Transition to New Technologies (TNT) Program

Nebojsa Nakicenovic Deputy Director General and Deputy CEO International Institute for Applied Systems Analysis Professor Emeritus of Energy Economics Vienna University of Technology

Why study Technology?

2

• Main mediator between humans and the environment

• Main source of productivity and welfare growth (development)

• Policy interest: “man-made resource”, but… – Change costly (investments!) – High uncertainty

(innovation and diffusion) – Large inertia for major transformations

(lock-in, path dependency) – Slow rates of change (systems/infrastructures)

Transitions to New Technologies

3

• Point of departure Scenarios/impacts of diffusion of new technologies (clusters): ICT, transport, energy, and impact on environment (e.g. climate).

• Strategic Goal Furthering the understanding of patterns, dynamics, and constraints of

technological change, and its drivers for global sustainability conditions. • Research Goal Focusing on the systemic aspects, understanding the evolution of entire

technology systems. • Research

- Drivers beyond “black box” - Models (uncertainty, increasing returns, agents) - Heterogeneity (time, space) - Impacts (scenarios) - Synthesis (metastudies)

Networks for Policy Relevant Research • Major International Assessments:

– Global Energy Assessment (GEA) (coordination, CLAs for 4 chapters) – IPCC AR5 (4 chapters, synthesis report)

• Global Fora: – Sustainable Energy For All (SE4ALL) TNT provides methodological frameworks, policy advice on technology strategies, roadmaps, urbanization patterns, and national scale modeling to support the SE4ALL 2030 goals:

• Universal access to modern energy • Doubling energy efficiency improvement rates • Doubling the share of renewable energy

– International Council for Science (ICSU)

– Future Earth Initiative

– Sustainable Development Solutions Network (SDSN)

– Global Carbon Project (GCP)

– German Government’s Advisory Council on Global Change (WBGU)

4

Data: UN, ITU, World Bank, 2010

Global Access to Technologies (Lorenz Curves)

5

Global – Historic Primary Energy Transitions (changeover time Δt: 80-130 years)

Source: Chapter 24: GEA, 2012

traditional biomass

coal

modern fuels:oil, gas,electricity

traditional biomass

coal

modern fuels:oil, gas,electricity

Begin of energy policy focus: Δts >2000 yrs

Δt -130 yrs

Δt -80 yrs

Δt +90 yrs

Δt +130 yrs

6

TNT Niche: Technology & Innovation Systems • Transformative change needs systemic understanding & policies on:

all innovation phases, processes, and energy systems components: – R&D, niche markets, diffusion, obsolescence – learning, actors/institutions, resources, technology (hard+soft-ware) …and yet…

Important biases at all stages: – R&D: supply side bias (nuclear, fossil) – Niche markets: supply side bias (solar/wind) – Diffusion: huge distortions via fossil fuel subsidies – Obsolescence: “grandfathering” of old/”dirty”

• New framework for analysis and GEA policy design criteria: Energy Technology Innovation Systems (ETIS) • Modeling endogenous evolution of technology systems • ABM of technological complexity • Technology “meta-studies”

(metrics & determinants of change, past and future scenarios) 7

Market Formation R,D&D

(public $) Diffusion Support

Social Rates of Return

Analysis & Modelling

Future Needs

supply : end-use (relative effort)

ACTO

RS &

INST

ITUT

IONS

TECHNOLOGY CHARACTERISTICS

KNOWLEDGE

RESOURCES

learning generation sh

ared

expe

ctatio

ns

entre

pren

eurs

/ risk

tak

ing cost

resource inputs

public policy & leverage

performance

key

Roadmaps & Portfolios

Technology Collaborations

Learning Effects

Directable (Activities)

Non-Directable (Outputs) 8

The GEA ETIS Framework

Source: Chapter 24: GEA, 2012

CLIMATE MITIGATION

Current Public ETIS Policy Leverage/Focus (policy-induced resource mobilization, billion US$2005)

Source: Wilson et al. Nature Climate Change, 2012 9

Criteria for Case Study Selection

Them

atic

/ M

eta-

anal

ytic

Supp

ly T

echn

olog

ies

End-

Use

Tec

hnol

ogie

s

Sing

le C

onte

xt

Com

para

tive

Cont

ext

Curr

ent

Hist

oric

al

Deve

lope

d Co

untr

y(s)

Deve

lopi

ng C

ount

ry(s

)

Influ

entia

l Pub

lic P

olic

y

Syst

emic

1 Energy Transitions X X X X X X X 2 Technology Diffusion X X X X X X X 3 Assessment Metrics X 4 Technology Portfolios X X X X

