Liquidated Damages Enforceability by State

The clause that quietly becomes a penalty

The canonical illustration of how an LD clause goes from "agreed damages" to "void penalty" in a single opinion is Lake River Corp. v. Carborundum Co., 769 F.2d 1284 (7th Cir. 1985). Judge Posner walked through a bagging-services contract where the minimum-guarantee clause would have paid Lake River $241,000 for handling that, on the actual facts, would have cost it roughly $107,000 to perform.

The clause looked like a damages estimate. Once Posner laid the numbers side by side, the disproportion was the whole opinion. The clause was a penalty. The plaintiff went back to actual damages.

That is the failure mode liquidated-damages drafting keeps reproducing. Not "is LD enforceable in my state," because in every state except those few with specific statutory carve-outs the answer is yes.

The real question is whether the number on the page survives the penalty test once a court reads it side-by-side with the harm it was supposed to cover. Get that ratio wrong and the clause is decoration.

This piece is a 50-state map of where LD clauses are enforced, where statutes have layered specific rules on top, and where the construction-contract regime diverges from the general commercial standard. The drafting checklist at the end is the version I have used in working commercial-litigation defense files, edited down from clauses that actually survived motion practice in Texas, Massachusetts, and California.

Are liquidated damages clauses enforceable? The short answer

Yes. A liquidated-damages clause is enforceable in all 50 states and the District of Columbia when it passes a two-prong test: the amount has to be a reasonable forecast of the harm, and actual damages have to be hard to estimate at the time the contract is signed. Fail either prong and the clause is treated as an unenforceable penalty, and the non-breaching party falls back to proving actual damages.

The split is in when courts measure reasonableness. Most states are presumptively single-look (the majority rule): they judge the clause only at contract formation. A minority are second-look (retrospective): they also compare the stipulated number to the actual harm at breach, and strike it if the two are wildly apart even when the forecast looked fine on signing day. Texas (Atrium Med. Ctr., LP v. Houston Red C LLC, 595 S.W.3d 188 (Tex. 2020)), New Jersey (Wasserman's Inc. v. Twp. of Middletown, 137 N.J. 238 (1994)), and Kentucky (retrospective, reaffirmed in Louisville & Jefferson Cnty. Metro. Sewer Dist. v. T&C Contracting, Inc. (Ky. 2018)) are verified second-look states. Massachusetts (Kelly v. Marx, 428 Mass. 877 (1999)) is the cleanest single-look statement. The quick-reference table and the full liquidated damages by state breakdown, with the controlling authority for each, are below.

TL;DR

• Liquidated-damages clauses are enforceable in every US state under some version of the Restatement (Second) of Contracts § 356 two-prong test: the stipulated amount must be reasonable in light of anticipated or actual loss, AND actual damages must be difficult to estimate at contract formation.
• The biggest single failure mode is the "disguised penalty," where the LD number bears no proportional relationship to the harm. Lake River Corp. v. Carborundum Co., 769 F.2d 1284 (7th Cir. 1985), is the cleanest published example. The clause defaults back to actual damages.
• Three states run distinct statutory regimes. California (Cal. Civ. Code § 1671) splits the analysis between consumer (§ 1671(d)) and commercial contracts (§ 1671(b)). New York applies UCC § 2-718 in goods cases and a common-law penalty test elsewhere. Florida applies the Lefemine two-prong test under its own appellate line.
• Most states layer a construction-specific liquidated-damages regime on top of the general test, typically a daily-delay rate that has to bear a reasonable relationship to actual project harm, and public-works projects are governed by separate state public-contracting statutes.
• A well-drafted clause survives in most bucket-1 states with the same core language; California, New York, and Florida need targeted carve-outs. The drafting checklist at the end of this post is the version I have seen hold up in contested motion practice.

Part of our all-50-states legal reference series.

The Restatement § 356 standard, and why it matters

Restatement (Second) of Contracts § 356(1) is the closest thing American contract law has to a national rule on liquidated damages: "Damages for breach by either party may be liquidated in the agreement but only at an amount that is reasonable in the light of the anticipated or actual loss caused by the breach and the difficulties of proof of loss. A term fixing unreasonably large liquidated damages is unenforceable on grounds of public policy as a penalty."

