Modular is usually greener. It is not automatically greener, and the bylaws are about to check.

Construction produces about a third of the world's waste and, by the widest accounting, around 40 percent of its carbon emissions.⁵ Any method that promises to build the same thing with less of both deserves a hearing. Modular's promise on waste is well supported. Its promise on carbon is real but conditional, and the conditions are exactly the ones its marketing tends to skip. We take them in order of how strong the evidence is.

Waste: the strongest claim

The most-quoted sustainability figure in the sector is that volumetric off-site construction can cut on-site waste by up to 90 percent compared with traditional building. It originates in benchmarking by WRAP, the UK resource-efficiency body, and is reproduced in industry literature; one volumetric manufacturer, Yorkon, reported total manufacturing waste of 1.8 percent, of which 65 percent was recycled or reused.¹ The mechanism is not mysterious. A factory buys materials in optimised lengths, cuts them on jigs, stores offcuts for the next module and controls inventory the way any manufacturer does. A site orders with contingency, cuts on sawhorses and fills a skip.

The 90 percent figure is an upper bound, and it measures site waste, not total waste — some of what disappears from the skip reappears, in smaller quantity, at the plant. But the direction and rough magnitude are corroborated across studies going back to Quale and colleagues' 2012 life-cycle comparison of modular and conventional US homes in the Journal of Industrial Ecology, still the benchmark academic treatment.² If you cite one modular sustainability number, cite this one, and say "up to."

Site disruption and logistics

The second well-evidenced benefit is what does not happen around the site. The World Economic Forum reported on Camp Hill, a modular residential project in Birmingham, England, where off-site construction eliminated roughly 3,700 deliveries and cut vehicle movements by 56 percent against an equivalent conventional build.⁵ Fewer trucks means less diesel, less noise, less dust and shorter road closures — benefits that fall on neighbours rather than owners, which is why planning authorities in dense cities have started to notice them.

This is a single project and should be read as illustrative, not typical. The mechanism generalises: a module arrives on one truck carrying what would otherwise have been dozens of trade deliveries. The magnitude depends on the distance the modules travel, which brings us to the harder question.

Embodied carbon: the honest picture

Camp Hill also reported a roughly 35 percent embodied-carbon reduction, at 915 kgCO₂e per square metre — an improvement, though still above the RIBA 2030 target of 800.⁵ Results like that are common in the literature and they are real. But recent peer-reviewed work complicates the general claim. Life-cycle assessments of steel-framed modular high-rises in Hong Kong, published in Building and Environment in 2024, found higher initial embodied carbon than conventional construction, because each module must carry its own structure strong enough to be craned, transported and stacked, producing redundant steel at every mate line.³ A parallel study of concrete modules found similar upfront penalties. The same body of work finds the penalty partly or wholly recovered at end of life, when modules can be unbolted and reused rather than demolished — but only if someone actually does that, decades from now.³

The synthesis is not complicated, but it is rarely stated: low-rise timber modular built close to site is almost certainly lower-carbon than the conventional equivalent; steel modular towers shipped long distances may not be, at least until end of life is counted. The MDPI 2024 assessment of modular steel social housing in the UK is a useful open-access model of how to do this comparison properly, and a template for what North American manufacturers should be publishing about their own products.⁴

Policy is forcing the comparison

That forcing has begun. Since October 2023 Vancouver's Building By-law has required every Part 3 building — broadly, anything larger than a house or small residential — to report whole-building embodied carbon at permit and to stay below twice a baseline. From January 2025, projects must demonstrate a 10 to 20 percent reduction from that baseline, depending on building type, and meet one of three responsible-sourcing criteria. It is the first such bylaw in Canada.⁶ Federally, the Greening Government Strategy and the Buy Clean agenda push low-carbon procurement of steel and concrete for federal construction, although the thresholds specific to prefabricated buildings are not yet defined in a primary directive we can cite.⁷

For a modular manufacturer this is an opportunity disguised as a compliance burden. A factory building the same module repeatedly can produce an environmental product declaration once and reuse it; a site-built project must model every building from scratch. The manufacturer who has the numbers will win the permit race in Vancouver and, soon, elsewhere. The manufacturer who has only the brochure will not.

Worker safety

A benefit that rarely appears in sustainability discussions but belongs there. Historically, US construction has had a fatal-injury rate roughly four times that of manufacturing — 14.1 against 3.6 per 100,000 workers in the Bureau of Labor Statistics' comparative analysis, which is dated but structurally informative.⁸ Construction's nonfatal injury rate has improved, to 2.3 per 100 full-time workers in 2023, close to the all-industry average.⁸ Moving hours of work from a scaffold in the weather to a bench at waist height under a roof does not eliminate risk, but it moves it into an environment with a dramatically better record. We would like to see a modular-specific injury dataset; none that we could find exists.

Disaster rebuild: Altadena and Lytton

The most vivid current test of factory-built speed is disaster recovery, and two cases show its limits as well as its promise. After the January 2025 Eaton and Palisades fires destroyed more than 16,000 structures in Los Angeles County, Samara — the prefab company spun out of Airbnb — installed its first factory-built accessory dwelling for an Altadena survivor through the non-profit Steadfast LA: a 950-square-foot unit with fire-resistant fibre-cement cladding, a metal roof and smoke-filtering ventilation, targeting 90 to 120 days from permit to occupancy from a Mexicali factory producing about two units a day.⁹ That is the method at its best: a standardised product, a repeat client, a clear need.

Lytton, British Columbia, is the counter-example. The village burned in June 2021. Its first rebuilt municipal building opened only in 2024–25, and the community remains under financial strain.¹⁰ Factory speed was never the constraint; contaminated-site remediation, archaeology, insurance and permitting were. Manufacturing capacity is necessary for fast rebuilding and nowhere near sufficient. Anyone selling modular as a disaster solution should be asked what they plan to do about the eighteen months before the first foundation is poured.

What to do with this

Lead with waste and site disruption, where the evidence is strong, and be specific about carbon, where it is conditional. If you manufacture, commission a whole-life assessment of your standard module and publish it; Vancouver has already made that the price of a permit. If you develop, ask for it. If you make policy, note that embodied-carbon rules reward exactly the repetition that modular economics need — and that disaster-recovery programmes should be buying pre-approved designs before the fire, not after.