Three brothers run an electronics shop on Hall Road. Suppliers know them by the painted board above the shutter. When the eldest has a heart attack, they discover the problem: every account, every credit line, every tenancy is in his name. The business is him. If he goes, it goes.

Their lawyer explains what registration actually does. On the day the certificate issues, a new person exists — not any brother, not all three together. That person owns the shop, signs for the stock, sues the supplier who shorted them, and attends nobody's funeral. Every rule that follows exists to protect the strangers who will one day deal with a person they can never meet.

First, who is allowed to become an owner. They can keep a closed list — a capped number of members, no invitation to the public, and no brother free to sell his share to an outsider without the others' say. Two things do not count against that cap: a share held jointly counts as one member, and employees who took shares while employed stay outside the count even after they leave. They can instead open the list to the whole street. The unmarried cousin who wants his own registration takes a list of one, and must name in advance the person who steps in the day he dies. And their father, registering the family charity, learns it can never pay a rupee out to anyone — the profit stays inside forever, and the licence is withdrawn if it does not.

Then two sets of rules, and the difference between them is the whole topic. The board above the shutter is what the street reads: the name, the city, what business is done here, how much capital stands behind it. The notice taped inside the storeroom is for the brothers alone — who opens up, who signs cheques, how a disagreement is settled. Repaint the storeroom notice by family agreement and nobody outside is affected; the only thing the brothers may never do is write a rule forcing a member to put in more money than he promised. Repaint the board, and you have told an entire street something new about a person they have been trusting on credit. So the law makes the board harder to repaint than the notice: the members must resolve it by the heavy majority, and for the biggest lines on it — the province, the principal business itself — the Commission must confirm before the paint is legally dry.

The board is policed even at the start. They wanted Punjab National Electronics. The counter refused: a name that implies government backing is not theirs to take, and even a name already on the register can be ordered changed later.

Two years on they quietly move into property dealing. Family agreement, no resolution, no filing, board unchanged. A supplier who extended credit against "electronics" sues; the registrar fines; and in law the change never happened, because the change bites when it reaches the register and not when the brothers agree. The address on the board still stands, so the summons is validly served whether or not anyone bothers to read it.

A six-storey building on a commercial plot in Gulberg. On the approved plan sits one number nobody thinks about until it matters: the declared depth of the foundation. The bank that lent against the building read that number. So did the steel supplier who gave sixty days' credit. Neither has ever been down there to look.

The flats are not all the same kind. Most owners hold the ordinary sort — a vote at the owners' meeting and a share of the rent in proportion to what they put in. The aunt who financed the lift took a different kind: her rent is fixed, it is paid before anyone else's, and if the building is ever sold she is repaid ahead of the ordinary owners — but she does not argue in the meeting. Some units carry extra votes, some carry none, some can be handed back at an agreed price after a fixed number of years, some convert into ordinary units later. Every one of these kinds exists only because the plan and the owners' own rules allowed it from the start, and because the authority was told and its conditions met.

Adding is the easy direction. Sanctioning more units than were originally approved needs a resolution of the owners and a notice to the authority; so does merging two units into one, splitting one into four, or striking off units that were sanctioned and never sold. A few owners are still paying by instalment, and if one stops, the society may call the balance and, failing that, forfeit what he holds.

Changing one kind's terms is different. The aunt's fixed rent cannot be reduced by a show of hands among the ordinary owners. Her class must agree, in its own room, by itself. The upper floors cannot vote away the ground floor's parking.

And then the gravest thing anyone can propose: returning part of the money to the owners and writing the foundation down to a smaller number. The owners' heavy majority is not enough. The Court must confirm it, because the people who must be heard first are not in the room — they are the bank and the steel supplier, who parted with money against a number on a plan.

Day to day, units move. A sale runs on a signed instrument delivered to the society, which updates its register within the prescribed days; in a closed society the committee may refuse only on the grounds written into its own rules, and must say which ground. A death is different — nothing is signed and nothing is sold, and the law itself moves the unit to the heir.

One owner sells and never lodges the instrument. The register still shows him. So the meeting notice goes to him, the rent cheque goes to him, and the buyer — who has paid in full — discovers he owns something the society is not obliged to recognise. Whatever the register says is presumed true until somebody proves otherwise, and until then the paper beats the payment.

