Predicting the Market - Roughly

Imagine you ask two questions about the sky.

The first is: "Will it rain next Tuesday afternoon?" Nobody on earth can really answer that. The best weather person, with the biggest computer, is mostly guessing more than a few days out. Tuesday might be sunny, it might pour, and anyone who tells you they know is fooling you.

The second question sounds almost the same, but it isn't: "Roughly how much rain will this town get over the next ten years?" That one you can actually answer, pretty well. You don't know any single day, but you know the climate. A town in the desert will stay mostly dry over ten years. A town in the hills will get soaked most years. The exact days are a mystery; the rough shape of a whole decade is not.

The stock market works exactly like the sky. There are two questions people constantly muddle together. One is: "Will the market go up next week, or next month?" That question has no honest answer - it's the "rain on Tuesday" question, pure guessing dressed up in fancy words. The other is: "Roughly what kind of return should I expect from the market over the next ten years?" And that question - surprisingly, wonderfully - you can answer, not exactly, but roughly, in a way that's useful.

This whole chapter is about keeping those two questions apart. You will never learn to time the market, because nobody can. But you can learn to read, from the price today, a rough clue about the decade ahead - and then do the one sensible thing: stay invested, while expecting less when things are expensive and more when they're cheap. The price you pay today quietly sets your odds: buy when the market is dear and your next decade of returns tends to be thin; buy when it's cheap and scary and the decade ahead tends to be fat.

You might think, "If I can't know the exact number, what good is a rough one?" But a rough answer is worth an enormous amount - as long as you use it for the right job.

Here's the wrong job: using it to jump in and out. That never works, and we'll see why. Here's the right job: setting your expectations so you don't do something silly.

Think about what actually hurts ordinary investors. It's almost never that they picked a slightly worse fund. It's that they expected the wrong thing, got surprised, and panicked. Someone is promised 20% a year, earns 8%, feels cheated, and quits at the worst moment. Someone buys when everything is expensive, expecting the good times to roll on forever, and then feels betrayed when the decade turns out flat. The damage comes from the gap between what they expected and what was ever realistic.

A rough forecast closes that gap. If you know, going in, that a very expensive market is likely to give you a thin decade, you won't be shocked when it does - and you won't sell in disgust. If you know a frightening, cheap market is quietly handing you better odds, you'll keep buying while everyone else runs. The rough number doesn't tell you what to do minute to minute. It tells you what to expect over years, and expecting the right thing is half of not doing something stupid.

There's a second reason it matters, and it's about honesty. Once you can pencil a rough return yourself, you become very hard to fool. When a salesman promises you 18% a year, you can quietly check it against reality and see that the number is a fairy tale. You stop needing to trust anyone's story, because you can do the back-of-the-envelope sum yourself. That small skill - a rough sum on the back of an envelope - is a kind of armour.

So where does the rough number come from? To see it, we have to ask a simple question people almost never ask: when you own a piece of the whole market, why does your money grow at all? What is the return actually made of?

It turns out to come from just two everyday things added together.

The first is the dividend yield - the cash the companies hand back to you each year, as a slice of what you paid. Suppose the whole market costs ₹100 for a share of it, and over a year the companies pay you ₹2 in dividends. That's a 2% yield. It's real money, arriving whether the price wobbles or not - like rent from a shop you own.

The second is the growth - the fact that those companies, over the years, sell a bit more, earn a bit more, and so pay a bit bigger dividend next year than this year. If the dividend grows from ₹2 to a little over ₹2 each year, that steady climb, year after year, is the second engine of your return.

Add the two together and you have a rough estimate of what a whole market will give you over a long stretch: the yield you can see today, plus the growth of that yield. That's the whole recipe. It has a proper name - grown-ups call it the yield-plus-growth idea, or the Gordon equation - but the recipe is just: cash-you-get-now + how-fast-that-cash-grows. A market's long-run return is roughly the dividend yield you can buy right now plus the real growth of those dividends - no crystal ball needed, just two plain numbers added up.

Look closely at the two bars, because they hide the whole secret. The growth block is the same height in both - companies grow at roughly the same pace whether their shares happen to be cheap or dear that year. The thing that changes is the yield block. And the yield block is set by one thing: the price you paid. That's the hinge the entire chapter turns on, so let's make it plain in the next section.

Here is the part that feels almost magical the first time you see it: for the same stream of dividends, the more you pay, the less you earn - automatically, without anything about the companies changing at all.

Picture a small shop that reliably pays its owner ₹10,000 of profit a year. If you buy that shop for ₹1,00,000, your yield is 10% - ten thousand out of a lakh. Now suppose the shop is exactly the same, earns the exact same ₹10,000, but you get excited and pay ₹2,00,000 for it. Your yield just halved to 5%. The shop didn't get worse. You made your return worse by paying more for the same thing.