Know

ledg

e 5 Solar Water Heaters X X X X X 6 Heat Pumps X X X X X X 5 Knowledge Depreciation X X X 6 Nuclear Power (France) X X X X

Adop

tion

& U

se

7 Solar Thermal Electricity (US) X X X X X 8 Vehicle Efficiency X X X X X X 9 Hybrid Cars X X X X X X 10 Solar Photovoltaics X X X X X X

Acto

rs &

In

stitu

tions

11 Wind Power X X X X X X X 12 End-Use Efficiency (Japan) X X X X X X 14 Rural Solar (Kenya) X X X X X 15 Synfuels (US) X X X X

Reso

urce

s 13 Ethanol (Brazil) X X X X X X X 18 Global Financial Resources X X X X X X X X 19 R,D&D Investments (Emerging Economies) X X X X X X X 20 Global End-Use Investments X X X X X X

ETIS Case Studies

10 Source: Chapter 24: GEA, 2012

World Energy Technology Innovation Investments (Billion $)

innovation market diffusion(RD&D) formation

End-use & efficiency >>8 5 300-3500Fossil fuel supply >12 >>2 200-550Nuclear >10 0 3-8Renewables >12 ~20 >20Electricity (Gen+T&D) >>1 ~100 450-520Other* >>4 <15 n.a.Total >50 <150 1000 - <5000 non-OECD ~20 ~30 ~400 - ~1500 non-OECD share >40% <20% 40% - 30%

* hydrogen, fuel cells, other power & storage technologies, basic energy research

Source: Chapter 24: GEA, 2012 11

Knowledge Depreciation Rates (% per year) Degree of technological obsolescence (rate of innovation)

Deg

ree

of k

now

ledg

e st

ock

turn

over

(p

olic

y &

hum

an c

apita

l vol

atili

ty) PV Japan:

30% Wind US: 10%

Engineering designs US:

<5%

Service industries:

95%

Aircraft, Liberty ships manufct. US:

40%

Chemicals, Drugs: 15-20%

Computers: 32%

Electrical, Machinery:

32-36% Miscell. >20%

OECD nuclear R&D:

10 – 40%

France breeder reactors: 50-60%

High

High

Low

12 empirical studies reviewed GEA Chap 24 (2012) and modeled R&D deprecation in US manufacturing (Hall, 2007)

Post Fossil Energy Supply Technologies Cost Trends

Source: Grubler/Wilson: CUP, 2013 13

Learning rates and cumulative experience (# of units produced/sold) for energy technologies

Source: Nature CC, 2012, S1

category technology data for: cumulative production (units) # exp period rate

energy Transitors World >1 10^18 1960-2010 40 end-use DRAMs World >1 10^11 1975-2005 16 - 24 Automobiles World >2 10^9 1900-2005 9 - 14

Washing machines World >2 10^9 1965-2008 33 ± 9 Refrigerators World >2 10^9 1964-2008 9 ± 4 Dishwashers World >6 10^8 1968-2007 27 ± 7 Freezers (upright) World >6 10^8 1970-2003 10 ± 5 Freezers (chest) World >5 10^8 1970-1998 8 ± 2 Dryers World >3 10^8 1969-2003 28 ± 7 Hand-held calculators US >4 10^8 early 1970s 30 CF light bulbs US >4 10^8 1992-1998 16 A/C & heat pumps US >1 10^8 1972-2009 18 ± 1 Air furnaces US >1 10^8 1953-2009 31 ± 3 Solar hot water heaters US >1 10^6 1974-2003 -3

average for end-use technologies 10^9 20 energy supply PV modules World >1 10^10 1975-2009 18-24

Wind turbines World >1 10^5 1975-2009 10-17 Heat pumps S, CH <1 10^5 1982-2008 2 - 21 Gas turbines World >4 10^4 1958-1980 10-13 Pulverized coal boilers World >6 10^3 1940-2000 6 Hypropower plants OECD ~5 10^3 1975-1993 1 Nuclear reactors US, France <1 10^3 1971-2000 -20 - -47 Ethanol Brazil <1 10^3 1975-2009 21 Coal power plants OECD <1 10^3 1975-1993 8 Coal power plants US <1 10^3 1950-1982 1 - 6 Gas pipelines US <1 10^3 1984-1997 4 Gas combined cycles OECD <1 10^3 1981-1997 10 Hydrogen production (SRM) World >1 10^2 1980-2005 27 LNG production World >1 10^2 1980-2005 14

average for suppy technologies 8 average for supply, excluding nuclear 12 10^4

learning

14

Market Size (normalized index, Core markets) vs Diffusion Speed (Δt) of Energy Technologies

