1
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual SupplierCooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
3
Cooperate or Compete?
Insights from Simulating a
Global Oil Market with No
Residual Supplier
Bertrand Rioux, Abdullah Al Jarboua, Fatih
Karanl, Axel Pierru, Shahd Al Rashed, and
Colin Ward
August 2020
Doi: 10.30573/KS--2020-DP13
2
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
About KAPSARC
Legal Notice
The King Abdullah Petroleum Studies and Research Center (KAPSARC) is a
non-prot global institution dedicated to independent research into energy economics,
policy, technology and the environment across all types of energy. KAPSARC’s
mandate is to advance the understanding of energy challenges and opportunities
facing the world today and tomorrow, through unbiased, independent, and high-caliber
research for the benet of society. KAPSARC is located in Riyadh, Saudi Arabia.
This publication is also available in Arabic.
© Copyright 2020 King Abdullah Petroleum Studies and Research Center (“KAPSARC”).
This Document (and any information, data or materials contained therein) (the
“Document”) shall not be used without the proper attribution to KAPSARC. The
Document shall not be reproduced, in whole or in part, without the written permission
of KAPSARC. KAPSARC makes no warranty, representation or undertaking whether
expressed or implied, nor does it assume any legal liability, whether direct or indirect,
or responsibility for the accuracy, completeness, or usefulness of any information that
is contained in the Document. Nothing in the Document constitutes or shall be implied to
constitute advice, recommendation or option. The views and opinions expressed in this
publication are those of the authors and do not necessarily reect the ofcial views or
position of KAPSARC.
3
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual SupplierCooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
We investigate a transition to a competitive world oil market in which OPEC — led by Saudi
Arabia — stops acting as the primary residual supplier within the world oil market starting in
2020. Our modeling results include the following highlights.
In 2020, as OPEC ramps up production, Brent prices fall US$11.5/b below the World Energy Outlook
(WEO) stated policies scenario of the International Energy Agency (IEA 2019).
From 2020 to 2030, prices recover as a result of demand response combined with a need for sustained
investment in new long-term conventional production.
Only when capital approved for new conventional projects matches historic highs (present value of
about US$125 billion per year) do prices remain below the WEO stated policies scenario through 2030.
Crude prices recover faster and exhibit signicantly higher variability when investment in shale oil
production slows and output peaks at 12 million barrels per day (MMb/d) versus 16 MMb/d in 2025.
A decline in short-term tight oil projects limits its role in balancing the market as a source of marginal
production, compared to new conventional projects with longer lead times.
Saudi Arabia benets nancially by continuing to act as the primary residual supplier only with strong
cooperation from other producers in OPEC and the larger OPEC+ group.
Key Points
4
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
Summary
Structural changes in the global oil sector are
disrupting conventional market dynamics
and the roles played by competing and
cooperating producers. Industry players are
adjusting to the shale (or ‘tight’) oil revolution and
the possibility of plateauing or peaking global oil
demand. In particular, OPEC and Saudi Arabia,
its top producer, are reshaping the organization’s
role as the primary residual supplier to the world
oil market. In recent years, OPEC has invited other
major exporters, including Russia, to cooperate
under the OPEC+ production agreement in an effort
to stabilize prices.
Given these changes, what if OPEC, led by Saudi
Arabia, decided to cease organizing residual
production collectively, transitioning the world to a
more competitive oil market? We investigate such
a transformation by simulating the mid-term market
from 2019 to 2030. We construct scenarios where
OPEC, or OPEC members other than Saudi Arabia,
start behaving as competitive price takers in 2020,
and stop participating as part of a collective residual
oil supplier. Our analysis employs an economic
equilibrium model, and is calibrated to the World
Energy Outlook (WEO) stated policies scenario of
the International Energy Agency (IEA 2019) as a
reference for expected real future oil prices, demand
and income growth. In our alternative residual
supplier scenarios OPEC, or Saudi Arabia alone,
organize production targeting the WEO price levels.
In our competitive market scenario, prices
drop in 2020 by around 12 United States (U.S.)
dollars per barrel (US$/b) below the WEO
reference of US$66/b (Brent), accelerating global
oil demand growth. However, depending on
upstream investment trends, we nd signicant
variability in the mid-term price response. This
includes approvals for long-term conventional oil
developments, and shorter-term tight oil projects,
in line with historic trends and analyst projections.
Following the increase in production by OPEC
members and initial price decline, prices do recover
to WEO levels by the end of the decade in most
scenarios.
In our simulations prices only remain well below
the WEO outlook when the approval of new
conventional projects increases signicantly with
respect to the slowdown observed in recent years.
Given the initial lowering of prices when transitioning
to a competitive market, one might expect a
continued slowdown in investment. In addition, as
existing conventional projects are depleted over
the next decade supplies are expected to tighten,
assuming demand growth does not reverse. In
this case the additional capacity from OPEC is not
sufcient to keep prices below the WEO.
In a scenario where investment in new tight oil
projects drops by 50% and total shale oil output
reaches a peak of 12 million barrels per day (MMb/d)
in 2025 (instead of 17 MMb/d), prices recover quickly
to WEO levels after 2020. However, they exhibit
greater variability compared to more aggressive
growth scenarios for global tight oil projects. Under
this tight oil constraint, the standard deviation of
the annual change in competitive prices increases
by at least 200%. This holds even with accelerated
investment in conventional projects. In this case
the slower development of tight oil projects limits
its availability as a source of marginal production to
balance supply and demand in the short term.
From 2020 to 2030, our results show that average
competitive market prices only remain below the
WEO reference levels when the present value of
capital committed to investment in new conventional
projects exceeds US$125 billion per year on a
present value basis. This level is similar to the
average yearly investment from 2004 to 2014,
5
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual SupplierCooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
when project approvals reached a four-decade
peak. However, when the present value of projects
approved each year falls below US$100 billion,
the recent 10-year average, the three-year moving
average prices surpass the WEO reference by
mid-decade.
Under the more aggressive growth scenario for tight
oil (17 MMb/d peak in 2025), the decline in oil prices
persists with Brent falling an average of US$10-14/b
below the WEO reference, and demand exceeding
the WEO scenario by at least 3 MMb/d. Only if
annual conventional investments drop well below
their 2004-2014 average, as might be expected
given the decline in prices, do the competitive
market prices fully recover or exceed the WEO
outlook in the second half of the decade.
We compare the present value of prots generated
by Saudi Arabia under the competitive market and
the residual supplier scenarios. The latter assume
that the residual supplier is able to stabilize the
market in line with the price and demand levels from
the WEO. We nd that Saudi Arabia only benets
nancially by serving as a residual supplier, rather
than engaging in traditional competitive market
behavior, with strong cooperation from OPEC.
As a sole residual supplier, Saudi Arabia would
have to withhold large amounts of production,
compromising its relative market share, resulting in
lower prots. Cooperation has become even more
necessary since the emergence of U.S. shale oil has
introduced structural uncertainties (such as the price
responsiveness of non-OPEC oil supplies) into the
market.
Expanding the role of the residual supplier to other
producers (OPEC, OPEC+) reduces the sensitivity
of Saudi Arabia’s market share and revenues to the
growth in tight oil production within the expected
range of the long-run price elasticity of global oil
demand (between -0.2 and -0.4). Therefore, Saudi
Arabia should seek greater support from the rest
of OPEC as well as non-OPEC producers (such as
Russia) to ensure that the Kingdom’s oil revenues
exceed what could be earned in a market with no
residual supplier at all. In other words, it is in Saudi
Arabia’s interest either to abandon its role as primary
residual supplier or to share this responsibility with a
larger group.
Summary
6
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
1. Introduction
The oil market is undergoing profound
structural changes due to the shale oil
revolution and the prospect of plateauing or
peaking oil demand. These developments could
induce either more cooperation or more competition
between oil-producing countries. In this paper,
we simulate a global oil market without a residual
supplier that organizes production levels in an
attempt to manage the price of oil, as opposed to
behaving as a competitive price taker.
We develop an equilibrium model of the global oil
market through 2030 with a detailed representation
of oil-producing assets throughout the world. Rather
than apply a dominant rm model with a competitive
fringe, as is standard in oil modeling (Plaut 1981;
Rauscher 1988; Jones 1990; Behar and Ritz 2017;
Golombek, Irarrazabal, and Ma 2018; Volkmar
2018), we test cases under a competitive market
with no residual supplier scenario in which every
oil-producing country behaves as a price taker.
In this case, investment and production decisions
depend only on how their marginal production cost
compares to price. We analyze how market prices
would potentially materialize in such a scenario and
the corresponding revenues for Saudi Arabia as the
largest residual supplier within OPEC.
We then compare the outcomes of the hypothesized
competitive market to an alternative reference
residual supplier scenario. We examine two cases
within this scenario: in the rst, OPEC members
collectively operate as a residual supplier; in the
second, Saudi Arabia acts as the only residual
supplier, and other OPEC members join the fringe.
In both instances, the residual supplier follows a
price targeting strategy: it adds or subtracts oil from
the market to achieve the desired global market
price, based upon how much is being supplied by
the fringe.
These two residual supplier cases reect different
perspectives within the research community. While
some studies treat OPEC as the world’s dominant
oil rm (Rauscher 1988; Jones 1990), others (Plaut
1981; Adelman 1995) argue that Saudi Arabia
performs the role of the dominant rm within
OPEC when its members fail to coordinate the
organization’s output. For example, in preparation
for expected oversupply in 2020, Saudi Arabia
reportedly encouraged OPEC to increase cuts
(Sheppard and Ravel 2019). In a review of the
evolution of OPEC models, Fattouh and Mahadeva
(2013) concluded that OPEC's market power has
varied over time and thus that no single model ts
OPEC’s behavior. Behar and Ritz (2017) suggest
that OPEC still operates as a dominant player but
prioritizes market share over deterrence strategies.
Following the oil crisis of the 1970s, the literature
on the strategies of oil-producers (especially
OPEC) regarding pricing and production decisions
has grown substantially (e.g., Powell 1990; Gately
1995). The geopolitical environment and oil market
structure are much different today than in the
1980s-90s. First, the market has transformed due
to increased shale oil production and reduced
production costs, weakening OPEC’s market
dominance and its ability to inuence prices.
Second, while non-OPEC oil supply continues
to grow, global oil demand is slowing. Third,
legislative attempts, such as the No Oil Producing
and Exporting Cartels (NOPEC) Act, which was
proposed by the United States (U.S.) Congress to
allow the national oil companies that make up OPEC
to be sued under U.S. antitrust law, may potentially
make it difcult for OPEC to engage in coordinated
cuts in world oil supply (NOPEC 2007; Rystad
Energy 2018b; Reuters 2019). To our knowledge,
no study exists that investigates the implications of
these changes in an oil market without a residual
supplier. We believe our analysis is timely and will
7
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual SupplierCooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
contribute to debates over the future of the global oil
market and the signicance of the residual supplier
role currently lled by Saudi Arabia and OPEC.
