Transition Strategies for Alternative Transportation Fuels ...web.mit.edu/jjrs/www/AETTPresentations/StrubenAETT200906.pdf · Use of petroleum product by sector in the US, 1900 –

© Jeroen Struben, MIT (2009)

This Research is sponsored by Shell Hydrogen, Ford Motor Company, National Renewable Energy Laboratories, and the MIT Sloan Project on Industry and Organization Studies. All statements are of the MIT authors. The authors claim no conflict of Interest. All experiments results are for qualitative interpretation only!

Transition Strategies for Alternative Transportation Fuels and Vehicles

Jeroen Struben [email protected] Sloan School of Management

Desautels Faculty of Management, McGill University

MIT/Ford/Shell Workshop Meeting,

June 9 2009

2


Many Project Contributors• Co-leads MIT energy transitions in transportation project

– Prof. Jeroen Struben (McGill; research affiliate @ MIT)– Prof. John Sterman (MIT)

• Students past and present– Derek Supple (MIT)– Qi Zhang (MIT)– Jessica Laviolette (MIT)– David Keith (MIT)– Katherine Dykes (MIT)

• Sponsors and informants– David Chock (Ford Motor Company)– Yimin Liu (Ford Motor Company)– Margaret Whalen (Ford Motor Company)– Sandy Winkler (Ford Motor Company)– Nikunj Gupta (Shell Company)– Henk Mooiweer (Shell Company)– Jooske van der Graaf (Shell Company)– Monisola Olaweraju (Shell Company)– Britta Gross (General Motors)– Cory Welch (National Renewable Energy Labs)

3


Summary

• Report latest results from the MIT alternative energy and transportation transition model

– New model enhancements• Additional AFV platforms

• Upstream fuel supply chains

• Enhanced consumer behavior

• Enhanced representation of OEM R&D, learning, technological spillovers

• Carbon Capture and Storage

– Analysis of platform and fuel diffusion• Fuel availability and price uncertainty

• Impact of consumer and OEM responses to carbon prices

• Impact of consumer and OEM responses to early/late breakthroughs with 2G biofuels and/or CCS

• Purpose is to– Further develop our conceptual models on transportation transition

challenges and opportunities

– Improve simulation model

– Explore further critical questions

4


2/3rds of all the world’s petroleum is used by transportation

Use of petroleum product by sector in the US, 1900 – 2005. Data source: US Department of Commerce, 1975. Historical Statistics of the United States, Colonial Times to 1970, Bureau of the Census, Washington, DC. US Energy Information Administration, 2007. Annual Energy Review 2006, Estimated Petroleum Consumption by Sector, US Department of Energy, Washington, DC. (From Schafer et al., MIT Press, forthcoming)

5


0

5

10

15

20

1880 1900 1920 1940 1960 1980 2000

US Petroleum Production and Consumption

Mil

lio

n B

arr

els

/Day

Alaska

Production,Lower 48 States

Consumption

Imports

Reducing Petroleum Consumption a Vital US National Interest

Imports(66%)

Alaska(4%)

Lower 48(30%)

US Oil Consumption, 2008

• Declining domestic production• High and rising import dependence• Vulnerability to supply disruption, geopolitical instability• Local air pollution and health effects• Greenhouse gas emissions

6


If the projected world population of 9 billion people in 2050 lived the way Americans do today...

• There would be 7.6 Billion motor vehicles on the roads

• Transportation alone would consume 440 million barrels of oil per day

– Total world oil production today is 82 million bbl/day

• CO2 emissions from transportation alone would be 62 billion metric tons/year

– Total world emissions from fossil fuels today ≈ 28 billion tons CO2/year

• The current model of development and transportation cannot scale to a world of 9 billion, all of whom aspire to live the way we do

– New energy supply technologies are necessary but not sufficient

– End-use efficiency improvements are necessary but not sufficient

• A new transportation system is coming.

– What pathway?How will be a leader?

– Note only of company but also of national Interest: leadership essential to preserve and enhance domestic innovation, investment, and jobcreation.

The current transportation model does not scale:

Adapted from: MIT Transportation Initiative, Profs. Barnhart, Jaillet, Sheffi, Sterman, Waitz, Zegras

7


Mass mobility wave in motion

The Tata Nano: a $2500 car made for a market of one billion people.

Source: MIT Transportation Initiative, Profs. Barnhart, Jaillet, Sheffi, Sterman, Waitz, Zegras

8


Cutting greenhouse gas emissions enough to stabilize atmospheric concentrations…

Source: Stern Review, Fig. 8.4

9


Is incompatible with this:

…unless we dramatically innovate in transportationaccessibility

AnnualTravel

per Capita(kilometers/year

per person)

Source: MIT Transportation Initiative, Profs. Barnhart, Jaillet, Sheffi, Sterman, Waitz, Zegras

10

© Jeroen Struben, MIT (2009)Many opportunities exist. Why not widespread?

