Title: Use of international energy statistic in research' Presentation of an oil market model
1Use of international energy statistic in
research. Presentation of an oil market model
- Presentation by Knut Einar RosendahlResearch
Department in Statistics Norway1st meeting in
the Oslo City group on energy statistics6-8.
February 2006Statistics Norway
2Introduction
- International energy markets are important to
understand - Particularly the oil market
- Affects global economy in general and national
economies in particular - Improved understanding of energy markets helps
decision makers wrt. investments, policies,
planning etc. - Construction of energy market models may help to
understand the markets - Put together central relationships like demand
functions for different consumers and supply
behaviour for different producers into a
consistent framework - Taking into account available reserves and costs
of production - Models will never be completely correct, e.g. due
to unpredictable behaviour
3- Numerical models need a quantitative basis
- Based on energy statistics
- A decent model requires decent statistics
- The value of a numerical model depends on the
quality of its data basis - Some statistics are more vital than others
- A numerical model may give valuable insight even
if (some of) the data are of poor quality - International energy statistics of good quality
are not easily accessible and applicable - Different definitions
- Lack of statistics for several countries
- Rest of presentation
- Brief presentation of oil market model and
analysis - Presentation of energy statistics used and
problems encountered
4FRISBEE a global energy market model
- FRISBEE is a partial equilibrium model of the
international energy markets - Developed in Statistics Norway over the last 3
years - Partly financed by Norwegian Research Council
- Emphasis on the oil and gas markets
- Applications
- Analysis of OPECs optimal oil price level
- Analysis of globalisation of gas markets
- Analysis of various energy and climate policies
on energy markets - Users
- Mainly own use for research
- Incl. presentations for e.g. Ministry of Energy,
Oil companies etc. - Several analyses for Statoil
5FRISBEE regional classification
Russia, Ukraine Belarus
Western Europe
Eastern Europe
Canada
China
Caspian region
USA
Rest-Asia
OECD-Pacific
OPEC-core
Africa
Latin America
OPEC-rest
6Important aspects of the FRISBEE model
- Production consumption of energy goods in each
region - Free trade on the world market
- Global equilibrium in each period (exogenous
stock changes) - Time periods 1 year
- Base year 2000
- Time horizon Normally 2030
- More flexibility in the long run
- Consumption of oil products in each region
- Three end-users Power production, Industry,
Others - Two oil products Transport oil and Oil for
stationary purposes - Depends on end-user prices of different energy
goods GDP per capita Population Energy
efficiency (exogenous) Demand in previous period
7Important aspects of the FRISBEE model (cont.)
- Transformation from primary to final energy goods
is not modelled (fixed add-on in unit costs) - Oil production in each region
- Four field groups (e.g., onshore/offshore)
- Distinguishes between fields in production,
undeveloped fields and fields to be discovered - Both investment and production decisions are
treated explicitly - Costs are divided into capital costs and
operating costs - Developed fields go through four phases
- Investment phase pre-peak phase peak phase
decline phase - Oil investments and discoveries
- At the end of each year oil companies and OPEC
invest in new fields and improved oil recovery
(IOR), and discover new fields
8Important aspects of the FRISBEE model (cont.)
- Oil production outside OPEC
- Production is set so that marginal costs equal
regional crude oil price (minus gross taxes) - Higher costs in the decline phase (before
shut-in) - Oil production in OPEC
- OPEC supplies residual demand for a pre-specified
oil price target - Investments outside OPEC
- Oil companies consider undeveloped oil fields and
IOR projects outside OPEC - Maximises net present value at discount rate set
to 10 p.a. - Adaptive price expectations based on recent years
price levels - Investments in new fields take time to get on
stream IOR-investments increase production the
next year
9Important aspects of the FRISBEE model (cont.)
- Investments in OPEC
- OPEC invests to maintain a certain excess
capacity - Investment costs
- Vary across regions and field groups
- High risk acts as actual costs
- Costs increase when remaining reserves decline
- Technological progress may compensate
- Short-term costs may increase due to high
activity level
10OPECs optimal oil price level analysis of the
FRISBEE model
- What is OPECs optimal oil price level in the
long run? - Discounted income over the period 2005-2030
- High prices High initial income lower market
share in the medium term (high Non-OPEC supply
and low demand) - Low prices Low initial income higher market
share in the medium term - Constraint Consider only constant oil prices
from 2010-2030 - Implicitly assume that OPEC is a profit
maximising cartel - Ignore internal conflicts or coordination
problems - Ignore other considerations for OPEC (e.g.,
geopolitical issues) - OPECs optimal oil price is not a prediction of
the actual oil price level in 2010-2030
11OPECs market share over time for different oil
price levels
12OPECs net cash flow over time
13OPECs optimal oil price level for different
discount rates (2000-)
14Energy statistics used to calibrate the FRISBEE
model
- Mostly international statistics
- IEA EIA (US DoE) BP USGS
- Energy demand and production in base year 2000
- Mainly IEA statistics
- Good coverage for most countries (incl. Non-OECD)
- Energy prices in base year
- Mainly IEA statistics
- End-user prices of energy goods in
households/industry/power prod. - OECD countries Large variation!
- Non-OECD countries More or less absent
- Forced to search elsewhere (research literature
etc.), and make rough assumptions
15Energy statistics used to calibrate the FRISBEE
model (cont.)
- Energy prices in base year (cont.)
- Wholesale prices
- OECD Mostly OK
- Non-OECD Little information
- Taxes on energy demand
- Not focused in this study (so far), but in other
studies - Difficult to obtain for several countries and
energy goods - Oil field information (resources, reserves,
geology, start-up) - Access to extensive database with detailed
information - Helped to sort oil reserves into
developed/undeveloped fields onshore/offshore
large/small etc. - Different opinions exist, especially wrt. size of
reserves
16Energy statistics used to calibrate the FRISBEE
model (cont.)
- Production costs
- Access to information about operating and
investment costs in different types of fields in
important countries - Used to calibrated cost functions in different
phases - Investment, peak phase, decline phase
- Undiscovered oil resources
- US Geological Survey (2000) Potential
discoveries next 30 years - Country specific
- Different opinions exist of course!
- Used to calibrate deterministic discovery
function - Projections of future economic growth and energy
demand - Mainly based on EIA
- Used to calibrate income elasticities
- Projections of energy demand vary a lot
17Energy statistics important needs
- More statistics for Non-OECD countries
- More and more important for the international
energy markets (e.g. China, Russia) - Very limited statistics, especially on prices
- More available statistics for fossil fuel
resources - Important in order to understand the development
of the energy markets (cf. e.g. the peak oil
debate) - Very costly today
- Different definitions about reserves, and
different evaluations - Predominance of unlisted companies
- Strategic importance for oil companies etc.
- Statistics on bio resources and other renewables
also important
18More information
- Aune, Glomsrød, Lindholt and Rosendahl (2005)
Are high oil prices profitable for OPEC in the
long run? Discussion Papers 416, Statistics
Norway(http//www.ssb.no/cgi-bin/publsoek?jobfor
sideiddp-416kodedplangen) - E-mail ker_at_ssb.no