Two prongs. First, reasonableness, measured either at the time of contracting (anticipated loss) or at the time of breach (actual loss). Second, difficulty of estimation, the idea that LD is appropriate only where actual damages would be genuinely hard to prove.

The two prongs are conjunctive in most states and somewhat looser in a minority that read the second prong as a soft factor rather than a gate.

Section 356 is not a freedom-of-contract rule. It is a public-policy override. Courts will strike a number the parties expressly agreed to, even where both sides were sophisticated and represented by counsel, if the number functions as in terrorem leverage rather than a damages estimate.

That is the entire game. A drafter who treats the LD clause as a freedom-of-contract exercise will lose in motion practice. A drafter who treats it as a forecast of harm, documented and proportional, will win.

Single-look vs second-look: the split that decides close cases

The doctrine has two camps, and which one your forum sits in changes how you draft. The disagreement is about timing.

Single-look (prospective), the majority rule. The court tests reasonableness only at the time of contracting. If the number was a reasonable forecast on signing day, it holds, even if actual harm later came in much lower. Massachusetts is the cleanest statement: Kelly v. Marx, 428 Mass. 877 (1999), expressly rejected the second look in favor of judging the clause at formation, and the Supreme Judicial Court reaffirmed the single-look rule in Cummings Properties, LLC v. Hines, 492 Mass. 867 (2023). NPS, LLC v. Minihane, 451 Mass. 417 (2008), applied the same Kelly framework. Most states presumptively sit here.

Second-look (retrospective), the minority. The court also looks at actual damages at breach and strikes the clause if the stipulated number turns out to be wildly disproportionate, even when the forecast was reasonable when signed. Three states are verified second-look: New Jersey (Wasserman's Inc. v. Twp. of Middletown, 137 N.J. 238 (1994), which measures reasonableness at both contracting and breach), Texas (Atrium Med. Ctr., LP v. Houston Red C LLC, 595 S.W.3d 188 (Tex. 2020), where the challenger can defeat the clause by proving an "unbridgeable discrepancy" between stipulated and actual damages), and Kentucky (Louisville & Jefferson Cnty. Metro. Sewer Dist. v. T&C Contracting, Inc. (Ky. 2018), reaffirming Kentucky's retrospective approach).

The Seventh Circuit's Lake River analysis, applying Illinois law, ran the actual-versus-stipulated ratio in the same retrospective spirit, which is why it reads as the canonical penalty case. The practical takeaway: in a second-look state, a number that was defensible at signing can still be struck if reality diverges, so the recital and the contemporaneous damages model matter even more. For every other state, treat single-look as the default unless you confirm otherwise for your forum and contract type.

The three-bucket framework

Bucket 1: Strict reasonableness, the Restatement majority rule

The largest bucket, roughly 45 states, applies § 356 substantially as written. The two-prong test (reasonableness plus difficulty of estimation) is the workhorse, and the controlling appellate authority in each state typically restates it in similar language.

Texas (Phillips v. Phillips, 820 S.W.2d 785 (Tex. 1991)), Illinois (XCO Int'l Inc. v. Pac. Scientific Co., 369 F.3d 998 (7th Cir. 2004), applying Illinois law), Pennsylvania (Pantuso Motors, Inc. v. CoreStates Bank, N.A., 568 Pa. 601 (2002)), Massachusetts (Kelly v. Marx, 428 Mass. 877 (1999)), and most of the rest run this framework with no statutory overlay.

The intra-bucket variation is in emphasis. Texas and Pennsylvania put heavy weight on the difficulty-of-estimation prong and will strike a clause if actual damages could have been reasonably forecast at contracting; Phillips expressly conditions enforcement on the harm being incapable of accurate estimation.

Massachusetts and Illinois weigh the reasonableness prong harder; Kelly v. Marx explicitly rejected the "second look" approach in favor of testing the clause at the time of contracting. The drafting consequence is small: a clause documented to satisfy both prongs is enforceable in every bucket-1 state.

How the bucket-1 penalty analysis actually reads on the page is worth understanding, because the move that loses cases is predictable. Judges open the contract to the LD clause, find the number, then look at the actual or anticipated damages and run a ratio.

