The Life Cycle of a Created Market: Redesigning Italy's Plastic Recycling Policy After Its Success

Some markets on which environmental policy relies were created by the state. The policy logic that justifies the creation of such a market is different from the logic that should govern it once it is mature. I study the principal case, plastic packaging recycling in Italy, which was created from nothing after 1997 by a per-tonne subsidy to separate collection, financed by a levy on packaging producers, together with public co-financing of reprocessing plant. I build a calibrated dynamic equilibrium model of the chain with two slow-moving stocks, the shared disposition of households to sort and reprocessing capacity, and obtain three results. First, the market could not have created itself. With the collection subsidy as the only instrument, no build-up plan from the empty state that I can find has positive present value at any credible subsidy, even under complete confidence that the market will mature. A documented capital co-financing share of 30–50 percent changes this result, as above a threshold share an investing equilibrium exists alongside the trap, and the credibility of the commitment selects between them (the confidence investors require decreases with the co-financing share). Second, run forward from the 1998 state under the instruments actually in force, with forward-looking investors and one number set to the history, the model reproduces the build-up of the following 27 years. Third, the optimal support path is self-eroding (it declines over time). For a decade it holds collection exactly to the capacity being built, as the binding constraint is the speed at which plant can be installed. It declines thereafter, because its own success reduces its justification, and it settles at 260 euro per collected tonne against the 322 actually paid. The actual instrument is indexed to collection costs and therefore does not decline, so that over-collection at maturity is the predictable consequence. The stakes are asymmetric. The missed reduction of the transfer costs the equivalent of 0.17 percent of consumption, and abolition costs twenty-five times that. I therefore recommend a redesign of the indexation that permits the transfer to decline.