Module VII – Taxes & Permits – Part II

This section looks at the practical application of economic instruments with a focus on efficiency as well as social cost minimization. It is assumed that targets are set through a political process using scientific inputs on likely damages and economic inputs on both damage costs and control costs. There exists 2 types of targets – target reduction in emissions output (for example, target reduction of 10 mi tonnes of SO2 levels) and target improvement in ambient environment quality (for example, fixing upper limits for lead content in drinking water). Also, a source is a point of discharge of pollution, a receptor is a point at which the level of ambient pollution is measured and the transfer coefficient is the ratio of change in pollution at receptor to the change in emission at the source. Some ways to reduce emissions include reducing output, changing the production process, using end-of-pipe technologies etc.



An important principle used in taxation for uniform pollutants is the Baumol Oates Least Cost Tax theorem which states that:

“A tax rate set at a level that achieves the desired reduction in the total emission of pollutants will satisfy the necessary conditions for the minimization of the programme’s cost to society.”

The implications of this theorem are that: (a) the firm has no price-setting power in the input or pollution abatement markets, (b) least cost tax = Marginal abatement costs at the target level of emission, and (c) for a given tax, the marginal abatement costs across all firms must be equal under cost-minimising solution.

In the case of a non-uniform pollutant, a single tax is no longer efficient since the tax rate will vary across sources. The target here will be to reduce ambient concentrations to some target ambient level. To achieve an efficient solution, each firm must face a different tax t*k which is determined by that firm’s degradation of environmental quality at each monitoring point (given by transfer coefficient) and by the ambient target. Separate tax rates for each monitoring point which are adjusted for each firm according to its transfer coefficient relating to that point can be levied, the only disadvantage being that it is administratively and politically difficult. In such cases, zonal taxes are preferred. A zonal tax is levied region-wise wherein each region or river basin is divided into zones and within each zone, the same emission fees applies. Different zones have different fees. Zonal taxes provide flexibility but are difficult to administer. Too many zones may lead to an ambient system being set in place.

Other kind of taxes include input taxes that tax firms based on the inputs they utilize. The downside of input taxes is that there is possibility of substitution which, if it goes undetected, may eventually be more harmful on the environment. Product taxes are another option if a predictable relationship between output and emissions can be found but taxes will be dependent on when the product is used and how it is disposed.


Problems with pollution taxes

It is important to consider several policy issues pertaining to pollution taxes. An initial incorrect tax rate can lock firms into incorrect investments in pollution control equipment. In such cases, costs are not minimised, as shown in Walker and Story (1977). A low initial rate may result in irreversible environmental damage. It is tricky getting the rate right due to the aggregate marginal abatement cost (MAC) function being unstable over time. This is caused by real-term fluctuations.

Another issue is that aggregate emissions increase with new entrants, as the aggregate MAC goes up. In the case of a basket of pollutants, regulating to set the correct taxes is a nightmare. An issue linked with equity is the undesirable redistributive effect pollution taxes have on households. Moreover, due to uncertainty over actions, the environmental target under a pollution tax regime is achieved only if: (a) all polluters are cost minimisers, (b) all producers are well-informed about their MAC schedules, and (c) no untaxed emissions are possible.


[Fig.1] Comparative regulatory analysis

Consider the figure above. The Y-axis is the ‘total cost of control’ mapped on the X-axis’ ‘particulate concentration’. This is based on a study done in St. Louis by Atkinson and Tietenberg (1982). They compared the control costs for these three regulatory systems. However, a limitation is that they looked not at marginal damage but standard ambient levels.



Emission Permit System (EPS) applies for uniform pollutants. Firm A reduces emissions by 1 unit and sells this permit to firm B which can increase its emissions by 1 unit. Ambient Permit System (APS) applies for non-uniform mixing pollutants each source’s marginal abatement cost is equal to weighted average of shadow cost of emission reductions needed to achieve target.

When there are multiple sources and receptors, market equilibrium exists in buying and selling the ambient pollution permits for any initial issuance of permits. After trading has occurred, the number of permits held by various firms for polluting each receptor must be less than or equal to the number of permits initially issued for that receptor.

Pollutants can be classified based on the concept of time into stock (pollutant that accumulates over time), flow (pollutant that fades with time) and temporal variability. It is important that marginal damage must include the damage that is over the time period when pollutant is resident in the environment.

The marginal savings from emitting a unit of pollution today equals the sum of all the marginal damage that may occur in the future. This marginal damage is discounted by 2 factors – the discount factor and the persistence of pollutant.

