Conditions Precedent or Subsequent: When Each Bites

A condition precedent must be satisfied before either side is obliged to complete. A condition subsequent must be satisfied after completion has already happened. The timing is the smaller half of the difference. The larger half is the remedy: an unsatisfied condition precedent leaves a buyer free not to buy, and an unsatisfied condition subsequent leaves it with a claim.

What is a condition precedent?

The mechanism is easier to see in a flat than in a company. A buyer has agreed to buy one. Buyer and seller have signed the agreement, the price is settled, the buyer's bank is ready, and a date has been written in for handing over the money and the keys. Written into that same agreement is one line: the money changes hands only once the seller produces the housing society's letter of no objection. Until that letter appears, nothing is due from the buyer. Because the one thing both sides agreed had to happen first has not happened, the date can arrive, the seller can be standing there with the keys, and still nothing is due.

The society's letter is a condition precedentSomething that must be true, or must have been done, before either side is under a duty to complete. Until it is satisfied or given up, the duty simply does not arise.. The word to hold on to is not the letter and not the society but the word obliged. A condition precedent does not make completing harder, slower or more expensive. A condition precedent makes the duty to complete conditional on a state of the world that nobody could settle on the day the agreement was signed.

A condition precedent sits between the parties and the obligation, so until it is satisfied or waived, nobody has to complete anything. That is a stronger position than it sounds. The buyer is not asking for permission to withdraw and is not claiming that the seller has done something wrong. The buyer is pointing at a line that both sides wrote, and observing that what it describes has not happened. The obligation to completeThe duty, once every condition is satisfied or waived, to actually hand over the money and take the shares on the day fixed for it. Before that point there is an agreement but no duty to perform it. was never triggered, so there is nothing to withdraw from.

Now put the company back. Harivansh Packaging Limited agreed to buy the whole of Sundarban Polymers Private Limited, and the agreement it signed carried three conditions: a regulatory approval, the absence of a material adverse changeAn agreed test, whose exact wording each transaction settles for itself, under which a serious deterioration in the business between signing and completion lets the protected party decline to go ahead. in the target between signing and completion, and consents from the two counterparties whose contracts change hands with the business. Each of the three protects somebody against something that could not be resolved on the day of signing. None of the three is a complaint about the other side.

Two more properties of a condition precedent matter. The first is that the party a condition protects can give it up. Giving a condition up is a waiverA decision by the party a condition was written to protect to go ahead without it. The condition is not satisfied; it is simply no longer insisted on., and it does not mean the condition was satisfied. A waiver means the protected party looked at what it was being protected against and decided to proceed anyway. So a condition ends in one of two ways, satisfied or waived, and completion waits until one of those two things has happened to every condition on the list.

The second is that the waiting has to stop somewhere. Suppose a condition is never satisfied and never waived. The two parties would be bound to each other forever, and for that reason the long-stop dateThe date written into the agreement after which either side may bring the arrangement to an end if the conditions are still not met. exists. How that date is fixed, and what either side may do once it passes, are set out under the long-stop date itself. One thing matters for the comparison ahead: for as long as a condition precedent is outstanding, the buyer still has its money and the seller still has its business, and both of them know it.

What is a condition subsequent?

Same flat, a few months on. The buyer has paid. The keys have changed hands. The buyer's name is on the agreement and the buyer sleeps there. But the electricity connection is still recorded in the seller's name, and written into the agreement is an undertaking that the seller will get it transferred within a month of the handover. The month passes. Nothing happens. The buyer telephones, writes, and is told it is in hand.

The limit on what the buyer can now do is the point. The flat cannot be handed back in exchange for the money. The duty to complete did arise, and the buyer performed it. Nobody can say the sale never became due. The buyer is left with a promise about a thing that has not been done, and whatever the agreement provides to do about it.

The seller's undertaking on the electricity connection is a condition subsequentSomething the agreement requires to be done after completion has already happened. It is tested on the far side of the transfer, when ownership and money have both already moved.. A condition subsequent is often written as a post-completion undertakingA promise by one party to do a specific thing within a stated time after the purchase has gone through, rather than before it., and the two names describe the same animal from different angles: one names when it is tested, the other names what it actually is.

