All You Can Drive Roads

The “all-you-can-eat” restaurant is popular in the US and Canada. Pay $14.95 and fill your plate over and over. Surely someone has studied the correlation between girth and frequency of patronizing such establishments, but who needs a science degree to describe the effect?


Free road access has a parallel. Fill your tank – i.e., pay fuel tax, the “all-you-can-drive” $14.95 – then drive what, where, and when you want. Congestion is the analogy to girth. When unlimited desserts are included some people eat six. When road access is free, too many drive in rush hour, on crowded roads, day-after-day. When road use is charged relative to value (higher demand, higher price) people shift their time or mode of travel. They will telework more often, commute at a different time, be more likely to carpool. They may even use transit, bike, or move closer to work.

And if you don’t shift your travel habits, you will enjoy less congested roads. Everyone wins.

No one is happy about congestion, and most of us realize it will be getting a lot worse. We generally admit automotive emissions are harming our cities, our health, and our planet. Less appreciated but easy to understand is that established cities have a hard time to add new travel lanes, that road building costs per lane-mile climb far faster than other costs, that each new lane mile finds as many or more citizens resisting it as clamoring for it.

The compendium of things wrong with adding cars and roads is long, but we still need to move people and goods as much or more than ever. We have a scarce resource: road-surface where and when we want it. The where and when are key.

In the world of how much to drive, fuel tax is a perfect mechanism: correlated, easily collected, hard to evade, somewhat hidden. But in the world of where and when to drive, fuel tax becomes the problem – silent about the difference between high demand and low-demand roads and silent about when fuel is used. But the value of access of many roadways during peak hours is far greater than those same roadways off-peak.

Paying a charge for road use rather than a tax on fuel consumption is the far smarter approach.

We now mostly operate “all-you-can-drive” roads. In fact, once you pay the London or Stockholm daily congestion charge, “all you can drive” still holds for the remainder of the day.

Therein lies the problem with flat fees. They reduce the number of vehicles that enter a cordon, but they do nothing to reduce miles traveled once inside.

The triple problem of road funding, congestion and emissions demands a variable, distance-based scheme for every mile traveled in lieu of the fuel tax. The most suitable technology for that is global navigation satellite systems such as GPS, and that is in development by several companies. There are ways to deploy without capital cost to government and unsightly infrastructure, unlike the current ‘tag-and-beacon’ systems. There are also ways to use this technology with absolute privacy protection.

The technology solution is understood, the economic reasoning is solid, governments are slowly gathering courage, only the switch-over still has a few kinks.

Something to ponder while you’re sitting in traffic.

A Canary in Dubai

Each new congestion charging system gets less attention than its predecessor. London’s was all the rage for a year before and 2 years after February 2003. Stockholm’s generated a blip by comparison even though it serves three times the cars and is a much slicker implementation. But hands up if you were even aware that Dubai just put one in?

I recently gave a talk to 45 engineers (only one of whom was in traffic). All had heard of London’s system, all but three of Singapore’s system, only two were aware of the Stockholm system and none was aware of Dubai’s. Sure, Dubai is further from Canadian consciousness than the others and sure Dubai’s was just installed, but there may be other factors at play.

Consider that congestion pricing is a tool whose time has come, that only journalists in affected cities will note new systems going in, that the international press has wearied of what is clearly a trend, and that your 3 year old will never drive on an unpriced road.

Consider that Dubai is also providing a more critical message. Dubai’s system is based on shortrange radio beacons and in-vehicle RFID tags. This gold-standard for electronic toll collection (ETC), works perfectly well for limited access highways in half the countries on the globe. In fact, it works for congestion pricing on limited access island cities like Stockholm pretty well, too. It works less well in Singapore because the central, congested areas are somewhat more diffuse and peter out onto untolled roads that motorists use instead. That flattens out peak traffic, but does not move people out of cars. It simply delays the confirmed solution.

Dubai’s system has experienced a stronger-than-usual backlash – much of it the usual suspects: distrust of government motives, an assumption of (and knee-jerk rejection of) taxes, entitlement to free access to roads, and misunderstanding of market pricing (but at a dollar each way, that is likely no where near market pricing). But there is another important element here: bad tools + bad design = poor results.

Only a small but critical fraction of the network was tolled which largely reroutes traffic rather than encouraging alternatives. This relocates the pain rather than relieving it. Like easing your headache by stubbing your toe. If you want to force consideration of alternate modalities, you have create a cordon. It also appears that a single price is in effect. This provides no signal about when not to drive – only where not to drive.

For once, I empathize with the local whiners – this does look more like tax collection than congestion abatement. Dubai’s system harms the fledgling reputation of congestion charging.