Chaudhry Textiles needs money for new looms. The CFO has found a rich outsider willing to write the cheque tomorrow. The company secretary stops him at the door with four words: pehle apne gharwalon se poochho.

The house has a rule. Anyone already inside gets first refusal, in proportion to what he already holds, before a rupee is taken from a stranger. The reason is not sentiment. If the outsider comes in at a price the family never saw, every existing member's slice quietly shrinks and somebody has taken value out of his pocket without asking.

So the family route runs first. The board announces it, and because the announcement moves the share price it goes to the exchange the same moment it goes to anyone else. Every member receives a letter of offer — the same price for all of them, cash only, no favours, and a window inside which each may take his portion, hand it to somebody else, or simply let it lapse. If a floor is set for how much must be taken up, that floor cannot be set low enough to be meaningless ⚑. Every substantial shareholder has to say in writing whether he is taking his share or arranging someone who will. And whatever the family leaves on the table is picked up by an underwriter — licensed by the Commission, and not a company connected to the issuer, because a rescue arranged with your own cousin is not a rescue.

Sometimes the family is genuinely the wrong answer. A strategic investor who brings buyers as well as money. A machinery supplier willing to take shares instead of cash. A lender willing to convert what he is owed into ownership. The law does not forbid any of it — it makes the house explain itself. The board must put down, in writing, exactly how many shares, what percentage of the company before and after, who this person is, why him, what the company gains, and what one share is actually worth on the books today. That paper goes to the members, who must approve by the heavy majority, and then to the Commission, which approves or does not. Where the payment is not cash, somebody independent values what is being handed over, because a machine is worth what a valuer says and not what the seller says.

Employees enter through their own side door — a scheme the Commission has approved, with its own pool and its own pricing, so that the people who run the looms can own part of them.

And then there is the issue where no money changes hands at all: the family capitalises reserves it already owns and turns them into new shares. Free and painless — except that the reserves must genuinely be there, certified, before anything is announced. A bonus declared out of reserves that exist only on a spreadsheet is not generosity. It is a misstatement with the members' names on it.

Not every fundraising is a public mela. Sometimes the seth calls a few people into his drawing room, closes the door, and raises what he needs over chai. The law allows this, and the whole regime is about that closed door.

The boundary sits in the definition itself. This is an offer to identified people — named, counted, and not exceeding the prescribed number ⚑ — with nothing said to the public at all. Cross either half of that sentence and you are no longer in the drawing room.

Who may be invited is not left to the seth. The invitees are people who can look after themselves: institutions and sophisticated investors who read balance sheets for a living, who can price the risk, and who can survive being wrong. The retired schoolteacher is not on that list, and the reason is exact — she is the person the prospectus regime exists to protect, and the drawing room has no prospectus in it.

How the invitation travels is also scripted. It goes as an information memorandum: who the issuer is, what the security is, what could go wrong. Not a newspaper advertisement. Not a tout at the chowk. Not a message forwarded onward.

And the seth must keep the guest list. Who was offered, who subscribed, on what terms, in writing — because the day SECP knocks, he does not get to say the gathering stayed inside. He has to show it.

Here is how it usually breaks. One invitee, meaning no harm, forwards the memorandum to a WhatsApp group of forty investors so a friend does not miss out. Nobody advertised anything. Nobody stood at a chowk. But the offer has now been made to people the seth never identified and never assessed, and it has been made by general circulation. In law that is not a private placement that went slightly wrong. It is a public offer that was made without approval — which drags in the whole prospectus regime the seth was trying to avoid: approval defects, liability for anything in the memorandum that turns out to be untrue, and penalties on the company and its directors.

That is the trade the Rules offer. Obey the closed door and you get speed: no prospectus, no book building, no exchange, money raised in an afternoon. Ignore it and every cup of that chai becomes evidence.

A large land parcel outside the city is being sold in plots, and the seller cannot simply announce a price. He does three things in order, and a company coming to market does exactly the same three.

First, the full history. Every survey, every dispute, every mortgage, every case pending — written down, approved by the authority, and handed to anyone who asks. Not because buyers demand it, but because of what happens if it is wrong: a buyer who paid on the strength of a false line in that document can recover from the seller, from every partner who signed it, and from the surveyor whose report was attached. If the falsehood was deliberate, the file goes further than compensation. This is why the document is long and boring. Length is the liability speaking.

