environmental economics, eco 565/365

Session 1, Week June 28 – July 2, 2010

1

Anglo-American University School of Business Administration

Environmental Economics, ECO 565/365 Instructor: Jana Krajcova

Office hours: Thursdays, 5:30-6 p.m.

Email: [email protected]

Web: http://home.cerge-ei.cz/richmanova/Teaching.html

What is your background?

environmental?

economic (statistical/econometric analysis)?

environmental economics?

experimental economics?

What do you expect from this course?

OUTLINE OF THE COURSE:

The aim of this course is to introduce students to some basic economic principles and theories

explaining environmental issues and problems today and to explore existing policies at the

national, international, and world level. Students will learn about concepts such as externalities,

the tragedy of the commons, enforcement as a public good, interventionist solutions to the

externality problem such as taxes and marketable pollution permits, as well as non-

interventionist solutions to the externality problem such as the Coasian solution and self-

regulation. Students will also review the debate over the environmental Kuznets curve. Because

experimental evidence complements theoretic insights, field data and simulating models nicely,

we will review some research articles that draw on the experimental methodology. We will also

review environmental policies and the field data of the Czech Republic.

Learning Objectives: The students should understand basic economic principles behind

environmental policies and should be able to evaluate, from the economic point of view, various

approaches to environmental protection. They should also get familiar with environmental

protection and the state of the environment in the Czech Republic and the role of European

Union in nowadays regulation.

Teaching Methodology: Students will be expected to read the course materials and we will meet

once a week to discuss. Every week, students will receive worksheets which should help them

focus on the most important topics. The students will also be required to submit a term paper at

the end of the session. The term paper should demonstrate their understanding of the covered

material.

mailto:[email protected]

http://home.cerge-ei.cz/richmanova/Teaching.html


2

Course Schedule:

Week 1

introductory meeting (Monday, June 28)

first discussion session (Thursday, July 1)

o Introduction (history/outline)

o Market failures: externalities, tragedy of the commons, enforcement as public

good (river experiment)

o Interventionist solutions to the Externality problem – Pigouvian taxes and

standards and charges

first worksheet assigned, due in next discussion session

Week 2

second discussion session (Thursday, July 8)

o Interventionist solutions to the Externality problem – Marketable pollution

permits

o Non-Interventionist solutions to the Externality problem – The Coasian solution

and Self-regulation

second worksheet assigned, due in next discussion session

Week 3

third discussion session (Thursday, July 15)

o Environmental Policy in the Czech Republic – History and current issues, EU

context

discussion about possible topics for term paper

third worksheet assigned, due in next discussion session

Week 4

fourth discussion session (Thursday, July 22)

o Environmental Policy in the EU – History and Current problems

o Environmental Policy in the world context – History and Current problems

term paper consultations

fourth worksheet assigned, due on July 27

Week 5

- term paper (and fourth worksheets) due on July 27

GRADING POLICY:

Worksheets: 10% each (total 40%)

Term paper: 40%

Class participation and activity: 20%


3

Introduction

Why do we care about the environment?

a) life supporting function (location and the basic conditions for the existence of life)

b) natural resources (inputs for consumption and production)

c) amenity values (natural beauty)

Natural resources:

a) flow resources (solar radiation, wind or water energy – the current use does not affect

the future availability)

b) stock resources (the current use affects future availability)

o renewable resources (forests, stock of fish, etc…)

o non-renewable (fossil fuels, mineral ores)

Nature of environmental problems

a) nature degradation due human activity (deforestation, pollution)

b) conflicting usage of the natural resource (e.g. amenity vs. production)

c) distribution of usage over time (this or future generation?)

d) distribution among agents (too many fishermen)

Current issues in Europe and the CR

a) water and air pollution, greenhouse effect (industry, transportation)

b) soil pollution (industrial fertilizers)

c) energy intensity

d) noise (transportation air/road/railway)

e) waste management

f) decrease in biological diversity and ecological stability (due agricultural production

and fragmentation of the landscape due transportation)

Instruments of environmental protection

a) regulations

b) economic and financial (standards and charges, marketable pollution permits, taxes,

fines, tax reliefs and subsidies, property rights)

c) voluntary programs (environmental labeling, self-regulation), environmental

education and public awareness

ENVIRONMENTAL ECONOMICS AND USE OF EXPERIMETAL METHODS

From Wikipedia, the free encyclopedia

Environmental economics is a subfield of economics concerned with environmental issues.

Quoting from the National Bureau of Economic Research Environmental Economics program:

Environmental Economics undertakes theoretical or empirical studies of the economic effects of

national or local environmental policies around the world. Particular issues include the costs and

http://en.wikipedia.org/wiki/Economics

http://en.wikipedia.org/wiki/National_Bureau_of_Economic_Research


4

benefits of alternative environmental policies to deal with air pollution, water quality, toxic

substances, solid waste, and global warming.

Use of Experimental Methods

Based on

(G&G) Greenstone, M., Gayer, T., (2007), Quasi-Experimental and Experimental Approaches to Environmental Economics, RFF Discussion Paper 07-22. (L&L) Levitt, S., D., List, J., A. (2009), Field experiments in economics: The past, the present, and the future, European Economic Review 53, 1-18

Wikipedia:

In scientific inquiry, an experiment (Latin: ex- periri, "to try out") is a method of investigating particular

types of research questions or solving specific problems. An experiment is a cornerstone in the empirical

approach to acquiring deeper knowledge about the world and is used in both natural sciences as well as

in social sciences. An experiment is defined, in science, as a method of investigating less known fields,

solving practical problems and supporting or negating theoretical assumptions.

Brief History (L&L)

1) Natural science experiments: Among the first experimenters

o Galileo Galilei in the 17th century: The falling bodies experiment (dropping objects from the leaning tower of Pisa in order to prove that all objects fall at the same rate, whatever their mass to disprove Aristotle's assertion that heavier bodies fall faster than light ones.),

o Sir Isaac Newton showed that the white light is a mixture of colored lights and shattered down another Aristotle’s theory that the white light is equal to purity,

o Pasteur rejected the theory of spontaneous generation with an experiment – he showed that microorganisms grow in boiled nutrient broth when exposed to the air, but not when exposed to the carefully filtered air.

2) “The dawn of field experimentation” in 1920s - 1930s: Experiments were used to help to answer important economic questions. None of those studies involved human subjects. Via experimentation with agricultural plots Neyman and Fisher conceptualized RANDOMIZATION as a mean to achieve identification.

3) Large-scale social experiments in mid 20th century conducted by government agencies that

involved individuals to evaluate employment programs, electricity pricing schemes, housing allowances … Experiments were used to test new programs later also reforms to existing programs – important influence on policy making.

Social Experiment: Ferber and Hirsch (1982,p.7) ‘‘ a publicly funded study that incorporates a rigorous statistical design and whose experimental aspects are applied over a period of time to one or more segments of a human population, with the aim of evaluating the aggregate economic and social effects of the experimental treatments.’’


5

Greenberg and Shroder (2004) define a social experiment as having at least the following four features: (i) random assignment, (ii) policy intervention, (iii) follow-up data collection, and (iv) evaluation.

4) Growing popularity and expansion to diverse areas of interest over the past decade

economists have increasingly used the field experiments and controlled small(er)-scale experiments to explore economic phenomena.

Use of experimental methods in economics:

education and training, public finance, industrial organization, labor and public economics, consumer behavior, game theory, development economics, environmental economics (see more in L&L and G&G)

Benefits of employing Experimental Methods

- A new drug is tested to make sure that it has the expected effect and at the same time that it is not outweighed by possible side-effects – to minimize potential cost on public health

- The effect of planned policy change can be tested at relatively low cost (compared to allocation of much larger resources to an inefficient program; e.g. training program for the unemployed, new pricing scheme for electricity,…)

- Explaining or predicting non-experimental outcomes (e.g. Barr and Serneels 2004: correlation of wage outcomes of employees with their behavior in a trust game experiment) – again, relevant policy/strategy implications at relatively low cost

- Testing theoretical predictions at relatively low cost (economic theory) - Help to generate the data which are difficult to be obtained from “the field” - Estimation of a cost that the firm which produces pollution should internalize so that the

(socially) more efficient outcome can be achieved -> ENVIRONMENTAL ECONOMICS Externalities -> correction? -> Environmental Economics (G&G)

air or water pollution as a byproduct of the production of marketable good

imposes health costs on inhabitants and/or costs on the down-the-river company not internalized by the firm which is responsible for producing the pollution

government intervention might help to maximize net (social) benefits/welfare – require reliable estimates of the costs and benefits => ENVIRONMENTAL ECONOMICS

EE addresses the inefficiencies resulting from production externalities –> experimental and quasi-experimental methods

hinge upon proper design, implementation, appropriate approach to the data analysis

Possible Difficulties when employing Experimental Methods “The aim of the researcher is to estimate a causal effect of some action (a new government program,

change in price,…), i.e. how outcomes differ when the action is taken vs. when it is not.”


