What is a performance bond in a public contract?

A performance bond, also called performance security, is a financial guarantee issued by a bank or surety that compensates the contracting authority if the contractor fails to perform the contract according to its terms. It is distinct from a bid bond, which only covers the tendering phase.

When does a performance bond start in public procurement?

A performance bond typically becomes effective on, or shortly before, the signing of the contract, once the contracting authority has issued the notice of award. Many notices require the bond to be submitted within 10 to 28 days of award, before the contract enters into force.

How long does a performance bond last?

Duration varies by contract type. Works contracts often keep performance security active through the defects liability or warranty period, which can extend 12 to 24 months beyond physical completion. Supply and services contracts usually release the bond at final acceptance.

What triggers a call on bond?

A call on bond is a formal demand by the contracting authority to the guarantor, usually following documented breach, non-performance, or failure to remedy defects within a notice period defined in the contract conditions.

Is a performance bond refundable?

Performance bonds are not deposits; they are guarantees, so there is nothing to 'refund' in the traditional sense. If no claim is made, the instrument is released or expires at the end of the warranty period, and any collateral the contractor posted with its bank is returned.

Can a performance bond be reduced during the contract?

Some contracts allow a reduction, often to 50% of the original value, once provisional acceptance is achieved, with the balance held through the warranty period. This depends entirely on the specific conditions of contract published by the authority.

Performance bonds in public contracts: when they start and how they end

Performance bonds in public contracts are financial guarantees that protect a contracting authority if a winning bidder fails to deliver according to the signed agreement. Unlike a bid bond, which only secures the tender phase, a performance bond (also called performance security) stays active from contract signature through some or all of the warranty period, and it can be enforced through a formal call on bond if the contractor defaults. Understanding exactly when this instrument starts, how its value may step down, and under what conditions it is released is essential before you commit working capital or bank credit lines to a public contract. This explainer walks through the lifecycle stage by stage, using terminology found in EU, World Bank, and UNCITRAL-aligned procurement frameworks.

What is a performance bond and why contracting authorities require it

A performance bond is a guarantee — issued by a bank, insurance company, or surety — that compensates the contracting authority for losses if the contractor does not fulfil its contractual obligations. It is a standard risk-allocation tool referenced across procurement systems tracked on TED, in World Bank-financed projects, and under the UNCITRAL Model Law on Public Procurement.

Authorities typically ask for performance security sized between 5% and 10% of the contract value, though high-risk infrastructure works can require more. The bond gives the buyer recourse without having to sue the contractor in court first — the guarantor pays on a valid demand, subject to the bond's own terms (conditional or on-demand).

This differs from the bid bond discussed in our guide on bid bonds and tender guarantees, which only covers the risk that a bidder withdraws or refuses to sign after winning. Once the contract is signed, the bid bond is released and the performance bond takes over.

When does a performance bond start

The starting point is usually fixed in the contract conditions, not in the tender notice itself. Common triggers include:

Notice of award — the authority informs the winning bidder, starting a countdown (commonly 10–28 calendar days) to submit the performance bond.
Contract signature — in many jurisdictions the bond must be in the authority's hands before or at the moment the contract is signed; without it, the award can be revoked.
Effective date / commencement order — for works contracts, the bond may need to remain valid from the commencement date stated in the letter of acceptance, not from signature.
First payment milestone — some service contracts condition the first invoice payment on proof that the performance bond has been issued and accepted.
Advance payment release — if an advance payment guarantee is also required, the authority may withhold any mobilisation advance until both instruments are lodged.

Because timing differs by authority and by contract type, always check the specific special conditions of contract (SCC) rather than assuming a generic rule. Our article on how to read a tender notice explains where these deadlines are usually disclosed before you even bid.

Financial data analysis on screen

How the warranty period affects the bond's duration

For works and some supply contracts, the performance bond does not expire at physical completion — it typically continues through a defined warranty period (also called defects liability period or maintenance period). This protects the authority against latent defects that surface after handover.

Contract stage	Typical bond status	Common duration
Contract signature to provisional acceptance	Full value active	Contract execution period
Provisional acceptance	Often reduced (e.g., to 50%)	Immediate to 30 days
Warranty / defects liability period	Reduced value active	12–24 months
Final acceptance	Released or expired	End of warranty period

Note that these figures are illustrative patterns seen across many jurisdictions, not a universal rule — the governing document is always the specific contract and the applicable national procurement law. Development-bank financed projects, for instance, often follow FIDIC-based conditions that spell out exact percentages and dates; see the World Bank procurement framework for reference language.

