A subdivision can be legally approved long before every street, pipe, sidewalk, drain, tree and streetlight is finished.
That creates a dangerous gap.
Homes may be sold. Lots may change owners. The developer may run into financial trouble. Construction costs may rise. A contractor may disappear. A partially built road may wash out. A utility trench may remain unfinished. The municipality can suddenly face the worst possible sentence in land development: somebody still has to finish this.
Performance security exists to keep that unfinished obligation from quietly becoming the public’s unfunded problem.
This article has one reader job: explain how subdivision and site-development bonds, letters of credit, cash deposits and related guarantees bridge the period between approval of public improvements and their verified completion. It does not re-own subdivision design, which belongs to TPW-0115 — The Subdivision Plat. It does not re-own development exactions, which belong to TPW-0116 — The Development Exaction. It owns the narrower question: who carries completion risk when required infrastructure is not finished yet?
1. The basic problem: development and infrastructure do not finish at the same moment
A developer may need to record lots, obtain building permits, close financing or begin sales before every public improvement is complete. Requiring every last tree and pavement marking before any economic activity can proceed may be unnecessarily rigid.
But allowing development to move forward with no protection is equally dangerous. Once lots are sold, the leverage available to force completion can weaken. The developer may have extracted much of the site’s value while the public works remain unfinished.
Performance security is the bridge. The authority allows a defined development milestone to proceed because another reliable source of value stands behind the unfinished works.
2. What “bond” means in planning conversations
People often use “bond” as shorthand for several forms of financial security.
- Surety bond: a surety promises performance or payment subject to the bond terms if the developer defaults.
- Irrevocable letter of credit: a bank commits funds that the authority may draw under specified conditions.
- Cash or escrow: money is deposited or controlled directly.
- Other approved security instruments: some jurisdictions permit additional forms under statute or local rules.
The legal mechanics differ. The planning purpose is the same: convert a developer’s promise to finish required improvements into a credible completion resource.
3. Performance security is not a development fee
A fee is ordinarily paid and spent for a defined public purpose. A performance security is held as protection against non-performance and may be reduced or released as the secured work is completed.
Confusing the two obscures the economics. A $1 million security does not necessarily mean the developer has paid the city $1 million as a permanent charge. It means the city requires access to sufficient secured value if the developer fails to perform the bonded work.
4. Performance security is also not a substitute for plan approval
A developer cannot normally avoid engineering review by posting more money.
The authority first needs an approved site-related or subdivision plan defining what must be built. The security then guarantees completion of that defined work. Money without an approved scope does not tell anyone which road section, pipe diameter, sidewalk alignment or landscape item is required.
5. A current real-world example: Fairfax County
Fairfax County, Virginia provides a useful modern example because its land-development system publishes both the bond process and current cost schedule.
The county’s New Bond or Agreement guidance explains that a developer enters into an agreement and posts security to guarantee that development is completed in accordance with the approved site-related plan and within a reasonable timeframe. If the developer defaults, the county may use bond funds to complete public improvements.
Its 2026 Comprehensive Unit Price Schedule, effective March 1, 2026, applies to new bond requests, replacement bonds, bond extensions and bond reductions, among other deposits. Fairfax notes that bondable public infrastructure can include streets, sidewalks, utilities and other required items.
The details are local. The architecture of the system is widely useful.
6. The secured scope must be explicit
What exactly does the security cover?
Potential items include:
- public or private streets;
- curbs and gutters;
- sidewalks and shared paths;
- storm drains and detention facilities;
- sanitary sewer;
- water mains;
- streetlights;
- traffic-control devices;
- grading and stabilisation;
- street trees and required landscaping;
- monuments, survey markers or other subdivision requirements;
- restoration of disturbed public land.
The list should come from the approved plan and governing ordinance, not from a vague promise to “complete infrastructure”.
7. Why the authority needs a cost estimate
The security amount must be connected to the cost of completing the secured work.