6

“The fundamental difficulty that arises is that either the action is taken or it is not—we never directly observe what would have happened in an alternative universe where a different action is taken. Thus, the construction of a control group becomes critical. Although we cannot observe what your outcome would have been had you not been treated, we can, for instance, observe outcomes for other similar individuals who were not treated.” Illustrative Examples: 1) observational study analyzing the use of estrogen replacement therapy (ERT) to maintain the

menopausal symptoms and their potential dangers such as higher incidence of heart disease 2) testing impact of a new regulation which restricts pollution that can be produced by a company;

what is the impact on health

want to test a treatment effect (like e.g. receiving drug vs. placebo, exposure to high vs. low pollution)

outcome may or may not respond to the treatment(=drug/high pollution) effect (heart disease/ other health problems) - > every individual has two potential outcomes but only one can be in fact observed

to isolate the effect of treatment – all other factors need to be held constant (ideally, we would want to observe the outcome for the same individual in both treatments – with and without drug/ exposed to high and to low pollution – not possible -> Fundamental problem of Causal Inference)

can observe the health outcome for treated individuals (with ERT/high pollution = TREATMENT

GROUP) and for not-treated (no ERT/low pollution = CONTROL GROUP) -> average difference in health outcome treated vs. untreated

PROBLEM -> SELECTION BIAS: our individuals might have some “special characteristics” that affect both, selection to treatment AND the outcome of the treatment (women with healthier lifestyles more likely to participate/ people with lower income living in more polluted areas) -> the effect of special characteristics can be, in some situations validly assumed zero, in other situations it can be controlled for -> the researchers need to be aware of it to be able to make valid inferences!

SELECTION BIAS: The researcher also compares the outcomes between the treatment and the control group. The problem is when the selection for the treatment is not up to the researcher – individuals are exposed (to treatment) by nature, politics, accident … -> NON-RANDOM ASSIGNMENT => possible source of SELECTION BIAS Still can make VALID inferences under the assumption that the assignment to the treatment is

not related to any determinant of the outcome o Neg. Exmample 1: Observational studies of ERT concluded no direct causality between

ERT and heart disease. Problem: Maybe women with healthier life style were more likely to participate, take ERT and therefore per se less likely to have heart problems (reasons to believe that -> self-selection -> special characteristics that might affect the results of an observational study) Solution: A randomized study -> Concluded that ERT substantially contributes to heart disease


7

o Neg. Example 2: similarly, in the pollution example if the housing prices are significantly

lower in the affect area and therefore it is colonized by poorer people with less healthy lifestyles, less resources to spend on healthcare

o Pos. Example: If e.g. the government decides to enroll the unemployed in a special

training program and selects randomly (or by some other rule, completely unrelated to their profession, abilities, education… anything that might affect their probability of success on the job market after the training) a half of the currently unemployed to receive the training (only a half for e.g. budgetary reasons, randomly to avoid e.g. accusations of discrimination) – their success after the training (if received) is not correlated with their selection for the group even though the assignment to treatment is not in the control of the researcher who will analyze the data

Possible threats to the validity of the experimental methods and generalization of the results:

- Internal validity: treatment status can be related to the outcome for reasons other than treatment (selection bias)

- External validity: Heterogeneity of the treatment effect – possible problems with generalization of the result and application to other context. The estimated treatment effect might be different…

o … from the overall population (perhaps the subjects are more sensitive to pollution) o … across geographic settings o … across institutional settings o … across years

Example: If a government implements a program that improves the quality of air only in some regions, some individuals might change their location. If we estimate a short-run treatment effect before the re-location that would likely be different from the estimated long-run treatment after the re-location (large-scale experiment).

- Construct validity: the researcher must understand the experiment and the possible effects very well. In the pollution example, if the implemented program reduces the concentration of a specific air pollutant but at the same time has some other effect on health, positive or negative, the researcher is not aware of then he cannot separate the two (or more) effects by looking at the impact on health. Still can analyze the effect of the program on overall improvement in health. The researcher just has to be very careful in the specification of his research question and choose an appropriate methodology to get valid answer to that question.

Other possible biases (=problems)

RANDOMIZATION BIAS (L&L)

- some individuals might be reluctant to subject themselves to a random assignment =>

experimental sample might differ from the population of interest because of randomization. For example, in medical trials, it is typically more difficult to persuade patients to participate in


8

randomized than non-randomized studies; in social experiments difficult (this could be a problem in both large and small-scale field experiments)

- participants in small-scale experiments might not be representative of individuals that would participate in a large-scale study; Heckman (1992), Heckman and Smith (1995), Manski (1995) (lab experiments, not natural field exp, when subjects are not aware of their participation)

SUBSTITUTION BIAS (L&L)

- subjects in the control group might seek available substitutes for treatment (large-scale experiments, NOT lab or framed experiments)

ATTRITION BIAS (L&L)

- within-subject design => some social experiments can be going on for several years during which subjects are surveyed – subjects might become tired of keeping detailed records, some might move,... (large-scale social experiments, not lab or framed experiments which are typically short-term)

GENERALIZATION OF THE RESULTS (L&L)

- even with proper estimation of the treatment effect, the generalizations of the results to other domains might prove difficult (lab experiment, framed experiments, NOT natural field experiments)

o lab experiments: student subjects, relatively small-scale

o subjects are aware that they are monitored and recorded

o psychological effect of being in the experiment, expecting the experimenter to expect specific result (see e.g. List 2006, or Benz and Meter 2008 for the difference in behavior when subjects are and are not aware they are participating in an experiment)

PUBLICATION BIAS (G&G)

- researchers more likely to submit, and journals are more likely to accept, for publication the studies that confirm the “expected results” (e.g. pollution is detrimental to health) – solution in leading medical journals, the researchers have to register their clinical trials, their study before knowing the results

REGULATORY BIAS (G&G)

- regulators put more weight on results that find a negative impact on health (to protect the public they require stronger evidence to support the “ no risk to health” than the “risk to health” results -> overestimated risk than reduces the chances to achieve the most efficient outcomes (risk-aversion is reflected in the willingness to pay and thereby effect the policy benefit calculations -> over-regulation -> E.g. pollutant A may be more risky than pollutant B, but if studies over-estimate the riskiness of B at the end the policy maker might not choose the most efficient allocation of resources to reduce the pollution.


9

MARKET FAILURES: EXTERNALITIES, TRAGEDY OF THE COMMONS, ENFORCEMENT AS PUBLIC GOOD

Externality:

(some bits and pieces here are from Wikipedia)

In economics, an externality or spillover of an economic transaction is an impact on a party that is not directly involved in the transaction. The basic idea is that an externality exists when a person makes a choice that affects other people that are not accounted for in the market price and thus the prices do not reflect the full costs or benefits in production or consumption of a product or service. An advantageous impact is called an external benefit or positive externality, while a detrimental impact is called an external cost or negative externality. Producers and consumers in a market may either not bear all of the costs or not reap all of the benefits of the economic activity.

o Positive externalities - an action that imposes a positive side effect on a third party o Negative externalities - an action that imposes a negative side effect on a third party; many

negative externalities are related to the environmental consequences of production and use. Can you think of couple of examples on

a) positive externalities b) negative externalities

Examples of positive externalities:

o A beekeeper keeps the bees for their honey. A side effect or externality associated with his activity is the pollination of surrounding crops by the bees. The value generated by the pollination might be even more important than the value of the harvested honey.

o An individual planting an attractive garden in front of his or her house may provide benefits to others living in the area, and even financial benefits in the form of increased property values for all property owners.

o Home ownership creates a positive externality in that homeowners are more likely than renters to become actively involved in the local community.

o Education creates a positive externality because more educated people are less likely to engage in violent crime, which makes everyone in the community, even people who are not well educated, better off.

Examples of Negative Externalities

o Transportation: drivers imposing congestion cost on other drivers o Industrial Production: producing (as a by-product) greenhouse gas emissions from burning oil,

gas, and coal -> climate change imposing cost on whole society o Water pollution by industries that adds poisons to the water, which harm plants, animals, and

humans. o Industrial farm animal production – farm that were maybe more efficient as regards the

production costs/revenues but they contributed to the increase in the pool of antibiotic-resistant bacteria because of the overuse of antibiotics + air quality problems + the contamination of rivers, streams, and coastal waters with concentrated animal waste + animal welfare problems, mainly as a result of the extremely close quarters in which the animals are housed.

o Fishing: harvesting by one fishing company in the ocean depletes the stock of available fish for the other companies and overfishing may be the result. This is an example of a common property resource, sometimes referred to as the Tragedy of the commons.


