CAC vs EIs in Air Pollution Control; International Experience

    By Daanish Padha & Shubhankar Yadav

This article is a brief commentary on the common mechanisms employed for air pollution control. Part one of the article is a delineation of the properties of these mechanisms and the subsequent part two outlines the international experience of these regulatory instruments.  The instruments for Air pollution regulation are broadly classified based on two criteria:

(i) Whether they require firms to adhere to certain technological standards and conform to certain emission limits, or the ones which are not dictatorial and rely on creating incentives to induce firms to reduce emission levels. The former is known as CAC (command and control) instruments and the latter are called as EI (economic incentives) instruments. Common examples of the EI instruments are tradable permits, environmental taxes and CAC instruments examples primarily feature emission standards and legal laws requiring firms to choose from a restricted, usually incommodious set of abatement technologies.

(ii) Whether they require the regulator to monitor emissions or not. Those which require monitoring are called direct instruments and those which do not are called indirect instruments. Examples of the former include emission standards, tradable permits etc and those of the later are taxes on output of polluting firms.  The initial part of the paper features an explication of some instruments in each of the categories elucidated above.

EI instruments: 

An Indirect EI instrument: Environmental Taxes 

Indirect instruments, like environmental taxes can be more effective than their direct counterparts especially in the cases where monitoring costs are high. By placing appropriate taxes on the output produced or the inputs bought by the polluting firms, the regulatory body impacts their cost schedules adversely and the result is therefore truncated supply of the polluting firms corresponding to a particular price. The impact of the tax depends on the elasticity of demand for the good produced by the polluting firms. If the demand is inelastic, such a tax will not affect consumption much whilst having a considerable fiscal impact (In terms of the revenue generated from the tax). The elasticity of demand depends on the availability of substitutes for the good being taxed. Narrowly targeted taxes, thence are expected to have a larger environmental impact as long as the untargeted substitutes are relatively less hazardous. In conjunction to the aforementioned fact, it is also noteworthy that demand, for most goods, is typically more elastic in the long run as people take time to substitute over the goods either because they are information constrained or due to habit persistence. Therefore, the environmental tax is expected to add more to the government coffers in the short run with relatively less impact on pollution whereas over the long run the effects are likely to be the opposite.  Environmental taxes are also relatively easy to administer as compared to various direct instruments. The production of a firm or its input purchase can be easily monitored as compared to monitoring emission levels as in the case of direct instruments. Moreover, if the regulatory body wishes to reduce emission levels by a certain amount through emission fees, it needs information regarding the firms marginal cost of reducing pollution whereas to set the right tax the regulatory body needs information about the elasticity of demand for the good in question which can be easily obtained through bygone market surveys.  Environmental taxes can also have adverse distributional impacts especially when the taxed good is a necessity for the less well-off households. E.g. a tax on gasoline can potentially impact a large mass of the poorer population especially in the developing countries. The distributional impact can however be assuaged through appropriate employment of the tax revenue and curtailing some of the other regressive taxes.

Direct EI instruments: Emission fees and Marketable Permits 

Direct EI and CAC instruments are more demanding in terms of the informational requirements but can achieve more targeted results in juxtaposition to its indirect alternatives. Among the class of direct instruments, economists have long argued that EI instruments are better than CAC instruments in terms of both dynamic and static efficiency. The former refers to the impact of the instrument on ‘cost of emission reductions’ in the short run when the set of feasible production technologies are fixed while the latter refers to the impact on the per-unit cost of aggregate emissions reductions attributable to a regulatory instrument in the long run when innovation in abatement technology is possible.

Static Efficiency: Direct EI instruments leave each regulated plant free to choose the least expensive means of cutting pollution. By contrast, CAC technology standards dictate the firms to choose only certain technologies, which leave them with dampened flexibility to cut costs expediently.

Dynamic Efficiency: Direct EI instruments provide firms with incentives for pollution reducing innovation in technologies whereas CAC instruments dampen the long run incentives as well due to enforcement risks associated with using non-approved technologies. A firm would not like to spend much on research and technology since it is not sure whether it would be permitted to use any new technology it develops.

