Question 3 of 6: Contract Administration — Design-Bid-Build versus CM at Risk, and screening unbalanced bids

Question 3: Contract Administration — Design-Bid-Build versus CM at Risk, and screening unbalanced bids (20 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

Design-Bid-Build

Design-Bid-Build (DBB) is the traditional delivery method and remains the default for Canadian public work. The owner engages a consultant to complete the design, tenders the completed documents, and awards a construction contract — normally CCDC 2 Stipulated Price — to the lowest compliant bidder. The three phases are sequential and the two contracts are separate: the owner holds one with the designer and one with the constructor, and the constructor has no contractual relationship with the designer at all.

Its advantages begin with price certainty and transparency. Because every bidder prices an identical, complete set of documents, the resulting bids are directly comparable and the owner obtains a firm lump sum before committing to construction. That comparability is also what makes DBB the easiest method to defend against a disappointed bidder, which is why public agencies favour it: the award criterion is objective, the process is auditable, and the procedure satisfies open-tendering obligations under trade agreements such as the Canadian Free Trade Agreement. The owner retains full control of the design and can specify exactly what it wants, since the design is finished before anyone prices it. Roles are unambiguous, the standard documents are mature and heavily litigated (so their meaning is well settled), and the method requires the least owner sophistication of any delivery model — a small municipality can run a DBB project with one consultant and a payment certifier.

Its disadvantages are the mirror image of those strengths. It is the slowest method, because design must be fully complete before tendering can begin and construction cannot start before award; there is no overlap of design and construction, so fast-tracking is impossible without abandoning the model. It provides no constructability input during design: the party who knows most about how the building will actually be built, and what it will cost to build, is absent from the room until the design is frozen. The result is a design that may be buildable only expensively, and a stream of change orders once the contractor's knowledge finally arrives. The relationship is adversarial by structure — the contractor's margin improves by finding gaps in the documents, and the owner's position improves by denying them — which drives the claims culture discussed in Question 2. Lowest-price award encourages bid shopping down the subcontract chain and rewards the bidder who has most aggressively assumed away risk, so the low bid is frequently the bid containing the largest error. The owner also retains the Spearin-type implied warranty that its documents are adequate for construction, so design errors return to the owner as cost. And because the two contracts are separate, a defect that lies between design and workmanship leaves the owner litigating on two fronts against parties who each blame the other.

Construction Management at Risk

Under Construction Management at Risk (CM at Risk, or CMAR), the owner engages a construction manager early — typically at 30 % design — initially for pre-construction advisory services, and then converts that engagement into a construction contract in which the CM guarantees the price and delivers the work through trade subcontracts it holds itself. The defining feature, and the reason for “at risk”, is the Guaranteed Maximum Price: once the GMP is set, the CM bears the cost of any overrun. This is what distinguishes CMAR from CM-for-fee (agency CM), where the construction manager is purely an advisor, holds no trade contracts and carries no price risk. In Canada the arrangement is documented by CCDC 5B (Construction Management Contract for Services and Construction), with CCDC 5A covering the agency variant.

Its advantages start with early constructability input: the CM contributes to the design while it can still be changed cheaply, advising on methods, sequencing, market availability of trades and materials, and value engineering. Second, schedule compression through fast-tracking: because the CM holds the trade contracts, packages such as excavation, foundations and structural steel can be tendered and started while the interior fit-out is still being designed, which can remove months from a programme that DBB would deliver sequentially. Third, cost visibility and transparency: the CM prepares successive estimates as the design develops, so the owner learns early — rather than at tender opening — that the design is over budget, and the open-book trade-tendering process shows the owner the actual subcontract prices, with savings against the GMP typically shared. Fourth, improved risk allocation and a less adversarial relationship: the CM is on the owner's side of the table during pre-construction, and selection is qualifications-based, so the owner chooses a team rather than a number. Fifth, flexibility to accommodate change: scope can evolve during design without the change-order machinery that DBB requires.

Its disadvantages are equally structural. The most serious is that the owner commits to a builder before the price is known — the GMP is negotiated with a party already in place and with no competitor, so its competitiveness depends on the owner's own ability to benchmark it. The method therefore demands a sophisticated, well-resourced owner able to audit open-book costs, review trade tenders and manage the GMP negotiation; an unsophisticated owner is at a permanent information disadvantage. The final cost is not known at commitment, since fast-tracking means later packages are priced against an incomplete design, and the contingency inside the GMP is a matter of negotiation rather than of market test. There is a role conflict: the CM advises the owner during pre-construction and then contracts against it, and the CM's estimate becomes the benchmark against which its own GMP is judged. Fast-tracking itself carries real risk — a design change late in the process can invalidate work already built, and errors are discovered after the fact rather than before. Public-sector acceptability is lower: qualifications-based selection is harder to defend against a challenge than lowest-price award, and some agencies are constrained by procurement rules or by statute from using it at all. Finally, the CM's pre-construction fee and general conditions make the method more expensive in soft costs, which is difficult to justify on a small or simple project.

