Can Markets Set Caps for Cap-and-trade Pollution Permits?

1
Can Markets Set Caps for Cap-and-trade
Pollution Permits?

• Ted Bergstrom
• Economics Department
• University of California Santa Barbara

2
Alternate Title
Arrow, Coase, Missing Markets,
and the Easter Bunny
3
Economists have lots of good ideas that
politicians and public dont like as well as we
do.

• Gasoline Tax
• Congestion Tolls
• Legalize and tax recreational drugs
• Legalize trade in body organs
• Subsidize surstromming?

4
Tradable pollution permits an economists idea
that got popular

• Endorsed by Clinton, Edwards, Obama, and McCain
• By major energy companies
• Some Environmental Groups
• Environmental Defense Fund, Union of Concerned
  Scientists
• But not Sierra Club

5
A Coasian Solution to Externalities?

• A fond hope of many economists is that
  externalities can be internalized by establishing
  property rights and market institutions.
• Achieved with a minimum of government regulation.
• Reduce politicians chances for corruption and
  bribery.
• Is this Easter Bunny economics?

6
What is cap-and-trade?

• Government requires a permit for every unit of
  pollution produced.
• Issues a fixed number of tradable permits.
• Permits typically issued to original polluters in
  proportion to historic pollution.
• Wouldnt have to be this way. Could be auctioned.
• Number of permits decreases over time.

7
Los Angeles Air Quality

• Typical example Southern Californias RECLAIM
  market started in 1992.
• Tradeable permits for SOx and NOx issued to large
  polluters in L.A. in proportion to emissions they
  produce at full capacity.
• Number of permits issued reduced by 8 per year
  1992-2003.
• Large brokerage firms handle trades
• Coasian Transaction costs? Commissions 1-3.5
  percent
• L.A. air quality has improved drastically.

8
Other Cap-and-trade markets

• Acid Rain SOx markets in Eastern US States
• Water pollution in Wisconsin and Colorado
• Fisheries in Australia and New Zealand
• Tradable permits to own cars in Singapore.
• Number increases by 3 per year.
• New permits auctioned in Walrasian auction

9
Alternatives to Cap-and-Trade

• Command and control measures
• Pigovian Pollution tax
• Why do energy companies like cap-and-trade?
• Preferable to C and C or pollution tax if they
  get the permits.
• Especially if abatement technologies are cheaper
  for them than for others.

10
Advantage of Cap-and-Trade

• Decentralizes firms decisions about whether to
  install abatement devices, change technologies,
  cut back output, move away.
• Marginal benefits from polluting are about the
  same across all firms and equal to marginal
  costs of abatement.
• Conditional on aggregate level of pollution,
• if market is competitive, outcome is Pareto
  efficient.

11
Further advantage

• Endowing current polluters, reduces opposition,
  makes change nearly Pareto improving
• Except for grouchy people at Sierra Club
• Even grouches may concede that they need to buy
  off polluters to get reform.

12
What is missing

• Cap-and-trade policies do a good job of
  allocating chosen level of cap, but cap is
  currently chosen by political bodies.
• Can the caps also be determined by markets?

13
Arrows suggestion

• Arrow (1969) stated
• By a suitable and not unnatural reinterpretation
  of the commodity space, externalities can be
  regarded as ordinary commodities and all of the
  formal theory of competitive equilibrium is
  valid, including its optimality

14
Arrows formalism

• Treat pollution of each victim as a separate
  commodity.
• These are joint products of pollutant produced by
  a polluter.
• Competitive equilibrium would exist in this
  imaginary economy and would be the same as a
  Lindahl equilibrium, which would be Pareto
  optimal.

15
Does this mean that decentralized markets can
optimize externalities ?
16
Not so fast, says Arrow

• The existence of Lindahl equilibrium prices for
  pollution does mean that they can be implemented
  in a working market.
• Two problems.
• 1) Free-riders cannot be excluded from benefits
  of air quality purchased by others.
• 2) Personalized markets are thin markets.
• Competitive outcome is unlikely.

17
Do Arrows 2 problems doom a full market solution
for pollution control methods?

• Rays of hope from mechanism design.
• Weakly dominant Groves Clarke Mechanism
• Nash implementation
• Groves-Ledyard
• Mark Walker and others
• Not like markets, but maybe some kind of market
  will work.

18
Is there a way around these problems?
19
Multiple polluters, single victim

• Assume there are many firms, each produces a
  single pollutant as a bi-product and could reduce
  the ratio of pollutant to output at some cost.
• There is only one pollution victim.
• Government monitors pollution and requires a
  permit for each unit of pollutant emitted.
• Government issues enough permits for total
  pollution produced before permits required.

