Research and version note This is a version 0.1 research draft in The Future Has No Representative. “Eighty years” exposes the temporal problem of long-lived infrastructure; it is not a universal design-life claim for bridges. The article concerns governance rather than an engineering safety determination. Evidence was reviewed to August 2026.
The Future Has No Representative · Article 5
Quick read
A bridge expected to serve for decades will not be designed, built, inspected, repaired, and retired by the same group. Engineers leave, contractors reorganise, departments change names, and technical standards develop. If responsibility in year eighty means that one original person must remain responsible throughout, it cannot survive the relevant period.
Long responsibility has to be carried by institutions. The project should leave readable design assumptions, material and inspection records, identify the asset owner and maintainer, and include whole-life expenditure in the budget. Every later custodian updates condition information and transfers incomplete responsibilities to the next. A change in ownership or outsourcing cannot be allowed to dissolve the chain.
OECD evidence from 2025 and 2026 on life-cycle asset management indicates that many countries still lack legally required asset plans or a complete central register. New construction is politically visible. Maintenance is dispersed and repetitive, and when it succeeds, nothing dramatic happens. That asymmetry encourages building while converting deterioration into more expensive repair for a later government.
Continuity of responsibility does not mean an unlimited warranty imposed on the original designer. Knowledge and control move. When a new risk becomes reasonably knowable, the current owner, maintainer, and regulator must act within their respective powers. A good handover enables their judgment; it does not complete that judgment in advance. Nor should spending alone be mistaken for stewardship, because emergency repair can cost more while revealing poorer maintenance.
Climate change also means that maintaining an asset “as designed” may cease to be adequate. Rainfall, heat, floods, and traffic can move beyond historical conditions. Responsibility includes testing whether the service and risk assumptions still hold. It may call for strengthening or restrictions, and eventually retirement.
The provisional conclusion is that infrastructure responsibility takes the form of a traceable chain of public stewardship. Present builders must make future maintenance possible. Each period’s owner has to monitor, fund, and hand over. Government remains responsible for ensuring that public safety is not fragmented by contract. Future managers retain authority to change the asset, but the present should not leave them a structure without information, maintenance capacity, or an identifiable final responsibility.
An opening ceremony begins the responsibility
The political account of major infrastructure centres on approval, commencement, and opening. Once complete, a commitment appears to have been fulfilled and the project moves from news to an asset register. For users, the service has only begun. Waterproofing, expansion joints, bearings, drainage, coatings, and structural members age differently. Vehicle weights and traffic patterns change, while severe weather may alter exposure.
Maintenance is not an occasional repair. Inspection has to follow risk and standards, defects must be recorded and tracked, components are renewed, and the whole facility may eventually be strengthened, replaced, or retired. If the initial business case compares construction prices while making no arrangement for this later work, approval has already created an unstated fiscal claim on future government.
Eighty years is an analytical horizon here. It does not establish that every bridge is safe for that period or automatically unsafe after it. Actual service life depends on design, materials, environment, use, maintenance, and regulation. The essential point is that it exceeds the lives of project teams, contracts, and electoral terms.
The question “who is responsible in year eighty?” cannot be answered with one name. Responsibility needs to remain recognisable despite continuing personnel replacement. Legal identity, budget mechanisms, technical records, professional capacity, and public accountability have to operate together.
How responsibility passes through changing personnel
Designers at the construction stage are responsible for applying the contemporary standard to reasonably foreseeable conditions and documenting critical assumptions. Contractors have to build to contract and quality requirements and preserve material and inspection evidence. At acceptance, the asset owner should verify documentation, defects obligations, and maintenance requirements rather than receiving only the physical work.
The centre of responsibility changes during operation. A managing entity inspects actual condition and decides repairs or restrictions from the evidence. Regulators establish and enforce a safety framework, while government ensures that ownership and public-safety duties remain clear. After a failure, historical design defects, construction error, neglected maintenance, and regulatory weakness may coexist. It would be wrong to transfer everything backward to the first engineer or erase every earlier cause simply because time has passed.