The EI direct instruments are more flexible and can easily accommodate changes as compared to the CAC instruments. In a CAC system, a regulator must formulate a prolific set of rules pertaining to various polluting entities, adherence to which has to be monitored at a high cost whereas in an EI system, firms retain control over their internal decisions and the regulator is only required to set the permitted emission level or fee. As a result, firms respond to changing economic environment and new technologies in an unprompted way in an EI system whereas the firms have a constricted set of responses in a CAC system.


CAC instruments: The discussion above suggests that CAC instruments are unambiguously worse when juxtaposed with their EI direct counterparts. However, in certain contexts EI instruments, although being relatively efficient and stable are demanding in terms of administrative costs as compared to CAC instruments. If we consider the case where there are certain firms with both high marginal damages and high marginal costs and certain firms with low marginal damages, an EI scheme would require fine tuning emission fees to various polluting sources or issuing permits tradable among similar firms. The administrative requirements for these modifications would render them impractical in many developing nations.  Also, in the cases where environmental regulations need to be constantly tailored to random shocks, CAC instruments are a better choice as compared to their EI counterparts because firms‟ responses to changes in emission fees are uncertain. Using historical data to determine the expected response of a firm to a given change in emission fee in this case might not be the correct thing to do because the firms‟ responses are contingent on the state of the world at that period of time.

What is Socially Efficient? 

The foregone discussion discusses the various pathways to achieve ‘socially efficient’ levels of pollution. In theory, socially efficient levels are characterized by the levels of the pollution which maximize the net benefit (i.e. Total Social Benefit – Social costs) of the society. How do we measure these benefits and costs? Many agents in the society, particularly those who make up the lower strata of the society would appropriate very little pertinence to the costs of pollution perhaps because they have very little information to make any rational choices for themselves or even if they had the requisite information, living near subsistence makes them willing to trade-off cheap commodities for increased pollution. E.g. Consider a society composed of many individuals unable to afford a full meal in a day. If a firm asks them for permissions to pollute in their neighbourhood and agrees to provide them with a meal 3 times a day, many would agree to the contract even if the pollution levels of the firm were very high. The scenario would completely change when we consider a well-off society. So quite clearly peoples‟ choices are characterized by their circumstances and if peoples‟ preferences are what comprise social costs, the optimal pollution levels would be contingent on how well off the society is. By this principle it would be okay to say that different levels of pollution are optimal for different societies. The statement seems morally displaced and poses an ethical conundrum. The divergence in what different societies perceive as optimum is often visible in public discussions in various countries. Any government’s primary objective is to please its electorate, so even under circumstances when the government has access to a larger set of information (as compared to the information set of an average individual), the government’s optimal actions would be characterized by the average individual’s optimal choice. The government would voluntarily neglect the extra information because factoring in that information doesn’t give them any additional benefit. Moreover, in democratic regimes government spending is biased towards programs which provide instant gratification to the public rather than those which have long-term benefits.

How feasible is regulation? 

All types of regulation require a public institution capable of formulating rules of conduct for the polluting parties, monitoring these entities and enforcing compliance. In many developing countries, institutional and financial constraints undermine these capabilities. There are commonly cited four key constraints. First is that Public in these countries generally favours economic development over environmental protection. The private sector is less concerned about pollution in developing countries as compared to the industrialized nations. As a result, it is often difficult to muster political support for regulations. The second reason is that environment regulatory institutions along with complementary judicial, legislative and other institutions are usually much weaker in developing countries. Third is that fiscal and technical resources for regulation are in short supply. Finally, the production is dominated by hard to monitor small-scale firms.

