Pay as you drive 2

Norwich Union has suspended its Pay-As-You-Drive insurance scheme [BBC News, 14 June 2008], announced here two years ago [Pay As You Drive]. I am disappointed at this news, because PAYD was my favourite example of differentiated pricing, using telematic information about driving patterns to determine the insurance premium paid by a driver.

This is not the end of differentiated pricing of course, and may not even be the end of PAYD. According to the BBC, there is one other insurance company in the UK (MoreTh>n) offering PAYD insurance, but all I could find on their website was a product called GreenWheels. This is not a PAYD insurance scheme, but uses similar technology to provide detailed information to the driver to help improve the driver's fuel and maintenance costs as well as safety.

Why did the Norwich Union scheme fail, and what does this tell us about the viability of such schemes in general? The primary reason is that the scheme failed to attract enough drivers to cover the fixed costs of administering the scheme. But in the longer term, a scheme like this was only going to be viable if the technology became cheaper and more widely available - in other words, compatible telematics devices fitted as standard to factory model automobiles. And this in turn was only going to happen if PAYD insurance was offered by several major insurance companies (perhaps across Europe rather than just in the UK, with appropriate "roaming charges" as with mobile phones) or if there was some other purpose for the technology and infrastructure. For example, providing information to support eco-friendly driving (Green Wheels). But the big opportunity was PAYD road pricing.

However PAYD road pricing is highly unpopular. There was a major campaign against PAYD road pricing in the UK, orchestrated by the road lobby. The UK government may have hoped that commercial PAYD products, including insurance, might have paved the way for PAYD road pricing, but this now looks less likely than ever.

Meanwhile, we already have some forms of road pricing. London already has a congestion charging scheme, and other cities are likely to follow. But these schemes, along with parking and traffic violations, are currently based on photographing the car plates, and this is of course vulnerable to identity theft. (You just need to put fake plates on your car, and someone else will get the penalty notice.)

At the core of PAYD needs to be a robust and reliable way of identifying vehicles and recording their behaviour, and this raises obvious privacy concerns. All differentiated service depends on an appropriate identity scheme, and such schemes are politically charged whenever governments get involved. So perhaps widespread adoption of differentiated service will have to wait until Identity 2.0 is more mature?