Accumulation of pollutants also differs with respect to whether they are being emitted during the day or during night times, which season, and the accumulative capacity such as whether it is a windy space or still space. The damage differs depending on these above mentioned factors. The analysis is however, similar to that of space.

When multiple pollutants are considered, the case is no longer linear. Permits issued for a certain amount of acid deposition can be quite effective and they can be converted into emission rights based on the different transfer coefficients aij (here, the ratio of increased pollution of type j from increased emissions of compound i)


Basic Debates

Some important design issues include whether firms should be allowed to ‘bank’ emission reduction credits and whether trade should be allowed between point and non-point sources of pollution. Banking emissions can be seen as a hedge against uncertainty and would be helpful in reducing pollution but may also imply that it may lead to an increase in pollution at some other point of time. Allowing trade between point and non-point sources of pollution may lead to a welfare gain but one must keep in mind that the marginal abatement cost for point sources is higher than that for non-point sources.


Prices versus Quantities

In a situation wherein the government has complete knowledge about aggregate costs and benefits of reducing greenhouse gases and producers know their abatement costs, i.e., there exists perfect certainty, taxes and permits achieve exactly the same abatement at the same cost. In reality, the government knows less than the producers about costs and the producers have imperfect knowledge and hence no one can say with certainty whether a price instrument (tax) or a quantity instrument (permit) will achieve higher abatements. While a tax limits the marginal cost of abatement, it offers more economic certainty but offers no guarantee regarding the emission levels, a permit, by limiting total emissions, offers more environmental certainty but does not necessarily limit the marginal cost of abatement.

Weitzman discusses this very issue in his paper Price Vs. Quantities (1974). He concludes that when the marginal benefit of a curve is steep, quantity instruments such as permits are more effective, whereas when the marginal benefit curve is flat, a price instrument like the tax is preferred.

Taxes yield lower expected losses than permits and provide certainty regarding the quantity of money that would be spent. Taxes are also relatively easier to administer. But choosing a price instrument can be more disastrous than a quantity instrument as crossing a particular threshold, say for emissions, when a price instrument is in place can turn out to be a catastrophe. Quantity instruments, on the other hand, can prevent breaching of known threshold limits. It also helps in preventing under or overpaying. The disadvantage regarding quantity instruments is that they may lead to big price swings.

Consider a situation where a country must bring about an 83 percent reduction in greenhouse gas emissions (compared with 2005 levels) by 2050. It could adopt two different mechanisms: a quantity instrument such as cap-and trade or a price instrument such as a carbon tax. In a cap-and-trade mechanism, since the price of the permit is not known in advance, it creates uncertainty in the cost of compliance for firms whereas the use of emission taxes may lead to an environmental outcome that is not guaranteed. If a particular threshold is crossed, this would lead to horrific runaway green house gas effect. Using Weitzman’s result, one may conclude that, if the threshold is known, then a quantity instrument is best utilized whereas if it is not known and if the situation is not that risky, a price instrument is the option to choose.

Recently, hybrid models, such as a cap-and-trade system with a ‘safety valve’ are in vogue. These limit the maximum market price of emission permits, simultaneously mimicking the impact of an emission tax. Using an emission cap, combined with a tradeable permit system with the maximum price capped would help overcome fundamental disadvantages prevailing in both systems and permits flexibility.

Further examples of price and quantity controls being used in the field of renewable resources would be in Boston where quantity instruments have been introduced to ensure that 15% of electricity is obtained from renewable resources by 2020 whereas in Europe, price mechanism is being utilized to encourage the development of expensive solar power by guaranteeing high payments to producers of renewable energy.


Implementing marketable permits

The relevant concerns linked with initial permit issuance include the possibility of a huge resource transfer from firms to the government. This can be avoided by means such as a zero revenue auction. The initial allocation is key, and it is possible to achieve the same effect as an efficient fee. Another problem is posed by the question of permits for new entrants, which a good design will accommodate.


Problems with Tradable Pollution Permits (TPP)

[Fig.2] Monopsonist

The first major problem is market power. Consider the figure above, mapping Y-axis ‘total cost of control’ on to X-axis ‘particulate concentration’. Dominant firms (monopsonists) bring about a manipulated emission level rather than the efficient one. A correct initial allocation is necessary to attain proper emission levels. There are two kinds of manipulation that firms engage in – cost-minimising manipulation and exclusionary manipulation, the latter to disadvantage rivals in the product market. In cases of market thinness, equalizing marginal control costs across firms is hard, since transactions are infrequent and permit costs are high.

