Chapter 2 · Indian Markets and Instruments

2.1 How the Market Works: Exchanges, Indices and Corporate Actions

What actually happens when you press Buy: the order book, bid–ask spread, market and limit orders and T+1 settlement; how a free-float market-cap index such as the Nifty 50 is built and kept continuous with a divisor; price vs total-return indices; and how splits, bonuses, dividends and rights issues change prices and holdings.

📍 The situation

Kavya, 24, has just opened a trading account and a demat account. On her broker's app the hypothetical Alpha Bank Ltd shows two prices — "Bid ₹812.20" and "Ask ₹812.40" — and a box of numbers called the market depth. She wants 1,500 shares and wonders whether to press "Market" or "Limit". The evening news says "the Nifty fell 1.2% today, dragged down by banks", although most of the stocks she follows rose. A week later her holding in the hypothetical Delta Motors announces a 1:1 bonus; on the ex-date the price on her screen halves and her app's "day's change" shows −50%. She panics. And a friend tells her that an index fund "should earn what the Nifty earns" — but which Nifty: the one quoted on the news, or something called the "TRI"?

This section explains the machinery behind those screens. How does an order find a counterparty, and what does it really cost? How is an index number computed from 50 share prices, and why does it not jump when a company splits its shares or leaves the index? What is a share worth after a bonus, dividend or rights issue? The mathematics is light — weighted averages, ratios and one conservation idea, "value does not appear from nowhere" — but using it correctly separates people who read market data from people who are misled by it.

⚠️ About the numbers on this page

All companies (Alpha Bank, Beta Infotech, Delta Motors…) and all prices, order books, index levels and corporate actions on this page are hypothetical, chosen to make the arithmetic realistic. Tick sizes, trading hours, settlement cycles, index methodology, market-cap category cut-offs, taxes and charges are set by SEBI, the exchanges and index providers and can change — check the current NSE/BSE circulars, NSE Indices methodology documents and SEBI/AMFI notices. Nothing here is a recommendation to buy or sell anything.

Exchanges, the Order Book and How a Trade Happens

India's two main stock exchanges are the National Stock Exchange (NSE) and BSE. Both are fully electronic: there is no trading floor, just a computer that keeps, for every share, a list of everyone's outstanding buy and sell instructions and matches them automatically. You cannot send an order to the exchange yourself; you send it through a SEBI-registered broker. Once a trade is matched, the exchange's clearing corporation becomes the buyer to every seller and the seller to every buyer, so neither side has to trust the other, and the shares move into and out of your demat account, held with one of the two depositories (NSDL and CDSL).

Settlement: T+1. Indian equity trades settle on a T+1 cycle: if you buy on trading day $T$, the money leaves your account and the shares reach your demat account on the next working day; if you sell on day $T$, the proceeds arrive on day $T+1$. A trade on Friday settles on Monday (if Monday is a working day). Settlement rules, holidays and timings are set by SEBI and the exchanges — check the current calendar.

📘 Definition 2.1 — Order book, bid, ask and spread

The order book of a share is the list of all unexecuted orders, arranged by price. Buy orders are bids; sell orders are asks (or offers). The highest bid is the best bid $b$ and the lowest ask is the best ask $a$; in a normal book $b\lt a$. Then

  • the bid–ask spread is $a-b$, and the mid price is $m=\tfrac12(a+b)$;
  • the relative spread is $(a-b)/m$, usually quoted in basis points (1 bp $=0.01\%$);
  • the quantities waiting at each price are the market depth. Prices move in steps of the tick size set by the exchange (we use an illustrative ₹0.05).

The last traded price (LTP) shown on a screen is simply the price of the most recent trade; it is not a price at which you are guaranteed to trade now.

📘 Definition 2.2 — Market order, limit order and price–time priority
  • A market order says "buy (or sell) $Q$ shares now at the best available prices". A market buy takes the asks from the cheapest upwards until $Q$ shares are filled; it is certain to execute (if the book is deep enough) but its price is not known in advance.
  • A limit order says "buy $Q$ shares at a price of at most $L$" (or "sell at no less than $L$"). Any part that cannot be matched at $L$ or better waits in the book. Its price is controlled, but it may fill only partly, or never.
  • Price–time priority: an incoming order is matched first against the best price on the other side; among orders at the same price, the one that arrived first is filled first.

(A stop-loss order is a conditional order that becomes a market or limit order once a trigger price is touched; we meet it in risk management, §7.2.)

📐 Theorem 2.1 (Formula) — The real price of a market order

Suppose a market buy of $Q$ shares takes $q_1,q_2,\dots,q_k$ shares at ask prices $p_1\lt p_2\lt\dots\lt p_k$, with $q_1+\dots+q_k=Q$. Then the average fill price is the quantity-weighted mean $$\bar p=\frac{\sum_j p_jq_j}{\sum_j q_j},$$ and the execution cost measured against the mid price is $(\bar p-m)\,Q$ rupees, or $(\bar p-m)/m$ as a fraction. For an order small enough to fill at the best price, $\bar p=a$ and the cost is $\tfrac12(a-b)$ per share — half the spread; buying and immediately selling such an order costs the whole spread, $(a-b)Q$ (before brokerage and charges).

Why the mid price? If you could trade at no cost, the fair reference price is halfway between what buyers bid and what sellers ask. A buyer who pays the ask gives up $a-m$ per share to the seller who was waiting in the book; that is the price of immediacy. A large market order is worse still, because after the best ask is exhausted it continues to the next, higher levels: $\bar p$ is a weighted average of increasingly bad prices, so it rises with $Q$. This is the simplest model of market impact, and it is why traders split large orders and why illiquid shares are expensive to trade even when brokerage is zero. ∎

