1

Peak Oil through the lens of a general equilibrium assessment

Henri WAISMAN, Julie ROZENBERG, Olivier SASSI and Jean-Charles HOURCADE

(waisman@centre-cired.fr)

34th IAEE International ConferenceStockholm - 19/23 June 2011

Motivation

Peak Oil = impending stagnation and decline of world oil production

Determinants of Peak Oil : what drives the occurrence of Peak Oil?

Date of Peak Oil : a signal for the urgency of changes in energy uses

Economic consequences : sudden shocks and long-term effects

Motivation

2 largely disconnected strands of literature

Geological-based analyses of Peak Oil Hubbert-like approach = bell-shaped production curves at the field level Extrapolation at a global level is questionable

heterogeneous oil reserves and producers’ rationale

Economic dimensions of oil markets Short-term effects of oil disruption

econometric analyses after first oil shock no explicit accounting of resource depletion

Long-term effects of exhaustible resources exploitation production profile under perfect foresight (Hotelling) scarce representation of peaks

Objectives

Representing the macroeconomic dimensions of long-term oil markets

Endogenizing Peak Oil from the interplay between determinants of long-term oil markets under inertia and imperfect anticipationsOil supply, Liquid Fuels demand, Alternatives to oil

Representing the impact of the economy on oil markets Consumer behavior driving oil demand in function of price signals Investment decisions in production capacities of oil and alternatives in

function of profitability prospects and capital availability

Capturing the feedback effects of oil markets on macroeconomy Energy trade (energy security for oil-importers, exportation revenues) Technical and structural change ( oil-dependency, industrialization) Aggregate effects on economic activity and welfare

Approach (1) A CGE approach embarking lessons from economic assessments of

oil shocks

mark-up pricing instead of competitive markets to capture a set of market imperfections

partial utilization rate of capital due to the complementarity between energy and capital services

putty-clay description of capital (with fixed energy intensity for old vintages) capturing the effects of inertias in the renewal of productive capital

frictions in reallocating capital and labor across heterogeneous sectors (in terms of sensitivity to oil prices) causing differentiated levels of idle production capacities and unemployment.

Approach (2)

A hybrid bottom-up/top-down approach

sectoral modules embarking expert-based information on the direction and magnitude of technical change, including the determinants of oil markets

a dialogue between technical/behavioral determinants and macroeconomic interactions

A representation of « second-best » economies  

missing, imperfect or distorted markets noncompetitive behaviors technical and institutional inertias asymmetry or incompleteness of information

The IMACLIM-R model

Economic signals

(prices, quantities, Investments)

Static Equilibrium (t)

under constraints

Dynamic sub-modules

(reduced forms of BU models)

Static Equilibrium (t+1)

under updated constraints

Technical and structural parameters

(i-o coefficients, population, productivity)

Hybrid matrixes in values, energy and « physical » content (Mtoe, pkm) Secure the consistency of the engineering based and economic analyses Explicit accounting of inertias on equipement stocks Technical asymptotes, basic needs

Solowian growth engine in the long run but transitory disequilibrium Unemployment, excess capacities Investments under imperfect foresight (informed by sectoral models) Trade and capital flows under exogenous assumption about debts

Modeling oil supply geological dimension

In each region, 7 categories of conventional + 5 categories of non-conventional reserves size of the reservoir Q∞,i (ultimate reserves, including past production) threshold selling price pi

(0) above which production is profitable (proxy for costs of exploration/exploitation and accessibility)

Maximum rate of increase of production capacity for each category, given geological constraints

bi: steepness of the bell-shape profile (default value: b=0.061)t0,i : expected date of the maximum for oil category i, given past production

0,

0,

( )

max( )

. 1( , )

( , ) 1

i i

i i

b t ti

b t t

b eCap t i

Cap t i e

Modeling oil supply producers’ decisions

All regions except Middle-East = “Fatal producers” maximum deployment for profitable categories (poil > pi

(0) ) investment stopped for non-profitable categories (poil < pi

(0) )

Middle-East = “Swing producers” Fill the gap between demand and other suppliers World price depends on the utilization rate of production capacities ME deployment of production capacities in function of their price objective

Total production capacity= sum over all categories and regions

time

Production Capacity of oil category i

t t+1

Cap(t,i)

Cap(t,i)+Δcapmax(t,i)

Modelling oil demand

Liquid fuels’ demand (residential, industry, transport)

Utility and profit maximization under constraintsShort-term : inertia in the renewal of equipments and LBDLong-term : consumption styles (preferences), technical potentials