Second, the gate. Not everyone gets to sell land this way. The seller, his backers and his managers must clear the authority's fitness tests and show a track record, and the seller's own retained plots are frozen ⚑ for a fixed period — so the family cannot sell the front row, pocket the money, and vanish before anyone finds out what the back row is like.

Third, the price, and this is where people expect an auction and get something else entirely. There is no shouting. A floor is published, below which no bid counts. Serious institutional buyers submit sealed bids at or above it, each with earnest money deposited upfront. A bidder may raise his number while the book is open. He may not take it back, and he may not take his earnest money back either — which is the whole reason the process produces an honest price. In an open auction you can bid high to excite the room and quietly walk away. Here there is no walking away, so nobody bids a number he is not willing to pay. And the seller may not put his own relatives in the queue to make demand look bigger than it is.

When the book closes the bids are stacked from highest downward until the plots on offer run out, and the number where they run out is the price everyone pays. Ordinary buyers then come in at that price, or below a cap on it, for the portion set aside for them ⚑. If the sealed bids never reach the required multiple of what was offered ⚑, there is no price to strike, the whole thing is cancelled, and the earnest money goes back inside the timeline.

Succeed, and allotment, refunds and delivery all run on the clock, the plots are registered at the exchange, and the seller acquires a permanent obligation he did not have before: from that day on he must tell the whole market whatever he tells anyone, keep enough plots in public hands to make a real market, and he cannot simply walk off the register when he tires of it.

The mohalla tandoor sells brass tokens; one token, one roti, redeemed whenever you like. Hundreds are out in the mohalla. After a good year the owner decides to buy some of them back.

Notice what that is. He is not selling anything. He is spending real money to take his own promises out of circulation, so that fewer claims exist against the same tandoor. Everyone still holding a token now owns a larger share of whatever the tandoor is worth — but the tandoor itself has less cash in the box than it had this morning, and that is exactly why nobody is allowed to do this quietly.

Two people can be hurt. The man who supplies the flour on credit, because the cash that would have paid him has just gone out to token-holders. And the token-holders themselves, because an owner who buys tokens back can push their price up on the day he chooses to buy.

So every step is chaperoned. The owner decides, and because his decision moves the token price he tells the whole mohalla the moment he tells anyone. The token-holders then bless it by the heavy majority, and the resolution has to say the specific things: how many tokens, at what price or by what formula, over what period, and by which of the two methods. There are only two. He can offer to all of them at once, same price, same window, so nobody is favoured. Or he can buy in the open market at whatever the going rate is.

Before any of that, he opens the box. The money must come out of genuine distributable profit ⚑ — not from the flour fund, not from borrowing. He must be solvent, current on what he owes, and inside the prescribed limits on his borrowings and his financial condition ⚑. If the box does not pass, the resolution does not save him.

The tokens he buys go one of two ways. Cancelled, and the tandoor genuinely has fewer promises outstanding forever. Or put in a locked drawer as treasury tokens, up to the permitted ceiling ⚑, where they sleep: no vote, no share of profit, no entitlement when new tokens are issued. A token in the drawer is not a token in the mohalla. Waking one up and selling it again is not a small administrative act — it is issuing a token to somebody, and it goes back through the whole further-issue framework as if it were new.

During the window the people who know most stand back from trading, the tandoor may not talk its own price upward, and when it is done the purchases are reported and filed. The float shrinks in daylight, token by token, on a record anyone can read afterwards.

Gulshan Colony has one rule that everyone signs on entry: the colony must always know who is amassing it. Malik Sahab has been buying plots for a year — one in his own name, two through a nephew, one through a firm his wife owns. On paper, four unrelated buyers. In fact, one man.

The first thing the rule asks of him is arithmetic. He must add up everything held by himself and everyone acting with him, and the moment that total crosses the marked lines ⚑ he tells the colony office, the exchange and the Commission what he now holds. Not what he bought this week. What he holds altogether.

The second line is different in kind. When his total would cross thirty parts in a hundred of the voting plots — or when, at any level, he can name the majority of the committee and set the colony's policy — the quiet buying is over. He is no longer a large resident. He is taking the colony, and everyone else living in it is entitled to leave at the same price he thought it was worth.