10

o Consumption of alcohol in some cases leads to drinking and driving accidents which injure or kill pedestrians and other drivers.

In these situations the marginal social benefit of consumption is less than the marginal private benefit of consumption [more detail to follow]. This leads to the good or service being over-consumed relative to the social optimum -> Market failure. Without any corrective measure, the good or service will be under-priced and the negative externalities will not be taken into account (recall interventionist vs. non-interventionist solutions). Market failure In economics, a market failure occurs when there is an inefficient allocation of goods and services in a market. That is, there exists another outcome where market participants' overall gains from the new outcome outweigh their losses (even if some participants lose under the new arrangement). Market failures can be viewed as scenarios where individuals' pursuit of pure self-interest leads to results that are not efficient – that can be improved upon from the societal point-of-view.

Pareto Efficiency

Pareto efficiency, or Pareto optimality, is a concept in economics with applications in all areas of the discipline as well as engineering and other social sciences. The term is named after Vilfredo Pareto, an Italian economist who used the concept in his studies of economic efficiency and income distribution. Informally, Pareto efficient situations are those in which it is impossible to make one person better off without necessarily making someone else worse off.

Given a set of alternative allocations of goods or outcomes for a set of individuals, a change from one allocation to another that can make at least one individual better off without making any other individual worse off is called a "Pareto improvement". An allocation is defined as "Pareto efficient" or "Pareto optimal" when no further Pareto improvements can be made.

Allocative efficiency also referred to as Pareto Efficient Allocation

a) consumption side: resources cannot be re-allocated to make one consumer better off (in terms of utility) without making another worse off; or

b) production side: allocation of production inputs (capital and labor) is Pareto-efficient if it is not possible to re-allocate these inputs and produce more of at least one good in the economy without decreasing the amount of some other good that is produced

A simple illustrative example:

Imagine that Robinson Crusoe has invented a machine that can make two mangoes out of one coconut. Conversely, the machine can make one coconut out of two mangoes. Assume that Crusoe‘s utility is U(c,m)=c*m (and thus marginal utilities are U

‘m=c and U

‘c=m. Suppose Crusoe

has, initially, four mangoes and four coconuts. Is that Pareto-optimal allocation? If not, what would Crusoe has to do to get a P-O allocation?

he‘ll end up with Robinson‘s final utility (c*m)

doing nothing 4 coconuts + 4 mangoes 16

converting 1 coconut into 2 mangoes 3 coconuts + 6 mangoes 18

converting 2 coconuts into 4 mangoes 2 coconuts + 8 mangoes 16

converting 2 mangoes into 1 coconut 5 coconuts + 2 mangoes 10


11

converting 4 mangoes into 2 coconuts 6 coconuts + 0 mangoes 0

It is easy to see in which case his utility function will be the highest… Robinson would obviously end up with 3 coconuts and 6 mangoes in the P-E allocation.

Note that more formally, you can solve the problem using the concepts of Marginal rate of transformation… those interested can find the solution in Schotter‘s textbook, Chapter 15, Solved Problem 15.1 (p. 581 in the 3

rd edition)

But, this is just a simplest case, with just one individual, Robinson Crusoe. What if we take into account also his ―Man Friday‖, whose utility over coconuts and mangoes might be different? Or even a larger economy with number of consumers and producers…. with potential externalities…

→ Social efficiency – total social (also external) costs are accounted for

Causes of market failures:

i) externalities ii) public goods or common goods (―the tragedy of the common‖) iii) market power (imperfect/no competition)

i) and ii) are interesting from the point of view of environmental economics [more details to follow], iii) is not so important for us now…

Public good

(From Wikipedia, the free encyclopedia)

In economics, a public good is a good that is non-rivalrous and non-excludable. This means,

respectively, that consumption of the good by one individual does not reduce availability of the good

for consumption by others; and that no one can be effectively excluded from using the good. In the

real world, there may be no such thing as an absolutely non-rivaled and non-excludable good; but

economists think that some goods approximate the concept closely enough for the analysis to be

economically useful.

Non-rivalness and non-excludability may cause problems for the production of such goods. Specifically,

some economists have argued that they may lead to instances of market failure, where uncoordinated

markets driven by parties working in their own self interest are unable to provide these goods in

desired quantities. These issues are known as public goods problems, and there is a good deal of debate

and literature on how to measure their significance to an economy, and to identify the best remedies.

These debates can become important to political arguments about the role of markets in the economy.

More technically, public goods problems are related to the broader issue of externalities.

Examples of public goods (can you think of any?):

a) light houses (cannot exclude shops from using it)

b) defense and law enforcement

c) fireworks

http://en.wikipedia.org/wiki/Economics

http://en.wikipedia.org/wiki/Good_%28economics%29

http://en.wikipedia.org/wiki/Rivalry_%28economics%29

http://en.wikipedia.org/wiki/Excludability

http://en.wikipedia.org/wiki/Market_failure

http://en.wikipedia.org/wiki/Markets

http://en.wikipedia.org/wiki/Externalities


12

d) streetlights

e) roads

f) informational goods (software development, authorship, invention)

g) environmental goods (clean air, clean water….environmental protection in general)

Some goods are ―mixed‖ in a sense that they have the properties of both, private and public goods

a) excludable but non-rival (like cable TV)

b) non-excludable but rival (like public park… when too many visitors it becomes less enjoyable)

Free rider problem

Public goods provide a very important example of market failure, in which market-like behavior of

individual gain-seeking does not produce efficient results. The production of public goods results in

positive externalities which are not remunerated. If private organizations don't reap all the benefits of a

public good which they have produced, their incentives to produce it voluntarily might be insufficient.

Consumers can take advantage of public goods without contributing sufficiently to their creation. This is

called the free rider problem, or occasionally, the "easy rider problem" (because consumer's contributions

will be small but non-zero).

The free rider problem depends on a conception of the human being as homo economicus: purely rational

and also purely selfish—extremely individualistic, considering only those benefits and costs that directly

affect him or her. Public goods give such a person an incentive to be a free rider.

For example, consider national defense, a standard example of a pure public good. Suppose homo

economicus thinks about exerting some extra effort to defend the nation. The benefits to the individual of

this effort would be very low, since the benefits would be distributed among all of the millions of other

people in the country. There is also a very high possibility that he or she could get injured or killed during

the course of his or her military service.

On the other hand, the free rider knows that he or she cannot be excluded from the benefits of national

defense, regardless of whether he or she contributes to it. There is also no way that these benefits can be

split up and distributed as individual parcels to people. The free rider would not voluntarily exert any extra

effort, unless there is some inherent pleasure or material reward for doing so (for example, money paid by

the government, as with an all-volunteer army or mercenaries).

To establish a national defense system, the government needs to determine how much money to spend

on it – small vs. huge military complex? Need to know the cost plus the maximum willingness of each

member of the society to pay these costs -> how to find out? Well, the government could try to ask…

Suppose you know that everyone is reporting their true willingness to pay. You also know that there are

so many people in the society that your response, however small, will not affect the level of national

defense. In that case, you have no incentive to report truthfully -> your ―economically rational‖ response

would be to say that you are not willing to pay for national defense. But if everyone would do so…..

Questions for an economist? (Schotter, Chapter 18)

i) what is the optimal amount of public good to produce , and what conditions must be satisfied

at such optimum?

ii) How can economy achieve that optimum?


13

iii) Will free markets be able to achieve that optimum, or must the government help the economy

to coordinate its activities?

Solved problem 18.1. (Schotter, Chapter 18, p. 655 in the 3rd

edition)

Ten years ago, most commercial software application were copy protected, which made them rather

difficult to copy. Nowadays software packages are hardly ever copy protected; that is, you can very easily

copy software applications that others paid for and use them at home for free. On the other hand, most

software manufacturers offer generous ―upgrade policies,‖ whereby legal owners of software can get

more recent versions of that software at a special discount, and ―support services,‖ whereby legal owners

get technical assistance from the manufacturer‘s technical experts. These facilities are, of course, not

available to users of illegal copies. Explain this business practice in terms of public goods and

externalities. What would be the consequence, if no one were to buy the legal copy? How would you say

―support services‖ and ―upgrade services‖ compare as an inducement to pay for one‘s software?

