00.1 · What fundamental investing is — and isn't

Why this matters: Every later module (accounting, ratios, valuation, forensics) is a tool for one question: is this price a good deal for the cash this business will produce over its life? Without that question the tools turn into trivia. This lesson also sets honest expectations. It covers where an edge can come from, why it takes years to show up in your results, and what returns are realistic.

Learning objectives — after this lesson you can:

• Explain the difference between price and value, and value a simple stream of cash flows as a claim on a business.
• Apply the three questions (what is the business, is it good, what is it worth compared with the price) to a real or fictional company.
• Describe seven schools of fundamental investing and the kind of mispricing each one looks for.
• Classify a claimed edge as behavioural, structural, informational or analytical, and state the efficient-markets objection and its best rebuttal.
• Calculate why a fundamental investor's skill takes years to show up statistically, compared with a market-maker's.
• Set realistic return expectations and spot survivorship bias in "multibagger" stories.

Prerequisites: none. This is the first lesson. · Time: ~100 min

---

1. Price is what you pay; value is what you get

Warren Buffett credits Ben Graham with the line "Price is what you pay; value is what you get" (Berkshire Hathaway 2008 letter). The whole discipline rests on the gap between those two words, so we define them carefully.

• Price: the last price at which a buyer and a seller agreed to exchange one share. It is set by whoever happens to be trading, for whatever reason they have, and it changes every second.
• Market capitalisation (market cap): price × number of shares outstanding. It is what the market charges for the whole of the company's equity.
• Intrinsic value (or just value): the present value of all the cash the business will hand to its owners over its remaining life. You cannot observe it. You can only estimate it, and two careful analysts will get different answers.
• Fundamental investing: buying a security (or selling it short) because your estimate of its intrinsic value differs from its price by enough to pay you for the chance that you are wrong. You then hold it until the gap closes or your estimate changes.

Look at what that definition leaves out. It says nothing about predicting next month's price, or about charts, macro calls or momentum. It does not say that low P/E stocks are good. It does not even say "long-term". A fundamental investor can hold for six weeks if the gap closes in six weeks. Holding is a consequence, not the goal.

1.1 What fundamental investing is not

Approach	What it bets on	Typical horizon	How it differs from fundamental investing
Technical analysis / chart trading	Patterns in past prices and volumes	Days–weeks	Ignores what the business earns; the price series is the whole input
Macro trading	Rates, currencies, commodities, policy	Weeks–months	Trades economies, not individual businesses
Quant / factor investing	Statistical regularities across thousands of stocks (value, momentum, quality)	Months	Uses many of the same signals, but systematically; rarely judges an individual business
Index investing	Nothing: owns the market at low cost	Decades	Accepts the market's price as the best estimate of value
Market-making / volatility trading	Spread capture, flow, the difference between implied and realised volatility	Seconds–weeks	Does not need a view on what the business is worth
"Tips" and narratives	A story, or someone else's conviction	Unpredictable	No independent estimate of value, and so no way to know when you are wrong

Graham described the market as a moody business partner, "Mr. Market" (The Intelligent Investor, 1949). Every day he offers to buy your share of the business or sell you his, at a price that depends on his mood. You are free to ignore him. His offers are useful only when they are foolish. That is the fundamental investor's relationship with price: price is information about other people's expectations, not a verdict on value.

2. A share is a claim on future cash flows

A company has two broad kinds of financiers. Debt holders (lenders) are promised fixed interest and repayment. Equity holders (shareholders) are the residual claimants: they get whatever is left after lenders, suppliers, employees and the tax authorities have been paid. Lesson 01.1 covers this in detail. For now the key idea is that a share is a claim on a stream of residual cash that runs for years.

The value of any claim on future cash is its present value (PV). You shrink each future rupee by a discount rate \(r\), the annual return you require for bearing the risk and waiting:

\[ V_0 \;=\; \sum_{t=1}^{\infty} \frac{CF_t}{(1+r)^t} \]

In words: value today equals each year's cash flow to owners, \(CF_t\), divided by one-plus-the-required-return compounded \(t\) times, summed over the life of the business. If the cash flow starts at \(CF_1\) and then grows at a constant rate \(g < r\) for ever, the sum collapses to the Gordon growth formula. You will derive it in 01.5:

\[ V_0 \;=\; \frac{CF_1}{r-g} \]

Worked example 1 — a toy business, and where its value comes from

A small, debt-free business will produce ₹10 Cr of cash for its owners next year. That cash will grow at 5% a year for ever, and you require 12% a year.

\[ V_0 = \frac{10}{0.12 - 0.05} = \frac{10}{0.07} = ₹142.9\text{ Cr} \]

Now ask how much of that ₹142.9 Cr comes from the next few years. Add up the discounted cash flows for the first \(N\) years:

Horizon	PV of cash flows in years 1…N (₹ Cr)	Share of total value
1 year	8.9	6.2%
3 years	25.2	17.6%
5 years	39.4	27.6%
10 years	67.9	47.6%
20 years	103.6	72.5%
30 years	122.3	85.6%

Less than a fifth of the value comes from the next three years, and more than half comes from beyond year ten. A share is mostly a claim on the distant future. That is why a bad quarter often matters less than the market's reaction suggests. It is also why a lasting change in the long-run outlook matters much more.

Now change the inputs slightly:

Required return \(r\)	Growth \(g\)	Value (₹ Cr)	Change vs base
12%	5%	142.9	–
12%	6%	166.7	+16.7%
11%	5%	166.7	+16.7%
11%	6%	200.0	+40.0%
13%	5%	125.0	−12.5%
13%	4%	111.1	−22.2%

A one-point change in either input moves the value by 12–17%. Change both together and the value moves by 22–40%. Any single "intrinsic value" figure is really the centre of a wide range. Module 06 turns this into sensitivity tables and scenario analysis (06.4).