If the LD is roughly proportional, the clause holds; the analysis stops at the first prong. If the LD is wildly larger than the harm (the Lake River posture, where the contract would have paid the plaintiff more than twice its actual cost of performance), the disproportion alone gets the clause struck and the second prong is rarely reached.

If the LD is roughly proportional but actual damages are easy to calculate (the bucket-1 second-prong attack), Texas and Pennsylvania will strike on summary judgment; Massachusetts and Illinois will usually let the clause stand. Knowing which prong your forum weighs, and whether it takes a second look, is half the drafting exercise.

Bucket 2: Specific statutory regimes

Three states have layered statutes that change the analysis materially.

California. Cal. Civ. Code § 1671 splits LD enforcement on the type of contract. Section 1671(b) governs commercial contracts and provides that an LD clause is "valid unless the party seeking to invalidate the provision establishes that the provision was unreasonable under the circumstances existing at the time the contract was made." This is a freedom-of-contract presumption that flips the burden onto the party challenging the clause.

Section 1671(d) governs consumer contracts (residential leases, retail purchases, contracts for personal/family/household use) and presumes the LD clause is void unless the party seeking enforcement proves it was impractical or extremely difficult to fix actual damages. Two regimes, one statute, opposite default rules.

In California, identify which subsection applies before you write the clause. A consumer-facing SaaS terms-of-service is a different drafting exercise than a B2B vendor MSA, and the same LD number behaves differently in each.

The California Supreme Court's analysis in Ridgley v. Topa Thrift & Loan Ass'n, 17 Cal. 4th 970 (1998), held a late-charge clause unenforceable under § 1671(d) where the lender could not prove difficulty of estimation, and is the cleanest illustration of how the consumer subsection actually plays out at the trial level.

New York. Two layers. NY UCC § 2-718(1) governs sales of goods and adopts the Restatement-style test with the additional language that "a term fixing unreasonably large liquidated damages is void as a penalty."

For non-goods contracts, New York applies a parallel common-law test articulated most clearly in Truck Rent-A-Center, Inc. v. Puritan Farms 2nd, Inc., 41 N.Y.2d 420 (1977). The two prongs are similar but New York courts read the difficulty-of-estimation prong strictly and will strike clauses where actual damages were calculable.

Florida. Florida applies its own two-prong test, controlled by Lefemine v. Baron, 573 So. 2d 326 (Fla. 1991), which restates § 356 in Florida-specific language. The Florida overlay matters most in construction, where the state's public-works statutes add daily-delay damage rules on top of Lefemine.

Bucket 3: Public-policy override

A handful of jurisdictions will void LDs viewed as penalty even where the clause facially satisfies § 356. This is not a separate test; it is an aggressive application of the penalty doctrine, usually triggered by a glaring disproportion between the LD number and the actual harm at breach.

Lake River is the federal-court canonical example applying Illinois law. Wassenaar v. Towne Hotel, 111 Wis. 2d 518 (1983), is the cleanest state-court illustration: the Wisconsin Supreme Court struck an employment-contract LD as a penalty under a totality-of-circumstances analysis that looked specifically at the mitigation-of-damages question.

The lesson for drafters is the same in every state: even where the local rule is freedom-of-contract-friendly, a number that looks like leverage rather than a forecast will get struck.

Construction contracts: the separate regime in most states

Construction is where LD doctrine gets genuinely state-specific. A substantial majority of states run a construction-contract overlay that applies in addition to (and sometimes instead of) the general commercial test.

The shape is consistent. Construction LDs are typically expressed as a daily-delay rate ($2,500 per day of delay past substantial completion is a common example), and the enforceability question becomes whether that rate bears a reasonable relationship to actual project harm: extended general conditions, lost use of the facility, lender carry costs, lost revenue from delayed occupancy.

The leading appellate authority in most states is a construction-specific case, not the general commercial line.

Public-works projects add a second layer. Most states' public-contracting statutes prescribe LD frameworks for state and municipal construction (often through the state's general procurement code), and many of those statutes incorporate AIA-form contract provisions by reference.

