May 2000 - As instruments for controlling pollution, how do emissions taxes and emissions trading compare in terms of the incentives they create to adopt cleaner technologies? Emissions taxes may have a slight advantage over emissions trading. Kennedy and Laplante examine policy problems related to the use of emissions taxes and emissions trading, two market-based instruments for controlling pollution by getting regulated firms to adopt cleaner technologies. By attaching an explicit price to emissions, these instruments give firms an incentive to continually reduce their volume of emissions. Command-and-control emissions standards create incentives to adopt cleaner technologies only up to the point where the standards are no longer binding (at which point the shadow price on emissions falls to zero). But the ongoing incentives created by market-based instruments are not necessarily right, either. Time-consistency constraints on the setting of these instruments limit the regulator's ability to set policies that lead to efficiency in adopting technology options. After examining the time-consistency properties of a Pigouvian emissions tax and of emissions trading, Kennedy and Laplante find that: If damage is linear, efficiency in adopting technologies involves either universal adoption of the new technology or universal retention of the old technology, depending on the cost of adoption. The first-best tax policy and the first-best permit-supply policy are both time-consistent under these conditions; If damage is strictly convex, efficiency may require partial adoption of the new technology. In this case, the first-best tax policy is not time-consistent and the tax rate must be adjusted after adoption has taken place (ratcheting). Ratcheting will induce an efficient equilibrium if there is a very large number of firms. If there are relatively few firms, ratcheting creates too many incentives to adopt the new technology; The first-best supply policy is time-consistent if there is a very large number of firms. If there are relatively few firms, the first-best supply policy may not be time-consistent, and the regulator must ratchet the supply of permits. With this policy, there are not enough incentives for firms to adopt the new technology. The results do not strongly favor one policy instrument over the other, but if the point of an emissions trading program is to increase technological efficiency, it is necessary to continually adjust the supply of permits in response to technological change, even when damage is linear. This continual adjustment is not needed for an emissions tax when damage is linear, which may give emissions taxes an advantage over emissions trading. This paper - a product of Infrastructure and Environment, Development Research Group - is part of a larger effort in the group to study the economics of pollution control. Copies of the paper are available free from the World Bank, 1818 H Street NW, Washington, DC 20433. Benoit Laplante may be contacted at blaplante@worldbank.org.