TDCR Technical debt and combined ratio

Issue 26 · Economics over fashion

A Construction Dashboard Cannot Turn a Bond into Insurance

Construction systems often bring course-of-construction coverage and surety bonds into one project view. That is useful, provided the system preserves the distinct obligee, principal, notice, indemnity, and damage requirements instead of presenting a fictional combined protection.

One system

surety bonds

The other

builders risk

Surety bonds and course-of-construction insurance meet at one construction project, but the first guarantees the contractor's obligations while the second responds to covered physical damage to the work.

The project has one schedule, but it does not have one risk-transfer contract.

A construction dashboard is a reasonable thing for us to want. We give the project one budget. One critical path. One set of draws. One unhappy executive when the date moves. It also has several contracts we did not write together.

The project has one schedule and no single risk-transfer contract.

A surety bond guarantees the contractor’s obligation. A course-of-construction policy, the Canadian name for builders risk, responds to covered physical damage and to defined extensions like soft costs or delay in start-up. One can be quiet while the other is in trouble. Both can be live in one month, answering different questions. We break that distinction with a modern interface. Our platform sees a contractor, an owner, a lender, a budget, a date, and a pile of documents. It wants one coverage status, preferably green. But they do not form one product we can sell.

CCDC 220, 221, and 222 do not become insurance forms because they share a project identifier with CCDC 2, CCDC 41, and a manuscript wording.

The timeline is worth building and the combined status is not

So here is the line I would hold.

Build the shared timeline.

Don’t build the fictional combined coverage engine.

The timeline earns its place by making our promises diverge.

A delivery can trigger inception before construction begins.

A tender can demand a bid bond before any contract.

Ready-for-Takeover can end property coverage.

A default leaves the performance bond open through warranty. A useful system puts those events beside each other, then keeps the separate tests underneath them.

The bond starts long before anybody breaks ground

The surety decision starts months before bid day.

A contractor gets a facility, not approval for one job. The surety reads financial statements, a work-in-progress schedule, equipment and encumbrances, bank operating-line details, and a CCDC 11 qualification statement. It sets a single-job limit and an aggregate for backlog.

That is a credit decision, not ours.

The surety is not pricing expected physical loss.

It is asking whether the contractor can perform, and what it recovers if the contractor can’t. The facility runs under a General Indemnity Agreement, signed once and used across every bond.

It binds the company, affiliates, and personal indemnitors.

It can assign contract proceeds and book debts.

It allows a collateral demand before the surety pays anything.

No equivalent hidden contract sits behind the policy.

That alone should stop us saying “coverage” as though it meant one thing. At tender, the bid bond answers one narrow question. If the low compliant bidder won’t sign at its price, the bond covers the owner’s difference to the next accepted bid, capped at the bond amount. In Canada, R. v. Ron Engineering supplies the Contract A doctrine that makes a compliant bid irrevocable on its terms. The bid bond is not a promise to build. At award, the performance and payment bonds take over.

In Ontario, public contracts above the prescribed threshold need both, at no less than half the contract price, on forms substantially like CCDC 221 and 222.

The performance bond protects the owner.

The payment bond protects unpaid subtrades and suppliers.

Our dashboard may show both as “bonded.”

Their obligees, claimants, and notice paths all differ.

The penal sum is a ceiling.

It is not money waiting to be released.

The surety picks a remedy its form allows: support the contractor, take over, arrange completion, or pay the cost to complete up to that ceiling.

The owner does not choose a completion contractor and invoice the surety like a contingency line.

The policy follows the work, and the work has named parties

Course-of-construction insurance starts from a different event.

It protects a different interest too.

We attach cover when the insured has an interest in materials.

In a typical placement, inception is the first delivery, which can precede the first day of construction.

It sits at full completed value from day one.

It does not climb with the progress-draw total.

That distinction belongs in the data model.

Current spend is project accounting, not ours.

Full completed value is the limit we wrote.

Treat one as a proxy for the other and the limit looks adequate until a loss prices the difference.

The named parties are another thing we flatten.

Owner, contractor, and subcontractors of every tier are normally named as their interests may appear.

Under Commonwealth Construction v. Imperial Oil, we cannot subrogate against a party insured under the same builders risk policy.

The question isn’t only who can claim.

It is who we can no longer pursue after paying.

The lender belongs on a separate track, as mortgagee or loss payee under a standard mortgage clause.

Architects and engineers stay outside the named-insured group, so subrogation against professional liability survives a design-caused loss.

A system with one participant table and one status called “insured” has thrown a decision away.

The wording answers to three rulebooks: the construction contract, the manuscript policy, and provincial insurance law.

CCDC 2 and 41 allocate that obligation and tie it to Ready-for-Takeover.

