There is a peculiar ritual that plays out with remarkable consistency across the world’s capital cities. A government, freshly energised by a new administration or a once-in-a-generation development mandate, commissions an international benchmarking study. Consultants are hired, flights are booked, PowerPoint decks multiply. Somewhere in the resulting report, Singapore is praised. Copenhagen’s cycling infrastructure receives a respectful paragraph. Seoul’s transit system is invoked as an aspirational benchmark. And then the report is shelved, or worse, selectively mined for the conclusions that were politically convenient before the study began.
This is not a cynical reading of how governments work. It is simply an accurate one, and understanding why it happens is essential if cities are ever going to get better at learning from each other.
The core problem is not a lack of information. Governments today commission more urban analysis than at any previous point in history. The real problem is a structural mismatch between what advice is designed to deliver and what decision-making systems are actually capable of absorbing. Advice tends to be comprehensive. Politics tends to be selective. The gap between those two tendencies is where good ideas go to die.
When governments commission benchmarking studies on cities, they typically ask the wrong questions first. The instinct is to find the best example of something – the best transit system, the best affordable housing model, the best approach to waterfront regeneration – and then figure out how to import it. This produces a kind of planning tourism, an endless circuit of study visits and case studies that treats urban policy as though it were a consumer product that can be lifted from one context and installed in another. The consultants who deliver these studies are rarely to blame. They produce what is asked of them. The problem is the commissioning logic itself, which tends to search for models rather than mechanisms.
The distinction matters enormously. A model is a finished outcome – Amsterdam’s canal-side density, Medellín’s cable cars, Tokyo’s punctual trains. A mechanism is the underlying institutional, financial, or political process that made the outcome possible. Governments almost always want the model. What they actually need is the mechanism. And mechanisms are far harder to transfer because they are embedded in specific legal traditions, specific fiscal relationships between central and local government, specific cultures of administrative professionalism, and specific histories of how trust was built or destroyed between the state and its citizens.
Consider how often London is cited in discussions of urban transport governance. The creation of Transport for London in 2000 under a directly elected mayor is held up as a template for integrating fragmented transit systems under unified political authority. What gets lost in the retelling is that it required decades of advocacy, a failed predecessor body, a specific devolution settlement negotiated with central government, and a political moment in which both the institutional design and the political personality of Ken Livingstone aligned in ways that were difficult to predict and impossible to simply replicate. None of that context travels in a benchmarking table.
There is also a deeper epistemological problem with how governments relate to urban advice, which is that they tend to commission it at the wrong moment. The ideal commissioning moment, in theory, is before decisions are made – when options are genuinely open and evidence could shape direction. In practice, advice is most commonly commissioned after the political direction has already been set, serving primarily to legitimate choices already made rather than to interrogate them. This is not conspiracy. It is a predictable consequence of how political mandates work. By the time a government is in a position to act on something, it has usually already decided what it wants to do. The advice that follows tends to be retrofitted around the decision.
This produces a particular kind of useless document: one that is technically rigorous, internationally grounded, and fundamentally irrelevant to the actual choices being made. Governments often sense this, which is why benchmarking reports so frequently gather dust. The officials who commissioned them move on. The political moment that prompted them passes. The institutional memory required to act on their recommendations disperses. And the next administration commissions a new study, which travels the same circuit and arrives at the same destinations.
What would it look like for governments to commission urban advice well? It would start with a more honest account of what the advice is actually for. Is it genuinely exploratory, meant to surface options that haven’t been considered? Is it politically protective, providing cover for decisions that have already been made? Is it capacity-building, designed to leave officials with better analytical tools rather than a static document? Is it diagnostic, focused on understanding the specific constraints of a particular context rather than describing what other places have done? These are not the same exercise, and conflating them produces advice that is too general to be diagnostic, too local to be exploratory, and too politically entangled to be genuinely protective.
The governments that get the most out of urban advice are generally those that have already invested in the internal capacity to challenge it. A city with a strong planning department staffed by people who have spent careers thinking about urban form will read an international benchmarking study very differently from one that depends entirely on the consultants to do the analytical work. In the first case, the advice becomes an input into a sustained institutional conversation. In the second, it substitutes for one. The paradox of commissioning advice is that the governments that need it least tend to use it best.
There is also the problem of prestige bias, which distorts the geography of learning in ways that are rarely acknowledged. The cities that are most frequently cited in urban benchmarking studies are those that have invested heavily in exporting their models – through international advocacy, through the travel schedules of their planning commissioners, through the conferences that shape the professional consensus about what good urbanism looks like. Singapore, Copenhagen, and Curitiba are not necessarily the most instructive cases for a mid-sized African or South Asian city grappling with informal settlement growth and constrained municipal finance. They are simply the most legible, the most photographed, the most articulate about their own success. The cities that might offer more genuinely transferable lessons – those that have managed incremental improvement under conditions of scarcity, political instability, or rapid demographic change – are systematically underrepresented in the literature that informs how governments commission advice.
This matters because the implicit aspiration embedded in most benchmarking studies is one of convergence: the idea that cities everywhere are on a spectrum from less developed to more developed, and that advice is a form of acceleration along that spectrum toward a shared destination. The reality is considerably more plural. Cities are not converging on a single model of good urbanism. They are managing different tensions, serving different populations, operating within different constitutional arrangements, and responding to different climate and ecological pressures. Advice that doesn’t take this seriously tends to produce recommendations that are technically impressive and contextually hollow.
None of this means that governments should stop commissioning urban advice, or that international benchmarking is without value but they should focus more on suitability and effectiveness within the existing capacity. What makes a spatial strategy effective is not its content but its governance, and that governance arrangements are not transferable abstractions but specific configurations of authority, legitimacy, monitoring, and time that have to be built from within.
That insight applies equally to the business of commissioning advice itself. The most useful thing a government can do before calling the consultants is to ask what it would actually do with the answer. If that question doesn’t have a clear institutional home, a committed champion, and a realistic implementation pathway, then the most sophisticated analysis in the world will remain exactly what so many benchmarking reports already are: a beautifully produced account of what other people have managed to figure out.
cover image: pxhere.com