That sounds simple until default occurs. The public authority may have to finish a project under worse conditions than the original developer: a new contractor, mobilisation costs, incomplete records, disturbed site conditions, inflation, damaged work, legal expenses and no commercial incentive to minimise cost.
For that reason, security schedules often include contingencies or administrative components rather than merely copying the developer’s lowest construction bid.
8. Unit-price schedules create consistency
One way to reduce argument is for an authority to publish standard unit prices for common improvement items.
Fairfax County’s 2026 schedule illustrates why such systems exist. The schedule is updated for current construction-cost conditions and is used in bond calculations, extensions and reductions. A consistent schedule reduces the temptation to understate completion cost by using unusually low internal estimates.
A unit-price schedule is not merely accounting. It is a risk-control instrument.
9. Inflation matters because unfinished infrastructure has a future price
A security posted three years ago may no longer cover the same work.
Fairfax’s 2026 schedule increased in line with its stated construction-cost methodology and requires supplemental security in certain extension or replacement cases where the updated estimate exceeds the existing posted amount.
The broader rule is important: security must be monitored against current completion exposure, not treated as correct forever because it was correct on the day it was posted.
10. The agreement defines the developer’s promise
The financial instrument normally sits beside a development or performance agreement identifying the obligations, deadlines, plans, inspection rights, extension process, default triggers and release conditions.
This separates two questions:
- What must the developer do? The agreement and approved plans define performance.
- What happens financially if the developer does not do it? The security instrument backs the obligation.
For the broader use of long-term agreements in planning, see TPW-0109 — The Development Agreement.
11. The security needs a reliable issuer
A promise is only useful if the party behind it can perform.
Authorities may therefore regulate acceptable sureties, banks, letter-of-credit wording, expiration dates, renewal provisions, forms and execution requirements. A beautifully calculated bond amount is little protection if the instrument expires unnoticed or cannot be drawn when default occurs.
12. Irrevocability matters
A letter of credit that can disappear at the developer’s request is not useful security.
The instrument must survive the period of risk under rules the authority can enforce. That is why local procedures often specify exact language and notice periods for expiration or non-renewal.
Administrative teams need a calendar for these dates. A security instrument can fail through paperwork long before the road fails physically.
13. When must security be posted?
The timing depends on the legal system. Common gates include final subdivision approval, plat recording, building permits, land disturbance, occupancy or another development milestone.
The policy principle is that security should exist before the municipality gives up the leverage that the security is meant to replace.
If lots can be sold first and security is requested later, the risk has already migrated.
14. Why bonds often appear around final plats
Recording a final plat can create legally transferable lots. That is a major economic event. Once lots are sold to individual owners, assembling land and responsibility becomes harder.
Performance security allows a jurisdiction to record a plat before every improvement is physically finished while preserving a funded route to completion.
This is one reason subdivision regulation is not just map-making. It is sequencing the creation of private property with the creation of the public systems that make those properties usable.
15. The inspection loop
Security is not released because the developer says the work is 80 per cent complete.
Inspection verifies installed quantities, quality and conformity with approved plans. Depending on the improvement, this may involve engineering inspections, compaction tests, pressure tests, CCTV of sewers, pavement checks, as-built surveys, landscape inspections or agency sign-off.
The bond ledger should be connected to verified field progress.
16. Bond reduction is risk reduction
Keeping the full original security until the final day can tie up developer credit unnecessarily. Releasing too much too early leaves the public exposed.
Many systems therefore permit periodic reductions as work is satisfactorily completed. The remaining security should correspond to the remaining completion risk, often with required retainage or contingency.
Fairfax County’s current 2026 unit-price schedule explicitly applies to bond reduction requests, illustrating that reductions are recalculated rather than treated as informal percentages.
17. Do not reduce against work that is merely installed but not acceptable
A road can be paved and still fail specification. A sewer can be buried and still leak. A tree can be planted and still be dead three months later.