NegaWatts (Negative Costs)

11


Technology Disruptions: Traditional Perspectives

time

Lev

el

of

ad

op

tio

n

Alternative Technology

ConventionalTechnology

Level of adoption

Cu

mu

lati

ve

N

et

Eff

ort

Personal Stereo

Cameras

12


Regulation

Fuel supply &infrastructure

Service

Repair

Conversion

Stationary applications

Canonical Diffusion Examplesversus Network Technologies

Technological improvements

Consumer acceptance

Complement productor market

Color television

Laser printer

Refrigerator

Walkman

VHS player

Document reader/writer

Alternative Fuel Vehicle

13


Year

Sh

are

of

To

tal A

do

pti

on

Diffusion Patterns of Alternative Vehicle Platforms

0 10 20 30 400

0.1

0.2

0.3

0.4

0.5

DOE High

DOE Low

UK

Sweden

US

HFCVDiesel

Year

Sh

are

of

To

tal A

do

pti

on


0 10 20 30 400

0.1

0.2

0.3

0.4

0.5

DOE High

DOE Low

ARG

NZL (Conversions)

HFCVCNG

Year

Sh

are

of

To

tal A

do

pti

on


0 10 20 30 400

0.1

0.2

0.3

0.4

0.5

DOE High

DOE Low

HFCV

Year

Sh

are

of

To

tal A

do

pti

on


0 10 20 30 400

0.1

0.2

0.3

0.4

0.5

DOE High

DOE Low

HFCVHybrid - USSUV - US

Alternative Fuel Vehicle Diffusion: Slow and Fragile

Diesel

Scenarios

14


Alternative Fuel Vehicle Diffusion: Slow and Fragile

Year

Sh

are

of

To

tal In

sta

lled

Base


0 10 20 30 400

0.1

0.2

0.3

0.4

0.5

HFCVHybrid - USSUV - USDieselCNG

15


• Compressed Natural Gas – So far so good: Argentina

– Low penetration: Italy

– Sizzle and fizzle: Canada, New Zealand

– Stalled: California, Europe (excl Italy)

• Diesel – High/self-sustaining: Austria, Germany, France

– Sizzle and fizzle: USA

– Low penetration: Sweden, Ireland

• Ethanol – Sizzle and fizzle: Brazil (100% ethanol)

– So far so good: Brazil (flex fuels)

• Gas-electric hybrid (e.g. Prius)– So far so good USA

• Electric– Sizzle and Fizzle USA: EV1, other pure electrics

• Plug-in Hybrids– Too soon to tell Various (Ford, Toyota, Chevy Volt)

Many programs to introduce Alternative Fuel Vehicles (AFVs) fail

16


Creating a Market That Does Not Exist: Research Question

• How do AFVs enter the market, gain traction, and sustain themselves?

• What are viable pathways?

• Where are important pitfalls?

• Where lie important policy leverages?

• What level of coordination is needed ? Who? What kind of coordination? How long?

• What portfolios to build?

17


MIT System Dynamics Group Approach

• Suite of simulation models of AFV introduction, diffusion, competition

– Dynamic

– Spatially explicit

– Behavioral (realistic depiction of decision making)

• Broad model boundary to avoid unanticipated “side effects”

– Integration of vehicle technology, competition among AFVs and with ICE, fuel supply technology, consumer behavior, government policies, other key actors and factors

– Counterfactual analysis

• Grounded in detailed empirical study, quantitative and qualitative data

– Case studies of prior AFV programs and policies

See: Struben J, Sterman J, 2008, "Transition challenges for alternative fuel vehicle and transportation systems" Environment and Planning B 35(6) 1070 – 1097

18


A Broad Boundary

Consumers• Vehicle replacement• Platform consideration • Platform choice

- new/used, car/truck

• Trip choice• Refueling choice• Topping-off behavior

FuelRetailing• Entrance/exit• Location selection• Fuel station capacity

Stacked boxes imply multiple vehicle and fuel platforms (Internal combustion engines, HFCVs, hybrids; gasoline / biodiesel/ electricity etc..)

Vehicle Installed Base

Fueling Infrastructure

Vehicle Installed Base

Fueling Infrastructure

Vehicles In use,Fuel

Supply Chain

Fuel Suppliers & Producers• Entrance/exit• Price Setting• Experience• CCS Investment• Electric Power Mkt

Automotive Producers• Platform portfolio• Production capacity• Experience from R&D and spillovers

• R&D investment, incl.fuel efficiency

• Marketing

Policy makers• Supply subsidies

/taxes• Demand subsidies

/taxes• Campaigns• Pilot programs• CO2 taxes/price

• Spatial disaggregation• Electricity Market

SocioEconomic

Sector Interactions

19


Principal feedbacks

Vehicle Choice Attributes

• Purchase price

• Performance

• Driving Convenience

• Safety

• Operating Cost

• Ecological impact

.

Success

Stagnation

Failure

Time

Share of Vehicles in Use

Vehicles InVehicles InVehicles InVehicles InUseUseUseUse

FuelingFuelingFuelingFuelingInfrastructureInfrastructureInfrastructureInfrastructure

Fuel

Consumption

+

VehicleVehicleVehicleVehicleSalesSalesSalesSales

VehicleVehicleVehicleVehicleDiscardsDiscardsDiscardsDiscards

+

VehicleVehicleVehicleVehicleChoiceChoiceChoiceChoice

Fuel AvailabilityFuel AvailabilityFuel AvailabilityFuel Availability

R1R1R1R1

x

y

DriveDriveDriveDriveAttractivenessAttractivenessAttractivenessAttractiveness +

+

+

VehicleVehicleVehicleVehicleAttractivenessAttractivenessAttractivenessAttractiveness

+

+Vehicles InVehicles InVehicles InVehicles InUseUseUseUse


Fuel

Consumption

+



+


ConsumerConsumerConsumerConsumerConsiderationConsiderationConsiderationConsideration

+

+


SocialSocialSocialSocialExposureExposureExposureExposure

R3R3R3R3

R1R1R1R1

x

y


+

+


+



Fuel

Consumption

+

AutomotiveAutomotiveAutomotiveAutomotiveManufacturerManufacturerManufacturerManufacturerCapabilitiesCapabilitiesCapabilitiesCapabilities



