A performance test has a price. Most operators never compute it.
They treat the test as a threat to survive, not a number to manage. But the price is knowable. It often runs to six or seven figures. And it turns on something most operators never think to check: how replaceable they are.
The companion piece, Performance Tests as Option Contracts, set up the structure. A performance test is an option the owner holds — the right, not the duty, to replace the operator when the test fails. The owner uses it rarely, because using it is expensive. In calibration research, a GOP prong that failed about 92% of the time produced an estimated owner termination rate of about 11%. That gap is the signature of an option: the owner holds the right and pulls the trigger only when the payoff clears the cost.
The option is legitimate. An owner needs a real exit against genuine underperformance, and the performance test is how the contract provides one. This piece takes that as given. The question isn’t whether the owner should hold the right. It’s what the right is worth — and what the operator’s exposure to it costs, in dollars.
The answer splits in two, on two separate ledgers. The owner’s payoff from firing, which explains why firing is rare and why the rate changes with the operator. And the operator’s exposure, which is the dollar the operator actually carries and negotiates. Most operators negotiate blind to both. And for the same deal, an easy-to-replace operator can carry several times the exposure of a hard-to-replace one — not because it loses more when fired, but because it’s fired more often.
The Framework: Two Ledgers, Two Numbers
A performance test is a real option, but not a traded one. There’s no market price and no clean Black-Scholes answer. The valuation is a transparent expected-value walk, not false precision.
The real-options literature gives the intuition. Trigeorgis, in Real Options: Managerial Flexibility and Strategy in Resource Allocation (1996), and Dixit and Pindyck, in Investment Under Uncertainty (1994), make the core point: an option holder facing a near-certain cost and an uncertain payoff will usually hold rather than exercise. That’s why failure is common but firing is rare. The owner carries the option across many failures and acts only when a specific case makes the payoff big enough to beat the cost.
The valuation produces two numbers, computed separately.
The Owner’s Ledger — Why the Owner Fires, or Usually Doesn’t
The owner’s decision to fire is a cost-benefit call.
The benefit is real but uncertain. The owner hopes a better operator lifts performance over the remaining term. That’s a hope, not a given. The owner has to believe a replacement will do materially better — and that belief has to survive the property’s constraints: location, physical product, market, comp set, and whether a credibly better operator will even take the deal. If the contract makes key money non-refundable on a for-cause exit, the owner also keeps that. The retained key money is certain if firing happens. The performance lift isn’t.
The cost is large and near-certain. The big one is the transition valley — 12 to 24 months of depressed performance after any operator change, as accounts, group pipelines, rankings, and staffing reset under the new manager. It’s a well-documented cost in the practitioner literature on HMA terminations (Pryor; DLA Piper). On top of it sit search and installation costs: finding the replacement, negotiating the new HMA, running the transition.
The owner fires only when the expected benefit beats the cost. Because the benefit is uncertain and the cost near-certain and large, the math usually says don’t. That’s the 92%/11% gap. The test fails 92% of the time. The owner fires 11% of the time. The difference is the cost-benefit call coming out negative in most cases.
And the rate changes with the operator. A hard-to-replace operator — a brand on a complex asset, a specialist, an operator with deep market presence — leaves a deeper valley and a smaller credible upside. The owner can’t easily find someone to match the incumbent, let alone beat it, because the incumbent’s edges (brand distribution, corporate accounts, systems, staff) are hard to copy. The firing rate is very low. An easy-to-replace operator — select-service, smaller regional, commoditized — leaves a shallower valley and a more credible upside. The owner can find a plausibly better replacement. The firing rate is higher.
The owner’s ledger sets the firing rate. That rate feeds the operator’s number.
The Operator’s Ledger — The Number the Operator Carries
The operator’s exposure is the expected value of what it loses if fired for cause.
What’s lost if fired. The operator gives up the remaining base fees and whatever incentive it would have earned across the rest of the term. That’s the certain core. If the contract makes key money non-refundable on a for-cause exit, the operator also forfeits the unamortized balance. That structure isn’t universal, but where it exists it raises the loss a lot.
And the operator collects no liquidated damages on a for-cause exit. In HMAs, LD and termination fees run owner to operator, and they trigger on convenience or sale — the owner-exit cases, not operator fault (DLA Piper; JMBM). A performance-test firing is for cause. No LD. The operator just loses the remaining fees, plus any unamortized key money.
The exposure itself is three things multiplied together: how often the test fails, how often the owner fires given a failure, and how much the operator loses if fired.
How often the test fails comes from the threshold set against the property’s realistic performance. How often the owner fires given a failure is the archetype rate from the owner’s ledger — it already includes the transition valley, so the operator doesn’t count the valley again; the operator doesn’t pay for the new manager’s ramp. And the loss if fired is the remaining fees plus any unamortized key money.
That exposure shrinks across the term as the remaining fees shrink. It isn’t a flat yearly number times the test years. In Year 3 the operator has seven years of fees at risk; in Year 8, two. Year 8 exposure is much lower than Year 3.