The federal public-works equivalent is FAR 52.211-12, the standard liquidated-damages clause for federal construction contracts, with its supporting cost-principles framework at FAR Part 11.5.

For a corporate lawyer drafting a private commercial construction contract, confirm two things: the state's general LD test (almost always bucket 1 or bucket 2 above), and whether the state has a construction-specific appellate line that adds additional gates. Most do.

The construction-specific authority is what controls the daily-delay rate's enforceability, and the math the court runs is whether the daily rate can be tied to extended general conditions and documented delay damages.

What actually gets clauses struck: language to flag on every redline

Three patterns predict almost every LD strike in the published opinions. They are also the cheapest things to fix at the redline stage.

"As a penalty" or "by way of penalty." The single word "penalty" appears in a striking number of form clauses, usually as legacy language no one updated. Courts treat it as a near-confession.

The fix is mechanical: change "penalty" to "liquidated damages" or "agreed damages" on every pass. If opposing counsel insists on "penalty" language during negotiation, that is the moment to push back hard; you are negotiating against your own clause.

A single number for "any breach." When the LD covers "any breach of this Agreement," the court has to assume it covers every breach equally, from a missed status meeting to total non-performance. That is the structural disproportion that doomed the Lake River clause. Tiered LDs scaled to specific breach categories survive; flat numbers across all breach types rarely do.

LD numbers that exceed the contract value. This is the single most reliable predictor of a strike. Courts run the ratio in the first paragraph of the analysis, and a number larger than the deal is the strongest possible evidence of a penalty.

Cap the LD at total contract value or a defensible fraction of it. If the documented harm legitimately exceeds the contract value (rare but possible in some IP-licensing or healthcare data contexts), put that documentation in the recitals; otherwise cap.

A pattern worth flagging because it does not show up in checklists: courts read contract recitals as substantive evidence of the parties' damages forecasting. Wassenaar v. Towne Hotel makes this explicit; Wasserman's Inc. v. Twp. of Middletown relies heavily on the parties' reasoning at contracting.

A clause with two sentences of "the parties acknowledge that actual damages from [specific breach] would be difficult to ascertain because [specific reason]" is meaningfully stronger than the same number without that recital. The recital is the single highest-leverage edit a drafter can make, and it is the edit form-library clauses almost never include.

The corollary point about contemporaneous damages modeling: a clause whose number can be matched to a spreadsheet, an underwriting memo, or a board-deck slide drafted around the same time as the contract is meaningfully harder to strike.

In commercial-litigation defense, the first discovery request after a penalty challenge is "produce all documents reflecting the parties' analysis of the LD amount at contracting." Documents that show a real damages forecast (extended general conditions, lost-revenue model, fraud-investigation cost analysis) win these motions.

Absence of any contemporaneous documentation is what plaintiffs' counsel uses to argue the number was arbitrary. The drafting move that follows: when LD numbers are being negotiated, save the modeling in a file you can produce later, and reference it in the recital.

Draft this in Vaquill AIBook a demo

The full 50-state list

Bucket assignment plus the controlling authority or statute, with a flag for construction-specific appellate authority where I have located it. A few methodology notes before the list.

"Construction overlay applies" means I have identified state appellate or controlling federal authority specifically addressing construction-contract LDs in that state; absence of the flag does not mean none exists, only that the dominant analysis tracks the general commercial line.

"Limited direct appellate authority" means the state supreme court has not squarely addressed contractual LDs in recent decades and federal courts sitting in diversity have applied the Restatement framework, which is the most defensible reading. Treat the list as a starting map; confirm current authority for your forum and contract type before citing anything.

On timing, I flag "second-look" only where I have located controlling authority that compares the stipulated number to actual damages at breach. The other states are presumptively single-look under the majority rule (reasonableness judged at contracting), but a few have language that could support a retrospective check, so confirm before relying on a single-look posture in litigation.

Quick reference: the states that need special handling

The states below depart from the plain Restatement § 356 majority rule, either with a statute or with a verified second-look timing rule. Everything not listed here sits in bucket 1 and is presumptively single-look; see the full list for each state's controlling authority.

