In the preceding discussion, a competitive equilibrium in the world economy was assumed. However, OPEC may be able to exercise a degree of monopoly power over the supply of oil. The issue has been raised in the literature as to the possible nature of an OPEC response to reduced demand for oil as a result of Annex I abatement. If in the short term OPEC were to reduce production to maintain prices in the face of lower demand, the time path for Annex I carbon taxes may need to be modified. See also Chapter 9.
A number of theoretical papers examined how a carbon tax might alter the optimal timing of extraction of given reserves of oil and, symmetrically, how significantly the potential supply response could alter the optimal time path of the price of carbon tax (Sinclair, 1992; Ulph and Ulph, 1994; Farzin and Tahovonen, 1996; Hoel and Kverndokk, 1996; Tahvonen, 1997). However, the severity of the potential problem depends on a number of key parameter values and implementation issues. Although it has been assumed that OPEC can Granger cause the world price of oil (Güllen, 1996), there is some question about the degree of cartel discipline that could be maintained in the face of falling demand (Berg et al., 1997a). Any breakdown in the cartel would tend to increase the supply of oil on the market, which in the short term may require a higher carbon tax to meet a given abatement target. On the other hand, Bråten and Golombek (1998) suggest that implementing an Annex I climate change agreement might be seen by OPEC members as a hostile act and could strengthen the resolve to maintain cartel discipline. The OPEC response is likely to be related to the size of its potential loss in revenue to OPEC and these potential losses would be smaller under Annex I emissions trading than under independent abatement.
A number of empirical studies have tried to assess the significance of the potential OPEC response within a game theoretic framework. To do so, Berg et al. (1997b) resorted to a CournotNash dynamic game in which parameter values are based on empirical estimates. They also identify (non-OPEC) fringe oil producers and other fossil fuel sources. A scenario is examined in which a carbon tax is maintained at a level of US$10 per barrel of oil. Initially, OPEC cuts back on production to try to maintain price, but this is partly offset by increased production by the fringe. Bråten and Golombek (1998) derive a similar pattern of OPEC response in a static model. Berg et al. (1997b) found that the optimal OPEC policy is not heavily influenced by intertemporal optimization in shifting supplies from one time period to another to maximize discounted net revenue.
If OPEC acts as a cartel, the extent of emissions leakage in response to Annex I abatement may be reduced (Berg et al., 1997b), because the resultant higher price for oil reduces the incentives for increased emission-intensive activity in non-Annex I regions. Lindholt (1999) examined the Kyoto Protocol in an enhanced version of the same model and assumed that an efficient tradable permit scheme is established between Annex B countries. Whether or not OPEC acts as a cartel does not affect the shape of the time path of permit prices, only their level according to Lindholt (1999). A permit price of US$14/tCO2 would be required in 2010 if OPEC acts as a cartel, whereas it would be US$24/tCO2 in a competitive oil market. The lower permit price when OPEC acts as a cartel stems from OPEC cutting back production to maintain a higher oil price, which slows the growth in emissions in Annex B countries.
These studies mentioned demonstrate that whether or not OPEC acts a cartel will have a modest effect on the loss of wealth to OPEC and other oil producers and the level of permit prices in Annex B regions. A natural extension of this research would be to trace through all the ramifications of cartel behaviour by OPEC in the more complex CGE models discussed in this section.
In a dynamic context, a progressive outward shift in the production possibilities frontier occurs over time as a result of technical change. A strand of literature (Goulder and Schneider, 1999) argues that climate policies will bias technical change towards emissions savings. In that case, there will be an outwards shift in the production possibilities frontier at some points, and an inwards shift at other points relative to the baseline.
One potentially important related issue not captured in the above models is that cleaner technologies, developed in response to abatement measures in industrialized countries, tend to diffuse internationally. The question is to what extent this will offset the negative aspects of leakage noted above and to amplify positive spillover. Grubb (2000) presents a simplified model, which represents this spillover effect in terms of its impact on emissions per unit GDP (intensities). The results suggest that, because the impact of cleaner technologies is cumulative and global, this effect tends to dominate over time, provided the connection between industrialized and developing country emission intensities is significant (higher than 0.1 on a scale where 0 represents an absence of connection and 1 a complete convergence of intensities by 2100). At this stage, empirical analysis is still lacking to derive a robust conclusion from this result. A recent work by Mielnik and Goldemberg (2000) suggests that the potential for technological leap-frogging in developing countries is important, but to what extent climate mitigation in Annex B accelerates this leap-frogging is still unclear. However, this demonstrates that the trickling down of technical change across countries deserves more attention in modelling exercises, all the more so since theoretically it (see Chapter 10 of this report) demonstrates that technological spillovers may be a major stabilizing force of any climate coalition.
Other reports in this collection