Solutions:

a) Lindahl ―free market solution‖ – relies on everyone‘s truthfully revealing their preferences for

public good; then the government serves as a ―coordinator‖ (no intervention) – sets everyone‘s‘

share on the total cost if the good is provided →people will face prices and the market will take

care of the rest: people will maximize their utility and state their demand for the public (as well as

private) good. In the equilibrium, prices of private goods and shares on cost of public good are set

such that no one wishes to change his/her demand for private and for public goods + supply of

private good equals the demand + everyone consumes the same amount of public good (due to

non-excludability).

Problem: incentives not to be truthful in revealing one‘s preferences.

Proposed solution:

i) a demand-revealing mechanism (imagine a dark street and three equally costly plans

to install streetlights (one very bright streetlight or combinations of less bright streetlights)

– then ask inhabitants, how much they are willing to pay for each of the proposed plans

and implement the one that maximizes the total willingness to pay) → still there is no

guarantee that collected contributions will cover the total cost of implementing the

streetlight plan.

ii) an auction election mechanism: people submit their bids (bidding the money one is

willing to pay and the quantity demanded); then if public good is produced, everyone

pays the difference between the cost and sum of the bids made by other people.

Everyone has a right to refuse his or her cost share. If all people agree to pay their costs

share the demanded quantity is produced. If no agreement is reached, public good is not

produced – the experimental evidence suggests that truth-telling does not seem to be the

general rule (Smith, 1977).

b) Coase argument: with no transaction cost, most conflicts could be resolved by private bargaining

(more on that later)

c) Government provision (then financed by tax revenues) - it might be difficult to ensure the

government has an incentive to provide the optimum amount even if it were possible for the

government to determine precisely what amount would be optimum

d) A government may subsidize production of a public good in the private sector; unlike

government provision, subsidies may result in some form of a competitive market. Principal-agent

problems can arise between the citizens and the government or between the government and the

subsidized producers.


14

e) an exclusion mechanism (club goods) is another solution, which has evolved for information

goods, is to introduce exclusion mechanisms which turn public goods into club goods. One well-

known example is copyright and patent laws. These laws, which in the 20th century came to be

called intellectual property laws, attempt to remove the natural non-excludability by prohibiting

reproduction of the good. Although they can address the free rider problem, the downside of

these laws is that they imply private monopoly power and thus are not Pareto-optimal.

f) support public mindedness by tradition and social norms (a non-market solution)

Tragedy of the commons (Hardin + Wiki)

Tragedy of the commons

(From Wikipedia, the free encyclopedia)

"The Tragedy of the Commons" was an influential article written by Garrett Hardin and first published in the journal Science in 1968. The article describes a dilemma in which multiple individuals acting independently and solely and rationally consulting their own self-interest will ultimately destroy a shared limited resource even when it is clear that it is not in anyone's long term interest for this to happen. More usually, the phrase does not refer to the article per se, but to the dilemma itself, typically in application to some circumstance to which it is thought to apply. Many, perhaps most, who use it are not aware of, nor have read, Hardin's essay, but are looking at conceptually parallel situations.

Central to Hardin's article is an example, a hypothetical and simplified situation from medieval land tenure in Europe, of herders sharing a common parcel of land (the commons), on which they are each entitled to let their cows graze. In Hardin's example, it is in each herder's interest to put the next (and succeeding) cows he acquires onto the land, even if the carrying capacity of the commons is exceeded and it is damaged for all as a result. The herder receives all of the benefits from an additional cow, while the damage to the commons is shared by the entire group. If all herders make this individually rational economic decision, the commons will be destroyed to the detriment of all.

- the problem arises when property rights are not well defined - private property then provides a mechanism to avoid externalities – he one who owns, cares

about the property and controls its use + can exclude others from overusing it (see the discussion in Hardin as well)

- private property is not the only available mechanism – regulations work as well (with legal system to enforce them)

Examples (can you think of any?):

a) over-herding cows (see Hardin) b) overfishing (each fisherman has a negligible impact on the total fish stock… but too many

fisherman might result in serious depletion) c) automobile pollution – each automobile lower the air quality and it is not likely that the free market

would result in the optimal amount of pollution → emission standards for automobiles → Illustration:

- 1963 Clean Air Act and its amendments set automobile emission standards for the manufacturers of vehicles in the US and Lawrence White examined the costs and benefits of this program

- cost of emission control equipment is estimated at $600 per car, extra maintenance cost at $180 per car, the cost of reduced gasoline mileage and the necessity of unleaded gasoline at about $670 per car → total coast at $1450 (in 1981 dollars)

http://en.wikipedia.org/wiki/Garrett_Hardin

http://en.wikipedia.org/wiki/Science_%28journal%29

http://en.wikipedia.org/wiki/Commons


15

- White identifies following problems: everyone who buys a car must pay extra $1450, whether they live in high

pollution area or not most of the responsibility falls on the manufacturer, only little on the user → car

owners have little incentives to keep the pollution control equipment in working order unless they are inspected

no incentive to economize driving – people who drive 2000 miles in less polluted areas pay exactly the same amount of money as people who drive 500,000 miles in heavily polluted areas → it would make sense to encourage people to drive less (at least in heavily polluted areas)

- Alternative solution that White offers: effluent fees annual inspection of all vehicles estimating the car‘s likely emissions during the

past year different communities (areas) could levy different fees → people would face the

true cost of generating pollution, which would encourage them to generate ―socially optimal amount of pollution‖ (or, well, at least closer to it)

Why should it work better? Any idea? the system would encourage the owners to search for low-cost ways of reducing

their emissions, including changing their driving habits and type of vehicle (more eco)

Some experimental evidence:

Cotten, Ferraro, Vossler, Can public goods experiments inform policy? Interpreting results in the presence of confused subjects (Cherry, Chapter 10)

Public policy – how to induce individuals to contribute to public goods when it may be in their private interests to free-ride off the contributions of others VCM (= voluntary contributions mechanism) is the cornerstone of experimental investigations on the private provision of public goods

Standard experimental investigation places individuals in a context-free setting where the public good, which is non-rival and non-excludable in consumption, is simply money

Specifically, endowed ―tokens‖ have to be divided between a private and a public account (contributions to the public account yield a cash return to all group members)

Typically, parameterized/designed so that each player has a dominant strategy of not contributing (to the public account) but the social optimum is realized when everyone contributes their entire endowment

In one-shot (single-round) VCM experiments, subjects contribute – contrary to the theoretical prediction – about 40% - 60 %

In finitely-repeated VCM experiments, subjects contribute about the same initially but contributions then decline towards zero (but rarely ever zero)

―Thus, there seem to be motives for contributing that outweigh the incentive to free ride‖ (CFV 194)

Possible ―motives‖: ―pure altruism‖, ―warm-glow‖ (also called, ―impure altruism‖), ―conditional cooperation‖, ―confusion‖

―Confusion‖ describes individuals‘ failure to identify (in the laboratory set-up) the dominant strategy of no contribution

Could it be that these findings are the result of confusion that ―confounds‖ the interpretation of behavior in public good experiments? (p. 195)

One new experiment, two old ones

Using the ―virtual-player‖ method to sort out pro-social motives such as altruism

http://home.cerge-ei.cz/ortmann/UpcesCourse/Cotten%20Can%20public%20goods%20experiments%20inform%20policy.pdf

http://home.cerge-ei.cz/ortmann/UpcesCourse/Cotten%20Can%20public%20goods%20experiments%20inform%20policy.pdf


16

Virtual players (that are preprogrammed to execute decisions that are made by human players in otherwise identical treatments)

Split-sample design (where each participant is randomly assigned to play with humans or (human condition) with virtual players (computer condition)

A procedure that ensures that human participants understand how the non-human virtual players behave.

Random assignment of subjects to the human condition or the computer condition – important assumption here that subjects are drawn from the same population.

―human twins‖ in multiple-round public goods games where the group contributions are announced after each round

→ to neutralize other-regarding and strategic motives (contributions in all-human minus contributions in virtual ≈ confusion contributions)

GHL experiment :

individual dominant strategy is to contribute nothing

full endowment contribution maximizes group earnings

varying internal and external rates of return (m‘s) and group sizes

individuals exhibiting pure altruism should increase their contributions when me increases

considerable contributions uncorrelated with me and n point at warm glow

- Finding:

pure altruism and confusion are important motives whereas warm glow is not

―The level of confusion in all experiments is both substantial and troubling.‖ (p. 196) ―Assuming that other-regarding preferences and confusion are present in all-human, but only confusion is in virtual treatment, this suggests that an alarming 75% of all-human contributions stem from confusion.‖ (p. 202)

―The experiments provide evidence that confusion is a confounding factor in investigations that discriminate among motives for public contributions,… ― (p. 196)

More evidence:

- Andreoni (1995) first to argue that (parts of) what looks like kindness in VCM experiments is really confusion. Andreoni finds that other-regarding behavior (kindness, altruism) and confusion are ―equally important‖

- Houser & Kurzban (2oo2) did the same thing but they used a different setup: o a ―human condition‖ (the standard VCM game) o a ―computer condition‖ (the standard VCM game, played by one human player and three

non-human (or, ―virtual‖) players. o Each round, the aggregate computer contribution to the public good is three-quarters of

the average contribution observed for that round in the human condition. o Basic result: Confusion accounts for about 54 percent of contributions to all public

good contributions. - Ferraro et al. (2003) and Ferraro & Vossler (2005), with designs similar to Houser & Kurzban find

that 54 and 52 percent contributions come from confused subjects. - Palfrey & Prisbey (1997) find a similar result in their own experiment (not using virtual players)

and estimate with their model that ―well over half‖ of the contributions in the classic VCM experiments by Isaac et al. (Public Choice 1984) are attributable to error.

