an imaginary share, "Sagar Textiles." On most days it moves up or down by about 1%. Using the calm bell curve, he works out that a fall of 20% in a single day is so rare it should happen roughly once every few thousand years. So he decides he never needs to worry about it. He borrows money to buy more, sure that a giant one-day fall simply cannot happen in his lifetime.

Now watch what fat tails do. The bell curve said a 20% crash was a once-in-thousands-of-years event. But because real markets have fat tails, a shock hits - a sudden panic, everyone selling at once - and Sagar Textiles falls 22% in a single day. On the calm maths this was "impossible." In the real, fat-tailed world it was simply one of those big-tail days that come around now and then. Arjun, who borrowed money believing the crash could never come, is wiped out - not because he was unlucky beyond all reason, but because he trusted a shape that was too thin at the ends. The move was rare, yes. But "rare" and "impossible" are not the same word, and the fat tail is exactly the space between them.

Fat tails do not tell you when. This is the most important limit. Knowing the tails are fat tells you a giant move is more likely than the calm maths says - it does not tell you it will happen tomorrow, or this year, or in ten years. People who hear "big crash coming" and sell everything can sit out of the market for years while it keeps rising. The idea is a warning about possibility, not a calendar of dates.

Not every scary story is a fat tail. Some people use "anything can happen" as an excuse to panic at every wobble. But most days really are ordinary and sit near the middle of the hill, just as the bell curve says. Fat tails are about the edges - the truly huge events - not about every 2% dip. Treating every small drop as the start of doom is its own mistake.

You cannot measure the tail exactly. Nobody has a perfect number for how fat the tails really are, because the giant events are, by their nature, few. So this is a way of thinking, not a precise formula. It tells you to leave room for the huge and rare. It does not hand you the exact odds.

This idea needs no special maths to feel true - you can see it in the monsoon. Most years the rain is ordinary, close to the average. But every so often a flood comes that is far bigger than anything in the last fifty years, and it does far more damage than all the ordinary years put together. A village that built its houses only for "average" rain gets swept away. That flood is a fat tail.

Our markets have fat tails just like every other market in the world. There have been days when Indian shares fell hard and fast, far more than any calm bell curve would predict, when panic spread and everyone rushed for the exit at once. The lesson for an Indian reader is not to guess the date of the next big fall - nobody can. It is to build your house for the flood, not just the average rain: do not borrow so much that one huge day can sink you, and do not believe anyone who says a crash is "practically impossible." In a fat-tailed world, the impossible-sounding day is the one that eventually arrives.

• Ask what shape they assumed. When someone says a disaster is "almost impossible," check whether they quietly used the calm bell curve. If so, their comfort is worth little.
• Watch for the word "never." Big crashes are rare, not never. Anyone promising a giant fall simply cannot happen is trusting tails that are too thin.
• Separate "rare" from "impossible." The fat tail is the space between those two words - events that seldom happen but absolutely do.
• Do not size your bets for the calm middle. If a single huge day would wipe you out, you have built for average rain in a land of floods.
• Remember the middle is still ordinary. Most days are normal and near the centre. Fat tails are about the edges, not about panicking every week.