But the construction contract doesn’t bind the insurer.

A certificate of insurance is a summary, not coverage.

The operative document is the wording: exclusions, sublimits, occupancy clause, conditions.

Which matters where our dashboard is least useful.

Defect, testing, and use.

LEG 1, 2, and 3 differ in how much of the cost of remedying a defect they exclude.

DE3 and DE5 use another series and another boundary.

Cold testing generally stays inside the grant.

Hot testing can be restricted, require consent, or push machinery breakdown into its own section.

No bond status or milestone decides any of it.

A water loss and a contractor collapse are two different files

Consider a water loss and a contractor collapse.

One project, two entirely different routines.

A water file begins with mitigation and causation, and we own both.

Drying has to start within two or three days, or the mould-remediation scope grows enormously.

The contractor needs separate cost codes and a change directive, not remediation folded into ordinary draws.

The adjuster needs a scope, a proof of loss, and a way to separate insured repairs from work the contractor owed anyway.

Soft costs and delay don’t make this a bond claim.

Both need covered physical damage under the material-damage section.

The material damage proviso means a labour action, a permit refusal, or a supplier insolvency triggers neither.

Their own mechanics still apply.

A waiting period in days. An indemnity period from projected completion. A schedule for construction-period interest, extended consultant and supervision fees, financing charges, or rental income under signed leases.

We should keep that arithmetic in the shared record.

On the corpus project, a thirty-five-day certified delay meets a thirty-day waiting period, leaving five indemnified days.

The policy hasn’t failed by declining to restore all thirty-five.

It applied the delay deductible somebody bought.

“DSU: active” hides the economic test entirely.

A performance-bond file begins somewhere else entirely.

The owner declares default under the construction contract.

It gives the cure notice, terminates the contractor’s right to continue where required, and notifies the surety in writing.

Under CCDC 221 the owner must not itself be in default, and must make the contract balance available.

Those are conditions precedent on the owner, not on us.

They are not workflow preferences.

The surety investigates percentage complete, remaining cost to finish, unpaid subtrade position, and contract balance plus holdback.

It may retain counsel and a construction accountant, because it is choosing a remedy rather than adjusting property damage.

After payment the indemnity agreement keeps running.

Indemnity, assigned receivables, trust claims, and salvage can keep the file alive for years after substantial performance.

The payment bond opens a third route.

Under CCDC 222, a claimant notifies principal, surety, and obligee within a hundred and twenty days of last supplying work or materials, then sues within the year.

The claimant is a beneficiary, not a party.

A generic “claim opened” stamp won’t carry those dates.

Build the timeline, then expose every handoff underneath it

The design is modest, which is why nobody sells it.

One shared project timeline.

Then separate records for bond form, obligee, principal, penal sum, facility capacity, indemnity agreement, default notice, cure status, contract balance, takeover, salvage.

And separate records for wording, full completed value, named insureds, lender clause, deductible, defect clause, extensions, loss notice, material-damage status, delay in start-up, soft costs, testing, occupancy.

The value isn’t a better combined score.

It is asking the right question at the right event.

Is the bid still open under Contract A?

Has the contractor consumed its aggregate bonding capacity?

Has the owner preserved its CCDC 221 conditions?

Does the loss arise from covered physical damage?

Is the delayed revenue beyond the waiting period?

Has a named subcontractor removed our subrogation route?

Did partial occupancy get prior written consent?

Those aren’t embellishments to bolt on afterwards.

They are the controls we operate.

Ontario’s lien-preservation clock starts at substantial performance.

Our cover can end earlier: occupancy, intended use, termination, expiry.

The permanent property policy needs coordinating before the handoff, not an alert after a tenant has keys.

This is board material. It is not a board system.

Executives want a disciplined view of default, damage, and delay.

We need the evidence and notices that make it true.

One red-green indicator answers a question nobody should ask.

AI helps here, in one specific place.

It reads certificates, pulls dates out of CCDC forms, compares a policy schedule to a project plan, and surfaces a missing consent.

It can help us find the contract.

It should not decide that a default notice worked, that a defect is covered, or that an extension survived occupancy.

“The dashboard thought the bond was insurance” is not a sentence anybody wants to read back in a dispute.

One project record earns its keep by showing where the promises stop.

What the issue establishes

  1. distinguish a bid bond, performance bond, and labour and material payment bond
  2. identify the surety facility and General Indemnity Agreement behind every bond request
  3. trace contractor default through cure, notice, takeover, indemnity, and salvage
  4. set course-of-construction insurance at full completed value rather than spend to date
  5. separate covered physical damage, soft costs, and delay in start-up from contractor default
  6. preserve distinct occupancy, testing, naming, and notice controls in a shared project record

The unfashionable recommendation

Build a shared project timeline rather than a combined coverage engine; the value is in exposing different triggers and notice obligations, not in erasing them.

