Skip to content

Hill – Mankiw 9th Edn Chapter 10 – Externalities

A commentary on Mankiw 9th Edn Chapter 10 – Externalities (Mankiw 9th edition)

Mankiw, N. G. (2021) Principles of microeconomics (9th ed.)
Principles of economics (9th ed.)
Mason, OH: South-Western Cengage Learning.

Rod Hill

University of New Brunswick, Saint John campus

Saint John, New Brunswick, Canada


Chapter 10 – Externalities

Here are some things to consider when reading this chapter.


  1. How important are externalities in practice?

This amounts to asking: (a) how frequently and to what extent do production and consumption activities affect third parties? And (b) to what extent are externalities adequately addressed by public or private action?

(i) Are externalities the rule or the exception?

In a 1972 lecture, Joan Robinson commented on “the notorious problem of pollution”. She remarked that the “distinction that [Arthur] Pigou made between private costs and social costs was presented by him as an exception to the benevolent rule of laissez faire. A moment’s thought shows that the exception is the rule and the rule is the exception. In what industry, in what line of business are the true social costs of the activity registered in its accounts?” She accused economists of managing “to hush it up” by presenting externalities as “a few minor points… that could easily be put right” (1972, p.7).

A good case can be made that important externalities are pervasive and can have effects beyond the economic system itself. For example, no standard text, including Mankiw’s, refers to the concept of meta-externalities. These are the side effects “of the economic system, as a whole, on the social and ecological environment in which it is embedded — and on whose health the continuing health and vitality of the economy depends”, as Neva Goodwin of Tufts University explains. As examples, she gives “the toxins and non-biodegradable wastes building up in huge quantities throughout the Earth’s ecosystems. No single corporation has major responsibilities for these; they come from the whole system” (2010, pp. 29-30).

Perhaps the largest meta-externality results from all the economic activities that contribute to climate disruption, with its resulting negative effects on future social and economic conditions. In an economic system still largely dependent on the use of fossil fuels, the energy required to produce virtually all goods and services will be mispriced if they do not incorporate these external costs. Firms’ marginal private costs will not reflect social marginal costs.

Meta-externalities can also be positive. Goodwin gives the example of the increase in literacy rates that accompanied industrialization and economic development.

There is ample evidence that people care about how their consumption and income compare to those of other people. As a result, changes in one individual’s consumption or income affects many other people’s feelings of well-being as well as the choices they make. Commentaries on later chapters will return to this type of externality, neglected in standard texts.

Except for this chapter and the next, externalities largely disappear from the book. The default assumption is that marginal private costs equal marginal social costs. This could be justified if externalities were (a) relatively rare and unimportant (which does not seem to be the case); or, (b) where significant external benefits and costs were appropriately addressed by public or private action to bring external costs/benefits to socially optimal levels.

In this case, corrective actions would bring marginal social costs into equality with marginal social benefits only if there were no other sources of market failure. For example, in the imperfectly competitive models of markets introduced later in the book, there is no reason to suppose that marginal social costs will equal marginal social benefits even in the absence of externalities.

(ii) What values should be assigned to marginal external costs and benefits?

It’s easy to draw a diagram illustrating the idea of an optimal tax/subsidy in a supply and demand framework, given the implicit assumption of perfect information. But it’s quite another matter to determine what that tax/subsidy should be in practice, even if the supply and demand framework were appropriate.

In the case of many pollutants, scientific understanding of their effects remains far from complete. For example, the consequences of polluting the planet with microplastics and (even smaller) nanoplastics is a topic of ongoing research. At the time of writing, microplastics have been found deep in people’s lungs and in their blood, perhaps not surprising because they are in the air we breathe, the food we eat and the water we drink. The health consequences are unknown, although microplastics have been found to damage human cells in laboratory settings. In such cases, with the costs unclear but potentially large, the problem is one of dealing with risk, or perhaps uncertainty would be a better term because there is not enough information to attach probabilities to future outcomes.