Trader's lens — equity is a long-duration asset with an implied growth rate

Differentiate the Gordon value with respect to \(r\): \(-\frac{1}{V}\frac{\partial V}{\partial r} = \frac{1}{r-g}\). The toy business has an equity duration of \(1/0.07 \approx 14.3\) years. It behaves like a long-dated bond with uncertain coupons. The response is convex: +1 pt on \(r\) costs 12.5%, −1 pt gains 16.7%. Turn the formula round and the market price tells you what growth the market expects: \(g = r - CF_1/P\). At a price of ₹200 Cr and \(r = 12\%\), the market expects \(g = 12\% - 5\% = 7\%\). This is the same move as backing implied volatility out of an option price. You are not asking "what is it worth?" but "what does this price assume, and do I disagree?" Module 06.6 formalises this as the reverse DCF.

Worked example 2 — what do you actually get for ₹390 of Kaveri Pumps?

The course's running example is Kaveri Pumps & Motors Ltd (fictional), a Coimbatore pump and motor maker. Its full data is in the reference page. On 18-Sep-2026 its share price was ₹390, with 6.00 Cr shares outstanding.

Item	Arithmetic	Result
Market cap	₹390 × 6.00 Cr shares	₹2,340 Cr
FY26 profit after tax (PAT)	from the income statement	₹90.5 Cr
Earnings yield (PAT ÷ market cap)	90.5 ÷ 2,340	3.9%
P/E (price ÷ earnings)	2,340 ÷ 90.5	25.9x
FY26 dividend per share	from the ratio table	₹4.0
Dividend yield	4.0 ÷ 390	1.0%
FY26 free cash flow (FCF) = operating cash flow − capex	65.1 − 48.0 − 4.0	₹13.1 Cr
FCF yield	13.1 ÷ 2,340	0.6%

(PAT is profit after all costs, interest and tax. Operating cash flow is the cash the business actually collected from operations. Capex is capital expenditure on plant, equipment and software. All three are built from scratch in Module 02.)

Someone paying ₹390 today gets about 1% a year in dividends and a business that produced only 0.6% of its price in free cash last year. Almost all of the return has to come from future growth in cash flow. The course's reference valuation (kaveri-valuation) makes this concrete:

• The present value of the forecast cash flows for FY27–FY29 is 102.4 + 96.1 + 93.4 = ₹291.9 Cr, only 13.9% of the ₹2,100 Cr enterprise value. (Enterprise value is the value of the whole business before subtracting debt; see 01.2.)
• The first ten years together are worth ₹947.7 Cr (45%). The terminal value, everything after FY36, is 55%.
• The base-case value is ₹320 per share, against a price of ₹390. The probability-weighted value across bull, base and bear scenarios is ₹306. On those numbers the price is 22% above the base case (390 ÷ 320 = 1.22) and value sits 18% below price (320 ÷ 390 − 1 = −17.9%).

This does not mean "sell Kaveri". It means that at ₹390 the market expects more than the course's base case. Module 06.6 shows that ₹390 implies about 14.1% revenue growth a year for ten years, against a base case of 10.8%. Deciding whether that expectation is too high is the fundamental investor's actual job. (The valuation is dated 31-Mar-2026. Rolling it forward to September at the cost of equity gives about ₹338. We ignore that refinement until Module 06.)

3. The three questions

Every piece of analysis in this course serves one of three questions. Ask them in this order. Most bad investments come from skipping the first two because the third looked exciting.

Q1. What is the business? How does it make money, from whom, and against whom? What is sold (a pump, a loan, a subscription)? Who pays, how often, and on what credit terms? What does it cost to make one more unit? If you cannot explain it in three plain sentences, you do not understand it well enough to value it. → Taught in 05.1 Business models and 05.2 Industry analysis.

Q2. Is it a good business, run by people you can trust? "Good" has a precise meaning. The business earns a return on invested capital (ROIC) above its cost of capital (WACC) and can keep doing so. ROIC is operating profit after tax divided by the capital tied up in the business. WACC is the blended return its lenders and shareholders require. Behind that sit the moat (a durable competitive advantage that protects those returns), the balance sheet, and the honesty and skill of management. → Modules 04 (returns, cash conversion), 05 (moats, management, governance) and 09 (forensics).

Q3. What is it worth, compared with the price, and what does the price already assume? Estimate a range of values, understand what the market is implicitly forecasting, and decide whether the gap is big enough. → Modules 06–07.

Apply them quickly to Kaveri (we do this properly in 00.3):

1. Business: agricultural, domestic and industrial pumps and motors sold through ~1,800 dealers, plus a fast-growing solar-pump segment sold to state agencies under a government scheme. Solar was 27% of FY26 revenue, up from 3% in FY21.
2. Good? Mixed. FY26 ROIC was 12.0%, almost exactly equal to the reference WACC of 12.19%. At today's returns, growth barely creates value (Lesson 06.1 proves why). Receivables are ballooning and the promoters have pledged some shares.
3. Worth vs price? Base value ₹320 and probability-weighted ₹306, against a price of ₹390. The price implies more growth than our base case.

Point 2 leads to one of the most important ideas in the course. Growth is only worth paying for when the business earns more on new capital than that capital costs. A company growing at 20% with ROIC below WACC destroys value faster the faster it grows.

4. The schools of fundamental investing

All fundamental investors compare price with value. The schools differ in which mistakes they think the market makes and which businesses they are willing to own.