By the time a condition subsequent is tested, ownership has moved and the money has moved, and what is left is an obligation supported by whatever the agreement provides. That last clause is doing an enormous amount of work. A retention, a right of set-off or a plain claim stands behind such an obligation, and the limits and the security themselves are read where the paper is read.

The shape of the difference deserves a blunt statement. A condition precedent is a gate. A condition subsequent is a promise. A gate either opens or it does not, and while it is shut the buyer keeps everything it had. A promise is only ever as good as what stands behind it, and by the time it is clear whether it will be kept, what the buyer was holding has already been handed over. Two ideas that sit inside the same list in the same agreement, under headings that look alike, and they are not the same kind of thing at all.

One consequence follows immediately and is worth saying before the comparison proper. A buyer reading a conditions list can count the items and feel it has counted its protections. It has not. The list holds a set of gates and a set of promises mixed together, and only the gates protect a buyer in the way it thinks they do.

The answer settles what has to be asked next. A condition subsequent leaves a buyer with a claim, so a condition moved from one side of completion to the other has not been rescheduled. Moving it has converted it.

What does each one do to the obligation to complete?

The real distinction lies here, and the timing is only its visible half. A condition precedent sits between the parties and the obligation. A condition subsequent sits after the obligation has been performed. Everything else that separates the two flows from those two positions.

Take the outstanding condition first. While a condition precedent is unsatisfied, the buyer is not refusing anything. The buyer has no duty to refuse. If the day fixed for completion arrives and the condition is still outstanding, the buyer simply does not complete, and it still has its money. The seller still has its business. Both parties are exactly where they were when they signed, minus the time. Nobody is claiming against anybody.

Now take the same condition on the far side. Completion has happened. The buyer paid, the shares transferred, the business is now run by somebody else. The condition is not satisfied. The purchase it might have declined has already been done, so the buyer is not in a position to decline anything. The buyer holds instead a remedyWhatever the agreement gives a party when a promise made in it is not kept. It is a route to compensation, not a route back to the position before the purchase., which is to say a route to being compensated for the fact that a promise was not kept.

The first gives a right not to buy and the second gives a right to be compensated, and those are not two versions of one thing. A right not to buy is exercised by the buyer, on a day the buyer knows about, over money the buyer is still holding. A right to be compensated is exercised against somebody else, on a timetable the buyer does not control, over money that other side is already holding, and its value depends on limits and security that were written into the agreement long before anybody expected to need them.

There is a household version of this that costs people real money every year. Somebody is engaged to do work on a house. If payment falls due on completion of the work, the money is the householder's instrument: while it is still in hand, the work tends to get finished. If the whole amount is paid in advance, the promise is exactly the same on exactly the same terms, and the instrument is now a conversation. Nothing in the contract changed. The change is which side of the transfer the householder is standing on.

The whole mechanism sits there, dressed in a purchase agreement instead of a painter. And it explains why a comparison of these two ideas cannot stop at when each one is tested. When each is tested is the cause. The effect is what each leaves the buyer holding, and the effect is the part that is worth money.

Why is anything ever left until after completion?

The honest answer has two halves. Given only the second half, the practice sounds merely cynical and the whole idea becomes hard to trust. So take the first half properly.

Some things need the new owner to exist, so they genuinely cannot be done before completion. A share transfer cannot be recorded in the buyer's name until the buyer is the holder. A notice that only an owner may give cannot be given by somebody who is not yet the owner. A filing that follows a transfer cannot be made before the transfer it follows. A licence or a registration recorded in the seller's name cannot be recorded in the buyer's name on a day when the buyer holds nothing. In each case the sequence is fixed by the nature of the thing, not by anybody's convenience.

