The most expensive sentence in M&A is not a number. It is: “We agreed on the price, the rest is boilerplate.” Owners say it, or quietly believe it, at the exact moment their position is weakest, after exclusivity has been granted and the competing bidders have gone home. The letter of intent set the headline. The purchase agreement decides how much of that headline you keep, how long you remain on the hook for a business you no longer own, and who writes the check when a problem surfaces 14 months after closing. Chapter 12 covered how deal structure moves the economics; this chapter covers how the definitive agreement moves the risk.
One line before the machinery: this chapter is educational, not legal advice. Its purpose is to make your first conversation with experienced M&A counsel an intelligent one, not to replace it.
From LOI to Definitive Agreement
By the time an owner signs a letter of intent, the deal feels done. Price agreed, structure sketched, closing date penciled in. Legally, almost none of it is done. Most LOI provisions are expressly non-binding: the price is a stated intention, the structure an outline. What actually binds is typically exclusivity (the owner agrees to stop talking to other acquirers, commonly for 45 to 90 days), confidentiality, and sometimes expense allocation. Notice the asymmetry. The owner gives up the one binding thing that matters, competition, in exchange for non-binding statements about everything else.
The purchase agreement is where those statements become enforceable, and where every term the LOI left vague gets negotiated with the owner’s alternatives gone. This is why experienced advisors push to settle more than price at the LOI stage: basket, cap, escrow, and survival terms agreed in two sentences before exclusivity are worth more than twenty pages of lawyering after it.
The document itself, a stock purchase agreement or asset purchase agreement depending on the structure chosen in Chapter 12, commonly runs 60 to 120 pages plus schedules and exhibits. Strip away the definitions and mechanics and it has five working parts: the purchase price and its adjustments, the representations and warranties, the covenants, the closing conditions, and the indemnification provisions. Each part answers one question: if something turns out to be different from what everyone believed at signing, who bears the cost? A purchase agreement is not a description of the deal. It is a risk-allocation machine, and every page either moves risk toward you or away from you.
Key point. The purchase agreement is a risk-allocation machine, not boilerplate. The price determines what you are owed; the reps, schedules, and indemnity terms determine what you keep. Owners who spend ten weeks negotiating price and ten days negotiating the agreement have the ratio backwards.
Reps and Warranties: The Statements That Carry the Risk
A representation and warranty is a statement of fact about the business, made in the agreement as of signing and usually repeated (“brought down”) as of closing. “The financial statements fairly present the financial condition of the company.” “There is no pending or threatened litigation.” “The company owns, or holds valid licenses to, all intellectual property used in the business.” A typical agreement contains 20 to 40 pages of them.
Reps do two jobs. The first is a truth-forcing function: an owner asked to certify in writing that there are no undisclosed liabilities develops an excellent memory, which is exactly the point. The second is risk allocation: if a rep proves false and the acquirer suffers a loss, the indemnification provisions make the owner pay for it. A rep is not a promise that nothing will go wrong. It is an agreement about who pays if something already was wrong and nobody knew.
The standard categories are stable across deals: organization and authority to sell; capitalization (who actually owns the equity); financial statements; absence of undisclosed liabilities; taxes; material contracts; litigation; intellectual property; employment and benefits; compliance with laws; environmental matters; customers and suppliers; insurance; and brokers’ fees. Acquirers draft them broad. Owners’ counsel narrows them. The narrowing happens through two dials that look like drafting trivia and are actually the whole negotiation.
The first dial is the knowledge qualifier. “There is no threatened litigation” makes the owner liable even for threats nobody had heard of. “To the knowledge of the owner, there is no threatened litigation” shifts the risk of the unknown to the acquirer. Even the word “knowledge” gets negotiated: whose knowledge counts (usually a defined list of named executives), and whether it means actual awareness or extends to what a person would have learned after reasonable inquiry.
The second dial is the materiality qualifier. “In compliance with all laws” is absolute; “in compliance in all material respects” filters out trivia. Acquirers often respond with a materiality scrape, a clause that reads the materiality qualifiers back out of the reps when calculating indemnifiable losses, on the theory that the basket (covered below) already filters small claims. Every qualifier added moves risk to the acquirer; every qualifier struck moves it back to the owner. That is why the reps take weeks: the parties are setting dozens of small dials, and the sum of the dial positions is a real number of dollars.
