Joshua Gans, like unaccountably many econobloggers, gets rather over-excited by Apple products. But, stripped of its Jobsianist overtones, Gans has an interesting idea – give people car navigation devices that, rather than deciding on the fastest route on an individual basis, coordinate traffic planning to spread traffic out over roads, and thus minimise overall congestion levels and travel times. According to Gans’ intuition:

Now you wouldn’t be compelled to follow the instructions handed to you but if it was meaningful you would follow them anyway as it will likely make you at least weakly better off by doing so.

Unfortunately, this isn’t always the case – the interest of the individual driver, and the collective interest of everybody on the roads, don’t necessarily coincide.

When choosing a route from A to B, you don’t generally have the opportunity to discuss it with other road users first. At best, you’ll know what the traffic (and thus the “segment” travel times) are on the various alternatives are. As a guy called John Wardrop put it:

Wardrop’s first principle states: The journey times in all routes actually used are equal and less than those which would be experienced by a single vehicle on any unused route. Each user non-cooperatively seeks to minimize his cost of transportation. The traffic flows that satisfy this principle are usually referred to as “user equilibrium” (UE) flows, since each user chooses the route that is the best. Specifically, a user-optimized equilibrium is reached when no user may lower his transportation cost through unilateral action.

In other words, “user equilibrium” is reached when no driver can change their route so as to get where they want to go faster.

This is a Nash equilibrium in game theory – nobody can unilaterally change their strategy (route) to improve their payoff (travel time).

The question is then whether the “user equilibrium” flow collectively minimises travel times. The answer, in busy cities, appears to be “no”.

The Prisoner’s dilemma is the classic illustration of this kind of problem: a situation where individuals, acting in their individual self-interest, end up with outcomes that are worse for them both than if they cooperated. If you’re not familiar with the dilemma, it goes like this:

Two suspects are arrested by the police. The police have insufficient evidence for a conviction, and, having separated both prisoners, visit each of them to offer the same deal. If one testifies (defects from the other) for the prosecution against the other and the other remains silent (cooperates with the other), the betrayer goes free and the silent accomplice receives the full 10-year sentence. If both remain silent, both prisoners are sentenced to only six months in jail for a minor charge. If each betrays the other, each receives a five-year sentence. Each prisoner must choose to betray the other or to remain silent. Each one is assured that the other would not know about the betrayal before the end of the investigation. How should the prisoners act?

Looking at it from the perspective of an individual prisoner (call them prisoner A), whatever the other prisoner (B) does, they are better off “defecting” than they are remaining silent. So, if they both follow this logic and talk to police, both receive a five-year sentence. This is clearly a much worse outcome for them than if they’d both chosen to remain silent, where they would have both received a six-month sentence.

Similar things can happen on the roads, as shown in an illustration of Braess’s Paradox – essentially, adding a road, even without adding any more traffic, can increase travel times if people are behaving according to Waldrop’s first principle and choosing routes that minimize their own travel time without consideration of the network-wide effects.

The standard example of Braess’s paradox is very artificial, but some clever people have done some analysis of the road networks of major cities. Their calculations suggest a couple of things. The first is that there are some roads in major cities that, if closed (or limited to local traffic) would actually decrease congestion. The second, and the one directly relevant here, is that the extra time spent on the roads by people choosing their own routes – the “price of anarchy” as the paper puts it, is around 30% at busy times. Yep, people spend nearly one-third longer in their cars because people only consider their own travel time when choosing a route.

If we really want to improve traffic flow in our cities, not only would we need coordinated navigation systems, we’d need some way to encourage as many people as possible to actually take the recommended route, rather than a route that improves their own travel time at the expense of the efficiency of the overall system. At this point you could send computer scientists, economists, and mathematicians into a frenzy of geekery to devise such a scheme.

Or, alternatively, you could encourage more public transport, where the routes and timetables are centrally planned already and this conflict doesn’t arise. But that’s not nearly as much fun for nerds…

Postscript: Thanks to commenters Dave (Bath, I presume) and Jacques Chester over at Core Economics for reminding me about network flow algorithms and Nash Equilibria.

I’m not sure what the category “supercar” means, but I gather it is a very expensive, high-performance car that would set you back a couple of hundred grand.

First there is this little beauty from Porsche.

