The Treasury first drafted a carbon tax bill in 2010. Since then it has been discussed ad nauseam, commented on, redrafted and rediscussed. The latest gabfest was in Parliament on March 14. They still haven’t got it right. Much of the problem comes from the fact that they have completely forgotten the principles of such a tax.

These principles were developed about 80 years ago by the British economist Arthur Pigou. He pointed out that the price of goods may not include their social costs — what are called "negative externalities" by economists.

The tax on plastic bags is an example. If the bags are discarded after use, they form nigh on indestructible litter. If the tax is set at the cost of collecting the litter, it pays the user to reuse the bags, and litter is reduced. Economically, this is a far more efficient solution than paying someone to gather the rubbish.

There is a slight problem; it is a regressive tax. Its impact is the same on the rich as on the poor, and the poor are therefore harder hit. However, Pigou found the answer to this. Reduce other taxes by the amount collected to offset the negative externality, and equity prevails. Pigouvian taxes are now applied by many governments, and when properly applied they have proved eminently workable.

There were many counter examples. India introduced a heavy coal tax; it lasted six years and had no effect on emissions

How could anyone get such simple principles wrong? The first challenge the Treasury faces is estimating the social cost of carbon emissions. It is predicted that burning fossil fuels will cause an increase in hot weather, droughts, floods, violent winds, cyclones and sea levels. But when you dig out the evidence for these increases, you find remarkably little support for them.

To the best of our knowledge, it has been warming for at least 180 years. Surely this should be long enough for the predicted increases to be seen? Surprisingly, it isn’t. Yes, it has become warmer, and glaciers are melting. But as the ice disappears on alpine passes, so footpaths appear that were last in use 800 years ago, when it was warmer than today.

Droughts and floods? Turn to the authority on such matters, the Intergovernmental Panel on Climate Change. It said: "There is medium confidence that since the 1950s some regions of the world have experienced a trend to more intense and longer droughts, but in some regions droughts have become less frequent, less intense or shorter." The trouble is that if you measure over, say, 50 years, it seems some places experienced an increase in droughts, but if you measure over 150 years, the places that first seem affected now seem normal, and new areas seem more prone to drought.

"There is limited to medium evidence available to assess observed changes in the magnitude and frequency of floods. Furthermore, there is low agreement in this evidence, and thus overall low confidence at the global scale regarding even the sign of these changes." Translated, this means the panel cannot say if floods are getting more or less frequent, a conclusion those who have not looked at the data find incredible.

Tropical cyclones? Tornadoes? Gales?

"There is low confidence that any observed long-term increases in tropical cyclone activity are robust. There is low confidence in any observed trends in ... tornadoes and hail. There is generally low confidence in projections of changes in extreme winds."

Sea level? "Sea level rise for 2081–2100 (relative to 1986–2005) will probably be 0.26m to 0.82m." Tide gauges show a rise of less than 0.2m per century since the early 1800s, so even if the panel’s guess is right, it seems unlikely to be a disaster. We have coped with sea level rises of this magnitude for centuries without difficulty.

The panel concludes that "uncertainty in projections of changes in large-scale patterns of natural climate variability remains large".

Most of the purported increases in extreme climate can barely be detected. It is true that some of the models the panel employs predict significant increases, but history fails to show the modelled behaviour and there are very strong grounds for believing the models do not reflect reality. For instance, they fail to predict the rate of warming of the atmosphere up to about 10km.

It follows that it is nigh on impossible to estimate a social cost, the value of the negative externalities, associated with greenhouse gas emissions. And if you cannot determine the social cost, you cannot follow Pigouvian principles and set the tax equal to the social cost. There is, therefore, absolutely no basis for a carbon tax.

Other nations have tried to resolve this challenge. Britain persuaded a Treasury official, Nicholas Stern, to try. One economist scorned his efforts: "An insurance salesman trying to sell me a policy for my great, great grandchildren against an injury that might be caused by an unlikely event in 200 years’ time."

I studied 19 jurisdictions that had introduced a carbon tax. In only one — British Columbia — had the tax had the desired effect of cutting emissions. I sought reasons for its success. First, the tax was revenue neutral; British Columbia had cut taxes elsewhere in the fiscus to equal the carbon tax. Secondly, most of British Columbia’s electricity was hydro-powered; most emissions were from liquid fuels, so the carbon tax was in effect a fuel tax. Use of public transport grew.

There were many counter examples. India introduced a heavy coal tax; it lasted six years and had no effect on emissions. Australia introduced a tax in 2010; costs shot up so rapidly that the government fell and the tax was removed after only two years. In none of the jurisdictions that traditionally depended heavily on carbon as their primary source of energy did a carbon tax have any effect except to increase the costs of energy.

A carbon tax only works if there are sources of energy services that do not rely on fossil fuels. If there is little alternative, as is the case in SA, the tax is nothing but a cost with no benefit. Energy producers cannot reduce their emissions, and the tax merely adds to their costs.

In this century, SA’s emissions have barely changed. In contrast, China’s and India’s emissions have both nearly doubled. The Department of Environmental Affairs believes we should make a "fair" contribution to the control of global emissions. What is "fair" when we are comfortably within our Paris Agreement commitments, and our partners in the Brics grouping of leading developing economies continue to grow their emissions regardless?

We do not need a carbon tax.

• Lloyd is with the Energy Institute at the Cape Peninsula University of Technology.