Disclosure Schedules: The Owner’s Shield
Every rep is really two documents. The rep in the agreement states the clean rule; the disclosure schedule lists the exceptions. “There is no litigation except as set forth on Schedule 3.11.” The agreement gets signed with dozens of these schedules attached: every material contract, every permit, every benefit plan, every open dispute, listed item by item.
Schedules are the unsung workhorse of the whole document, and for the owner they are the primary defense. A matter properly disclosed on a schedule generally cannot support an indemnity claim: the acquirer took it with open eyes and had the chance to price it. The same matter left off a schedule is a breach. This turns schedule preparation, tedious as it is, into the cheapest liability protection in the deal. Over-disclosure costs almost nothing. Under-disclosure costs exactly what the problem costs, plus legal fees, plus credibility. The practical rules: disclose specifically (vague references may not count as disclosure), disclose early (a schedule item raised in the first draft is diligence; the same item surfacing in week nine is a repricing event), and keep schedules updated between signing and closing under whatever notification mechanism the agreement sets.
Common pitfall. Thin disclosure schedules. An owner sprinting to close treats the schedules as an administrative chore and lets them be assembled from memory in a weekend. A customer dispute the owner considered resolved never makes Schedule 3.11. Fourteen months later it resurfaces as a claim, and what would have been a priced-in disclosure item becomes a six-figure indemnity payment out of escrow. Hours spent on schedules are the highest-return hours in the closing process.
Sandbagging needs a plain answer. The question: if an acquirer learns before closing that a rep is false, closes anyway, and then brings a claim for the breach, does it win? A pro-sandbagging clause says yes: the reps are a priced term of the contract, enforceable regardless of what the acquirer knew. An anti-sandbagging clause says no: no recovery for matters the acquirer actually knew about. Silence leaves the answer to state law, which varies; Delaware, the most common governing-law choice in these deals, has generally leaned toward enforcing the contract as written. Acquirers usually win this clause, and owners’ counsel usually trades it away for something better. The practical lesson is not to fight the doctrine but to moot it: nothing properly disclosed on a schedule can sandbag you.
Indemnification: Who Pays, How Much, For How Long
If the reps are the statements, indemnification is the enforcement. It answers four questions: who pays whom, for what, up to how much, and out of which pot. Four concepts do nearly all the work.
Survival periods set how long each rep can support a claim after closing. General reps commonly survive 12 to 18 months, long enough for one full audit and tax cycle to flush out problems. Fundamental reps (ownership of the equity or assets, authority to do the deal, capitalization, brokers’ fees) survive far longer, often to the statute of limitations, because an acquirer who did not actually get what it bought should not lose that claim on a timer. Tax reps typically track the tax statute of limitations. For the owner, survival is the answer to a personal question: when can I stop looking over my shoulder?
Baskets set the threshold below which the owner owes nothing, sparing both sides from disputes over small dollars. The distinction that matters is deductible versus tipping. A true deductible works like insurance: the owner pays only losses above the threshold. A tipping basket (also called a first-dollar basket) works differently: once cumulative losses cross the threshold, the basket “tips” and the owner pays from dollar one. On identical facts, the difference between the two is exactly the size of the basket, which is why counsel fights over one word. Baskets commonly run under 1% of purchase price, often with a per-claim minimum (a “mini-basket”) that keeps trivial items out of the count entirely.
Caps limit the owner’s total exposure. The general cap covers most rep breaches. Fundamental reps typically carry a higher cap, often the full purchase price. Fraud is carved out and uncapped everywhere: no agreement protects a deliberate lie.
Escrows and holdbacks are the funding source. An escrow places part of the price with a neutral agent at closing; a holdback simply lets the acquirer keep it. Owners should prefer escrow: when a claim is resolved in your favor, you collect from an escrow agent rather than chasing the acquirer. The escrow amount and release date are negotiated alongside the survival period, usually with the balance released when the general reps expire. The companion prize for owners is exclusive-remedy language, which makes the indemnity provisions (and, for general claims, often the escrow itself) the acquirer’s only recourse, instead of one option among lawsuits.