From that point the order of events is the law, and the order is the thing candidates get wrong. His own board resolves to make the acquisition. He appoints a licensed manager to run the offer — someone answerable, who is not him. Then the announcement of intention: the colony office, the exchange and the Commission are told, and it is published within two working days in English and Urdu dailies circulating in every province, so that a plot-holder in Quetta learns it the same morning as one in Karachi. Only then may he sit down with the seller and sign anything. A man who signs first and announces afterwards has committed the one breach the whole regime exists to prevent, because he has bought control with information nobody else had.

Then the offer itself, published with its disclosures and sent as a letter to every remaining plot-holder. It must be for at least half of what is left after his own holding — not half the colony — and at a price no worse than the regulations' floor, which is the highest of what he negotiated with the seller, the market average over the look-back window, and whatever he himself paid while accumulating ⚑. He cannot pay the seller generously and the neighbours meanly.

The colony may not stay quiet either. A rival has twenty-one days from the first announcement to put up a competing offer, and it must be at least as good in both volume and price. Malik Sahab may then raise his own, up to seven working days before closing; or he may walk away, but only within seven working days of the rival's announcement. Let both windows pass and he is bound — and his offer now stays open until the rival's closes.

There are narrow exits — a fresh public offer inside the last twelve months, arrangements among sponsors already inside ⚑ — and they are read narrowly, because every exemption is a plot-holder who did not get to leave.

An apartment block's owners' association meets once a year. Half the owners live abroad, most of the rest come only if something concerns them, and the whole system rests on one thing: the notice.

The annual meeting is not optional and it is not at anyone's convenience. A newly formed association must hold its first within sixteen months ⚑ of coming into existence, and after that within a hundred and twenty days of each year-end, in the town where its registered office is — with a listed company also required to let owners in other cities take part from there ⚑. Twenty-one days' notice goes to every owner and to the registrar, and a listed company publishes it in English and Urdu newspapers besides. Four things are expected business: the accounts, the auditors, the dividend, and the election of the committee.

Everything else is special business, and special business carries a passenger. With the notice must go a statement of material facts — the full picture of what is being proposed and, in particular, which committee member has an interest in it and how much. This is the provision the whole topic turns on, and here is why. At one meeting, item seven reads simply "approval of maintenance contract." It is passed in ninety seconds. It emerges later that the contractor is the chairman's brother-in-law and the rate is double. Nobody in that room voted for that; they voted for a line of text that concealed it. The resolution is challengeable — not because the contract is bad, but because the owners were asked to consent to something they were never told. Consent obtained without disclosure is not consent. That is the sentence to carry into the exam.

Between annual meetings, urgency has its own door. Owners holding the requisite voting power ⚑ may requisition a meeting, and if the committee sits on it beyond the statutory days, the requisitionists may call it themselves and recover the cost from the association.

A meeting also needs enough people in the room. If the quorum ⚑ is not present within the grace period, the meeting adjourns by operation of law — nobody has to move anything. An owner who cannot attend sends a proxy on the prescribed instrument, deposited before the deadline, and the proxy need be an owner himself only if the association's own rules say so ⚑.

Decisions come in two weights. Ordinary matters pass on a simple majority of those voting. The heavy ones need three-fourths of the votes cast, on twenty-one days' notice, which may be shortened only with the prescribed consent ⚑. Voting is by show of hands until somebody demands a poll, and then it is by shares. The chairman runs it.

And afterwards, the minutes. Signed and kept, they are the association's memory, and the law treats what they record as true until somebody proves otherwise. An owner who says "that is not what we decided" is arguing against a document, which is a much harder thing than arguing against a recollection.

The family that owns the mill is scattered — Dubai, Toronto, Karachi — and the room where decisions get made is in Faisalabad. For years that meant one thing: whoever lives near the room decides, and whoever does not signs a proxy and hopes.

The postal ballot ends that arrangement. For the matters the Regulations specify ⚑, the vote does not happen in the room at all. Every member votes directly, by post or electronically, wherever he is. For other matters the company may choose to run it this way. And in a listed company, when members holding the requisite shareholding ⚑ ask for it, even the election of directors moves to electronic voting.

The reason this exists is worth its own mark, so understand it before the procedure. A general meeting is a room, and rooms can be managed. Hold it far away, hold it in the middle of a working week, gather proxies from members who barely read them, and the sponsor arrives with the outcome already in his pocket. The postal ballot takes the count out of the room. Every share votes directly and every vote leaves a trace. It is minority protection, and it is used precisely on the questions where the majority has something to gain.