International Experience 

The second part of this article discusses the experience of various nations with Command and Control Policies (CAC) vs Economic Incentives (EIs) in the context of air pollution control.  Recapping briefly, CAC policies generally set standards of emissions or specify the technology of abatement that a firm must use. On the other hand, EIs use market based instruments and incentivize abatement. Typical examples are Discharge Fees or Marketable Permits. There are many advantages of using EIs to control pollution levels, such as Static and Dynamic Efficiency (i.e. allowing the least cost firm to abate the most and leaving room for innovations in abatement), Flexibility (firms retain complex decisions) and Revenue Use (can be earmarked for further environment related expenditure, or correct market failures elsewhere). But not all is hunky dory. Efficiency comes at the cost of implementation constraints, especially in developing countries. The Public sector there is generally inclined to focus on development over environmental concerns. This trade-off is only exacerbated in the presence of tight budgets which get stretched when establishing, monitoring and enforcing programs based on EIs. Small scale firms in these countries are hard to monitor and the judicial legislature is weak. As we shall see, EIs often fail to achieve desirable outcomes when they’re superimposed over an existing CAC skeleton and thus, such a „Two-Tiered system‟ cannibalizes its own efficiency by complicating the implementation.  Having given a brief idea of what the two competing mechanisms to tackle pollution are, let us move to some examples of how these programs were implemented in particular nations; how they became either success stories or costly failures; and more importantly, why.

The United States 

Let’s begin with arguably the biggest nation in the world, the United States. Two of the most famous Trading Permits programs implanted in the US are the ETP and the Sulphur Dioxide Program. The Emissions Trading Program was a move to EIs but was superimposed on the existing CAC program in place. Thus, new firms could set up but only if there were offsetting emissions reductions by the existing firms. It allowed for sophisticated transactions for permits such as old firms setting up new branches, called “Netting”; firms holding permits intertemporally rather than exhausting them at the outset, called “Banking”, and treating close enough pollution sources as just one entity and giving permits accordingly, called “Bubbles”. Despite being a move to EIs, this program had a limited environmental impact. There were major cost savings but low trading of permits. As mentioned before, a primary reason for this was the superimposition upon an existing CAC program. As a result, many permit trades were disallowed if they increased firm emissions even if total emissions declined (A firm couldn’t exceed its “commanded‟ limit even if it was buying permits from a separate firm that was shutting down). The market for trading was not well-developed and in the face of uncertainty about future regulation, there was low participation. The other major Permit Trading Program implemented in the US is the SO2 program. It was implemented to reduce SO2 emissions from electrical power plants, in the 1990s. Without going into the minute details of the program, we can broadly judge the difference between it and the ETP, and where this succeeded. Plants were grandfathered permits (allocated free of cost to begin with) according to historical levels of emissions and there was aggregate emissions cap. Firms were audited annually and severe penalty was imposed if found emitting in excess of the cap. Moreover, auctions were held too for tradeable permits allowing new firms to enter the industry without needing older firms to severely cut down emissions. This avoided the problem faced by the ETP where it disallowed numerous trades. The program led to major cost savings, as well as large environmental impacts. One of the lessons that can be learnt from this is the ease with which trades were allowed. Moreover, effective monitoring and severe punishment helped build market confidence among firms and reduced the incentive to cheat. Most importantly, the ETP was overlaid on an existing CAC program whereas the SO2 program substituted the existing program, and thus achieved efficiency without being hampered by the continuing inefficacy of a CAC scheme.  But herein lay the suffocating trade-off that a developing country would face. Having observed the success of the SO2 program, a developing country must face high administrative costs, in terms of monitoring trading, auditing, transaction costs, for both the regulator and the firm. Given a budget to run social programs with, it wouldn’t be politically palatable to overlook social programs elsewhere to implement such an expensive trading scheme, which relies on First Best monitoring (instruments like the CEM (Continuous Emissions Monitor) and flow monitors; all expensive transactions for firms involved).