+

+



+

+AutomotiveAutomotiveAutomotiveAutomotive

Economies of Scale,Economies of Scale,Economies of Scale,Economies of Scale,Scope, RDScope, RDScope, RDScope, RD



R3R3R3R3

R2R2R2R2

R1R1R1R1

x

y

+


+

+


+



Fuel

Consumption

+




+

+



+





R3R3R3R3

R2R2R2R2

R1R1R1R1

R4bR4bR4bR4b

x

y

R1bR1bR1bR1b

+


+

+

+


+



Fuel SupplyFuel SupplyFuel SupplyFuel SupplyChainChainChainChain

CapabilitiesCapabilitiesCapabilitiesCapabilities

Fuel

Consumption

+

+




+

+



+



Fuel AffordabilityFuel AffordabilityFuel AffordabilityFuel Affordability



R3R3R3R3

R2R2R2R2

R1R1R1R1

R4bR4bR4bR4b

x

y

InstitutionalInstitutionalInstitutionalInstitutionalCoevolutionCoevolutionCoevolutionCoevolution

R1bR1bR1bR1b

+

+


+

+

+

R4R4R4R4


+

+

Vehicles in

Use

20


Creating an AFV market: California as Laboratory for Experimentation

• Focus on Central/Southern California– 13.5 Million households– 13 Million ICE vehicles– 6,500 gasoline fuel stations

• Behavioral Dynamics– Willingness to consider an

AFV in purchase decision depends on marketing, social exposure to AFVs, word of mouth from others (favorable and unfavorable)

– AFV purchase decision also conditioned by inconvenience of fuel search and risk of no fuel

– Drivers will go out of their way for fuel – up to a point

– Drivers worried about fuel availability may top off tanks

Population

21


Liquid/gasfossil

gasoline,ethanol, diesel

fuel-cell

hydrogen

gasoline

conventional ICE/gasoline,hybrid

$ $

OE

M r

ein

vestm

ent,

le

arn

ing,

scale

, scope

$ $ $ $ $

Consumer Acceptance

Experiment 1: HFCV Diffusion

22


Example: HFCV• 2006 ICE/Gasoline Technical Parameters

• Hydrogen Fuel Cell Vehicles compared to current ICE: Initial Mature

– $35,000 production cost 2.25 1

– Equal Initial performance 1 1.25

– 35 mi/gge fuel economy 1.67 3

– 6 gge tank capacity 0.3 0.5

• Hydrogen Fuel Stations

– H2 Produced at Station Forecourt via Steam Reformation of Natural Gas

– $2.10 variable cost per gge H2 output (~$9/mcf natural gas, 70% efficiency)

– Selection, permitting, construction delays total 2 years

• Aggressive, coordinated, and persistent policies across the system:

– Intensive 15 yr marketing program to build awareness

– Fleet program involving 500,000 vehicles

– Full subsidy of HFCV vehicle price difference with ICE

– Intensive R&D programs to lower AFV cost and boost performance prior to roll out

– Fuel station rollout totaling about 800 stations

– Fixed $2.50 gge alt fuel retail markup for 10 years, gradual deregulation thereafter

– $0.50/gallon additional gasoline tax

– Cost of R&D, marketing program, fleet program, AFV subsidies, fuel station rollout shared between government, auto OEMs and fuel providers

• Assume no Hindenburgs

Max Range

210 miles

23


2010 2025 2040 2055 2070 20850

0.25

0.5

0.75

1

Adoption Fraction

ICE

HFCV

5,700

Base Case

Spatial snapshot

Spatial snapshot

Fuel Stations

(Max Scale =

23,000 stations)GasolineHydrogen

Rollout StationsTotal 785Urban 87%

SubUrban 5%Rural 8%

Market collapses after subsidies end

Hydrogen Fuel

Cell Vehicles on

the Road

Hydrogen Fuel

Stations

Fuel

Demand

+

Fuel

Availability

+

+

HFCV

Sales

Fuel Station

openings

+

Chicken and Egg

Problem

R

24


Some adoption in urban areas, but poor rural, exurb fuel availability leads to market collapse

Sacramento

Fresno

Bakersfield

San

Jose

San Diego

Los Angeles

To Reno

To

Las Vegas

Tahoe

10

5

99

101

Death Valley

Yosemite

Sequoia

Sacramento

Fresno

Bakersfield

San

Jose

San Diego

Los Angeles

To Reno

To

Las Vegas

Tahoe

1010

55

99

101

Death Valley

Yosemite

Sequoia

4040

1515

1515

Adoption,2030 Sacramento

Fresno

Bakersfield

San

Jose

San Diego

Los Angeles

To Reno

To

Las Vegas

Tahoe

10

5

99

101

Death Valley

Yosemite

Sequoia

Sacramento

Fresno

Bakersfield

San

Jose

San Diego

Los Angeles

To Reno

To

Las Vegas

Tahoe

1010

55

99

101

Death Valley

Yosemite

Sequoia

4040

1515

1515

Fuel Availability,2030

25


Topping Off

1. Rigid

• refill at buffer• buffer fixed

2.�Flexible• refill on average at buffer• buffer fixed

3.�Adaptive• refill on average at buffer• buffer adapts to

perceived fuel availability�

We examine the following behavioral assumptions for driver refueling behavior:

What is the impact

on AFV diffusion?

26


AFV Driver “Topping off”Creates Self-Reinforcing

Fuel Shortages

Gas line during 1979 crisis

27


Topping Off

• Topping off is individually rational when fuel availability is uncertain.