• Alabama. Bucket 1. Sutton v. Epperson, 631 So. 2d 832 (Ala. 1993). Construction overlay applies.
• Alaska. Bucket 1. Restatement § 356 framework; limited direct appellate authority. Construction overlay applies.
• Arizona. Bucket 1. Pima Sav. & Loan Ass'n v. Rampello, 168 Ariz. 297 (Ct. App. 1991). Construction overlay applies.
• Arkansas. Bucket 1. Alley v. Rodgers, 269 Ark. 262 (1980). Construction overlay applies.
• California. Bucket 2. Cal. Civ. Code § 1671(b) (commercial) and § 1671(d) (consumer). Two regimes split by contract type. Construction overlay applies, with substantial public-works statutory layering.
• Colorado. Bucket 1. Rohauer v. Little, 736 P.2d 403 (Colo. 1987). Construction overlay applies.
• Connecticut. Bucket 1. Hanson Dev. Co. v. East Great Plains Shopping Ctr., Inc., 195 Conn. 60 (1985). Construction overlay applies.
• Delaware. Bucket 1. Brazen v. Bell Atl. Corp., 695 A.2d 43 (Del. 1997) (M&A termination-fee context). Heavy weight on sophistication of commercial parties.
• Florida. Bucket 2. Lefemine v. Baron, 573 So. 2d 326 (Fla. 1991). Construction overlay applies, with significant public-works statutory framework.
• Georgia. Bucket 1, with statutory codification at O.C.G.A. § 13-6-7. Southeastern Land Fund, Inc. v. Real Estate World, Inc., 237 Ga. 227 (1976). Construction overlay applies.
• Hawaii. Bucket 1. Restatement framework; limited direct appellate authority.
• Idaho. Bucket 1. Graves v. Cupic, 75 Idaho 451 (1954). Construction overlay applies.
• Illinois. Bucket 1. XCO Int'l Inc. v. Pac. Scientific Co., 369 F.3d 998 (7th Cir. 2004) (applying Illinois law); Lake River Corp. v. Carborundum Co., 769 F.2d 1284 (7th Cir. 1985). Construction overlay applies.
• Indiana. Bucket 1. Gershin v. Demming, 685 N.E.2d 1125 (Ind. Ct. App. 1997). Construction overlay applies.
• Iowa. Bucket 1. Rohlin Constr. Co. v. City of Hinton, 476 N.W.2d 78 (Iowa 1991). Construction overlay applies and Iowa appellate authority is unusually developed.
• Kansas. Bucket 1. White Lakes Shopping Ctr., Inc. v. Jefferson Standard Life Ins. Co., 208 Kan. 121 (1971). Construction overlay applies.
• Kentucky. Bucket 1, second-look (retrospective). Louisville & Jefferson Cnty. Metro. Sewer Dist. v. T&C Contracting, Inc. (Ky. 2018), reaffirming Kentucky's retrospective approach; United Servs. Auto. Ass'n v. ADT Sec. Servs., Inc., 241 S.W.3d 335 (Ky. Ct. App. 2006). Construction overlay applies. (Reporter page for the 2018 opinion not independently confirmed here; verify before citing.)
• Louisiana. Bucket 1, civil-law overlay. La. Civ. Code art. 2005 governs "stipulated damages" and uses the civilian "manifestly unreasonable" standard rather than the common-law penalty test. Construction overlay applies.
• Maine. Bucket 1. Interstate Indus. Uniform Rental Serv., Inc. v. F. R. Lepage Bakery, Inc., 413 A.2d 516 (Me. 1980).
• Maryland. Bucket 1. Bd. of Educ. of Talbot Cnty. v. Heister, 392 Md. 140 (2006). Construction overlay applies.
• Massachusetts. Bucket 1, single-look (prospective). Kelly v. Marx, 428 Mass. 877 (1999), reaffirmed in Cummings Properties, LLC v. Hines, 492 Mass. 867 (2023). One of the cleaner single-look statements in the country. Construction overlay applies.
• Michigan. Bucket 1. St. Clair Med., P.C. v. Borgiel, 270 Mich. App. 260 (2006). Construction overlay applies.
• Minnesota. Bucket 1. Gorco Constr. Co. v. Stein, 256 Minn. 476 (1959). Construction overlay applies.
• Mississippi. Bucket 1. Limited direct appellate authority; federal courts apply Restatement framework. Construction overlay applies.
