After Density: Urban Form Is Becoming the Next Frontline of Climate Policy

Public discussion about cities and climate has long been dominated by two metrics: the urbanization rate and population density. The former answers "how many people are moving into cities," while the latter answers "how tightly packed people are." Both matter, yet together they sidestep a third question—what shape cities have actually taken in space.

This study, published in Scientific Reports, brings this long-treated-as-background variable to the foreground. Using a time span from 1975 to 2020, the researchers constructed a long-term panel covering more than 11,000 urban centers worldwide, placing harmonized definitions of urban boundaries, high-resolution residential and road transport emissions data, and multiple sets of urban geometry metrics into the same analytical framework. The conclusion is straightforward: more compact urban form is associated with lower per capita emissions in the two sectors. The authors also make clear that this is a conditional association, not a causal effect.

The real weight of this study lies not in the judgment that "compact cities are lower-carbon"—which is hardly novel to the instincts of planning—but in what it reveals: the strength of this association is not a universal constant, but changes systematically with stage of development. In low- and middle-income countries, the association is stronger and more robust; in high-income regions such as Europe and North America, it is weaker, or even statistically insignificant.

If this asymmetry holds, then "compact cities" is not a universally applicable policy doctrine, but a strategic asset with a clearly defined range of applicability and period of validity.

Form Is Not Background, but an Asset Locked In by Planning

Urban form is shaped jointly by land-use regulation, road networks, parcel subdivision, and building permits, making it one of the few emissions dimensions that policy levers can directly reach. More critically, it is highly path-dependent: a building's lifespan is measured in decades, and once a low-density suburb takes shape, its commuting structure, pipeline network costs, and energy-consumption profile tend to keep replicating themselves for the next half century. The land-use red lines drawn today are the climate legacy of the mid-century.

Questions Density Cannot Answer, Geometry Can

Density is a scalar. It cannot distinguish "high-density but highly dispersed clusters of towers" from "high-density and continuously compact neighborhoods," nor can it depict how built-up areas extend through space or their degree of fragmentation. The study therefore introduces multiple sets of geometry metrics, attempting to separate urban form from density and city size, and to observe its relationship with residential energy consumption (building heating and cooling) and road transport emissions in isolation.These two sectors were not chosen by chance. They are precisely the two types of emission sources most susceptible to urban form: the former depends on the relationship between building surface area and volume, as well as shading and ventilation conditions; the latter depends on travel distance, mode choice, and road network efficiency. The spatial logic of industrial emissions and electricity production is more determined by energy systems and industrial chains, and its connection to urban geometry is much more indirect.

Asymmetric evidence points to stage of development, not statistical noise

The study’s most important finding is the regional differentiation in the strength of the association: it is stronger and more robust in low- and middle-income countries; in high-income regions it is weaker or even insignificant.

This is not hard to understand. In high-income cities, the urban form dividend has largely already been realized and locked in. Their current marginal room for emission reduction comes more from power grid decarbonization, energy efficiency retrofits of the building stock, and transport electrification—the effectiveness of these levers is partly substituting for the effectiveness of form. When a city’s electricity mix continues to decarbonize, the marginal impact of commuting distance on per capita emissions naturally shrinks.

By contrast, the situation is exactly the opposite for economies in the rapid urbanization stage. Their built environment is still being written on a large scale, and form’s ability to shape cooling and heating loads and total travel demand is at its peak. The same planning decision produces non-equivalent emission consequences at different development stages. This constitutes a clear but often overlooked judgment: urban form is a stage-sensitive policy resource, whose marginal effectiveness declines as income levels rise.

An underestimated premise: the city itself has no uniform definition

To understand why such conclusions have been late in coming, one needs to see a more fundamental problem: globally, there is no uniform definition of “city.”