School	Core belief	What they buy	Key figures / texts	Typical failure mode
Deep value	The market overreacts to bad news and neglects dull, cheap assets	Stocks below liquidation value, low P/B, low P/E, "net-nets"	Benjamin Graham & David Dodd, Security Analysis (1934); Graham, The Intelligent Investor (1949)	Value traps: cheap because the business is shrinking or badly run
Quality / compounders	The market underprices how long great businesses can reinvest at high returns	High-ROIC businesses with moats and long reinvestment runways, bought at fair prices	Philip Fisher, Common Stocks and Uncommon Profits (1958); Warren Buffett & Charlie Munger	Overpaying, and moats that erode quietly
Growth	The market underestimates how large a winner can become	Fast-growing revenue and earnings, often at high multiples	Fisher; many modern tech investors	Mistaking growth for value; buying at bubble prices
GARP (growth at a reasonable price)	You can get growth without paying bubble prices	Growth stocks with P/E not far above their growth rate (PEG = P/E ÷ growth ≈ 1)	Peter Lynch, One Up On Wall Street (1989)	PEG ignores return on capital, risk and duration
Special situations	Corporate events create forced or uninterested sellers	Spin-offs and demergers, buybacks, open offers, delistings, mergers, restructurings	Joel Greenblatt, You Can Be a Stock Market Genius (1997)	Deals fall apart; small, illiquid positions
Activism	Value exists but management won't release it	Large stakes, then pushing for change (board seats, payouts, sales, governance)	Many US hedge funds; rarer in India	Expensive, public and slow; control shareholders can block it
Long/short	You can be right about relative value while hedging the market	Long undervalued, short overvalued stocks	Alfred Winslow Jones's partnership, founded in 1949 and usually called the first hedge fund (Institutional Investor)	Shorts have unlimited loss and can squeeze; borrowing costs

A few notes on each:

Deep value. Graham's cleanest test was the net-net: a stock trading below its current assets minus all liabilities, so you get the plant, the brand and the future for less than nothing. His margin of safety idea is to buy only when price is far enough below your value estimate that even a large error leaves you whole. That idea runs through every school, and we restate it as a probability distribution in 06.8. The weakness of deep value is that cheapness is often deserved. A stock at 0.5x book value earning 4% on that book is not obviously a bargain.

Quality and compounders. Fisher added scuttlebutt: talking to customers, suppliers, competitors and former employees to understand a business beyond its accounts (05.7). Buffett, pushed by Munger, moved from Graham-style bargains towards great businesses. In his 1989 letter he preferred "a wonderful company at a fair price" to a fair company at a wonderful price, and said Munger understood this long before he did (Berkshire 1989 letter). The maths in Worked example 1 explains why. If most of the value lies beyond year ten, the quality and length of the reinvestment runway dominate everything else.

Growth and GARP. Growth investors accept high multiples because they expect earnings to outgrow them. The Cisco case (G3) shows what happens when the price already assumes more than even a great company can deliver. GARP tries to control this with ratios such as PEG, but those shortcuts break down once you understand ROIC and duration.

Special situations. These are about the event, not the business's long-run quality. Index funds sell a spun-off subsidiary they never wanted. A buyback tender has a calculable acceptance ratio. An open offer or delisting sets a price floor. Indian versions of these trades are covered in 12.3.

Activism is structurally harder in India, where most listed companies have a controlling promoter (the founding family or group, often owning 50–75%). A contested example: in September 2021, Invesco and OFI Global China Fund, together holding about 17.9% of Zee Entertainment, asked for a shareholder meeting (EGM) to remove the MD & CEO and reshape the board. The company rejected the request and there was litigation. Invesco withdrew in March 2022 (Business Standard, Sep-2021; Business Standard, Mar-2022). In India, "activism" usually means institutions voting against resolutions, often guided by proxy advisers, rather than takeover fights.

Long/short. Jones's idea was to keep the stock-picking skill and remove the market's direction by pairing longs with shorts. In India, short selling in the cash market must be covered by delivery at settlement; naked shorting is not allowed. Institutions may not square off shorts within the day. Most practical short exposure comes through single-stock futures, which exist only for F&O-eligible stocks, or through securities lending and borrowing (SEBI's short-selling framework circular of 5-Jan-2024; verify current rules before acting). 11.7 goes further.

The schools overlap more than their followers admit. Buffett is a value investor who buys quality. A growth investor who insists on ROIC above WACC is doing GARP with better tools. The common core is this course: understand the business, judge its quality, estimate value, compare with price, size for uncertainty. Quantitative factor investing is a close cousin. It harvests the average mispricing across thousands of stocks systematically, where a fundamental investor tries to be right about a few in depth.

5. Why would anything be mispriced?

5.1 The efficient-markets objection

The efficient market hypothesis (EMH), associated with Eugene Fama (Journal of Finance, 1970), says prices already reflect available information. It comes in three strengths:

• Weak form: past prices can't predict future prices (charts don't work).
• Semi-strong form: prices reflect all public information (reading the annual report doesn't help).
• Strong form: prices reflect even private information (nothing helps).

If the semi-strong form held exactly, this course would be pointless. The evidence against active managers is serious. In S&P's SPIVA India Year-End 2025 scorecard, 76.3% of Indian active large-cap equity funds underperformed their benchmark (S&P India LargeMidCap) over the ten years to December 2025. The figures were 82.9% for tax-saving ELSS funds and 79.0% for mid/small-cap funds. Even in 2025, a good year for mid/small-cap funds, 75.0% of large-cap funds lagged (SPIVA India Year-End 2025, data as of 31-Dec-2025). These are full-time professionals with research teams. Take the number seriously.

