Lump sum contracts: price certainty and what it costs

A lump sum is the contract most owners reach for first. The contractor agrees to deliver a defined scope for a fixed price, and the owner knows at award what the works will cost. For a small business with a fixed approval and a board that dislikes surprises, that certainty is valuable. It is not free, and it is not as fixed as it looks.

This chapter builds a fair lump sum for the distribution centre’s works from its parts, shows what the contractor earns or loses as its own costs vary, and follows the price through variations to an expected outturn against the works allowance. It then explains why a contractor that is losing money behaves differently from one that is not, and ends with a tender comparison in which the cheapest headline price was the dearest real one.

What a lump sum buys

Under a lump sum, the owner pays an agreed price for an agreed scope, adjusted only for changes the owner makes and for events the contract puts at the owner’s risk. The contractor carries the cost of doing the work: if its own cost is lower than it allowed, it keeps the difference, and if higher, it absorbs it. The sharing rule of Chapter 2, How contracts shape behaviour, puts the lump sum at one end: the contractor keeps every dollar saved and pays every dollar lost.

That rule gives the owner three things:

• A known price at award, which the board can compare with the approved budget.
• A contractor with a strong reason to control its own cost, by planning well, buying well and avoiding rework.
• A simple administration: the owner pays progress claims against the price and manages variations, and does not need to audit the contractor’s costs.

It also gives the owner three obligations. The scope must be defined well enough to price, because anything not in it becomes a variation. Changes must be priced as they arise, usually without competition. And the owner must accept that the price includes the contractor’s allowance for the risks it is carrying, whether or not they occur.

In Australia, many building lump sums are let under general conditions published by Standards Australia: AS 4000-1997, general conditions of contract for construct-only work, and AS 4300-1995, general conditions of contract for design and construct. Standards Australia now lists AS 4000-1997 as superseded, so readers should check the current edition before relying on either form, and in practice both are almost always amended by special conditions. The NEC4 Engineering and Construction Contract separates the same pricing choice into main options: Option A, a priced contract with an activity schedule, and Option B, a priced contract with a bill of quantities. The forms differ in their mechanisms; the commercial logic of a fixed price for a defined scope is the same. The collection chapter Matching the contract to the work covers how far a contract’s structure should scale with the work.

Building up a fair price

This is an illustration. All amounts are in A$m excluding GST. A fair lump sum for the works is built from three parts.

• Direct cost, the contractors’ own cost of labour, materials, plant, subcontracts and site overheads, at the median estimate: 13.50.
• Fee, the contractor’s corporate overhead and profit at 8% of direct cost: 0.08 × 13.50 = 1.080.
• Risk premium, the contractor’s allowance for carrying the risks of a fixed price, at 4% of direct cost: 0.04 × 13.50 = 0.540.

The fair lump sum is 13.50 + 1.080 + 0.540 = 15.120.

Figure 1The fair lump sum of A$15.120m

1. Direct cost, median estimateA$13.50m · 89%labour, materials, plant, subcontracts, site overheads
2. Fee, overhead and profit at 8%A$1.080m · 7%the contractor's business costs and return
3. Risk premium at 4%A$0.540m · 4%ground, productivity, weather, subcontractor failure, estimating error

The risk premium is the price of certainty. It covers the contractor’s exposure to everything the contract puts at its risk: ground conditions it has been given information on, its own productivity, weather within normal bounds, subcontractor failure, and errors in its own estimate. The 4% is a teaching value for this case, not a market rate. A real premium depends on how complete the design is, how much risk the contract transfers, and how hungry the market is for work.

The owner pays the premium whether the risks occur or not. If the contractor’s direct cost turns out to be exactly 13.50, the contractor keeps the 0.540 on top of its 1.080 fee. That is not a windfall; it is the fair price of having carried a risk that, this time, did not occur.

The contractor’s margin at five outcomes

Once the price is fixed at 15.120, the contractor’s margin, its fee, premium and any saving combined, is simply the price minus its own direct cost.