Conditions subsequent of that kind are proper, and the honest list of them is short, closed and boring, and the dullness is exactly what makes it trustworthy. The list can be written out in full. Registrations in a new name, notices that only an owner may give, filings that follow the transfer, records that carry the new holder's name. When somebody proposes a condition subsequent that belongs to that list, there is nothing to argue about. No version of the transaction ever existed in which the item could have been done first.

Now the second half. Everything else moved to the far side of completion was moved for a different reason, and the reason is almost always a date. A consent has not arrived. A document has not been produced. An approval is still being processed. The completion date is fixed, people have arranged their lives around it, funding has been lined up against it, and the outstanding item is the only thing standing between a room full of people and a finished transaction.

The second category has no end to it. Anything at all can be moved after completion once a date is under enough pressure, and that is what separates the two lists. The first is bounded by the nature of the items in it. The second is bounded only by how badly somebody wants to be finished. A buyer that cannot tell which list an item belongs to has lost the only test that matters here, and the test is a single question: could this have been done before completion? If it could, then moving it is not a scheduling decision.

What does moving a condition actually change?

Here is the part that decides whether the distinction has been understood. When somebody in a transaction proposes moving a condition from before completion to after it, they almost never describe it as a change of remedy. The proposal is described as a way of holding a date.

The words used are usually some version of this: everybody is agreed on everything, the consent is only a formality, it will come, and there is no sense in the parties standing around waiting for it. Every clause of that sentence can be true. The parties may be agreed. The consent may well arrive. Waiting may genuinely cost money on both sides. And the proposal is still not what it appears to be.

The party protected by the condition gives up the ability to decline to complete and receives in exchange a claim, whose value rests on limits and security that were negotiated when nobody expected to need them. That is the exchange, stated plainly. The proposal is not a change to the calendar. The proposal swaps one kind of protection for a different kind, and the two kinds are not priced the same way by anybody who has ever had to use one.

Each protection behaves differently when it is actually needed. The right to decline needs nothing from anybody: no proof, no negotiation, no forum, no delay. The buyer simply does not complete, and every rupee it was about to part with stays where it is. A claim needs everything: it needs the loss to be established, it needs to fall inside whatever the agreement allows, it needs somebody on the other side who can pay, and it needs time. Two protections written into the same document, and one of them works by itself while the other one has to be made to work.

There is also an asymmetry in what each side of the table is giving up, and it is worth naming because it explains why these proposals so often come from the side they come from. The seller giving up nothing but a delay is giving up very little. The buyer giving up a gate in return for a promise is giving up the strongest instrument it holds in the whole process. A proposal that costs one side a fortnight and the other side its leverage is not a compromise, however evenly it is phrased.

Who carries the risk in each case?

Before completion, the party protected by a condition carries almost none of the risk that the condition describes. Carrying none of that risk is the whole function of the condition. If the consent does not arrive, the buyer does not complete, and the consequence lands on the seller, who wanted to sell and now has not sold. The buyer has lost the time it spent and the costs it ran up. The loss is real. But its money is intact and its position is the one it started from.

After completion, that reverses completely. The buyer has paid, holds the business, and carries two distinct risks rather than one. The first is that the undertaking is simply not performed. The second, and the one people forget, is that even if it is performed badly or late, the compensation available is worth less than performance would have been. The money arrives after the customer has gone, so a contract a business needs is not fully replaced by a sum of money.

Moving a condition across completion is a transfer of risk between the parties, and a transfer of risk should carry a price like every other transfer in a transaction. That is not a moral point. The same arithmetic logic runs through the rest of this purchase. When the working capital position at completion came in above the level the agreement had set, the price moved up by Rs 12 crore. When net debt came in above the level the transaction had assumed, the price moved down by Rs 15 crore. Both were positions changing hands, and both carried a figure.

So a risk transferA change that leaves one party carrying something the other was carrying before. In a transaction, most of these are recognised by adjusting the price. that moves no figure at all is the odd one out. The risk was not found to be worthless. Nobody put the question at all. The discussion was framed as a question about a date, and the people in the room answered the question they were asked.

