How much contingency on a GMP is enough is the wrong question in the form it is usually asked. A percentage describes the size of a fund and says nothing about the demand it has to meet, and the demand is knowable on the day you sign.
Why percentages are unhelpful
Every owner has heard a number. Three percent, five percent, seven on a renovation. They are rules of thumb and they come from somewhere reasonable, which is that somebody averaged a lot of projects.
The trouble with an average is that it describes a population and you are building one building.
Two projects at the same contract value with the same contingency percentage can be in completely different positions. One has a complete design, four allowances and no uncapped exclusions. The other has a seventy percent design, twenty three allowances and three uncapped exclusions on a site nobody has surveyed.
The same percentage protects the first generously and the second not at all.
What is more useful is to turn the question around. Instead of asking how big the fund should be, list what is going to draw on it and add that up. The list exists at signature and every item on it is in a document you already hold.
That is the whole method and the rest of this note is how to build the list.
Answering how much contingency on a GMP by building the demand
Five categories, each with a low and a high.
Unresolved allowances. For each one, the figure carried against a realistic landing point given the current design and current pricing. On most projects this is the largest single contributor.
Uncapped exclusions. For each, the probability it is encountered on this site and the cost if it is. An exclusion with a quantity cap contributes a bounded amount; one without contributes a much wider band.
Unbought packages. The value still priced from an estimate rather than an award, with a reasonable band for market movement between now and award.
Outstanding owner decisions. Every decision not yet made is a cost not yet fixed, and late decisions price worse than early ones.
Expected change volume. Based on the delivery method and on how complete the design is, rather than on an industry rule.
Add the lows and the highs. That range is the demand, and the question is now whether the fund covers it.
Reading the answer
Compare the owner controlled fund against the demand range and one of three things is true.
The fund exceeds the high. Comfortable, and possibly over provided. On a well documented project this is achievable and it frees capital elsewhere.
The fund covers the middle. The usual position and a reasonable one. It means a normal project completes within budget and an unlucky one requires a decision.
The fund is below the low. A capital call or a scope reduction is already implied on the day you sign, and the only open question is when somebody notices.
The third case is more common than owners expect, and it is almost always caused by two things: counting the contractor contingency as though it were available, and sizing the fund from a percentage rather than from the list.
The first of those is covered in contractor contingency and owner contingency, and it is worth checking before anything else, because on many budgets it moves the answer by a third.
A worked example
Illustrative figures. Not taken from any client project and not a quotation.
A $64 million guaranteed maximum price. The budget shows contingency of $2.9 million, which is 4.5 percent and reads as adequate.
Of that, $1.2 million is contractor contingency inside the contract price. The owner controlled fund is $1.7 million.
The demand. Seventeen unresolved allowances carrying $4.1 million, realistic range $4.4 million to $5.3 million, contributing $300,000 to $1.2 million. Two uncapped exclusions contributing $0 to $700,000. Unbought packages of $18 million contributing $360,000 to $900,000. Two outstanding owner decisions contributing $150,000 to $400,000. Expected change volume $280,000 to $450,000.
Demand range: $1.09 million to $3.65 million, midpoint $2.37 million.
Fund to demand at the midpoint: 0.71. Against the high: 0.47.
The budget said 4.5 percent and looked fine. The list says a capital call is more likely than not, and every figure in it came from documents the owner already held on the day of signature.
What to do when the fund is short
Four options, and they are not equally available at every stage.
Reduce the demand. Cap the uncapped exclusions, resolve the allowances with the largest ranges, and make the outstanding owner decisions. Every one of these is cheapest before signature and each closes a line in the list.
Increase the fund. Move money from elsewhere in the budget, or from the equity. Honest and unpopular.
Reduce the scope. Identify what comes out if the fund runs short, in advance, rather than deciding under pressure in month fourteen. A written deferral list is a genuine instrument and almost nobody has one.
Accept the position knowingly. Legitimate, provided it is a decision rather than an oversight, and provided whoever funds the capital call has seen the arithmetic.
The third is the most undervalued. A deferral list agreed at signature, naming the scope that comes out in order, converts a panic in month fourteen into an item that was already decided.
It also tends to improve the negotiation before signature, because a list of things you are prepared to remove is a list of things somebody would rather you kept.
The order of the four matters as well. Reducing the demand is always tried first, because every dollar removed from the demand is worth a dollar added to the fund and it costs a negotiation rather than capital.
The second option, more money, is the one owners reach for and the one with the highest cost of capital attached. It is the right answer when the demand genuinely cannot be reduced, and it is frequently used when nobody tried the first option because nobody built the list.
What to do before you sign
- Separate the owner fund from the contractor fund and use only the first in the comparison.
- Build the demand from the five categories, each with a low and a high.
- State the ratio of fund to demand at the midpoint and at the high.
- Close the widest lines first: uncapped exclusions and the largest allowance ranges.
- Make the outstanding owner decisions, because those are the ones you control.
- Write a deferral list naming what comes out, in order, if the fund runs short.
- Rebuild the demand whenever the allowance schedule or the buyout position moves materially.
Item three is what turns this into a number you can report. A ratio moves for reasons, improves when allowances resolve well and deteriorates when an exclusion is encountered, and every movement has a cause you can name.
Item seven keeps it true. A demand built at signature and never updated is a snapshot, and the whole value of the method is that it narrows as the project settles.
The scope options for this reading sit in the review packages.
What we do
We build the demand from the allowance schedule, the qualifications page, the procurement log and the outstanding owner decisions, and set it against the fund you actually control. The output is a ratio with the arithmetic shown and the three widest lines named, so the conversation is about specific items rather than about a percentage. The work is the readiness review.
Questions people ask
Is five percent contingency enough?
The percentage cannot answer that on its own. Five percent against a complete design with four allowances is generous. The same five percent against seventeen unresolved allowances and two uncapped exclusions is short before anybody starts. Adequacy is a relationship between the fund and the demand.
How do you size the demand?
From five categories: unresolved allowances, uncapped exclusions, unbought packages, outstanding owner decisions and expected change volume. Each gets a low and a high, and the totals give a range. Every input is in a document you already hold at signature.
What if the fund is clearly short?
Four options: reduce the demand by capping exclusions and resolving allowances, increase the fund, reduce scope using a written deferral list agreed in advance, or accept the position knowingly. The third is undervalued because it converts a later panic into a decision already made.
This is general information about construction contracts and is not legal advice.