Cities count the emissions they make and not the ones they buy

Everything here earned its place by changing an outcome. Nothing about consumption emissions accounting is included to round the number up.

Two ways of drawing the boundary

Territorial accounting counts emissions physically released within a defined area, which is how national and most city inventories work. Consumption accounting attributes emissions to the place where the goods and services are finally used, wherever they were produced.

A city that has lost its industry and imports its goods looks much better under the first method than under the second. Neither method is wrong, and they answer different questions and support different policies. Problems arise when a figure produced by one method is used to make a claim that only the other could support.

What territorial accounting captures well

It measures things a city can directly influence, such as heating fuel, local transport, waste treatment and municipal operations. Because those are largely within local control, the figure connects to levers a city actually holds.

Across a network, it is also comparable between places using the same protocol, which supports benchmarking and target setting. Electricity is usually included even where generated elsewhere, which is a partial step towards consumption accounting. For an authority planning building retrofit, transport and waste policy, this is the appropriate frame.

What it leaves out

Food, clothing, electronics, construction materials and aviation are largely produced or consumed outside the boundary and are frequently excluded or partially counted. For wealthy cities these categories are substantial, and their exclusion produces a figure that understates the total footprint considerably.

Over a decade, a city can reduce its territorial emissions by losing manufacturing while its residents consume the same goods made elsewhere. That outcome is a genuine reduction within the boundary and no reduction at all in atmospheric terms. Consumption accounting exists precisely to make this visible, and it requires data that is harder to obtain and more uncertain.

The methods have real limitations

Consumption accounting relies on economic input-output models with significant uncertainty, particularly for imported goods. Attribution across long supply chains involves choices that different practitioners make differently, so figures are not straightforwardly comparable.

At street level, updating the underlying data is slow, so consumption inventories often lag several years behind the period they describe. These are reasons to treat the numbers as indicative rather than precise, not reasons to ignore the perspective. Reporting both figures with their uncertainties stated is more honest than choosing whichever supports a preferred narrative.

Policy implications differ

Territorial thinking leads to heat networks, building retrofit, transport policy and waste treatment, which is where municipal power lies. Consumption thinking leads towards procurement, construction material choices, food policy and reuse, which cut across many departments.

Public procurement is the strongest consumption lever most cities hold, since it directly controls large purchases of goods and construction. Requiring embodied carbon assessment for new buildings is a consumption measure applied through a territorial planning system. Cities that report both figures tend to develop policies in both categories, which is the practical argument for doing so.

The data here is patchy — most cities do not publish it consistently.

Reading a city climate claim

Check which accounting method produced the figure, since the method determines what the number can support. Look at whether aviation, shipping and construction materials are included, as these are the usual omissions.

Note whether reductions came from measures taken or from changes in the local economy, which are different achievements. See whether electricity is counted at generation or at use, since this changes attribution significantly. A claim without a stated method is not assessable, and that is worth saying rather than assuming good faith either way.

Everything above, in order of what to do first

1. Two ways of drawing the boundary. Territorial accounting counts emissions physically released within a defined area, which is how national and most city inventories work.
2. What territorial accounting captures well. It measures things a city can directly influence, such as heating fuel, local transport, waste treatment and municipal operations.
3. What it leaves out. Food, clothing, electronics, construction materials and aviation are largely produced or consumed outside the boundary and are frequently excluded or partially counted.
4. The methods have real limitations. Consumption accounting relies on economic input-output models with significant uncertainty, particularly for imported goods.
5. Policy implications differ. Territorial thinking leads to heat networks, building retrofit, transport policy and waste treatment, which is where municipal power lies.
6. Reading a city climate claim. Check which accounting method produced the figure, since the method determines what the number can support.

The takeaway

Ask where the boundary was drawn. Every city climate number begins with that choice.

Measure the walk, not the map.