Taxing CO₂ emissions and its ancillary health benefits: a computable general equilibrium analysis for Tunisia

Abstract

Since the middle of the last century, Tunisia has embarked on an ambitious trade reform program aimed at improving its integration in the world economy, boosting growth through valorizing comparative advantages, and reducing unemployment among its population. However, and despite the positive role that trade may play in improving growth through better allocation of domestic resources and lower costs of imported equipments and raw materials, the risk is to amplify output in sectors intensive in energy in a country where energy is still subsidized. Introducing pollution abatement taxation has been suggested as a way to achieve ancillary benefits from reduced local air toxics. The highest level of local air pollution is found in heavily populated cities where labor is concentrated and where labor health is believed to have been significantly impacted. The objective of this paper is to address this important issue. It identifies the optimal and ‘no regrets’ pollution abatement tax on a net welfare function, which integrates both net health benefits and adjustment costs. The paper uses a Dynamic Computable General Equilibrium (CGE) model for the assessment that allows the health benefits to feed back into the economy. A health effects sub-model takes the local air emissions output from the CGE model and assesses the implications for ambient air concentration levels and health effects. The results suggest an ‘optimal’ abatement rate in 2020 of around 25% of CO₂ reduction compared with the baseline 2020 emissions. However, the most significant impact concerns the relatively small aggregate cost of pollution abatement in terms of forgone real average growth rate of GDP between 2010 and 2020 for the trade scenario with ‘optimal’ climate policy. Finally, the major consequence of pollution abatement policies is the reduction of production generated by polluting activities against a higher production of less-polluting activities.

Keywords:

• environmental economics

JEL Classifications:

• D58
• Q51
• F1

Acknowledgements

This work was supported by the Economic Research Forum (grant number 2009–015).

Notes

1. Bussolo and O'Connor (2001) estimate the health impact of environmental degradation in India, while Chemingui (2002) estimates the economic costs of a pollution abatement policy in Tunisia in the context of its FTA with the EU.

2. These effects are scale, technology, and composition effects. A scale effect is observed when greater economic activity raises demand of all inputs and increases emissions. A technological effect exists and tends to reduce emissions when higher effluent charges encourage firms to shift toward cleaner production processes. Finally, a composition effect is observed when income growth shifts preferences toward cleaner goods and the share of pollution-intensive goods in output falls.

3. A useful reference for the ELES approach is found in Lluch (1973).

4. In the utility function, S needs to be deflated by an appropriate price, which would represent the consumer spot price of future consumption. This price does not need to be specified for the model since household saving can be derived as a residual from the budget constraint. For welfare calculations, the consumer price index, cpi, has been chosen as the saving deflator since there is no forward-looking behavior in this model.

5. Armington (1969).

6. Increased import demand, due to trade liberalization must be financed by increased exports, and these can expand owing to the improved resource allocation. Price decreases in importables drive resources towards export sectors and contribute to falling domestic resource costs (or real exchange rate depreciation).

7. The model's long-run properties are discussed in the technical document.

8. Garbaccio et al. (2000) and Cifuentes et al. (1999) provide a detailed description of this approach.

9. CGE models taking into account technological standards are based on the notion of product differentiation. Usually, this type of model integrates many types (or quality) of the same commodity, such as green and classical products through specific production and consumption functions for each type (Schubert & Zagame, 1998).

10. Dessus et al. (1994) describe the used methodology.

11. Tunisia has already implemented the FTA with the European Union and with Arab countries. Some FTAs are at the end of their implementation period, such as with Turkey, and the European Free Trade Area member countries, while others are under negotiation, such as with COMESA.

12. There are three options of implementing abatement: delayed to the end of the simulation period, immediate, or gradually over the entire simulation period. Each option has its own advantages and disadvantages as with the case when implementing trade reform. Arguments for delay usually center on the scope of lowering abatement costs by waiting for new technologies to become available. Arguments for early action usually cite the costs averted from premature obsolescence of polluting capital equipment. In economic modeling, it is preferred to opt for a medium option which is between not early and no delay, which means in the middle of the simulation period.