Self-reported city emission inventories are usually compiled by local governments using different methods, and the administrative boundaries of the “city” or “municipality” on which they are based are themselves products of historical, economic, and political processes—they may include large tracts of farmland, or cover only a corner of the core built-up area. Comparisons between cities using administrative boundaries as units lack comparability from the outset. This is precisely why researchers have turned to harmonized definitions of urban centers and distinguish direct emissions from consumption-based carbon footprints. Consumption-based footprints require environmentally extended input-output tables to track emissions from production processes and attribute them across countries; the concept is more complex, and the attribution chain is longer; existing studies have used this to estimate the carbon footprints of tens of thousands of cities worldwide.

The debate over accounting scope appears to be a technical issue, but in fact concerns governance: if cities are not comparable, policy learning, performance comparison, and horizontal accountability among cities lack a foundation. Cities have repeatedly been confirmed as actors in climate action—relevant United Nations assessments hold that cities contribute more than 70% of global carbon dioxide emissions—but the measurement system supporting their actions is still in a patchwork stage.

Cities are already climate actors, but the toolbox is not completeThe difficulty in advancing compactness policies lies not in engineering but in political economy. It touches land property rights, housing affordability, local fiscal structures, and commuting radii, directly affecting the prices of existing assets and established constellations of interests. Yet cities’ authority over land use, energy, transport, and finance is often dispersed between national and local levels, and any adjustment to urban form must cross multiple levels to be implemented.

This creates an intriguing gap: cities are expected to shoulder a key share of global emissions reductions, but the tools that shape their form are precisely the most fragmented and the most easily diluted by local politics.

The Global South’s Window and the North’s Marginal Problem

Over the coming decades, the net increase in the global urban population will fall mainly in cities in Asia and Africa. This means that the decisive writing of this century’s urban form is taking place in the South. If, during this window, the suburbanization path characterized by low-density sprawl is replicated, its lock-in costs will be borne by several generations; if a contiguous, mixed-use, transit-oriented spatial order can be established during the construction peak, the benefits will likewise accumulate over the long term.

But compactness in the South cannot simply copy the North’s normative template. Informal settlements, the spatial dispersion of employment, housing affordability, and existing social networks all mean that “compactness” must be premised on affordability and accessibility of livelihoods. Density increases detached from these conditions will only convert the emissions reduction agenda into a distributional agenda.

For the North, the conclusion is closer to a reminder: the form dividend has become marginal, and the focus of climate policy should be more clearly placed on energy systems and retrofitting the existing stock, rather than treating “greater compactness” as a panacea. At the same time, the main tension in Northern cities over the issue of form is shifting from emissions reduction efficiency to housing supply, fiscal sustainability, and spatial equity.

Long-Term Judgments

First, land-use regulation should be formally incorporated into the climate policy toolbox, rather than continuing to be treated as a technical matter for planning departments.

Second, comparable urban data are a public good. Without harmonized city-boundary and sectoral emissions data, urban climate governance can only remain at the declaratory level.

Third, as the weight of Southern cities in population and construction scale rises, the center of gravity of global urban knowledge will shift accordingly. With regard to the relationship between urban form and emissions, the most explanatory lessons may increasingly come from rapidly expanding low- and middle-income cities, rather than from highly established Northern metropolises.

Fourth, restraint is needed. This study provides conditional associations rather than causal evidence, and there may still be endogeneity between urban form and emissions that is difficult to fully eliminate. Reading it as a universal law that “compactness equals low carbon” would precisely repeat the most common error in urbanization discussions of recent decades—compressing complex local processes into a portable indicator.

Cities are, after all, long-term assets. Their shape is neither an aesthetic preference nor a short-term policy variable, but a spatial commitment spanning political cycles. At a time when the climate issue is moving from the commitment stage to the implementation stage, whoever preserves the geometric order of cities during the construction peak is setting an upper bound in advance for the emissions curve of the second half of this century.Source: Urban compactness and carbon emissions: global evidence over the period 1975–2020, Scientific Reports, https://www.nature.com/articles/s41598-026-54461-9