The drill is short. The notice goes out carrying the ballot paper itself and the login details for voting electronically. A scrutinizer is appointed — independent of the company, holding custody of the ballots, the neutral man who counts because nobody trusts the interested one to. Members return their chitthi inside the window, electronic votes travel encrypted, and a member who has voted electronically cannot then post a paper vote and be counted twice. The scrutinizer reconciles, reports to the chairman, the result is declared, filed and put on the website. And what emerges stands exactly as if it had been passed by hands raised in the room.

Where it goes wrong is at the one point everything rests on. At one company the scrutinizer turned out to be a firm that did most of its work for the sponsor. Nothing in the count was ever proved wrong. It did not matter — the entire mechanism is a promise that the person holding the ballots has no stake in the answer, and once that promise is broken the result is worth nothing, however accurate it may have been. An independent scrutinizer is not an administrative formality in this regime. It is the regime.

A listed company has seven seats to fill. The sponsor family holds sixty parts in a hundred; the rest is spread among small holders, and one investor holds twelve. Every year the family takes all seven seats, because sixty beats forty in every single contest.

Then the election is run the way the law actually requires, and the arithmetic changes shape. Each member gets votes equal to his shares multiplied by the number of seats, and he may pile every one of them onto a single candidate. The family, spread across seven contests, is strong everywhere. The investor with twelve, stacking everything on one name, is unbeatable in one place. Seven seats, so a bloc needs a little more than one-eighth of the votes to guarantee a seat — and twelve is more than one-eighth. He gets his director, not by permission, but by arithmetic. That is the whole point of cumulative voting, and it is why candidates who reproduce the formula without the reason lose the marks that follow it.

The same arithmetic guards him afterwards. When the family later moves to remove him, the resolution fails: if the votes cast against removal would have been enough to elect him cumulatively, he stays. A seat won by stacking cannot be taken away by a simple head-count.

Around that election sit the ordinary bones. A board is at least one for a single member company, two for a private, three for an unlisted public, seven for a listed one ⚑. Some people simply cannot sit: minors, the unsound, undischarged insolvents, the fraud-convicted, with extra bars for listed companies. The board fixes the number of seats before the election, retiring directors step down, and anyone who wants to stand files notice fourteen days ahead ⚑. If a seat empties mid-term the board fills it for the remainder — within the prescribed days ⚑ in a listed company — and some seats empty by themselves, without anyone voting, when a director misses meetings or becomes disqualified.

Being elected is not the same as being unlimited. The biggest decisions are not the board's to take at all: selling the undertaking, disposing of a substantial part of it ⚑, and the other reserved matters go to the members. Lending money to a director is chaperoned in its own section. And in every meeting there are two separate duties that candidates keep merging: a director must disclose any interest he has in a contract, and an interested director must then not vote on it. Disclosure alone is not compliance.

Above the board sits the chief executive, first appointed by the directors within days of incorporation, re-appointed by each new board for a term that ends with that board's own. Removing him early takes three-fourths of the directors or a special resolution of members. And the bar is the same at the top as at the bottom: a person who could not be a director cannot be chief executive either.

A match between two mohallas needs an umpire with no cousin on either team. Everyone understands this instinctively, and nobody argues that a fair-minded man with a brother batting is good enough.

A listed board needs the same person, and the law defines him entirely by what he is not. No material money relationship with the company, its sponsors or its management — nothing that could bend a judgment. Not someone who worked there recently, not someone who audited it, not a shareholder large enough to have his own position ⚑, and not connected through a relative to any of those. Independence is a description of absences, which is why it is proved by producing a list of things that are not true.

It is also not a status conferred once. It is a condition tested at appointment and every day afterwards.

Selecting him is a documented process, not a name suggested over tea. Candidates may come only from the databank maintained by the institute the Commission has authorised. The company searches it, evaluates candidates against criteria it has actually written down, takes the candidate's own declaration of independence, and then — the step everyone forgets — the board makes its own assessment of whether the declarant genuinely qualifies and records that it did. A declaration is the candidate's opinion of himself; the board is not permitted to simply accept it. And the directors' report then tells the members how the process was run.

How many umpires a board needs, and how many sit on each committee, is not this regulation's business at all — that belongs to the Code. This regulation owns the manner and the selection. Keeping those two apart is itself examinable, because candidates routinely cite the wrong instrument.