- Goeree et al. (2002) find in their own experiment (not using virtual players) both a positive and significant effect on coefficients that correspond to (pure) altruism and decision error (confusion);


17

- Fischbacher & Gaechter (manuscript 2004) find in their own experiment (not using virtual players) that ―at most 17.5% ― are contributed by confused subjects; they also argue that none of their subjects exhibits altruism or warm-glow (no subject stated they would contribute if other group members would not). In Fischbacher & Gaechter‘s view, all non-confused subjects are conditional cooperators‖

- Summary: every study that looks for confusion finds that it plays a significant role in observed contributions. (at least half of contribution plus confusion does not go away over the course of many repeated rounds)

- Some graphs:

Solutions:

Increase monetary rewards in VCM experiments! (inadequate monetary rewards having been identified as potential cause of contributions provided out of confusion)

Make sure instructions are understandable! (poorly prepared instructions having been identified as possible source of confusion)

Make sure, more generally, that subjects manage to identify the dominant strategy! (the inability of subjects to decipher the dominant strategy having been identified as a possible source of confusion)

―Our results call into question the standard, ―context-free‖ instructions used in public good games.‖ (p. 208) ―Subjects may seem like zero intelligence agents when they are placed in the unfamiliar and abstract context of the experiment, even if they function quite adequately in familiar settings.‖ (Loewenstein, 1999)


18

INTERVENTIONIST SOLUTIONS TO THE EXTERNALITY PROBLEM

A. Theoretical background

Schotter, Microeconomics, A Modern Approach (Second edition) Section 17.3

- problem of externality

- interventionists vs. free market advocates

Interventionist solutions:

1. Pigouvian Taxes

2. Standards and Charges

3. Marketable Pollution Permits

1. PIGOUVIAN TAXES

the society produces paper and clean water

paper mill dumps waste into the river and thereby increases the cost of cleaning it => externality

(as this cost is external to the mill, it is borne by the WT plant) => not taken into account when

making production decision

Say the mill is producing

o 10 tons of paper (1 ton=2,000 pounds, 10 tons… 20,000 pounds)

o with a (private) MC (of labor and capital) of $0.005/pound.

o In a competitive market p=MC

Water treatment plant‘s MC

o when the mill is idle is $.50/1,000 gallons;

o when the mill is active, additional cost of $.05/1,000 gallons for each ton of paper

produced

o given the current mill‘s production, the total MC is $.50 +10(tons)*$.05=$1 per 1,000

gallons

In a competitive industry the price of water will be $1 per 1,000 gallons. Assume at such price

1 mil. gallons of water is demanded => Society spends $1,000 on water

Can we expect the society to produce a PARETO OPTIMAL amounts of water and paper?


19

Intuitively, we might expect the answer to be NO.

The paper mill is imposing an additional cost on the water treatment plant, but there is no mechanism to make the mill accountable for this cost, so it seems unlikely that the outcome for society will be Pareto-optimal. Indeed it is not …

Three conditions must be fulfilled for a perfectly competitive economy to produce Pareto-optimal outcomes: d) MRS1 = MRS2=… the marginal rate of substitution (the ratio of the MU (paper) to the MU

(water), which in equilibrium has to be equal to the price ratio of the price of paper and the price of water) of paper for water has to be the same for all consumers. Where MRSw for

p=MUp/MUw=pp/pw=$0.005/$0.001=5/1 => all consumers maximizing their utility will set their MRSs such that this condition is met

e) MRTSp=MRTSw - the marginal rates of technical substitution of paper mill and water treatment plant ought to be the same

f) MRSw for p= MRTw for p= MCp/MCw - the marginal rate of substitution of water for paper must be equal to the marginal rate of transformation of water for paper, which in equilibrium is supposed to be equal to the ratio of the marginal cost of producing paper to the marginal cost of producing clean water

But … the (private) marginal costs of paper and water are not what their (social) marginal costs, and the marginal utilities are … PROBLEM

g) (MRTw for p would have to be 5/1= we must give up 5 gallons of water to obtain 1 more pound of paper.)

h) take away $1 from production of water (MCw=$0.001 => 1,000 fewer gallons are produced) => 999,000 gallons are produced

i) give that $1 to the mill (MCp=$0.005 => 200 more pounds of paper can be produced) => so far so good BUT

j) extra 200 pounds => 10.1 tons are produced => increase in MCw (each ton brings extra $0.05 in MC) => 0.1 tons brings about extra $0.005 => MCw = $1.005 =p => with $999 available on water, the society will only purchase 999/1.005= about 994,000 gallons of water. Thus the society has to give up almost 6,000 gallons, not 1,000 to obtain 200 more pounds of paper => the MRTw for p= MRTw for p= 6,000/200=30/1 rather than 5/1.

k) At (competitive) production levels of 10 tons of paper and 1 million gallons of water, the

society would be producing too much paper and not enough water => MARKET FAILURE


20

point A – the level of production of paper resulting from a competitive market -> Not Pareto

Optimal. Why?

assume mill would reduce its production by 200 pounds (0.1 ton). Given the market price

that would mean a loss of 200x$.005=$1 in revenues

cost of producing clean water is now reduced by (200p/2000p)=1/10x$.05=$.005 per

1,000gal. => 1 mil. gallons would be produced at a cost of $995 instead of $1,000 -> $5

saved for the treatment = Pareto Improvement

the cost savings of the WT plant are sufficient to allow it to produce more water and to

compensate the mill for its lost revenues

the ―pollution‖ cost is external to the mill, so it does not affect the production decision

from the social point of view => Social Marginal cost MC‘ of the paper production (=production

cost + pollution cost)

point A is not optimal for society -- BC (social MC)>BA (social marginal benefit) => point

D is the social optimum

Pigou – TAX the mill by the amount of the marginal externality (EF) in order to internalize the externality

and directly affect the mill‘s paper production => point D

PROBLEM – To set the tax, the government needs to know the exact amount of the externality

(the cost). The afflicted party, however,

o might not be able to estimate accurately

o might have incentives to exaggerate (both, the mill and the WTP)

2. STANDARDS AND CHARGES

The government sets the standard – the amount of externality considered acceptable and then charges in

order to induce the agents to reduce the externality to the acceptable level.


21

Single firm

the government conducts a study to determine how much pollution is acceptable

charge on each gallon of waste to induce the mill to reduce the pollution to the acceptable level

when the mill‘s cost is MC => it will produce at point A

charge=> MC‘=> it will produce at point B

Ideally, with qc the production of waste is at the STANDARD

Two or more firms

2 firms: mill A 70 gallons of waste a day, mill B 30 gallons. STANDRD= 50 gal.

an across-the-board 50% cut not the most efficient (different MCs for waste reduction)


22

o A would have to reduce by 35, B by 15 gal. Say A‘s cost of reducing by additional 1 gal.

is $5, B‘s is $8 => if A‘s total abatement is 36 gal. and B‘s is 14 gal. the total abatement is

same but the society could save $8-$5=$3.

o Firms with lower cost should reduce by more and firms with higher cost by less!

o Figure 17.4 – once the environmental charge (per unit of pollution) is set, each firm will

reduce by the corresponding amount. alow+ahigh=atotal (MCA=MC

B=charge; STANDARD is

induced)

PROBLEM – even more difficult to administer, need to know the exact damage to society to set

the STANDARD + the cost of abatement for each firm (guess and verify, don‘t want the firms to

reduce neither too much nor too little)

3. MARKETABLE POLLUTION PERMITS

For each unit of produced waste the firm pays not only the cost of labor and capital, but also a

permit that will allow to produce that unit. A firm with higher MC of abatement is willing to pay

more for the permit that the firm with lower MC of abatement (up to its cost of abatement for the

corresponding number of units)

The government first finds an acceptable level of pollution and then offers for sale the

corresponding number of permits

The firm can only pollute with the permit. The government directly controls the amount of pollution

without having to know any specific about the firms‘ marginal costs of abatement or about the

social marginal cost of pollution

we will talk more about this next week.