The next section describes the representation of
demand, the decision rules for producers, and other
features of the model. Section 3 details scenarios
that explore the consequences of structural changes
in the world oil market, including to the role of the
residual supplier, and different scenarios regarding
investment in new oil production capacity. Section 4
discusses the results and their interpretation. Finally,
Section 5 offers concluding remarks.
1. Introduction
8
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
!!!
"!
#
!!$%!!!
'
!
!
!
"
!
#
!
!!
!"!
!#!!
"!
#
!!!
"!
#
!!$%!!!
'
!
!
!
"
!
#
!
!!
!"!
!#!!
"!
#
We propose an aggregate model of the
global oil market within an equilibrium
framework in which price clears the
market. The discussion below describes the supply
and demand representations and provides a
comparison to techniques used in other world oil
market models. Appendix A provides the complete
mathematical formulation with equations.
The model simulates the medium-term
consequences of a market with and without a
residual supplier. First, to calibrate our demand
outlook, we establish a reference residual supplier
scenario in which the residual supplier targets
a given oil price by increasing or decreasing
production in response to the total output of all
other suppliers, treated as fringe competitors.
Then we solve for the market equilibrium under the
competitive scenarios where all producers behave
as price takers (no residual supplier). The model is
solved to nd a new competitive price equilibrium
in the midterm. It also estimates the nancial
consequences for different suppliers in terms of oil
revenues.
The model is dynamic and captures the transitional
adjustments that occur when an alternative market
structure is assumed. The world oil market clears,
period-by-period, with demand balancing supply
on an annual basis, for all hydrocarbon liquids
including crude oil, condensates, natural gas liquids
(NGL), renery gains and other liquids (biofuels
and alcohols destined for the same market as
petroleum products). We represent global demand
and supply as a single node and do not account for
regional crude ows.
2. Model Description
2.1. Representation of global
oil demand
The following equation species total world oil
demand in each year t (Dt)
(1)
where At is a scaling variable, or a variable
capturing the effects of all other exogenous factors,
and p
t is the three-year moving average market
price with pt representing the market-clearing price
in year t. We consider a single global oil price
based on the Brent Crude oil marker. Yt is the
current global gross domestic product (GDP), ε
is the long-run price elasticity of oil demand, and
γ is the long-run income elasticity of oil demand,
or the impact of global GDP. The price elasticity
of demand is applied to the three-year moving
average price to reect the lag on the impact of oil
prices on demand (Hamilton 2003; Kilian 2008).
Current global GDP is given by
(2)
where gt is a reference GDP growth rate. The ratio
p
t ⁄pt in equation (2) represents the impact of prices
on GDP growth, where pt is the three-year average
reference price linked to the reference GDP growth.
The parameter θ can also be interpreted as the
elasticity of real economic growth with respect
to variations in the price of oil. For instance, if
the moving average oil price (p
t ) is higher than
that of the reference level (pt), then for θ<0 the
economic growth from t-1 to t will be slower than
the reference (gt), and vice versa.1
A straightforward way to specify equations (1) and
(2) is to use elasticity estimates available in the
1 James L. Smith suggested this demand-side representation and has our gratitude.
9
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual SupplierCooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
!!!
"!
#
!!$%!!!
'
!
!
!
"
!
#
!
!!
!"!
!#!!
"!
#
!!!
"!
#
!!$%!!!
'
!
!
!
"
!
#
!
!!
!"!
!#!!
"!
#
2. Model Description
literature. Then, At can be calibrated based on
projections for a desired reference case (i.e.,
using the expected oil demand D
t, the moving
average oil price pt, and global GDP Y t). However,
the exact value of θ cannot be estimated based
on available data. We show in Appendix A.1 that
an approximate value of θ can be obtained by
formulating a time series equation that relates the
growth rate of global GDP to that of oil prices.
2.2. Calibrating the world oil
demand
We calibrate the demand curve to replicate a
reference scenario that projects annual world
demand, average oil price and global GDP growth.
We investigate the period from 2019 to 2030, using
the 2019 World Energy Outlook (WEO) from the
International Energy Agency (IEA 2019).
The scaling coefcient At from equation (1) is set as
follows:
(3)
In other words, At guarantees that the price and
GDP outlook replicate the global oil demand
projected in the reference scenario.
(4)
Regarding the price and income elasticities
of oil demand, the relevant literature offers no
consensus. The estimates range widely, from -0.01
to -0.58 for price elasticity (ε) and 0.24 to 1.32 for
income elasticity (γ) (Javan and Zahran 2015).
For the simulations in this paper, we select -0.25
as the long-run price elasticity of oil demand and
0.75 for the long-run income elasticity. We also run
sensitivity analyses by calibrating the model across
a range of price elasticities from -0.1 to -0.6.
We source world oil prices for the years 2017 and
2018 from Reuters to construct the three-year
moving average prices during the rst two years
of the study period. All prices are adjusted to 2019
real terms.
WEO publishes several outlooks for the global
oil market. We focus on the organization’s
stated policies scenario, which accounts for new
environmental measures that target a gradual
reduction in oil demand growth. Under this
scenario, annual demand growth slows to an
average of 0.8% as global demand rises from
98.8 MMb/d in 2019 to 107.7 MMb/d in 2030. This
scenario assumes Brent prices steadily increase
from 61 U.S. dollars per barrel (US$/b) in 2019 to
US$88/b in 2025, and US$96/b in 2030. Over this
period GDP growth averages 3.6%, and oil demand
does not peak.
As part of our sensitivity analysis, discussed in
Section 4.3, we also calibrate our model to the
International Energy Outlook (IEO) of the Energy
Information Administration (EIA 2019). The IEO
projects weaker average price and demand growth,
with Brent hitting a maximum of US$76.9/b and
global demand reaching 105.8 MMb/d in 2030,
reecting annual growth of 0.5%; GDP also
expands more slowly at an average of 3.3%. Table
A.2 in Appendix A.1 presents the values from WEO
and IEO.
2.3. Global oil supply model
In each period, suppliers organize to maximize
their prots in two ways: selling oil produced from
existing projects at the market price and investing
in new ones. The model assumes fringe suppliers
behave competitively, selling oil at the market price
that clears demand, corrected to account for crude
quality and regional price markers (e.g., Brent versus
WTI). Appendix A.2 provides a detailed
10
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
mathematical formulation of the supplier’s problem.
To explicitly model supply decisions at the asset
level, we use a set of linear activities built using a
detailed database of production cost and capacity
projects, as described in Section 2.4. Supplies are
differentiated by quality, eld type, location, and
ownership, providing a detailed representation of
different supply categories. As we represent oil
demand at the global level, we do not explicitly
model regional transportation.
An alternative approach would be to employ
a reduced-form supply curve. For example,
Huppmann and Holz (2012) develop a model to
investigate market power in the global oil market.
Each node in their model represents a single
continuous non-linear Golombek supply curve
(Golombek, Gjelsvik, and Rosendahl 1995).
However, their model only considers a single year
and neglects investment decisions and depletion.
One can also design a structural equation based
on the price elasticity of supply rather than using
a supply curve. This provides a more aggregate
representation of production and investment
decisions and can be useful for investigating
longer-term trends and time horizons, and when
detailed supply data may not be available. The
dominant rm-competitive fringe model developed
by Golombek, Irarrazabal, and Ma (2018) employs
this technique to investigate the exercise of market
power by OPEC from 1986 to 2016, and drivers of
long-run oil price trends, including GDP and supply
depletion.
Our model includes investments as additional
linear activities and categorizes them as either
short-term (shale, or ‘tight,’ oil projects) or long-term
(all other developments, including conventional
oil, oil sands, heavy oil, NGL, and condensates).
U.S. shale oil projects are generally characterized
by fast decline rates, and short development lead
times and production cycles; the majority of a
single well’s total production occurs within a year
(Kleinberg et al. 2016). Other developments, such
as conventional onshore and offshore, and oil
sands, generally have multi-year investment lags
and production proles.
For tight oil, we assume that on an annual scale,
the time between project approval and start of
production is negligible. Therefore, producers
effectively make only one decision: to develop new
capacity if the total unit production cost (operating
and development) is lower than the current price
at the time. We then embed capital development
costs within the marginal production cost, dened as
the breakeven cost of the shale project. The model
decides whether to make investments in tight oil
projects based on the current equilibrium price and
assumes they will be available for production within
the same year that the investment decision was
made.
For longer-term investments, producers make two
decisions based on different cost curves: the rst
for production from existing capacity, when the price
exceeds the variable operating cost, and the second
for capacity expansion. As detailed in Appendix A.2,
the model initiates conventional investments if the
forecasted present value (PV) of oil revenues for all
future production exceeds the PV of the total capital
commitment of the project. All PVs are calculated
using a discount rate of 10%, a standard value
applied in the oil industry and the related literature
(e.g., Powell 1991; Gately 1995).
We calculate future oil revenues for production
beyond the model horizon by extrapolating forward
oil prices. We x prices to the level output by the
model at the end of the horizon, and production to
the expected annual output reported by Rystad.
2. Model Description
11
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual SupplierCooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
2. Model Description
The model can be solved using either myopic or
forward-looking supplier behavior. In the latter we
solve the model as a single problem, assuming
suppliers have perfect foresight over the model
horizon. The myopic approach uses a rolling horizon
that provides suppliers with limited or imperfect
information on current and future market conditions.
In this case, we solve the model using the recursive
method described in Appendix A.5.
A myopic supplier does not factor in the
longer-term exhaustibility of reserves.
However, work by Hart and Spiro (2011) nds
that scarcity or Hoteling rents that would result
from resource depletion have historically been
marginal or absent in oil markets, and that
other factors play a stronger role in shaping oil
prices.
2.4. Supply calibration
To calibrate oil supplies, we employ Rystad
Energy’s UCube upstream oil and gas database,
which represents each OPEC member, including
Saudi Arabia, on a stand-alone basis. This allows
us to capture country-level supply changes and
disruptions, including the attack on Saudi Arabia’s oil
production in September 2019. For each country, the
database provides distinct resource endowments,
cost structures, and nancial, technical and
geopolitical constraints. Rystad production data
extracted for existing projects include projected
annual output, marginal costs, capital development
cost, and breakeven oil prices (for tight oil). All costs
are reported in real 2019 terms.
We also extract data on crude quality ranked
by the American Petroleum Institute (API)
gravity scale, sulfur content, and regional price
differentials, such as Brent-WTI spread. These
values are used to adjust the price paid for
different grades of oil in different regions. Table
1 displays a summary of the quality indices
and examples of regional Brent/WTI spreads,
providing an overview of the indices evaluated
at the asset level. First prices for a given crude
type are determined by dividing the projected
annual revenues of each asset by the production.
The prices are then compared to Rystad’s Brent
outlook to determine a price index that reects
crude quality, regional logistics, and other factors.