Midway through the season the umpire's brother joins one of the teams. Nothing the umpire has done is wrong. It does not matter. The moment independence is lost it must be disclosed, the board must reassess him, and if his loss breaks a composition requirement the board has to be reconstituted.

And this is why the examiner buries the disqualifying relationship deep in a long set of facts and says nothing about it. An umpire who was never neutral does not just spoil one decision. Everything he blessed is contaminated — the audit committee's findings, the related-party approvals he signed off, the remuneration he helped set. Finding him is the question. The consequences are the marks.

The school committee needs a caterer for four hundred children, every day, all year. The chairman's brother runs the best kitchen in town and quotes three hundred rupees a head. Two outside caterers quote two hundred and twenty.

Nothing here is forbidden. The law never asks whether the brother's food is good. It asks one question and then makes the committee answer it on paper: would this committee have signed this contract with a stranger?

The first move is to name the relationship, and the net is wider than people expect. It catches directors and their relatives, the people who actually run the place, associated companies and undertakings, and entities sitting under the same influence. The chairman's brother is caught. So would his son's company be, and so would a supplier owned by a director's wife. Read the definition before you decide there is no related party — that is where scenarios are built to trap you.

The second move is price, and here the extra eighty rupees is the whole problem. The transaction must be at arm's length, and arm's length is not a feeling, it is a demonstration. The committee must show its working by a recognised method: what an unrelated caterer charges for the same thing, or what the brother charges his other customers less his margin, or what it costs him plus a defensible return. If the eighty rupees buys something real — a nutritionist, a second kitchen, delivery twice a day — the file says so. If it buys nothing, the price is not at arm's length and no amount of procedure fixes that.

The third move is who may approve it. The board approves, but the chairman must first disclose his interest and then leave the vote alone — disclosing and voting anyway is not compliance, it is two breaches. And when a majority of the committee is interested, which is exactly what happens in family companies, the board cannot cure itself: the matter climbs to the members in general meeting. So too where the terms are not at arm's length. Where these transactions recur, the company's policy frames them in advance rather than approving each one from scratch.

The fourth move is the paper. A maintained record of every related party transaction — who, what relationship, what terms, and why the price is defensible — with the particulars placed before the board and kept available for inspection. The audit committee reads that register the way a nosy but honest phuppo reads a wedding guest list: not to stop anything, but so that nothing passes unnoticed.

The trap in the exam is a committee that did everything except the second move. Disclosed, abstained, minuted, filed — and never once showed why three hundred was the right number. Procedure without pricing is not compliance. It is a well-documented breach.

The cousin's factory is struggling and he needs fifty million. The family company has it sitting in the bank. Around the table everyone already agrees, which is precisely the situation the law distrusts.

The first question is not whether to lend. It is whether he is family for this purpose at all, because the answer decides which rulebook applies. Common directors above the threshold, twenty parts in a hundred or more of the shareholding, or common control — meet any of these and the cousin's factory is an associated undertaking, and everything that follows is compulsory. Miss the test and you have answered a different question from the one asked.

If he is family, the money cannot move on the board's say-so. Every kind of transfer is caught — buying shares, lending, advancing, guaranteeing his borrowing from somebody else — and each needs the members' approval by the heavy majority. Not a vague blessing either. The resolution has to state how much, in what form, for how long, for what purpose, and what the company expects to get out of it.

And then the provision that carries most of the marks, and the one the story exists to explain. The cousin proposes to pay nothing for two years while he turns things around. He cannot. A loan to an associate must carry a return not below the prescribed floor ⚑, anchored to what money actually costs the company itself. Ask whose money is being lent, and it becomes obvious. The company is either paying interest on its own borrowings or earning something on that deposit. Lend below that line and the shortfall is not the cousin's discount — it is a transfer of value out of every shareholder's pocket, including the ones who never met him. Shareholders' money does not subsidise cousins.

The homework is forced into daylight before the members vote. The statement of material facts that goes with the notice sets out the associate's financial position, where the money for repayment is going to come from, and — if the associate is loss-making — a justification that every member gets to read. Due diligence and the prescribed auditor certifications ride alongside ⚑, so nobody can later say the numbers were nobody's job.

Two years on, the cousin cannot pay. The directors quietly reschedule at a lower rate. That is a second breach, not a workout: terms once approved by the members cannot be softened by the board, and any change needs to go back to the same room that granted them.