• Missouri. Bucket 1. Grand Bissell Towers, Inc. v. Joan Gagnon Enters., Inc., 657 S.W.2d 378 (Mo. Ct. App. 1983). Construction overlay applies.
• Montana. Bucket 1. Restatement framework; limited direct appellate authority.
• Nebraska. Bucket 1. Growney v. C M H Real Estate Co., 195 Neb. 398 (1976). Construction overlay applies.
• Nevada. Bucket 1. Joseph F. Sanson Inv. Co. v. 268 Ltd., 106 Nev. 429 (1990). Construction overlay applies.
• New Hampshire. Bucket 1. Orr v. Goodwin, 157 N.H. 511 (2008).
• New Jersey. Bucket 1, second-look (retrospective). Wasserman's Inc. v. Twp. of Middletown, 137 N.J. 238 (1994), reaffirmed in MetLife Capital Fin. Corp. v. Washington Ave. Assocs. L.P., 159 N.J. 484 (1999). The court measures reasonableness at both contracting and breach. Construction overlay applies.
• New Mexico. Bucket 1. Restatement framework; limited direct appellate authority.
• New York. Bucket 2. NY UCC § 2-718(1) for sales of goods; Truck Rent-A-Center, Inc. v. Puritan Farms 2nd, Inc., 41 N.Y.2d 420 (1977), for non-goods contracts. Difficulty-of-estimation prong applied strictly. Construction overlay applies.
• North Carolina. Bucket 1. Knutton v. Cofield, 273 N.C. 355 (1968). Construction overlay applies.
• North Dakota. Bucket 1. N.D. Cent. Code § 9-08-04 codifies the rule that contracts fixing damages for breach are void with exceptions for cases where actual damages are impracticable to fix. Statutory codification but bucket-1 in operation.
• Ohio. Bucket 1. Lake Ridge Acad. v. Carney, 66 Ohio St. 3d 376 (1993). Construction overlay applies.
• Oklahoma. Bucket 1. Sun Ridge Inv'rs, Ltd. v. Parker, 956 P.2d 876 (Okla. 1998). Construction overlay applies.
• Oregon. Bucket 1. Illingworth v. Bushong, 297 Or. 675 (1984). Construction overlay applies.
• Pennsylvania. Bucket 1. Pantuso Motors, Inc. v. CoreStates Bank, N.A., 568 Pa. 601 (2002). Construction overlay applies.
• Rhode Island. Bucket 1. Psaty & Fuhrman, Inc. v. Hous. Auth., 76 R.I. 87 (1949). Construction overlay applies and the appellate line is notably old but still good law.
• South Carolina. Bucket 1. Tate v. LeMaster, 231 S.C. 429 (1957). Construction overlay applies.
• South Dakota. Bucket 1. S.D. Codified Laws § 53-9-5 codifies the LD framework. Construction overlay applies.
• Tennessee. Bucket 1. Guiliano v. Cleo, Inc., 995 S.W.2d 88 (Tenn. 1999). Construction overlay applies.
• Texas. Bucket 1, second-look (retrospective). Phillips v. Phillips, 820 S.W.2d 785 (Tex. 1991) (two-prong test), as applied in Atrium Med. Ctr., LP v. Houston Red C LLC, 595 S.W.3d 188 (Tex. 2020), where a challenger can defeat the clause by proving an unbridgeable discrepancy between stipulated and actual damages. Heavy weight on the difficulty-of-estimation prong. Construction overlay applies, and Texas has one of the more developed construction-LD appellate records.
• Utah. Bucket 1. Reliance Ins. Co. v. Util. Trailer Sales of Salt Lake, Inc., 690 P.2d 545 (Utah 1984). Construction overlay applies.
• Vermont. Bucket 1. Restatement framework; limited direct appellate authority.
• Virginia. Bucket 1. Brooks v. Bankson, 248 Va. 197 (1994). Construction overlay applies.
• Washington. Bucket 1. Watson v. Ingram, 124 Wash. 2d 845 (1994). Construction overlay applies.
• West Virginia. Bucket 1. Wheeling Clinic v. Van Pelt, 192 W. Va. 620 (1994). Construction overlay applies.
• Wisconsin. Bucket 1. Wassenaar v. Towne Hotel, 111 Wis. 2d 518 (1983). Construction overlay applies.
• Wyoming. Bucket 1. Restatement framework; limited direct appellate authority.