The Two Ledgers Together
The owner’s ledger sets the firing rate. The operator’s exposure uses that rate. For the same deal, the easy-to-replace operator carries more exposure — not because it loses more when fired (the loss is the same), but because it’s fired more often. Termination risk scales with replaceability.
A Worked Example: Two Archetypes
Take an illustrative property: a 180-room select-service hotel on a 10-year HMA. Base fee 3% of revenue. A modest, partly earnable incentive. Performance test: GOP margin at or above 31%, tested each year from Year 3. Key money: $500,000, straight-line over 10 years, non-refundable on a for-cause exit. From underwriting, the property’s GOP margin centers around 29%, clearing 31% in strong years. Revenue: $7 million a year, stable.
How often the test fails
The Monte Carlo puts GOP margin below 31% in about 65% of test years. So the test fails roughly two years in three. That’s this property’s own distribution — different from the 92% calibration in the opening, which was a different property, threshold, and performance band.
How often the owner fires — by archetype
Hard-to-replace operator. A brand on a complex asset. The owner’s upside is modest and uncertain — the brand already has the systems, presence, and distribution a replacement would struggle to match. The cost is a deep valley: 18 to 24 months of depressed performance, an illustrative 18% revenue dip recovering slowly, plus search and installation — call it $2.0 to $2.8 million all in. The benefit rarely beats that. Firing rate: about 6% of failures.
Easy-to-replace operator. A select-service regional operator. The owner’s upside is more credible — a larger or sharper operator might really do better, and the incumbent’s edges are shallower. The valley is smaller: 12 to 15 months, an illustrative 12% dip recovering faster, with lower search costs because replacements are easy to find — call it $1.2 to $1.6 million. The benefit beats that more often. Firing rate: about 16% of failures.
The 11% from the opening sits between the two. It’s not a constant. It’s one property’s archetype rate.
What’s lost if fired
Say the firing lands in Year 5, with five years left. Remaining base fees: 3% of $7 million for five years, about $1.05 million. Remaining earnable incentive: modest, call it $150,000. Unamortized key money, non-refundable on for-cause: half of $500,000, so $250,000. Total loss if fired: about $1.45 million.
The exposure, each archetype
Hard-to-replace. Multiply it out: a 65% failure rate, a 6% firing rate, and a $1.45 million loss give about $56,500 of exposure in that year. Across the eight test years (Years 3 to 10) the term total is larger — but it doesn’t just multiply by eight, because the loss shrinks each year as fewer fees remain: seven years at risk in Year 3, one by Year 9. Rolled up over that declining stream, the term exposure runs on the order of $400,000. A rough figure — each year’s inputs are estimates.
Easy-to-replace. Same walk, higher firing rate: a 65% failure rate, a 16% firing rate, the same $1.45 million loss give about $151,000 in that year. Rolled up at the higher rate across the eight test years, the term exposure runs on the order of $1 million. Again approximate — and again driven by the firing rate, not the loss, which is identical to the hard-to-replace case.
The punchline
Same deal. The easy-to-replace operator carries about 2.5 times the exposure of the hard-to-replace one. Not because it loses more when fired — the loss is identical. Because it’s fired more often: 16% versus 6%. Termination risk scales with replaceability.
Why does the firing rate differ? The hard-to-replace operator leaves a deeper, near-certain valley and a smaller credible upside, so the owner rarely fires. The easy-to-replace operator leaves a shallower valley and a more credible upside, so the owner fires more. The operator’s exposure tracks that difference directly.
How the Levers Move the Exposure
Different terms move the exposure by different amounts. Some are high-impact, some aren’t. Knowing which is which changes the negotiation.
Threshold
Drop the GOP threshold from 31% to 29% and the failure rate falls hard — for this property, from about 65% to about 25%. The exposure falls by more than half. One point of threshold is worth more than most fee concessions. It’s the highest-impact lever in most deals, because it moves the failure rate directly, and the failure rate is the first thing the exposure multiplies. A threshold two points below the property’s realistic band erases most of the exposure. Two points above it creates exposure even a hard-to-replace operator can’t ignore.
Key money
Non-refundable key money on a for-cause exit raises the loss directly — the operator forfeits the unamortized balance. Refundable, or pro-rata on for-cause, cuts it. On a deal with real key money, the difference is six figures. Resist the non-refundable structure unless it’s offset by a softer threshold or other protection. It benefits the owner — a bigger payoff on firing — and the operator pays for it in exposure.
Remaining term
A longer remaining term means more fees at risk, so more exposure. Front-loading fees — more base early, less late — cuts the loss in the later test years. Ramp-period fee floors cut exposure during stabilization, when the test is most likely to fail. The exposure falls naturally over the term anyway, so threshold calibration matters most when the remaining term is long.
Liquidated damages
LD caps protect the operator on a convenience or sale exit. They do nothing for a for-cause firing. A performance-test firing is for cause, so no LD. LD caps are still worth negotiating — they cover the owner’s sale or convenience exit, which are separate risks — but they don’t touch performance-test exposure. Operators sometimes treat LD as general termination protection. It isn’t. The distinction matters.