The teeth are personal. Directors who invest without the resolution, or on sweeter terms than the members approved, are liable to make the loss good themselves. Help your cousin — through the front gate, on recorded terms, at a rate that respects whose money it really is.

Four hundred workers at a hosiery unit in Korangi. Every year before Eid there is a bonus, and every year it has exactly the same shape.

Nobody can demand it. A worker who walks into the office in Ramzan and says "give me my bonus" is sent back to his machine, and he has no answer, because a bonus is not a wage. It exists only if the owner decides there is something real to give it out of. And it must come out of something real: not borrowed against next season's yarn, not taken from the money already set aside for raw material. Some units are required to put aside a fixed slice into reserves before anything else, and that slice goes first. If the year was bad there is no bonus, and nobody has been wronged.

The general manager works out what the unit can afford and puts a number to the owner. The owner may cut it. He may not raise it — the man on the floor knows what the floor can carry, and the whole system depends on the number coming up from the floor rather than down from the office.

Then the manager stands on a crate in the yard, in front of four hundred men, and says: one month's salary, into your accounts, by Friday.

Everything changes in that sentence. Before it, a hope. After it, a wage. The owner can no longer say sales were slow — slow sales were a reason not to climb onto the crate; they are not a reason to climb down from it. He cannot announce next week that he has reconsidered. He cannot pay a favoured foreman first and the rest in March. And "your account" now means the account: the money goes down the same pipe the salary goes down, not as a cheque handed to a nephew that will be lost for a year.

Only a few reasons excuse a delay, and they are narrow — the law forbids the payment, the worker himself has asked for it to be held, or someone else claims the same money. Cash-flow is not on that list.

And when Friday comes and the money does not, the four hundred men do not go looking for the owner, who is in Dubai. They surround the man who stood on the crate. The law does exactly the same thing: it puts the default on the chief executive personally — a fine, and disqualification from that chair. This is the provision that turns a cash-flow problem into a career problem, and it is why "can we defer?" is never a finance question.

Separately, and outside all of this, the owner sometimes hands out an advance mid-season on his own authority without waiting for the yard. That one he may decide alone. Zakat, tax withholding and money nobody ever came to collect each run on their own plumbing ⚑ — but none of it moves Friday.

Every month the mill's workers put part of their wages into the provident fund. That box holds retirements and widows' futures, so the law does not trust anyone's judgment about where it may sleep — it draws four ceilings around it, and the investable amount is whatever the lowest of the four allows.

The trustees have money to place and a proposal in front of them, and the mistake they are about to make is the one the exam is built on: they check the ceiling that is easiest to check, find room under it, and invest. Four ceilings means four calculations, and passing three of them is failing.

The first is a roof over the whole idea of market risk: only so much of the fund may sit in listed securities at all ⚑. Everything else belongs in safer places, and government securities are the one home with no ceiling over it.

The second is a smaller room inside that roof: listed equity gets its own, tighter limit ⚑, because shares behave differently from rated debt and the fund is not a place to be brave.

The third is a wall inside that room. No single sector may take more than its share of the equity money ⚑ — and note the base carefully, because it is a proportion of the equity sub-limit, not of the fund. A trustee who applies it to the whole fund gets a number several times too large and never notices.

The fourth is a single door: exposure to any one company is capped at the lower of a proportion of the equity money or a proportion of that company's own paid-up capital ⚑. Two different bases again, and the answer is whichever binds harder.

Then compute each ceiling as remaining space after what the fund already holds, and take the smallest of the four. That number is the answer. The rest of the working is there to show why the other three were not.

Two things sit outside the arithmetic and are quietly worth marks: money-market scheme units and term finance certificates are not equity and do not enter the equity computation at all. Candidates who fold them in overstate the exposure and lose the number. Quality gates sit alongside — eligible equities meet the prescribed track record ⚑ and debt instruments carry the required rating.

And the fence with a moral behind it: the employer may not park the workers' box in its own shares or its associates' beyond the permitted sliver. That is the conflict the entire regulation exists to prevent — a company using its employees' retirement money to support its own price.

Last, the difference between a ceiling breached by the market and one breached by a decision. If prices move and the fund drifts over a limit, there is a window to rebalance. If the trustees knowingly crossed it, there is no window — there is personal liability.