District of Columbia sits in bucket 1 and applies the Restatement § 356 framework.

Common drafting failures

Across the contested-LD cases I have read in the last decade, four failure modes do most of the work.

The penalty disguised as LD. Round numbers ($100,000, $500,000, $1,000,000) and identical amounts for any breach are the giveaway. A clause that fixes the same number for late delivery, defective performance, and total non-performance is not a damages estimate; it is leverage.

Courts catch this on the first read. The Lake River numbers ($241,000 stipulated for what would have cost $107,000 to perform) are the published illustration.

One-size-fits-all clauses across multiple breach types. A vendor MSA that applies the same LD to any breach, regardless of whether the breach was a missed milestone, a confidentiality violation, or a failure to perform, will struggle in motion practice. Different breach types cause different damages. The drafted clause should reflect that.

No nexus to actual damages. Courts run the proportionality math and strike clauses where the ratio is absurd. The cleanest fix is documenting the rationale for the number in the contract recitals: "the parties have determined that actual damages from breach of [specific clause] would be difficult to ascertain because [specific reason], and have agreed that [amount] is a reasonable estimate of such damages."

Stacked damages. LD plus actual damages plus lawyers' fees, all from the same breach, is the move that loses you the entire clause. LD is an alternative to actual damages, not a supplement. Stacking either gets the clause struck or gets the actual-damages award reduced by the LD amount, and either way the drafting effort was wasted.

The drafting checklist

Every state enforces a reasonable LD clause under Restatement § 356; three buckets (majority rule, statutory regimes, public-policy override) decide how the penalty test is run.

The clause that has survived in every bucket above looks substantially the same. The structure that holds up:

1. Tie the amount to a difficult-to-prove harm. Identify the specific breach the LD covers and state why actual damages would be hard to calculate. Lost goodwill, brand harm, lost use of a facility, lender carry costs in a construction context are all defensible. "Damages from any breach of this Agreement" is not.
2. State the rationale in the contract recitals. Two sentences that document the parties' reasoning for the number. This is the single highest-leverage edit a drafter can make. Courts read it. It satisfies the difficulty-of-estimation prong on its own in most jurisdictions.
3. Different amounts for different breaches. Tiered LDs scaled to the harm of each breach type look like damages forecasts, not penalties. A flat number for any breach looks like leverage.
4. Cap at total contract value. The math test courts run first: is the LD bigger than the contract? If yes, the disproportionality presumption kicks in. Capping the LD at contract value forecloses the easiest motion-practice attack.
5. Avoid the word "penalty." It sounds obvious, and it matters. Some form clauses still use "penalty" loosely; strike it on every redline. Use "liquidated damages" or "agreed damages" instead.
6. Carve-out for fraud and willful misconduct. LD should not cap damages for intentional bad acts. The carve-out is standard, and most courts will read it in even if it is missing, but the better practice is explicit drafting.

How that plays across deal types in practice.

SaaS vendor MSA. Three potential breach types. Late delivery of a milestone: LD of $5,000 per business day past the deadline, tied to documented lost productivity of the customer's deployment team. Data confidentiality breach: LD of $50,000 per incident, tied to documented investigation, notification, and remediation costs, with a recital citing average per-record breach costs from a named source.

Total non-performance: refund of fees paid for the breaching service tier, plus a 20% premium representing transition costs. The tiered structure tracks Phillips in Texas and survives § 1671(b) in California because the bargaining position is commercial-to-commercial.

Construction subcontract. Daily-delay rate tied to documented general-conditions cost. The math the court will run is set out in a recital: extended general conditions of $3,200 per day, lost use of the owner-occupied facility of $4,000 per day, totaling roughly $7,000 daily. The LD is set at $5,000 per day, below the documented harm. That is the Watson v. Ingram posture in Washington and the Phillips posture in Texas, and it survives.