Continuity therefore attaches to roles, not to a person remaining forever. When a department is renamed or merged, or operation is outsourced, obligations should transfer by law with a clear interface. If a private contract operates for twenty years while the bridge is expected to last eighty, its ending condition, records, and renewal funding have to be considered at the contract’s beginning.
“Government is ultimately responsible” is not enough as a vague backstop. It may lead each party to believe that losses will be socialised and weaken incentives for maintenance. A defensible allocation specifies routine work, major renewal, independent inspection, emergency response, and final retirement, while government retains oversight and last-resort protection within an explicit public framework.
Information is part of the bridge
The physical structure does not explain itself. A future engineer needs design loads, geology, material batches, construction deviations, earlier defects, repair techniques, and sensor histories. Drawings without context may persist and still fail to support a reliable judgment.
Digitisation does not automatically preserve this capacity. Software formats become obsolete, fields are lost during database migrations, and a contractor’s platform may close with the contract. There can also be too much information. Millions of monitoring observations without metadata, selection logic, or a responsible interpreter are difficult to use. A durable archive preserves raw evidence, significant explanations, and the history of changes.
Handover should be a tested activity rather than the transmission of a folder. The receiver needs to confirm that records open, versions are understood, unresolved risks are identifiable, and the next inspection is known. After major works, the record must represent the actual structure rather than allowing the original design to stand in for present reality.
Knowledge here is institutionally maintained. No individual knows the entire eighty-year history; safety judgment depends on many people preserving and interpreting partial information at different dates. An institution that retains data while losing the capacity to read it leaves later managers an informational ruin rather than usable knowledge.
Why maintenance loses to construction
A new project is concentrated, named, and attached to an opening date. Maintenance is distributed and repetitive; successful work often produces only an absence of interruption. Political rewards can favour construction, while maintenance becomes an early target under budget pressure.
Deferral is not costless. Early treatments can slow deterioration, and missing the relevant window may require a much more expensive renewal. A backlog also becomes a distributional issue. Facilities in remote or lower-income communities may struggle for funding until closure or an accident attracts attention.
The OECD’s 2025 life-cycle asset indicator reports that only eight of thirty-three countries for which information was available required asset-management plans by law or regulation, and ten had no centralised fixed-asset register. Its 2026 working paper on management throughout the asset life cycle emphasises integration from planning and construction through operation, maintenance, and end of life as a foundation for value and resilience.
These indicators do not determine whether any particular bridge is safe or produce a simple ranking of countries. They expose a governance prerequisite. If a government does not know what it owns, its condition, or its renewal point, budgeting cannot readily move from post-failure reaction to risk-based planning.
Maintenance performance cannot be inferred from spending alone. High expenditure may reflect a backlog and emergency repair; low expenditure may represent efficiency or neglect. Assessment should connect cost with condition, service interruption, risk, and completed work, with independent audit testing the quality of the underlying information.
Whole-life budgeting cannot remain one estimate
Business cases often estimate life-cycle cost, although no precise eighty-year figure can be reliable. Materials, technologies, climate, and use will change. The estimate remains valuable because it forces the construction decision to acknowledge later obligations and allows designs with different maintenance requirements to be compared.
Whole-life budgeting should not end in one discounted calculation. During operation, plans need updating from condition evidence, distinguishing routine maintenance, major rehabilitation, and replacement. Funding should be identified. General revenue, user charges, dedicated funds, and contract payments have different distributional effects. Earmarking is not always best because excessive lock-in reduces flexibility; reliance on future annual appropriation alone can make upkeep a residual.
Public–private partnerships can place maintenance performance in a long contract, encouraging the builder to consider later cost. They can also hide complicated risk in contracts and off-budget obligations. If the private party exits mid-project, the public sector still has to sustain an essential service. Risk transfer is real only when duties can be enforced, information handed over, and government retains supervisory competence.
