Defining a Better Transport System, Part 1: Getting Productivity Right
Key Takeaways
- This is part of a series of posts on what a better transport system would look like from different perspectives, starting with economic productivity.
- Economic productivity in transport isn't about how fast one trip is; it's about how efficiently the whole system moves people and goods.
- Cars can harm productivity in multiple ways beyond congestion, including through land used for parking, crash costs, pollution, and sedentary-lifestyle health effects.
- Avoiding the productivity argument on transport cedes it to those who wrongly use it to defend car-centric status quos.
- Travel time savings became the default productivity measure, but it is flawed.
- We misuse travel time savings in four ways: treating origins and destinations as fixed, assuming travel time is dead time, optimising individual trips instead of the system, and ignoring demand management.
- Three structural failures block a genuinely productive system: subsidising driving, under-investing in more space-efficient modes, and restrictive land use.
- The transport system a productivity-first analysis leads to closely resembles the one people reach when they start from environmental, wellbeing, or livability goals instead.
What next?
Are you falsely associating economic productivity with a motor vehicle-centric transport system?
Enjoying this? Here's how you can help keep it going.
If this blog has given you something useful (or just made your inbox a little better), here are a couple of ways to support it:
📤 Forward it to a friend
Know someone who'd get value from this? Forward this email their way. Word of mouth is still the best way for a newsletter like this to grow.
💛 Chip in, if you're able
If this blog is useful to you and you'd like to support it financially, you can do that here. Totally optional, always appreciated.
Whichever you choose (or if it's none of the above, no pressure!), thank you for reading. It means a lot.
|
Introduction
I am often asked what I do. I say I help people build better transport systems. It sounds simple, but the question underneath it isn't: better by what measure? Economically productive? Good for the environment? Good for taxpayers, for health, for equity, for freedom, for livability?
Over the next few weeks, I'm going to work through what a good transport system looks like from each of these angles, including the trade-offs between them, and whether transport has an everything-bagel problem, loading a public policy with so many competing goals that it becomes too slow, too expensive, and too complex to achieve any of them well.
I'll finish by pulling it together: where we are today, where we need to get to, and how I think we get there.
I'm starting with economic productivity, because it's possibly the most misunderstood of the lot and is often the underlying basis for transport systems around the world. So before anything else, it's worth being precise about what economic productivity actually means in a transport context, and where people’s intuitions about it go astray.
What is Economic Productivity?
Economic productivity measures how efficiently we turn inputs into outputs; in transport terms, how easily we can move goods and people from A to B.
That's why transport economics has traditionally focused on travel time savings: shave a minute off a trip, and in theory you've improved productivity. This logic became so embedded that travel time savings sit at the core of most transport business cases.
It's also intuitive. Turn a dirt track into a one-lane road, and speeds go up, travel times fall, and productivity rises. The trouble starts when we scale it. When that new road congests, the standard response is to add another lane, on the assumption that more capacity means more productivity. But this ignores induced demand: extra capacity generates extra traffic, so we can spend heavily on road widening for very little lasting benefit, the opposite of productive.
Before going further, it's worth clarifying why productivity matters in the first place. Paul Krugman put it bluntly: “productivity isn't everything, but in the long run it's almost everything, since a country's ability to raise living standards over time depends almost entirely on raising output per worker”. Most economists would agree. Productivity matters a lot, which is why it is so important to understand it properly in transport.
Why Faster Isn't the Same as More Productive
So travel time matters, which makes the common logic that follows seem obvious: to grow the economy, we should minimise travel time by favouring the fastest mode and building infrastructure for it. In most contexts, that fastest mode is the car, so the conclusion is to keep building roads.
This logic is flawed, because what's true for one journey doesn't hold once everyone acts on it. Cars are an extremely space-inefficient way to move people, and in cities, space is scarce. In a city, lots of people driving, especially at similar times, creates congestion, which erodes the travel-time gains the logic promised.
Cars carry other productivity costs beyond congestion. They need to be stored, which consumes land that could otherwise house people or businesses. They crash, at a cost that goes well beyond the tragedy of lives lost: emergency response, hospital treatment, and the congestion a single crash can cause on its own. They pollute, degrading the health and value of everything nearby. And a sedentary, car-dependent lifestyle carries its own long-run health costs, which show up in the economy as lost productivity, too. On top of that, cars competing for the same road space slow freight vehicles moving the goods we all require, increasing the cost of goods in shops and lowering welfare.
Put simply, a motor-vehicle-centric transport system is not economically optimal.
The Costs of Avoiding The Productivity Debate
Given the productivity narrative of motor vehicles, some critics of the car-centric system respond by avoiding the productivity argument entirely, making the case for reform on other grounds instead. I suspect this partly comes from an underlying assumption: that talking productivity inevitably leads back to car-centric solutions. If that's the reasoning, it's itself a symptom of car-brained thinking: we need to stop treating car-centric solutions as the most productive ones.
The cost of ceding the productivity ground is high. It leaves the productivity argument uncontested for the people who do want to make it, and they use it to defend the motor-vehicle-centric status quo in the name of jobs and growth. That case isn't actually true, as the mechanics above show. But it too often goes unanswered when the people best placed to rebut it opt out of the argument altogether.
