4/23/2015

1

1

Costs and Benefits of Mitigating Global Climate Change

Some slides adapted from: Maureen Cropper, University of Maryland and William D. Nordhaus, Yale University

1. Assessments of climate change2. Costs of mitigation

▫ What factors drive estimates of the costs of mitigation and why do estimates differ?

▫ How costly is it to stabilize at 450 vs. 550 ppm?

3. Benefits of mitigation▫ Overview of benefits estimation▫ How benefits vary geographically

4. Balancing costs and benefits▫ How to deal with timing issues, uncertainty? What does

this imply about the optimal stabilization level?

2

“Most of the observed increase in global average temperatures since the mid-20th century is very likely due to the observed increase in anthropogenic greenhouse gas concentrations.” (IPCC 2007)

4/23/2015

2

IPCC AR4 Model Results: History and Projections

3

4

Sources of GHG Emissions

Source: IPCC (2007)

4/23/2015

3

5

What Determines Fossil Fuel Emissions?

• Emissions = Population x (GDP per capita)

x (Energy/GDP) x (CO2/Energy)

• Policies to reduce emissions must reduce:▫ Energy per unit of GDP (become more energy efficient)

▫ Carbon intensity of energy (switch from coal to gas; fossil fuels to renewables and nuclear)

• Carbon tax or other explicit policies are needed to promote move to a low carbon economy

6

How Economists Model Mitigation Costs

• “Top-down” computable general equilibrium models ▫ Divide the world into economically important regions▫ Model demand and supply for commodities in all sectors of

the economy

• Estimate costs of mitigation by imposing a worldwide carbon tax.▫ Tax causes substitution of low carbon fuel.▫ The cost of doing this represents the cost of mitigation.

4/23/2015

4

7

What Determines the Cost of Reaching Particular GHG Targets?• Business-as-usual concentrations of GHGs

• Ease of substituting energy for other inputs

• Ease of substituting low-carbon for high-carbon fuels

• Assumptions about rate of technical progress

8

Estimated Mitigation Costs

Source: den Elzen et al. (2007)

Pathways: default (black), delayed (red) and early action (green); envelopes (set of grey lines)

Miti

gatio

n co

sts

as %

of G

DP

450 ppm 450 ppm

550 ppm 550 ppm

A1b (Moderate Growth) B1 (Low Growth)

4/23/2015

5

9

What Are the Benefits of Mitigating GHG Emissions?• Stabilizing at 550 ppm implies:

▫ 99% chance of exceeding 2°C

▫ 69% chance of exceeding 3°C

▫ 24% chance of exceeding 4°C

▫ 7% chance of exceeding 5°C

• What does this mean in terms of impacts?

• Benefits of mitigation = value of avoided damages

1°C 2°C 5°C4°C3°C

Sea level rise threatens major cities

Falling crop yields in many areas, particularly developing regions

Food

Water

Ecosystems

Risk of Abrupt and Major Irreversible Changes

Global temperature change (relative to pre-industrial)0°C

Falling yields in many developed regions

Rising number of species face extinction

Increasing risk of dangerous feedbacks and abrupt, large-scale shifts in the climate system

Significant decreases in water availability in many areas, including Mediterranean and Southern Africa

Small mountain glaciers disappear – water supplies threatened in several areas

Extensive Damage to Coral Reefs

Extreme Weather Events

Rising intensity of storms, forest fires, droughts, flooding and heat waves

Possible rising yields in some high latitude regions

4/23/2015

6

11

Sectors for Which Impacts Have Been Studied

• Agriculture• Forestry• Coastal Infrastructure• Energy▫ Supply & Demand

• Water▫ Supply & Demand

• Health Impacts▫ Malaria▫ Malnutrition▫ Heat Stroke, Floods

• Species Loss• Ecosystem Changes

12

Climate Change Damages

• Estimating the impact of climate change on society is extremely challenging.

• Damages in developing countries—especially low latitude developing countries—are often a larger percent of GDP than in high income countries

• To compare with global costs of mitigation, need to aggregate benefits of mitigation across countries

• How this is done has a big impact on results:▫ Aggregate by GDP or population?

4/23/2015

7

13

The damages of climate change with increasing global temperatures

-12

-10

-8

-6

-4

-2

0

2

4

0 2 4 6

Pe

rce

nt

of

wo

rld

GD

P

Global mean temperature

Hope

Mendelsohn

Nordhaus, output

Nordhaus, population

Tol, output

Tol, equity

Benefits

Costs

14

Balancing Costs and Benefits• Integrated Assessment Models (IAMs)

▫ Need to consider: impact of emissions on temperature damages of temperature on GDP costs of mitigating emissions

▫ Path of emissions chosen to maximize present discounted value of utility of per capita GDP

4/23/2015

8

15

Balancing Costs and Benefits• Key Integrated Assessment Models

•William Nordhaus’s DICE/RICE models

• Richard Tol’s FUND model

• Chris Hope’s PAGE model

16

Fossil fuel usegenerates CO2emissions

Carbon cycle: redistributes around 

atmosphere, oceans, etc.