(technology availability, asymptotes), location patterns

Alternatives to oil

BiofuelsCompetition over oil-based fuels: supply curves increasing with oil priceAsymptotes on BF production at a given year (competition of land uses)Evolve in time to represent technical progress

Coal-To-Liquidbackstop technology with capacity constraintsenter the market at high oil priceproduction limited by the cumul of past investments

Results

Two counterfactual scenarios of the world economy over 2010-2050 different production capacity expansion in the short term

Market Flooding scenario

ME expands production capacities to maintain oil price at 2009 level Intense demand in the short-termTechnical change towards oil-intensive patterns in the long term

Limited Deployment scenario

ME restricts capacity expansion to let short-term prices riseModeration of oil demandTechnical change towards oil-free patterns in the long term

ResultsPeak Oil profiles

Level

Date

Post-PO decrease

0

20

40

60

80

100

120

140

2010 2020 2030 2040 2050

Oil production (Million b/d)

Limited Deployment (LD) scenarioMarket Flooding (MF) scenario

Close dates but very different time profiles!

0

0,5

1

1,5

2

2,5

3

3,5

4

4,5

5

2010 2020 2030 2040 2050

World oil price (index 1=2009 level)

Limited Deployment (LD) scenarioMarket Flooding (MF) scenario

ResultsWorld oil prices

Controlled by OPEC in the short-term

Sudden rise at the Peak Oil date

Continuous increase in long term due to constraints on CTL

ResultsTime profile of oil revenues

Short-term revenues controlled by price targets

Bubble of long-term profits triggered by price increase after PO

0

200

400

600

800

1000

1200

1400

1600

2010 2020 2030 2040 2050

Middle-East annual oil profits (Billion $)

Limited Deployment (LD) scenarioMarket Flooding (MF) scenario

Room for Short-term vs. Long-term tradeoff!

ResultsTradeoff for oil producers

MF scenario profitable for oil producers at discount rates lower than 6%

Discount rateLimited Deployment

ScenarioMarket Flooding

Scenario

0% 38.9 43.6

1% 28.9 31.8

2% 21.9 23.6

5% 10.6 10.8

6% 8.7 8.6

7% 7.2 7.0

15% 2.4 2.2

Results Impacts on oil-importers (OECD)

Close average growth but different time profiles

Average(2010-2050)

Short-term Period(2010-2025)

Peak Oil Period(2025-2040)

Long-term Period(2040-2050)

Natural growth rates 1.42% 1.69% 1.30% 1.19%

Effective growth rates

Limited Deployment scenario

1.57% 1.93% 1.43% 1.24%

Market Flooding scenario

1.53% 2.00% 1.29% 1.18%

ResultsA sensitivity analysis on oil

determinants Amount of reserves Low bound: 1.6Trillion bbl conventional+ 0.8Trillion bbl non-conventional High bound: 2.3Trillion bbl conventional+ 1.2 Trillion bbl non-conventional

Inertia on non-conventional production Low inertia: b=0.04 vs. high inertia: b=0.07

2015

2020

2025

2030

2035

2040

0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1

Date (years)

m (reserves)

Date of the Peak Oil

LD ; b=0.041

LD ; b=0.07

MF ; b=0.041

MF ; b=0.07

Increasingreserves

60

80

100

120

140

160

180

200

220

0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1

Price ($/bbl)

m (reserves)

Post -Peak Oil price($/bbl)

LD ; b=0.041

LD ; b=0.07

MF ; b=0.041

MF ; b=0.07

Increasing reserves

Conclusion

A model to frame the debate on Peak Oil and long-term oil markets

representing the interplay between geological and macroeconomic dimensions of long-term oil markets

endogenizing the date of Peak Oil from market interactions between oil supply, liquid fuels’ demand and alternatives to oil

analyzing the effects of alternative pricing strategies and various assumptions on reserves and non conventional production

assessing the indirect macroeconomic effects of Peak Oil on long-term economic growth trajectories

Conclusion

Some important messages

Oil pricing trajectories hardly affect the date of Peak Oil … but have important consequences on its macroeconomic effects

(oil revenues, economic activity in oil importing countries)

strong variations of Peak Oil and related macroeconomic effects according to assumptions on oil reserves

Middle-East producers may prefer moderate short-term prices if they are able to bias technical change under imperfect foresight

High short-term oil prices may be beneficial in oil importing countries as a hedging strategy against long-term oil scarcity

20

Thank you for your attention!

waisman@centre-cired.fr

http://www.centre-cired.fr

http://www.imaclim.centre-cired.fr/

34th IAEE International ConferenceStockholm - 19/23 June 2011