Screening unbalanced bids and selecting a responsible winner

A public agency awarding to the “lowest bid” is in fact awarding to the lowest bid that is both responsive and from a responsible bidder, and unbalancing is the principal way a bid can be low on its face and expensive in outturn.

An unbalanced bid is one in which the total is competitive but the individual unit prices do not reflect the actual cost of the corresponding work. Two forms are distinguished, and only the second is a ground for rejection. Front-end loading (mathematical unbalancing) inflates the prices of early items — mobilisation, excavation, foundations — and deflates those of later items. The bid total is unchanged, but the contractor is financed by the owner: it draws cash early, improving its own cash flow at the owner's cost in interest and, more seriously, eroding the owner's security, because if the contractor defaults at 60 % completion it may already have been paid 80 % of the price. Quantity-error exploitation (materially unbalancing) is the more dangerous form: the bidder judges that a unit-price item's estimated quantity is understated and prices it very high, while pricing an overstated item very low. The bid wins on the engineer's estimated quantities but, when the true quantities are measured, the contract cost rises well above the apparent low bid. This second form makes the award itself unsound, because the bid was never really the lowest.

The tests an agency applies to detect unbalancing are essentially arithmetic and comparative. The primary screen is a unit-price comparison against the engineer's estimate, item by item, flagging any line deviating beyond a stated tolerance — commonly ±25 % on significant items, with agencies also flagging any item whose extended value exceeds a threshold share of the bid. A second screen is a cross-bid comparison: the same item priced by all bidders, with outliers identified by their departure from the median rather than from the mean, which is far more robust when there are only four or five bids. A third is a front-end loading test that compares the cumulative cash flow implied by the bid prices, run against the tendered schedule, with the cash flow implied by the engineer's estimate; a bid that is paid materially earlier for the same work is front-end loaded whatever its total. A fourth is the quantity-sensitivity test that catches the materially unbalanced bid: the bids are re-evaluated using the agency's own best estimate of the likely final quantities rather than the tendered ones, and if the ranking changes, the apparent low bid is not in fact low. Fifth, an agency will look for nominal, zero or negative unit prices, which are almost always evidence of unbalancing and are prohibited outright by many tender documents. Sixth, the bidder is asked to justify the flagged prices in writing with a cost breakdown; an explanation that stands up (an item genuinely covered by plant already mobilised, for example) may resolve the flag, and the absence of one confirms it.

The corresponding defences written into the tender documents are worth stating alongside the tests, because prevention is cheaper than rejection: an express right to reject materially unbalanced bids; a cap on the mobilisation item, commonly at 3–5 % of the contract price; a “price adjustment for significant quantity variation” clause that re-negotiates the unit rate where the measured quantity departs from the estimate by more than, say, 25 %; a requirement to submit a schedule of values for owner approval before the first progress claim; and holdback and progress-payment provisions consistent with the applicable lien statute — in British Columbia the Builders Lien Act, which requires a 10 % holdback — which limits the exposure that front-end loading can create.

The second half of the enquiry is responsibility, which is about the bidder rather than the bid. An agency assesses financial capacity (audited statements, working capital, bank references, and above all the ability to furnish CCDC 221 performance and CCDC 222 labour-and-material payment bonds, since a surety's willingness to bond at 50 % or 100 % of the contract price is an independent underwriter's opinion of the bidder's solvency and competence); bonding and insurance (the bid bond that accompanied the tender, plus the required liability and course-of-construction coverage); experience and past performance on comparable work, verified through references, together with any record of default, litigation, liquidated damages or termination; technical and managerial capacity — the qualifications of the named project manager, superintendent and key staff, plant and equipment available, and current workload against bonding capacity; the quality of the proposed subcontractors, which is why many agencies require named subcontractors at bid closing (a device that also suppresses bid shopping); safety performance, typically the WorkSafeBC experience rating or equivalent provincial factor, plus a Certificate of Recognition and a reviewed safety programme; and legal and statutory standing — corporate registration, WorkSafeBC clearance letter, tax compliance, and eligibility under the agency's supplier-integrity or debarment policy.

Two Canadian points close the answer. First, the Ron Engineering line of cases established the Contract A / Contract B doctrine: submitting a compliant bid to a tender call forms an immediate contract (Contract A) obliging the owner to treat bids fairly and in accordance with the stated terms, with the construction contract (Contract B) awarded on acceptance. An agency that rejects a low bid as unbalanced or a bidder as not responsible is exercising a right that must have been reserved and disclosed in the tender documents, applied against criteria stated in advance, and documented — otherwise it is in breach of Contract A to the disappointed bidder. Second, the familiar “privilege clause” reserving the right not to accept the lowest or any tender does not license an arbitrary award; the courts have made clear it must be exercised in good faith and consistently with the disclosed evaluation criteria. In practice this means that a public agency's protection against an unbalanced bid is built at the drafting stage, not at the opening.