✏️ Example 1 — Kavya's 1,500 shares at market
The book of Alpha Bank shows asks ₹812.40 × 300, ₹812.45 × 500, ₹812.60 × 1,200, ₹812.90 × 2,000 and bids ₹812.20 × 400, ₹812.10 × 700, ₹812.00 × 1,500, ₹811.80 × 900. Find the spread, the mid price and the relative spread. Kavya sends a market buy for 1,500 shares: which orders fill it, at what average price, and what does the order cost her relative to the mid price?
Spread $a=812.40$, $b=812.20$: spread ₹0.20, mid $m=\text{₹}\,812.30$, relative spread $0.20/812.30\approx0.0246\%\approx2.5$ bp.
Walk the book 300 at ₹812.40, then 500 at ₹812.45, then the remaining 700 from the 1,200 at ₹812.60. Total paid $$243{,}720+406{,}225+568{,}820=12{,}18{,}765.$$
Average price $\bar p=12{,}18{,}765/1{,}500=\text{₹}\,812.51$ per share — ₹0.11 above the best ask and ₹0.21 above the mid.
Cost $0.21\times1{,}500=\text{₹}\,315$, i.e. $0.21/812.30\approx0.026\%$ of the trade value of about ₹12.2 lakh. Small here, because Alpha Bank's book is deep; in a thinly traded share the same order could walk through many levels.
✏️ Example 2 — A limit order waits its turn
Instead, Kavya places a limit buy for 600 shares at ₹812.10. (a) Does it execute immediately? (b) Where does it sit in the book? (c) A few minutes later someone sends a market sell for 1,500 shares, and no other orders have arrived. How many of Kavya's shares are bought, and at what price?
(a) No: the cheapest seller asks ₹812.40, above her limit of ₹812.10, so nothing can be matched.
(b) It joins the bids at ₹812.10, behind the 700 shares already waiting there (time priority). The bid side now reads ₹812.20 × 400, ₹812.10 × 1,300 (700 + her 600), …
(c) The market sell of 1,500 takes 400 at ₹812.20, then 700 at ₹812.10 from the earlier orders, and the remaining $1{,}500-400-700=400$ from Kavya's order, at ₹812.10.
Result 400 of her 600 shares are bought at exactly ₹812.10 (₹0.41 per share cheaper than her market order's average); 200 still wait. A limit order trades price certainty for execution uncertainty.
Interactive 1 — Inside the Order Book: How an Order Walks the Book

A hypothetical order book: red bars are sellers' asks, blue bars buyers' bids, longer bars mean more shares waiting. Choose buy or sell, market or limit, and an order size. The amber parts are the shares your order takes; the dashed amber line marks your average price. With a limit order, any unfilled remainder joins the book as the purple bar. Press Play to grow the order and watch the average price climb.

1,500—
best bid / best ask—
spread · mid—
shares filled—
average price—
cost vs mid—
left over—

Market Capitalisation, Free Float and Building an Index

"The Nifty fell 1.2%" summarises the prices of 50 companies in one number. How the 50 prices are combined — which stocks count for more — decides what the number means. Most important indices in India and abroad weight each company by the value of its shares that are actually available to investors.

📘 Definition 2.3 — Market capitalisation, free float and size categories

For a company with $Q$ shares outstanding and share price $P$:

  • its market capitalisation (market cap) is $PQ$ — the market value of all its shares;
  • its free float is the part of its shares available for trading by the public, i.e. excluding holdings such as those of promoters, the government as promoter, and other strategic or locked-in holders; the free-float factor or investable weight factor (IWF) $f\in(0,1]$ is the free-float fraction;
  • its free-float market cap is $PQf$.

Size categories. For mutual-fund purposes SEBI currently classifies listed companies by full market-cap rank: the 1st–100th companies are large cap, the 101st–250th mid cap, and the 251st onwards small cap; AMFI publishes the list periodically. These are ranks, not rupee thresholds, so a company's category can change as prices move — check the current SEBI rules and AMFI list.

📘 Definition 2.4 — Free-float market-cap index, base value and divisor

A free-float market-cap weighted index of $n$ stocks has level $$I_t=\frac{M_t}{D_t},\qquad M_t=\sum_{i=1}^{n}P_{i,t}\,Q_i\,f_i,$$ where $M_t$ is the total free-float market cap of the constituents. On the base date the index is set to a round base value $B$ (such as 100 or 1,000) by choosing the divisor $D_0=M_0/B$. The divisor is then kept fixed, except when the index basket changes for reasons that are not market movements (Theorem 2.3).

The Nifty 50 (50 stocks) and the Sensex (30 stocks) are free-float market-cap weighted indices of this kind; the index providers' methodology documents also set eligibility, rebalancing dates and limits on individual weights.

📐 Theorem 2.2 (Formula) — Weights and the daily index return

In a free-float market-cap index the weight of stock $i$ at time $t$ is $$w_{i,t}=\frac{P_{i,t}Q_if_i}{M_t},\qquad\sum_i w_{i,t}=1.$$ If the divisor does not change between $t-1$ and $t$ and stock $i$'s price return is $r_i=P_{i,t}/P_{i,t-1}-1$, then the index return is the weighted average of the constituents' returns, using the previous day's weights: $$\frac{I_t}{I_{t-1}}-1=\sum_{i=1}^{n}w_{i,t-1}\,r_i .$$

Derivation. With the same divisor, $I_t/I_{t-1}=M_t/M_{t-1}$, and $P_{i,t}=P_{i,t-1}(1+r_i)$, so $$\frac{M_t}{M_{t-1}}=\frac{\sum_iP_{i,t-1}Q_if_i\,(1+r_i)}{M_{t-1}}$$ $$=\sum_i w_{i,t-1}(1+r_i)=1+\sum_iw_{i,t-1}r_i,$$ because the weights add up to 1. ∎ So a 2% move in a stock with weight 30% moves the index by 0.6%, while a 10% move in a stock with weight 1% moves it by only 0.1%. That is how "the Nifty fell although most stocks rose" can happen: a few heavyweights can outweigh many small risers. It also tells an index fund what to hold: holding each stock in proportion to its weight makes the portfolio's return equal to the index's.