Price and yield sit on opposite ends of a seesaw. Push the price up, and the yield you get goes down. Let the price fall, and the yield you get goes up. Nothing about the underlying business has to move - the seesaw is pure arithmetic. And since your long-run return is yield plus growth, and growth barely changes, this means the price you pay is the single biggest lever on the return you'll earn.

This is why "the market is expensive" is not just gossip - it's a genuine warning about your future returns. When the whole market is dear, the yield side of the seesaw is pressed low, so the total return you can expect from here is thin. When the market is cheap and frightening, the yield side is pushed high, so the return waiting for you is fat. The scary moment and the generous moment are the same moment.

Hold on to the seesaw. It's the reason today's price is a rough clue to tomorrow's decade - and the reason nobody needs a crystal ball to make that clue.

Let's put the recipe to work with real rupees, the way you actually would on a scrap of paper. illustrative

Meet Rohan, who has been putting money into a plain index fund that tracks the whole Indian market. One evening a cousin corners him at a wedding and swears he can get Rohan 18% a year, easily, forever. Rohan doesn't argue. He just quietly does the sum in his head that this chapter has taught him.

He looks up two boring numbers. First, the market's dividend yield right now is about 1.5% - that's the cash the companies are handing back, as a slice of today's price. Second, he thinks about growth: over long stretches, Indian companies have grown their earnings and dividends at a healthy real pace, and he pencils in something sensible like 5% a year after inflation. He doesn't pretend to know it to the decimal; he picks a fair, un-greedy number.

Then he simply adds them: 1.5% + 5% = 6.5% real return a year, roughly, over a long stretch. That's his back-of-the-envelope anchor for what the whole market is likely to hand him over the next decade - a fair, believable number he built himself from two facts anyone can look up.

Now hold that ₹6.5% up against the cousin's promise of 18%. The gap is enormous. For the cousin to be right, either the yield would have to leap far higher than it is, or the companies would have to grow more than three times faster than they ever have for long - neither of which is going to quietly happen. Rohan doesn't need to prove the cousin is lying. He just smiles, because his own scrap of paper already told him the promise is a fantasy. That's the quiet power of the recipe: it makes you impossible to dazzle. Rohan keeps his money in the plain fund and enjoys his dinner.

Notice what Rohan did not do. He didn't predict next year, or next month. He didn't decide to jump in or out. He just set an honest expectation for a decade and used it to spot a lie. That's the whole job.

Now let's feel the seesaw bite, using two savers who buy the very same companies but at different prices. illustrative

Meet Aarvi and her friend Haridya. Both decide to invest ₹5,00,000 in a fund holding the whole market. The only difference is when the price happens to find them.

Aarvi buys during a calm, cheerful stretch when everyone is optimistic and prices are stretched high. At that price the market's dividend yield is a skinny 1%. She adds the same 5% growth, and her honest anchor for the decade ahead is about 6% a year. Perfectly fine - but thin, because she paid up.

Haridya, by luck of timing, puts her money in a year later, right after a frightening fall when the headlines are all doom and prices have been crushed. At her entry the yield has jumped to 3.5%, because the seesaw tipped when prices dropped. She adds the same 5% growth, and her anchor for the decade is about 8.5% a year.

Same companies. Same growth engine. Same ₹5,00,000. Yet Haridya's likely decade is meaningfully richer than Aarvi's - and the only reason is the price each paid. Over ten years, that gap between roughly 6% and 8.5% compounds into a strikingly different pile of rupees, even though neither of them was cleverer than the other and neither picked a single stock.

Here's the twist that catches people out. Haridya's better deal came wrapped in fear. She bought when the news was ugly and it felt reckless; Aarvi bought when everything felt safe and sunny. The comfortable moment gave the worse return; the scary moment gave the better one. That's not bad luck - it's the seesaw doing exactly what it always does. The scariest, cheapest moments quietly carry the best long-run odds, and the calm, expensive moments carry the worst - so a low price bought in fear is usually a gift, not a warning.

Let's zoom out from single people to the whole market, and watch how faithfully the rough clue plays out over many decades of history - because this is where the idea earns its keep. illustrative

Imagine we lined up hundreds of past ten-year stretches and sorted them by one thing: how expensive the market was at the start of each stretch. A simple way to measure "expensive" is to ask how many rupees you had to pay for each rupee of company earnings - a high number means dear, a low number means cheap.

When you do this, a clear, stubborn pattern appears. The decades that began with the market very expensive tended to deliver thin returns over the ten years that followed. The decades that began with the market beaten-down and cheap tended to deliver generous returns. It isn't a perfect line - some expensive starts did okay and some cheap starts disappointed - but the tilt is unmistakable and it shows up again and again, across countries and across eras.