Taking up the problem of transaction costs: this negates the main benefits (trading permits and pricing them equal to the opportunity cost of emitting). Stavins (1995) lists three types of sources for such costs: (a) search and information costs, (b) bargaining and decision making costs, and (c) monitoring and enforcement costs. Foster and Hahn (1995) note that transaction costs work against small trades. Tietenberg (1990) gives examples of an offsetting scheme (1977) and a bubble scheme (1979), and the EU Emissions Trading System (ETS), contending that its competitiveness and rent-seeking is due to grandfathering, and noting that national non-compliance of member states is an issue.

Trading rules and non-uniform mixing poses another problem for TPP. Emission Permits System (EPS) must be played off versus Ambient Permits System (APS); while the latter are complicated, they are more appropriate for non-uniformly mixed pollutants. The rules for offsets are simple: the permits are in units of emissions, and their trade is governed to stop violation of ambient quality charges. There are three types: (a) pollution offset, in which trade must not violate ambient levels at receptor points, (b) non-degradation offset, with the additional condition that there should be no increase in emissions due to trade, and (c) modified pollution offset, which ensures that neither pre-trade nor target ambient levels are violated. Atkinson and Tietenberg (1987) used the cases of St. Louis and Cleveland to compare that the Least-cost solution (APS) versus the State Implementation Plan (command & control), which was exorbitantly expensive in comparison. There is, notably in this regard, also a sulphur trading case study that refers to the 1990 Clean Air Act Amendments in America.

Another problem is whether to favour grandfathering or auctions. While the former is politically feasible, it creates rents and runs the risk of moral hazard, with firms increasing emissions over cost-minimising levels in the run-up to reissue. An auction system can be designed in various ways, such as single price, but should preferably follow the incentive-compatible Groves mechanism, for instance using a Vickrey second-price auction adaptation. Another problem is that of sequential trading, for which Atkinson and Tietenberg have outlined four scenarios: (a) simultaneous, full information, no increase in total allowed, (b) sequential, full information, biggest cost-saving trade downward, (c) partial information, lowest cost firm is the first seller, and (d) partial information, randomly selected firm is the first seller. The penalty is greater for stricter targets in a static one-shot formulation, as opposed to a dynamic market process. The price formation process can be separated from finalising contracts, allowing for a bilateral sequential process, which has been shown to replicate the least-cost solution. Moreover, the role of strategic behaviour, cheating and market power must also be considered specifically in sequential trading.


Taxes versus Permits

[Fig.3] Savings under innovation diffusion and regulatory response (Y-INR, X-emission reductions).

[Fig.4] Savings under innovation with pollution tax (Y-INR, X-emissions).

Considering innovation and cost savings over time, the graphs are self-explanatory. Requate and Unold (2003) put forward that  due to the likelihood of individual firms tending to be free riders, taxes work better than permits for innovation. The regulator’s anticipation of innovation and pre-commitment to change in policy is required.



Pizer (1999, 2002) notes that we have little experience with large emission cuts, do not know future technological options, nor the ‘do nothing’ level of emissions. Montero, on a different note, adds that there is uncertainty on the part of firms over whether the regulator will approve trade or not. When it comes to the use of economic instruments for pollution control, or rather the lack thereof, most of it can be attributed to policymakers being unaware of the potential, practical problems such as few TPP traders and lack of political acceptability, and also institutional problems, with cost effectiveness not necessarily being the primary objective, and also pertaining to the ethical implications of economic instruments. Ultimately, there is a trade-off between increased financial burdens on firms but reduced abatement cost for society, and to make this less of a departure from the status quo, partial grandfathered schemes may be used, though it must be noted that all sizable tradable permit systems implemented to date are grandfathered. Little surprise then that there is resistance to significant changes in pollution control policy. In the light of this, gradual introduction is essential.


Why market policies fail to address basic environmental objectives

Ackerman and Gallagher contend that “the market is a reasonable policy tool but not a reasonable blueprint for society’s goals”, and outline five forms of failure, which are: (a) large irreversible damages must be prevented, (b) outcomes for the future are important ($1@5%pa = $%17,000 in 200 yrs and $2m in 300 yrs), (c) many environmental values are not commodities that can be priced, (d) volatile market prices can cause wasteful misallocation of resources (early ‘80s oil price and small cars, 1995 recycled paper and 1997 mills shutdown), and (e) “if it’s not broken, don’t fix it”. They conclude by saying that the market mechanism decentralizes information processing and decision making, allowing each firm to analyze and respond to the data that affects its operations, and with complex technical choices requiring site-specific information, broad standards allow firms to choose the most cost-effective way of meeting standards. Hence market based policy is important. But they qualify this by pointing out that other approaches must be re-legitimised to enable a broader dialogue about the full range of options for environmental policy.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: