long-term strategy toward deep emission reductions under … · 2018-02-13 · long-term strategy...

Long-term strategy toward deep emission reductions under several

kinds of uncertainties

International Symposium−Prospect of Decarbonization after the Paris Agreement

February 8, 2018; University of Tokyo

Keigo AkimotoSystems Analysis GroupResearch Institute of Innovative Technology for the Earth (RITE)

1. Required long-term goals and uncertainties in short/mid-term pathway

2. Mitigation costs − the gaps between the idealistic mitigation costs and real costs

3. Climate change mitigation measures under different socioeconomic conditions

4. Co-benefit and trade-off between climate change and other sustainable development goals

5. Innovations and emission pathways

6. Conclusions

Contents2

1. Required long-term goals and uncertainties

in short/mid-term pathway

4

Relationship between cumulative CO2 emissions and temperature rise

Source) Synthesis report of IPCC AR5

- Approximately linear relationship between cumulative CO2 emissions and temperature rise can be observed.- Nearly net zero CO2 emissions are necessary for the stabilization of global temperature at any level.

Temperature response to emissions in 2010; the responses are normalized by the amount of contribution of CO2 emission after 100 years past

History of climate sensitivity judgment by IPCC and the sensitivity employed in the scenario assessments of the

IPCC WG3 AR5

♦ The equilibrium climate sensitivity, which corresponds to global mean temperature increase in equilibrium when GHG concentration doubles, is still greatly uncertain.

♦ AR5 WG1 judged the likely range of climate sensitivity to be 1.5−4.5 °C, in which the bottom range was changed to a smaller number than that in the AR4, based not only on CMIP5 (AOGCM) results but also other study results.

♦ AR5 WG3 adopted the climate sensitivity of AR4, which has the likely range of 2.0−4.5 °C with the best estimate of 3.0 °C, for temperature rise estimates of long-term emission scenarios.

Equilibrium climate sensitivityLikely range (“best estimate” or “most likely value”)

Before IPCC WG1 AR4 1.5−4.5°C (2.5°C)

IPCC WG1 AR4 2.0−4.5°C (3.0°C)

IPCC WG1 AR5 1.5−4.5°C (no consensus)

Global mean temperature estimations for the long-term scenarios in the IPCC WG3 AR5 (employing MAGICC)

2.0ｰ4.5°C（3.0°C）[Based on the AR4]

5

[The related descriptions of the SPM of WG1 AR5]Likely in the range 1.5 °C to 4.5 °C (high confidence)Extremely unlikely less than 1 °C (high confidence)Very unlikely greater than 6 °C (medium confidence)No best estimate for equilibrium climate sensitivity can now be given because of a lack of agreement on values across assessed lines of

evidence and studies.

Same “likely” range

Source) Interagency working group on social cost of carbon, 2016

6Social Cost of Carbon (SCC)

- Social cost of carbon is the marginal damage costs of CO2 emissions.- The estimation methods are very debatable, and the estimated distributions of the damage costs vary widely depending on the estimated models, climate sensitivity, discount rate etc. Therefore, it is not easy to determine the optimal temperature level.

-10

0

10

20

30

40

50

2010 2060 2110 2160 2210 2260

CO

2em

issi

onis

[GtC

O2/

yr]

Business as Usual (BaU)

Submitted NDCs

+2℃ stabilization under climate sensitivity of 2.5℃ (No exceedance of 580 ppm)+2℃ stabilization under climate sensitivity of 3.0℃ (No exceedance of 500 ppm)450 ppm CO2eq stabilization

Submitted NDCs23002050 2100

7

Global CO2 emission profiles toward 2300for the 2 °C targets

- The global CO2 emissions should be nearly zero for a long-term period in the far future in any pathway to achieve temperature stabilization.- On the other hand, the allowable global CO2 emissions toward the middle of this century have a wide range according to the uncertainties in climate sensitivity (or achieving probability) even when the temperature target level is determined as a 2 °C. We should use this flexibility to develop several kinds of innovative technologies and societies.