Where these terms actually settle is one of the few places real data exists. In the ABA’s 2025 Private Target Deal Points Study (139 acquisitions of private targets signed in 2024 and early 2025), reps and warranties did not survive closing at all in 41% of deals, up from 30% in the prior study, driven by the adoption of reps and warranties insurance. Where that insurance is used, median indemnity caps collapse to roughly a quarter of one percent of deal value. In traditional non-insured middle-market deals, indemnity caps around 10% of purchase price, escrows near 9-10%, and survival periods of 12-18 months remain common reference points (Seyfarth 2024/2025 Middle Market M&A SurveyBook). Read those numbers with care: the ABA sample skews toward deals meaningfully larger than most lower-middle-market transactions, where insured structures are routine. In the size range this guide covers, the traditional pattern (cap, basket, escrow, survival) is still the norm, and the figures above are reference points for negotiation, not entitlements.
One more mechanism matters whenever the consideration includes a seller note or an earnout: set-off rights, which let the acquirer withhold those future payments against pending indemnity claims. Unbounded set-off turns the owner’s deferred consideration into a hostage, and a creative acquirer can always find a claim. Owners should negotiate to prohibit set-off, cap it, or at minimum require that claims be finally resolved (agreed or adjudicated) before a payment is withheld.
The indemnity package is a small set of interlocking levers; the table below maps the architecture.
| Concept | What it does | The negotiation lever |
|---|---|---|
| Survival period | Sets how long each rep can support a claim after closing | Duration by category: general reps 12-18 months is a common reference point; fundamental and tax reps run longer |
| Basket | Sets the loss threshold before the owner owes anything | Size (commonly under 1% of price) and type: true deductible vs tipping (first-dollar once crossed) |
| Cap | Limits the owner’s total exposure for general rep breaches | Percentage of price; around 10% is a common non-insured reference point |
| Fundamental reps | A short list of core reps with higher caps and longer survival | Which reps make the list; acquirers try to promote ordinary reps onto it |
| Escrow / holdback | Sets aside part of the price as the funding source for claims | Size, duration, release schedule, and whether it is the exclusive remedy |
| Set-off rights | Lets the acquirer net claims against seller notes or earnouts | Prohibited, capped, or allowed only for finally resolved claims |
Illustrative example. An acquirer buys a company for $20 million. The agreement sets a general indemnity cap of 10% ($2M), a basket of 1% ($200,000), a 10% escrow ($2M), and 15-month survival on the general reps. Eight months after closing, a misclassified-contractor problem costs $120,000. That sits below the basket: no recovery, the acquirer absorbs it. At month 11, an undisclosed customer dispute settles for $180,000. Cumulative losses are now $300,000 and the basket has been crossed. If the basket is a tipping basket, the owner owes the full $300,000, paid out of escrow, because crossing the threshold triggers liability from the first dollar. If it is a true deductible, the owner owes only the $100,000 above the threshold. Same facts, same agreement, $200,000 apart: precisely the size of the basket. Either way the owner’s aggregate exposure for general claims stops at $2 million, the escrow funds the payment without a collection fight, and at month 15 the unclaimed balance is released to the owner.
Reps and Warranties Insurance
Reps and warranties insurance (RWI) moves most of this risk to an insurer. The acquirer buys a policy (most policies are buy-side, even when the owner effectively funds part of the premium), and rep breaches become insurance claims rather than clawbacks of the owner’s proceeds. The effect on deal architecture is dramatic: escrows shrink to a sliver, the owner’s post-close exposure for general reps largely disappears, and survival becomes a function of the policy rather than the agreement. That 41% of ABA-study deals with no post-closing rep survival is RWI at work.
The honest lower-middle-market caveat comes with the same data. RWI appeared in 63% of deals in the ABA’s 2025 Private Target Deal Points Study, but premiums typically run 3-4% of the coverage limit with retentions around 1-2% of deal value, and as a practical matter RWI rarely pencils below roughly $20-25 million in enterprise value. That is why many lower-middle-market deals still rely on traditional escrows and indemnities. If your deal is large enough to qualify, raise RWI at the LOI stage, because it reshapes the entire indemnity negotiation. And understand what it never covers: issues known before closing (everything on your disclosure schedules and everything found in diligence), purchase price adjustments such as the working capital true-up, and deal-specific exclusions the underwriters carve out. RWI insures the unknown. The known stays with the parties, which keeps the schedules just as important as ever.