We calibrate OPEC production data to reect
the total sustainable production capacity of each
member country, dened as the capacity that can
be put into production within 90 days, based on
data from IEA’s monthly oil market report (IEA
2019). The model denes spare capacity as total
sustainable capacity minus annual production
for each member country. Our analysis excludes
some OPEC members (Ecuador, Congo, Gabon,
Equatorial Guinea, Nigeria, and Venezuela) because
of discrepancies between projected production
in the IEA’s monthly oil market and self-reported
production levels, noting that they do not hold
signicant spare capacity.
Appendix A.5 provides additional descriptions of the
supply data, including the characterization of tight
oil elds, gas condensates, and other liquids. We
also describe data aggregation methods used to
reduce the number of supply activities in the model
and improve model performance, without severely
compromising the model resolution.
2.5. Calibrating investments
We utilize Rystad UCube data on new oil projects
planned between 2020 and 2050, including the
projected approval year, production start year,
and annual capital development costs. We initially
calibrate the model so that all projects can be
approved in any year within the horizon, if they are
protable, with cost and production proles
12
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
2. Model Description
adjusted to the year selected by the model.
However, under this assumption, the total capital
committed in a given year can greatly exceed the
range of values observed in the oil market.
(a) Price correction by product quality
Quality Markdown
Light Crude 1
Regular Crude 0.993
Condensate 0.99
Heavy Oil (API 20 – 23) 0.957
Sour Crude 0.936
Synthetic Crude 0.908
Heavy Oil (API 15 – 19) 0.907
Extra Heavy Oil 0.901
Bitumen 0.624
NGL, renery gains, and other liquids 0.512
Table 1. Examples of quality and regional price markers used in the model.
Sources: KAPSARC analysis, Rystad.
(b) Brent-WTI price spread
Year Brent/WTI
2019 1.15
2020 1.15
2021 1.18
2022 1.12
2023 1.08
2024 1.08
2025 1.08
2026 1.08
2027 1.07
2028 1.07
2029 1.07
2030 1.07
Although many projects may be protable within
an expected price range, numerous factors
exogenous to our basic optimization problem
constrain potential investments. The amount of
capital available globally for the development of
new oil projects is limited and can be inuenced
by social and political factors. Investors may react
to environmental concerns related to oil projects,
choosing to allocate funds to other markets.
To address this, we introduce annual investment
constraints shared by all suppliers,2 as detailed in
the following scenario design section. Within our
equilibrium framework, they introduce an additional
cost to the development of oil projects that are
protable but exceed the investment cap. Appendix
A.2 presents the mathematical formulation of the
constraints.
Figure 1a plots the PV of capital (in real 2019
dollars) approved annually for new long-term
projects between 1980 and 2030, calculated using
reported and projected development costs for all
global oil projects included in the UCube Database
through 2100. The values vary signicantly during
this period, averaging US$46 billion during the
1980s and 1990s and US$145 billion from 2004
to 2014, before the mid-decade crash in oil prices.
Based on Rystad’s outlook, capital commitments
are expected to reach similar levels after 2020.
2 This representation allows us to simulate the oil price for a given level of investment in the upstream sector. The
model assumes a scarcity premium on the investment cost and applies this for all producers. Investment happens
in countries that have the lowest production costs.
13
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual SupplierCooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
Figure 1a. Aggregate PV of capital committed to new long-term oil projects globally.
0
20
40
60
80
100
120
140
160
180
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
2012
2014
2016
2018
2020
2022
2024
2026
2028
2030
Billion US$
Figure 1b. Total annual capital expenditures on short-term tight oil projects.
Note: Values beyond 2018 (patterned area) based on projections from Rystad.
Sources: Rystad UCube, KAPSARC Analysis
0
20
40
60
80
100
120
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
2012
2014
2016
2018
2020
2022
2024
2026
2028
2030
Billion US$
2. Model Description
14
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
2. Model Description
Figure 1b shows the annual capital expenditures
on tight oil investments from 1980 to 2030, rather
than the discounted lifetime project costs, reecting
the short-term nature of these projects, and the
extensive funds required to keep them running. The
rapid expansion beginning around the year 2000
reects the U.S. tight oil boom and the expected
future capital required to continue growth based on
Rystad’s projections.
Compared to other industry research rms, Rystad
predicts more aggressive growth for tight oil. It
forecasts that U.S. tight oil output will more than
double from an average output of 6.4 MMb/d in 2018
to a peak of about 17 MMb/d in 2025. The outlook
is based on Rystad’s assessment of commercially
viable projects, at a WTI price in the range of
US$55/b to US$70/b. The 2017 World Oil Outlook
(OPEC 2017), and the WEO stated policies scenario
(IEA 2019) both project slower growth in tight oil
production, with global output reaching 12 MMb/d in
2025. Reducing the annual capital invested in tight
oil projects to 50% of the levels projected by Rystad
generated similar production levels.
Using our detailed supply representation, one
could apply different assumptions at the country or
rm level, based on past performance or current
geopolitical constraints, for example. Technical
and logistical bottlenecks can be included by
constraining new capacity to a fraction of the
capacity existing in the previous period. A maximum
expansion rate could be assigned to different
producer groups based on historical and projected
expansion rates. Investment behavior could also
be adjusted by altering the discount rate or using
different techniques to extrapolate future oil prices.
15
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual SupplierCooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
We design scenarios to assess the
medium-term consequences (2019
to 2030) for the world oil market. This
includes the competitive scenarios assuming all
oil producers behave as price takers (no residual
supplier), and reference scenarios where OPEC,
or Saudi Arabia alone, continues operating as a
residual supplier.
The competitive market model is calibrated to the
WEO stated policies scenario (IEA 2019). The model
is run rst with no investment constraints, and then
under different constraints on capital available for
new long-term conventional and short-term tight oil
projects. These scenarios are used to analyze how
development of these different types of production
impact the dynamics of the competitive scenarios.
We also compare Saudi Arabia’s market share and
revenues in the competitive and residual supplier
scenarios under identical investment constraints.
Under the residual supplier scenario, a select group
of producers (e.g. OPEC) implicitly target a world oil
price by adding or removing capacity. We use the
price, and corresponding global demand, from WEO
as the target for the residual supplier; however,
alternate targets could also be tested. Appendix
A.3 describes how we solve our equilibrium model
under the residual supplier scenario. The model
determines production and investment decisions
of all countries, except the residual supplier, by
assuming they behave as competitive price takers.
The residual supplier then sets its production in
order to clear the market.
First, we assume that the group of OPEC members
with production quotas in 2018 (Algeria, Angola,
Iran, Iraq, Kuwait, Libya, Saudi Arabia, and the
United Arab Emirates) collectively serve as
the residual supplier, and that they coordinate
production in proportion to their total capacity.
Given the history of oil markets and that Saudi
Arabia is the largest oil exporter and maintains
most of the world’s spare capacity, we run an
alternative case in which Saudi Arabia acts as a
residual supplier without support from OPEC. In
this situation, OPEC members are assumed to stop
negotiating production quotas, which could happen
under international pressure, such as the U.S.
NOPEC bill. (We note that our analysis is purely
hypothetical and does not reect policies of Saudi
Arabia’s government.)
3.1. Financial constraints for
new long-term projects and
tight oil
For long-term projects we simulate various
investment constraints within the range of historic
values reported in Figure 1a. Between 2020 and
2030, the Rystad data projects that the average
annual capital committed to new projects is
US$132 billion in present value terms. In our Rystad
investment plan scenario the model can invest in
new projects on or after their projected approval
year.
Next, we simulate alternative scenarios in which
new projects can be developed up to an annual cap
on the present value of approved capital. Instead of
following the project approval years from Rystad,
any project approved between 2020 and 2050 can
be built. The constraints and corresponding scenario
labels are US$75 billion, US$100 billion, US$125
billion, and US$150 billion cap. These differ from the
standard Rystad plan because we assume suppliers
can prioritize projects according to protability. This
provides a straightforward approach to simulate
a range of constraints, while allowing exibility in
project plans, including accelerated development of
new protable projects by OPEC members.
3. Scenario Design
16
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
Short-term tight oil production and investment
decisions are based on the commercial outlook
provided by Rystad, reaching a peak of about 17
MMb/d in 2025. We assume suppliers can develop
any tight oil eld, up to the maximum production
level, as long as prices exceed the break-even point.
To restrict tight oil production to near 12 MMb/d we
implement a tight oil cap scenario limiting annual
investments to 50% of the Rystad projections.
3. Scenario Design
17
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual SupplierCooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
In this section, we will analyze the market
equilibrium in the competitive market scenario
with different investment constraints, and
conduct a sensitivity analysis with respect to the
model demand parameters. We also examine the
impact of these scenarios on the production and
revenue of the primary residual supplier (i.e., Saudi
Arabia) with and without support from other OPEC
members.
4.1. Oil price and demand
dynamics: cap on long-term
investments.
Figure 2 compares oil price (2a) and demand (2b)
trajectories from the models of perfect competition
(no residual supplier scenario) with the reference
WEO levels used to construct the residual supplier
scenario. We include results for the competitive
market model with no investment constraints under
both the forward-looking and myopic supplier
assumptions. We then simulate the competitive
market under the different investment constraints
described in section 3.1; the Rystad investment
plan and the three average investment caps
(US$75 billion, US$100 billion, US$125 billion and
US$150 billion). We apply the forward-looking
supplier assumption in these cases; however, given
binding investment constraints in these scenarios,
the results do not differ signicantly with myopic
suppliers.
Under the competitive market scenarios with
unconstrained investments (forward-looking and
myopic in Figure 2), the present value of capital
approved for new projects exceeds US$1.6 trillion
in 2020. This surpasses historic investment levels
reported in Figure 1 and drives prices below
US$55/b after 2023 with a rapid acceleration
in demand growth. It is very unlikely that such
aggressive investments would materialize on the
basis of protability alone. These two scenarios
illustrate the impact of applying myopic versus
forward-looking supplier behavior without binding
investment constraints. In the latter, producers
expect a downward trend in prices, withholding
approval for projects that become unprotable below
about US$45/b.
Under the constrained scenarios, the
difference between the myopic (not shown) and
forward-looking assumptions become much less
pronounced because the investment constraints are
binding in both cases, resulting in nearly identical
sets of approved projects and available capacity.
The level of the constraint does alter the resulting
equilibria signicantly after 2024 as new capacity
ramps up and replaces declining production from
existing projects. Under the Rystad investment plan
and the US$150 billion cap, prices between 2020
and 2025 average US$11/b, or 14%, below the WEO
reference prices. However, if the current slow-down
in long-term project approvals persists (below
US$100 billion), prices could exceed the WEO
reference.
4.2. Oil price and demand
dynamics: tight oil
investments
In Figure 3 we show price (3a) and demand (3b) for
scenarios where tight oil investments are capped
at 50% of the levels projected by Rystad. Figure
4 depicts global tight oil production levels before
and after applying the US$150 billion investment
constraint.