There is a household version here too. The flat is empty and the tenant needs somewhere to be, so a landlord asks the tenant to move in a week before the paperwork is finished. Everybody agrees it will be sorted out. The tenant is now living somewhere without the signed agreement that would settle what happens if there is a disagreement. Nothing was priced. Nothing was even discussed. The risk simply moved, in a conversation about convenience.

How is a condition subsequent enforced at all?

A reader who has followed the argument this far should be asking a sharp question. If a condition subsequent leaves the buyer with a claim, what actually gives that claim any force? A promise on its own is a sentence in a document. Something has to turn a sentence in a document into a promise a seller has a reason to keep.

Three answers. The first is a retentionPart of the price held back at completion and released once the promised thing has been done, so that money stays on the buyer's side of the line.. Part of the price does not travel at completion, and the held-back part is released once the promised thing is done. The second is a right of set-offA right to keep back an amount otherwise payable to the other side, and apply it against what they owe instead., where the buyer keeps back something it would otherwise have paid. The third is an ordinary claim on the undertaking, with no money held anywhere. An ordinary claim is the weakest of the three and by some distance the most common.

How much is held, for how long, against what limits and with what security are questions about the document, and they are read where the paper is read. That ordering rule holds throughout: what is agreed comes before what is drafted.

The shape of each arrangement settles the rest. A retention is strong because the money has not crossed the line, so the position is closer to a gate than to a promise. A set-off is somewhere in between: it depends on there being something else still to pay. An ordinary claim carries nothing behind it except the seller's willingness and ability to answer it. The identity of the seller therefore matters as much as the wording of the undertaking. A promise from somebody who will still be there in two years is a different asset from a promise from somebody who will not.

Which brings the sequence to its last practical instruction. A buyer that accepts a condition subsequent without knowing which of the three it has, and what stands behind it, has accepted a position it cannot describe. The position is not necessarily bad. Nobody in the room can put a value on that position, and something nobody can value has almost certainly been priced at nothing.

How do this purchase's conditions sort, and what does one swap change?

Take the transaction that has run through this sequence and do two things with it. Sort its conditions first. Then run one swap and be exact about what moves and what does not.

The sort

Harivansh Packaging Limited agreed to acquire the whole of Sundarban Polymers Private Limited, and the agreement carried three conditions. A regulatory approval. The absence of a material adverse change in the target between signing and completion. Consents from the two counterparties whose contracts change hands with the business.

All three are conditions precedent, and not one of them was left until after completion. That is the ordinary starting position for a purchase of this kind, and it is worth noticing because a reader who has just learned that two categories exist tends to assume that real transactions carry a mixture. Many do. Harivansh Packaging's list did not, and a conditions list with no conditions subsequent on it at all is perfectly normal.

The three conditions occupied the conditions period, and the conditions period ran nine weeks. The whole span from the term sheet to completion on this transaction was twenty two weeks, so the conditions period was 40.9 per cent of it, and the thirteen weeks before signing went on documentation. Elapsed time varies enormously from one purchase to the next, so a nine week conditions period on one transaction predicts nothing about the length of another.

One caution about what can be read off that. The record fixes the nine weeks and it fixes the three conditions. The record does not fix which of the three took the longest, so no claim about the individual items can rest on it. For 40.9 per cent of the elapsed transaction the buyer had signed an agreement, was not obliged to hand over the Rs 1,137 crore, and knew it.

The swap

Now suppose one of the two counterparty consents was slow. The completion date is close, the funding for the Rs 1,137 crore is arranged against it, and to hold the date the parties agree to move that one condition: completion goes ahead, and the seller undertakes to procure the consent afterwards. Moving that one condition is the whole swap. Everything else on the transaction is left exactly as it was.

Work through what has changed and what has not, and be exact. The exactness is the teaching. The price has not changed. The equity value agreed at signing was Rs 1,140 crore. The sellers of Sundarban Polymers Private Limited actually receive that amount, and it is not the enterprise value of the purchase. Against that, the working capital position at completion moved the price up by Rs 12 crore and the net debt position at completion moved it down by Rs 15 crore, a net of minus Rs 3 crore, so Rs 1,137 crore was paid. All of that is unchanged by the swap.