Sit with the scatter for a second, because it's honest in two directions at once. The cloud slopes clearly downward - that's the real, useful signal that expensive starts hand you thinner decades. But the dots are a cloud, not a tidy line - that's the equally-real warning that the clue is rough. It tells you the odds, the way knowing a town's climate tells you the odds of a wet decade. It does not tell you any single year, and it certainly can't tell you what happens next month.

This is exactly how Aarvi, from earlier, should hold the number. Suppose the market she bought into looks expensive, so her honest anchor is a thin 6%. That doesn't mean she should sell and hide. It means she should keep her monthly investing going while quietly expecting less for a while - and if a scary fall comes and prices drop, she should feel something close to relief, because the seesaw has just tipped the odds in her favour for every rupee she invests from now on. She lets the valuation gently tilt her expectations; she never lets it flip her in or out.

The great mistake with this whole idea is beautifully simple: people take a rough clue about a decade and try to use it to time a day. They hear "the market is expensive, expect thin returns," and their brain instantly rewrites it into "sell now, buy back at the bottom." And that rewrite quietly ruins them.

Here's why it fails. A cheap market can get cheaper and stay cheap for years before it rewards you. A dear market can get dearer and keep climbing far longer than feels sane. So the person who sells because "it's expensive" often watches it rise for another two or three years, feels foolish, and buys back in higher than they sold - having paid a fee in both directions and missed the dividends in between. To win by timing you have to be right twice - the moment to leave and the moment to return - and being right twice, again and again, is something essentially nobody manages. The rough clue was never a timing signal. It's a slow dial for expectations, and treating a slow dial as an on/off switch is how careful people blow themselves up.

Now for the other half of the truth, the half that keeps you humble. Everything above is about a rough clue for a long stretch. The instant someone gives you a precise number for a short stretch, you are no longer looking at analysis. You are looking at fortune-telling in a suit.

Turn on any business channel near the new year and you'll hear confident people announce exactly where the market will be twelve months from now - a specific number, said with a straight face. It sounds scientific. It is not. Nobody has ever been able to do this reliably, and the ones who happen to be right one year are usually wrong the next, because they were guessing all along. The precision is the tell: the more exact and short-term the prediction, the more certainly it's invented. Real knowledge about markets is fuzzy and long-range, like climate. Fake knowledge is sharp and short-range, like a horoscope that names the day. A confident, precise, short-term market forecast is fortune-telling dressed up as expertise - the sharper and sooner the number, the more surely it was made up.

And here's the thing to hold in both hands at once, because it's easy to get confused: the rough decade-clue is useful, and the precise year-forecast is nonsense, and they are not in conflict. It's exactly like the sky again. "This region gets heavy rain most years" is useful climate knowledge you can plan a whole life around. "It will rain 4mm next Tuesday between two and three o'clock" is a made-up claim, no matter how confidently the man on television says it. Knowing the difference is the entire art. Trust the fuzzy long clue that comes from plain arithmetic; laugh gently at the sharp short prophecy that comes from someone's imagination. The moment you mix them up - either by ignoring the honest rough clue, or by believing the dishonest precise one - is the moment the market starts taking your money.

The rough clue is genuinely useful, but honesty demands we show you its edges, because a tool you trust blindly will eventually cut you.

First, the recipe leans on dividends behaving normally, and sometimes companies change their habits. Increasingly, firms hand cash back not by paying dividends but by buying back their own shares - which rewards owners in a different way that the plain dividend yield doesn't fully capture. When that happens, the raw yield understates what's really coming back to you, and a strict yield-plus-growth sum can read too gloomy. The fix isn't to throw the recipe out; it's to remember it's an estimate built on an assumption, and to sense-check whether the companies are returning cash in ways your simple yield number is missing.

Second, the growth number is a guess, and it's the part people cheat on. It is horribly tempting, when you want a market to look attractive, to quietly dial the growth assumption up - pencil in 9% instead of a sane 5% and suddenly the sum promises a lovely return. But wishing doesn't make companies grow faster. The discipline is to keep the growth number honest and un-greedy, anchored to what businesses have actually managed over long stretches, and to be suspicious of yourself whenever a high growth figure happens to justify a price you already wanted to pay. A recipe is only as truthful as the numbers you feed it.

Third, and most important, remember what the clue is for. It is a slow, decade-scale anchor for expectations - not a forecast for any single year, and never a timing switch. Cheap can get cheaper for years; dear can stay dear for years. If you demand that the clue tell you when, it will fail you every time, because that was never its job. Used as a dial - expect less when dear, more when cheap, and keep steadily investing through both - it's one of the most trustworthy ideas in all of investing. Used as a switch, it's just another way to lose. The tool is excellent; the danger is always in asking it a question it was never built to answer.