A project can hold every approval the law requires and still be stopped, delayed or redesigned because the public will not have it. Practitioners have a name for the missing ingredient, borrowed from mining: a social licence to operate. The term is now used freely in infrastructure, including in Australian government contracts, but it was defined for a different kind of industry and has never been worked out properly for this one. In a paper presented at EURAM 2025 in Florence, written with Khalil Rahi, Jin Xue and Xu Yang, we set out what the concept has to look like when it is moved into infrastructure, and proposed a way of measuring it (Ke, Rahi, Xue, & Yang, 2025).
Why the mining version does not transfer
In the extractive industries, the social licence describes a relationship between one company and one community, usually at one site, sustained over decades of continuous operation. The company that seeks approval is the company that keeps it, and it is present from exploration to closure.
Infrastructure works differently. A project passes through planning, procurement, construction and operation, and the organisation in charge changes at each transition. A government agency runs the approvals, a contractor builds the asset, an operator runs it, and each of them inherits a public that formed its views long before they arrived. The affected public is also broader: not only neighbours, but users, non-users, taxpayers, councils and advocacy groups, whose interests do not align. Applying a company-centred concept to this setting produces the odd result that legitimacy resets every time a contract changes hands.
Four questions
The framework we proposed is organised around four questions. Social contract theory answers the first two, since it explains why societal approval is owed at all. Stakeholder theory answers the last, since it explains how approval is earned in practice.
Who owns the licence
We argue that the licence is granted to the project rather than to any single organisation, and is held jointly by the lead parties across the lifecycle. This has a practical consequence. Public trust built during planning carries into construction and shapes how the contractor is received; so does public hostility. Treating the licence as continuous and cumulative means a new lead party starts from the project's existing social standing rather than from nothing.
The licence attaches to the project, not to whoever is currently holding the contract.
If that is right, the obligations need to be written down and passed on. Contracts are the obvious vehicle, and some already do this. The Sydney Roads Asset Performance Contract between Transport for NSW and ConnectSydney includes a stewardship principle requiring the service provider to “put TfNSW customers first and maintain social licence to operate”.
Who grants it
Approval comes from a network rather than a single community: local residents concerned with noise, traffic and property; interest groups concerned with environmental or sectoral outcomes; civil society organisations concerned with transparency and equity; and local councils, which are both regulators and affected parties. These groups apply different tests, so support is often partial and conditional. A project can satisfy one constituency and lose another, and the balance shifts as the project moves through its phases.
What its status is
Following the established social licence literature, we treat the status as a continuum running from rejection through tolerance to approval and active identification, resting on four elements: whether the project is seen to deliver real benefits (economic legitimacy), whether it fits local values and expectations (socio-political legitimacy), whether the project team is trusted in day-to-day dealings (interactional trust), and whether durable institutional relationships exist between the project and community bodies (institutionalised trust). Each is a precondition for the next. A project may hold economic legitimacy and still stall at tolerance because it handles complaints badly.
How it is gained and kept
Five principles follow from stakeholder theory: include affected parties throughout the lifecycle rather than at consultation milestones; communicate openly, including about uncertainty; respond to concerns and adjust when the evidence warrants it; act ethically in matters the contract does not cover; and commit for the long term, since relationships outlast delivery. None of this is novel as advice. What the framework adds is a reason these activities belong to project governance rather than to communications.
A refined framework
Working with the four questions in front of audiences, it became clear that they carry two different kinds of content, and that keeping them in one list obscures the difference. The framework has since been restructured across two dimensions.
The normative dimension states what the licence is, and holds the first three elements: ownership, grantors and status. The operational dimension states what a project team has to do, and holds three more: gaining, maintaining and repairing.
Separating repair from maintenance is the substantive change. In the original version, both sat inside a single question about obtaining and keeping the licence, which implied that a project losing support should do more of what it was already doing. That is usually wrong. Once trust has been broken, by a broken commitment, an accident or a decision made without notice, further engagement of the ordinary kind tends to confirm the grievance rather than settle it. Repair is separate work: acknowledging what happened, accepting the part of it the project owns, remedying it in a way the community can verify, and then changing the behaviour that caused it. Treating repair as its own element also makes it something a project can prepare for rather than improvise.
The six elements are then read through a set of contextual modifiers, including the type of asset, its scale, the governance and delivery model, and the political setting. These do not change what the licence is, but they change what any of it requires in practice. A motorway widening through established suburbs and a regional water treatment upgrade are not held to the same standard by the same people, and a framework that specified one standard for both would be of little use.
Sydney Metro Northwest
A social licence cannot be applied for or issued, which makes it awkward to manage. The established instrument is a survey of stakeholder representatives, which works but is slow, costly at scale, and gives a reading of one moment. We proposed public sentiment on digital platforms as a complementary measure, on the grounds that it is continuous, inexpensive, and captures not only the level of support but the events that move it.
To show how this works, we collected posts about the Sydney Metro Northwest project (and its earlier name, the North West Rail Link) from Platform X, using the full project titles as search terms. That query is deliberately narrow, and it returned 3,057 posts spanning May 2008 to December 2024. Each was classified as positive, negative or neutral using ChatGPT-4o.
Across the whole period, 36.6 per cent of posts were positive, 53.2 per cent neutral and 10.2 per cent negative. Plotting the quarterly balance of positive over negative posts, alongside the cumulative balance, shows sentiment that stayed positive throughout the lifecycle. Discussion volume peaked twice: from 2011 to 2015, covering tendering and construction, and again from mid 2018 to mid 2019, around trial running and opening.
The value is in reading the peaks rather than the average. Looking only at the second quarter of 2019 shows what the opening actually generated: praise for reduced travel times, station design and driverless operation, alongside specific complaints about station amenities and early service problems. That is a usable list. It tells an operator which issues to fix and which features to build on, while the attention is still there.
What the paper does not settle
The framework is conceptual. It has not been tested across a range of projects, and the Sydney Metro Northwest analysis is a demonstration of the method rather than evidence that the framework holds.
Two limits on the measurement matter more. First, sentiment cannot yet be mapped to the levels of the licence. We can say sentiment improved; we cannot say the project moved from tolerance to approval. Establishing that would require projects where survey data and sentiment data exist together, so the two can be calibrated against each other. Second, the people who post about a metro line on a social media platform are not the community. Digital sentiment under-represents older, poorer and less connected residents, which is to say some of the people most affected by infrastructure. The approach supplements conventional engagement and cannot replace it.
Reference
Ke, Y., Rahi, K., Xue, J., & Yang, X. (2025). Social licence for infrastructure: A conceptual framework and a sentiment-based measurement approach. Paper presented at the 25th Annual Conference of the European Academy of Management (EURAM), University of Florence, Italy, 22–25 June 2025.