Similarly, understanding of the harmful consequences of air pollution from the burning of fossil fuels continues to advance. This leads to ever increasing estimates of the millions of premature deaths annually that are attributable to it. As a result, the World Health Organization revised its guidelines in 2021, cutting its recommended maximum amount of exposure to small particulate matter (i.e. less than 2.5 µm in size) in half (from an annual average of 10 mcg/m³ to 5). These particles have been found in human organs, including the heart. One example of the effects of this was seen from the reduction in air pollution during the Covid 19 pandemic because of decreased commuting. There was a measurable decrease in heart attacks in the United States; a pollution reduction of 10 mcg/m³ was associated with 6 percent fewer heart attacks.

Compared with air pollution, understanding of the effects of chemical pollutants is still in its infancy.

More than 140,000 new chemicals and pesticides have been synthesised since 1950. Of these materials, the 5,000 that are produced in greatest volume have become widely dispersed in the environment and are responsible for nearly universal human exposure. Fewer than half of these … have undergone any testing for safety or toxicity, and rigorous pre-market evaluation of new chemicals has become mandatory in only the past decade and in only a few high-income countries (Landrigan et al. 2018, p. 462).

What the appropriate Pigouvian tax should be on greenhouse gas emissions has been a source of ongoing controversy. Even if the consequences faced by current and future generations were fully known, which is far from the case, how should dollar values be attached to them? After all, the negative consequences include illness and premature deaths, coastal flooding and displacement of millions, reduced agricultural production in many areas and species extinctions, to mention just a few. And how much should future benefits and costs be discounted? Clearly an ethical judgement is required in weighing how much to invest now to reduce emissions to change the well-being of people in the future, including those who do not yet exist.

To summarize: for many important externalities, the information does not exist to allow policymakers to enact the optimal Pigouvian taxes recommended in the textbooks.

(iii) Are important externalities adequately addressed by private and/or public action?

Each of Mankiw’s examples is accompanied by possible solutions using taxes, subsidies, and various kinds of regulation, again implicitly assuming perfect information. In his summary of the chapter, he writes: “Governments pursue various policies to remedy the inefficiencies caused by externalities.” He adds: “Those affected by externalities can sometimes solve the problem privately”, although “[i]n many cases, however, reaching a bargain among the many interested parties is difficult”. Presumably government could act in such situations. Mankiw offers no evidence that public and private action adequately addresses externalities in practice.

The only indications he gives that public action may be inadequate are in the examples of the gasoline tax and the carbon tax. In the US, gasoline taxes are only 1/6 of his reported estimate of the optimal tax (p. 195). His only explanation: the gasoline tax “has never been politically popular”, but this fails to explain why much higher gasoline taxes have been enacted in other countries. Might the political power of the oil industry play a role?

His discussion of a carbon tax is contained in a newspaper editorial in a box (p. 198), which cites Mankiw himself as a co-author of a proposal for such a tax. The editorial says that this carbon tax proposal has the support of “oil giants ExxonMobil, Shell and BP.” Not mentioned in the editorial is the part of the proposal that reads: “Much of the EPA’s [Environmental Protection Agency’s] regulatory authority over carbon dioxide emissions would be phased out, including an outright repeal of the Clean Power Plan. Robust carbon taxes would also make possible an end to federal and state tort liability for emitters” (Baker et al. 2017, p. 3; my emphasis).

Immunity from liability would be attractive for oil companies who are facing lawsuits brought by state governments. Among other things, the companies are accused of systematically misleading the public about the role of their products in causing climate change. This illustrates another way in which those causing negative externalities may, to some extent, take external costs into account in their decisions. If the law assigns property rights to those harmed, they may have to pay damages in the future if lawsuits are successful.

As well, oil companies’ public expressions of support for carbon taxes can ‘greenwash’ their public image while they work behind the scenes to make sure that carbon taxes don’t happen. (For example, see this report on ExxonMobil). The influence of corporate power on public policy is systematically overlooked in economics principles textbooks, including Mankiw’s, although it is an obvious feature of the political landscape, particularly in the United States.

