A Short History of Prefab

The modern history of prefabrication has a spectacular opening act. For the Great Exhibition of 1851, Joseph Paxton designed the Crystal Palace in London — an enormous exhibition hall of cast iron, wrought iron and glass, covering some nineteen acres, erected in Hyde Park in roughly nine months and later dismantled and rebuilt elsewhere. Its secret was prefabrication and standardisation: a modest kit of standard components — identical cast-iron columns and girders, a standard pane of glass, repeating bays — manufactured off-site in quantity and assembled on site at extraordinary speed. It proved, more than a century before the word 'DfMA' existed, the core off-site proposition: design around a small set of repeated, factory-made parts and you can build big, fast and well.

The 19th century had other lessons too. The same industrial logic produced prefabricated iron buildings shipped around the British Empire — churches, warehouses, even whole cottages sent in pieces to the colonies and the goldfields, where skilled labour was scarce and speed mattered. Cast-iron building fronts were mass-produced and bolted onto commercial streets. The railways, themselves a triumph of standardised iron parts, made both the manufacture and the distribution of such components possible. Prefabrication, in other words, was a natural child of the Industrial Revolution: once you could make identical metal parts in a factory and move them cheaply by rail and ship, extending that logic to buildings was almost inevitable.

But already the limits showed. The Crystal Palace worked brilliantly because it was a highly repetitive, dimensionally disciplined, single-purpose structure — exactly the kind of project off-site rewards. Exported iron buildings worked where conventional building was impractical. What did not follow was a general transformation of ordinary building. Houses, shops and offices — varied, one-off, rooted in local craft and materials — largely went on being built the traditional way. The 19th century thus established both halves of the prefab story that would repeat ever after: a dazzling demonstration of what standardised, factory-made parts can do for the right kind of building, and a quiet failure to displace conventional construction for everything else.

If the Crystal Palace showed prefab for the monumental, the early 20th century showed it for the ordinary home. Between roughly 1908 and 1940, Sears, Roebuck and Co. sold tens of thousands of mail-order kit houses across the United States — the 'Sears Modern Homes'. A customer chose a design from a catalogue; the company shipped the pre-cut lumber, fittings, fixtures and a thick instruction book by rail; and the house was assembled on the buyer's lot. These were not volumetric modules but kit-of-parts prefabrication: standardised, pre-engineered packages that cut waste and skilled-labour needs and brought decent housing within reach of ordinary families. Other firms ran similar catalogues. The kit house proved that off-site logic could democratise the home — and also that it depended utterly on reliable demand and distribution; when the Depression and changing markets hit, the programmes wound down.

The two world wars gave prefab its greatest spur and its most lasting image. War demanded huge quantities of buildings fast — barracks, hospitals, factories, and then, after 1945, homes for bombed-out populations with construction labour still in uniform. Britain's post-war 'prefabs' — factory-made emergency bungalows — were rolled out in hundreds of thousands as a stopgap; many were genuinely liked by their residents and some outlived their intended short lives by decades. Around the world, wartime and post-war programmes produced vast amounts of quick, factory-assisted housing and huts.

But this era also fixed prefab's reputational problem, one the field still fights. Because the most visible prefabs were emergency, temporary, cheap and uniform, 'prefab' came to mean, in the public mind, exactly that: a flimsy stopgap, a hut, a downgrade from a 'real' house. The association was unfair to much of the work — and it ignored that the wartime drivers (desperate speed, scarce labour, mass need) were precisely the conditions off-site rewards — but it stuck. Much of the modern industry's preference for the neutral term 'off-site construction' over 'prefab' is an attempt to escape this mid-century shadow. The deeper lesson is subtler: prefab surged when the conditions fit (overwhelming demand, labour scarcity, a premium on speed) and receded when they passed — a pattern of fit, not of steady triumph.

The 1960s and early 1970s gave prefab its most architecturally ambitious chapter — and its most instructive over-reach. In Japan, the Metabolist movement imagined cities as living, growing organisms: permanent cores serving interchangeable, factory-made capsules that could be plugged in, replaced and renewed like cells in a body. Its built icon was Kisho Kurokawa's Nakagin Capsule Tower in Tokyo (completed 1972): a pair of concrete cores onto which were bolted scores of small prefabricated living capsules, each a compact fitted room craned into place. It was a genuine realisation of volumetric, replaceable, plug-in modular architecture — and a vivid demonstration of the idea.

In Montreal, for Expo 67, Moshe Safdie built Habitat 67: a startling housing complex of stacked, staggered precast concrete boxes, each a module, arranged so that roofs became terraces and every dwelling got light, air and a garden. It aimed to bring the qualities of a suburban house to dense urban housing through volumetric prefabrication, and it remains one of the most admired and studied modular buildings ever made.

And yet the utopian moment did not scale. Nakagin's capsules were, in the event, almost never actually replaced as the theory promised; the building aged, its services dated, and after long decline it was dismantled in 2022 — the 'renewable' architecture proving, ironically, hard to renew. Habitat 67 was brilliant but expensive, and the mass, affordable, repeatable version it was meant to prototype never followed; its modules were less repetitive and its construction more costly than the industrialised dream implied. Both buildings were triumphs of vision and real achievements of volumetric prefab — and both showed the recurring gap between the manifesto and the economics. The plug-in city is a seductive idea; but replaceability, true repetition and affordable scale are hard, and prefab's history is littered with beautiful demonstrations that the market did not, or could not, reproduce at volume. That gap — between what prefab can *show* and what it can *sustain* — is the hinge of the whole story.

Step back from the individual buildings and a pattern emerges that matters more than any of them: prefab moves in boom-and-bust cycles. A wave of enthusiasm and investment builds — a new technology, a housing crisis, a burst of capital, confident forecasts that off-site will soon take a large share of construction. Factories are built or expanded on those forecasts. Then demand proves thinner, lumpier or more bespoke than promised; the factories, which need steady high-volume throughput to cover their fixed cost, run under-utilised; margins collapse; and firms fail, sometimes spectacularly — including, in recent years, well-funded modular-housing companies in the US, UK and elsewhere that promised to revolutionise homebuilding and then went under. The lesson is relearned, the field retrenches, and a few years later a new generation announces off-site as the future once more.

Why does this keep happening? Several honest reasons. First, a factory is a fixed-cost machine: it is only cheap when it runs full and steady, so a modular business lives or dies by a reliable pipeline of repetitive work — and construction demand is notoriously cyclical, fragmented and bespoke. Second, prefab's benefits are conditional, not automatic: they require repetition, volume, early design freeze and good logistics, and when a project or a market lacks those, off-site is not cheaper or faster and can be worse. Third, the industry repeatedly confuses a technical demonstration (yes, you can build a brilliant modular tower) with a viable business model (can you sell enough near-identical modules, steadily, to keep the factory full and the warranties honoured?). Fourth, switching an entire construction ecosystem — trades, financing, regulation, supply chains, client habits — is slow and hard, whatever the factory can do.

None of this means prefab is a failure; it means prefab is a tool with a specific shape, and the busts happen when enthusiasm forgets the shape. The genuine, durable successes — precast for repetitive structures, bathroom pods, modular hotels and student housing, rapid institutional buildings — are exactly the cases where repetition, volume and speed align with the method. The modern revival (next lesson) is real and has new drivers; but the single most useful thing this history gives you is a calibrated scepticism: when someone says off-site will transform all of construction soon, remember 1851, 1945, 1972 — and ask where the repetition, the steady demand and the frozen design are going to come from.