Estimated by RITE using MAGICC and DNE21+

(the middle socioeconomic scenario)

0

10

20

30

40

50

60

70

1990 2010 2030 2050 2070 2090

GH

G e

mis

sion

s[G

tCO

2eq/

yr]

Business as Usual (BaU)

Submitted NDCs

+2℃ stabilization under climate sensitivity of 2.5℃(No exceedance of 580 ppm)+2℃ stabilization under climate sensitivity of 3.0℃(No exceedance of 500 ppm)450 ppm CO2eq stabilization

Historical

Submitted NDCs

2100

+14%

-42%

-74%

8

Global GHG emission profiles toward 2100 for the 2 °C target

- The corresponding GHG emission trajectories for the 2 °C target vary widely particularly in 2050. - There are large gaps between the expected emissions under the submitted NDCs and the 450 ppm CO2eq pathway.

Estimated by RITE using MAGICC, DNE21+ and non-CO2 GHG models

2. Mitigation costs − the gaps between the idealistic

mitigation costs and real costs

10Huge costs are estimated for achieving the 2 °C target

- According to the IPCC AR5, the CO2 marginal abatement costs (carbon prices) for the 430-530 ppm CO2eq (which are consistent with the 2 °C target) are about 1000-3000 $/tCO2 (25-75 percentile) and 150-8000 $/tCO2 (full range) in 2100.- About 25% of the analyzed scenarios estimate global GDP losses of over 10%.- The feasibility of such scenarios should be carefully examined in terms of various constraints in the real world.

Corresponding to 2 °C target Corresponding to 2 °C target

Source) IPCC WG3 AR5

0

0

0

0

1

4

12

14

27

33

54

58

70

85

95

144

166

210

378

380

0 50 100 150 200 250 300 350 400

ChinaUkraine

IndiaTurkey

South AfricaRussia

BelarusKazakhstan

MexicoAustraliaThailand

East Europe (Non-EU member)Norway

United StatesNew Zealand

KoreaCanada

EU28Japan

Switzerland

CO2 marginal abatement cost ($/tCO2)

11

CO2 marginal abatement costs of the NDCs

Source: J. Aldy et al., Nature Climate Change, 2016

Source: K. Akimoto et al., Evol. Inst. Econ. Rev., 2016

2030 (2025 for the U.S.)【World GDP loss due to mitigation】

NDCs:0.38%; the global least cost：0.06%The least cost (equal marginal abatement costs)：6$/tCO2

Average of 2025-2030

- The estimated marginal abatement costs of NDCs are largely different among countries, and the mitigation costs are much larger than those under the least cost measures due to such large difference in marginal abatement costs.- The difference will induce carbon leakages, and the leakages will reduce the effectiveness of global emission reductions.

12

CO2 marginal abatement cost for the U.S, EU and Japan considering several kinds of policy constraints

0

100

200

300

400

500

CO

2m

argi

nal a

bate

men

t cos

t ($/

CO

2)

I. US II. EU III. Japan

I-a

III-a

I-a: -26%; the least costI-b: -28%; the least costI-c: -26%; power sector

according to CPPI-d: -28%; power sector

according to CPP

I-b

II-a: the least costII-b: Brexit (-40% for UK)II-c: splitting into ETS and

non-ETS sectors

III-a: the least cost under nuclear of maximum 20%

III-b: the least cost undernuclear of maximum 15%

III-c: following the NDC including the energy mix (nuclear of 20%)

III-d: following the NDC including the energy mixbut nuclear of 15%

I-c

I-d

II-a

III-cIII-d

III-bII-b

II-c

Source: estimated by RITE DNE21+

- It is not easy to achieve the least cost measures because there are several kinds of social and political constraints in each nation.- The mitigation costs constrained by other policies can be much higher than those under the least cost measures.

* CPP: Clean Power Plan

3. Climate change mitigation measures under different

socioeconomic conditions

14Overview of Shared Socioeconomic Pathways (SSPs)

Tech. improve: low;Population: low;GDP: low

Tech. improve.：high; Public acceptability of large-scale tech.: low;Population: low; GDP: high

Governance: low;Price distribution of fossil fuel energy prices: big

Fossil fuel price: low;Fossil fuel resources: high;GDP: very high

15Relationship between SSPs and RCPs

K.Riahi et al., Global Environmental Change 42 (2017) 153–168

RC

P

Note 1) 2.6 W/m2 corresponds to below 2 °C in 2100 with >66% achieving probability; 3.4 W/m2 corresponds to below 2 °C in 2100 with >50% probability, and 4.5 W/m2 corresponds to below about 2.5 °C with >50% probability.Note 2) Carbon prices are shown as the converted values in 2010 by employing discount rate of 5%/yr. The carbon price of 20 $/tCO2 as the 2010 value corresponds to about 1800 $/tCO2 for 2100.

16Global CO2 emissions in Baseline

- Baseline emissions are very different depending on the future socioeconomic conditions including technology improvements.

Estimated by RITE DNE21+ model

0

20

40

60

80

100

120

1990 2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100

Ener

gy-re

late

d C

O2

emis

sion

[GtC

O2/

yr] Historical SSP1 SSP2 SSP5

17

Marginal CO2 abatement costs (Carbon prices) for the 2 °C target

Source) estimated by RITE DNE21+

SSP2 (Middle of the Road) SSP1 (Sustainability)+2°C stab. under climate sensitivity of 2.5°C

+2°C stab. under climate sensitivity of 3.0°C

450 ppm CO2eq stab. (climate sensitivity of 3.4°C)



450 ppm CO2eq stab. (climate sensitivity of 3.4°C)

2050 12 135 604 14 117 518

2100 408 427 457 134 140 143

Unit: $/tCO2 (real price); Uniform carbon prices among all nations are assumed.

- The marginal abatement costs (carbon prices) for the 2 °C target are huge even under the global least cost measures (uniform carbon prices) except in the case of low climate sensitivity (2.5 °C) and by 2050.

- The carbon price in SSP1 in which energy demands in the end-use sectors are much smaller than in SSP2 is much lower than that in SSP2.

- Technological and social innovations are definitely required for the 2 °C target to be achieved in harmony with other SDGs. (Newly emerging technologies such as AI, IoTetc. will induce social changes which may lower the energy demand.)

SSP: “Shared Socioeconomic Pathways”

0

5000

10000

15000

20000

25000

30000

SSP1 SSP2 SSP5 SSP1 SSP2 SSP5 SSP1 SSP2 SSP5 SSP1 SSP2 SSP5 SSP1 SSP2 SSP5 SSP1 SSP2 SSP5

[3] +2℃stabilization under

C.S. of 2.5℃

[1] 450 ppm CO2eqstabilization


C.S. of 2.5℃



C.S. of 2.5℃


2010 2030 2050 2100

Prim

ary

ener

gy s

uppl

y （M

toe/

yr）

Coal (w/o CCS)

Coal (w/ CCS)

Oil (w/o CCS)

Oil (w/ CCS)

Gas (w/o CCS)

Gas (w/ CCS)

Biomass (w/o CCS)

Biomass (w/ CCS)

Hydro + Geothermal

Nuclear

Wind power

Solar PV

18Global primary energy supply

- The energy supply is very different in 2050 according to the uncertainty in the climate sensitivity and different socioeconomic scenarios.- The total amount of energy supply in the SSP1 world in 2100 is much smaller than that in the SSP2 and SSP5.


0

20000

40000

60000

80000

100000

120000

140000


[3] +2℃ stabilization under C.S. of 2.5℃






2010 2030 2050 2100

Elec

tric

ity g

ener

atio

n （TW

h/yr）

Coal (w/o CCS)

Coal (w/ CCS)

Oil (w/o CCS)

Oil (w/ CCS)

Gas (w/o CCS)

Gas (w/ CCS)

Biomass (w/o CCS)

Biomass (w/ CCS)

Hydro + Geothermal

Nuclear

Wind power

Solar PV

Solar PV (Water electrolysisw/o grid connection)Hydrogen (H2 from otherproduction proccess)Hydrogen (H2 from waterelectrolysis)

19Global electricity generation

- CO2 emissions from power sector in most of the scenarios for the 2 °C goals are nearly zero.- The total amounts of electricity for the 2 °C target will increase with deeper emission reductions due to substitution of fossil fuel use in other sectors.