The business has not changed either. The same plant, the same customers, the same contracts, the same people. The consent that has not arrived is the consent that had not arrived before the swap. Nothing about the thing being bought is different at half past four from what it was at half past three.

Before the swap Harivansh Packaging Limited could decline to complete and keep its money, and after it Harivansh Packaging has paid Rs 1,137 crore and holds a promise about a contract it needs. Read that as an exchange and the size of it becomes visible: risk moved from the seller to the buyer and no rupee moved the other way. A transfer of risk that does not move a price has not been traded. The risk has been given away.

And notice where the two figures that did move came from. Both were positions changing hands at completion, so the Rs 12 crore and the Rs 15 crore were computed, argued over and recorded. The condition that crossed the line was also a position changing hands at completion, and it is the only one of the three that nobody put a figure on.

Where the sequence closes

The worth of the remedy for that broken promise, the limits that apply to it and the security standing behind it are questions about the document, and they are read where the paper is read. The ordering rule holds here as everywhere: what is agreed comes before what is drafted, and the two are separate things.

Agreement and paper can still be seen as separate at this point, and not for much longer. Once drafting begins, the paper carries everything, and it becomes very easy to believe that the words in the document are the deal. The words are not the deal. The deal is the set of positions the parties agreed to hold, and the paper is the record of them. A condition that moves across completion changes the positions, and it does so quietly, in a document that looks almost identical either way.

How does a transaction team use this in the room?

The distinction is not an academic one, and the people who use it hardest are the ones sitting in the last fortnight before a completion date. Take three of them.

Ashwin Rege, who leads the transaction team at Harivansh Packaging Limited, keeps a conditions list. Everybody keeps a list that names the items. The useful part of Ashwin Rege's list is a further column, recording which side of completion each item is tested on and what happens if it is not satisfied. When somebody proposes moving an item, the column has to change, and a change to that column is escalated rather than noted. The list is designed so that the conversion cannot happen invisibly.

Devyani Kulkarni, the chief financial officer, reads the same list from one question: what does this company hold if that item never happens? For a condition precedent the answer is short, and it is the money, all Rs 1,137 crore of it. For a condition subsequent the answer has to name the retention, the set-off or the plain claim, and if it cannot, the item goes back for an answer before it is agreed. A finance function that can state what it holds against every outstanding item has priced the conditions list; one that cannot has only counted it.

A lender funding a purchase reads it from the other end. Its money is meant to be drawn against a completion that actually happens, so it cares about which conditions stand between signing and drawdown and which have been pushed past it. A condition moved after completion does not disappear from a lender's view. The condition reappears as a question about the business the lender has just lent against, and that business now depends on a promise rather than a document.

And an analyst reading a completion announcement, with none of the paper in front of them, still has one honest question to ask: was anything left outstanding at completion, and what stands behind it? The answer is not always disclosed and the analyst may not get it. But asking it separates a reader who understands what completion means from one who has read the word and moved on.

What do a condition precedent and a condition subsequent have in common?

More than the contrast has suggested, and it is worth closing the comparison honestly rather than leaving the impression that one is respectable and the other is a trap. Both are conditions in the same agreement, drafted by the same people, in the same list, often in adjacent lines. Both exist because something about the transaction could not be settled on the day of signing. Both attach to a real thing that somebody needs done. Neither is a criticism of the other side, and neither is unusual.

Both also share the same underlying source of difficulty. The world does not arrange itself around a completion date. Approvals take the time they take. Counterparties answer when they answer. A business can deteriorate between two dates for reasons nobody controls. Conditions of both kinds are the parties' way of saying what happens in a world that will not be hurried, and any agreement without them would simply be an agreement that had not thought about it.

The two are the same tool pointed at the same problem, and everything that separates them follows from which side of completion the tool is pointed at. That is why the comparison matters. Nothing in the wording of a condition shows which of the two it is. Only its position does, and its position decides everything that happens afterwards.