What is a call on bond and when can it happen

A call on bond is the formal demand the contracting authority sends to the guarantor requesting payment under the performance security. It is one of the most consequential events in a contractor's relationship with a public buyer, because it can affect future eligibility and banking relationships.

Typical grounds for a call on bond include:

Failure to start works or deliveries within the contractually agreed period.
Abandonment of the contract or persistent non-performance after a documented notice to cure.
Failure to remedy defects identified during the warranty period within the stated notice window.
Termination of the contract for default by the authority.
Breach of specific obligations the bond was explicitly written to cover (e.g., advance payment misuse, if a separate advance payment guarantee was not used).

On-demand bonds (common in many jurisdictions) allow the guarantor to pay on first written demand, without needing to verify the underlying breach — the contractor's recourse is then a separate dispute with the authority, not a defense against the bank. Conditional bonds require the authority to prove default before payment is released. Knowing which type is required in the notice materially changes your risk exposure, so this should be checked during the documentation-preparation stage described in how to prepare tender documentation.

How performance bonds end: release, expiry, and reduction

There are generally three ways a performance bond's obligation ends:

Release upon final acceptance — the authority issues a certificate confirming the works, supplies, or services meet contract requirements, and formally authorises the guarantor to release the bond.
Natural expiry — if the bond has a fixed validity date and the authority does not extend or call it before that date, it lapses automatically. Some contracts require the contractor to extend validity if completion is delayed.
Step-down / reduction — rather than a single release, many contracts reduce the bond value at defined milestones (e.g., 100% to 50% at provisional acceptance, 50% to zero at final acceptance), reducing the cost of maintaining the guarantee over time.

Contractors should track these dates as carefully as they track delivery milestones, since banks typically charge ongoing fees on the outstanding bond value and may require collateral or credit-line allocation for the full validity period.

Savings and investment growth concept

Risks and limits you should plan for

Performance bonds carry real financial and operational risk, and no outcome — financing approval, bond issuance, or contract completion without a claim — can be guaranteed. Before committing to a bid that requires performance security, consider:

Bank capacity and collateral. Your bank may require cash collateral, a counter-guarantee, or a charge over assets before issuing the bond, tying up working capital for the full contract and warranty period.
On-demand exposure. If the bond is on-demand, a dispute over quality or delay can result in payment to the authority before you have a chance to contest the underlying claim.
Multi-contract exposure. Running several public contracts simultaneously means several bonds drawing on the same credit facility — plan capacity well before bidding, not after award.
Currency and jurisdiction risk. Cross-border contracts may require the bond to be issued by a locally licensed bank or in local currency, adding cost and lead time.
Impact of a call on bond on future eligibility. A documented call can affect your standing in future tenders, since many eligibility checklists ask about past performance and claims history.

None of this is intended as financing or legal advice — verify every bond condition against the specific contract's special conditions and your own bank's terms, and cross-check eligibility requirements using our bid compliance checklist.

Practical checklist before you commit to a bond

Use this sequence when assessing a tender that requires performance security:

Confirm the required bond percentage and whether it is fixed or tiered against contract value.
Identify whether the bond is on-demand or conditional, and read the exact wording required by the authority.
Check the submission deadline relative to contract signature and the award notice date.
Ask your bank, in advance, about available capacity, collateral requirements, and issuance timelines.
Map the warranty period and any step-down milestones into your contract calendar.
Confirm the release/expiry mechanism — automatic, certificate-based, or requiring an authority letter.
Record the governing law and dispute venue in case a call on bond is contested.

Building this into your bid/no-bid decision — alongside the broader sourcing work covered in how to find public tenders worldwide — helps you avoid late surprises after an award is already confirmed.

If you want a second set of eyes on a specific notice's bonding requirements or on structuring your documentation file, leave your contacts for a consultation on tender selection and documentation preparation.

This material is for informational and educational purposes only. It is not legal, tax, or financial advice and it is not an official statement of any contracting authority. B2G Global Services Corp. is not a government agency. Outcomes in public procurement depend on published criteria, local law, and the bidder's own evidence. Readers should verify primary sources as of the action date.