Figure 2Contractor's margin under the lump sum at five direct-cost outcomes

Direct cost 12.80A$2.320m15.3% of price

Direct cost 13.50, the medianA$1.620m10.7% of price

Direct cost 14.00A$1.120m7.4% of price

Direct cost 14.40A$0.720m4.8% of price

Direct cost 15.12A$0.000m0% of price

At the median direct cost of 13.50, the margin is 1.620, 10.7% of the price: the fee plus the unused premium. If the contractor works efficiently and its cost falls to 12.80, the margin rises to 2.320, 15.3% of the price. If its cost rises to 14.40, the margin falls to 0.720, 4.8%, and at 15.12 it is zero. Every dollar of the contractor’s cost moves its margin by a dollar and leaves the owner’s payment unchanged.

That is the lump sum’s great strength while the contractor is in the upper half of the chart. A contractor earning a healthy margin has every reason to keep earning it: to finish on time, avoid rework and protect its reputation.

Variations and the expected outturn

The price of 15.120 covers the scope as tendered. The owner will change some of it. For the distribution centre, the owner’s team expects changes of about 0.50 of direct cost over the project, from layout refinements, extra dock equipment and adjustments to the conveyors once the order profiles are final.

Under a lump sum, those changes are priced by the contractor as variations. The contractor is already on site and no competitor can offer a price, so variations commonly carry a margin above the tendered fee. At a 15% margin on cost, the owner’s changes cost 0.50 × 1.15 = 0.575. The difference between 15% on variations and 8% in the tendered fee is not profiteering: a variation disrupts planned work, needs design and pricing effort, and is priced without the competitive pressure of the tender.

An owner can reduce that margin before award rather than argue about it afterwards. The tender can ask for rates for foreseeable changes: a price per additional pallet position, per dock door with leveller, per metre of conveyor, per square metre of office fit-out. Those rates are then priced under competition and become part of the contract, so the most likely changes are valued by schedule rather than by negotiation. The owner can also state in the tender how variations will be valued where no rate applies, for example at the contractor’s verified cost plus a stated margin, which removes one source of dispute. Neither step stops the owner changing its mind; both make the price of changing it predictable.

The other lever is definition. A lump sum is only as firm as the scope it prices, and Scoring project definition with a weighted checklist in Front-End Planning by the Numbers shows one way to measure how firm that is before tender. For the distribution centre, the building was fully documented at tender and the racking layout was fixed, but the conveyors were specified by performance, a throughput and a set of functions, because the integrator’s detailed design depended on the WMS configuration. The more of the price that rests on a performance specification, the more of the final cost will be settled through variations and their margin, and the less the headline price means.

The expected contract outturn is the price plus variations: 15.120 + 0.575 = 15.695. The owner also pays for external contract administration, the quantity surveyor who checks progress claims and prices variations, and legal advice when needed: 0.15 for a lump sum, the lowest of the models in this book, because there are no costs to audit. The expected cost against the works allowance is 15.695 + 0.15 = 15.845.

Figure 3The lump sum's expected outturn against the works allowance

Item	A$m	Running total
Fair lump sum	15.120	15.120
Expected variations, 0.50 at cost plus 15%	0.575	15.695
External contract administration	0.150	15.845
Works allowance	15.600	over by 0.245
Owner's contingency remaining	0.400 − 0.245	0.155

The expected outturn of 15.845 is 0.245 over the A$15.60m works allowance. It fits within the owner’s A$0.40m contingency, leaving 0.155. That is thin: the contingency also has to cover everything else that might go wrong in the owner’s scope, from design development to the WMS. The owner should know this before it signs, not discover it from the variation register. How a contingency is sized from a forecast distribution, and who holds which part of it, is set out in Uplifts and contingency: from a percentile to a budget in Reference Class Forecasting by the Numbers.

When the contractor is losing

Now consider the lower half of the margin chart. If the contractor’s direct cost heads towards 15.12, its margin heads towards zero, and beyond it the contractor is paying to do the work. Its behaviour changes, and the change is predictable.

