Chapter 12
Emission Taxes and Subsidies
If we want to build a house, we have to buy building materials; nobody is likely to give them to us free. If we want to have architects and carpenters work on the house, we will have to hire them; they won’t work for nothing. In other words, in order to use the services of these inputs, we have to pay for them. We are used to doing this, because these goods and services are bought and sold in well-developed markets. The fact that we have to pay for them gives us an incentive to use the inputs as sparingly and efficiently as possible. The economic-incentive approach to environmental policy works in much the same way. Until recently people have been able to use the waste-disposal services of the environment virtually without cost, so there has been little incentive for them to think about the environmental consequences of their actions and to economize on the use of these environmental resources. The incentive approach seeks to change this situation.
There are basically two types of market-based incentive policies: (1) taxes and subsidies and (2) transferable emission permits. Both require a regulator to put the program into effect and to monitor outcomes, so they are less decentralized than liability laws or letting parties bargain over emission levels. Regulators set a price for pollution via taxes and subsidies and set quantities of allowed emissions with transferable emission permits. The market determines the price of pollution under the permit approach. Under each policy, polluters make their own decisions about the amount of pollution to emit based on the prices per unit pollution they face. Governmnets worldwide are increasingly turning to economic incentives, including some examples in Canada.
In the United States and Europe, emission markets have been in place for a number of years; for sulphur dioxide in the US and greenhouse gases in Europe. This chapter examines the economic theory of emission charges and subsidies; Chapter 13 covers the technique of using transferable emission permits. Chapters 17 and 20 look at policies in Canada and other countries in practice.
Economists have long promoted the idea of incorporating incentive-based policies more thoroughly into environmental policies.1Incentive-based policies can be more cost effective than standards and provide more stimulus for polluters to seek cost-reducing abatement strategies. can serve to put more teeth into environmental policies in many cases and substantially improve the cost-effectiveness of these policies. But keep in mind something we have said before: No single type of policy is likely to be the best in all circumstances. Incentive-based policies are no exception. They have strengths and they have weaknesses. The strengths are sufficiently strong to encourage greater reliance on them in many circumstances. But there are types of environmental problems where they may not be as useful as other approaches, and politically, they may be more challenging to install than standards. Chapter 14 looks in more detail at the tradeoffs among policies and why polluters and regulators may favour one type over another
1. An economist who emphasized the role of taxes as a method of internalizing externalities was A.C. Pigou, way back in the 1930s. Environmental taxes are sometimes called Pigouvian taxes after him.
Emission Taxes
The most straightforward incentive-based approach to controlling emissions of a particular residual is to have a public agency offer a financial incentive to change those emissions. This can be done in two ways: by taxing each unit of emissions, or by giving a subsidy for each unit of emissions that the source cuts back.
We deal first with emission taxes, sometimes also called emission charges. Emission taxes imply that polluters are able to discharge any amount of the taxed pollutant they wish, but they will be required to pay a tax for every unit (e.g., tonne) discharged. For example, the British Columbia government has imposed a carbon tax in 2008 on over 75 percent of the greenhouse gases emitted in the province as a means of reducing carbon dioxide emissions and ameliorating global warming. BC’s goal is to reduce carbon emissions by 1/3 of their 2007 level by 2020 and to be 80% below 2007 levels by 2050. When an emission tax is put into effect, those responsible for emissions must pay for the services of the environment—transportation, dilution, chemical decomposition—just as they must pay for all other inputs or goods they use. Once pollution is “priced” by the tax, those who release it will have an incentive to release less of it; that is, to conserve on their use of environmental services. How do they do this? Any way they wish (within reason). This may sound flippant, but in fact it represents the main advantage of this technique. By leaving polluters free to determine how best to reduce emissions, they can use their own energy and creativity, and their desire to minimize costs, to find the least-cost way of reducing emissions. It could be any combination of pollution abatement, substitution of one good for another, internal process changes, changes in inputs, recycling, or shifts to less-polluting outputs. In the case of BC’s carbon tax, people may reduce their dependence on motor vehicles by driving less, taking public transit, car pooling, adding more insulation to their homes to reduce heating costs. Industries may shift from higher carbon-intensive fuels such as petroleum and coal to less carbon-intensive fuels such as natural gas or electricity, which in BC is predominately produced by hydro-power and hence, carbon free. More detail on BC’s carbon tax is provided below.