The omission is even more mysterious because business firms in the textbooks are assumed to have one goal: maximizing profits. For some firms, it may pay to find ways to lower costs by externalizing them onto others. Aside from political lobbying, firms can attempt to influence (or ‘capture’) government regulatory agencies to permit the production of hazardous products and to allow pollution above levels that would be socially efficient. Indeed, such activities take place in plain sight and have been the subject of many investigative reports and books.

In short, the difficulties of putting a dollar value on externalities, as outlined in the previous section, and the realities of corporate power make it unlikely that the textbook story of optimal Pigouvian taxes describes what really happens. The result: pervasive inefficiency even in absence of other sources of market failure.

Here is one important example. The failure to adequately price fossil fuels – with the resulting excessive costs of air pollution and climate disruption – is estimated every year by the International Monetary Fund (IMF). As of 2020, the IMF estimated that fossil fuel subsidies were $5.9 trillion, 6.8% of world GDP. These consist of explicit subsidies, keeping the selling price below cost, and the far larger implicit subsidies by failing to impose the IMF’s estimate of adequate Pigouvian taxes to address the external costs of air pollution, climate disruption, road congestion and vehicle accidents.


  1. The Coase Theorem

In considering private solutions to externalities, Mankiw writes: “A famous result, called the Coase theorem after economist Ronald Coase, suggests that it can be very effective in some circumstances. According to the Coase theorem, if private parties can bargain over the allocation of resources at no cost, then the private market will always solve the problem of externalities and allocate resources efficiently” (p. 201, my emphasis). He notes that this does not happen if there are significant transactions costs as is so often the case. For example, when large numbers of people are involved, negotiations are costly. Monitoring compliance with contracts and enforcing them is also costly. Despite its apparent irrelevance in dealing with important externalities, there are repeated references to the Coase theorem, including in the conclusion and in the ‘Chapter in a Nutshell’ summary.

Coase himself referred to it as “the notorious Coase Theorem” because he felt that his basic message was often misunderstood (1994, p. 10). He believed that economics should incorporate positive transactions costs in its analysis. In that case, the rights assigned to people by the legal system “will have a profound effect on the working of the economic system and may in certain respects be said to control it” (1994, p. 11).

In contrast, standard economic theory assumes zero transactions costs, although that is rarely emphasized. But if they really were zero, Pigouvian taxes and subsidies would be unnecessary: people would have already costlessly negotiated an efficient outcome that could be costlessly enforced. In effect, the textbook presentation contains an internal contradiction: Pigou’s prescription would be needed in a world of positive transactions costs, yet it’s presented in a zero transactions cost model.

The example Mankiw gives in a box entitled “The Coase Theorem in Action” is, ironically, a good example of how the Coase theorem does not apply (p. 203). In the case of airline seats, people have the right to recline their seat; people behind them have no veto power that would be enforced by the airline. The significant transactions costs are (incompletely) described in one sentence, but it’s hardly surprising that the author of the article has never had anyone offer to pay him not to recline his seat. After all, who would enforce the agreement? Perhaps more importantly, there are social situations in which an offer of money is acceptable and others in which it is inappropriate or offensive. In the case of reclining airline seats, a polite request and an explanation would be appropriate. (The same point applies to the detailed example of a neighbour with a barking dog in Section 10.3b.)


  1. Is environmental quality and regulation just a reflection of people’s preferences?

Mankiw writes: “Like all normal goods, it [environmental quality] has a positive income elasticity: Rich countries can afford a cleaner environment than poor ones and, therefore, usually have more rigorous environmental protection” (p. 200). This view attributes differences in environmental regulation to people’s preferences as average income changes, with governments responding to these preferences.

A political economy view would explain the extent of environmental protection provided by government regulations as the outcome of a struggle for influence between those who benefit from pollution and those who are harmed by it. The beneficiaries of pollution are the owners of firms who would otherwise have to incur costs to reduce it and consumers of the product if suboptimal environmental protection results in lower prices. Who is harmed by the pollution varies on a case-by-case basis; it could range from people in the vicinity of a dirty industry to everyone in the world. (Where an individual’s interests lie depends on that person’s net benefit from pollution, which could be positive or negative. This includes any weight that people give to the effects of pollution on others, not just on themselves.)