B. Experimental Evidence

Plott, Externalities and Corrective Policies in Experimental Markets, also Schotter, Section 17.4

A series of experiments to evaluate how the interventionist solutions work

the subjects buy and sell units of a fictitious good using a double oral auction (In such a double

oral auction any potential buyer (or, seller) can make a verbal bid (offer) to buy a unit of the good

at a specified price. Any seller (buyer) can accept a bid. If a bid is accepted a binding contract is

closed for a single unit at the specified price. Any ties are resolved randomly).

each buyer is paid a redemption value for every purchased unit according to a predetermined

redemption schedule induced demand curve

each seller must pay a premium for each unit he sells according to a predetermined cost

schedule private marginal cost curve

every completed transaction imposes an additional cost in all subsequent transactions ; the cost

increases with the number of units sold externality => social marginal cost curve.


23

Note that after 6 units sold, the marginal externality cost is $.24, after 43 transactions, it is $.42

Pareto optimal solution -- point A (13 units at price $2.69)

without intervention -> theory predicts the competitive outcome ―as with no externality‖ -> point B

(24 units at price $2.44)

Charles Plott:

―Do markets with externalities behave in accordance with the law of supply and

demand?‖

―How do pollution tax, pollution standard and pollution licenses compare as methods for

correcting the externality?‖

4 treatments, 2 sessions for each, 6 buyers and 6 sellers in each market


24

individual demands and costs are assumed to be unknown, only the optimum level of pollution and

marginal social cost at the (social) optimum are known for the license and the tax policy

1. Market with externality (no policy, 5 periods in each)

- benchmark, to see the market solution

2. Pigouvian Tax policy (6+7 periods)

- the amount of marginal social cost is calculated at the optimum quantity Q0, and is

imposed on sellers as a per unit tax. Tax revenues are then redistributed back.

3. Standards policy (9+7 periods)

- the ABCD area is the ‗optimum‘ value of pollution damage => STANDARD limits the

amount of admissible pollution such that imposed damage is ABCD (so here, number of

trades is limited such that the total environmental damage equals ABCD, 13 units in fact

– thus only the first 13 trades)

4. Permits policy (10+12 periods)

- only Q0 permits exist and only licensed unites can be produced, 13 licenses

- EQ: price of license = BC; market price of the good = Pe‘; quantity = Q0; licenses should

be held by the low cost sellers


25

1. Market with externality

at the top of each graph, see the mean price and the number of units sold in each period

in both sessions

o the volume sold tended to move toward the competitive eq. of 24 units

o price close to the competitive equilibrium level of $2.44

the market failed => the theoretical prediction confirmed: subjects ignored the externality

competitive rather than Pareto optimal outcome


26

2. Pigouvian tax policy

at the top of the graph, see the mean price and the number of units sold in each period

cost schedule is increased by a tax equal to the amount of the marginal externality

the imposition of tax simply becomes a change in supply

TAX effective in pushing the volume down to the Pareto optimal level of 13, and price up to eq.

level of 2.69

3. Permits policy

at the top of the graph, see the mean price and the number of units sold in each period


27

secondary market for permits is created: in order to sell 1 unit of the good on the primary market,

a seller first had to purchase a permit on the secondary market

PERMITS effective in pushing the volume down to the Pareto optimal level of 13, and price up to

eq. level of 2.69

ALSO the price per permit converged to the equilibrium level of $.36 (look at the picture in the

original paper, p.110)

Session 8 convergence more obvious, session 7 series more stable, close to eq. levels

more efficient than TAXES in terms of the surplus captured by subjects (efficiency – maximizing

the total earnings of subjects)

4. Standards policy

at the top of each graph, see the mean price and the number of units sold in each period

the least efficient way of intervention

because the total number of permits was limited to 13, the subjects rushed into concluding the

deals => dispersed prices, means close to the levels with no intervention

CONCLUSION:

the LEAST efficient is the unregulated market

the MOST efficient is the permits policy


28

ENVIRONMENTAL LABELING AND INCOMPLETE CONSUMER INFORMATION IN LABORATORY MARKETS Product certification

From Wikipedia, the free encyclopedia

Product certification or product qualification is the process of certifying that a certain product has passed

performance and quality assurance tests or qualification requirements stipulated in regulations such as a building

code and nationally accredited test standards, or that it complies with a set of regulations governing quality and

minimum performance requirements.

Certification of ENVIRONMENTAL QUALITY….

Cason, Gangadharan, Environmental labelling and incomplete consumer information in laboratory

markets

survey evidence exists suggesting that the consumers care for the environment and are willing to

pay a higher price for the more environment friendly products

they study a market with incomplete information – prior to purchase the consumer is unaware of

the product‘s (environment-related) quality (moral hazard problem)

no signaling, no reputational concerns => market failure

various treatments to remedy the market failure: cheap talk signals, seller reputation, (costly)

certification (―eco-label‖)

http://en.wikipedia.org/wiki/Quality_assurance

http://en.wikipedia.org/wiki/Building_code

http://en.wikipedia.org/wiki/Building_code


29

Experimental Design and Implementation

21 sessions, 20 periods each (except 1st); 5 sellers+6 buyers randomly assigned

neutral wording of instructions

sellers can sell up to 2 units of REGULAR or 2 units of SUPER grade in each period

SUPERs more expensive to produce than REGULARs (120 exp. francs vs. EF 20, common

knowledge)

buyers‘ resale value of SUPERs > of REGULARs which is common knowledge but buyers‘

marginal values are private info [=> induced demand curve, see the fig. below]

o SUPERs: 1st unit EF 330, 2

nd unit EF 300, 3

rd unit EF 270

o REGULARs: 1st unit EF 180, 2

nd unit EF 165, 3

rd unit EF 150

buyers prefer to buy SUPERs unless they are by EF 120-150 more expensive

than REGULARs

o Each SUPER costs the seller 120 EF and each REGULAR 20 EF

all buyers and sellers have identical cost and value schedules

Efficient equilibrium all (10) SUPERS are produced and traded at EF 300

Inefficient equilibrium all (10) REGULARS are produced and traded at EF 165

TREATMENTS

BASELINE

o the sellers are asked to indicate privately the number of units they want to sell, the offer

price per unit and the grade of the units at the beginning of each period.

o the price offers by the sellers are posted on the board in a random order to hide the seller

identity (no reputations)

o buyers are randomly selected to take turns accepting the offers

o after all the buyers have an opportunity to purchase, or all the units are sold, the grades

of the units are written next to each price offer (in all treatments, grade info of all sellers is

revealed publicly at the end of each period)


30

o in all treatments sellers must commit to a specific quality level privately to the

experimenter at the start of the period.

REPUTATIONS ONLY

o the same trading procedure as in BASELINE, except that here the first seller‘s price offer

is always written in the first row on the board, the second seller‘s offer in the second row,

etc.

o allows the buyers to track the sales record of each seller and identify if a particular seller

has a history of selling REGULARs or SUPERs

THE CHEAP TALK SIGNALLING o prices written on the board in the specific order to identify sellers‘ history

o whether unregulated claims could by themselves help in increasing the number of

SUPERs sold

o sellers have the following 2 options:

1) indicate no grade information to buyers (thus, only the price and the number of

units offered for sale appear on the board)

2) indicate a grade to be shown on the board, although this need not correspond to

the actual grade offered

o the 2nd

option represents the unregulated environmental quality claims -> so-called cheap

talk (claims made by producers that have not/cannot be verified by a third party)

CERTIFICATION

o prices written on the board in the specific order to identify sellers‘ history

o would sellers choose the option of certifying their product at extra cost of EF 30?

o sellers have the following 3 options:

1) indicate no grade information to buyers (thus, only the price and the number of

units offered for sale appear on the board)

2) indicate a grade to be shown on the board, although this need not correspond to

the actual grade offered

3) sellers can pay 30 francs to certify that the product they are offering is a SUPER

(thus, the buyers would be sure they are buying a SUPER)

o the certification is indicated by a ‗‗star‘‘ next to the price offer, and it corresponds to third-

party verified environmental labeling schemes

o the certification is costly (in practice real resources are needed to test products)

o the cost of certification is set such that it can lead to the efficient equilibrium


31

TESTED MODELS (HYPOTHESES)

Lemons Model

When sellers face buyers who cannot distinguish between REGULARs and SUPERs, they will only offer

REGULARs. Buyers observe only REGULARs delivered and so they will behave as if they expect only

REGULARs. Hence in equilibrium, only REGULARs will be delivered and the price prevailing in the

market will be PR (= EF 165) This equilibrium is particularly likely when sellers cannot establish

reputations, as in the BASELINE treatment.