The essence of the tax approach is to provide an incentive for the polluters themselves to find the best way to reduce emissions, rather than having a central authority determine how it should be done.
The Basic Economics of Emission Taxes
The essential mechanics of an emission tax are depicted in Figure 12-1. The numbers refer to a single source of a particular pollutant who has a marginal abatement cost function of MAC = 200 – 4E. Assume that the regulator has set the emission tax at $100 per tonne per month. The top panel shows the analysis numerically, while the bottom shows the same information graphically. The second column of the table shows the firm’s marginal abatement costs and the third column shows total abatement costs at each emission level. The last two columns show the total monthly tax bill the firm would pay at different emission levels and the total private cost of compliance.
Figure 12-1: The Basic Economics of an Emission Tax
An emissions tax of $100 per tonne of pollutant released per month is levied on a polluter. The table shows marginal and total abatement costs, the polluter’s tax bill, and total costs. Total costs are minimized at discharges of 25 tonnes/month. This is shown graphically as the point where the tax rate intersects the polluter’s MAC curve. Area a is the tax bill; area b shows the total abatement costs.
Total private cost of compliance of an emission tax is defined as the sum of abatement costs and the tax bill for the polluter.
As we’ll see, these are not the same as social costs. The minimum total cost of $3,750 occurs at an emission rate of 25 tonnes/month. The logic behind this can be seen by considering marginal abatement costs. With no regulation, the polluter emits at E0 = 50 tonnes/month and pays a tax bill of $5,000 (i.e., 50 tonnes times $100); if it were to cut emissions to 45 tonnes it would cost $50 in abatement costs, but on the other hand it would save $500 in taxes—clearly a good move. Following this logic, it could improve its bottom line by continuing to reduce emissions as long as the tax rate is above marginal abatement costs. The rule for the firm to follow is this: reduce emissions until marginal abatement costs are equal to the emissions tax rate. This is shown diagrammatically in the bottom part of Figure 12-1. To reiterate,
polluters will minimize their total private costs by reducing emissions until the tax rate equals their marginal abatement cost.
After the polluter has reduced its emissions to 25 tonnes/month, its total (monthly) tax bill will be $2,500. Its monthly abatement costs will be $1,250. Graphically, total abatement costs correspond to the area under the marginal abatement cost function, labelled b in the figure. The total tax bill is equal to emissions times tax rate, or the rectangle labelled a. Total private cost is thus area (a + b).
Suppose the polluter is a firm. Why wouldn’t the firm simply disregard the tax, continue to pollute the way it has been, and just pass the tax on to consumers in the form of higher prices? If the firm stayed at 50 tonnes of emissions, its total outlay would be $5,000 per month, consisting entirely of tax payment. This is much higher than the $3,750 it can achieve by cutting back to 25 tonnes/month. If a firm operates in a perfectly competitive environment, it survives by maximizing its profits. Emission taxes raise the costs of the firm. Therefore, to maximize profits, the firm must do whatever it can to minimize its total costs inclusive of the emission taxes. The response will depend on several factors. The higher the tax, the greater the reduction, and vice versa. In the example of Figure 12-1, a tax of $50 would have led the source to reduce emissions only to 37.5 tonnes/month, while a tax of $180 would have produced a cutback to 5 tonnes/month. Also, the steeper the marginal abatement cost function, the less will emissions be reduced in response to a tax. We will come back to this below. Recognize, however, that if firms do not operate in perfectly competitive markets, a tax will not work in the way we have shown. Also, firms that sell their products in international markets and compete against others who do not pay environmental taxes may be unable to pass any of the tax along to consumers.
Emission Taxes vs. a Standard
Compare the tax approach with an emission standard. With the tax, the firm’s total outlay is $3,750. Suppose that, instead, the authorities had relied on an emission standard to get the firm to reduce emissions to 25 tonnes/month. In that case, the firm’s total outlay would be only the $1,250 in abatement costs. Thus, the tax system ends up costing the firm more than the standards approach. With a standard, the firm has the same total abatement costs as in the tax system but it is still essentially getting the services of the environment free, while with a tax system it has to pay for those services. But while polluting firms would thus prefer standards to emission taxes, there are good reasons, as we shall see, why society would often prefer taxes over standards.
The Socially Efficient Tax
In competitive situations, higher taxes will bring about greater reductions in emissions, but just how high should the tax be set? If we know the marginal abatement cost and marginal damage function, the economist’s answer is to set the tax so as to produce the efficient level of emissions, as in Figure 12-2. A marginal damage function, MD = 4E, is added to the MAC curve from Figure 12-1. Equating MD to MAC yields the socially efficient tax rate of $100 per tonne. If the regulator knows both functions, the tax per unit pollution is thus readily calculated.