Cure mechanics
Cure rights, windows, and caps change the paths open to the owner. A higher cure cap keeps cure attractive to the owner and lowers the firing rate — the owner takes a cure payment instead of eating the valley. Longer windows give the operator time to fix performance before paying. Unlimited rights keep the contract from turning into sudden death late in the term. Cure is one input that lowers the firing rate. It doesn’t erase the exposure — the test still fails, and cure payments still cost — but it makes the owner likelier to cure than to fire. The compound-option structure of cure, an option nested inside the termination option, is its own piece. Here it’s a modifier that pulls the firing rate down.
Replaceability
Hard-to-replace operators face lower firing rates; easy-to-replace operators, higher. Replaceability isn’t a contract term — the operator can’t negotiate it. But knowing where the operator sits tells it which levers to prioritize. An easy-to-replace operator should fight harder for a softer threshold, because the higher firing rate amplifies every point of the failure rate. A hard-to-replace operator has more room to accept a tighter threshold, because the low firing rate holds the exposure down even when the test fails often.
What the Operator Does With It
Price the exposure
Run the expected-value walk for the specific deal. Inputs: the property’s realistic performance from underwriting or Monte Carlo, the threshold, the archetype firing rate, the remaining fees, the key money structure. Output: dollar exposure across the term, declining as the fees shrink. The number isn’t decorative. It’s the dollar the operator carries. It turns “we’re worried about the performance test” into “the test as drafted exposes us to about $1 million over the term, and here’s the lever that halves it.”
Identify the archetype
Hard-to-replace — branded, complex assets, specialized — means a lower firing rate, so lower exposure at the same threshold. Easy-to-replace — select-service, regional, commoditized — means a higher firing rate, so higher exposure at the same threshold. If the operator is easy-to-replace, threshold calibration matters even more: the higher firing rate amplifies every failure, and a threshold one point too high can double the exposure. If it’s hard-to-replace, there’s more room on threshold, because the low firing rate holds the exposure down even when the test fails often.
Target the highest-impact lever
A softer threshold usually moves the number most, because it cuts the failure rate directly — one point can cut exposure 40 to 60% when the property’s performance sits near the threshold. If the threshold is fixed — the owner won’t move, or market convention locks it — go after key money refundability or cure mechanics. Refundable key money cuts the loss; cure mechanics cut the firing rate. Both matter, though neither moves the number as much as threshold. If threshold, key money, and cure are all fixed, front-load the fees to cut later-year exposure — more revenue early, when the test is less likely to fail, and less late, when failure is likelier but the loss is smaller.
Use the number in negotiation
“The test as drafted exposes us to about $1 million in expected termination cost over the term. We’ll take that at a 29% threshold. Or we need key money refundable on for-cause. Or the threshold at 28%.” The owner probably hasn’t computed the operator’s exposure — most don’t. But the owner understands cost-benefit, and understands an operator pricing a risk and asking for a specific trade. That’s a negotiable conversation. “We’re worried about the performance test” isn’t. The operator knows the number; the owner likely doesn’t. That’s the edge.
Portfolio view
Across many HMAs, the exposure compounds. An easy-to-replace operator with 10 to 15 HMAs can carry millions in aggregate. One bad cycle — a recession, a demand shock, a regional downturn — can push several properties into failure at once, so the risk is correlated across the book. Standardizing threshold and key money terms across the portfolio cuts the aggregate more than a fee concession on any single deal.
The Strategic Shift
Most operators negotiate performance tests as threats to avoid. Pass, or lose the contract. That framing produces defensive behavior — fighting every provision without knowing which ones matter. The better frame: a performance test is a priced option to manage. Compute the exposure, identify the archetype, target the highest-impact lever, and use the number to frame the trade. Once the price is on the table, the operator can decide whether to pay it, cut it, or trade it away.
Two Numbers, One Negotiation
A performance test has two prices. The owner’s payoff from firing explains the firing rate. The operator’s exposure is the negotiable number.
The owner’s ledger: uncertain benefit against a near-certain, large cost. The valley is real; the upside is a hope. So firing is rare, and rarer still for hard-to-replace operators. That’s the 92%/11% gap. The owner holds the option and mostly doesn’t use it.
The operator’s ledger: the exposure is the failure rate times the firing rate times the loss if fired. For the same deal, easy-to-replace operators carry more — not because they lose more, but because they’re fired more. Termination risk scales with replaceability.
The move is to compute both numbers before negotiating. Identify the archetype. Target the lever that moves the exposure most. Use the number to frame the trade.
Performance tests aren’t binary thresholds. They’re options with prices. The owner’s price explains why firing is rare. The operator’s price is what gets negotiated. So on your current deals, what is the performance test actually worth — and are you negotiating the number you carry, or just the one you’re afraid of?
Dash Decisions provides vendor-side financial deal desk services for hotel operators bidding on HMAs.