✏️ Example 3 — Building the "Mini-5" index
Build a free-float market-cap index of five hypothetical stocks with base value 1,000. Shares $Q$ are in crore, so market caps come out in ₹ crore.
stockprice $P$ (₹)$Q$ (crore)IWF $f$
Alpha Bank1,600500.80
Beta Infotech3,200200.50
Gamma Power2501200.40
Delta Motors900400.60
Epsilon FMCG2,500120.35
(a) Find each full and free-float market cap, the weights and the divisor. (b) Next day the prices move by $+2\%$, $-1.5\%$, $+4\%$, $-3\%$ and $+1\%$ respectively. Find the new index level directly and check it with Theorem 2.2.
Market caps Full: 80,000; 64,000; 30,000; 36,000; 30,000 (total ₹2,40,000 crore). Free-float ($\times f$): 64,000; 32,000; 12,000; 21,600; 10,500 — total $M_0=\text{₹}\,1{,}40{,}100$ crore.
Weights $w=M_i/M_0$: Alpha 45.68%, Beta 22.84%, Gamma 8.57%, Delta 15.42%, Epsilon 7.49%. (By full market cap Gamma and Epsilon would each have 12.5%; their low free float cuts their influence.)
Divisor $D_0=M_0/1{,}000=140.1$ (₹ crore per index point), so $I_0=1{,}000$.
Next day New free-float caps: $65{,}280+31{,}520+12{,}480+20{,}952+10{,}605=1{,}40{,}837$, so $I_1=140{,}837/140.1\approx1{,}005.26$.
Check $\sum w_ir_i=0.4568(0.02)+0.2284(-0.015)+0.0857(0.04)+0.1542(-0.03)+0.0749(0.01)\approx0.00526$ — a 0.526% rise, as found.
✏️ Example 4 — Same prices, three different "markets"
Using the prices and moves of Example 3, compute the day's return of (a) a price-weighted index $\sum P_i/D$ (each stock counts in proportion to its share price), (b) an equal-weighted index (each stock 20%), (c) a full market-cap index. Compare with the free-float index. Then suppose Beta Infotech splits each share into two at the end of the day: what happens to each index if nothing is adjusted?
(a) Price weights are $P_i/\sum P=P_i/8{,}450$. Return $$\frac{32-48+10-27+25}{8{,}450}=\frac{-8}{8{,}450}\approx-0.095\%.$$ Beta, the highest-priced share, dominates.
(b), (c) Equal weights: $(2-1.5+4-3+1)/5=+0.50\%$. Full market cap: weights $\tfrac13,\tfrac4{15},\tfrac18,\tfrac{3}{20},\tfrac18$, return $\approx+0.442\%$. Free float: $+0.526\%$.
Lesson Same day, same prices: the "market" fell 0.09%, or rose 0.44%, 0.50% or 0.53%, depending only on the weighting. An index is a definition, and the headline number is meaningful only if you know it.
The split Beta's price halves ($3{,}152\to1{,}576$) and its share count doubles, so its market cap is unchanged: the two market-cap indices do not move. The price-weighted sum falls from 8,442 to 6,866, so with an unchanged divisor that index would drop by about 18.7% although nobody lost a rupee. Its divisor must be cut in the same ratio (Theorem 2.3). A share price by itself says nothing about a company's size — which is the main reason serious indices are cap-weighted.
📐 Theorem 2.3 (Formula) — Divisor adjustment keeps the index continuous

Suppose that at the close of day $t$ the index basket changes for a non-market reason — a constituent is replaced, a company issues or buys back shares, or a free-float factor is revised — so that the total free-float market cap at unchanged prices changes from $M$ to $M'$. Setting the new divisor $$D'=D\cdot\frac{M'}{M}$$ leaves the index level unchanged, $M'/D'=M/D$, and from the next day the index return is again the weighted average of constituent returns (Theorem 2.2) with the new weights. Splits and bonus issues do not change $M$ (price falls in the same ratio as the share count rises), so a market-cap index needs no adjustment for them; a price-weighted index does.

Why. $M'/D'=M'\cdot M/(D\,M')=M/D$. The rule makes the index measure only price changes of the basket: money that enters (new shares paid for in cash) or a stock that is swapped in should not register as a market move. Without the adjustment, an index would jump up every time a large company replaced a small one. ∎

✏️ Example 5 — Epsilon leaves, Zeta joins
At the close of the day in Example 3(b) ($M=1{,}40{,}837$, $D=140.1$, $I=1{,}005.26$), Epsilon FMCG (free-float cap ₹10,605 crore at its closing price ₹2,525) is replaced by the hypothetical Zeta Pharma: price ₹1,200, 30 crore shares, IWF 0.45. (a) Find the new divisor. (b) Next day only Zeta moves, by +5%. Find the index level. (c) What level would have been published on the change day if the divisor had not been adjusted?
(a) Zeta's free-float cap is $1{,}200\times30\times0.45=16{,}200$. At unchanged prices $M'=1{,}40{,}837-10{,}605+16{,}200=1{,}46{,}432$, so $$D'=140.1\times\frac{146{,}432}{140{,}837}\approx145.666.$$ Check: $146{,}432/145.666\approx1{,}005.26$ — unchanged.
(b) Zeta's cap rises by $16{,}200\times0.05=810$: $I=147{,}242/145.666\approx1{,}010.82$, a rise of $5.56$ points. Zeta's weight is $16{,}200/146{,}432\approx11.06\%$, and $11.06\%\times5\%\approx0.553\%$ of 1,005.26 is indeed about 5.56 points.
(c) $146{,}432/140.1\approx1{,}045.20$ — a fake "rally" of almost 4% caused purely by swapping a smaller company for a larger one.
Interactive 2 — Build an Index: Weights, Divisor and Corporate Actions

The five Mini-5 stocks follow simulated price paths for 120 trading days (not real data). Choose how the index weights them and which events happen: on day 40 Beta Infotech splits each share into two; on day 80 Epsilon is replaced by Zeta Pharma. The coloured strip at the top shows the current weights. Switch the divisor to "not adjusted" to see the jumps a careless index would publish — the dashed line is the correct index. Press Play to run the 120 days.

day—
index level—
divisor—
largest weight—
change since day 0—

Price Index and Total Return Index

When Alpha Bank pays a dividend of ₹16 a share, its price falls by about ₹16 on the ex-date — the cash has left the company and gone to the shareholders. A price index registers that fall as a loss, although shareholders are no worse off. Over decades the dividends are a large part of what shareholders earn, so an index that ignores them understates the return of actually owning the stocks.