Figure 12-2: A Socially Efficient Emission Tax
The socially efficient equilibrium is reached with a tax set equal to $100 per tonne. This is the “price” at which MD = MAC. The polluter’s private costs of compliance are its total tax bill paid, area (a + b + c + d), plus its total abatement costs, area e. Total social costs of compliance are just the TAC. The net benefit of the tax is the total damages forgone, area (e + f) net of TAC. This is area f.
What are the compliance costs of the tax policy? We must distinguish between private and social costs. We have already defined private costs of compliance as the polluter’s total abatement costs plus its tax bill. In Figure 12-2, private costs are, respectively, area e plus areas (a + b + c + d). But private costs of compliance do not represent the real resource cost society incurs as a result of levying the emission tax. It is social costs that society is interested in.
Social costs of compliance include only the real resources used to meet the environmental target; they do not include the tax bill.
Taxes are actually transfer payments, payments made by the polluters to the public sector and eventually to those in society who are benefited by the resulting public expenditures. The polluter itself may be a recipient of some of these benefits. Transfer payments are therefore not a social cost of the policy. Thus, the social costs of compliance are area e, the polluter’s total abatement costs.
Society is also interested in the net social benefits from the tax policy.
Net social benefits of a policy are defined as the total damages forgone net of the social costs of compliance.
Example: Compute net social benefits for Figure 12-2
The steps are as follows:
1.Compute total damages forgone.
Total damages forgone is measured by the area under the MD curve from the initial level of emissions to the socially efficient level, E*.
This is areas (e + f) = $3,750.
2.Compute total abatement costs.
TAC = area e = $1,250.
3.Compute net social benefits.
Net social benefits = total damages forgone minus total abatement costs.
Net social benefits = areas (e + f) minus area e = $2,500.
Compare the net social benefits of the emission to a standard set at the socially efficient level of emissions, 25 tonnes per month. The net benefits of the standard are identical. Thus a tax and standard, set at the same level, yield identical net benefits to society. What differs is the impact on the polluters.
The reduction of emissions from E0 = 50 to E* = 25 tonnes per month has eliminated damages of (e + f). Remaining damages are (b + d), an amount less than the firm pays in taxes. This underscores the idea that the emission tax is based on the right to use environmental resources, not on the notion of compensation. But a “flat tax” like this (one tax rate for all emissions) has been criticized because it would often lead to situations where the total tax payments of firms would substantially exceed remaining damages. A way around this is to institute a two-part emission tax. We allow some initial quantity of emission to go untaxed, applying the tax only to emissions in excess of this threshold. For example, in Figure 12-2 we might allow the firm E1 = 10 tonnes of emissions free of tax, and apply the tax rate of $100 per tonne to anything over this. In this way the firm would still have the incentive to reduce emissions to E*, but its total tax payments would be only (c + d). Total abatement costs, and total damages caused by the E* units of emissions, would still be the same.
How could regulators introduce an emission tax if they do not know the marginal damage function? We know that emissions are connected to ambient quality; in general, the lower the emissions the lower the ambient concentration of the pollutant. So one strategy might be to set a tax and then watch carefully to see what effect this had in terms of improving ambient quality levels. The regulator would have to wait long enough to give firms time to respond to the tax. If ambient quality did not improve as much as desired, increase the tax; if ambient quality improved more than was thought appropriate, lower the tax. This is a successive approximation process of finding the correct long-run emissions tax. It might be a good idea, however, for regulators to give polluters some advance warning of any rate changes. In responding to a tax, polluters might invest in a variety of pollution-control devices and practices, many of which would have relatively high upfront costs. This investment process could be substantially upset if, shortly afterward, the authorities shift to a substantially different tax rate.2 The setting of the tax could become politicized as a result. While it is better to find the correct tax rate when the policy is introduced, taxes at least allow for the possibility of iterating to the socially efficient tax rate. There is no way to do this with a standard. This issue will be examined in detail in Chapter 14.
2. Note, however, that firms and consumers deal daily with prices that can change considerably. A good example is the retail price of gasoline. Adjustments that regulators make to pollution tax rates would probably be far less volatile than prices in many markets.
Emission Taxes and Cost-Effectiveness