4. Model Analysis and Results
18
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
4. Model Analysis and Results
Figure 2b. World oil demand, IEA WEO versus ‘no residual supplier’ scenario.
95
100
105
110
115
120
125
130
135
2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
Demand, MMb/d
Competitive (forward-looking) Competitive (myopic) Competitive US$75 billion cap
Competitive US$100 billion cap Competitive US$125 billion cap Competitive US$150 billion cap
Competitive Rystad investment plan Residual supplier (WEO stated policies)
Sources: IEA World Energy Outlook, KAPSARC analysis.
Figure 2a. World oil price, IEA WEO versus ‘no residual supplier’ scenario.
35
45
55
65
75
85
95
105
115
125
2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
Price, US$/b
Competitive (forward-looking) Competitive (myopic) Competitive US$75 billion cap
Competitive US$100 billion cap Competitive US$125 billion cap Competitive US$150 billion cap
Competitive Rystad investment plan Residual supplier (WEO stated policies)
19
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual SupplierCooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
4. Model Analysis and Results
Figure 3a. World oil prices: residual supplier (WEO stated policies) versus the competitive scenarios, assuming a
50% reduction in annual capital expenditures on short-term tight oil projects.
50
60
70
80
90
100
110
120
130
140
2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
Price, US$/b
Competitive US$75 billion cap Competitive US$100 billion cap Competitive US$125 billion cap
Competitive US$150 bil. cap Competitive Rystad plan Residual supplier (WEO)
Figure 3b. World oil demand: residual supplier (WEO stated policies) versus the competitive scenarios, assuming a
50% reduction in annual capital expenditures on short-term tight oil projects.
98
100
102
104
106
108
110
112
2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
Demand, MMb/d
Competitive US$75 billion cap Competitive US$100 billion cap Competitive US$125 billion cap
Competitive US$150 bil. cap Competitive Rystad plan Residual supplier (WEO)
Sources: IEA World Energy Outlook, KAPSARC analysis.
20
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
In Figure 3a, the slowdown in the growth of tight oil
production causes both average prices and price
variation to increase. The standard deviation of
the annual rate of change in competitive market
prices increases by at least 120% in the presented
scenarios. This holds even when assuming stronger
investments in conventional long-term production.
What we observe is that a decline in the capacity
of short-term tight oil projects restricts its ability
to balance the market as a source of marginal
production, compared to new conventional projects
with longer lead times.3
Under the range of investment assumptions
considered, prices recover much faster than in
Figure 2. As a result, the market could experience
a ramp-up in investment in long-term projects,
in response to increasing prices. Only under the
accelerated approval of new long-term projects (i.e.,
the US$150 billion cap), do the competitive prices
remain below the WEO reference after 2025.
In Figure 3b total demand drops well below the
reference scenario after 2024, when the average
annual capital committed to long-term projects
is capped below levels projected by the Rystad
investment plan (i.e. US$75 billion and US$100
billion). We observe peak oil demand in both these
scenarios, as well as under the Rystad investment
plan.
Note that under the residual and no residual
supplier scenarios we apply the same constraints
on investments in new projects, and nd them to be
binding in both cases. Under the lower investment
constraints (e.g. US$75 billion and US$100 billion)
the total sustainable production (demand) in the
competitive scenarios is less than the demand in
WEO. In this case the residual supplier would have
to produce above its sustainable production levels at
additional capital cost, to balance the market at the
price and demand projections of the WEO. Under
these capital constraints it appears that the WEO
market equilibrium would not be sustainable.
Finally, Figure 4 displays the global tight oil
production in different scenarios, with and without
investment caps. The black dotted and dashed lines
illustrate how tight oil production responds to the
increased production from conventional projects
and decline in prices. The dashed line (competitive
w/o investment cap) represents the scenario with
myopic suppliers and no investment caps. It shows
the sensitivity of tight oil production as prices
dip below US$60/b after 2022. Below US$45/b
production drops rapidly, with output in 2026 falling
to pre-2018 levels but recovering as price recover
to US$ 55/b by the end of the decade. The other
scenarios in Figure 4 are for the Rystad investment
plan. In this case the smaller reduction in prices in
the competitive results in a smaller reduction in tight
oil production, between 1 and 2 MMb/d.
4.3. Sensitivity analysis:
demand parameters
The price elasticity of demand plays a central role
in calibrating how consumers respond to a change
in the supply structure. To investigate the sensitivity
of our model to our assumptions, we run several
scenarios calibrated across a range of elasticities.
We also investigate sensitivity with respect to
income elasticity, but nd that for the range of
values of interest the price elasticity of demand has
a much stronger impact on the equilibria.
3 A Rystad study (2018) nds amplied short-term price volatility when U.S. tight oil capacity increases. However,
the methodology and data frequency differ from those employed in this study.
4. Model Analysis and Results
21
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual SupplierCooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
Figure 4. Global tight oil production in the residual and the competitive scenarios under the Rystad investment plan
(dashed line is without constraints on conventional and unconventional projects).
4
6
8
10
12
14
16
18
2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
MMb/d
Residual supplier Competitive Residual supplier tight oil cap
Competitive tight oil cap Competitive w/o investment cap
Sources: Rystad, KAPSARC analysis.
The price and demand results under three different
price elasticity assumptions (-0.1, -0.25, and -0.5)
are presented in Appendix B, gures B.1 and
B.2. As expected, with consumers more (less)
responsive to the change in price in the competitive
market, demand and prices recover faster (slower).
Figure 5 plots the average prices from the
competitive scenario (solid lines) from 2020 to 2030
across a range of demand elasticities: the US$150
billion cap (5a), US$125 billion cap (5c), and US$100
billion cap (5e), and cases including the 50% cap on
tight oil (5b), (5d) and (5f), respectively. Dotted lines
show the maximum and minimum values, and the
dashed lines the average prices from WEO.
In the scenarios with more relaxed investment
constraints (US$150 billion cap) average prices
respond more to changes in the absolute price
elasticities, or demand response. As the price
elasticity increases, demand reacts stronger
to lower prices created by additional OPEC
production, leading to higher equilibrium prices.
As the investment constraints, and supplies, are
tightened average prices atten out across different
elasticities, but exhibit larger variability (maximum
and minimum price spread) as observed in Figures
2 and 3. Here the results reect the calibration,
with average prices converging towards the WEO
reference, while oscillating with greater amplitude
due to the higher scarcity premiums on investments.
The trend of attening of average prices reverses
as the investment constraints are increased further,
Figure (5f) US$100 billion and tight oil cap. In this
case unresponsive demand does not catch up with
tighter supplies, causing more frequent price spikes
(see Appendix B Figure B.2) and higher average
prices for lower elasticities.
In Appendix B, we present additional scenarios
calibrated to the reference demand and price
projections of IEO (EIA 2019).
4. Model Analysis and Results
22
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
Compared to WEO, calibrating the model to IEO
data produces reduced prices and price variability,
due to the slower price and demand growth
projections.
Figure 5. Average prices (solid lines) from the competitive scenarios between 2020 and 2030 across a range of
long-run price elasticities of demand. Dotted lines represent maximum and minimum prices in the competitive
scenario, dashed lines the average price from the WEO reference (2019).
50
70
90
110
130
150
0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60
Price, US$/b
| |
(e) US$100 billion cap
50
70
90
110
130
150
0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60
Price, US$/b
| |
(f) US$100 billion and tight oil cap
50
60
70
80
90
100
0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60
Price, US$/b
| |
Average Average (WEO) Max/Min
(a) US$150 billion cap
50
60
70
80
90
100
0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60
Price, US$/b
(b) US$150 billion and tight oil cap
||
50
60
70
80
90
100
110
120
130
0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60
Price, US$/b
| |
(c) U$125 billion cap
50
60
70
80
90
100
110
120
130
0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60
Price, US$/b
(d) U$125 billion and tight oil cap
| |
Sources: KAPSARC analysis.
4. Model Analysis and Results
23
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual SupplierCooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
4.4. Supply dynamics of the
residual supplier
Figure 6 plots the annual liquids production
for Saudi Arabia (a) and OPEC (b) under the
competitive market scenarios with different
investment constraints. These scenarios
Illustrate the growth in Saudi Arabia and OPEC’s
total production, as well as available capacity
assuming accelerated approval of new projects
based on their protability relative to other
producers. Note that under the Rystad plan
many projects organized by OPEC members are
scheduled for approval after 2030.
The lines in Figure 6 reect total production
assuming OPEC coordinates residual production
under the Rystad investment plan, with dashed
lines including the cap on tight oil. See Appendix
B (Figure B.4) for the residual supplier’s
production under different investment constraints.
Saudi Arabia’s production falls to 11.4 MMb/d,
about 2 MMb/d below its capacity, coinciding
with tight oil production peaks in 2025. In this
case OPEC production falls to 35.3 MMb/d, with
participating members contributing an amount
of residual production proportional to their total
capacity. Historically, Saudi Arabia has shouldered
the largest share of production cuts compared to
other members. If Saudi Arabia were to organize
the majority of withheld production (about 6.4
MMb/d), with limited to no support from OPEC, it
would face a signicant reduction in market share.
Under the 50% reduction in the amount of capital
invested in tight oil projects market share of the
residual supplier increases signicantly, exceeding
40 MMb/d after 2020. In this case Saudi Arabia
may be better positioned to operate as a residual
without support from OPEC. Also under this tight
oil constraint, the residual supplier’s production
exceeds OPEC capacity under the Rystad
investment plan by 2.2 MMb/d. This would require
members to produce above sustainable production
levels, or accelerate project approvals, such as
under the US$75 billion cap investment cap.
In a world with strong tight oil growth, clearly
Saudi Arabia would require stronger support from
other producers to maintain production above 10
MMb/d at the stated price target. This might include
countries outside OPEC, such as Russia and other
producers participating in the OPEC+ group (Gnana
2019). The idea that OPEC may require support
from other countries outside of the organization has
been explored in the study titled “Is OPEC Dead
Without Russia?” by Volkmar (2018).
4.5. Economics of the
residual supplier
In light of the ndings above, we investigate whether
serving as a residual supplier can increase Saudi
Arabia’s oil revenues relative to a purely competitive
market behavior, assuming the kingdom can
independently maintain its market share as the
primary residual supplier, or do so with support from
OPEC (and other partners).
To answer this question, we estimate the relative
value of Saudi Arabia’s oil prots, dened as net
revenues less annual capital expenditures, under
the competitive and residual supplier scenarios.
We calculate the difference between the annual
prots in the residual supplier and the competitive
scenarios in two cases: OPEC jointly acting as
residual supplier and Saudi Arabia acting alone.4
4 We assume that the residual supplier targets the same reference price regardless of whether all OPEC or only Saudi
Arabia serves as residual supplier. Although Saudi Arabia might revise (reduce) the price target to preserve its market
share if operating solo, we do not include such an adjustment in our estimates of the relative value of the different markets.
