Schedule is proven. Cost is conditional. Financing is the real constraint.

Every modular pitch deck has the same slide. Two numbers, usually side by side: projects delivered 20 to 50 percent faster, costs reduced by up to 20 percent. Both come from McKinsey's 2019 report, and both are real figures from that document.¹ But they are not the same kind of claim, and the report itself says so. Treating them as a pair is the most common analytical error in the sector.

Two claims, two evidence bars

Schedule compression is a physical consequence of the method. Foundations and site work proceed while modules are built indoors; weather stops mattering for 60 to 90 percent of the work; the critical path is parallelised.⁶ This shows up on almost every project that is competently run, and it has for decades. It is robust.

Cost savings are a consequence of scale, repetition and logistics. They appear when a factory is busy, a design is repeated, transport is short and the site is simple. McKinsey said that savings above 20 percent were achievable for "scaled players" but were "the exception, not the norm," and that poorly planned logistics or material costs could produce cost increases of up to 10 percent.¹ That caveat is in the same report as the headline. It rarely makes the slide.

The schedule case

The Terner Center at UC Berkeley, which has studied factory-built housing more carefully than anyone in North America, cited 833 Bryant Street in San Francisco — a 145-unit supportive housing project — as achieving roughly 30 percent time savings and 25 percent cost savings against comparable multifamily construction.² British Columbia's temporary modular housing programme delivered 606 supportive homes across ten Vancouver sites in about eighteen months from commitment to final opening. Hong Kong stacks public-housing floors in days.

The schedule advantage has a hard edge, though. It is only realised if the site is ready when the modules are. Toronto's Auditor General found that the city's Modular Housing Initiative — designed to deliver supportive housing quickly — took between five and thirteen months from construction start to occupancy across its sites, with delays driven by incomplete site due diligence rather than anything in the factory.⁷ The factory kept its schedule. The project did not.

The cost case, honestly

Terner's own framing is the most useful we have found: hard-cost reductions of 10 to 25 percent are achievable "under the right conditions," and mature markets like Japan and Sweden see systemic savings of 10 to 20 percent because factory-built is standard practice there rather than a project-by-project experiment.² The savings are a property of the market's maturity, not of the method in isolation.

In North America today, the conditions that produce savings are the exception because the market is immature. A developer commissioning one modular building pays for a factory's learning curve, absorbs one-off transport engineering, and carries financing costs designed for a different draw schedule. The same developer commissioning the tenth building of the same design, from a factory an hour away, gets a very different number.

The productivity trap

The macro argument for modular is not any single project's savings. It is that construction is the one large sector whose productivity has not improved. McKinsey's 2017 analysis found construction labour productivity grew about 1 percent a year over two decades, against 2.8 percent for the whole economy and 3.6 percent for manufacturing.³ Closing that gap, they estimated, would be worth $1.6 trillion a year globally, and manufacturing-style production could deliver five- to ten-fold productivity gains in specific applications.³

Those are modelled potentials, not observed outcomes, and we flag them as such. But the direction is not in dispute. A factory that builds the same module a thousand times gets better at it in a way a site crew building a different building every eighteen months cannot. Modular's cost case rests on getting to a thousand. Most North American factories have not.

Financing is the binding constraint

If you want to understand why adoption sits near five percent, look at the capital stack rather than the crane. The Construction Financial Management Association describes the typical modular payment schedule: a 5 percent deposit with the letter of intent around six months before production, a material deposit of 25 percent or more about three months out, then progress billing of 30 to 35 percent at module start and again at completion, invoiced every 15 to 30 days, with a final 5 to 10 percent on transport and set.⁴

Construction lenders were not built for that. Their draw schedules assume work happens on the collateral — the site — where an inspector can verify it. Work happening in a factory two states away, on modules the developer does not yet own, is frequently treated as unsecured and underwritten against the sponsor's own credit. The result, per CFMA, is loan-to-cost ratios 5 to 10 points lower than for comparable site-built projects, higher rates, and manufacturers who discourage retainage.⁴ Every one of those pushes equity forward and upward.

A developer choosing between a modular project needing 35 percent equity and a stick-built one needing 25 percent is not being irrational when they choose the stick.Volumetric Building editorial position

This is fixable, and the fixes are institutional rather than technical: factory inspection regimes lenders trust, title mechanisms for modules in production, and — most powerfully — repeat buyers with balance sheets, which is what public housing agencies in Hong Kong, Singapore and British Columbia turned out to be.

Logistics: the 300% surprise

The industry repeats a rule of thumb that modules should travel no more than 300 to 500 miles. We looked for a primary source for that figure and could not find one; treat it as lore. What we can source is what happens when transport is planned late. MBI documented an Idaho project where oversized loads requiring pilot cars and police escorts increased shipping costs by 300 percent — an expense that earlier route planning could have cut by three-quarters.⁵

Transport is where modular's cost case most often quietly dies. Module dimensions are set by road rules, not by architecture. Permits, escorts, bridge clearances and crane access are project costs that stick-built construction never incurs, and they are set by decisions made in schematic design. A modular project whose transport engineer joins after the modules are sized has already lost money it does not know about.

What to do with this

Justify a single modular project on schedule certainty and quality, and treat cost parity as the target rather than a discount. If your business case needs a 20 percent saving to work, you need a repeat programme, not a project. Bring the transport engineer and the lender into the room before the modules are dimensioned. And if you are a public agency, understand that your greatest contribution to modular economics is not a grant but a pipeline: a credible commitment to buy the same thing many times.