📘 Definition 2.5 — Price return index and total return index

The index $I_t=M_t/D_t$ defined so far is a price return index (PRI): it follows prices only. A total return index (TRI) also assumes that every dividend is reinvested in the index on its ex-date. If constituents worth $d_t$ index points of dividend go ex on day $t$ (a dividend of ₹$\delta$ per share on stock $i$ is worth $\delta Q_if_i/D_t$ points), the TRI moves by the factor $(I_t+d_t)/I_{t-1}$ instead of $I_t/I_{t-1}$. Indian index providers publish TRI versions of their indices, and mutual-fund performance is compared with the TRI of the benchmark.

📐 Theorem 2.4 (Formula) — The total return index

Starting from $\text{TRI}_0=I_0$, $$\text{TRI}_t=\text{TRI}_{t-1}\cdot\frac{I_t+d_t}{I_{t-1}}.$$ Over a year in which the price index rises by $g$ and dividends equal to a fraction $y$ (the dividend yield) of the year-end level are reinvested, the total return is $(1+g)(1+y)-1\approx g+y$, and after $T$ such years $$\frac{\text{TRI}_T}{\text{PRI}_T}\cdot\frac{\text{PRI}_0}{\text{TRI}_0}=(1+y)^T.$$ The gap between the two indices compounds.

Why. An investor holding the index portfolio worth $I_{t-1}$ ends day $t$ with shares worth $I_t$ plus cash worth $d_t$; reinvesting the cash, her wealth grows by $(I_t+d_t)/I_{t-1}$, and chaining these daily factors gives the TRI. Over a year, wealth $(1+g)$ per rupee plus dividends $y(1+g)$ reinvested gives $(1+g)(1+y)$; the ratio of the two indices therefore picks up a factor $1+y$ every year. ∎

✏️ Example 6 — The dividend that the price index forgets
The Mini-5 price index (divisor 140.1) closes at 1,000.00, 1,004.00, 998.50 and 1,002.00 on four days. On day 2 Alpha Bank (50 crore shares, IWF 0.80) goes ex-dividend with ₹16 a share. (a) How many index points is the dividend worth? (b) Compute the TRI on days 1–3 and the 3-day returns of both indices. (c) Over 20 years, with the price index rising an illustrative 9% a year and a dividend yield of 1.3%, how much larger is the TRI's growth?
(a) Dividend paid on free-float shares: $16\times50\times0.80=\text{₹}\,640$ crore, i.e. $640/140.1\approx4.568$ points.
(b) $\text{TRI}_1=1{,}000\times\frac{1{,}004}{1{,}000}=1{,}004.00$; $\ \text{TRI}_2=1{,}004\times\frac{998.50+4.568}{1{,}004}\approx1{,}003.07$; $\ \text{TRI}_3=1{,}003.07\times\frac{1{,}002}{998.50}\approx1{,}006.58$. Over three days the PRI rose 0.20%, the TRI 0.66%.
(c) PRI: $1.09^{20}\approx5.60$; TRI: $(1.09\times1.013)^{20}\approx7.26$. The ratio is $1.013^{20}\approx1.295$: the TRI ends almost 30% higher. A 1.3% yield sounds small, but compounding makes it a big share of long-run wealth.
Moral An index fund's NAV includes the dividends it receives, so comparing it with the price index flatters the fund; the fair benchmark is the TRI (see tracking difference in §2.2).

Corporate Actions: Splits, Bonuses, Dividends and Rights

A corporate action is an event decided by the company that changes the number of shares, the price per share or both. The single principle behind all the formulas is conservation of value: at the instant of the action, the total value of a shareholder's position (shares plus any cash received, minus any cash paid in) is unchanged. Real prices then move with the market, but the theoretical ex-price is fixed by this bookkeeping.

📘 Definition 2.6 — Split, bonus, dividend, rights issue; record date and ex-date
  • Stock split: each share is divided into $k$ shares (for example, face value ₹10 split into five shares of ₹2, $k=5$).
  • Bonus issue $a{:}b$: $a$ free new shares for every $b$ held (a "1:1 bonus" doubles the share count; a "1:2 bonus" gives one new share for every two held), paid for by moving reserves into share capital.
  • Dividend: a cash payment of ₹$\delta$ per share.
  • Rights issue $a{:}b$ at price $S$: existing shareholders may buy $a$ new shares for every $b$ held at the subscription price $S$, usually below the market price. In India the rights are credited to the demat account as rights entitlements, which can be traded on the exchange or allowed to lapse.

To receive the benefit you must be a shareholder on the record date. The ex-date is the first trading day on which a buyer no longer gets it; under T+1 settlement the ex-date is normally the same day as the record date, so you must buy before it. On the ex-date the exchange adjusts the reference price.

📐 Theorem 2.5 (Formula) — Theoretical ex-prices and adjustment factors

Let $P$ be the cum (pre-event) price and $N$ the shares held. By conservation of value:

  1. Split into $k$: $N'=kN$, $P'=P/k$.
  2. Bonus $a{:}b$: $N'=N\cdot\dfrac{a+b}{b}$, $\ P'=P\cdot\dfrac{b}{a+b}$.
  3. Dividend $\delta$: $P'=P-\delta$ (before tax).
  4. Rights $a{:}b$ at $S$: the theoretical ex-rights price is $$\text{TERP}=\frac{bP+aS}{a+b},$$ and the theoretical value of the right to buy one new share is $\text{TERP}-S$.

The adjustment factor is $\phi=P'/P$. To make a price history comparable across the event, multiply every price before the ex-date by $\phi$ (and divide old share counts by $\phi$); returns computed from adjusted prices are then correct.