5.2 The best rebuttal: Grossman–Stiglitz

Sanford Grossman and Joseph Stiglitz ("On the Impossibility of Informationally Efficient Markets", American Economic Review 70(3), 1980; AEA PDF) pointed out a paradox. If prices reflected all information perfectly, nobody would be paid for gathering it. Then nobody would gather it, and prices could not reflect it. So in equilibrium there must be just enough mispricing to pay the people who do the work. Three things follow:

1. Edge exists, but it is a competitive business with costs: time, data, skill and bearing risk. On average, active managers earn roughly what it costs them, and after fees most lose to the index.
2. Your edge has to be specific. "I'm smart and I read annual reports" is the price of entry, not an edge.
3. Mispricing is most likely where research is costly, unrewarded or constrained: small, dull, complicated, or temporarily unpopular companies, and situations where natural holders are forced to sell.

(Robert Shiller's 1981 work showing that stock prices move far more than later dividends justify is the other classic challenge to strong efficiency. Fama and Shiller shared the 2013 Nobel memorial prize, which neatly captures where the debate stands.)

5.3 The four sources of mispricing

Source	Mechanism	Example	How you'd know it's real	Where taught
Behavioural	Predictable human errors: extrapolating recent trends, anchoring on past prices, loss aversion, preferring lottery-like stocks, herding	After three bad quarters a cyclical is priced as if the downturn were permanent	You can show the base rate (how often similar companies recovered) differs from what the price implies	11.6, 12.2
Structural / time-horizon	Rules, mandates or incentives force other investors to buy or sell regardless of value; many are judged quarterly	Index deletions, fund-category limits, margin calls on pledged promoter shares, career risk	You can name the constrained seller and explain why they can't wait	01.4, 12.3
Informational	Public but costly information that few people process: dense notes to accounts, regional data, primary research	A receivables-ageing note in a small cap that no analyst covers	You can point to the specific document and show the price hasn't reacted	03.3, 05.7
Analytical	Same facts, better interpretation: seeing the economics behind the accounting	Recognising that heavy spending on customer acquisition is investment, not cost; or that profit growth funded by rising receivables isn't turning into cash	You can explain the market's current model and exactly where yours differs	Modules 04, 06, 09

Two warnings. First, structural edges are the most reliable because they don't depend on your being cleverer than anyone. Keynes observed that it is "better for reputation to fail conventionally than to succeed unconventionally" (The General Theory, 1936, ch. 12), and fund managers still behave that way. A fund manager who must beat the index every quarter cannot hold a stock that will look bad for two years. You can. In India, fund-category rules (a large-cap fund must keep at least 80% in large caps, the top 100 companies by market cap; see SEBI's categorisation circular of 26-Feb-2026) and promoter-pledge margin calls both create sellers who are not asking what the business is worth.

Second, an "informational edge" must never mean unpublished price-sensitive information (UPSI). Trading on material non-public information is insider trading under SEBI's rules. If your edge is a phone call from someone at the company, you do not have an edge; you have a legal problem (05.7).

Trader's lens — whose constraint are you harvesting?

A market-maker already understands structural edge: you are paid for providing liquidity to people who need it and for absorbing flow that is not price-sensitive. SEBI's own data shows the scale in Indian derivatives. Over FY22–FY24, 93% of more than 1 crore individual F&O traders lost money, with aggregate losses above ₹1.8 lakh crore. In FY24 alone, proprietary traders and FPIs booked gross trading profits of about ₹33,000 Cr and ₹28,000 Cr respectively (SEBI press release, 23-Sep-2024). The fundamental version of the same question is: who is on the other side of my trade, and why are they willing to sell to me at this price? If the answer is "someone who knows more", walk away. If it is "a fund that must sell because of a rule", "a promoter facing a margin call", or "people extrapolating a bad quarter", you may have something.

The edge test. Before any position, write down one sentence for each of these:

1. What does the price imply? (the market's model)
2. What do I believe instead, and why? (my model)
3. Why does the market believe what it does, and why is it wrong? (the source of the mispricing, from the table above)
4. What will make the price move towards my value, and roughly when? (the catalyst)
5. What would prove me wrong? (the thesis-breaker)

If you cannot fill in line 3, you probably don't have an edge. You have an opinion.

6. Your edge as a trader vs your edge as a fundamental investor

If you come from market-making or volatility trading, you already know what a real edge feels like: small, measured, repeated thousands of times, with the P&L confirming it within days. Fundamental investing's edge is statistically different in almost every way, and ignoring that difference is the most common failure of smart traders who move to long-horizon investing.

Dimension	Market-maker / vol trader	Fundamental investor
Horizon of each bet	Seconds to weeks	1–5 years
Independent bets per year	Thousands to millions	5–20 new theses, and they are correlated (same market, often same factors)
Edge per bet vs noise per bet	Tiny edge, tiny noise; the law of large numbers works for you	Large hoped-for edge (20–50% mispricing) against huge noise (a single stock often moves 30–50% a year)
Feedback loop	Same-day P&L; realised vs theoretical edge measured continuously	Years; in the first 1–3 years price noise usually swamps the signal
Shape of outcomes	Aggregated P&L close to normal	Positively skewed: a few big winners, many mediocre outcomes, some near-zeros
How you know you were right	P&L attribution (spread, vega, gamma, theta), realised vs expected edge	Did the business do what you predicted (revenue, margins, cash, returns)? The price confirms late and noisily
Main risk	Inventory, jumps, model error, tail events	Permanent loss of capital: wrong business judgement, fraud, too much debt, overpaying
Capacity limit	Available flow	Liquidity, especially in small caps

Worked example 3 — how long before skill shows up in the data?