• But topping off lowers effective vehicle range, – Increasing trips to and congestion at fuel stations

– Increasing chances of “bank run” dynamics (gas lines and panic as in USA 1979, UK/France 2000)

◊ Result: Lower AFV attractiveness and adoption

28

© Jeroen Struben, MIT (2009) 2010 2025 2040 2055 2070 20850

0.25

0.5

0.75

1

2010 2025 2040 2055 2070 20850

0.25

0.5

0.75

1

Adoption Fraction

ICEHFCV

5,700

Fuel Station Rollout, Rural Emphasis

Spatial snapshot

Spatial snapshot

Fuel Stations

(Max Scale =


Rollout StationsTotal 785

Urban 87%

SubUrban 5%

Rural 8%

Adoption Fraction

ICEHFCV

Fuel Stations

(Max Scale =


5,700

Spatial snapshot

Spatial snapshot

Total 785

Urban 33%

SubUrban 26%

Rural 41%

Rollout Stations

0

0.25

0.5

0.75

1

Base Case

Successful Policy: Subsidize fuel stations in rural areas

29


AFV Adoption Fraction in 2030

Sacramento

Fresno

Bakersfield

San

Jose

San Diego

Los Angeles

To Reno

To

Las Vegas

Tahoe

10

5

99

101

Death Valley

Yosemite

Sequoia

Sacramento

Fresno

Bakersfield

San

Jose

San Diego

Los Angeles

To Reno

To

Las Vegas

Tahoe

1010

55

99

101

Death Valley

Yosemite

Sequoia

4040

1515

1515

Sacramento

Fresno

Bakersfield

San

Jose

San Diego

Los Angeles

To Reno

To

Las Vegas

Tahoe

10

5

99

101

Death Valley

Yosemite

Sequoia

Sacramento

Fresno

Bakersfield

San

Jose

San Diego

Los Angeles

To Reno

To

Las Vegas

Tahoe

1010

55

99

101

Death Valley

Yosemite

Sequoia

4040

1515

1515

Base Case with Rural Fuel StationRollout Emphasis

Base Case

30


2010 2025 2040 2055 2070 20850

0.25

0.5

0.75

1

Successful Diffusion: OverviewInstalled Base Fraction

ICE

HFCV

Vehicles in Use (LDVs)

(Max Scale = 25M Vehicles)ICEHFCV

Fuel Stations

(Max Scale = 23,000 stations)GasolineHydrogen

13.3M

5,700

0

0.25

0.5

0.75

1

Population

(Max Scale = 60M people)

26.9M

2010 2025 2040 2055 2070 20850

0.25

0.5

0.75

1

2010 2025 2040 2055 2070 20850

0.25

0.5

0.75

1

2010 2025 2040 2055 2070 20850

0.25

0.5

0.75

1

31


2010 2025 2040 2055 2070 20850

0.25

0.5

0.75

1

2010 2025 2040 2055 2070 20850

0.25

0.5

0.75

1

2010 2025 2040 2055 2070 20850

0.25

0.5

0.75

1

2010 2025 2040 2055 2070 20850

0.25

0.5

0.75

1

0

0.25

0.5

0.75

1

year

2010 2025 2040 2055 2070 20850

0.25

0.5

0.75

1

Fuel Consumption and Emissions

Fuel Consumption

(Max Scale = 12B gge)GasolineHydrogenBaU

Cumulative GHG Emissions

Well-to-Wheels Per 2010

(Max Scale = 10B T.C02 equiv)Gasoline Supply ChainHydrogen Supply ChainBaU

GHG Emissions per Vehicle

Mile Traveled, Well-to-Wheels

(Max Scale = 6e-4 T.C02/mile)Gasoline AverageHydrogen BaU

32


0

0.25

0.5

0.75

1

2010 2025 2040 2055 2070 20850

0.25

0.5

0.75

1

2010 2025 2040 2055 2070 20850

0.25

0.5

0.75

1

New Vehicle attributes

Vehicle Performance

Vehicle Fuel Efficiency HFCV Tank Capacity (max scale = 30 kg)

20,000 $/vehicle

6 Kg

35,000 $/vehicle

21 m/g

35 m/gge

1

ICE

HFCV

ICE

HFCV

ICE

HFCV

Vehicle Production Cost

33


2010 2025 2040 2055 2070 20850

0.25

0.5

0.75

1

2010 2025 2040 2055 2070 20850

0.25

0.5

0.75

1

Consumer consideration and vehicle sales

HFCV Market share (New)

ICE, Used Car Fraction

HFCV, Used Car Fraction

• Assumes aggressive, sustained marketing effort, demonstration fleet program to build awareness, increase consumer consideration of AFVs

• Surplus used conventional vehicles depress used car prices, delay AFV adoption

• “Cash for Clunkers” can speed AFV market success, stimulate new car sales

Equilibrium used car fraction

HFCV

ICE

Willingness to Consider a Platform Vehicle Sales Overview

34


0 3 6 9 120

10

20

30

40

0 3 6 9 120

10

20

30

40

Experiment 2: Waiting for the oldies to retire: vehicle life strongly affects AFV adoption

• Market penetration increases rapidly with a reduction in the replacement time (dynamics result from social exposure effects, infrastructure, technology learning effects ignored)

• Policies that reduce the effective replacement rate of vehicles have a large impact on moving us over the tipping point: “Cash for Clunkers”

• IMPORTANT: “Clunkers”must not only be deregistered but shredded, with materials recycled. Cannot be sold into used markets in the US or other countries.