4. Model Analysis and Results
24
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
Then we calculate the net present value (NPV) of
the difference from 2020 to 2030 using a discount
rate of 4%, as opposed to the 10% rate used to
evaluate upstream investment decisions. This
reects the lower discount rates typically applied
by governments, and is within the range of risk
premiums derived by Pierru and Matar (2014)
for the evaluation of oil-related public investment
projects in Saudi Arabia.
Our NPV calculation provides an estimate of the
prots gained (or lost) by Saudi Arabia when
transitioning to the competitive market structure.
However, one should be careful in interpreting
this as a measure of the additional market value
that can be achieved by the residual supplier.
First matching the price and demand equilibrium
from the WEO stated policies assumes perfect
coordination of residual production by different
producers (e.g. OPEC). This may not be feasible.
In addition, the residual supplier could adopt a
different production target that does not reect
the WEO levels. Our analysis is simply used
to identify a directional shift in prots of the
competitive scenario compared to the WEO
reference case.
Figure 7a and 7b plot the difference in Saudi
Arabia’s prots for different price elasticities of
demand, for scenarios without and with the 50%
cap on tight oil, respectively. This illustrates how
the consumer response to a change in price can
impact the relative value of transitioning to a more
competitive market structure. Solid lines represent
cases where Saudi Arabia coordinates production
with OPEC as the residual supplier and dashed
lines with Saudi Arabia acting alone.
As shown in Figure 5 when consumers
are less price responsive (lower absolute
elasticity) average prices tend to decrease
in the competitive market, resulting in lower
overall prots. This trend is reected in Figure
7a. Under these scenarios when Saudi Arabia
coordinates production cuts with OPEC, the NPV
of its prots are greater than the competitive
scenarios for price elasticities of demand below
0.35 (in absolute terms), depending on the
investment constraints applied. When acting as
a residual supplier without support from OPEC,
Saudi Arabia’s prots are always lower than the
competitive scenarios (> US$300 billion) due to a
signicantly lower market share.
When applying the tight oil cap in Figure 7b the
market share of the residual supplier increases.
This increases the prots of the residual supplier,
and under the US$150 billion cap this pushes the
NPV curve to the right, crossing at a higher price
elasticity. However, the supply constraints also
result in higher average prices in the competitive
market that balance the increased production by
the residual supplier.
As the investment cap on new conventional
projects drops to US$100 billion the slope of the
NPV curve changes. Under this scenario average
competitive prices decrease as the elasticity
increases (Figure 5e). Despite the declining
prices in the competitive market they are still
on average higher than the WEO reference,
and the prots made by a residual supplier. We
exclude results where Saudi Arabia acts as the
only residual supplier, because, in this scenario
Saudi Arabia produces well above its sustainable
production capacity after 2024 (see Appendix B
Figure B.4a).
4. Model Analysis and Results
25
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual SupplierCooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
Figure 6b. Same as Figure 6a but for Liquids production by OPEC.
34
36
38
40
42
44
46
48
50
2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
MMb/d
Competitive US$150 billion Competitive US$125 billion Competitive US$100 billion
Competitive US$75 billion Competitive Rystad plan
Residual supplier (all OPEC)
Residual tight oil cap (all OPEC)
Sources: Rystad, KAPSARC analysis.
Figure 6a. Liquids production by Saudi Arabia in the competitive scenarios (shaded areas) and OPEC as the
residual supplier (lines). Dashed lines are for the 50% cap on tight oil investments.
10
11
12
13
14
15
16
2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
MMb/d
Competitive US$150 bil. Competitive US$125 bil. Competitive US$100 bil.
Competitive US$75 bil. Competitive Rystad plan Residual supplier (all OPEC)
Residual tight oil cap (all OPEC)
4. Model Analysis and Results
26
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
Figure 7a. NPV of Saudi Arabia’s prots (residual supplier minus competitive scenarios) versus the price elasticity of
demand. Residual supplier as OPEC (solid lines) or only Saudi Arabia (dashed lines).
-1200
-800
-400
0
400
0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60
NPV of profits, billion US$
Price elasticity of demand (absolute value)
Residual supplier - All OPEC (US$125 bil) Residual supplier - Saudi Arabia (US$125 bil)
Residual supplier - All OPEC (US$100 bil) Residual supplier - Saudi Arabia (US$100 bil)
Residual supplier - All OPEC (US$150 bil) Residual supplier - Saudi Arabia (US$150 bil)
Figure 7b. NPV curves as described in 7a including the 50% cap on tight oil investments.
NPV, billion US$
Price elasticity of demand (absolute value)
Residual supplier - All OPEC (US$150 bil) Residual supplier - Saudi Arabia (US$150 bil)
Residual supplier - All OPEC (US$100 bil)
500
250
0
-250
-500
0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60
Source: KAPSARC analysis.
4. Model Analysis and Results
27
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual SupplierCooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
This study introduces a competitive market
model for the supply-demand equilibrium
of the global oil market, based on a novel
approach that analyzes mid-term oil market
dynamics with and without a residual supplier.
The model uses detailed linear supply activities
and explicit nancial investment constraints, and
differentiates supplies and investment as either
long-term conventional or short-term tight oil
projects, and can further categorize them based
on political, geographic, product quality and other
factors.
We calibrate the model to a given reference outlook
relating oil prices, global demand and GDP growth
from 2019 to 2030, including global price elasticity
of demand and income elasticity assumptions.
The reference represents an idealized view of
the current market structure with Saudi Arabia,
supported by OPEC, operating as the primary
residual supplier.
This study presents several scenarios across a
range of investment assumptions for long-term
and tight oil projects, solved from the year 2019 to
2030. Our competitive scenarios demonstrate how
prices and demand could respond, relative to the
reference scenario, in a market with no residual
supplier. Under our standard demand elasticity
assumption (ε=-0.25) prices decline signicantly
under the Rystad investment plan and the US$150
billion cap for long-term investments, average
US$14/b less than the reference WEO prices from
2020 to 2025.
Our analysis indicates that prices under our
competitive market scenarios have a high
sensitivity to growth in tight oil production. Price
variability, measured as the annual change in
prices, increases substantially (by at least 200%)
when capping tight oil investments to 50% of the
levels projected by Rystad. In this case the ability
of new short-term tight oil projects to balance
the market as a source of marginal production is
reduced compared to conventional projects that
involve longer lead times.
The study nds that Saudi Arabia only benets
nancially by serving as a residual supplier
(following the WEO reference demand and prices)
with strong coordination from other OPEC and
possible assistance from OPEC+ members. Also,
this only holds when the long-run price elasticity
of demand is low, less than about 0.35 in absolute
terms.
The results suggest that cooperation between
Saudi Arabia and other producers can reduce the
sensitivity of the Kingdom’s oil revenues to tight oil
production growth. Given that the long-run price
elasticity of global demand is likely to be close to
the values mentioned above, this nding supports
the view that collectively enlarging the function of
the residual supplier to other non-OPEC producers
(like Russia) may be necessary for Saudi Arabia
to maintain higher oil revenues than it could in a
market without a residual producer.
The emergence of U.S. shale oil has introduced
structural market uncertainties (such as the
price responsiveness of non-OPEC oil supplies),
making cooperation even more necessary. Even
in cases where tight oil production growth slows,
Saudi Arabia’s does not clearly benet from
acting alone as a residual supplier. Therefore,
it is in Saudi Arabia’s interest either to abandon
the role of residual supplier, or to jointly perform
this function as part of a larger — and more fully
cooperative — group.
5. Conclusion
28
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
References
Adelman, Morris Albert. 1995. Genie Out of the Bottle:
World Oil Since 1970. Cambridge, M.A.: MIT Press.
Behar, Alberto, and Robert A. Ritz. 2017. “OPEC vs US
Shale : Analyzing the Shift to a Market-Share Strategy.”
Energy Economics 63 (March): 185–98. https://doi.
org/10.1016/j.eneco.2016.12.021
Dickey, David A., and Wayne A. Fuller. 1981. “Likelihood
Ratio Statistics for Autoregressive Time Series with a
Unit Root.” Econometrica 49 (4): 1057–72. https://doi.
org/10.2307/1912517
EIA. 2019. “International Energy Outlook 2019.” Outlook
484 (July): 70–99. https://www.eia.gov/outlooks/ieo/pdf/
ieo2019.pdf
Energy Information Administration (EIA). 2019.
“International Energy Outlook 2019.” September 24,.
https://www.eia.gov/outlooks/ieo/pdf/ieo2019.pdf
Fattouh, Bassam, and Lavan Mahadeva. 2013. “OPEC:
What Difference Has It Made?” Annual Review of
Resource Economics 5 (1): 427–43.
Gnana, Jennifer. 2019. “What Is the Opec+ Alliance
and Its Impact on the Oil Industry?” The National. 2019.
https://www.thenational.ae/business/energy/what-is-the-
opec-alliance-and-its-impact-on-the-oil-industry-1.838468
Gately, Dermot. 1995. “Strategies for OPEC's Pricing
and Output Decisions.” The Energy Journal 16 (3): 1-38.
https://www.jstor.org/stable/41322604
Golombek, Rolf, Eystein Gjelsvik, and Knut
Einar Rosendahl. 1995. “Effects of Liberalizing
the Natural Gas Markets in Western Europe.”
The Energy Journal 16 (1): 85–112. https://doi.
org/10.5547/ISSN0195-6574-EJ-Vol16-No1-6
Golombek, Rolf, Alfonso A. Irarrazabal, and Lin
Ma. 2018. “OPEC’s Market Power: An Empirical
Dominant Firm Model for the Oil Market.” Energy
Economics 70 (February): 98–115. https://doi.
org/10.1016/j.eneco.2017.11.009
Hamilton, James D. 2003. “What Is an Oil Shock?”
Journal of Econometrics 113(2): 363–98. https://doi.
org/10.1016/S0304-4076(02)00207-5
Hart, Rob, and Daniel Spiro. 2011. “The Elephant in
Hotelling’s Room.” Energy Policy 39(12): 7834–38. https://
doi.org/10.1016/j.enpol.2011.09.029
Huppmann, Daniel, and Franziska Holz. 2012. “Crude Oil
Market Power—A Shift in Recent Years?” Energy Journal
33 (4): 1–22. https://doi.org/10.5547/01956574.33.4.1
IEA. 2019. “World Energy Outlook 2019.” https://www.iea.
org/countries/united-states
International Energy Agency (IEA). 2018. “Oil Market
Report - October 2018.” https://www.iea.org/reports/
oil-market-report-october-2018
———. 2019. “World Energy Outlook 2019.” https://www.
iea.org/reports/world-energy-outlook-2019
Javan, Afshin, and Nahl Zahran. 2015. “Dynamic
Panel Data Approaches for Estimating Oil Demand
Elasticity.” OPEC Energy Review 39(1): 53–76. https://
doi.org/10.1111/opec.12040
Jones, Clifton T. 1990. “OPEC Behaviour Under
Falling Prices: Implications For Cartel Stability.” The
Energy Journal 11(3): 117–29. https://doi.org/10.5547/
ISSN0195-6574-EJ-Vol11-No3-6
Kilian, Lutz. 2008. “The Economic Effects of Energy Price
Shocks.” Journal of Economic Literature 46(4): 871–909.
https://doi.org/10.1257/jel.46.4.871
Kleinberg, Robert L., Sergey Paltsev, Charles K.