Measure skill with the information ratio, \(IR = \alpha / TE\). Here alpha (\(\alpha\)) is your average annual return above the benchmark, and tracking error (\(TE\)) is the annual standard deviation of that excess return. After \(T\) years your estimated alpha has a standard error of \(TE/\sqrt{T}\), so the t-statistic is \(t = IR\sqrt{T}\). For \(t = 2\) (roughly "95% confident it isn't luck") you need:

\[ T = \left(\frac{2}{IR}\right)^2 \]

A very good fundamental investor: \(\alpha = 3\%\) a year after costs, \(TE = 8\%\) (a concentrated portfolio).

\[ IR = \frac{0.03}{0.08} = 0.375 \quad\Rightarrow\quad T = \left(\frac{2}{0.375}\right)^2 = 28.4 \text{ years} \]

Even an excellent investor needs about 28 years of results to prove skill at conventional significance. With \(\alpha = 5\%\) and \(TE = 10\%\) (\(IR = 0.5\)) it still takes 16 years.

The same investor will underperform in any given year with probability \(\Phi(-IR\sqrt{T})\), where \(\Phi\) is the standard normal cumulative distribution:

Window	P(underperform the index) with \(IR = 0.375\)
1 year	35.4%
3 years	25.8%
5 years	20.1%
10 years	11.8%

A genuinely skilled investor trails the index about one year in three, and one five-year stretch in five.

A market-maker: expected P&L of ₹1 lakh a day with a daily standard deviation of ₹2 lakh gives a daily IR of 0.5. Then \(T = (2/0.5)^2 = 16\) trading days, and the annualised Sharpe ratio is \(0.5 \times \sqrt{252} \approx 7.9\).

The contrast is the whole point. In fundamental investing, results are an extremely noisy measure of skill. Three practical consequences run through the course:

1. Judge the process, not the outcome. Keep a decision journal that records why you acted and what you expected, before you know the result (11.6).
2. Shorten the feedback loop with intermediate predictions. "Kaveri's receivable days will fall towards 80 by March 2027" can be checked in months. "Kaveri is worth ₹320" may never be checked cleanly. Good theses are built from falsifiable business predictions (11.3).
3. Size for being wrong. When you can't tell skill from luck quickly, overconfidence is expensive. Position sizing (11.4) uses fractional Kelly for exactly this reason.

What does carry over from trading: thinking in probabilities and expected value, treating price as an implied forecast, respecting sizing and ruin, and emotional discipline around P&L. What doesn't: treating mark-to-market as truth, the urge to act every day, and hedging away the very risk you are paid to take.

7. What returns are realistic?

Some anchors, each with its date:

Benchmark	Figure	As of / source
Nifty 50 Total Return Index (dividends reinvested), annualised since 3-Nov-1995	12.38%	31-Aug-2026, NSE Indices factsheet
Nifty 50 price index, same period	10.86%	same
Nifty 50 TRI, last 5 years (annualised)	8.32%	same
Nifty 50 annualised volatility since inception	22.4%	same
India 10-year government bond yield	≈7.05%	21-Sep-2026, Trading Economics
CPI inflation, August 2026	4.82%	Trading Economics (MoSPI data, released 14-Sep-2026)
RBI inflation target	4% ± 2%, retained to March 2031	Business Standard, Mar-2026
Indian large-cap active funds lagging the benchmark over 10 years	76.3%	SPIVA India YE2025
Berkshire Hathaway per-share market value vs S&P 500 (with dividends), 1965–2025	19.7% vs 10.5% a year	Berkshire 2025 letter (Feb-2026)

The Total Return Index (TRI) assumes dividends are reinvested. It is the fair comparison for anyone's performance, and the gap to the price index (about 1.5 points a year here) is the dividend yield. Note that the 5-year figure (8.32%) is well below the 30-year figure (12.38%). Returns come in lumps, and a five-year window tells you little.

Berkshire's 9-point annual margin over 61 years is the best-known long record in public markets. Treat it as the right tail of the distribution, not a target.

Worked example 4 — what realistic numbers compound to

Invest ₹10 lakh for 20 years. These are illustrations of compounding, not forecasts.

Scenario	Annual return	Growth multiple (20 yrs)	Ending value
Index at its since-inception rate	12.38%	1.1238²⁰ = 10.32x	₹1.03 Cr
Same, less 1.5% a year of costs (fees, brokerage, impact)	10.88%	7.89x	₹78.9 lakh
Index + 3% a year of genuine alpha	15.38%	17.48x	₹1.75 Cr

Three points of alpha kept for 20 years is 1.7x the index outcome (17.48 ÷ 10.32). Leaking 1.5% a year in costs takes away almost a quarter of it (7.89 ÷ 10.32 = 0.76). Anyone who can reliably add 2–4% a year over a decade after costs is very good. Promises of 30–40% a year, sustained, should make you ask what risk is hidden or how short the sample is.

Taxes are part of the return. As of FY27, listed Indian equity gains held more than 12 months are long-term capital gains (LTCG), taxed at 12.5% above an annual exemption of ₹1.25 lakh. Shorter holdings are short-term (STCG) at 20%. These rates took effect on 23-Jul-2024 and Budget 2026 kept them. The Income-tax Act, 2025 replaced the 1961 Act from 1-Apr-2026 without changing them (Business Standard on the 2024 change; PIB on the new Act; detailed treatment and current rules in 11.5). Take a 13% pre-tax return for 20 years, ignoring the exemption, surcharge and cess:

Behaviour	Arithmetic	Multiple	Post-tax CAGR
No tax (reference)	1.13²⁰	11.52x	13.0%
Buy and hold; pay LTCG once at the end	1 + (11.52 − 1) × (1 − 0.125)	10.21x	12.3%
Realise gains every year as LTCG	(1 + 0.13 × 0.875)²⁰	≈8.6x	11.4%
Churn: realise every year as STCG	(1 + 0.13 × 0.80)²⁰	7.23x	10.4%

Deferring tax by holding is worth almost two points a year over churning. That is a structural edge available to every patient individual, and it is invisible in pre-tax performance tables.