25% installed base

15% installed base

Years

to

in

sta

lled

base f

racti

on

Years to replacement

i-pods automobiles

9 Years

4 Years

AFV diffusion vs. average vehicle life

35


Experiment 3: Multiplatform Competition

Wind, Solar

Liquid/gasfossil

Coal

power grid

Biomass

gasoline, dieselethanol,

hydrogen

$ $

Fuel pro

vid

ers

rein

vestm

ent,

le

arn

ing,

exp

ansio

n,

entr

y/e

xit

OE

M r

ein

vestm

ent,

le

arn

ing,

exp

ansio

n,

scale

, scope

$ $ $ $$ $ $

Nuclear Consumer Acceptance

Conventional ICE/gasoline, hybrid

PHEV / PBEV

(P)HFCV

?

?

36


Exploring PHEV and PHFCV Introduction:Basic Assumptions and Data

• Launch in California

• Initial PHEV costs high (realistic), but…– Charge-at-home capability

– Extensive PHEV R&D, marketing. Funding from various parties.

– Multistakeholder commitment to deploying PHEVS

• Calibration from established data sources; fuel supply chain and (public) PHEV data from Ford and Shell Hydrogen– For illustrative purposes: electricity predominantly

derived from fossil inputs (coal, natural gas)

37


Installed Base Dynamics for Multiplatform Competition

Time (Year)2010 2020 2030 2040 2050

PHEV Small

Large

PHFCV Small

Large

25% in stalled base share

10

8

6

4

2

0

2010

Insta

lled

Base [M

Vehic

les]

Population Growth = 1.5%

38


38

Conditions favor successful PHEV diffusion

Attractivenes ofAttractivenes ofAttractivenes ofAttractivenes ofMarket ShareMarket ShareMarket ShareMarket Share

PHEVsPHEVsPHEVsPHEVs----PHEVPHEVPHEVPHEVSalesSalesSalesSales++++

AttractivenessAttractivenessAttractivenessAttractivenessof PHEVsof PHEVsof PHEVsof PHEVs

++++

ImprovementImprovementImprovementImprovementRate of PHEVsRate of PHEVsRate of PHEVsRate of PHEVs

++++

++++

R1R1R1R1

PHEVs use existing fuel infrastructure

Sustained diffusion through self-reinforcing learning, scale effects

39


Platform Installed Base Distribution in 2050 for Different Competition Scenarios

Households Using Vehicle is[r1u,PTPlus] @ 2050

CON S PHEV LSmall Large Small Large Small Large

Insta

lled

Base [M

Vehic

les]

8

4

0

BasePHEVPHFCV_S (250 station rollout)PHFCV_L (500 station rollout)PHEV+PHFCV_L

ICE PHEV PHFCV

40


• Simulations suggest a viable path to widespread, self-sustaining PHEV diffusion– PHEV has advantages relative to e.g. HFVC:

• Fueling infrastructure already deployed; can transition to carbon-neutral via biofuels

• Self-sustaining diffusion much easier to achieve

– Nevertheless, diffusion is slow, consistent with history of other automotive technologies

– Significant investment still required to pass tipping point

• Marketing; consumer acceptance

• Cost reduction and reliability improvement through learning, R&D, scale

– Nontrivial risks:

• Technical (e.g. battery reliability)

• Economic (cost)

• Social (willingness to consider)

PHEV versus (P)HFCV Diffusion: Preliminary Insights

41


Experiment 4: Biofuel Pathways

Wind, Solar

Liquid/gasfossil

Coal

power grid

Bio (Corn)


hydrogen

$ $

Fuel pro

vid

ers

rein

vestm

ent,

le

arn

ing,

exp

ansio

n,

entr

y/e

xit

OE

M r

ein

vestm

ent,

le

arn

ing,

exp

ansio

n,

scale

, scope

$ $ $ $$ $ $



PHEV / PEV

(P)HFCV

Bio (Cellulosic)

42


US Corn Production and the Renewable Fuel Standard Projection 1988-2030

Source: USDA (ERS; 09-09 based on ProExporter Projections); Tradex; Renewable Fuels Association, EISA 2007

88 1990 1995 2000 2005

20

15

10

5

0

Year

Yield [10BU/acre]

Area Harvested [10M acres/year]

Production [1000MBU/year]

Corn Price [$/BU]

Ethanol Production [1000MBU/year]

2010 2015 2020 2025 2030

Projected Ethanol Supply

[100 MBU/year] (RFS, EISA 2007)

Ligno-Cellulosic(and other)

Corn

43


Biofuels Experiment, Main Assumptions• Crop yield grows with 1% per year

• Fleet grows, new conventionals are 100% FFV

– As well as full awareness by their drivers etc..

• Market conditions

– Mandate is pursued by indicated blending level that changes over time. Producers understand demand as indicated by mandate.

– Ethanol production for blending and E85 is market driven, but producers receive 50c/gallon subsidy

– Corn price start in equilibrium at 4.5 $/Bushel

– Initialized and calibration to US corn land use, yield, price and production data

• Demographics:

– Population growth 1.5%

– CA to US extrapolation to examine fuel demand impact

• Timing of 2G Biofuels depends on scenario

– One scenario (2G 2015) involves highly optimistic successful availability for commercialization as of 2015 (after which 2G market performance depends on endogenous scaling up and improvements by learning)Experiment is for qualitative interpretation only!