A buyer in Multan is about to pay eight million rupees to a company in Lahore he has never visited, run by people he has never met. He does one thing first: he looks the company up on the register. Directors, capital, registered office, charges over its assets, resolutions filed. Ten minutes, a small fee, and he knows more about a stranger than most people know about their neighbours.

That is the whole point of the registry, and it explains every rule inside these Regulations. A public record that anyone may inspect and take certified copies from is what allows commerce between people who have no reason to trust each other. Everything else is machinery protecting that record's reliability.

The machinery starts before the company exists. Names are reserved, and a long catalogue of them is refused: deceptively similar to something already registered, suggesting government patronage, offensive, or needing a ministry's clearance first. If a name slips through, there is an objection and rectification route to take it back — which is where Punjab National Electronics died in the first story. The incorporation submissions themselves have their own prescribed shape.

After incorporation, the record has to keep up with the company, and this is the part that leaks into every other answer in the paper. Allot shares, and a return of allotment goes in. Transfer them, change a director or an officer, move the registered office, pass a special resolution, create a mortgage or a charge — each has its own form and its own filing window, and the windows are short. The annual return refreshes the whole picture once a year, and the statutory registers the company keeps at its own office run alongside the registrar's copy.

Notice the pattern in all of it. The deed is done when it is done — the members really did resolve, the charge really was created. But it is invisible until filed, and an invisible fact cannot protect anyone who relied on the register. That gap between doing and filing is where the buyer in Multan gets hurt: a charge created six weeks ago and never registered does not appear, so he lends against an asset that is already pledged.

The desk also punishes lies. A false statement in a filing is not a clerical matter; it carries its own sanction, because the register is only worth what the filings are worth.

For the exam this topic rarely stands on its own. It arrives as the last line of somebody else's answer, and it is worth a mark almost every time. Train the reflex: whatever corporate action you have just described, finish it at the registrar — the right form, inside the window.

A fabric mill outside Faisalabad stops running. What happens next turns on one question: can it pay everyone it owes?

If the owners are sure it can, they may close it themselves — but only after the directors sign a written declaration, on their own responsibility, that every debt will be paid in full within the stated period. That signature is the fork in the road. Sign it honestly and the owners keep control of the closing and appoint their own man. Sign it when it is not true and the men who signed answer for it personally. If they cannot sign it, the closing is still voluntary but it is no longer theirs: the creditors meet, their choice of liquidator prevails, and the board's powers stop the day he walks in. And if the owners will not close at all, a creditor who served a formal demand and was ignored for the prescribed period ⚑ can have the gates closed over their objection — as can a member who shows the business has become impossible to carry on. Inability to pay is also presumed where an execution comes back unsatisfied, or where insolvency is proved counting contingent liabilities.

The Court's man takes the keys. He takes custody of everything, demands a sworn statement of affairs from the officers, files his own report on how it came to this, and then builds two lists. One is of everyone the mill owes. The other surprises people: everyone who still owes the mill — shareholders who never paid their shares up in full, including some who sold out during the past year ⚑ and believed they had walked away. He can summon anyone holding mill property, examine promoters and directors publicly, and have an absconding contributory arrested.

Then the queue at the gate, and this is where most candidates reason wrongly.

The finance company holding a registered charge over the generator is not in the queue. It never was. It takes the generator, or its value, and leaves. That value was never part of what the mill has to share out — a secured creditor realises outside the estate, and only what remains after him is the estate at all.

Of that remainder, government dues and employees' claims go first, each worker capped at the prescribed amount per head ⚑, with anything above his cap dropping back into the general pile. Then everyone else — suppliers, the bank's unsecured balance, the landlord — and here is the correction that earns the marks: they are not paid in order of arrival, and they do not take fixed shares. They abate together. If there is thirty paisa in the rupee, every one of them receives thirty paisa. The first man at the gate takes no more than the last. Members take the residue, which is almost always nothing, and that is what limited liability was always for.

Last, the liquidator looks backwards. The clock started the day the petition was presented, not the day the Court ordered ⚑. In the weeks before it the owner signed a truck over to his brother-in-law and gave a floating charge to a friend's finance company. The truck comes back as a fraudulent preference; the charge is void except to the extent of whatever fresh money the friend actually advanced that day. Dissolution finally closes the grave — and even that can be reopened within the window if fraud surfaces.