However, the idea of being forced to pay to enter a city has not often proved popular with the public. In 2005, 74 per cent of people voted to reject a scheme in Edinburgh in a referendum.
The proposal was put to the vote two years after congestion charging was introduced in London in 2003 by the then mayor, Ken Livingstone, and while he was re-elected the following year, it is fair to say it was a highly controversial issue even in a city as crowded as London.
But what if there was a way to introduce a scheme that, amazingly, was actually well liked by a very large majority of the public, while still solving those perennial blights on modern society? And what if this scheme could also increase the turnout at elections and provide financial help to those on low incomes?
In this article, I describe how just such a system could be introduced not just in cities but across the whole country.
At present, entering the London congestion charge zone for one minute costs the same as 15 hours. London’s current mayor, Sadiq Khan, who is currently considering a number of changes including increasing the size of the zone, says he ultimately wants to bring in a pay-per-mile system.
But this would not actually measure the problem that the charge is designed to address – congestion – and it is technically possible to design a system that does this almost exactly.
My proposal is for every vehicle to carry a box of electronics glued to the inside of the windscreen behind the rear-view mirror. Every few seconds it would transmit a short burst of radio signals of different amplitudes. In between transmissions, it would ‘listen’ for signals from other vehicles and count the number.
It would pick up only the very largest signals from distant vehicles, all of them from ones close by, and most from those in between. The total count would therefore be an accurate measure of how long each vehicle had been close to others – and this is exactly how we define ‘traffic congestion’.
From time to time, perhaps at MoT tests, service intervals or ownership changes, the accumulated count would be read with an instrument at the service station.
Like sunglasses, radio signals can be polarised and the angle of polarisation can be measured, which would enable the signal count from cars travelling in the same direction to be greater than those going in the opposition direction, so motorists going the opposite direction from everyone else would not be penalised for simply driving past standing traffic.
An important question is what to do with the revenue. It is claimed that the money raised goes to worthy traffic reduction schemes chosen by unknown officials even though congestion is getting worse and traffic moving more slowly, which is one reason why congestion charging is seldom seen as a voter winner.
The money raised by my system could be claimed by central government or the relevant local authority.
But the most politically popular way would be to introduce a negative poll tax in which everyone turning up to vote at an election would get a share of all the revenue since the last election.
The London congestion charge of £15 a day adds up to more than £20,000 if drivers pay it every day for four years, so payouts that were even a moderate fraction of this sum would be extremely popular and would also give the incoming government an economic boost.
The big lump sum coming in might seem better to motorists than the slow, long-term trickle going out. There might even be fond memories of the political party which introduced the scheme!
Vehicle owners might compete with one another to produce less than average congestion by switching from their normal routes to quieter ones. And people too poor to own a car would get a life-changing windfall.
Those living in isolated rural communities, who need a car much more than city dwellers, would be paying much less by not being close to other cars and it would also be possible to give them an initial ‘ration’ of negative counts.
The present London scheme does allow lower payments for environmentally benign vehicles but with rather coarse steps. My pulse-counting system would allow a completely smooth range based on car-by-car emissions and even driver-by-driver charges depending on occupation.
We could also fit static signal generators to discourage people from driving past particular congestion hotspots and also to charge for parking with the payment in direct proportion to the time in which a parking space is occupied, removing anxieties over punishment for parking one minute too long or not knowing how long a meeting will take.
Overseas tourists, who could be easily identified by their number plates, could be allowed free movement for a limited period to avoid dissuading them from coming.
Some might hit upon the idea of cheating the system by wrapping aluminium foil around the signal boxes, which would reduce the incoming radio bursts and therefore the charge.
However, this would also reduce the outgoing signal strength so traffic wardens and police cars could carry transmitters which send a message saying "report my signal strength” to the devices on passing vehicles and take action if the reply is lower than it should be. This means fraud prevention would be at least as good as the prevention of number-plate doctoring, such as changing a P to an R, or an I to a T.
A congestion charge system that achieves the same societal benefits as any other but that people might actually come to love? There’s an idea.
Stephen Salter is emeritus professor of engineering design at University of Edinburgh