Funding should remain connected to service value. Not every old asset should be preserved indefinitely. Population and traffic move, and another connection may become safer and more economical. Responsible management includes orderly retirement rather than continuing repair because past expenditure was large.
Climate changes the meaning of normal maintenance
Maintenance has traditionally referred back to historical environment and design assumptions. Climate change can alter rainfall intensity, flood frequency, heat, and fire conditions. A bridge in sound physical condition may no longer meet the new exposure. Traffic growth, vehicle technology, and surrounding land use can change loads and consequences too.
Present design cannot know the exact climate in year eighty, but it can use scenarios, margins, and upgradable forms. Components may be made inspectable and replaceable. Monitoring can track critical developments, and planning can reserve alternative routes. A higher initial cost that preserves capacity to adapt may be more robust than choosing one future and optimising around it.
Responsibility moves with this evidence. An operating institution that mechanically follows an old checklist while ignoring new knowledge cannot rely entirely on compliance with the original design. Conversely, the first designer should not remain liable for every condition that could not reasonably have been foreseen. Whoever has information about risk and control at a particular time carries a duty to explain action or inaction.
The public should be informed about service constraints. Load reductions, lane closures, or early retirement may be unpopular and still more responsible than concealing risk. Long-lived infrastructure promises a governed service, not the preservation of one object unchanged under every condition.
Users also have responsibilities
Infrastructure governance extends beyond government and contractors. Overloading, collision, vandalism, and disregard for restrictions accelerate damage, and users can carry legal and practical duties. Communities also possess knowledge about interruption and repair timing that can improve maintenance.
“Everyone uses it” should not dilute professional responsibility into a claim that everyone is responsible. Ordinary users cannot assess concealed structures and do not control contracts or budgets. Public bodies must provide reporting channels, explain condition and restrictions, and remain accountable for professional judgment.
Transparency has to be fitted to purpose. Public inspection conclusions, risk ratings, programmes, and backlogs support accountability, while detailed security-sensitive material may need protection. An institution should neither invoke safety to refuse all explanation nor publish masses of technical data as a substitute for clarity. An intelligible condition statement and the specialist record perform different functions.
Services and responsibilities can cross territorial levels. Central government may fund construction, transfer operation to a local authority, and leave it with an inadequate revenue base. That creates a structural maintenance deficit. An asset transfer should establish the recipient’s capability and long-term income rather than treating changed ownership as vanished cost.
Provisional judgment: responsibility is a traceable chain of stewardship
The original team will not remain collectively responsible in year eighty. Long responsibility can still exist because it does not require one subject to persist in the same form. It is renewed through ownership, regulation, archives, budgets, and professional roles.
Present builders have a distinctive duty not to make future upkeep impossible from the outset. The structure should be inspectable, assumptions and actual construction documented, and life-cycle cost and transfers of duty included in approval. Every later owner monitors in the light of current knowledge, funds work, adjusts service, and hands unresolved risk over explicitly.
Institutional failure often occurs at interfaces: construction is separated from operation, capital budgets from maintenance, and a contract’s duration from the life of the asset. Each participant may complete a narrow task while the bridge receives no continuous care. Accountability therefore has to examine the chain, not merely search for the last individual to make an error.
Future managers should not be compelled to preserve everything we build. They need authority to strengthen, alter, or retire it according to changed need. What the present owes them is an asset capable of judgment: its condition can be known, its risks explained, maintenance remains feasible, and routes for change exist. A bridge of this kind is not a completed object handed to the future. It is a public relationship in which the future will have to participate.
Primary sources and further reading
- OECD, Management of Assets throughout Their Life Cycle, 2026.
- OECD, “Management of asset performance throughout the life cycle”, Government at a Glance 2025.
- OECD, Recommendation of the Council on the Governance of Infrastructure, 2020.
- ISO, ISO 55000 family — Asset management.
Series navigation: The Future Has No Representative — series overview
Discover more from Geoffrey Chen
Subscribe to get the latest posts sent to your email.