Climate Change 2001: Working Group III: Mitigation

	Other reports in this collection

8.2.2.1.4 Conclusions: Interest and Limits of Aggregate Analysis

A lesson from this section is that, despite their great diversity, the findings of empirical models confirm the theoretical diagnosis. Revenue-raising instruments such as carbon taxes or auctioned emissions permits are, if properly utilized, the most efficient instrument for minimizing the aggregate welfare losses (or maximizing the welfare gains) of climate policies.

It should be noted however, that, even if the only one available study for China suggests that opportunities for revenue recycling exist in developing countries, no swapping generalization can be made at this stage. While theoretical modelling and empirical evidence suggest that such opportunities are available in many OECD countries, developing countries in many cases start from a different fiscal baseline (e.g., fewer entrenched distortionary payroll taxes). They also have other potentially underused tax bases that may become more developed as their economies grow at rates that typically exceed growth rates in OECD countries. In developing countries, direct welfare losses associated with a carbon tax may, therefore, reduce opportunities for mitigation within the fiscal reform policy envelope. At this stage, however, insufficient evidence exists either to confirm or to substantiate these hypotheses; studies to date have mainly concentrated on developed countries and their conclusions may not be directly transferable.

Beyond controversies about the capacity of government to warrant fiscal neutrality, that is the fact that the total fiscal burden remains unchanged, the adoption of carbon taxes or auctioned permits confronts the fact that their enforcement must be done in the heterogeneity of the real world, and can have very significant distributive implications:

• Across economic sectors. The carbon content of the steel, aluminium, cement, basic chemical, and transport industries are, indeed, four to five times higher per unit of value added than for the rest of industry. For unilateral initiatives, carbon taxes drastically impact the competitiveness of these sectors (with potential economic shocks at the regional level); even with an internationally co-ordinated policy, their equity value will be lowered compared with the rest of industry.
  
• Across households income groups. Carbon taxation increases the relative prices of energy services such as heating, lighting, and transport. The resultant impact on welfare is then more negative for low income levels and people living in cold areas and in low density areas. It is also higher for high income groups and more beneficial for medium income groups in case of swap with other taxation.

Economic analysis can define the compensation necessary to offset these negative distributional effects but, in the real world, winners cannot (or are not willing to) compensate losers. This is especially relevant when the losers suffer heavy impacts and the winners enjoy only marginal gains, which leads to the so-called political mobilization bias (Olson, 1965; Keohane and Nye, 1998) when the losers are more ready to organize a lobbying and incur mobilization costs than the winners (Williamson, 1996). Under such circumstances, policies yielding the largest aggregate net benefits may prove very difficult to enforce. Economic models provide no answer to this issue, but can try to frame the debate by providing the stakeholders with appropriate information. This is the objective of Sections 8.2.2.2 and 8.2.2.3.