Why the predicted Portland ‘Carmaggedon’ never arrived — and what London can learn about traffic evaporation

For weeks, officials at the Oregon Department of Transportation warned that closing a mile-long section of Interstate 5 through Portland’s Rose Quarter would produce “Carmaggedon.” Travel times, they said, could double or triple. Congestion might back up seven miles into Washington state. Broadcast news ran segments titled “MONDAY GRIDLOCK.” The Portland Oregonian repeated the dire forecasts.

The closure began on 11 September 2026. By 14 September, the first full commute day, the reality was strikingly different.

Google Maps data tracked a 9-mile southbound morning trip from Vancouver City Hall to Portland City Hall. Before the closure, that journey took about 26 minutes during the 8-9am peak, with a normal range of 20 to 40 minutes. On Monday 14 September, the same trip took 28 minutes. By Thursday 17 September, it had stabilised at roughly 31 minutes — an increase of about five minutes, well within the pre-construction variation.

ODOT’s own spokesperson, Rose Gerber, told Oregon Public Broadcasting that traffic levels remained “mostly normal.” Carmaggedon, once again, did not materialise.

The phenomenon is called traffic evaporation. When road capacity is reduced, a proportion of vehicle trips simply disappear: people change their departure time, switch mode, combine trips, or skip non-essential journeys. The effect is the mirror image of induced demand, the well-documented process by which new road capacity fills up with additional traffic. Both are expressions of what economists call the fundamental law of road congestion.

For London readers, the Portland case is not a distant curiosity. It directly challenges assumptions that underpin much of the capital’s transport planning and public debate.

Key facts

Why traffic evaporates

The mechanism is straightforward. Urban road networks are not fixed systems; they are elastic. When a link is removed, some travellers reroute, but many others adjust their behaviour in ways that reduce total vehicle miles travelled. A 2018 study by the UK Department for Transport, summarised in its “Evidence on the impacts of road capacity reduction,” found that traffic evaporation typically accounts for 15-25 percent of displaced trips in urban areas, with the remainder dispersing across time, mode or destination.

In Portland, the five-minute addition to the morning commute is consistent with this pattern. The feared spillover onto adjacent highways did not occur. If anything, traffic flowed more smoothly than usual, suggesting that the reduction in overall vehicle kilometres travelled eased pressure on the wider network.

The same logic applies in reverse. The Katy Freeway in Houston, Texas, widened to 23 lanes at a cost of $2.8 billion, now moves traffic more slowly than before the expansion. More supply generated more demand.

What this means for London

London has its own history of predicted gridlock that failed to arrive. When the Congestion Charge was introduced in 2003, opponents forecast economic collapse and perpetual traffic jams. Instead, traffic entering the charging zone fell by about 30 percent, journey times improved, and bus ridership rose sharply. The charge has since been expanded and adjusted, and while its effects have evolved, the principle holds: road pricing and capacity reduction do not produce the chaos that is commonly predicted.

More recently, the Low Traffic Neighbourhoods (LTNs) programme, accelerated during the pandemic, provoked fierce opposition from drivers who claimed that filtered streets would cause gridlock on boundary roads. The evidence, as reviewed by Transport for London and academic researchers, is more nuanced. Some boundary roads saw increased traffic; others did not. Overall, traffic within LTN areas fell by roughly 40-60 percent, while total vehicle miles travelled across the affected neighbourhoods typically declined by 5-15 percent, as some trips evaporated or shifted to walking, cycling or public transport.

The Portland case reinforces a point that transport planners have made for decades but that rarely penetrates public debate: traffic is not a fixed quantity. It is a behaviour, not a law of nature.

The limits of the comparison

There are important differences. Portland’s I-5 closure is temporary and partial; London’s road-space reallocations are often permanent and cover larger areas. The American context — low-density suburbs, high car dependency, limited public transport alternatives — is not a direct analogue for London’s denser, multi-modal network. A five-minute delay on a 26-minute commute in a car-dependent city is not the same as a five-minute delay on a 45-minute bus journey in a city where mode shift is more feasible.

Nevertheless, the underlying elasticity is the same. If a major freeway closure in a car-oriented US city produces only a modest increase in travel times, the case for London’s more modest road-space reallocations — cycle lanes, bus priority, filtered neighbourhoods, pedestrian improvements — is strengthened, not weakened.

What happens next in Portland

The I-5 Rose Quarter widening project, of which this closure is a part, remains controversial. The project’s stated aims include improving safety and relieving congestion, but critics argue that the induced demand effect will quickly absorb any new capacity. The Carmaggedon non-event provides fresh ammunition for opponents, who can point to the evaporation effect as evidence that the widening is unnecessary.

ODOT, for its part, has retreated from its apocalyptic language. Spokesperson Gerber now describes conditions as “mostly normal.” The media, which amplified the warnings, has been largely silent about the non-event.

For London’s transport policymakers and campaigners, the lesson is clear. The next time a road closure, cycle lane, or congestion charge expansion is announced, the inevitable claims of gridlock should be treated with scepticism. The evidence, from Portland and elsewhere, suggests that traffic finds a way to adapt — and that the forecast Carmaggedon rarely arrives.

Source: City Observatory, “Where’s my Carmaggedon?” https://cityobservatory.org/wheres-my-carmaggedon/