Completion for security purposes needs a defined acceptance standard. Otherwise the authority converts construction risk into repair risk while releasing the money that protected it.
18. Extensions are not clerical renewals
If the developer needs more time, an extension should trigger a status review.
- Which improvements remain incomplete?
- Is the security still valid?
- Has construction cost changed?
- Are there unresolved inspection defects?
- Has the approved plan changed?
- Is supplemental security required?
- Is the project still progressing credibly?
An extension is therefore a risk-refresh event.
19. The default scenario is the reason the system exists
Most developers complete their work. A performance-security system is designed for the case where one does not.
Default may arise from insolvency, abandonment, refusal to correct defective work, expired deadlines or another breach defined in the agreement. The authority then follows the required notice and enforcement process and may call or draw security according to the governing documents.
The objective is not punishment. It is completion of the public improvements that were part of the development bargain.
20. Calling a bond is not magically equivalent to finishing a road
Financial recovery is only the beginning.
The authority may need to establish site control, procure designers, verify incomplete work, tender a completion contract, coordinate utilities, resolve defects and manage residents living beside an unfinished project.
This is why security sizing should consider public completion conditions rather than assuming the municipality can inherit the developer’s contractor and prices.
21. Default exposes documentation quality
When the original developer is gone, the authority needs the approved plan, inspection history, test results, as-built information, utility coordination records and a clear schedule of unfinished items.
A bond cannot compensate for total information loss. Good recordkeeping is part of completion security.
22. Public acceptance is a separate gate
A road that looks finished is not necessarily a public road.
Formal dedication or acceptance may require inspections, as-built drawings, legal conveyance, correction of punch-list items, utility sign-off, final pavement or another statutory process. Until then, responsibility may remain with the developer or private owner.
Subdivision buyers should therefore ask who owns and maintains streets and utilities at each stage—not assume public maintenance begins when residents move in.
23. Performance bond versus maintenance bond
Completion and durability are different risks.
A performance security protects completion. After improvements are completed or accepted, a maintenance or warranty security may continue for a defined period to protect against defects, settlement, failed planting or other performance problems that appear after construction.
Training materials used by planning organisations commonly distinguish performance security from later maintenance guarantees. The handoff between the two should be explicit so there is no unprotected gap.
24. The punch-list problem
Near completion, dozens of small defects can remain: cracked curb, missing sign, dead tree, incorrect valve cover, unfinished pavement marking, blocked drain, unseeded slope.
The final security amount should be sufficient to correct the outstanding list if the developer walks away. Releasing security because “the project is basically done” defeats the purpose precisely when leverage is most needed.
25. Landscaping is not decorative when it is required mitigation
Trees, buffers, screening and stabilisation may be required for environmental, visual or erosion-control reasons. They should not automatically be treated as optional finishing touches.
Because planting is seasonal and biological, security procedures may need special timing or replacement provisions. A tree is not successfully delivered merely because an invoice says it was planted.
26. Phased projects need phased security architecture
A large subdivision may have several plats and infrastructure phases. One trunk sewer or arterial road can serve multiple later phases.
The authority should identify which security protects which improvement and what happens if later phases never proceed. Critical shared infrastructure should not disappear into a spreadsheet organised only by individual parcel release.
This is where subdivision security intersects with infrastructure sequencing and adequate-public-facility controls, but it should not replace them.
27. Security and concurrency solve different problems
Concurrency asks whether sufficient infrastructure capacity exists or will exist as growth occurs. Performance security asks whether a developer will actually complete specific required improvements.
A bond cannot create water-treatment capacity that does not exist. Conversely, a citywide capacity plan does not guarantee that the sidewalk in front of a particular subdivision will be built.
See TPW-0110 — Concurrency and Adequate Public Facilities.
28. Security and exactions also solve different problems
An exaction asks what land, improvement or mitigation the developer may lawfully be required to provide in response to development impacts. Security asks how completion of an already established obligation is guaranteed.