0

5

10

15

20

25

30

35

40

45

SSP1

SSP2

SSP5

SSP1

SSP2

SSP5

SSP1

SSP2

SSP5

SSP1

SSP2

SSP5

SSP1

SSP2

SSP5

SSP1

SSP2

SSP5

[3] +2℃stabilization

under C.S. of 2.5℃

[1] 450 ppmCO2eq

stabilization



[1] 450 ppmCO2eq

stabilization



[1] 450 ppmCO2eq

stabilization

2030 2050 2100

CO

2st

orag

e[G

tCO

2/yr

]

BECCS

Iron and Steelsector, HydrogenproductionGas power

Oil power

Coal power

20Global CO2 capture and storage (CCS)


- The total amount of CO2 storage by CCS is also very different in 2050 according to the uncertainty in the climate sensitivity and different socioeconomic scenarios.- In 2100, large amounts of CCS including BECCS are required for all of the emission pathways for 2 °C goal.

0

10

20

30

40

50

60

70

80

90

100



C.S. of 2.5℃

[1] 450 ppmCO2eq

stabilization


C.S. of 2.5℃

[1] 450 ppmCO2eq

stabilization


C.S. of 2.5℃

[1] 450 ppmCO2eq

stabilization

2010 2030 2050 2100

Pass

enge

r tra

nspo

rtat

ion

serv

ice

dem

and

(bill

ion

p-km

/yr)

ICEV

ICEV (High-concentrationbio fuel)

HEV

HEV (High-concentrationbio fuel)

Plug-In HEV

Plug-In HEV (High-concentration bio fuel)

BEV

FCV

21Global transportation (automobile)


- The technology options in automobile are also very different in 2050 according to the uncertainty in the climate sensitivity.- In 2100, large shares of EVs and FCVs are required as well as HVs for all of the emission pathways for 2 °C goal.

4. Co-benefit and trade-off between climate change and other

sustainable development goals

23

Harmonization among climate change issues and other SDGs needed

- We have multiple agendas to be tackled. Harmonization among climate change issues and other SDGs are necessary.

0

50

100

150

200

250

300

350

400

A-b

asel

ine

A-C

P6.

0A

-CP

4.5

A-C

P3.

7A

-CP

3.0

A-b

asel

ine

A-C

P6.

0A

-CP

4.5

A-C

P3.

7A

-CP

3.0

A-b

asel

ine

A-C

P6.

0A

-CP

4.5

A-C

P3.

7A

-CP

3.0

A-b

asel

ine

A-C

P6.

0A

-CP

4.5

A-C

P3.

7A

-CP

3.0

A-b

asel

ine

A-b

asel

ine

A-C

P6.

0A

-CP

4.5

A-C

P3.

7A

-CP

3.0

A-b

asel

ine

A-C

P6.

0A

-CP

4.5

A-C

P3.

7A

-CP

3.0

A-b

asel

ine

A-C

P6.

0A

-CP

4.5

A-C

P3.

7A

-CP

3.0

A-b

asel

ine

A-C

P6.

0A

-CP

4.5

A-C

P3.

7A

-CP

3.0

A-b

asel

ine

A-C

P6.

0A

-CP

4.5

A-C

P3.

7A

-CP

3.0

A-b

asel

ine

A-C

P6.

0A

-CP

4.5

A-C

P3.

7A

-CP

3.0

2000 2050 2100 2000 2050 2100 2000 2050 2100 2000 2050 2100 2000 2050 2100

China India Other Asia Sub-saharan Africa L. America

Food

Acc

ess

Inde

x[a

mou

nts

of fo

od c

onsu

mpt

ions

/GD

P]

(US

.Y20

00=1

00)

Bioenergy and forestation effects

w.o. Bioenergy and forestation effects

686 916 1315651

655715

745754

Vulnerable

801

Climate Change Mitigation & Food Access

- Vulnerabilities of food access will decrease in most countries and regions in the long-term under any emission scenarios, because future incomes are expected to increase.- Large-scale forestation and bioenergy use slightly increase vulnerabilities of food access.

24

Source) K. Akimoto et al., Natural Resources Forum, 36(4), 231-244, 2012

Food access index (Amounts of food consumption／GDP)

Deeper emissionreductions

An example of the synergy and trade-off among SDGs: Climate Change Mitigation and Food Access 25

- Factor decomposition shows that climate change mitigation brings about small positive impacts on the food access index in certain aspects, but worsens the index in total.