Derivation of TERP. Take $b$ old shares worth $bP$. Subscribing to $a$ new shares adds $aS$ in cash, giving $a+b$ shares. If no value is created or destroyed at that instant, each share is worth $(bP+aS)/(a+b)$. A right lets you buy at $S$ a share worth TERP, so it is worth $\text{TERP}-S$. The other three cases are the same argument: a split or bonus changes the number of slices but not the size of the cake, and a dividend moves $\delta$ per share from the company's cash to your bank account. ∎

✏️ Example 7 — Kavya's Delta Motors after a bonus and a split
Kavya bought 40 Delta Motors shares at an average ₹2,100. Delta announces a 1:1 bonus; the cum price is ₹2,400. Months later it splits each share into five (face value ₹10 to ₹2) when the price is ₹1,300. Today the share trades at ₹275. (a) Track her holding and its theoretical value through both events. (b) Her app shows her cost as ₹2,100 and today's price as ₹275. What is her true gain?
Bonus $N'=40\times\frac{1+1}{1}=80$ shares, $P'=2{,}400\times\tfrac12=\text{₹}\,1{,}200$. Value before and after: $40\times2{,}400=80\times1{,}200=\text{₹}\,96{,}000$. The "−50%" on her app on the ex-date is just the price per share halving.
Split $N''=80\times5=400$ shares, $P''=1{,}300/5=\text{₹}\,260$; value $\text{₹}\,1{,}04{,}000$ before and after.
Adjusted cost Combined factor $\phi=\tfrac12\times\tfrac15=\tfrac1{10}$: her cost per current share is $2{,}100\times\tfrac1{10}=\text{₹}\,210$. Total cost $40\times2{,}100=400\times210=\text{₹}\,84{,}000$.
Gain Value today $400\times275=\text{₹}\,1{,}10{,}000$: a gain of ₹26,000, or $275/210-1\approx+30.95\%$. Comparing 275 with 2,100 ("−86.9%") mixes prices of different-sized shares.
✏️ Example 8 — A rights issue: subscribe, sell or ignore?
The hypothetical Theta Telecom announces a 2:7 rights issue at ₹150 when the share trades at ₹220. A shareholder owns 700 shares. (a) Find the TERP and the theoretical value of each rights entitlement (RE). (b) Compare her wealth if she subscribes, if she sells her REs at their theoretical value, and if she lets them lapse.
(a) $$\text{TERP}=\frac{7\times220+2\times150}{9}=\frac{1{,}840}{9}\approx\text{₹}\,204.44.$$ Each RE is worth $204.44-150\approx\text{₹}\,54.44$. She receives $700\times\tfrac27=200$ REs.
Subscribe She pays $200\times150=\text{₹}\,30{,}000$ and holds 900 shares worth $900\times204.44=\text{₹}\,1{,}84{,}000$ — exactly her ₹1,54,000 before plus the ₹30,000 paid in.
Sell the REs 700 shares worth $700\times204.44\approx\text{₹}\,1{,}43{,}111$ plus RE sale proceeds $200\times54.44\approx\text{₹}\,10{,}889$: total ₹1,54,000, unchanged.
Lapse Only ₹1,43,111: she loses ₹10,889 (7.1%). A rights issue below the market price is not a gift; it is an option that must be used or sold. The adjustment factor is $\phi=204.44/220\approx0.9293$.
⚠️ Common mistakes and myths
  1. "A ₹3,000 share is expensive and a ₹30 share is cheap." Price per share says nothing about value; a company's size is its market cap, and valuation needs earnings and cash flows (Chapter 5). A split makes a share cheaper to buy in small lots, not better value.
  2. "My stock fell 50% on the bonus date." After a 1:1 bonus you hold twice as many shares at half the price (Example 7). Use adjusted prices for returns.
  3. "A market order gets the price on the screen." It gets the best available asks or bids, walking the book if necessary (Example 1); the last traded price is history.
  4. "A limit order always executes." Only if the market reaches the limit — and even then perhaps only partly, behind earlier orders (Example 2).
  5. "The Nifty rose 1%, so my stocks rose 1%." The index return is a weighted average dominated by heavyweights; a portfolio with different weights earns something different (Example 4).
  6. "The price index shows what an index investor earned." It leaves out dividends; the TRI includes them, and the gap compounds (Example 6).
  7. "Ex-dividend, the price fell, so I lost the dividend." The price falls by about the dividend you received; before tax your wealth is unchanged. Dividends are taxable income (check current rules), so the after-tax effect can be negative.
  8. "Market cap is the cash a company has." It is what the market values all its shares at — an opinion priced every second, not money in a bank.
  9. "Rights at a discount are free money." The share price falls to TERP; a shareholder who neither subscribes nor sells the entitlements loses (Example 8).
  10. "An index jumps when a big company joins." Not a properly maintained one: the divisor is adjusted so that changes in the basket do not show up as market moves (Example 5).
📈 Market case — Running a ₹500 crore Mini-5 index fund

A hypothetical index fund tracks the Mini-5 index of Example 3 with ₹500 crore of investors' money. Follow the fund manager through two days, a split and a constituent change, using the numbers above. (Trading costs are illustrative.)

  1. Day 0: invest by weight. She buys each stock in proportion to its weight: Alpha 45.682% → ₹228.41 crore (about 14.28 lakh shares at ₹1,600), Beta ₹114.20 crore, Gamma ₹42.83 crore, Delta ₹77.09 crore, Epsilon ₹37.47 crore (about 1.50 lakh shares at ₹2,500).
  2. Day 1: the fund moves with the index. Each holding moves by its stock's return, so the fund grows by $\sum w_ir_i=0.526\%$ to ₹502.63 crore — exactly $500\times1{,}005.26/1{,}000$ (Theorem 2.2).
  3. A split needs no trade. If Beta splits 1:2, the fund simply holds twice as many Beta shares at half the price; its weight is unchanged, and so is the index (Theorem 2.3).
  4. The constituent change needs trades. Epsilon (weight 7.53% at the close) leaves and Zeta enters with weight $16{,}200/146{,}432=11.063\%$. The new weights of the others shrink: Alpha $46.351\%\to44.580\%$, Beta $22.380\%\to21.525\%$, Gamma $8.861\%\to8.523\%$, Delta $14.877\%\to14.308\%$. On a ₹502.63 crore fund she must sell all of Epsilon (₹37.85 crore) and trim Alpha (₹8.90 crore), Beta (₹4.30 crore), Gamma (₹1.70 crore) and Delta (₹2.86 crore), and buy Zeta for ₹55.61 crore (about 4.63 lakh shares at ₹1,200).
  5. What the change costs. Total trading is ₹55.61 crore of sales and ₹55.61 crore of purchases — ₹111.21 crore, 22% of the fund. If each rupee traded costs an illustrative 0.05% (half the spread plus charges), the cost is about ₹5.56 lakh, or 0.011% of the fund. The index pays nothing; the fund pays this.