Do not use a bond to hide an unlawful scope. The requirement itself must be valid before its performance is secured.
29. Security and development agreements solve different problems too
A development agreement may coordinate long-term rights, infrastructure, phasing and public benefits. Performance security can sit inside or beside that agreement as the financial assurance for selected obligations.
The agreement is the architecture of promises. The security is one of the mechanisms that makes a promise credible under failure.
30. The developer’s financing perspective
Security is not free.
A surety requires underwriting. A letter of credit can consume banking capacity and collateral. Cash held in escrow cannot be used elsewhere. Excessive security can therefore increase development cost.
That is why a competent system combines conservative initial protection with timely, evidence-based reductions. The goal is not to trap capital; it is to match public security to actual remaining exposure.
31. The municipality’s perspective
The public authority is protecting future residents and taxpayers from an unfunded handoff.
If streets and utilities are required as part of development approval, the city should not have to choose later between abandoning residents beside incomplete infrastructure and paying public money to finish a private developer’s obligation.
Security exists because the best time to allocate default risk is before default.
32. The homebuyer’s perspective
A purchaser sees a finished house and assumes the surrounding street system is similarly settled. It may not be.
In a new subdivision, roads may still be privately maintained pending acceptance. Final surfacing may be deferred until heavy construction ends. Street trees may remain under a developer maintenance period. A stormwater facility may not yet have completed final certification.
Transparent disclosure about the infrastructure stage helps residents understand what remains temporary and what security exists behind it.
33. The replacement-developer problem
Sometimes a project is sold before bonded improvements are complete.
The authority must control how responsibility and security move. It may require a replacement agreement or substitute bond before releasing the original security. Otherwise the first developer may be released before the second developer is fully bound.
This is the infrastructure version of a relay race: the baton cannot be dropped between owners.
34. Do not release the original security first
A clean substitution sequence is typically: approve the new responsible party, approve the replacement instrument, confirm it is effective, then release the old instrument according to procedure.
Reversing that order can leave a moment—sometimes a catastrophic one—when no valid security exists.
35. The annual security audit
Authorities with many active developments should not manage security only when developers request reductions.
A periodic audit can identify instruments nearing expiry, projects with no recent progress, undersecured balances, old punch lists, changes in responsible parties and completed projects still carrying unnecessary security.
This is an operational-control problem as much as a planning-policy problem.
36. A practical security ledger
| Field | Why it matters |
|---|---|
| Project / phase | Identifies the secured development |
| Approved plan reference | Defines the required work |
| Security type | Surety, letter of credit, cash or other instrument |
| Issuer | Identifies financial counterparty |
| Original amount | Initial completion protection |
| Current amount | Remaining secured value |
| Expiration / renewal date | Prevents accidental lapse |
| Completed work | Basis for reductions |
| Outstanding work | Remaining completion exposure |
| Defects / punch list | Prevents premature release |
| Acceptance status | Shows whether public handoff is complete |
| Maintenance security | Tracks post-completion risk |
37. A worked example: 80 homes, unfinished road
A subdivision has 80 lots. Underground utilities, base pavement and drainage are complete, but final asphalt, sidewalks and street trees remain. The jurisdiction allows lot sales because valid security covers the unfinished improvements.
As verified work is completed, the developer requests bond reductions. The authority inspects the work, recalculates remaining exposure and retains enough security for outstanding construction and required contingency. After final work passes inspection, the street is accepted through the required legal process. A separate maintenance security may then protect the warranty period.
The bond did not build the street. It made the incomplete period tolerable because failure had a funded backstop.
38. A worked example: developer insolvency
A developer stops work after selling half the lots. Stormwater works are incomplete and the final road surface has not been placed.
Without security, residents and government must negotiate from crisis. With enforceable security, the authority has a defined financial resource. It still needs engineering, procurement and administration, but it does not begin with an empty account.
