The Financial Times, reported on 3 July 2006 - Breathing in exhaust fumes in central London as I cycle to work, I can easily understand why carmakers get so much of the blame for global warming. Few people own anything that produces more visible and smelly pollution than their car, so it is no surprise that public and political pressure is focused on making them cleaner.

But just because cars churn out nasty gases does not mean that everything done to make their engines produce less of the carbon dioxide responsible for global warming is automatically a good thing.

This point seems to have been lost on environmental lobby groups and politicians during the past two years of near-panic over the twin problems of soaring oil prices and mounting evidence that the planet is heating up.

Governments in much of the developed world have begun pushing the car and oil industries to invest in technologies hybrid cars and biofuels in particular that are not yet properly developed. Hybrids have become synonymous with green awareness while, in the imaginations of both public and policymakers, biofuels, made from plants, must be a good thing. But neither makes environmental sense and the government-sponsored rush to adopt them pushes up the cost of both cars and petrol.

It is not that hybrid petrol-electric cars or biofuels are bad for global warming, although some biofuel, made in coal-burning facilities, produces more CO2 than petrol. The problem is that both are expensive and far more so than other ways of reducing CO2.

After more than 100 years of development, engines burning petrol or diesel are pretty economical. Hybrids, which add a heavy battery and electric motor, are much more expensive, even with big government subsidies. Biofuels cost more than petrol even with subsidies and the high price of oil.

"There's a reason why we use a lot of oil," Rick Wagoner, chief executive of General Motors, the world's largest carmaker, said recently. "Because it is the cheapest and most efficient way to power an engine."

Research from the European Commission backs up this claim, which would otherwise be dismissed by the environmental movement. The research shows that it will cost from 207 (GBP143) to 239 a tonne to reduce CO2 by using biofuels made from the most common European sources, sugar beet and wheat, in 2010. Put another way, biofuels cost the economy just under 1 extra per gallon, mostly hidden by tax breaks and subsidies.

Hybrid cars transport of choice for environmentally-minded government ministers will cost 1,062 per tonne of CO2 saved, as well as being loss-making for almost all manufacturers. As I sat on my bike watching the haze over a queue of traffic, this did not immediately bother me and few in Brussels, Washington or Westminster are overly concerned about the profitability of car and oil companies. But if the money is spent on making engines produce a little less CO2, it is not being spent on other ways of reducing CO2 that are more effective and cheaper. Chief among these are combined heat and power plants that burn straw or wood, which are alternatives to coal-fired power stations and to fuel oil used for heating, particularly in industry.

According to one of the authors of the Commission's study, the straw-fired power station at Ely, in East Anglia, costs just 35- 100 per tonne of CO2 saved. Europe's carbon trading scheme valued cuts of a tonne of CO2 at a peak of 31, because there are still so many cheap ways to save energy in less efficient industries. This is not to say that carmakers should be free to ignore carbon emissions their vehicles produce a quarter of man-made CO2 emissions, after all. But they should be free to choose how to reduce carbon, not be locked into certain technologies by governments. (One of the technologies being pushed by Washington and Toyota a "plug-in" hybrid with an external recharger actually results in more carbon emissions than a plain petrol car because so much US electricity comes from coal.)

A well-designed trading scheme would give manufacturers the choice of meeting CO2 reduction targets in the most cost-effective way. This could involve paying to clean up other industries by buying their carbon credits, funding carbon capture systems to store CO2 in undersea tunnels or, if it made financial sense, new car technologies. Unlike existing average emission goals, it would also provide an incentive to outperform targets as excess carbon credits could be sold at a profit. Carmakers would still have a reason to research hybrids or biofuels, as both promise to come down in cost over the next decade.

Prices of carbon credits from other industries will also rise as they take advantage of the easiest ways of reducing CO2, often a proxy for energy use. But instead of rushing to sell unprofitable vehicles to meet arbitrary carbon reduction targets, they would have an incentive to reduce carbon emissions by as much as possible. If the EU study's numbers are right, between twice and 30 times as much carbon could be saved by replacing current technologies with a trading scheme. Automotive carbon trading might not provide politicians with the image boost they get from driving a Toyota Prius hybrid or filling up a car with ethanol from Iowa's cornfields, but it would be far more effective at fighting global warming. Or you could just get on your bike.