Ebinger, David Hobbs, and Tim Boersma. 2018. “Tight
Oil Market Dynamics: Benchmarks, Breakeven Points,
and Inelasticities.” Energy Economics 70 (February):
70–83. https://doi.org/10.1016/j.eneco.2017.11.018
No Oil Producing and Exporting Cartels Act of 2007
(NOPEC). 2007. H.R. 2264, 110th Cong.
OPEC. 2017. 2017 OPEC World Oil Outlook. https://www.
opec.org/opec_web/static_les_project/media/down-
loads/publications/WOO%20%202017.pdf
Pierru, Axel, and Walid Matar. 2014. “The Impact
of Oil Price Volatility on Welfare in the Kingdom of
Saudi Arabia: Implications for Public Investment
Decision-making.” The Energy Journal 35(2): 97–116.
https://www.jstor.org/stable/24695762
29
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual SupplierCooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
Phillips, Peter C B, and Pierre Perron. 1988.
“Testing for a Unit Root in Time Series
Regression.” Biometrika 75 (2): 335–46.
http://www.jstor.org/stable/2336182?se-
q=1&cid=pdf-reference#references_tab_con-
tents%0Ahttp://about.jstor.org/terms
Plaut, Steven E. 1981. “OPEC Is Not a Cartel.” Challenge
24(5): 18–24. http://www.jstor.org/stable/40720003
Powell, Stephen G. 1990. “The Target
Capacity-Utilization Model of OPEC and
the Dynamics of the World Oil Market.” The
Energy Journal 11(1): 27-63. https://www.
jstor.org/stable/41323435
Powell, Stephen G. 1991. “A Risk Analysis of Oil
Development in the Arctic National Wildlife Refuge.”
The Energy Journal 12 (3): 55-76. https://www.jstor.org/
stable/41322428
Rauscher, Michael. 1988. “OPEC Behaviour and the
Price of Petroleum.” Journal of Economics 48(1): 59–78.
http://www.jstor.org/stable/41793887
Reuters. 2019. “U.S. House Panel Passes ‘NOPEC’ Bill,
Full House Vote Uncertain.” February 7, 2019. https://
af.reuters.com/article/commoditiesNews/idAFL1N2021LF
Rystad Energy. 2018a. “A System Dynamics Approach to
Model Future Oil Price Variability and the Impact of Swing
Producers.” https://communications.rystadenergy.com/
acton/fs/blocks/showLandingPage/a/12327/p/p-0060/t/
page/fm/0
———. 2018b. “NOPEC and Its Potential Impact on the
US E&P Sector.” https://www.rystadenergy.com/?cli-
ents-login&ReturnUrl=/clients/articles/oil-markets/2019/
nopec-scenario
Sheppard, David, and Anjli Ravel. 2019. “Saudi Arabia
to Urge Opec to Make Deeper Cuts in Production.”
Financial Times, December 5, 2019. https://www.ft.com/
content/900ed958-16b9-11ea-9ee4-11f260415385
Volkmar, Peter. 2019. “Is OPEC Dead Without
Russia? Shedding Light on the Question Using
Game Theory.” In Essays on Crude Oil Markets
and Electricity Access. Rice University. https://
scholarship.rice.edu/handle/1911/105919
References
30
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
⊂
⊂
⊂
⊂
⊂
Appendix A. Model Formulation
Indices
iAll projects existing and new
jAll projects with existing production capacity
kAll new projects that can be built
kL kAll new projects with long-term production cycles (no tight oil)
iR iProjects of the residual supplier, existing and new
iS iAll short-term tight oil projects existing (jS j)and new (kS k)
t,t' Years in the model {ts,ts+1,… tN} where ts is the start year and tN the last year modeled
(horizon). t' indexes years when new projects are built.
τAll years for projects operating beyond the model horizon {ts,…t∞}
∆t Shift in the projected production proles for new projects k, ∆t'=ts-t'. As a convention
the approval years of all new projects k are set to ts
Table A.1. List of indices, variables and parameters used in the model.
Coefcients
Supply coefcients
Ciτ Production cost proles projected for all projects in USD/bbl
Eiτ Projected production proles for existing projects in MMbbl
Fiτ Projected production proles for new projects in MMbbl
Kiτ
Annual projected capital development cost for each project i and year τ in million
USD
Κ tCap on the total discounted capital commited to all new long-term projects in year t
�
tCap on the total capital expenditures on all short-term tight oil projects in year t
rInterest rate used to discount future cash ows
Miτ
Index to correct the global market price (calibrated to Brent) for each asset based on
quality and regional markers
TkMinimum year when investment decision for new asset k can be made
Demand Curve
tReference GDP growth rate projected for all future years
D t Reference oil demand for each year in MMbbl
P t Moving average reference oil price projected for all future years in USD/bbl
tReference world GDP for the model start year in million USD
Variables
Primal
bkt' New project k built in year t' (scaled by the projected production prole)
qjt Quantity produced from existing asset j in year t in MMbbl
xkt' t Quantity produced in year t from new project k built in year in t' in MMbbl
31
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual SupplierCooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
"!
"!#"!
"
$!$&
),
∑!"!+ ∑#"!"
#"!≥ " ⊥ "≥ 0 ∀ (A.1)
$"
$"#$
= %"
%"#$
+ " (A.1)
Appendix A. Model Formulation
Dependent variable
Dt
Demand for oil as a function of the oil price, world GDP and GDP growth rate in
MMbbl
YtWorld GDP per year in million USD
P tMoving average of the market clearing price in USD/bbl
Here we complete the mathematical formulation of
the demand and supply models used to construct
an equilibrium model of the global oil market
formulated as a mixed complementarity problem
(MCP). All indices, variables and parameters used
in the model are shown in Table A.1.
First we introduce the independent demand
constraint in equation (A.1), complemented by
the market-clearing price pt. It sets the demand
equations from equation (1) as the lower bound on
the total supplies from existing and new projects,
qjt and xkt' t, respectively. The index j prepresents all
existing oil projects, and k ll new oil projects, while
t is the production year, and t' the build approval
year for new projects.
A.1 Estimation of θ and
calibration of At in the
demand equation
To estimate an approximate value of θ in Eq. (2)
1 In both ADF and PP tests the Akaike information criterion (AIC) is used to choose the lag length. The OLS
regression includes time dummies for the years 2009 and 2010 to control for the effects of the global nancial crisis.
While time dummies are signicant, the parameter θ is found to be insignicant at conventional levels. To ensure the
goodness of t of the model, a series of diagnostic and stability tests are also conducted. To conserve space, we do
not report the econometric results here. All unreported results are available from the authors upon request.
Dual
PtMarket clearing price, the marginal value of oil demand in USD/bbl
YkMarginal value on the constraint on investment in new projects i' (A3.1)
λjt Marginal value on the supply constraint for existing projects i used in t (A3.2)
μkt' t
Marginal value on the supply constraint for new projects i' built in t' and used in t
(A3.3)
∑!"!+ ∑#"!"
#"!≥ " ⊥ "≥ 0
∀
(A.1)
we consider the following equation:
(A.1)
where ut is an error term. To get Eq. (A.1) we
consider the growth equation given in Eq. (2)
(i.e. gt= ,take the natural logarithm of
both sides and replace g
t and by gt-1 and pt-1,
respectively. To avoid spurious regression, we test
for stationarity of the variables in Eq. (A.1) using the
augmented Dickey-Fuller (Dickey and Fuller 1981)
and the Phillips and Perron (1988) unit root tests.
As expected, both series are found to be stationary,
which allows us to estimate Eq. (A.1) by means of
ordinary least squares (OLS). The estimated value
for θ from the OLS regression is -0.2, which will
thus be used in the following simulations to replace
θ in Eq. (A1.1).1 For the sake of completeness,
the sensitivity of the results are evaluated against
a range of plausible values for θ (such as -0.5
and -0.8). The results of the sensitivity analysis
(unreported here) show that our simulation results
are robust to different values of θ.
32
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
Table A.2. IEA WEO 2019 oil demand, oil price (Brent) GDP, and corresponding calibration coefcients At .
Year Oil Demand, D
t (Mbbl/d) Oil Price, Pt ($/bbl) GDP PPP Growth (%) At ε=-0.25, γ=0.75
IEA EIA IEA EIA IEA EIA IEA EIA
2017 96.60 97.48 54.16 54.16 3.7 3.7
2018 99.20 99.90 70.81 70.81 3.7 3.7
2019 98.75 99.83 61.04 61.04 3.6 3.2 0.0390 0.0380
2020 99.83 102.20 65.53 63.00 3.6 3.5 0.0391 0.0384
2021 100.90 102.56 70.03 67.0 0 3.6 3.5 0.0385 0.0373
2022 101.98 102.92 74.52 68.00 3.6 3.3 0.0386 0.0368
2023 103.05 103.28 79.01 69.00 3.6 3.3 0.0386 0.0363
2024 104.13 103.64 83.51 70.00 3.6 3.2 0.0386 0.0357
2025 105.20 104.00 88.00 71.15 3.6 3.3 0.0385 0.0352
2026 105.70 104.36 89.60 72.31 3.6 3.2 0.0382 0.0346
2027 106.20 104.72 91.20 73.46 3.6 3.2 0.0377 0.0340
2028 106.70 105.08 92.80 74.61 3.6 3.2 0.0371 0.0335
2029 107.20 105.44 94.40 75.77 3.6 3.1 0.0365 0.0330
2030 107.70 105.80 96.00 76.92 3.6 3.1 0.0359 0.0325
A.2 The supplier’s optimization problem
(A3)
(A3.1)
(A3.2)
(A3.3)
(A3.4)
(A3.5)
(A3)
(A3.1)
(A3.2)
(A3.3)
(A3.4)
(A3.5)
The reference prices, demand, and GDP from the
WEO state policies scenario (IEA 2019) and IEO
reference scenario (EIA 2019) used to calibrate the
scaling coefcient At with a price elasticity of -0.25
and income elasticity of 0.75 are provided in Table
A.2. Notice that the slower average demand and
price growth in the IEO results in reduced scaling
coefcients, At.
Appendix A. Model Formulation
33
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual SupplierCooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
(A3)
(A3.1)
(A3.2)
(A3.3)
(A3.4)
(A3.5)
Appendix A. Model Formulation
The supplier’s optimization problem is represented
in equation block (A3). Oil producers make two
decisions for conventional oil, based on two cost
curves:
I. A cost curve for production qjt from existing
projects j over the period t constrained by the
existing capacity Ejt.