44


Time (Year)2010 2020 2030 2040 20500

0.05

0.1

0.15

Time (Year)2010 2020 2030 2040 20500

0.05

0.1

0.15

Ethanol (Energy) share in gasoline blend

Mandate Blend Target

Blend share with mandate

Blend share without mandate

Blend share – no producer subsidy

E10 ��

E5 ��

45


Principle feedbacks of corn price dynamics

Corn Demand From

Transportation

Mandated Ethanol

Content in

Transportation

Corn Share in

Biofuel Demand

Corn

Price

Corn

Demand

+

+

CornLanduse

+

+

Marginal

Yield

Corn

Supply

+

-

Corn Demand

Non Transport

-

+

-

Innovation

Efforts

+-

Expected Revenue

per Acre

+

BB

+

Biofuel

Demand

-

+

+

DemandSupply

R

Shortage

+

+

B

Innovation

B

Saturation

46


Time (Year)2010 2020 2030 2040 20500

3

6

9

12

Time (Year)2010 2020 2030 2040 20500

3

6

9

12

Corn Price under Mandate

Mandate

No Mandate$/B

ushel

Counterfactual:Fixed ethanol demand

47


US Corn Demand for Transport Biofuel

Time (Year)2010 2020 2030 2040 20500

3000

6000

9000

12000

Mandate

No Mandate

MB

U/y

ear

48


Time (Year)2010 2020 2030 2040 20500

0.08

0.16

0.24

0.32

E85 Prevalence for Flex FuelsE

85 R

efill

share

under

FF

V d

rivers

Mandate

No Mandate

E85 consumption suppressed under mandate

49


Time (Year)2010 2020 2030 2040 20500

2.5

5

7.5

10

Time (Year)2010 2020 2030 2040 20500

2.5

5

7.5

10

0

4

8

12

16

No Mandate - Corn

Only

Mandate - Corn Only Mandate - Corn/2Gbio0

4

8

12

16

No Mandate - Corn

Only

Mandate - Corn Only Mandate - Corn/2Gbio

Total US Ethanol ProductionE

tha

no

l P

rod

uc

tio

n [Q

ua

ds/y

ea

r]

Mandate

Mandate + 2G

No Mandate

Ethanol Price [c/kwh]

2015

2040

NM M M+2G

2G

Bio

Bre

akth

rough

Actual Projected Total

50


Year2010 2020 2030 2040 20500

10000

20000

30000

Sensitivity Analysis: Biofuel Production

20152025n.a.

g*Eth = .2 [1/yr]; gyield= .01 [1/yr]g*Eth = .3 [1/yr] ; gyield=.01 [1/yr]g*Eth = .3 [1/yr] ; gyield=.02 [1/yr]

2GBio Introduction

Targets and Productivity

Co

rn E

qu

ivale

nt

Pro

du

cti

on

[MB

Ue]

51


Petr

ol

Eth

an

ol (E

85)

No M Mandate (Corn) Mandate (Corn + 2G Bio)

Example Scenario Impact on GHG Emissions (2050)(No Carbon Pricing)

Em

iss

ion

In

ten

sit

y

[gC

O2

/mile

]

1200

900

600

300

0

Petr

ol To

tal

Note: Corn / 2G bio mix in ethanol evolves over time, depending on market conditions

Current Main Sources: - Tilman et al. 2006- Searchinger et al. 2008- Fargione et.al. 2008- EPA 2009

Seq.Credit

Seq.Credit

Seq.Credit

+LandU.Corn

Seq.Credit

+LandU.

All

Seq.Credit

+LandU.

Range, as estimates vary widely

We perform scenario analysis, with different emission intensities throughout the fuel supply chain, using inputs

from multiple expert studies. Note that 2G biofuel estimates in particular vary considerably

52


PHEV100%

0

HFCV

Fuel Choice and Efficiency by Platform

Petro Ethanol Electric Petro Ethanol Electric H2

Mile

s/k

wh

*0.0

1 M

iles/g

CO

2

6

4

2

0

Sh

are

of

tota

l m

iles b

y f

uel

ICE202

0

205

0

Scenario: Seq.Credit, LandU., No 2GBio

53


Experiment 5: The Effect of Oil Shocks on PHEV Diffusion

Wind, Solar

Liquid/gasfossil

Coal

power grid

Biomass


$ $

Fuel pro

vid

ers

rein

vestm

ent,

le

arn

ing,

exp

ansio

n,

entr

y/e

xit

OE

M r

ein

vestm

ent,

le

arn

ing,

exp

ansio

n,

scale

, scope

$ $ $ $$ $ $



PHEV / PEV

$

?

?

54


Hypothesis: Higher oil prices improve relative attractiveness of PHEVs

Attractivenes ofAttractivenes ofAttractivenes ofAttractivenes ofMarket ShareMarket ShareMarket ShareMarket Share

PHEVsPHEVsPHEVsPHEVs----PHEVPHEVPHEVPHEVSalesSalesSalesSales++++


++++


++++

++++

R1R1R1R1

By strengthening self-reinforcing learning, and scale effects

Attractivenes ofAttractivenes ofAttractivenes ofAttractivenes of

----Oil Price

+

55


Year

Rela

tive I

nsta

lled

Base

PHEV Diffusion Patterns

2000 2010 2020 2030 20400

0.25

0.5

0.75

1

1.25

Conventional ICEPHEV

Year

Rela

tive I

nsta

lled

Base


2000 2010 2020 2030 20400

0.25

0.5

0.75

1

1.25

Conventional ICEPHEV

Oil Shock Scenario

Oil Shock Scenario

Actual Effect: Negligible!