Just unwrapped at Geneva is this extraordinary Porsche 918 Spyder concept car, a mid-engined two-seater combining supercar performance with just 70 grams CO2 per kilometre emissions and a fuel consumption of 3.0 litres/100 kilometres (94 mpg imp). Under those curvaceous lines hides a 500 bhp V8 plus three electric motors totalling 215 bhp offering a wickedly fast 0-100 km/h time of 3.2 seconds, a top speed of 320 km/h (198 mph) and it’s already done a lap of the Nordschleife Nurburgring circuit under 7:30 minutes, which is faster than even the Porsche Carrera GT. Buyers will be queued up around the block if (or more likely, when) this goes into production. (Emphasis added)

And it’s seriously sexy hybrid!

Even more interesting is the Revenge Verde:

The mid-engined Verde offers three drive chain and power train options, including an HP2g V8 engine that runs on E-85 ethanol fuel and produces 400hp, goes from 0-60mph in 3.5 seconds with a top speed of 200mph+, while achieving an amazing 100mpg.

So even lower consumption, only slightly less acceleration, and 100% biofuel. You need to watch the video in the link.

The car designer is an Australian who made good in the US of A who describes the extensive use made of lightweight materials. The engine has an electric motor within it, so is “true hybrid”. The 8 cylinder motor can tick over on just one cylinder.

This mob are planning to use the supercar to bed down the concept and are later heading for the passenger market, SUVs and trucks. I assume they will have a straight fuel option down the track, rather than just ethanol.

Technologically these cars seem streets ahead of the Tata Nano which Robert posted about a couple of years ago.

What potential the technologies in the Verde Revenge and the Porsche have in the mass market I wouldn’t know, but to state the bleeding obvious, the future will be different from the past.

The actual vehicle doesn’t exactly fill me with confidence, either; it’s the replacement for the Daewoo Lacetti and the Chevrolet Cobalt in Asian and US markets. American-designed small cars have, historically, been sub-standard, something that most definitely applies to the current Chevy Cobalt which I had the misfortune to rent when last in the USA. By contrast, Ford’s upcoming domestically-produced small car, the Focus, is a Ford of Europe adaptation of a highly-rated Mazda design. History may not be destiny, but I know which car has the better small-car design heritage behind it…

Regardless of the merits of the specific vehicle, one still wonders just how much government money it’s taking for Australian-built small cars, churned out at a rate of maybe 40,000 a year if they’re lucky, to be competitive with factories in Asia who produce cars on a far, far larger scale. Enough to pay for a lot of homeless shelters, I’ll venture.

As has been said any number of times here on LP, the government’s election promise of 60% emissions cuts by 2050 has been left behind by events; the science is calling for steeper cuts, and any global deal is going to involve Australia making disproportionate cuts because of its enormous per-capita emissions. Absent cheap, environmentally benign carbon air capture and sequestration technology, this will necessarily involve the virtual decarbonization of Australia’s transport sector, and probably over the period of a couple of decades. Thankfully, this looks like it may actually be achievable, through technologies like the plug-in hybrid, the first production example of which will be the Chevrolet Volt.

But the review report’s treatment of the issue is incredibly casual. There’s no quantitative analysis of the effect of an ETS on fuel prices – a dead-easy calculation, and a fundamental part of predicting the likely effects of the ETS on the transport sector. Furthermore, the effects of the the more stringent emissions trajectories implied in the Garnaut Review are not even considered. Heck, they don’t even accurately describe the Chevy Volt itself; the review describes it as an electric car that may, at some point in its model life, be fitted with a range-extending petrol motor. 30 seconds of Googling would have told them that’s completely wrong – the Volt has been described from the ground up as a plug-in hybrid, designed to run about 60 kilometres on a full battery charge, then switch to a petrol motor. The nightmare scenario for the car industry – that they can’t develop affordable low-carbon cars, and hence there’s a switch to alternative means of transport – hasn’t even been mentioned. And there’s a thought that “by 2050″, the near-complete decarbonization of vehicular transport might be feasible, whether this is soon enough or not.

In some ways, this doesn’t matter very much; Green Car fund or no Green Car fund, Australia’s contribution to developing low-carbon personal transport will be a tiny fraction of the global effort, and the cost pressures of an ETS (and, possibly, peak oil) will drive a transition whether Kim Carr’s department is expecting it or not. But it’s further evidence that large parts of the bureaucracy and government still haven’t grasped what Australia will have to do in a carbon-constrained world.