8. Survivorship bias and the multibagger myth

Survivorship bias is the error of drawing conclusions from a sample that includes only the survivors. In investing it shows up everywhere:

• Articles on "stocks that turned ₹1 lakh into ₹1 crore" (a multibagger is a stock that multiplies many times; a "10-bagger" rises 10x). Nobody writes about the thousand stocks from the same starting year that went nowhere or were delisted.
• Fund advertisements that show the best schemes. SPIVA's year-end 2025 report found that only 79.4% of the 97 Indian large-cap funds in existence ten years earlier still existed. Across all categories, 27% of funds had merged or been liquidated over the decade. The dead funds' records disappear from the brochures.
• Back-tests on today's index members. The Nifty 50 is rebalanced semi-annually. Today's members are, almost by definition, companies that did well, so testing them over the past 20 years is biased. (Investing in the index itself is not biased. It really did hold the losers until it dropped them.)
• Data sets that quietly exclude delisted and suspended companies.

What the evidence says

Hendrik Bessembinder's study of all US stocks from 1926 to 2016 found that most individual stocks had lifetime buy-and-hold returns below one-month Treasury bills. The best-performing 4% of companies accounted for the entire net wealth created by the US stock market ("Do Stocks Outperform Treasury Bills?", Journal of Financial Economics 129(3), 2018; RePEc abstract). The cause is positive skewness. A stock can lose at most 100% but can gain 10,000%, and compounding stretches the right tail further.

A follow-up covering more than 64,000 stocks worldwide over January 1990 to December 2020 found the same pattern (Bessembinder, Chen, Choi & Wei, "Long-Term Shareholder Returns: Evidence from 64,000 Global Stocks", Financial Analysts Journal, 2023; SSRN 3710251). 55.2% of US stocks and 57.4% of non-US stocks underperformed one-month US T-bills, and the top 2.4% of firms accounted for all US$75.7 trillion of net wealth creation. The India rows (3,967 stocks, returns measured in US dollars) are even more extreme:

India, Jan-1990 to Dec-2020 (lifetime buy-and-hold, USD)	Value
Mean return per stock	+501%
Median return per stock	−33.6%
Share of stocks beating one-month US T-bills	36.3%
Share of stocks beating the value-weighted Indian market	24.2%
Share of gross wealth creation from the top 1% of firms (40 firms)	65.5%
Share of gross wealth creation from the top 5% of firms (199 firms)	92.1%

The average Indian stock multiplied about 6x, while the typical (median) stock lost a third of its value in dollar terms. Both statements are true at once.

Worked example 5 — why the median stock loses while the average stock wins

Model each stock's value as a lognormal random walk: log returns are normally distributed. Assume an expected (mean) return of \(m = 12\%\) a year, volatility \(\sigma = 50\%\) (typical of an Indian small cap), and a 15-year horizon. If \(\ln W_T \sim N(\mu T, \sigma^2 T)\) with \(\mu = \ln(1+m) - \sigma^2/2\), then the mean terminal wealth is \((1+m)^T\) and the median is \(e^{\mu T}\):

• \(\mu = \ln 1.12 - 0.5^2/2 = 0.1133 - 0.1250 = -0.0117\) a year, so \(\mu T = -0.1751\)
• \(\sigma\sqrt{T} = 0.5 \times \sqrt{15} = 1.9365\)

Quantity	Formula	Result
Mean multiple	\(1.12^{15}\)	5.47x
Median multiple	\(e^{-0.1751}\)	0.84x
P(lose money) = P(\(W<1\))	\(\Phi(0.1751/1.9365)\)	53.6%
P(do worse than the average stock)	\(\Phi\big((\ln 5.474 + 0.1751)/1.9365\big)\)	83.4%
P(10-bagger) = P(\(W>10\))	\(1-\Phi\big((\ln 10 + 0.1751)/1.9365\big)\)	10.0%
P(100-bagger) = P(\(W>100\))	\(1-\Phi\big((\ln 100 + 0.1751)/1.9365\big)\)	0.68%

Simulate 1,000 such stocks (Python, seed 42) and you get: 545 lose money, 105 are 10-baggers and 4 are 100-baggers. The top 10 stocks produce 47.9% of all the gains, and the top 40 produce 70.0%. The sample mean is 6.66x against a theoretical 5.47x. With a tail this heavy, even 1,000 draws give a noisy average.

import math, random
random.seed(42)
m, sig, T = 0.12, 0.50, 15
mu = (math.log(1 + m) - sig**2 / 2) * T
s = sig * math.sqrt(T)
W = sorted((math.exp(random.gauss(mu, s)) for _ in range(1000)), reverse=True)
gains = [w - 1 for w in W]
gross = sum(g for g in gains if g > 0)
print(sum(w < 1 for w in W), sum(w > 10 for w in W), sum(w > 100 for w in W))   # 545 105 4
print(round(sum(gains[:10]) / gross * 100, 1))                                    # 47.9

Now imagine a magazine in year 15 profiling the four 100-baggers: "Their founders were visionary; their moats were obvious in hindsight." Every sentence can be true, and the article still teaches you almost nothing. You would also need to know how many companies looked just as visionary in year 0 and how many of them ended among the 545 losers. That number is the base rate, and ignoring it is the core of survivorship bias.

