In reality, because of the importance of economies of scale, product differentiation and other considerations, producers in particular industries may have market power and use this power in determining decisions. This section considers potential qualifications to the distributional effects of policy interventions to reduce greenhouse gas emissions reported in the preceding section for a competitive model if producers use their market power in setting prices and quantities, and in particular in changing price and quantity decisions in response to the additional production costs of an emissions tax or a tradable-permit scheme.

While there are many different models of monopoly, oligopoly and monopolistic competition, they have some common properties which are germane to our questions. On the assumption that firms seek to maximise profits, they choose quantities and prices to equate marginal revenue, MR, with marginal private costs, MPC.

Assume initially that the firm demand curve has a constant elasticity of demand. A typical firm i has a MR_i function

MR_i = P_i (1 - 1/E_i) (1)

where, P_i is price on the firm’s perceived demand curve, and E_i is the absolute value of the elasticity of demand perceived by the individual firm i taking into account such considerations as the quantity and price-decision reactions of other firms in the industry. Note that profit-maximising firms choose an output where demand is elastic, that is E_i > 1, so that MR_i > 0. Equating (1) to the firm marginal cost, MC_i, the firm sets price as a mark-up over marginal cost, with the mark-up given by E_i/( E_i - 1) > 1, at

P_i = (E_i/(E_i - 1)) MC_i (2)

Note from (2) that the competitive model of the previous section is a special case of (2), since as the perceived firm demand elasticity becomes more elastic, and in the extreme perfect competition case E_i = ∞, the mark-up approaches unity.

More generally, when the perceived firm level demand elasticity E_i is very large, the competitive model assessment of the distributional effects of market measures to reduce greenhouse gas emissions will provide a good approximation. In these cases, P ≈ MC, and the industry supply curve is also approximately the MPC shown in Figure 1. In the case of monopolies, often the price is regulated to be close to MC. In most cases of monopolistic competition there are many firms with fairly similar or close substitute product options. In the case of oligopoly industries, for Cournot (or quantity setting) firms the perceived firm elasticity of demand increases with the number of firms, and for Bertrand (or price-setting) firms the price set approaches MC the closer are the firm product substitutes, and for perfect substitutes P = MC as under perfect competition. For many of the major greenhouse gas-emitting industries, there are similar quality products; for example, electricity is electricity. There are a large number of firms with differentiated products, but where some of the different products are close substitutes — for example, motor vehicles — this suggests the competitive model results will be a reasonable approximation of the distributional effects of market-based policy interventions to reduce greenhouse gas emissions.

Suppose instead that greenhouse gas-emitting firms are able to, and in practice do, exercise market power and set prices according to (2); that is the perceived firm demand elasticity E_i is, say, 5 or less elastic. Then, BAU output, including greenhouse gases, will be less than the competitive model, and the initial market price for the good goods will be above the competitive market price. More importantly, using (2) and assuming a constant marginal cost,^[6] the effect of a carbon tax or the opportunity cost of a tradable permit, T, to reduce greenhouse gas emissions on the consumer or market price P_c will be

dP_c/dT = E_i/(E_i - 1) > 1 (3)

That is, unlike the competitive model in which 100 per cent of the tax or permit price is passed forward to consumers, here more than 100 per cent of the additional cost is moved forward to consumers.

Now, rather than assuming, as was done above, that the firm demand curve has a constant elasticity at all price–quantity combinations, suppose instead that we assume a linear demand curve (with demand becoming more elastic at higher prices). In the special case of a monopoly, only a half of any marginal cost increase, including that associated with a greenhouse gas emission-reduction policy, would be passed on to consumers. With a Cournot oligopoly, the price increase enlarges with the number of firms and approaches 100 per cent for many firms (Smale, et al., 2006). In the case of monopolistic competition, Ng (1986) shows that more than 100 per cent of any cost increase will be passed forward to consumers as higher prices. Here, the emissions tax or the opportunity cost of the tradable permit increases both the average and the marginal cost, and the reduction in firm numbers (because of the higher price and less aggregate industry demand), combines to reduce the slope of the firm demand curve at which the new higher equilibrium price is established.

In principle, we can point to a wide range of different models of firm conduct with the use of market power, to differences in the shape of the demand curve facing each firm, and to differences with the shape of firm cost curves. Different combinations result in less than 100 per cent, about 100 per cent and more than 100 per cent of the increased costs to firms of policy interventions to reduce greenhouse gas emissions been passed forward to consumers as higher prices. But, in most cases, with the monopoly with a linear demand curve being the main exception, a cost pass-through of 100 per cent or more is the behavioural response. We now turn to some empirical evidence on the rate of pass-through of higher taxes or tradable-permit costs on greenhouse gas emissions on consumer prices for carbon-intensive products.