A losing contractor looks for revenue in the contract. Every ambiguity in the scope becomes a potential variation; every late drawing, access restriction or instruction becomes a potential delay claim. Notices are issued promptly and in writing, as the contract requires, and the owner’s team begins to spend its time responding to them. The contractor also looks for cost to cut: less supervision, the cheapest acceptable finish, the least experienced crew that will pass inspection, and less willingness to help with problems that are not strictly its own.

None of this is improper; the contract invites it. But it means that the lump sum’s price certainty is weakest exactly when the owner most needs it, and that the owner’s administration cost rises sharply on a losing job. A sound lump sum therefore depends on a complete definition, a price that is fair to the contractor, and a contingency large enough to absorb the claims that a tight price invites.

The error: comparing headline prices

This is an illustration. Three builders tendered for the works as a lump sum. The headline prices were:

• Tender A: 14.86
• Tender B: 15.08
• Tender C: 15.29

The first draft of the board paper recommended Tender A, A$220,000 below B and A$430,000 below C. Before it went to the board, the owner’s quantity surveyor completed the exclusions schedule: a line-by-line list of everything each tenderer had excluded or qualified, priced at what the owner would have to pay to add it back.

Tender A had excluded four items that the tender documents required:

• slab joint sealing under the racking, 0.11;
• temporary power for conveyor commissioning, 0.06;
• the stormwater detention tank, 0.24;
• the dock levellers’ pits, 0.11.

Together, those exclusions were worth 0.52. Tender B had one minor qualification worth 0.04, and Tender C had none.

Figure 4Three lump sum tenders, headline and normalised

Tender	Headline	Priced exclusions	Normalised
Tender A	14.86	0.52	15.38
Tender B	15.08	0.04	15.12
Tender C	15.29	0.00	15.29

Normalised to the same scope, the order changed. Tender B became the lowest at 15.12, matching the owner’s fair lump sum. Tender A, cheapest on headline by A$220,000, was dearest once normalised at 15.38, A$260,000 more than B.

The real cost of choosing A would probably have been higher still. Items excluded from a lump sum come back as variations, priced after award without competition and often at the 15% margin seen above, and some of them, such as the detention tank and the dock leveller pits, sit on the critical path of the civil works, where late addition would add delay as well.

The error was caught because the quantity surveyor’s exclusions schedule was a required step before the board paper was finalised, not an optional check. The corrected paper recommended Tender B and showed the normalisation table in full, so the board could see why the cheapest headline was not the cheapest price. Normalisation is one of the disciplines covered in Running a fair tender process, which also deals with clarifications, late bids and the rules an owner sets for itself.

Exercise: one variation, two ways to buy it

After award, the owner decides to add two dock doors with levellers on the east wall. The builder prices the variation at its direct cost of A$180,000 plus its 15% variation margin. The owner’s project director suggests a cheaper route: take the work out of the builder’s contract and obtain three quotes from specialist dock-equipment contractors. The lowest quote is A$186,000, and the owner’s team estimates A$12,000 for coordinating the separate contractor on the builder’s site. Work out both prices and discuss whether the cheaper route is worth taking.

Worked answer. The builder’s variation costs A$180,000 × 1.15 = A$207,000. The separate route costs A$186,000 + A$12,000 = A$198,000, a saving of A$9,000, about 4%. The saving is small against the interfaces it creates. A separate contractor on the builder’s site creates a new organisation pair and new interfaces: the wall openings and lintels, the pits and the slab, power to the levellers, and the programme for working in the builder’s area. The shared safety duties described in Chapter 1 mean the two businesses must coordinate their work, and the builder, as the contractor in charge of the site, must bring the new contractor under its site safety arrangements. If the separate contractor’s work delays the builder, the builder may have a claim for delay against the owner, and one week of delay to the distribution centre costs the owner A$80,000, almost nine times the saving. Warranty for the doors is also split from the warranty for the walls around them. Unless the owner has a reason beyond price, such as a preferred maintenance supplier for its existing dock equipment, it should negotiate the builder’s variation instead, by asking for the cost build-up and testing the 15% margin against the three quotes.