Food access index (amounts of food consumption/GDP) in 2050 by factor

-5%

0%

5%

10%

15%

20%

25%

30%

CP4.5 CP3.7 CP3.0 CP4.5 CP3.7 CP3.0 CP4.5 CP3.7 CP3.0 CP4.5 CP3.7 CP3.0 CP4.5 CP3.7 CP3.0

China India Midde-East &N.Africa

Subsahara Africa Latin America

Chan

ge ra

tes o

f foo

d ac

cess

(%, r

elat

ive

to th

e ba

selin

e)

－△(GDP)/GDP : due to climate damage

－△(GDP)/GDP : due to climate mitigation

△(Food consumptions)/(Food consumptions): due to land-use changes for "bioenergy production and afforestation"△(Food consumptions)/(Food consumptions): due to land-use changes for "food production"

△(Food consumptions/GDP)/(Food consumptions/GDP)

CP4.5: about 650 ppm CO2eqCP3.7: about 550 ppm CO2eqCP3.0: about 450 ppm CO2eq

Source) estimated by RITE (2012)

worse

Climate Change Mitigation & Air Pollution (PM2.5) Reduction Measures – Global Impacts

26

- Co-benefits of CO2 emission reduction by PM2.5 concentration reduction measures saturates as thelevel of PM2.5 mitigation deepens in the light of significant difference in the measure costs betweenPM(L) and PM(H).- Taking into account the sufficiently high measure costs for PM(H), the co-benefit of CO2 reduction byPM2.5 measures is limited to be about 9 Gt-CO2/yr within the realistic range of measure costs.

20

30

40

50

60

70

10 20 30 40 50 60 70

Glo

bal a

vera

ge p

opul

atio

n-w

eigh

ted

PM2.

5 [μ

g/m

3 ]

Global energy-related CO2 [Gt-CO2/yr]

REFCO₂ (L)CO₂ (H)PM (L)PM (H)

2010

2030

2050

2030

2050

2030

2050

2030 2050

20302050

0

50

100

150

200

250

2010 2020 2030 2040 2050

Carb

on p

rice

[US$

/t-C

O2]

CO₂ (H)

CO₂ (L)

0

10

20

30

40

2010 2020 2030 2040 2050

PM2.

5 pr

ice

[bill

ion

US$

/(μg

/m3 )

] PM (H)

PM (L)

Scenario assumptions*

*The emission factors by region and sector are assumed not to increase after 2010 than those in 2010.Estimated by RITE DNE21+

0

20

40

60

80

100

-20 0 20 40 60 80 100

PM2.

5 re

duct

ion

rela

tive

to R

efer

ence

(%)

CO2 emission reduction relative to Reference (%)

0

20

40

60

80

100

-20 0 20 40 60 80 100

PM2.

5 re

duct

ion

rela

tive

to R

efer

ence

(%)

CO2 emission reduction relative to Reference (%)

Climate Change Mitigation & Air Pollution (PM2.5) Reduction Measures – Co-benefits

27

- The co-benefit of CO2 emission reductions on PM2.5 reductions are larger than that of PM2.5reductions on CO2 emission reductions. Large co-benefits are not necessarily observed for allcountries but are observed particularly in India and China.- For PM2.5 reductions, relatively cheap end-of-pipe type measures exist (e.g., de-Sulfer, de-NOx); butfor CO2 reductions, the end-of-pipe type measures (e.g., CCS) are relatively expensive.

PM2.5 concentration reduction cases CO2 emission reduction casesSmaller co-benefit of PM2.5 reductions on CO2 reductions

Larger co-benefit of CO2 reductions on PM2.5 reductions

Net CO2 = (Net CO2／Gross CO2) × (Gross CO2／PE) × (PE／GDP) × (GDP)PM2.5 = (PM2.5／Gross PM2.5) × (Gross PM2.5／PE) × (PE／GDP) × (GDP)

Energy savingFuel switchingEnd-of-pipe measures (de-Sulfer, de-NOx etc.)

End-of-pipe measures (CCS)Kaya identity Co-benefit measures

adverseside-effects

Estimated by RITE DNE21+

5. Innovations and emission pathways

5th Science and Technology Basic Plan of Japan- “Society 5.0” (“Super Smart Society”) - 29

- Wide range of innovations of technologies and their integrations are required for improving our welfare and sustainable development.- AI, IoT, big data etc. will be able to stimulate such innovations.

Source) Japanese Government

AI + IoT + big data + ….

Operation ratio of automobiles is about 4%, for example. The large room for the improvement exists.