What it shows. An index is a calculation; an index fund is a portfolio that has to trade in a real order book. Trading costs at every rebalance, the expense ratio, cash waiting to be invested and dividends that arrive later than the TRI assumes all make the fund's return differ slightly from the index's. Measuring that difference — tracking difference and tracking error — is the subject of §2.2.

Exercises

All companies, prices and order books are hypothetical. Market caps are in ₹ crore when shares are given in crore. Give prices to two decimals and index levels to two decimals.

I
Order book, spread and orders.
  • 1 A share's best bid is ₹455.20 and best ask ₹455.35. Find the spread, the mid price and the relative spread in basis points.
  • 2 The asks are ₹455.35 × 600, ₹455.40 × 900, ₹455.55 × 1,500. Find the average price of a market buy for 2,000 shares and its cost relative to the mid price (in ₹ and in bp).
  • 3 With the same book, a limit buy for 2,000 shares at ₹455.40 is sent instead. How many shares fill, at what average price, and what happens to the rest?
  • 4 A trader buys 1,000 shares at the ask and sells them a second later at the bid; the spread is ₹0.15 and nothing else changes. What is the loss before charges, and what is it as a percentage of the trade value at the mid price of part 1?
  • 5 Bids at ₹455.20: order X for 300 shares (arrived 9:20), order Y for 500 (9:25) and your order for 400 (9:31). There are no higher bids. A market sell for 1,000 shares arrives. How many of your shares are bought?
  • 6 Under T+1 settlement, you sell shares on a Thursday that is a working day. When do you receive the money? You buy on a Friday; when do the shares reach your demat account if the next Monday is an exchange holiday?
  • 7 Explain why a large market order in a thinly traded small-cap share can be costly, and suggest two ways to reduce that cost.
1 Spread ₹0.15; mid ₹455.275; relative spread $0.15/455.275\approx3.29$ bp.
2 600 at 455.35, 900 at 455.40, 500 at 455.55: total ₹9,10,845, average ₹455.4225. Cost vs mid: $0.1475\times2{,}000=\text{₹}\,295$, about 3.24 bp.
3 Only the asks at ₹455.35 and ₹455.40 qualify: 1,500 shares at an average $(600\times455.35+900\times455.40)/1{,}500=\text{₹}\,455.38$. The remaining 500 rest as a bid at ₹455.40, which becomes the new best bid.
4 Loss $=0.15\times1{,}000=\text{₹}\,150$, which is $0.15/455.275\approx0.033\%$ — the whole spread, paid for immediacy.
5 Time priority: X (300) and Y (500) fill first, leaving $1{,}000-800=200$ for you; 200 of your 400 shares are bought at ₹455.20 and 200 remain.
6 Sale on Thursday: funds on Friday ($T+1$). Purchase on Friday: the next working day is Tuesday, so the shares arrive on Tuesday.
7 The book is shallow and the spread wide, so the order walks through many worse price levels (Theorem 2.1) and moves the price against you. Use limit orders, split the order over time, or trade when the book is deeper.
II
Market cap, free float and index construction. A three-stock index has X: price ₹500, 10 crore shares, IWF 0.6; Y: ₹1,200, 5 crore, IWF 0.4; Z: ₹80, 50 crore, IWF 0.9.
  • 1 Find each stock's full and free-float market cap and its weight in a free-float index.
  • 2 With base value 100, find the divisor.
  • 3 Next day X rises 3%, Y falls 2% and Z rises 5%. Find the new index level, and check it with the weighted-return formula.
  • 4 A price-weighted index of the same stocks starts at 100. Find its level after the same moves.
  • 5 After the close of that day Y splits each share into four. What happens to the free-float index and its divisor? Find the new divisor of the price-weighted index, and the level it would show if the divisor were not changed.
  • 6 At that close Z is replaced by W (price ₹300, 8 crore shares, IWF 0.5). Find the free-float index's new divisor (use the closing prices of part 3).
  • 7 Under SEBI's current rank-based categorisation, a company ranked 180th by full market cap is in which category? Why does a free-float index give a promoter-dominated company less weight than its full market cap suggests?
1 Full: 5,000; 6,000; 4,000. Free float: 3,000; 2,400; 3,600 (total 9,000). Weights 33.33%, 26.67%, 40.00%.
2 $D=9{,}000/100=90$.
3 $3{,}090+2{,}352+3{,}780=9{,}222$; index $9{,}222/90\approx102.47$. Check: $\tfrac13(3\%)+\tfrac4{15}(-2\%)+0.4(5\%)\approx2.467\%$.
4 $\sum P=1{,}780$, divisor $17.8$; new prices $515+1{,}176+84=1{,}775$, level $1{,}775/17.8\approx99.72$ — a fall, because the falling Y has the highest price.
5 The free-float index and its divisor are unchanged (Y's market cap is the same). Price-weighted: the sum falls from 1,775 to $515+294+84=893$; new divisor $17.8\times893/1{,}775\approx8.955$, keeping the level at 99.72. Unadjusted, it would show $893/17.8\approx50.17$.
6 Z's free-float cap at its closing price ₹84 is $84\times50\times0.9=3{,}780$; W's is $300\times8\times0.5=1{,}200$. $M'=9{,}222-3{,}780+1{,}200=6{,}642$, $D'=90\times6{,}642/9{,}222\approx64.821$; the level stays 102.47.
7 Mid cap (ranks 101–250). Shares held by promoters rarely trade, so investors — and an index fund — can buy only the free-float part; weighting by free float reflects the investable size.
III
Price and total return indices.
  • 1 A price index closes at 2,000 two days running; on the second day constituents worth 3 index points of dividend go ex. By what percentage does the TRI change on that day?
  • 2 In a year the price index rises 8% and dividends equal to 1.5% of the year-end level are reinvested. Find the total return.
  • 3 Over 15 years the price index grows at 9% a year and the dividend yield is 1.2%. By what factor does the TRI end up above the PRI (both starting equal)?
  • 4 A large-cap fund returned 11.2% in a year in which its benchmark's PRI returned 10% and its TRI 11.5%. Did the fund beat its benchmark? Explain.
  • 5 In the Mini-5 index (divisor 140.1), Alpha Bank pays a dividend of ₹20 a share. How many points does the price index lose on the ex-date (all else unchanged)?
  • 6 Explain why a price index understates long-term returns more for a high-dividend sector index than for a low-dividend one.
1 $(2{,}000+3)/2{,}000=1.0015$: the TRI rises 0.15% while the PRI is flat.
2 $1.08\times1.015-1=9.62\%$ (approximately $8+1.5=9.5\%$).
3 $1.012^{15}\approx1.196$: the TRI ends almost 20% above the PRI ($1.09^{15}\approx3.64$ vs $(1.09\times1.012)^{15}\approx4.36$).
4 No. The fund's NAV includes the dividends it received, so the fair comparison is the TRI: $11.2\%\lt11.5\%$, an underperformance of 0.3 percentage points. Against the PRI it would wrongly look 1.2 points ahead.
5 $20\times50\times0.80/140.1=800/140.1\approx5.71$ points.
6 The gap factor is $(1+y)^T$ (Theorem 2.4); a larger dividend yield $y$ makes it grow faster, so more of the sector's return is invisible in its price index.
IV
Corporate actions.
  • 1 You hold 120 shares at ₹3,600. The company splits each share into 10. Find your new holding, the theoretical price and your position's value.
  • 2 A 3:2 bonus is declared on your 50 shares, cum price ₹900. Find the new number of shares and the theoretical ex-bonus price.
  • 3 A share trading at ₹1,215 goes ex-dividend with ₹15 a share. Find the theoretical ex-price. You hold 200 shares; show your wealth is unchanged before tax, and find the cash you keep with an illustrative 20% tax on dividends.
  • 4 A 1:5 rights issue is priced at ₹400 when the share trades at ₹520. Find the TERP and the value of one rights entitlement.
  • 5 In part 4 you own 1,000 shares and let your rights lapse. How much do you lose, and how could you have avoided it without investing more money?
  • 6 A share closed at ₹1,500 the day before a 1:1 bonus and at ₹760 on the ex-date. Find the unadjusted and the correct (adjusted) one-day return.
  • 7 You bought 30 shares at ₹4,000. Later came a 1:1 bonus and then a split of each share into two. The price is now ₹1,150. Find your holding and your percentage gain.
1 1,200 shares at ₹360; value ₹4,32,000 before and after.
2 $50\times\tfrac{3+2}{2}=125$ shares; price $900\times\tfrac25=\text{₹}\,360$; value ₹45,000 both times.
3 Ex-price ₹1,200. Before: $200\times1{,}215=\text{₹}\,2{,}43{,}000$; after: $200\times1{,}200+3{,}000=\text{₹}\,2{,}43{,}000$. After a 20% tax on the ₹3,000 dividend you keep ₹2,400, so your wealth falls by ₹600 — the tax, not the ex-date drop, is the cost.
4 $\text{TERP}=(5\times520+400)/6=\text{₹}\,500$; one RE is worth $500-400=\text{₹}\,100$.
5 Your 1,000 shares fall from ₹520 to ₹500: a loss of ₹20,000 — exactly the value of your 200 REs ($200\times100$). Selling the REs on the exchange during the rights period would have recovered it.
6 Unadjusted: $760/1{,}500-1\approx-49.33\%$. Adjusted previous close $1{,}500\times\tfrac12=\text{₹}\,750$, so the true return is $760/750-1\approx+1.33\%$.
7 $30\to60\to120$ shares; adjusted cost per share $4{,}000/4=\text{₹}\,1{,}000$. Value $120\times1{,}150=\text{₹}\,1{,}38{,}000$ against cost ₹1,20,000: a gain of 15%.
🎯 Practice corner NISM-styleCase studyInterview