Asked in the meeting

Is a performance bond the same thing as builders risk insurance?
No. A performance bond guarantees the principal contractor's performance of the construction contract for the obligee owner, subject to the bond form and the owner's compliance with the contract. Course-of-construction insurance responds to covered direct physical loss or damage to the work, subject to its wording, exclusions, deductibles, and conditions. A surety expects indemnity from the contractor; an insurer does not convert a physical-damage claim into contractor debt.
What is the difference between a bid bond, a performance bond, and a payment bond?
A bid bond responds when a low compliant bidder refuses to execute the contract at its bid price, usually for the difference to the next accepted bid up to the bond amount. A performance bond guarantees the contractor's contractual performance for the owner. A labour and material payment bond protects unpaid subcontractors and suppliers, who may claim as beneficiaries despite not being parties to the bond.
What is a surety facility and why does it matter before a project starts?
A surety facility is approved bonding capacity established before a particular project is bid. The surety assesses financial statements, work-in-progress reporting, equipment, bank facilities, and contractor qualifications, then sets a single-job limit and an aggregate backlog limit. Each bond request consumes remaining capacity, so a contractor can be unable to obtain a new bond despite having qualified for a similar project earlier.
What does the General Indemnity Agreement do?
The General Indemnity Agreement is the master agreement under which the contractor, affiliates, and often personal indemnitors reimburse the surety for losses and expenses. It can assign contract proceeds and receivables, permit a collateral demand when loss is anticipated, and make the surety's payment records presumptive evidence of loss. It explains why a bond is a credit guarantee rather than an insurance policy for the contractor.
Can an owner simply call a performance bond when a contractor stops working?
No. A performance-bond response normally requires the owner to declare default under the construction contract, give the required cure notice, terminate the contractor's right to continue where the contract requires it, and notify the surety in writing. Under CCDC 221, the owner must also remain compliant and make the contract balance available. The surety then selects a remedy permitted by the bond.
What can a surety do after a valid contractor default?
A surety can remedy the default while supporting the original contractor, take over and complete through a completion contractor, arrange completion and pay the owner's excess cost over the contract balance, or pay the owner's cost to complete up to the penal sum. The obligee does not choose among those options merely because it has declared default. The bond form and the notice sequence govern the response.
Why is builders risk written for full completed value from the beginning?
Course-of-construction insurance is ordinarily written to the full completed value of the work from inception, rather than to construction spend to date. The insured exposure is the completed building that will exist during the project, and the premium is rated on that value. A project record that reports only accumulated spend can therefore look financially complete while hiding an inadequate insurance limit.
Does delay in start-up pay for every construction delay?
No. Delay in start-up and soft-costs coverage require an underlying covered physical loss or damage under the material-damage section. A labour action, permit refusal, or supplier insolvency does not trigger those extensions merely because it delays completion. The coverage also applies its own waiting period, indemnity period, and schedule of covered soft costs, rather than the performance-bond default test.
Why does a construction dashboard need separate notice controls?
Bond and insurance notices serve different contracts and occur at different events. A payment-bond claimant under the traditional CCDC 222 form gives written notice within 120 days of last supplying labour or material and starts suit within one year. A performance-bond file requires default, contractual cure, termination where required, and surety notice. A physical-damage loss requires preservation, mitigation, and insurer reporting under the policy.
What does named insured status do on a builders risk policy?
Named insured status on a builders risk policy is not simply permission to make a claim. Under Commonwealth Construction v. Imperial Oil, an insurer cannot subrogate against a party insured under the same policy. The usual structure names the owner, contractor, and subcontractors of every tier as their interests may appear. Design consultants are normally excluded so the insurer can preserve recovery against professional-liability coverage.
What changes when part of the project is occupied before the rest is finished?
Course-of-construction coverage commonly restricts or terminates at the earliest of substantial completion or Ready-for-Takeover, occupancy or intended use, contract termination, or policy expiry. Partial occupancy usually requires the insurer's prior written consent. A permanent property policy must be coordinated before tenants move in or the owner uses a finished area, because a dashboard cannot prevent the coverage gap created by an unapproved handoff.
Why do Canadian construction projects need their own rules in a project system?
Canadian projects use Canadian Construction Documents Committee forms and provincial construction regimes. Ontario public contracts above the prescribed threshold require performance and labour and material payment bonds of at least 50 per cent of the contract price on prescribed forms substantially the same as CCDC 221 and CCDC 222. CCDC 2 and CCDC 41 allocate property-insurance obligations, while provincial rules govern liens, trusts, procurement, and insurance conditions.