pigouvian-tax, damage-function, externalitiesWhen internalising an externality, one way to arrive at the level of a suitable Pigouvian tax is to estimating the damage function, and from there, to calculate the marginal damage cost of a unit of pollution: for example, Chris Hope’s work for the Stern Review on the climate damage function and the marginal damage cost of a unit of CO2 emissions.
Once we have that marginal damage cost, then it seems reasonable that a market at equilibrium would have a Pigouvian tax at that level. And if we were starting with a tabula rasa, a blank slate, then that would be the right level too.
However, such a tax (for example a carbon tax) is, in the real world, introduced into a market where capital investment has already taken place, leaving the market far from the equilibrium, and far from the optimal resource allocation that could be found if the externality had been internalised all along.
The marginal cost for incumbents will be lower than their average cost, in the short- to medium- term: they already have sunk costs, in their fixed asset base: factories, workforce skills, institutional structures. These incumbents with sunk costs slow down the move to a new equilibrium, and can act as a barrier-to-entry to disruptive new entrants who would otherwise catalyse the move to the new equilibrium.
So, if one didn’t want to model all capital investments, but wanted a less data-intensive means of estimating the appropriate level of Pigouvian tax to minimise unnecessary damage and bring the market to new equilibrium in an orderly, but prompt manner, what information would be needed, to trade off the cost of stranded assets versus the cost of delay?
There were no answers to this question.
All content is licensed under CC BY-SA 3.0.