Implications for the rest of the course:

1. Study failures as carefully as successes. That is why Module 13 has Satyam, IL&FS/DHFL, Yes Bank, Enron and Wirecard alongside Asian Paints and Bajaj Finance.
2. Stock-picking is hard mathematically, not only psychologically. A randomly chosen stock is more likely than not to lag the index. Owning the index gives you the right tail automatically. Picking stocks is only worth it if you can identify it better than chance, or avoid the left tail better than chance.
3. Concentration raises the variance of outcomes a lot. With a skewed distribution, a 5-stock portfolio usually lags and occasionally wins big. Sizing (11.4) is where this is dealt with.
4. The downside matters as much as the upside. Avoiding the permanent losers (fraud, over-borrowing, broken business models) is a large part of the edge. That is why forensics has its own module.

India notes

• Promoter control is the norm. Most listed Indian companies have a controlling promoter, so minority shareholders depend on governance more than on the market for corporate control. Governance analysis (05.6) carries more weight here than in US-style markets, and activism is rarer.
• Index concentration. On 31-Aug-2026, financial services were 36.5% of the Nifty 50 and HDFC Bank and ICICI Bank alone about 19% (NSE Indices factsheet). "Beating the Nifty" is partly a bet on or against Indian banks.
• Short selling is legal for all investor classes, but naked shorts are prohibited and institutions cannot day-trade. Practical shorting for most investors means single-stock futures on F&O-eligible names (SEBI circular of 5-Jan-2024). Long/short is therefore harder to run in Indian small caps.
• Structural flows from SEBI's market-cap-based fund categories, index inclusions and exclusions, and steady SIP inflows are real and recurring sources of forced buying and selling (see 01.4).
• Tax rewards patience: LTCG at 12.5% after 12 months against STCG at 20% (FY27; verify current rules in 11.5).

Common mistakes

• Confusing a great company with a great investment. Quality is question 2; price is question 3. A great business at a price that assumes perfection is a poor investment (see the Cisco case, G3).
• Treating a low P/E as "cheap". A low multiple is the market's forecast of low or risky future cash flows. Your job is to show that forecast is wrong, not to notice that the number is small.
• Using price moves to validate or reject a thesis within months. With an IR of 0.4, a year of underperformance is close to a coin toss. Watch the business KPIs instead.
• Learning from survivors only: studying 100-baggers without studying the companies that looked the same and failed.
• Mistaking information for edge. If everyone has read it, it is in the price. If almost no one legally could have it, it may be UPSI, and trading on it is illegal.
• Bringing a trader's clock to an investor's game: over-trading, reacting to daily P&L, and hedging away the exposure you are paid to hold.
• Ignoring costs and taxes, which compound as surely as returns.

Key terms

Term	Meaning
Price	The last traded price of a share; set by whoever is trading, for any reason
Market capitalisation	Share price × shares outstanding; the market price of the company's equity
Intrinsic value	The present value of all future cash the business will distribute to its owners; an estimate, never observed
Present value (PV)	A future amount shrunk by the required return for each year of waiting: \(CF_t/(1+r)^t\)
Discount rate / required return	The annual return an investor demands for the risk and the wait; turns future cash into present value
Gordon growth formula	Value of a cash flow growing at \(g\) for ever, discounted at \(r\): \(CF_1/(r-g)\), valid only if \(g<r\)
Residual claimant	Equity holders: paid last, after lenders, suppliers, employees and taxes, but own all the upside
Free cash flow (FCF)	Cash from operations minus capital expenditure; cash the business could distribute without shrinking
P/E and earnings yield	Price ÷ earnings per share; its inverse (earnings ÷ price) is the earnings yield
ROIC / WACC	Return on invested capital (after-tax operating profit ÷ capital employed in the business) / weighted average cost of capital (the blended return lenders and shareholders require). Growth creates value only if ROIC > WACC
Moat	A durable competitive advantage that protects returns on capital from competition
Margin of safety	The gap between estimated value and price, big enough to absorb estimation errors
Net-net	Graham's test: a stock priced below current assets minus all liabilities
Value trap	A stock that looks cheap on multiples but stays cheap or falls because the business is deteriorating
GARP / PEG	Growth at a reasonable price; PEG = P/E ÷ expected earnings growth (in %)
Special situation	An investment driven by a corporate event (demerger, buyback, open offer, delisting, merger)
Promoter	India: the founder/controlling shareholder group of a listed company
Efficient market hypothesis (EMH)	The claim that prices reflect available information (weak, semi-strong, strong forms)
Grossman–Stiglitz paradox	Perfectly efficient prices are impossible, because nobody would then be paid to gather information
UPSI	Unpublished price-sensitive information; trading on it is insider trading under SEBI rules
Catalyst / thesis-breaker	The event expected to close the gap between price and value / the evidence that would prove the thesis wrong
Alpha / tracking error / information ratio	Excess return over the benchmark / its standard deviation / their ratio, the standard measure of active skill
Total Return Index (TRI)	An index version that assumes dividends are reinvested; the fair benchmark for performance
LTCG / STCG	Long-/short-term capital gains; for listed Indian equity, the long-term threshold is 12 months
Survivorship bias	Drawing conclusions from survivors only, ignoring the failures that dropped out of the sample
Multibagger	A stock that rises many-fold (a "10-bagger" rises 10x)
Positive skewness	A distribution with a long right tail; for stocks, the mean is far above the median
Base rate	How often an outcome happens in the relevant reference class; the starting point before case-specific evidence

Check your understanding

1. A business will generate ₹20 Cr of owner cash next year, growing 6% a year for ever. You require 13%. (a) What is it worth? (b) The market values it at ₹400 Cr. What growth rate is the market implying at your 13% required return?