II. A cost curve for production xkt' t from the
development of new projects bkt' in year t', with
projected production capacity Fkt
Over the simulation period, dened by the index t,
suppliers maximize prots π of selling production
from existing and new projects. Net revenues are
calculated as the market price, pt, corrected by the
marker index Mit, less marginal production costs,
Cit. All revenues are discounted using the interest
rate, r, compounded for every year beyond the start
year of the simulation, ts. In the second revenue
term for new projects, we shift the projected
production cost prole Ckt+∆t' by ∆t'=ts-t'. By design
we assume the projected proles of all new
projects start in the rst year being modeled, year
ts, such that ∆t' is the same for all projects built in
t'. Note that the proles of new projects include the
approval year and development lead times ahead
of the production start.
In the third term we calculate the net revenues
from the production of new long-term projects
kc (excluding tight oil) beyond the model solve
period, τ>tN. Here we calculate the future revenues
assuming the oil price equals the equilibrium price
from the last year solved by the model, PtN. In the
last term we subtract out the corresponding total
discounted capital development cost Kkc of new
long-term project. The suppliers use this term to
select the projects that are economically viable; the
discounted net revenues exceed the discounted
capital costs over the lifetime of the project.
The investment variable bkt' is dened as a unit-less
non-negative number, approximating investment
decisions as a continuous variable. In Eq. (A3.1)
the upper bound on bkt is set to 1. This allows the
model to partially develop a new project in a given
year when it is the marginal supply unit, rather
than solving a more complex problem where bkt
is a binary decision equal to 0 or 1. Note that
new investments are limited to t'≥ Ti, where Ti
represents the minimum year when a new project
k can be approved for development, restricting the
start date of new projects.
The supply from existing and new projects are
constrained in (A3.2) and (A3.3), respectively.
Recall production and cost curves of new projects
are adjusted by the shift parameter, ∆t', to account
for delays in the decision to approve new projects.
34
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
To capture the short investment and production
cycles of new tight oil projects the model considers
that the producers make one decision only: to
develop new capacity if the full cost (operating
and development) are lower than the current price,
for this reason the capital development costs are
embedded in the cost parameter Cit dened as the
breakeven price for the tight oil play. Therefore, we
do not have explicit capital development costs for
tight oil projects in the objective function. Finally,
the approval year of these projects is assumed to
coincide with the rst year of production.
However, in some of the scenarios described in
Section 3 we do apply nancial constraints on
the total capital development costs for all tight oil
projects. This constraint is dened in equations
(A3.4), where �
t is the capital on total annual tight
oil capex. The two terms on the right-hand side
include the capex spent on the development of new
projects, and the capex spent on extracting tigh oil
from existing projects, respectively.
Equation (A3.5) adds a cap on the total
discounted capital that can be committed to all
new long-term oil projects in a given year, dened
by the coefcient Κ t.
A.3 Optimality conditions of the supplierʼs problem
(A4)
(A4.1)
(A4.2)
(A4.3)
(A4.4)
(A4.5)
(A4.6)
(A4.7)
(A4.8)
(A4.9)
Appendix A. Model Formulation
35
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual SupplierCooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
"!≥ ! ⊥
!
"≥ 0 ∀
+!≤ !
⊥
!
#≥ 0
∀
"!≥ !
⊥
!
"≥ 0
∀
+!≤ ! ⊥
!
#≥ 0 ∀
In order to solve the supply and demand
problems simultaneously we formulate an
equilibrium model consisting of the demand
function (A1) and demand constraint (A2) with the
optimality conditions of the supplier’s problem
dened in (A4). Equations (A4.6), (A4.7), (A4.8)
and (A4.9) represent the optimality conditions
for the primal variables qjt, xkt,t', bkL t' and bkS t'
respectively. They are expressed in terms of the
coefcients of the objective function and the dual
variables λjt, μkt' t, γk, φt and σt from the original
primal constraints, (A4.1), (A4.2), (A4.3), (A4.4)
and (A4.5), respectively.
When the capital constraints (A4.4) and (A4.5)
are binding the dual variables φt and σt represent
the additional cost of tight oil projects, and
new long-term investments, respectively. On
the left-hand side of optimality condition (A4.8)
σt is added to the discounted cost of capital,
representing an adjustment to the protability of
projects when the capital constraint is reached.
Similarily, φt is added to the production cost of
exiting tight oil projects for existing projects in
(A4.6) and development cost of new projects in
(A4.9).
When solving the residual supplier scenario we
use complementary slackness to incorporate
price ceiling (A4.10) and oor (A4.11) constraints
into the optimality conditions of the supplierʼs
problem. These constraints signal the residual
supplier to add or remove capacity, rt
+and rt
-, in
response to the production from the competitive
fringe to achieve the desired ceiling or oor,
represented by the coefcients Pt and P t,
respectively.
(A4.10)
(A4.10)
The variables rt
+and rt
- are dened as duals
on the price constraint, and are used to adjust
the total production required from the residual
suppliers assets. For example, rt
- represents
the aggregate spare capacity that can be
distributed across the supplierʼs assets in
several ways. For example, withhold the most
expensive assets (revenue maximization), or by
distributing it relative to the share of capacity by
each asset. The optimal strategy will depend on
various technical charateristics and operational
requirements of the supplier not accounted for
in our model. We choose a strategy to distribute
spare capacity across all assets, reecting the
average production cost of the residual supplier.
Given that the price target is xed, the fringe and
residual supplier’s production level can also be
evaluated using a pure accounting approach.
However, we leverage the competitive market
model to evaluate production and investment
decisions, including constraints, to avoid
developing additional accounting logic. This
approach also allows for exibility in how the
residual supplier targets the market price, for
example only setting a price oor.
A.4 The recursive problem
The full MCP consisting of (A1), (A2) and (A4)
over the time period t = {ts ,t s+1,…tN } can either
be solved as a single problem, or recursively
for several smaller time periods t of size n,
t ={ts ,t s+1,…ts+N }. All t and t' are replaced by t and
t ’, with n<N and s denoting the start year in each
recursive step. After solving the equilibrium model
for the rst period we perform the recursive
operations outlined in (A5) to update the model
coefcients, move forward the start year, s=s+1,
and repeat until we reach the last period covered
by the model, s+n=N.
Appendix A. Model Formulation
36
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
(A5)
= ∪ ; !"!
"> 0 (A5.1)
!" = (
.!"#∆""!"!
"!,"#∆""+ !"!")/(!"!
"!,"#∆"!
"+ !") ∀ ≥ & (A5.2)
!" = !" + !"!
"!,"#∆"" ∀ ≥ & (A5.3)
!= !+ !"!
" ∀ (A5.4)
= ;
!= 1
(A5.5)
= + 1 (A5.6)
(A5)
= ∪ ; !"!
"> 0 (A5.1)
!" = (
.!"#∆""!"!
"!,"#∆""+ !"!")/(!"!
"!,"#∆"!
"+ !") ∀ ≥ & (A5.2)
!" = !" + !"!
"!,"#∆"" ∀ ≥ & (A5.3)
!= !+ !"!
" ∀ (A5.4)
= ; != 1 (A5.5)
= + 1
(A5.6)
(A5)
= ∪ ; !"!
"> 0 (A5.1)
!" = (
.!"#∆""!"!
"!,"#∆""+ !"!")/(!"!
"!,"#∆"!
"+ !") ∀ ≥ & (A5.2)
!" = !" + !"!
"!,"#∆"" ∀ ≥ & (A5.3)
!= !+ !"!
" ∀
(A5.4)
= ; != 1 (A5.5)
= + 1 (A5.6)
(A5)
= ∪ ; !"!
"> 0 (A5.1)
!" = (
.!"#∆""!"!
"!,"#∆""+ !"!")/(!"!
"!,"#∆"!
"+ !") ∀ ≥ & (A5.2)
!" = !" + !"!
"!,"#∆"" ∀ ≥ & (A5.3)
!= !+ !"!
"
∀ (A5.4)
= ; != 1 (A5.5)
= + 1 (A5.6)
(A5)
= ∪ ; !"!
"> 0 (A5.1)
!" = (
.!"#∆""!"!
"!,"#∆""+ !"!")/(!"!
"!,"#∆"!
"+ !") ∀ ≥ & (A5.2)
!" = !" + !"!
"!,"#∆"" ∀ ≥ &
(A5.3)
!= !+ !"!
" ∀ (A5.4)
= ; != 1 (A5.5)
= + 1 (A5.6)
(A5)
= ∪ ; !"!
"> 0 (A5.1)
!" = (
.!"#∆""!"!
"!,"#∆""+ !"!")/(!"!
"!,"#∆"!
"+ !") ∀ ≥ & (A5.2)
!" = !" + !"!
"!,"#∆""
∀ ≥ & (A5.3)
!= !+ !"!
" ∀ (A5.4)
= ; != 1 (A5.5)
= + 1 (A5.6)
(A5)
= ∪ ; !"!
"> 0 (A5.1)
!" = (
.!"#∆""!"!
"!,"#∆""+ !"!")/(!"!
"!,"#∆"!
"+ !")
∀ ≥ &
(A5.2)
!" = !" + !"!
"!,"#∆"" ∀ ≥ & (A5.3)
!= !+ !"!
" ∀ (A5.4)
= ; != 1 (A5.5)
= + 1 (A5.6)
(A5)
= ∪ ; !"!
"> 0 (A5.1)
!" = (
.!"#∆""!"!
"!,"#∆""+ !"!")/(!"!
"!,"#∆"!
"+ !") ∀ ≥ & (A5.2)
!" = !" + !"!
"!,"#∆"" ∀ ≥ & (A5.3)
!= !+ !"!
" ∀ (A5.4)
= ; != 1 (A5.5)
= + 1 (A5.6)
(A5)
= ∪ ;
!"!
"> 0
(A5.1)
!" = (
.!"#∆""!"!
"!,"#∆""+ !"!")/(!"!
"!,"#∆"!
"+ !") ∀ ≥ & (A5.2)
!" = !" + !"!
"!,"#∆"" ∀ ≥ & (A5.3)
!= !+ !"!
" ∀ (A5.4)
= ; != 1 (A5.5)
= + 1 (A5.6)
in Eq. (A5.4) we introduce the unitless coefcient
Bk that keeps track of all new build decisions from
previous start years, initialized to 0. Any new
project built to completion (Bk=1) are removed
from the subset k in (A5.5). Under the recursive
solution approach we add Bkt' to the left-hand
side of (A4.3) since we are solving the model over
the reduced time period t that excludes past start
years. Finally, we move forward to the next start
year, solve the equilibrium problem, and repeat.
A.5 The supply data (Rystad
UCube)
The Rystad UCube upstream database is
constructed using a bottom-up approach based
on private sector and government reporting
and calibrated using various other sources for
country-level production. It includes more than
21,000 individual assets, with historical data
starting from the year 1900 and projected data
up to year 2100, including production proles,
operating costs, and investment plans. It also
includes lead times between approval year (rst
year of development) and start-up (rst year of
production).
