Reducing personal energy consumption

We live on the South Circular road, one of London’s busiest commuter routes. Every morning and evening there is a huge volume of traffic, mostly cars, taking people to and from work. There are hundreds (thousands?) an hour in each direction. The thing I find most disturbing about this is not just the sheer volume, but also the fact that 90% or more of the cars contain only one person.
And presumably most of them are the same people, driving the same route, every day.
On a simplistic level, the maths is pretty straightforward. If we could get each person sharing with one other, it would reduce the number of cars, and consequently energy consumption, pollution and congestion, by 50%.
And if we could actually get 4 people into each car, then we’d reduce each of those things to just ¼ of their current levels.
In reality, probably better than that because less traffic means less holdups, allowing faster smoother and more energy efficient driving. At a guess it would end up getting energy consumption, emissions and congestion down to well under 20% of what it is now.
Extend that to every car commuter in the world and our petrol reserves would last 5 times as long, journeys to work would be so much faster and easier (more time to enjoy other things) and each individual would save thousands of pounds a year (what does the average car commuter spend on petrol?).
Sharing your journey with others would also result in more social interaction & fulfilment, contributing to the resolution of a whole bunch of other social issues.
And that brings me to the bit I’m really interested in – the behavioural aspect.
There are very clear logical reasons for making these changes in our commuting behaviour – even to the climate change denier the money saving argument is undeniable. However, we’ve been doing the logical benefit explanation thing for energy reduction for a long time now and how much difference has it really made?
The evidence outside my door every day says very little.
And my own experience of causing behaviour change in others confirms that the logical explanation approach has very little real or lasting effect. After all, every smoker I’ve worked with has fully understood the logical benefits of being a non-smoker for years, and yet they’ve continued to smoke. The way I help people quit smoking does not involve giving them ever more detailed logical explanations of the benefits.
So how do we get people to make real, significant and lasting changes to their energy consumption behaviours?
I’ve got some ideas, but right now I’m more interested in yours.
What approach do you think would get people to make those changes?
Please post your thoughts and comments.