20
Permits given to Polluters

• Government issues permits to original polluters
  equal to their previous emissions.
• Pollution victim can buy permits and retire them.
• Outcome will be efficient and Pareto superior to
  status quo ante.
• Damage to victim of marginal unit of pollution
  equals marginal benefit of polluting and marginal
  cost of abatement for each polluter.

21
Permits given to victim

• We may think that the victim had a right to a
  clean environment and should not have to pay.
• Coasian argument No worries. Just endow the
  victim with the permits. Outcome will be the same
  except for income distribution.

22
Not so fast, Mr. Rabbit

• Victim will now be a monopoly seller of permits.
• Marginal cost of selling a permit is victims
  disutility of pollution.
• Firms demand for permits will be downward
  sloping. Marginal revenue will exceed price.
• Victim will sell too few permits for efficiency.
• We could use a Pigovian tax, but how high should
  the tax be set?

23
Many Polluters, Many Victims

• Assume many polluters as before, but now let
  there be many pollution victims.
• Suppose that pollution permits are required and
  permits issued to current polluters allowing
  pre-permit pollution levels.
• This establishes property rights to air quality.
  Victims can buy and retire permits. Markets give
  us efficiency, right?

24
Not so fast, Mr. Rabbit

• Nash equilibrium in this market will be the
  voluntary provision of public goods outcome.
• The equilibrium price in this market will be
  equal to the marginal cost of pollution to any
  individual victim who buys tickets.
• Each retired premit reduces pollution for all
  victims, yet purchasers consider only their own
  gains.
• Efficiency requires price of permit equals sum of
  marginal costs to victims. Equilibrium has far
  lower price, equal to maximum of marginal costs
  to victims.

25
Endowing the victims

• Can we cure the problem by endowing the victims
  rather than the polluters?
• Suppose the total supply of permits was divided
  equally among victims. Market would be
  competitive now many victims
• Victims would be willing to supply at their
  marginal damage. Equilibrium price would be
  marginal damage to a single person. Again far
  below efficient price. And total
• revenue received by victims far less than total
  damage caused by pollution.

26
Personalized permits

• Suppose that the government endows each pollution
  victim a number of pollution permits equal to
  the total amount of pollution produced before
  permits were introduced.
• For each unit of pollution that a firm produces,
  it must have one personalized permit from each
  victim.

27
Unanimous consent?

• What goes wrong when many victims are endowed
  with tradable permits?
• Each victim can sell permission to pollute not
  only himself but everyone in the community.
• Lets try market devices that require unanimous
  (or near unanimous consent) of victims.
• (Wicksell suggested that near unanimity might be
  required for tax-funded projects.)

28
Too many commodities?

• Think of a global market with a small number of
  nations acting on behalf of their citizens.
  Polluters need a permit from each country for
  each unit of pollution they produce.

29
Complementary monopoly problem

• Each victim is the monopoly supplier of his own
  permits.
• Permits are perfect complements.
• This is the complementary monopoly model of
  Cournot.
• Output is even smaller than with a single
  monopoly.
• Total markup as percentage of sum of prices is
  N/E, where E is elasticity of demand and N the
  number of victims.

30
Fixing Arrows Problem 2

• We could have personalized permits without
  monopoly by spreading initial endowments of each
  victims permits among many participants
• either victims or polluting firms, depending on
  distributional goals.
• There would be many demanders for these
  permitsseveral firms and one victim.

31
Lindahl equilibrium

• Endowment of X personalized permits for each
  person is distributed.
• In equilibrium, each victim buys the same
  quantity Q of his own permits and firms buy a
  total of X-Q of each kind of permit.
• Each persons personalized price is his marginal
  rate of substitution between pollution and
  private goods.
• Cost to a firm of producing a unit of pollution
  is sum of prices of personalized permits.

32
Can we implement Lindahl with competitive permit
market?

• The Lindahl equilibrium would be a competitive
  equilibrium if each victim experienced X-Qi units
  of pollution when he retires Qi permits.
• But in the actual market we have proposed, firms
  can only produce
• X-maxQj units of pollution.

33
Lindahl equilibrium is not supported by a
competitive equilibrium in pollution tickets

34
But is there some other kind of equilibrium and
is it any good?

• What if we make the amount of pollution that a
  producer is allowed some c.e.s function of the
  numbers of each type of ticket that he holds?

35
Hope springs eternal
36
Had enough?