So why does focusing on travel time savings often lead to car-centric solutions? We make a number of mistakes.
The Travel Time Trap: Four Common Mistakes
So what goes wrong when we use travel time savings as a measure? Four mistakes stand out.
Mistake 1: Treating A and B as fixed. For any single trip, the origin and destination are fixed. But at a system level, they're not; we choose them. Mixed-use development that puts homes near jobs shrinks the distance people need to travel in the first place, generating travel time savings without moving a single vehicle faster. Working from home takes this to its limit: A and B can be the next room.
Mistake 2: Assuming travel time is dead time. It usually isn't. Someone answering emails on a train is more productive than someone driving the same trip with no such option, even if both journeys take the same time. Someone cycling to work may be more productive still, folding in exercise they'd otherwise have to find time for elsewhere, and avoiding the productivity costs of poor health, even if the ride takes longer than driving would.
Mistake 3: Optimising the trip instead of the system. One person's drive barely affects travel times. Tens of thousands of people driving at the same time is a different problem entirely. Speed limits are a useful illustration: they can slow individual trips (although the delays are often small for many trips), but by reducing crashes and the gridlock they cause, they can make the system faster overall.
Mistake 4: Ignoring demand management. Most congestion is concentrated at peak times. Shifting even a fraction of trips outside the peak improves that traveller's journey and eases congestion for everyone still travelling at peak, without building any infrastructure.
Individually, each of these mistakes causes us to overrate travel time savings measured at the level of a single trip. Together, they point to the same conclusion: travel time is only one input into productivity, and it only makes sense to analyse it at the level of the whole transport and land-use system.
The problems with travel time savings as a measure are one of the reasons why accessibility, how easily people can reach jobs, services and opportunities, and how easily, quickly and affordably goods can reach their destination, has become an increasingly common measure in transport agencies.
How Would a More Productive Transport System Work?
So how would a transport system optimised for economic productivity actually work? It would allocate places, modes and travel times according to what's genuinely productive at the system level, not just at the level of one trip.
In practice, that means discouraging travel patterns that look efficient individually but aren't once everyone does them, a car on a congested road at peak time being the clearest example. The productive response isn't to add capacity for it; it's to shift it: re-time the trip to outside the peak, re-route it onto a less congested path, re-mode it onto public or active transport or freight rail, or in some cases remove the need to travel at all, as working from home does.
That's not what happens today. And the gap between that allocation and our current one is a large part of why so many cities carry chronic congestion, and why transport's contribution to productivity falls well short of what it could be.
Why a Productive Transport System Doesn't Happen by Itself
So why doesn't this productive allocation happen today? Three factors stand in the way.
We subsidise the modes that cause the most congestion. Motor vehicles get free or heavily discounted road use and parking, while the wider costs they generate, emergency services, pollution, and other knock-on effects are spread across all taxpayers rather than charged to the people driving. Subsidising a mode increases demand for it, and driving is no exception.
We build infrastructure around cars at the expense of more productive alternatives. Buses can move far more people per lane than cars at peak times, but only with dedicated lanes and enforced traffic priority (as well as getting the operational services right), both of which are routinely watered down because of the perceived impact on drivers. Cycling infrastructure faces the same resistance for the same reason. The result is that people stay in the least productive mode available, because the more productive ones were never properly supported.
We restrict land use in ways that force people to travel further than they need to. Zoning rules routinely keep homes away from jobs and transport hubs, often because of local objections to higher density. That pushes people into locations with fewer good alternatives to driving, baking car dependency into the built environment.
What the Productive Transport System Actually Requires
So what would this look like in practice? People would drive substantially less than they do today, and congestion and pressure for ever-bigger roads would ease accordingly. That doesn't mean eliminating driving. Motor vehicles would still play a significant role wherever they're genuinely the most productive way to meet a given travel need; this isn't a car-free vision, just a better-allocated one.
Getting there is, in policy terms, fairly well understood. Road user charging, varying by time of day, location, and vehicle size, would price driving closer to its true cost. Parking reform would price kerb space to actually manage demand rather than give it away. Sustained investment in public transport, walking, cycling and freight rail would give people productive alternatives to choose instead. And land use reform would reduce how far people need to travel in the first place, while improving access to the transport hubs already in place.
Alongside this, the tax base underpinning driving needs to shift too, replacing fuel taxes, which are already being eroded by more efficient and electric vehicles.
Conclusion
So where does this leave us? Economic productivity is not a reason to build more roads. Interestingly, once you account for induced demand, system-level congestion, and the true cost of a motor-vehicle-centric system, the productive answer looks a lot like the answer people already arrive at when they argue from environmental, health, or livability grounds: fewer cars, better alternatives, and land use that puts people closer to what they need.
For a system genuinely optimised for economic productivity, road user charging, proper parking pricing, and land use reform are what economic productivity requires.
At this point, I am not going to say that we should develop a transport vision based solely on economic productivity. First, we need to consider other aspects before drawing conclusions.
Next week, I'll turn to another lens for judging a good transport system. Once we've worked through each lens, I'll bring it together: what a good system looks like, where we currently fall short, and how I think we get there.