CO2 emission

Carbon price

Baseline scenario

Intervention scenario

Carbon price/Marginal abatement cost

Explicit high carbon prices of such as over 100$/tCO2 in real price are unlikely in a real world. Technology and social innovations resulting in low (implicit or explicit) carbon prices (including coordination of secondary energy prices) are key for deep emission cuts to be implemented.

Model world: Ordinary technology progress

CO2 emission

Carbon price

Baseline scenario

Intervention scenario

Implicit or explicit carbon price/Marginal abatement cost

By technology and social innovations

Realistic world requirement:Innovations stimulated & implemented

30

Image of standard scenario by models and real world scenarios for deep cuts

6. Conclusions

32

♦ Nearly zero CO2 emissions are required in the long-term.♦ But there are lots of uncertainties, and we should recognize these

uncertainties to manage the risks in a better way.♦ Potential increase in mitigation costs: political factors (large

differences in MAC across nations, Trump Administration etc.), social constraints of technology deployment, inefficient policies etc.

♦ Potential decrease in mitigation costs (future unknown innovations)♦ Pursuing co-benefits in line with several objectives of sustainable

development including PM2.5 reductions. But some are trade-offs. Our resources are limited and total risk management is required.

♦ Innovations are significant for achieving zero emissions. The demand side revolutions induced by IT, AI etc. will be highly expected as one of the innovations for reducing energy consumptions and toward deep emission reductions (but currently uncertain) .

♦ Total risk management with flexibilities reserved is critically important rather than pursuing rigid targets.

Conclusions

Appendix

Energy Assessment Model: DNE21+♦ Linear programming model (minimizing world energy system cost)♦ Evaluation time period: 2000-2100

♦ World divided into 54 regions

♦ Bottom-up modeling for technologies both in energy supply and demand sides (about 300 specific technologies are modeled.)

♦ Primary energy: coal, oil, natural gas, hydro&geothermal, wind, photovoltaics, biomass and nuclear power

♦ Electricity demand and supply are formulated for 4 time periods: instantaneous peak, peak, intermediate and off-peak periods

♦ Interregional trade: coal, crude oil, natural gas, syn. oil, ethanol, hydrogen, electricity and CO2

♦ Existing facility vintages are explicitly modeled.

Representative time points: 2000, 2005, 2010, 2015, 2020, 2025, 2030, 2040, 2050, 2070, 2100

Large area countries are further divided into 3-8 regions, and the world is divided into 77 regions.

- The model has regional and technological information detailed enough to analyze sectoral measures. Consistent analyses among regions and sectors can be conducted.

34

Region divisions of DNE21+ 35

Technology Descriptions in DNE21+ (1/2)

Fossil fuelsCoalOil (conventional, unconv.) Gas (conventional, unconv.)

Cumulative production

Unitproductioncost

Renewable energiesHydro power & geothermalWind powerPhotovoltaicsBiomass

Annual production

Unitsupplycost

Nuclear power

Energy conv. processes(oil refinery, coal gasification, bio-ethanol, gas reforming, water electrolysis etc.)

Industry

ElectricPower generation

CCS

Transport

Residential & commercial

Iron & steel

Cement

Paper & pulp

Chemical (ethylene, propylene, ammonia)

Aluminum

vehicle

Refrigerator, TV, air conditioner etc.

Solid, liquid and gaseous fuels, and electricity <Top-down modeling>



36

0

2,500

5,000

7,500

10,000

US W. Europe Japan China India and S. Asia

Ene

rgy

sec

urity

inde

x

2000

2050 A-Baseline

2050 A-CP4.5

2050 A-CP3.0

Vulnerable

37Climate Change Mitigation & Energy Security

While the energy security index of Japan decreases (less vulnerable) for CP3.0 (synergy effects), those of China and India increase (more vulnerable) for deeper emission reductions due to increase in imported gas shares (adverse side effects).

( ) ( )∑∑ ⋅+⋅=i

gasiigas

ioilii

oil SrTPES

cSr

TPEScESI 2

,2

,

Share of imported oil in TPES Political risks of region i Dependence on region iESI : energy security index, TPES: total primary energy supplyNote: index based on IEA, 2007

Source) K. Akimoto et al., Natural Resources Forum, 36(4), 231-244, 2012

long-term strategy toward deep emission reductions under … · 2018-02-13 · long-term strategy...

Documents