UCC § 2-718 goods supply. Limited-quantity custom-manufactured goods. The LD is a percentage of contract value (20%) for buyer's wrongful rejection, with a recital documenting the seller's reasonable forecast of resale loss and restocking cost. § 2-718(1)'s "reasonable in light of anticipated or actual harm" prong is satisfied on the recital alone; the difficulty prong is satisfied because the goods are custom and have a thin secondary market.

California consumer-facing terms. § 1671(d) inverts the burden, so the recital does double work here. Document the rationale and the difficulty of estimation in the terms themselves, and tie the LD number to a specific, defensible cost (e.g., chargeback fees, fraud investigation cost) rather than a flat dollar figure. Ridgley is the warning shot.

The clause that does not survive is the one with a single round number and no rationale. The clause that does survive is the one that reads like a damages forecast on its face.

In the contested-LD cases I have seen go to motion practice, that distinction wins or loses the clause in the first ten minutes of oral argument.

FAQ

Are liquidated damages clauses enforceable in every state? Yes. All 50 states and D.C. enforce a liquidated-damages clause that passes the two-prong test: a reasonable forecast of the harm plus genuine difficulty estimating actual damages at contract formation. No state bans LD clauses outright. What fails is the clause that reads as a penalty rather than a damages estimate.

What is the two-prong test for liquidated damages? First, the stipulated amount has to be reasonable in light of the anticipated or actual loss. Second, actual damages have to be hard to prove or estimate when the contract is signed. Both prongs trace to Restatement (Second) of Contracts § 356(1). Most states treat them as conjunctive, so failing either prong sinks the clause.

What is the difference between liquidated damages and a penalty? Liquidated damages compensate for a loss that is hard to measure; a penalty punishes the breaching party or pressures performance. Courts run a proportionality check: if the number bears no reasonable relationship to the harm (the Lake River posture, where the clause would have paid more than twice the cost of performance), it is a void penalty and the non-breaching party falls back to actual damages.

Which states use a single-look versus a second-look test? Most states are single-look (prospective), judging the clause only at contract formation, with Massachusetts (Kelly v. Marx) the cleanest statement. A verified minority take a second look at actual damages at breach: Texas (Atrium Med. Ctr. v. Houston Red C), New Jersey (Wasserman's v. Middletown), and Kentucky (Louisville MSD v. T&C Contracting). Treat single-look as the default for any state not confirmed second-look.

Are liquidated damages enforceable in California? Yes, but the rule splits by contract type under Cal. Civ. Code § 1671. Section 1671(b) makes the clause valid in commercial contracts unless the challenger proves it was unreasonable at formation. Section 1671(d) presumes the clause void in consumer contracts unless the party enforcing it proves actual damages were impractical to fix. Identify the subsection before drafting.

Can you recover liquidated damages and actual damages for the same breach? No. Liquidated damages substitute for actual damages; they do not supplement them. Stacking LD plus actual damages plus the same loss usually gets the clause struck or the award offset by the LD amount. Carve out fraud and willful misconduct so those stay uncapped, but do not draft the LD on top of compensatory damages.

How do you draft a liquidated damages clause that holds up? Tie the number to a specific, hard-to-measure harm, document the rationale in two sentences of recital, use tiered amounts for different breach types instead of one flat number, cap the LD at or below total contract value, and never use the word "penalty." Save the contemporaneous damages model so you can produce it if the clause is challenged.

For related state-by-state coverage, see Indemnification Caps by State: How 50 Jurisdictions Read the Same Clause, Jury-Trial Waivers: Enforceable in All 50 States?, Attorneys' Fees Provisions by State: When Contractual Fee-Shifting Survives, and Choice-of-Law Clauses That Quietly Fail: A State-by-State Breakdown. All sit in our all-50-states legal reference series.

When you are checking a draft LD number against the right state's test before signature, Vaquill AI's drafting workbench runs that state-by-state proportionality and penalty check against the clause in seconds.

Arshita Anand

Co-Founder & CEO · Attorney

Arshita leads product and strategy at Vaquill, building the legal AI suite that solo, small-firm, and in-house US lawyers use to run a matter end to end.