Answer

(a) $V = 20/(0.13 - 0.06) = 20/0.07 = ₹285.7$ Cr. (b) Rearranging $P = CF_1/(r-g)$: $g = r - CF_1/P = 13\% - 20/400 = 13\% - 5\% = 8\%$. The market expects 8% growth for ever. You disagree only if you can explain why growth will be closer to 6%, which is the price-implied-expectation thinking of the Trader's lens.

1. Classify each proposed edge as behavioural, structural, informational, analytical, or not an edge: (a) a small cap is being removed from a major index and passive funds must sell on the effective date; (b) in the notes to accounts you notice receivables grew 33% a year for three years while revenue grew 15%; (c) a friend in the company's finance team tells you the quarter was terrible, before results are published; (d) after three weak quarters, a cement stock trades as if today's depressed prices will last for ever.

Answer

(a) **Structural**: forced, price-insensitive sellers. (b) **Informational/analytical**: public but little-read data, plus the interpretation that growth is not turning into cash (exactly Kaveri's FY23–FY26 pattern: receivables compounding at about 33% a year against revenue at 14.7%). (c) **Not an edge.** It is UPSI, and trading on it is illegal insider trading. (d) **Behavioural**: extrapolation of a cyclical trough. The test is whether the base rate of cyclical recovery differs from what the price implies.

1. A manager's alpha is 4% a year with a tracking error of 10%. (a) How many years of data are needed for a t-stat of 2? (b) What is the probability she underperforms the index in a given year, assuming normal excess returns?

Answer

$IR = 0.04/0.10 = 0.4$. (a) $T = (2/0.4)^2 = 25$ years. (b) $\Phi(-0.4) = 34.5\%$. So about one year in three, even though she is genuinely skilled. That is why process, not short-run results, is the right thing to judge.

1. In the lognormal model of Worked example 5, the average stock grows 5.5x over 15 years, yet more than half the stocks lose money. Explain in two sentences why both are true, and what it implies for a 5-stock portfolio.

Answer

Compound returns are positively skewed. A few stocks multiply 50–300x and drag the mean far above the median, while the typical stock's compounded result is pulled down by volatility (the $-\sigma^2/2$ term makes the median log return negative here). A 5-stock portfolio will *usually* miss the rare big winners and lag the average, though occasionally it catches one and wins big. Concentration raises the variance of outcomes and needs genuine skill to justify it.

1. Kaveri's FY26 ROIC was 12.0% and the reference WACC is 12.19%. A friend says: "Management guides 15–18% growth, so the stock deserves a high multiple." What is wrong with the argument?

Answer

Growth creates value only when the return on new capital exceeds the cost of capital. At ROIC ≈ WACC, each rupee reinvested to grow earns about what it costs, so faster growth adds almost no value. Kaveri's growth also consumes working capital (receivables), which pushes returns on *incremental* capital lower still. The multiple should depend on growth *and* ROIC together, which Lesson 06.1 formalises.

1. A fund's marketing says: "Our five best picks since 2016 compounded at 40% a year." List at least three biases or missing facts.

Answer

(i) **Survivorship/selection**: five best out of how many? What did the whole portfolio return? (ii) No benchmark: 40% against what TRI over the same period? (iii) Position sizes: were these large positions or 0.5% holdings? (iv) Start and end dates chosen after the fact. (v) Funds that were merged or closed are left out; SPIVA found 27% of Indian funds across categories did not survive the decade to 2025. (vi) Risk: what drawdowns came with the return?

1. Which school would be most interested in each: (a) a company trading below the cash on its balance sheet, net of all liabilities; (b) a consumer brand with 40% ROCE and a long runway, at a fair multiple; (c) a conglomerate announcing the demerger of a subsidiary that index funds won't want; (d) a promoter-controlled company paying large royalties to a promoter entity, where institutions own 25%?

Answer

(a) **Deep value** (a net-net or near net-net). (b) **Quality/compounders**. (c) **Special situations** (demerger, forced selling of the spun-off entity). (d) **Activism / governance engagement**: realistically in India, voting against related-party resolutions and engaging through proxy advisers rather than a control fight.

1. Suppose you earn 15% a year pre-tax for 15 years. Compare buy-and-hold (pay 12.5% LTCG once at the end) with churning every year at 20% STCG. Ignore the exemption, surcharge and cess.

Answer

Pre-tax multiple $1.15^{15} = 8.14$x. Buy-and-hold: $1 + (8.14 - 1)(1 - 0.125) = 7.24$x. Churn: $(1 + 0.15 \times 0.80)^{15} = 1.12^{15} = 5.47$x. Holding ends about 32% richer (7.24 ÷ 5.47 = 1.32) with identical stock-picking. That is a structural edge from patience alone.

Go deeper

• Benjamin Graham, The Intelligent Investor (revised edition with Jason Zweig's commentary, 2003). The chapters on "Mr. Market" and "margin of safety" are the philosophical base of this course.
• Warren Buffett's shareholder letters (berkshirehathaway.com/letters), especially 1989 ("Mistakes of the First Twenty-five Years") on moving from cheap to wonderful businesses.
• Michael Mauboussin, The Success Equation (2012). Separating skill from luck, and why sample size matters so much in investing.
• Bessembinder, Chen, Choi & Wei, "Long-Term Shareholder Returns: Evidence from 64,000 Global Stocks", Financial Analysts Journal (2023), SSRN 3710251. Read the country tables, including India.
• Grossman & Stiglitz (1980), "On the Impossibility of Informationally Efficient Markets" (AEA). A short paper that explains why an edge must exist and why it must be costly.
• S&P Dow Jones Indices, SPIVA India scorecards, published twice a year. The standing evidence on how hard beating the index is.

---

← Previous: Syllabus · Module index · Next: How to use this course →