Reputation Model

In the presence of some imperfect information, even in finite period games sellers may establish

reputations for delivering SUPERs in sequential equilibrium. According to this model, for some early

range of periods some sellers will deliver SUPERs at a price of PS (EF 300). A buyer who observes a

seller delivering a REGULAR will update her beliefs and expect that seller to always deliver REGULARs

in the future. Therefore, in later periods more REGULARs will be delivered at PR.

Signaling Models with Unverifiable Signals

When it is not possible to verify product claims by sellers, then no cost differential exists between adding

the ‗‗SUPER‘‘ signal to SUPERs and to REGULARs. In a ‗‗babbling‘‘ cheap talk equilibrium, no seller

adheres to her signals, and buyers do not believe that the signals convey any information. Consequently,

market outcomes would be unchanged by the introduction of signaling. Signaling could, however, assist

sellers in establishing the reputations.

Signaling Models with Verifiable Signals

When the signaled product claims are verifiable, sellers would find it profitable to deliver certified SUPERs

since the certification cost is less than the marginal profit from delivering SUPERs rather than REGULARs

at their respective equilibrium prices. Buyers know that if the product is certified, they are guaranteed to


32

receive a SUPER and are thus willing to pay the higher equilibrium price PS. Hence when certification is

available, outcomes correspond to the full information equilibrium.

RESULTS – WHAT WOULD YOU EXPECT???

Market performance is measured by frequency of SUPERs delivered to consumers and by the overall

market efficiency.

1) Allowing seller reputations increases the rate at which SUPERs are traded, but Cheap Talk does not

affect the rate at which Supers are traded compared to the treatment with Reputations Only.

Certification is sufficient to increase the rate that SUPERs are traded.

in the BASELINE, SUPERs account for about 12% of total sales (all periods), whereas in the REPs

ONLY it is about 40% (significant)

in the CHEAP TALK, about 1/3 are SUPERs (not signif. different from REPs ONLY)

in the CERTIF., more than 2/3 are SUPERs (signif. more than in REPs ONLY)

the results also confirmed by econometric analysis

- subjects accumulate evidence from offering SUPERs and REGULARs and update their

beliefs about their expected profits

- in REPs ONLY, # of SUPERs rises over time

- initial periods of BASELINE are not sign. different than in REPs ONLY, in later periods the #

is lower in BASELINE

- early periods of CHEAP TALK – more SUPERs than in REPs ONLY, the difference

disappears

- CERTIF. not different in early periods, later on, signif. more SUPERs

- even though SUPERs tend to be delivered in later periods, in the final periods, most units

offered are REGULARs except in the CERTIF. (end-game effect)

2) Allowing seller reputations marginally increases efficiency, but neither Cheap Talk nor Certification

significantly improves efficiency compared to the treatment with Reputations Only.


33

efficiency measured as the fraction of the maximum possible gains from trade actually realized by

subjects

efficiency goes up from 65% to near 80% when sellers can establish reputation

the difference between BASELINE and REPs ONLY significant, between REPs only and CHEAP

TALK or CERTIFICATION not significant

efficiency tends to rise over time

3) Non-certified Super signals are frequently false.

about 22% in the CHEAP TALK are false. Non-ceritfied SUPER signals are more rare in the

CERTIFICATION, but still 33% of them is false.

REGULAR signals are much less common in both treatments, they are almost always truthful.

4) Given the opportunity in the Certification treatment, sellers frequently certify their units as Super.

in some sessions nearly all units are certified, e.g. average number of sellers who choose to certify is

4.75 out of 5 in all periods, rises to 5 in both the last 10 and 5 periods; in other sessions the

certification rate is 2-4 and tends to increase in time.

many sellers use certification to establish reputation and then, later, use cheap talk to obtain higher

prices; buyers, however, often refuse to buy uncertified units for SUPER prices (except of 1 session)

5) a) Reputations modestly impact transactions prices in certain conditions; b) signals have an impact

on transaction prices in the Cheap Talk treatment; and c) certification is necessary for sellers to sell at

substantially higher prices in the Certification treatment.

6) Sellers who reveal themselves as ‘‘cheaters’’ by delivering Regular units at ‘‘Super prices’’ can

frequently regain a positive reputation quickly – often in the next period. This seems to be due in part

to the lack of an explicit outside option for buyers in most of our sessions.

‗‗immediate reputation recovery‘‘ -- when a seller is able to sell an uncertified unit at a high price one

period after they sold a Regular at a high price


34

this puzzling high rate of reputation recovery could be due to the fact that buyers‘ only source of profit

is from making purchases in the market.

to test this conjecture, they conducted three additional sessions where buyers could choose not to

purchase from any seller and still earn 10 francs, focusing on the REPs ONLY as here the reputation

recovery rate was particularly high (70%)

the buyers opt for the no purchase option in exactly one-quarter of the periods

in general, the overall performance in this new treatment is similar to the five sessions with REPs

ONLY and no outside option.

Importantly, the reputation recovery rate declines substantially in these new sessions with an outside

option, to 33%..

The puzzling high reputation recovery rate in the REPs ONLY without the outside option treatment

could hence be explained in part by the inability of the buyers to exit the market profitably.

CONCLUSION

1) Seller reputations increase the number of high-quality goods delivered relative to the no-

reputation baseline.

2) Unverified claims are not sufficient to improve market outcomes.

3) Although certification is costly, sellers usually opt to certify; consequently, the number of high-

quality units increases, even though efficiency does not significantly increase due to the

certification costs. Certification appears sufficient to overcome the moral hazard problem.

4) Seller reputations modestly influence prices in some conditions, and signals and certification have

a significant impact on transaction prices.

5) Buyers are willing to forgive sellers quickly who previously deliver Regulars at Supers prices it is

explained at least in part by the absence of a profitably exit from the market.

6) laboratory results suggest that government regulators or non-governmental organizations can

improve environmental performance by providing the option of certified green labeling. (only a

single dimension of the product studied here)

ENVIRONMENTAL KUZNETZ CURVE Yandle, Vijayaraghavan, Bhattarai, The Environmental Kuznets Curve: A Primer. Kuznets (1955) hypothesized that income inequality first rises and then falls with economic growth – inverted U shape - hence the name EKC (Environmental Kuznets Curve) for similar pattern in environmental policy


35

Since 1991 when EKC first reported [Grossman and Krueger‘s analysis of air quality measures in a cross-section analysis of countries for different years, investigating the claim that economic growth accompanying the NAFTA would foster environmental degradation – they did indeed identify a turning point where higher income started to improve air quality (for two indicators SO2 and dark matter, or smoke)], it has become standard fare in technical conversations about environmental policy. Early estimates showed that some important indicators of environmental quality such as the levels of SO2 and particulates in the air actually improved as income levels and levels of consumption went up Meadows, Meadows, Randers, & Behrens (1972) – ―Club of Rome‖ study, dire predictions (economic growth is bad, a threat to the environment) (in Stern 2004) Grossman & Krueger (1991) - impact of NAFTA study, much more optimistic predictions (economic growth may be good); popularized by 1992 World Bank Development Report (IBRD 1992) the theoretical framework still in early stages (see e.g. Lopez 1994 or Munasinghe 1999, in Yandle et al.) What is the basic issue?

- Can economic growth be the means to *eventual* environmental improvement? - Can have humankind ―have our cake and eat it‖ – a prospect of achieving sustainability without a

significant deviation from business (Rees 1990, p. 435, in Stern) - Does the following result (of an ―inverted U-shaped function of income per capita‖) generalize to

other emissions?

Environmental impact

The burning of coal and/or petroleum by industry and power plants generates sulfur dioxide (SO2), which reacts with atmospheric water and oxygen to produce sulfuric acid (H2SO4). This sulfuric acid is a component of acid rain, which lowers the pH of soil and freshwater bodies, sometimes resulting in substantial damage to the environment and chemical weathering of statues and structures. Fuel standards increasingly require sulfur to be extracted from fossil fuels to prevent the formation of acid rain. This extracted sulfur is then refined and represents a large portion of sulfur production. In coal

http://en.wikipedia.org/wiki/Coal

http://en.wikipedia.org/wiki/Petroleum

http://en.wikipedia.org/wiki/Power_plants

http://en.wikipedia.org/wiki/Sulfur_dioxide

http://en.wikipedia.org/wiki/Oxygen

http://en.wikipedia.org/wiki/Sulfuric_acid

http://en.wikipedia.org/wiki/Acid_rain

http://en.wikipedia.org/wiki/PH

http://en.wikipedia.org/wiki/Soil

http://en.wikipedia.org/wiki/Environment_%28biophysical%29

http://en.wikipedia.org/wiki/Chemical_weathering

http://en.wikipedia.org/wiki/Fossil_fuel


36

fired power plants, the flue gases are sometimes purified. In more modern power plants that use syngas the sulfur is extracted before the gas is burned.