56


Year

New

Veh

icle

Sale

s S

hare


2000 2010 2020 2030 20400

0.2

0.4

0.6

0.8

1

Higher fuel prices lead consumers to choose smaller, more efficient vehicles

Large ConventionalVehicles

Small Conventional Vehicles

Oil Shock Scenario

75% Oil price increase120$/barrel

And there is more…

57


xxxx

OilOilOilOilPricePricePricePrice

Attractivenes ofAttractivenes ofAttractivenes ofAttractivenes ofConventionalConventionalConventionalConventionalVehiclesVehiclesVehiclesVehicles

Market ShareMarket ShareMarket ShareMarket SharePHEVsPHEVsPHEVsPHEVs----

PHEVPHEVPHEVPHEVSalesSalesSalesSales

++++Cost ofCost ofCost ofCost ofDrivingDrivingDrivingDriving

PetroleumPetroleumPetroleumPetroleum

++++ ++++


++++


++++

++++

R1R1R1R1

World OilWorld OilWorld OilWorld OilDemandDemandDemandDemand

World OilWorld OilWorld OilWorld OilSupplySupplySupplySupply

----

++++

Demand fromDemand fromDemand fromDemand fromEmergingEmergingEmergingEmergingEconomiesEconomiesEconomiesEconomies

++++

xxxx





++++

ConsumerConsumerConsumerConsumerChoice forChoice forChoice forChoice for

Efficient CarsEfficient CarsEfficient CarsEfficient Cars

Cost ofCost ofCost ofCost ofDrivingDrivingDrivingDriving


++++ ++++


++++

Efficiency ofEfficiency ofEfficiency ofEfficiency ofConventionalConventionalConventionalConventionalCars on RoadCars on RoadCars on RoadCars on Road

B1B1B1B1

++++++++

----


++++

++++

R1R1R1R1



----

++++


++++

xxxx





++++





++++ ++++

Efficiency ofEfficiency ofEfficiency ofEfficiency ofConventionalConventionalConventionalConventionalCars on MarketCars on MarketCars on MarketCars on Market

EfficientEfficientEfficientEfficientCar ModelsCar ModelsCar ModelsCar ModelsOfferedOfferedOfferedOffered

++++


++++


R2R2R2R2

B1B1B1B1 B2B2B2B2

++++++++

++++

++++++++

----


++++

++++

R1R1R1R1



----

++++


++++

xxxx





++++





++++ ++++


AutomotiveAutomotiveAutomotiveAutomotiveR&D in FuelR&D in FuelR&D in FuelR&D in FuelEfficiencyEfficiencyEfficiencyEfficiency

++++

++++


++++


++++


B3B3B3B3

R2R2R2R2

B1B1B1B1 B2B2B2B2

++++++++

++++

++++++++

----


++++

++++

R1R1R1R1



----

++++


++++

xxxx





++++





++++ ++++



++++

++++


++++

VehiclesVehiclesVehiclesVehiclesMilesMilesMilesMiles

TraveledTraveledTraveledTraveled

----

ReplacementReplacementReplacementReplacementRatesRatesRatesRates


++++


B3B3B3B3

R2R2R2R2

++++

++++

B1B1B1B1 B2B2B2B2

++++++++

++++

++++++++

----


++++

++++

R1R1R1R1



----

++++


++++

xxxx





++++





++++ ++++



++++

++++


++++

Cost of DrivingCost of DrivingCost of DrivingCost of Drivingon Electricityon Electricityon Electricityon Electricity

----

ElectricityElectricityElectricityElectricityPricePricePricePrice

++++

Cost ofCost ofCost ofCost ofProducing &Producing &Producing &Producing &DistributingDistributingDistributingDistributingElectricityElectricityElectricityElectricity

++++

++++



----



++++


B3B3B3B3

R2R2R2R2

++++

++++

B1B1B1B1 B2B2B2B2

++++++++

++++

++++++++

----


++++

++++

R1R1R1R1



----

++++


++++

xxxx





++++





++++ ++++



++++

++++


++++

NonTransportNonTransportNonTransportNonTransportDemand forDemand forDemand forDemand forElectricityElectricityElectricityElectricity


----


++++

++++

++++


++++

++++



----



++++


B3B3B3B3

R2R2R2R2

++++

++++

B1B1B1B1 B2B2B2B2

++++++++

++++

++++++++

----


++++

++++

R1R1R1R1



----

++++


++++

----

----

B7B7B7B7

xxxx





++++





++++ ++++



++++

++++


++++

NonTransportNonTransportNonTransportNonTransportDemand forDemand forDemand forDemand forElectricityElectricityElectricityElectricity


----


++++

++++

++++


++++

++++



----



++++


B3B3B3B3

R2R2R2R2

++++

++++

B1B1B1B1

B6B6B6B6

PetroleumPetroleumPetroleumPetroleumDemand fromDemand fromDemand fromDemand fromTransportationTransportationTransportationTransportation

++++

----

B2B2B2B2

B5B5B5B5

++++++++

++++

++++++++

----

PetroleumPetroleumPetroleumPetroleumCars on theCars on theCars on theCars on the

RoadRoadRoadRoad

----

++++

B9B9B9B9


++++

++++

R1R1R1R1



----

++++


++++ ++++

----

----

B7B7B7B7

Consumer Responses

Auto OEM Responses

Fuel Supply Chain Responses

PHEV Market Formation:

Overcoming an Established and Resilient Transportation System

-

58


Behavioral Responses Have Been Observed

Data: US Census; http://www.greencarcongress.com/, accessed June 2008; Transportation Energy Data Book, Ed. 26-2007 Table 4.6, Autodata and Ward’s

Vehicle Weight & Performance

US Vehicle Distance TraveledRoad Vehicle Miles (Trillions/year)

3

2.5

2

1.5

0 to 60(sec.)