Covenants, Conditions, and the MAC Clause
Many lower-middle-market deals sign and close on the same day, which makes this section short for them. When signing and closing are separated, usually to obtain third-party consents or complete financing, two more sets of provisions govern the gap.
Interim operating covenants require the owner to run the business in the ordinary course between signing and closing: no unusual contracts, no compensation changes, no asset sales, no settling litigation, often no capex beyond a threshold without the acquirer’s consent. These are reasonable in principle and suffocating when drafted badly. An owner still has a company to run, and Chapter 11’s warning holds: a business that drifts during the deal endangers the deal. Negotiate consent standards (“not to be unreasonably withheld”) and response deadlines so routine decisions do not queue behind the acquirer’s inbox.
Closing conditions define when the acquirer must actually wire funds: the reps remain accurate at closing (the bring-down), covenants have been performed, required consents have arrived, and no material adverse change has occurred. Consents deserve early attention: if key customer or lender contracts have change-of-control clauses, the parties must decide whose problem a withheld consent is, and that allocation belongs in the agreement, not in a closing-week scramble. Financing conditions, which let an acquirer walk if its lenders do, should be resisted and are one reason acquirer certainty gets screened before exclusivity.
The MAC (material adverse change) or MAE (material adverse effect) clause has a reputation that outruns its reality. It nominally lets an acquirer refuse to close if something devastating happens to the business before closing. In practice, courts have rarely allowed acquirers to walk on a MAC; the definitions are heavily negotiated, loaded with carve-outs for general economic conditions, industry-wide effects, and force-majeure-type events, and the threshold for “material” is high. Its real function is renegotiation pressure: an acquirer with cold feet invokes the specter of a MAC dispute to reopen price. Owners’ counsel fights over the carve-outs, and especially over “disproportionate impact” language that claws carve-outs back, precisely because the clause matters most as a bargaining position.
Where Your Negotiating Capital Belongs
A first-time owner staring at a 100-page draft wants to fight everything, and an acquirer’s first draft is written to invite it. Resist. Legal fees compound, momentum decays, and a counterparty watching you contest the notice provisions concludes you have never done this before. Most of the document genuinely is market: standard definitions, mechanical provisions, covenant language every deal carries. Experienced M&A counsel, and it must be M&A counsel, not the generalist who handles your leases, will tell you which twenty issues are worth raising and which five are worth real capital.
In a traditional non-insured deal, the five are these. First, the general cap: the single number that bounds your worst case. Second, the basket: size and, just as much, type, because tipping versus deductible is pure money. Third, survival periods: every month shorter is a month sooner your proceeds are truly yours. Fourth, the escrow: its size, its release schedule, and whether it is the exclusive remedy for general claims. Fifth, the fundamental-rep list: acquirers like to promote ordinary reps (intellectual property is the classic) onto the fundamental list, where higher caps and longer survival follow automatically; keeping that list short is worth more than winning any single qualifier. If your consideration includes a seller note or earnout, add set-off terms as the sixth.
Remember also that these terms trade against price and against each other. A larger escrow might buy a shorter survival period; a tighter cap might be worth a modest price concession; an acquirer who insists on a tipping basket can pay for it with a higher threshold. Negotiate the indemnity package as a package, the same expected-value habit Chapter 12 applied to the consideration stack. Owners who know which five levers matter, and arrive with schedules that leave nothing for those levers to catch, sign better agreements and sleep better afterward.
The Bottom Line
- The purchase agreement is a risk-allocation machine: the price sets what you are owed, and the reps, schedules, and indemnity terms decide what you keep.
- Settle the key indemnity terms (cap, basket, escrow, survival) in the LOI, while competition still exists; after exclusivity, every open term is negotiated from weakness.
- Disclosure schedules are the owner’s cheapest liability protection: what is properly disclosed generally cannot come back as a claim, whatever the sandbagging clause says.
- Know the difference between a tipping basket and a true deductible; on identical facts they diverge by exactly the basket amount.
- RWI has transformed indemnity terms in larger deals, but it rarely pencils below roughly $20-25 million in enterprise value, so most lower-middle-market deals still run on escrows and traditional indemnities.
- Spend negotiating capital on five terms (cap, basket, survival, escrow, the fundamental-rep list), hire specialist M&A counsel, and let the boilerplate be boilerplate.