E1. NISM-style A company has 100 crore shares trading at ₹500; 60% of the shares are held by its promoters and the rest are free float. Its free-float market capitalisation is (A) ₹50,000 crore (B) ₹30,000 crore (C) ₹20,000 crore (D) ₹3,000 crore.

Idea Free-float cap $=PQf$ with $f=1-0.60=0.40$.
Compute $500\times100\times0.40=20{,}000$. (A) is the full market cap; (B) wrongly uses the promoter share 0.60; (D) is a slip in the decimal place.
Answer (C).

E2. NISM-style An investor holds 300 shares priced at ₹1,200. The company declares a 1:2 bonus (one bonus share for every two held). Other things unchanged, after the bonus she holds (A) 450 shares at ₹800 (B) 600 shares at ₹600 (C) 450 shares at ₹1,200 (D) 900 shares at ₹400.

Compute Shares $300\times\tfrac{1+2}{2}=450$; price $1{,}200\times\tfrac23=\text{₹}\,800$; value ₹3,60,000 unchanged. (B) is a 1:1 bonus; (C) ignores the price adjustment and creates value from nothing; (D) is a 2:1 bonus.
Answer (A).

E3. NISM-style A constituent of an index splits each of its shares into five. Which index would show a false fall if its divisor were left unchanged? (A) a free-float market-cap weighted index (B) a full market-cap weighted index (C) a price-weighted index (D) none of these.

Reason A split leaves $PQ$ and $PQf$ unchanged, so (A) and (B) are unaffected. A price-weighted index adds the share prices themselves, and one of them falls to a fifth.
Answer (C).

E4. Case study Theta Telecom announces a 3:10 rights issue at ₹80; the cum-rights price is ₹110. An investor holds 5,000 shares and has no spare cash. (a) Find the TERP and the theoretical value of a rights entitlement. (b) Show that selling all her entitlements at that value keeps her wealth at ₹5,50,000. (c) On the exchange the REs actually trade at ₹20. What does she lose by selling there, and what could an investor with cash do instead?

(a) $\text{TERP}=(10\times110+3\times80)/13=1{,}340/13\approx\text{₹}\,103.08$; RE value $\approx\text{₹}\,23.08$.
(b) She gets $5{,}000\times\tfrac3{10}=1{,}500$ REs worth $1{,}500\times23.08\approx\text{₹}\,34{,}615$. Shares $5{,}000\times103.08\approx\text{₹}\,5{,}15{,}385$. Total ₹5,50,000 $=5{,}000\times110$.
(c) Selling at ₹20 brings ₹30,000, about ₹4,615 less than the theoretical value. An investor with cash could buy REs at ₹20 and subscribe at ₹80, getting a share for ₹100 that is theoretically worth ₹103.08 — cheaper than buying in the market, although the share price may move before the new shares are credited, so it is not risk-free.