This is the scenario every performance-security system is quietly designed around.
39. A worked example: the bond was not updated
A project stalls for four years. The original security was based on older construction prices. No extension audit recalculates the amount. The developer then defaults.
The security exists, but the completion contract now costs substantially more than the secured amount. The public is exposed to the difference.
This is why current cost schedules and extension reviews matter. A stale bond can create the appearance of protection without sufficient protection.
40. A worked example: the almost-finished project
A developer requests full release because every house is occupied. The road has its final surface, but several drainage defects remain and required street trees are failing.
Occupancy is not the release test. The authority should retain the amount required under its rules until secured obligations are satisfactorily completed or moved into an appropriate maintenance guarantee.
41. Common failure mode: security amount based on the developer’s optimistic estimate
The developer’s project budget may reflect package discounts, existing contracts or cost assumptions unavailable to a public authority after default.
Security should use a transparent approved methodology capable of funding completion under public takeover conditions, not merely reproduce the developer’s cheapest internal forecast.
42. Common failure mode: no expiry calendar
An otherwise strong letter of credit can become useless if it expires before renewal or draw action.
Security administration therefore needs automated reminders, escalation dates and clear responsibility for monitoring every instrument.
43. Common failure mode: releasing against sales rather than construction
Strong home sales do not reduce the cost of unfinished public improvements. In fact, more occupied lots can increase the consequences of default.
Bond reduction should track verified secured work and governing formulas, not commercial success.
44. Common failure mode: confusing dedication with acceptance
A plat may show land intended for a street or public facility, but that does not always mean the public authority has accepted maintenance responsibility.
The exact legal sequence varies. Every project should know when title, operation, maintenance and liability actually move.
45. Common failure mode: ignoring maintenance risk after completion
Fresh asphalt can settle. New sewers can reveal defects. Trees can die. Slopes can erode.
If the system releases performance security at completion, it should consciously decide whether a maintenance or warranty instrument is required next. Completion risk has ended; defect risk may not have.
46. The developer’s checklist
- Identify every bonded improvement from the approved plan.
- Use the authority’s current cost methodology.
- Select an acceptable security instrument and issuer.
- Confirm execution and posting before the relevant development gate.
- Track expiration and renewal dates.
- Request inspections as work is completed.
- Request reductions with evidence, not estimates.
- Correct punch-list work promptly.
- Complete public acceptance or dedication steps.
- Move into maintenance security where required before seeking final release.
47. The municipality’s checklist
- Tie security to an approved and inspectable scope.
- Use current, transparent cost estimates.
- Verify instrument form and financial counterparty.
- Calendar every expiration and renewal trigger.
- Inspect work before granting reductions.
- Recalculate exposure when projects are extended or revised.
- Preserve sufficient retainage for incomplete and defective work.
- Keep default procedures operational, not theoretical.
- Separate completion acceptance from maintenance warranty.
- Close and release security promptly when all requirements are genuinely satisfied.
48. The core idea
Town planning does not end when the subdivision plat is signed. The city still needs the physical systems that make the map real.
A line labelled “street” has to become pavement, drainage, lighting, access and maintenance. A line labelled “sewer easement” is useful only if the pipe is actually installed and functional. A tree symbol does not create a living canopy.
Performance security is the planning system’s answer to the time gap between legal approval and physical completion. It allows development to move before every public improvement is finished, but refuses to let the completion risk disappear with the developer’s promise.
The rule is simple: if the public is relying on an improvement that is not finished yet, somebody credible must still stand behind the cost of finishing it.
Authoritative reference points
- Fairfax County Land Development Services — New Bond or Agreement, explaining the use of security to guarantee completion of approved site-related public improvements and the county’s ability to use bond funds after default.
- Fairfax County — 2026 Comprehensive Unit Price Schedule, effective March 1, 2026 for new, replacement, extension and reduction bond requests and related security calculations.