Climate system: change in radiative warming, precipitation, ocean currents, sea level rise,…

Impacts on ecosystems,agriculture, diseases,

recreation, …

Measures to controlemissions (limits, taxes, 

subsidies, …)

Basic Model Structure

4/23/2015

9

17

What Drives IAM Results?

• Key factors affecting the optimal reduction in GHG emissions are:

▫ Shape of the aggregate damage function

▫ Discount rate used to express future damages and costs in present value terms

▫ Impact of emissions/atmospheric concentrations on temperature

18

Damage Function• Steeper in DICE and PAGE models than in FUND

4/23/2015

10

19

Importance of Discounting

• Future costs and damages associated with GHG emissions expressed in present value terms.

▫ Should discount rate reflect market opportunities or ethical concerns?

• Cost of mitigation begin today, whereas major climate damages occur in the future.

• Lower discount rates imply more mitigation

20

The Discounting Controversy

• DICE uses market interest rates (approximately 4.5%)

• PAGE uses a discount rate that reflects how we weight the preferences of future generations vs. ourselves▫ Allowing for the fact they will be better off▫ Discount rate approximately 1.4%

4/23/2015

11

• Impact of atmospheric CO2 on temperature is uncertain

• PAGE treats this parameter as uncertain; DICE and FUND do not

21

Importance of Uncertainty & Risk Aversion

▫ Climate sensitivity = impact on temperature of a doubling of atmospheric CO2 from pre-industrial levels

▫ 95% confidence interval = 2.5°C to 5.4°C

• Low abatement costs

• Increasing future damages

• High risk

Gradual Emissions Reduction

Uncertainty

• High abatement costs

• Moderate future damages

• Moderate risk

Aggressive Emissions Reduction

• Accounting for risk aversion increases the incentive for aggressive emissions mitigation

Importance of Uncertainty & Risk Aversion

4/23/2015

12

23

Mitigation Benefit and Cost Estimates Relative to No Mitigation – Nordhaus’s DICE Model

Benefits (Avoided Damages)

Abatement Costs

Benefits minus Costs

$ Trillion (US 2005) $Trillion

Nordhaus/DICE Optimal 5.2 2.2 3.1

420 ppm stabilization 12.6 27.2 ‐14.6

560 ppm stabilization 6.6 3.9 2.7

700 ppm stabilization 5.2 2.2 3.1

Note: DICE ignores uncertainty and uses a relatively high discount rate.

Source: Wheeler (2007)

24

What Can We Conclude About Costs and Benefits of Mitigation?

• Conservative model estimates suggest that stabilizing CO2 at 560 ppm passes the cost-benefit test.

• Accounting for uncertainty/risk or lowering the discount rate would increase the benefits, implying an even more aggressive target.

• So how do we get there? What policy measures are possible?

4/23/2015

13

A Price on Carbon

25

• Economic participants (millions of firms, billions of people, trillions of decisions) need to face realistic carbon prices if their decisions about consumption, investment, and innovation are to consider the “negative externality” of carbon emissions.

1. To be effective, we need a market price of carbon emissions that reflects the social costs.

2. Moreover, to be efficient, the price must be universal and harmonized in every sector and country.

• But what is the appropriate price of carbon? This question can also be addressed by integrated assessment models.

26

1. Baseline: No emissions controls.

2. Optimal policy: Emissions and carbon prices to maximize discounted economic welfare.

3. Limit to 2°C: Constraining global temperature increase to 2°C above 1900

4. Strengthened Kyoto Protocol: Modeled on US proposal with rich countries starting at same time and developing countries joining after 1-3 decades

Specific Policy Scenarios

4/23/2015

14

27

300

400

500

600

700

800

900

1000

1100

1200

1300

Car

bon

con

cen

trat

ion

s (p

pm

)Optimal

Baseline

< 2 deg C

Strong Kyoto

CO2 Concentration Projections

28

Temperature Projections

4/23/2015

15

Carbon Prices

29

0

100

200

300

400

500

600

700

800

900

1000

2005 2015 2025 2035 2045 2055 2065 2075 2085 2095 2105

Car

bon

pric

e (2

005

US$

per

ton

C)

Optimal

Baseline

< 2 degrees C

Strong Kyoto

What Would Carbon Prices Mean in Practice?

30

Carbon tax, 2010

Increase in price of energy, US

[$/tC] Gasoline

All energy 

expenditures

Kyoto Global Average $2 0.2% 0.3%

"Optimal" 35 3.3% 5.4%

2 deg. C Constrained 50 4.8% 7.7%

4/23/2015

16

Major Policy Approaches

• Internationally harmonized carbon tax is the economic ideal.• Raise fossil fuel prices proportional to carbon content• All countries would target a comparable tax• Level of tax set to meet emissions target • Uses consumption basis for tax• Raise revenues and have potential for reducing dead

weight losses of taxation.

