06 August 2012

The Leviathan Tax, A Guest Post by Richard Tol

Note: Richard Tol, an economist at the University of Sussex, has a new paper out on carbon taxes, with important implications for policy. He graciously agreed to write up this guest post with an overview of the paper and a discussion of its implications. Reactions and questions are welcomed in the comments.

There are regular calls for stringent climate policy. Long-term targets are cheap talk. It is not clear how stringent climate policy can be in the short run. Norway has had a $21/tCO2 carbon tax for two decades. Australia is having difficulty introducing a similar tax.

To provide some structure to the debate, in a recent paper (pre-print), I define and estimate the Leviathan tax. The state would grow, relative to the rest of the economy, if the carbon tax would exceed the Leviathan tax. The Leviathan tax is the maximum carbon tax that is budget-neutral. That is, all other taxes are reduced to zero and replaced by a carbon tax.

The Leviathan carbon tax equals the total tax take times the inverse of the carbon intensity of the economy. I used the World Bank’s Development Indicators. Economic data are in constant 2000 US dollars, using market-exchange rates. Carbon dioxide emissions are for fossil fuel combustion and cement production only. I used data for 2005, the most recent year with almost complete coverage. Data and graphs are here.

Table 1 shows the Leviathan tax, tax revenue, carbon intensity, and carbon dioxide emissions for the 20 largest emitters, together covering 85% of global emissions.


Leviathan tax
Tax revenue
Carbon intensity
Carbon dioxide emissions

$/tCO2
(% GDP)
(kg CO2/$)
(mln t CO2/yr)
Iran
23
7.9
3.34
459
Ukraine
23
17.1
6.28
339
China
29
8.7
2.76
5,790
Russia
36
16.6
4.06
1,616
India
45
9.9
2.08
1,411
Indonesia
77
12.5
1.58
336
Thailand
99
17.2
1.62
275
South Africa
106
26.9
2.42
408
Poland
109
16.7
1.39
304
Canada
202
13.8
0.63
563
South Korea
211
14.7
0.68
463
United States
223
11.2
0.48
5,595
Spain
243
12.6
0.49
353
Germany
267
11.1
0.38
810
Australia
330
24.7
0.73
367
Brazil
353
16.7
0.45
350
Japan
450
10.9
0.24
1,238
Italy
522
21.2
0.38
471
France
822
22.4
0.25
392
United Kingdom
855
27.2
0.29
542

The Leviathan carbon tax is less than $1/tCO2 for Liberia and Nigeria. Both countries collect less than 1% of GDP in taxes, getting most of their government revenue from state-owned oil companies. The Leviathan carbon tax for Iran and the Ukraine is $23/tCO2, $29/tCO2 for China, $36/tCO2 for Russia, and $45/tCO2 for India. The Leviathan carbon tax of the USA is $223/tCO2, $353/tCO2 for Brazil (recall that land use emissions are excluded) and $855/tCO2 for the UK. Iceland’s Leviathan carbon tax is the highest at $1367/tCO2.

What does this imply? Following the results of EMF22, a stabilization target of 650 ppm CO2 equivalent would require a 2015 carbon tax of $6/tCO2e, in all countries, on all emissions, of all gases. The carbon tax would rise over time with the interest rate plus 0.6%. A target of 550 ppm CO2e would require an initial tax of $29/tCO2e, and a 450 ppm target would need an initial $149/tCO2e tax.

If the carbon tax is $6/tCO2e, its revenue would exceed 100% (10%) of total tax revenue in countries that account for less than 0.5% (3.5%) of total emissions. In countries representing 73% of emissions, the carbon tax would yield less than 1% of current revenue. This is fiscally feasible.

The revenue of a $29/tCO2e carbon tax would exceed 100% (10%) of total tax revenue in countries that account for almost 2% (21%) of total emissions. The carbon tax revenue is greater than 1% of total revenue in all countries.

For $149/tCO2e, the carbon tax revenue is greater than 100% (10%) of tax revenue for more than 10% (100%) of emissions. Yet, this tax would lead us to 450 ppm CO2e, with a fifty-fifty chance of meeting the 2°C target.

Such a tax would imply a substantial extension of the government budget, even if all other taxes are abandoned, in China, India and Russia. Two-thirds of the US tax take would be from carbon. This is unlikely, and may be unwise. Fiscal experts agree that the tax base should be diverse.

Even a $29/tCO2e, and hence the 550 ppm CO2e strains credibility. Almost 100% of China’s tax revenue would be from climate policy, and almost two-thirds in India. If these countries impose a lower tax, other countries would have to levy a higher tax if the target is to be met.

There are a number of caveats. Carbon taxes have played a minor role in abatement policy, but other instruments can be associated with a carbon tax that has an equivalent effect on emissions (and necessarily a higher cost). The economy’s carbon intensity tends to fall over time, and would fall faster is climate policy is successful. On the other hand, I omit land use emissions and other greenhouse gases. I ignore wider economic effects, which could be quite substantial if all other taxes are abolished. I neglect distributional effects. Carbon taxes are regressive, and the maximum permissible tax would be lower if there were constraints on both the total revenue and the impact of the income distribution.