Imagine being stuck on the same highway every morning. Cars barely move, a journey that should take twenty minutes takes an hour, and the solution seems almost embarrassingly obvious. There are too many cars and not enough road. Add another lane, widen the highway, and the same traffic will finally have more space to move. For a while, that may be exactly what happens. The new road opens, journey times fall and congestion eases. Yet several years later, drivers may find themselves sitting in traffic again, except now on a larger highway. This seemingly contradictory result is explained by one of the most interesting ideas in transport economics: induced demand.

The basic idea is that the amount people drive is not fixed. It depends partly on how costly driving is, and one of the largest costs of travelling is time. Suppose commuting to work takes an hour by car but forty-five minutes by train. Even if you prefer driving, the extra time might persuade you to take public transport. Another commuter might still drive but leave home unusually early to avoid rush hour. Someone else might reject a job farther away because the daily journey would be exhausting. If a new road suddenly reduces the drive to thirty-five minutes, none of these people's preferences have necessarily changed. What has changed is the effective price of driving. The cost of petrol may be identical, but the opportunity cost of spending time on the road has fallen. Just as consumers tend to buy more of a product when its price falls, people tend to make more or longer car journeys when the time cost of driving falls.

This is why adding road capacity can produce traffic that did not previously appear to exist. Economists sometimes describe this as latent demand. Imagine a highway currently carries 100,000 journeys every day. Engineers calculate that increasing its capacity to 130,000 cars should eliminate the shortage of road space. But the original 100,000 trips do not represent every journey people would like to make. Perhaps another 10,000 commuters currently use trains because driving is too slow. Another group travels before or after rush hour. Some people avoid optional journeys altogether, while others choose homes, jobs or businesses partly because particular locations are too difficult to reach. Once the highway improves, some of these decisions change. Demand that congestion had previously suppressed begins to appear.

Economists Gilles Duranton and Matthew Turner found powerful evidence of this when they examined highways across American cities. Their research found that increases in interstate highway lane kilometres were associated with roughly proportional increases in the total kilometres driven by vehicles. They called this relationship the fundamental law of road congestion. Importantly, the additional traffic did not come from only one source. Existing residents drove more, commercial traffic increased and people migrated toward areas that had become easier to access. A road therefore does not simply respond to where economic activity already exists. By changing accessibility, it can influence where future economic activity occurs.

Think about what this might mean in a city such as Mumbai. Suppose a new road cuts a ninety-minute journey between two parts of the city to forty-five minutes. The immediate beneficiaries are obvious: drivers already making the trip save time. But the larger economic effects may only emerge later. A worker might now accept a job that previously seemed too far from home. A family may move farther from a commercial district because lower housing prices suddenly compensate for the commute. Delivery companies can serve a larger area, businesses may open in newly accessible locations, and some commuters who previously relied on public transport might start travelling by car. Each decision can be perfectly rational for the individual making it. Yet thousands of individually rational responses can collectively fill part of the capacity the new road created.

This does not mean that building roads is pointless. New transport infrastructure can connect previously isolated areas, remove genuine bottlenecks and create substantial benefits by reducing travel times. The more important point is that those benefits cannot be calculated by assuming people's behaviour remains unchanged. A 2026 review commissioned by the UK Department for Transport examined how additional road capacity generates extra travel and how this should affect traffic forecasts, project appraisal and estimates of value for money. If planners assume that today's number of journeys will simply spread across tomorrow's larger road network, they can overestimate how much permanent congestion relief the project will deliver.

This creates an unusual economic problem because congestion is partly a problem of scarcity without an effective price. Road space at 8:30 on a Monday morning is extremely scarce, while the same road at 2 a.m. may have enormous unused capacity. Yet in many cities, the monetary price of entering the road barely changes between those two periods. When a scarce resource is made available for little or no additional charge at the moment everyone wants it, quantity demanded can exceed available capacity. In most markets, a shortage might produce an actual queue outside a shop. On a highway, the queue is simply made of cars. Drivers pay for the scarce road space not primarily in money, but in the time they lose waiting.

Congestion becomes even more interesting because each driver creates a negative externality. When I enter an already crowded road, I consider my own costs: fuel, tolls and the amount of time I expect the journey to take. What I do not naturally account for is the small additional delay my car imposes on every other person using the road. One additional vehicle may add only seconds to another driver's journey, but when thousands of vehicles impose these costs on one another, the total social cost can become enormous. The private cost of driving is therefore lower than the full cost imposed on society, which can lead to more road use than would occur if drivers faced the entire cost of their decision.

This is the economic reasoning behind congestion pricing. Instead of responding to every traffic jam only by increasing the supply of roads, a city can attempt to manage demand by making it more expensive to use particularly congested roads at particularly congested times. Some drivers may switch their journey to another hour, use public transport or decide that a lower-value trip is no longer worth making. Economically, the idea is attractive because it places a price on a scarce resource and makes drivers face more of the external cost they create. Politically, however, it is much harder. If reliable public transport is unavailable, a congestion charge can fall heavily on commuters who have little practical choice but to drive. The policy therefore raises questions not only about efficiency, but also about equity and how the revenue should be used.

The larger lesson from induced demand reaches far beyond highways. Policies change the environment in which people make decisions, so they can change the behaviour the policy was originally designed around. More parking can encourage people to drive. Faster transport can influence where homes are built and where workers look for jobs. Better infrastructure can make previously unattractive locations economically valuable. None of this makes infrastructure undesirable. It simply means that policymakers cannot treat demand as a number written permanently into a spreadsheet.

Perhaps that is the real paradox of building more roads. A congested new highway may look like proof that nothing improved, when part of the traffic exists precisely because the road made travelling more attractive. The question for governments is therefore not simply, "How much road do we need for the traffic we have?" It is also, "How much traffic will exist once we build the road?" Economics becomes useful the moment we recognise that those are not the same question.

Sources

  • Gilles Duranton and Matthew A. Turner, "The Fundamental Law of Road Congestion: Evidence from US Cities," American Economic Review — aeaweb.org
  • UK Department for Transport, "How Additional Road Capacity Affects Travel Demand" (2026) — gov.uk
  • National Bureau of Economic Research, "The Fundamental Law of Road Congestion: Evidence from US Cities" — nber.org
  • Federal Highway Administration, "The Transportation and Land Use Connection" — environment.fhwa.dot.gov