31

Major Policy Approaches• Universal cap and trade is a close second (only if

well-designed)▫ The fundamental defect is the lack of a connection

between targets (emissions) and objective (climate or damages).

▫ Caps are troublesome in a world of differential economic growth and uncertain technological change.

▫ Cap-and-trade regulations show extremely volatile trading prices for carbon emissions.

▫ Systems with international trading are much more susceptible to corruption than tax regimes.

▫ Taxes are more familiar and reliable in every country of the world.

32

4/23/2015

17

Major Policy Approaches

• Voluntary measures (carbon offsets) are difficult to calculate and verify and probably useless in the long-term.

• Regulatory substitutes (CAFE standards, ban on light bulbs, …) are very inefficient approaches

33

34

Problems in Reaching an International Agreement

• Reducing GHGs is a global public good▫ If one country reduces, all countries benefit▫ Individual countries have an incentive to “free ride”

• Benefits of avoiding climate change occur in the future, and most benefits accrue to developing countries.

• Developing countries argue that they are not responsible for the majority of the stock of GHGs

• … and their emissions are very low on a per capita basis.

4/23/2015

18

The Externalities Game (TEG)

Motivation:• To obtain a sense of the ethical dilemmas involved in

climate change mitigation, our next case will involve the Externalities Game.

• An “externality” is a second-party effect arising from the production and/or consumption of goods and services for which no appropriate compensation is paid. ▫ For example, in climate change an externality would be sea level

rise or global temperature change.

▫ Like the noncooperative games examined in game theory (e.g. the Prisoners’ dilemma), externalities create tension between individuals who seek to achieve personal well-being at the expense of communal well-being.

• In this game, the outcomes that impact each player are determined, in part, by the actions of other players.

4/23/2015

19

Review of Rules:• Each player is randomly assigned one of three possible roles:

Luxury, Intermediate or Subsistence Producer

• Players will make production decisions that result in points (for themselves) and externality costs (shared by everyone).

▫ Whereas points accumulate linearly with production, externalities grow exponentially.

• Production Decisions will be kept confidential, unless players choose to reveal their decision to others

• Your score on this case study will be determined by the production points that you are able to accumulate individually, minus the externality costs that are incurred collectively.

4/23/2015

20

Review of Rules:• Randomly assigned roles:▫ Luxury Producer (L): gain the most points per

unit of production, but also emit the greatest amount of externalities Max production level = 10

▫ Intermediate Producer (I): gain the second most points per unit of production, and emit the second highest amount of externalities Max production level = 50

▫ Subsistence Producer (S): gain the least amount of points per unit of production, but emit the least amount of externalities Max production level = 240

• A round of game play consists of two steps: 1) production decisions by all players2) point-sharing negotiations.

• Players can only produce whole goods up to their maximum production capacity and not less than 0.

• Players can make deals during the game to share points or limit production for the greater good. However, the instructor will not enforce agreements. Players may lie to each other about their behaviors, and in many cases these lies may go undetected.

• Players do not have to cooperate with anyone if they do not choose to, but it is in their best interest to do so.

• The people they share with and the amount they share is solely at their own discretion and will be kept confidential.

Review of Rules:

4/23/2015

21

Review of Rules:• Players are allowed to numerically experiment with various production

choices. To experiment with possible choices, players can enter their production decision (as well as the assumed decisions of others) into their copy of the spreadsheet and view the consequences of that decision.

• The actual consequences of a player’s production choices will not be fully attained until after the round has been fully played.

• When all players have come to a final decision, or when the instructor states that deliberation must end, the players will submit their final production value on a note card. *THIS VALUE DOES NOT NECESSARILY HAVE TO BE WHAT WAS AGREED UPON WITH THE OTHER PLAYERS*

• The scores are then calculated by the instructor’s spreadsheet, and they are revealed to the students with an anonymous list of emissions generated by each player.

• A follow-up step of point sharing will then occur.

Susan Spierre

Example of Game Card

S28150 +83 Give 3 points

to J. Smith+80

ProductionDecision Resulting

Grade PointsPoint Sharing Final Grade

Points

Player IDPlayer Name

4/23/2015

22

Reflection Questions• How did you feel about the outcome? Why?

• How did you feel about your assigned role (L, I, S)? Why?

• Did your assigned role determine your behavior? How?

• How did you react to others’ behavior? Did you feel mad at anyone?

• Did you follow the determined optimal strategy? If not, why not?

Reflection Questions (cont’d)

• How did you decide who to share points with?

• Did you try to get others to share points with you? If so, how?

• What type of real-world situations can we relate this to?

• Are there such techno-scientific optima in the real world?

• How could we re-design the underlying system to solve the non-cooperation problem?