Yandle et al

- Where did the name ―Environmental Kuznets Curve‖ come from? - Why Kuznets? - What have we learned about the statistical relationships between various measures of

environmental quality and income? - Do all aspects of environmental quality deteriorate or improve systematically with economic

development? - Does the degree of property rights and contract enforcement make a difference?

So what is the intuition? ―At the low levels of per capita income found in pre-industrial and agrarian economies, where most economic activity is subsistence farming, one might expect rather pristine environmental conditions, relatively unaffected by economic activities—at least for those pollutants associated with industrial activity. The EKC statistical relationship suggests that as development and industrialization progress, environmental damage increases due to greater use of natural resources, more emission of pollutants, the operation of less efficient and relatively dirty technologies, the high priority given to increases in material output, and disregard for—or ignorance of—the environmental consequences of growth. However, as economic growth continues and life expectancies increase, cleaner water, improved air quality, and a generally cleaner habitat become more valuable as people make choices at the margin about how to spend their incomes. Much later, in the post-industrial stage, cleaner technologies and a shift to information and service-based activities combine with a growing ability and willingness to enhance environmental quality (Munasinghe, 1999).‖ ―Saying all this may tempt one to think that higher incomes alone will solve most environmental problems. Unfortunately, life is not that simple. If it were, transfers of income from richer to poorer societies—through foreign aid, for example—would enable the recipients to avoid environmental destruction. The movement along an environmental Kuznets curve is also a movement through a well-known set of property rights stations. In primitive societies managed by tradition or tribal rule, part of the resource base may be treated as a commons. With growing scarcity, however, a time comes when some aspects of the commons become defined as public or private property. As ―propertyness‖ expands—and private property is the most incentive-enriched form—individuals have a greater incentive to manage, to conserve, and to accumulate wealth that can be traded or passed on to future generations. … Eventually, when most aspects of the environment are defined as property, the community moves rapidly in the race to improve environmental life.‖ ―Thus, the Environmental Kuznets Curve is a proxy for a property rights model that begins with a commons and ends with private property rights.‖ Evidence:

- Grossman and Krueger 1991 – identified the turning point for SO2 and dark matter

- Shafik and Bandopadhay (1992) – for SO2, suspended particulate matter, fecal coliform [a

bacteria - its Aerobic decomposition can reduce dissolved oxygen levels if discharged into rivers or waterways]

- Hettige et al 1992 - using a composed production toxicity intensity index – found EKC for toxic intensity per GDP, not for ―per manufacturing output‖… Manufacturing, which is just one part of GDP, did not become cleaner or dirtier as income changed. Instead, manufacturing became smaller relative to services and trade in expanding economies. … This could mean that

http://en.wikipedia.org/wiki/Synthesis_gas


37

dirty production shifts elsewhere!!! They find that ―toxic intensity in manufacturing has grown much more rapidly in economies that are relatively closed to international trade‖

- Suri and Chapman (1998) - focused on energy consumption showed that as industrialized economies matured, they moved to services and then imported more manufactured goods from developing countries -> the global diffusion of manufacturing contributes to environmental improvements as incomes rise and development continues

- Gokany 2001 – ―Open economies improve their environments‖ (positive impact of international trade?)

- Cropper and Griffits (1994) – as income increases the rate of deforestation levels off - Panayotou (1995) – finds that the turning point for deforestations occurs much earlier than

for emissions,‖ because deforestation for either agricultural expansion or logging takes place at an earlier stage of development than heavy industrialization‖

- Shafik (1994) – four determinants of environmental quality -> mixed results - Grossman and Krueger (1995) – more extensive empirical study, focusing on water quality ,

found a turning point for 11 out of 14 selected indicators - following up, Selden and Song (1994) – 2 G&K‘s air pollutants + oxides of nitrogen and carbon

monoxide – found EKC for all 4, turning points for pollutants from G&K are significantly higher than G&K‘s estimates (they use readings from both urban and rural areas, G&K only urban)

- Cole et al – examined a wide range of indicators for different countries and On property rights…

- Panayotou (1997) examining EKC for sulfur dioxide found that faster economic growth and higher

population density do increase moderately the environmental price of economic growth, but better policies such as more secure property rights under a rule of law and better enforcement of contracts and effective environmental regulations can help flatten the EKC and reduce the environmental price of higher economic growth. Similar results obtained by Qin (1998), and Bhattarai (2000)

Conclusion

- there is no single EKC relationship that fits all pollutants for all places and times - The indicators for which the EKC relationship seems most plausible are local air pollutants such

as oxides of nitrogen, sulfur dioxide, and particulate matter. - there is no evidence to support the EKC hypothesis for gases such as carbon dioxide, which

cause no harm locally but may affect the global climate as they accumulate in the atmosphere - the evidence for water pollution is mixed - better policies can help to flatten ECK and perhaps to achieve an earlier turning point


38

Stern, The Rise and Fall of the Environmental Kuznets Curve

- currently, dispute over methodology and the reality of the EKC

―most of the EKC literature is econometrically weak‖ (Stern p. 1420) because it does not account properly for the statistical properties of the data used [serial dependence, stochastic trends] and issues of model adequacy [possibility of omitted variable bias]

―when we … use appropriate techniques, we find that the EKC does not exist‖ (Stern p. 1420)

―It seems that emissions of most pollutants and flows of waste are monotonically rising with income‖, with ―income elasticity‖ being less than 1 and moderated also by other factors (―income independent, time related effects [such as institutional quality] reduce environmental impacts in all countries at all levels of income‖ (Stern p. 1420))

→ In rapidly growing middle-income countries, scale effects tend to dominate time effects

→ Pure growth, without change in the structure or technology of an economy, leads to more pollution and other negative environmental impacts (scale effects)

→ As economies grow, output mix changes – from more pollution intensive to less pollution intensive industries -- input mix changes, emission specific regulations might change and actually hasten output/input mixes, etc.

In wealthy countries, time effects can dominate scale effects (partially because growth is slower)

- currently, also a dispute over the appropriate mix of metrics /statistics and theory, as well as the facts

―many environmental economists take the EKC as a stylized fact that needs to be explained by theory.‖ (Stern p. 1421)

―the EKC has never been shown to apply to all pollutants or environmental impacts and recent evidence.‖ (Stern p. 1421)

- A number of theoretical models have been developed of how preferences and technology interact

to result in different time paths of environmental quality. (Stern p. 1422) - Most of these studies can generate an inverted U-shape curve of pollution intensity but … the

assumptions made and the values given to particular parameters make all the difference whether indeed the EKC gets generated.(Stern p. 1422)

- ―Many EKC studies have also been published that include additional explanatory variables, intended to model underlying or proximate factors such as ‗political freedom‘ … or output structure … or trade … In general, the included variables turn out to be significant at traditional levels. … it is not clear what we can infer from this body of work [because of potential for omitted variable bias].

- Given these problems, I do not review these studies systematically here.‖ (Stern p. 1423) - Turning point estimate differ widely (see Table 1, Stern p. 1425, here reproduced in parts only): - ―The only robust conclusions from the EKC literature appear to be that concentrations of

pollutants may decline from middle income levels, while emissions tend to be monotonic in income. … ― (Stern p. 1426)

- How about the feedback from environmental damage to economic production (typically assumed away by most EKC studies)?

- Could a EKC-type relationship, if it does indeed exist, partly or largely be the results of the effects of trade on the distribution of polluting industries, as proposed by Arrow et al. (Science 1995) and Stern et al. (World Development 1996)? -> literature on pollution havens ―There is no clear answers on the impact of trade on pollution from the empirical EKC literature.‖ (Stern 1427)

- ―It seems unlikely that the EKC is an adequate model of emissions or concentrations. I concur with Copeland and Taylor (JEL 2004), who state that: ―Our review of both the theoretical and


39

empirical work on the EKC leads us to be skeptical about the existence of a simple and predictable relationship between pollution and per capita income.‖ (Stern p. 1435)

- ―The true form of the emissions-income relationship is likely a mix of two of the scenarios proposed by Dasgupta et al. (JEP 2002) illustrated in Figure 3. The overall shape is that of their ‗new toxics‘ EKC – a monotonic increase of emissions and income. But over time this curve shifts down, which is analogous to their ‗revised EKC‘ scenario. [innovations being adopted in high-income countries, and with a short lag in the majority of poorer countries.]‖ (Stern p. 1435)

environmental economics, eco 565/365

Documents