Source: Wilson, YSSP, 2008. E-Bikes & Cell Phones courtesy of IIASA Post-Doc Dr. Bento

CELL PHONES

15

Scaling patterns Past and Scenarios (IIASA GGI) (8 Scenarios: A2r/B1/B2 * base/670/480)

• Scenarios more conservative compared to past

• Closer relationship for “lumpy” power techs

• Method adopted in IAM community for sce4nario validation

1.E+00

1.E+01

1.E+02

1.E+03

1.E+04

1.E+05

1.E+06

1.E+07

0 25 50 75 100 125

Nor

mal

ised

K (i

ndex

)

Δt (yrs) of cumulative total capacity

Cumulative Total Capacity (OECD): normalised K vs ΔtALL TECHS: HISTORICAL & SCENARIOS - semi-log

SCENARIOS (All Techs)

Historical (Core)

Historical (Core) -POWER only - no WIND

Expon. (SCENARIOS (All Techs))

Expon. (Historical (Core))

Expon. (Historical (Core) - POWER only -no WIND)

Source: Wilson et al., Climatic Change, 2013 16

Cumulative Experience and Learning: The Importance of “granularity”

A

B

C

D

E

F

GH

IJ

KL

M

N

1

23

4

56

7

8

91011

12

-50.0

-40.0

-30.0

-20.0

-10.0

0.0

10.0

20.0

30.0

40.0

50.0

1.E+00 1.E+03 1.E+06 1.E+09 1.E+12 1.E+15 1.E+18

Lear

ning

rate

(% c

ost c

hang

e pe

r dou

blin

g)

Cumulative # of units produced

A TransitorsB DRAMsC AutomobilesD Washing machinesE RefrigeratorsF DishwashersG Freezers (upright)H Freezers (chest)I DryersJ CalculatorsK CF light bulbsL A/C & heat pumpsM Air furnacesN Solar hot water heaters

1 PV modules2 Wind turbines3 Heat pumps4 Gas turbines5 Pulverized coal boilers6 Hypropower plants7 Nuclear reactors8 Ethanol9 Coal power plants

10 Coal power plants11 Gas pipelines12 Gas combined cycles

Mean of “granular” end use technologies: LR=20% CumProd= 10e9

Mean of “lumpy” supply technologies: LR=10% CumProd= 10e4

Source: Wilson et al, Nature CC, 2012

TNT Collaboration - Publications

18

TNT Resources Pioneering: • Uncertainty IR (Gritsevskyi/Grubler) • Supercomputer runs for tech uncertainty in CC (Nakicenovic/Gritsevskyi) • Agent-based modeling of tech complexity (Ma/Grubler/Nakicenovic/Brian Arthur)

Collaboration/spin-offs: • Stochastic uncertainty modeling (w. ENE) • MCA multiple objectives (w. ENE) • Tech change in IAMs (w. ENE, RITE) • Web access/open source models (LSM)

On-line scenario and technology data bases • Scenario DBs IIASA GGI, GEA, IPCC-RCPs-SSPs (with ENE) http://www.iiasa.ac.at/web-apps/ggi/GgiDb http://www.iiasa.ac.at/web-apps/tnt/RcpDb • Energy & CO2 inventories Database

http://www.iiasa.ac.at/Research/TNT/WEB/Publications/Energy_Carbon_DataBase • Scaling Dynamics of Energy Technologies (SD-ET) on novel historical technology data

http://www.iiasa.ac.at/~gruebler/data.htm http://www.iiasa.ac.at/Research/TNT/WEB/Publications/Scaling_Dynamics_of_Energy_Technologies

• Primary Final and Useful Energy Database (PFUDB) http://www.iiasa.ac.at/web/home/research/researchPrograms/TransitionstoNewTechnologies/PFUDB.en.html

Models: • LSM2 Technological Growth & Substitution http://www.iiasa.ac.at/Research/TNT/WEB/Software/LSM2

19

The Evolution of Technological Complexity

• Agent-based simulation model of global energy system since 1800

• Random walk model of invention discovery and stochastic combination with other technologies into energy chains and systems

• Evolutionary selection environment - uncertain increasing returns - market share gains f(rel. advantage) - externalities (stochastic C-tax)

• Evolution of complexity is function of learning rate and innovation impatience

• Complexity lock-in requires “gales of creative destruction”

20