Take a typical example of a developing country, China. It faces the stereotypical trade-off between Environment and Development. In their bid to boost economic growth, they had 2.5 times the WHO standard of particulate matter in their atmosphere by 1980 itself. They put in a fee program to control emissions (another form of EIs) but it failed miserably due to insufficient monitoring, weak fee payment standards, and even refunds of the fee collection back to the firms. For a cash strapped regulatory authority, it meant an invitation for bribery and corruption and widespread noncompliance. To summarize briefly how the system worked, the firms had to pay an emission fee on levels above a set standard. This fee, however, varies for every province and needs to be paid only for the pollutant whose gap from the standard is worst. Monitoring, shockingly, was voluntary and intermittent at best. There was a very low probability of getting caught and even lower probability of being made to pay the nominal fine, if any. In light of such blatant loopholes, it’s unlikely that the fee program would have had any major impacts on emissions. Studies however showed a fall in emissions per unit output; and a negative correlation between fees and emissions. But rather than taking the regulatory authorities‟ word about the success of the program, it is quite possible that the first observation was due to the CAC program already in place upon which this fee was imposed; and the second was a spurious correlation as the lenient fee system (at levels well below the MC) was hardly a good proxy for regulation stringency. Not to forget the refunding of fees to the firms and the decreasing tax obligation for firms that paid this nominal fee, all of which led to firms‟ incentives to report truthfully being distorted.

A grim conclusion? 

Let us wrap up by discussing what essentially stops a developing country from achieving the same goals as a developed country and what may we do to move closer to the first best outcome.  Success of Taxes (Sweden’s Sulphur and Carbon Taxes lead the way, but haven’t been discussed here), Emission Fees (Sweden’s NOx fee), and Tradeable Permits (SO2 Program in the US) depends crucially on administrative ease, political palatability and a generous budget to effectively monitor and enforce regulations. A developing country generally finds it tough to generate political acceptability of environment related goals as opposed to those which would boost economic growth. Permit transaction, enforcement, monitoring and administration are costly and often lead to substitution with second best monitoring technology, revenue leakages and lax enforcement. So, what can a developing country do in face of such overwhelming odds against environment protection? Studying the successes and failures of these schemes in developed countries would give us an idea of which EI would work best in general and how to improve upon idiosyncrasies of various EIs. Let’s look at these two in turn. One, Tradable Permit schemes are unlikely to work because of the high costs of administration and monitoring of trades that take place. The need to establish First Best Monitoring mechanisms to build market confidence in the scheme is infeasible for a developing country with a constrained budget. The choice remains between Taxes and Emission Fees. Taxes are less politically acceptable but raise revenue and overcome institutional constraints such as setting up a separate regulatory body. Emission Fees are good to begin with and can be more easily changed over time, and raise revenue as well, but require the setting up of a separate environmental regulatory authority to oversee the workings of the scheme and revenue collection. Thus, the choice for a developing country becomes that of one between political acceptability and a new institutional set up, which would obviously need expenditure. Two, for either of these schemes, some helpful lessons would be setting the fees or the tax at a level close to the MC of the firms, which can be truthfully revealed by various auction mechanisms; refunds of revenue collected should be to firms which abated the most, thus incentivizing abatement. The workings of the regulatory authority should not depend on the revenue collected as this incentivizes corruption. Instead, there could be bonuses for aggregate emissions reductions in an area. This may however cause distortion of bargaining power between the firm and the authority. There must be a minimum level of institutional capability and political will to enforce and monitor effectively, for either of these programs to work.  This paper thus shows, using examples of the US and China, that the theoretical superiority of EIs on efficiency grounds may not agree with their realistic application in developing countries whose cash strapped regulatory authorities and development oriented governments face a trade-off between protecting the environment and focusing on economic growth.  Unsurprisingly, the choice is nontrivial.


Salvatore Di Falco (2012); Atkinson and Tietenberg (1991); Kete (1994); Blackman and Harrington (2000); Hahn and Hester (1989); Blackman (2010); Wang and Wheeler (2005); Yang et al (1997); Xu, Hyde and Ji (2010).


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 )

Google+ photo

You are commenting using your Google+ 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 )

Connecting to %s