The demand for stricter pollution regulation could increase with per capita incomes (as in the textbook story) or it could reflect the increased political influence of the groups harmed by pollution. Mariano Torras and James Boyce (1998) hypothesized that a society with a more equitable distribution of power would have stricter pollution controls and lower levels of air and water pollution, all else equal. To test this idea, they compared measures of air and water quality in samples of cities in many different countries. Per capita incomes varied across these countries, but so did other things that acted as rough measures of the distribution of power: levels of income inequality, literacy, political rights and civil liberties. Their statistical results were generally consistent with their hypothesis, with literacy and political and civil rights being strong predictors of pollution levels in low-income countries (1998, p. 158).

It’s important to note that environmental quality often differs significantly in different parts of a country. A recent report to the United Nations Human Rights Council described heavily polluted “sacrifice zones” in both poor and rich countries. Unsurprisingly, in all countries it is poor and marginalized communities, who have the least political power, that suffer the greatest harm. For example:

In the United States, cancer rates are far higher than the national average in predominantly Black communities … located in Louisiana’s “Cancer Alley”, which is home to more than 150 refineries and petrochemical plants, including the world’s largest producer of Styrofoam. Large polluting industrial facilities in the United States are disproportionately located in communities with the highest percentages of persons of African descent, the lowest household incomes and the highest proportion of residents who did not graduate from high school. A leading scholar wrote that, “[e]nabled by state zoning, a wave of chemical plants dropped on African American communities like a bomb”. Cancer Alley contains 7 of the 10 United States census tracts with the highest risk of cancer from air pollution. In 2020, air concentrations of cancer-causing chloroprene in St. John the Baptist Parish were 8,000 times higher than the acceptable level established by the United States Environmental Protection Agency. (Boyd 2022, pp 10-11)

Finally, the textbook claim that environmental quality is the result of informed choice also has implications for international trade. This view implies that the most polluting industries will (and should) be located in the lowest income countries where pollution standards will be lowest. High income countries with high pollution standards can “consume the products of polluting activities without having to host them, and the low-standard regions or countries would have more economic activity, in return for pollution to which they are, relatively speaking, indifferent”, as Martin Wolf writes, leaving “the world… unambiguously better off” (2004,  p. 191). These pollution externalities, and the illness and premature death that results for people in low-income countries, act as an export subsidy for which people in importing countries should be grateful, according to Mankiw’s view about export subsidies noted in the previous Commentary. Instead of his suggestion of thank you cards, perhaps flowers for the additional funerals would be more appropriate…



Baker, James, Martin Feldstein, Ted Halstead, N. Gregory Mankiw, Henry Paulson, George Schultz, Thomas Stevenson, and Rob Walton (2017) The Conservative Case for Carbon Dividends, Climate Leadership Council, available here.

Boyd, David R. (2022) The right to a clean, healthy and sustainable environment: non-toxic environment. Report of the Special Rapporteur on the issue of human rights obligations relating to the enjoyment of a safe, clean, healthy and sustainable environment, UN General Assembly Human Rights Council, January 12, A/HRC/49/53.

Coase, Ronald (1994) Essays on economics and economists, University of Chicago Press.

Goodwin, Neva (2010) ‘Good business’, Planet, February, 28–31, available here (PDF).

Landrigan, P., et al. (2018) ‘The Lancet Commission on pollution and health’, The Lancet, 391, 464–512.

Robinson, Joan (1972) ‘The second crisis of economic theory’, American Economic Review, 62(2): 1–10.

Torras, Mariano and James K. Boyce (1998) ‘Income, inequality, and pollution: a reassessment of the environmental Kuznets Curve’, Ecological Economics, 25, 147-60.

Wolf, Martin (2004) Why Globalization Works,: Yale University Press.

Respond to this commentary

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Please note that your email address will not be published.