E5. Case study In the Mini-5 index at the close of Example 3(b) ($M=1{,}40{,}837$, $D=140.1$, $I\approx1{,}005.26$), Delta Motors' promoters sell part of their stake and its IWF is raised from 0.60 to 0.75 at its closing price of ₹873. (a) Find the change in $M$ and the new divisor. (b) Next day only Delta moves, by +2%. Find the new index level and Delta's new weight. (c) What level would the index have shown on the change day without the divisor adjustment?

(a) $\Delta M=40\times873\times(0.75-0.60)=5{,}238$, so $M'=1{,}46{,}075$ and $D'=140.1\times146{,}075/140{,}837\approx145.311$.
(b) Delta's new free-float cap is $40\times873\times0.75=26{,}190$, weight $26{,}190/146{,}075\approx17.93\%$. A 2% rise adds $523.8$ crore: $I=(146{,}075+523.8)/145.311\approx1{,}008.87$, a gain of $17.93\%\times2\%\approx0.359\%$.
(c) $146{,}075/140.1\approx1{,}042.65$ — a false jump of 3.7%.

E6. Interview "The index rose 1% today, yet most of its stocks fell." Show with a three-stock cap-weighted index that this is possible, and say what it tells you about reading index moves.

Construct Weights 50%, 25%, 25%; returns $+3\%$, $-1\%$, $-1\%$. Index return $0.5(3)+0.25(-1)+0.25(-1)=+1\%$, while two of the three stocks fell.
General The index return is $\sum w_ir_i$ (Theorem 2.2); a few heavyweights can outweigh many small stocks. Market breadth (how many stocks rose) and the index return answer different questions; an equal-weighted index is one way to see breadth.
🧑‍🏫 For the teacher

Suggested plan — 2 lectures × 60 min.

  1. The order book and the index. Start with the 📍 situation: show a broker's market-depth screen (any share) and ask what "Bid" and "Ask" mean. Definitions 2.1–2.2 and Theorem 2.1 with Examples 1–2 on the board. Project Interactive 1: start with a market buy of 1,500, press Play, and ask why the average price curves upward; then switch to a limit order at ₹812.10 to show a partial fill. Then market cap and free float (Definitions 2.3–2.4) and build the Mini-5 index (Example 3) together.
  2. Keeping an index honest. Example 4 (three weightings, one split), Theorem 2.3 and Example 5. Project Interactive 2: price-weighted, "not adjusted", events "both" — let the class see the split crash; then switch to free-float. PRI vs TRI (Theorem 2.4, Example 6). Corporate actions (Theorem 2.5) with Examples 7–8; finish with the 📈 Market case.

Class activity. Groups of four each "own" one Mini-5 stock and announce a random daily return; the class computes the index by weights and by the full $M/D$ calculation and checks they agree. Then one group announces a bonus issue and another a constituent change; the class decides who must adjust what.

Misconceptions to watch for. Price per share as a measure of value; the LTP as a guaranteed price; "the index fell, so everything fell"; percentage changes across a bonus or split; ignoring dividends in index returns; rights issues as free money.

Suggested homework. I(2, 3, 5, 6), II(1–6), III(2, 3, 4), IV(3, 4, 5, 7); Practice corner E1, E4, E5.

🛠️ Do it yourselfSpreadsheet mini-project · about 60 minutes

Project. Build your own five-stock free-float index with a divisor, a price-return and a total-return column, and adjust a real share's price history for a split or bonus.

Why it matters. Every index level, every "stock fell 50%" headline and every fund comparison rests on this bookkeeping. Doing it once by hand makes the published numbers transparent.

What to do.

  1. Pick five listed companies. Note each one's shares outstanding and an approximate free-float fraction from its shareholding pattern (promoter holding is disclosed on the exchange websites); treat the IWF as $1-$ promoter share.
  2. Download about 60 days of closing prices for the five (the exchange websites provide historical data). Put dates in column A and prices in B–F; put $Q$ and $f$ in rows 1–2 above.
  3. Column G, total free-float cap: =B5*B$1*B$2+C5*C$1*C$2+… (or =SUMPRODUCT(B5:F5,B$1:F$1,B$2:F$2)). Divisor in H5: =G5/1000; index in I5 and below: =G5/$H$5. Check that =I6/I5-1 equals the weighted average return =SUMPRODUCT(B5:F5*B$1:F$1*B$2:F$2/G5, B6:F6/B5:F5-1).
  4. Add a dividend column J (in index points: dividend per share $\times Q f$ / divisor on its ex-date) and the TRI in K: =K5*(I6+J6)/I5.
  5. Find a company that had a split or bonus (the exchanges publish corporate-action lists). Download raw prices across the ex-date, compute the adjustment factor $\phi$ and an adjusted price column =IF(date<exdate, price*phi, price). Compare the one-day returns from raw and adjusted prices.

Python version: M=(P*Q*f).sum(axis=1); I=M/M.iloc[0]*1000; TRI=1000*((I+d)/I.shift(1)).fillna(1).cumprod().

Hand in: a chart of your index and its TRI, the weights on the first and last day, and the raw vs adjusted price chart around the corporate action.

Going further. Replace one constituent halfway through and adjust the divisor (Theorem 2.3); build an equal-weighted version and compare it with your cap-weighted index.

Instructor note: students' IWFs will differ from the official ones (index providers exclude more categories than promoters), so their index will not match the official Nifty weights — that is fine; the aim is the mechanics. A good check is that the index return equals the weighted return to rounding error; a mismatch usually means the divisor was recomputed every day instead of being held fixed.

🧪 Self-check quiz

0/5 correct

Q1. A share's best bid is ₹250.10 and best ask ₹250.30. Its relative spread is about

Q2. In a free-float index, which stock has the larger weight: P (full market cap ₹60,000 crore, IWF 0.25) or R (₹30,000 crore, IWF 0.90)?

Q3. When a constituent splits each share into five, a price-weighted index must

Q4. A 1:4 rights issue (one new share for every four held) is priced at ₹450 when the share trades at ₹600. The theoretical ex-rights price is

Q5. To judge how well an index fund has done, its return should be compared with