1983 Year 2008

Car and Light Truck Sales

Car

Truck

Annual Monthly

1980 Year 2008

Weight(.lbs)

59


Vehicle average fleet efficiency responds to real prices with a very long delay. Prices jump in 1973, but efficiency remains nearly constant until 1980. Prices fall to about $20/bbl by 1986 and efficiency gains stall in 1992 as automakers use technical improvements to boost performance instead of mileage, and as consumers switch to larger vehicles and SUVs.

0

20

40

60

80

1950 1960 1970 1980 1990 2000 2010

2007 $

/bb

lReal Petroleum Price, 2007$/bbl

10

12

14

16

18

20

1950 1960 1970 1980 1990 2000 2010

US Motor Vehicle Fleet Efficiency

Mil

es/G

all

on

Source: Energy Information Administrationwww.eia.doe.gov/emeu/aer/txt/ptb0208.html

60


Experiment 6: Varying CO2 Prices

Wind, Solar

Liquid/gasfossil

Coal

power grid

Biomass


hydrogen

$ $

Fuel pro

vid

ers

rein

vestm

ent,

le

arn

ing,

exp

ansio

n,

entr

y/e

xit

OE

M r

ein

vestm

ent,

le

arn

ing,

exp

ansio

n,

scale

, scope

$ $ $ $$ $ $



PHEV / PBEV

(P)HFCV

?

?

61


Automotive Responses to CO2 Price

Time (Year)2010 2020 2030 2040 20500

0.25

0.5

0.75

1

Small

Large

Relative Fuel Economy New Vehicles

CO2 Price200 $/tCO2

100500

The actual average fuel economy depends on consumer choice and fleet replacement

2

1.5

1

0.5

0

62


CON S

CON L

PHEV S

PHEV L

PHFCV S

PHFCV L

$50/tCO2 carbon tax has moderate impact on platform installed base shares

I. Mand+2GBio+AFVsII. As I +CO2_50, no 2GBioIII. As I +CO2_50IV. As I +CO2_100

Vehicle Installed Base in 2050 (Millions)

12

8

4

0

Note: PHFCV’s do best in absence of 2Gbiofuels

63


CO2 Emissions im[Fuel,Trans] @ 2025

400

300

200

100

0

CO2 Emissions from transportation under varying CO2 Scenarios

EL ETH

I. Mand+2GBio+AFVs+SeqCr/LanduCornII. As I +CO2_50III. As II +Hydro-El (California EL Mix)IV. As II + Late TaxV. As I +CO2_100VI. As I, no 2GBio

CO2 Emissions (California)

2020 2050

EL Petro H2 ETH EL Petro H2 ETH

[MTCO2e/year]

411 258 342 247 325 433

64


Other Analysis Performed, in Progress, and Planned

• Extensive model testing• Further PHEV analysis

– Base scenario using: CAFE, (CA) renewable fuel standards, Waxman-Markey, etc..– Sensitivity to technical, economic, behavioral uncertainties (e.g. CCS, biofuel

commercialization timing)– Policy analysis

• Competition with and interactions between PHEV & other AFVs– Alternative PHEV technologies– Conventional hybrids– Biofuels/biodiesel– HFCV, H-ICE– Pure electric (e.g. BetterPlace)

• Fuel supply chain scenarios:– Biofuels: 2nd Generation (cellulosic; waste inputs)– H2 from sustainable sources (e.g. Nocera process, biofuels)– Electricity peak and base load, battery supply chains, storage

• C-Price and CCS Scenarios• Interactions among all items above• Other regions

65


AFV Diffusion:Counterintuitive Dynamics

• Focusing initial fuel station rollout on urban areas, where initial AFV demand likely highest, leads to urban focus, market failure.

• More costly exurb/rural focus builds sustainable, profitable AFV and alt fuel market, with greater urban market share, larger NPV for all key actors (Auto OEMs, fuel providers, consumers, government and environment).

• A more efficient AFV can slow or prevents adoption due to negative impact of lower fuel demand on alt fuel profitability and infrastructure investment.

• Plug-in Hybrids not vulnerable to infrastructure dynamics; diffusion more rapid and durable, assuming technical risks overcome.

• Success rapidly reduces gov’t fuel excise tax revenues; fuel tax must rise over time to maintain revenues (and compensate for drop in world oil price induced by lower consumption).

• Faster AFV sales leads to surplus used conv. vehicles. Low used car prices limit AFV diffusion. Early decommissioning of conventional cars (Cash for Clunkers) a high-leverage policy.

• Others…

66


Summary: Transition strategies for alternative transportation fuels and vehicles

• Understanding AFV diffusion requires sensitivity to– Technical, economic and socio-behavioral factors

– Understanding counterintuitive dynamics

– Worse-before better dynamics

• Effective policy making and market success requires coordinated/shared understanding, long-term commitment– Value and challenge of coordination among key stakeholders

• Auto OEMs (entrants)

• Fuel providers, electric utilities, power producers

• Fed, state governments

• The modeling process is designed to enable learning with, and coordinate across different market players

Documents

Transition Strategies for Alternative Transportation Fuels ...web.mit.edu/jjrs/www/AETTPresentations/StrubenAETT200906.pdf · Use of petroleum product by sector in the US, 1900 –