For each asset, the approval year shows
which year the asset was, or is expected to
be, sanctioned for development in UCube.
OPEX values were extracted in USD/bbl. At the
beginning and end of the projected production
prole, OPEX values were abnormally high due to
dividing costs by expected low production in early
and late production years. These outliers were
replaced with average OPEX for each asset using
statistical analysis.
The number of years solved during each step
denes the horizon over which suppliers make
decisions, and can be adjusted to represent
myopic or forward-looking supplier behavior.
Under a myopic approach suppliers incorporate
a limited amount of information or expectations
about demand and production in future years.
In the extreme case, suppliers only consider
decisions made in the current start years, s=n.
(A5)
(A5.1)
(A5.2)
(A5.3)
(A5.4)
(A5.5)
(A5.6)
In (A5.1) we include all new projects i' that are
built in the current start period (bi' ts'>0) to the set
of existing projects. In (A5.2) the projected cost
proles for new projects built in t' are adjusted
based on the shift parameter ∆t'=ts-t'. We dene
Ckt as a copy of the original Ckt, to track cost
proles of new projects partially built in different
years. Notice the updated cost parameter is
a weighted average of the production prole
corresponding to the current build (bkts' Fk,t+∆t') of a
new project and the existing production resulting
from the partial build of the same project in past
years, where Ekt, is initially 0. In (A5.3) we add
the production proles for new projects built in
the current start year to the existing capacity
parameter Eit, again accounting for the shift
parameter ∆t'.
Appendix A. Model Formulation
37
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual SupplierCooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
Gas condensate elds
All hydrocarbon liquids are extracted from
UCube, including crude oil, condensate, NGLs,
renery gains, and other liquids. Liquids
produced from gas-condensate elds (gas
elds with condensate-to-gas-ratio exceeding
1 bbl/MMcf) are assumed to be a byproduct
with no additional costs in our model. In reality,
processing the liquids from gas elds will
have some operating cost and perhaps some
capital expenditure; however, as a simplifying
assumption, we disregard these costs.
Tight oil elds
Tight oil eld costs are treated separately in the
model. Tight oil elds represent an aggregation
of many wells drilled at a projected schedule
depending on the forecasted oil price. As such,
each new well can be treated as having its own
break-even price and as a new investment.
Due to the inability to disaggregate proles for
individual wells from the extracted UCube data,
we chose to use the full-cycle break-even prices.
Break-even prices are calculated on an asset
level by estimating the oil prices that give an NPV
of zero based on future free cash ow. Cash
ows incorporate all production costs (CAPEX
and OPEX) as well as any government taxes. A
discount rate of 10% is applied to calculate the
NPV. Decommissioning and abandonment costs
are not included in break-even calculations (see
Rystad).
We use the Rystad data to calculate a price
correction term for all hydrocarbon liquids at the
project level based on the API, other discount
elements (sulfur content, etc.), and regional price
markers. The term is calibrated with respect to
Brent prices used to calibrate the demand curve
used in our model. Condensate, NGL and gas
prices are estimated within UCube based on
dened links to oil prices.
Liquids production (condensate, NGLs, renery
gains, and other liquids) are also based on
UCube projections for existing producing elds
developed prior to 2018.
Appendix A. Model Formulation
38
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
Appendix B. Sensitivity Analysis: Price
Elasticity of Demand
The detailed results from the equilibrium problem
solved in the competitive scenario under the
Rystad investment plan are listed in Table B.1,
alongside the reference values from IEA’s stated
policies scenarios.
Table B.1. Scenario results for demand, price and GDP in the competitive market scenarios under the Rystad
investment plan.
Demand (MMbbl/d) Oil price ($/bbl) GDP Growth (%)
Year IEA
WEO
ε =
-0.1
ε =
-0.25
ε =
-0.5
IEA
WEO
ε =
-0.1
ε =
-0.25
ε =
-0.5
IEA
WEO Growth ε=
-0.1
ε =
-0.25
ε =
-0.5
2018 99.2 70.81 131,908
2019 98.8 98.9 99.0 99.2 61.0 57.9 58.8 59.5 136,657 3.7% 3.4% 3.4% 3.4%
2020 99.8 100.8 101.7 102.5 65.5 51.4 54.0 57.1 141,576 3.7% 3.5% 3.4% 3.4%
2021 100.9 102.8 104.0 104.6 70.0 56.9 61.7 66.4 146,673 3.6% 3.5% 3.5% 3.4%
2022 102.0 104.5 105.6 105.8 74.5 60.8 68.0 72.0 151,953 3.6% 3.6% 3.5% 3.4%
2023 103.1 105.8 106.9 107.1 79.0 59.9 65.0 69.2 157,424 3.6% 3.6% 3.5% 3.5%
2024 104.1 107.5 108.7 109.2 83.5 60.2 69.1 75.0 163,091 3.6% 3.6% 3.5% 3.5%
2025 105.2 109.2 110.4 110.8 88.0 64.8 75.9 82.5 168,962 3.6% 3.6% 3.5% 3.5%
2026 105.7 109.7 110.7 110.9 89.6 69.9 76.6 81.1 175,045 3.6% 3.6% 3.5% 3.5%
2027 106.2 110.0 110.9 110.9 91.2 73.3 78.9 84.3 181,346 3.6% 3.6% 3.5% 3.5%
2028 106.7 109.9 110.6 110.6 92.8 83.4 87.2 90.9 187,875 3.6% 3.5% 3.5% 3.4%
2029 107.2 109.8 109.8 109.8 94.4 88.4 93.3 92.2 194,638 3.6% 3.5% 3.4% 3.4%
2030 107.7 109.2 109.1 109.2 96.0 102.9 95.6 94.8 201,645 3.6% 3.4% 3.4% 3.4%
In Figure B.1 we present the results for total oil
demand and price under different assumptions
for the price elasticity of demand with no residual
supplier. As expected, we nd a much faster price
recovery in the no residual supplier scenario
when the absolute value of the elasticity is higher
and consumers are more responsive to the shift
in OPEC supplies. Notice that when consumers
are less responsive to the structural change, over
time the equilibrium price grows at a rapid rate
and surpasses reference. This occurs as a result
of the slow price recovery delaying the approval
of new projects as scheduled under the Rystad
investment plan, and therefore available capacity
towads the end of the decade.
Figure B.2 shows the same set of scenarios
as Figure B.1 after applying the 50% cap on
investments in tight oil projects. Notice that the
price volatility increases as the absolute value of
the elasticity is reduced. Since supplies are tighter
in these scenarios, reduction in consumer price
response and associated changes in production
decisions have a more pronounced impact on the
market equilbira over a shorter time period (e.g.
three years). Also, the total demand converges in
each case as a result of the elevated price levels
and nearly identical investment decisions.
39
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual SupplierCooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
Figure B.2. World oil price and demand for the no residual supplier scenario under the Rystad investment plan with
the 50% tight oil investement cap for different price elasticity assumptions.
45
55
65
75
85
95
105
115
125
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
Price, US$/b
98
100
102
104
106
108
110
112
114
116
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
Demand, MMb/d
Residual supplier (WEO stated policies)
Competitive, = -0.1
Competitive, = -0.25
Competitive, = -0.5
Residual supplier (WEO stated policies)
Competitive, = -0.1
Competitive, = -0.25
Competitive, = -0.5
To assess the sensitivity of our model to the
reference case (IEA NPS) we also calibrate the
demand curve to the EIA’s IEO. We compare the
results with the original IEA calibration in Figure
B.3, when applying the US$150 billlion cap on
long-term investments and the tight oil cap.
Clearly the EIA adopts a more moderate growth
trajectory overall with prices consistently below
WEO. Following accelerated demand growth in the
year 2020, it is consistently slower over the next
10 years, with total demand falling to 2 MMbbl/d
below WEO in 2030.
Appendix B. Sensitivity Analysis: Price Elasticity of Demand
Figure B.1. World oil price and demand for the competitive scenario under the Rystad investment plan for different
price elasticity assumptions.
45
55
65
75
85
95
105
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
Price, US$/b
Residual supplier (WEO stated policies)
Competitive, = -0.1
Competitive, = -0.25
Competitive, = -0.5
98
100
102
104
106
108
110
112
114
116
Demand, MMb/d
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
Residual supplier (WEO stated policies)
Competitive, = -0.1
Competitive, = -0.25
Competitive, = -0.5
40
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
Overall, the IEO calibration produces both lower
and slightly more stable oil prices. A contributor
to the lower price volatility is that demand is
slightly lower under the IEO calibration, and
therefore supplies are less tight. Keep in mind
this is assuming that investors spend on average
US$150 billion per year in both scenarios.
Figure B.4 depicts Saudi Arabia’s liquids
production under different investment constraints.
Notice that under the US$100 billion investment
cap (Figure 4a) with the tight oil cap (hollow
dashed line), by the end of the decade Saudi
Arabia’s production exceeds the total sustainable
production in the competitive scenario. As a lone
residual supplier, Saudi Arabia would have to
expand its existing production capacity to meet
the demand (and price) outlook of the WEO. A
more realistic expectation is that the residual
supplier would produce at the sustainable
production levels, and a different price equilibrium
would be reached.
Appendix B. Sensitivity Analysis: Price Elasticity of Demand
Figure B.3. World oil price (a) and demand (b): Comparing results under the IEA NPS and EIA IEO reference
calibration. All scenarios apply the US$150 billion cap on new long-term investment and the 50% cap on short-term
tight oil. Sources: IEA World Energy Outlook, KAPSARC analysis.
50
60
70
80
90
100
110
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
Price, US$/b
98
100
102
104
106
108
110
112
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
Demand, MMb/d
Residual supplier (EIA IEO ) Residual supplier (IEA WEO)
Competitive tight oil cap (EIA IEO)
Competitive tight oil cap (IEA WEO)
Competitive (EIA IEO)
Residual supplier (EIA IEO ) Residual supplier (IEA WEO)
Competitive tight oil cap (EIA IEO)
Competitive tight oil cap (IEA WEO)
Competitive (EIA IEO)
41
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual SupplierCooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
Figure B.4. Total liquids production by Saudi Arabia in the residual supplier scenarios, under different investment
constraints. Dashed lines depict production by Saudi Arabia when it is the residual supplier without support from
OPEC. Shaded areas show total production in the competitive scenarios.
Competitive
Residual supplier (all OPEC)
Residual supplier (only Saudi Arabia)
Residual supplier tight oil cap (only Saudi Arabia)
(a) US$100 billion cap
(b) US$125 billion cap
(c) US$100 billion cap
Competitive
Residual supplier (all OPEC)
Residual supplier (only Saudi Arabia)
Residual supplier tight oil cap (only Saudi Arabia)
Competitive
Residual supplier (all OPEC)
Residual supplier (only Saudi Arabia)
Residual supplier tight oil cap (only Saudi Arabia)
6
8
10
12
14
16
18
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
MMb/d
6
8
10
12
14
16
18
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
MMb/d
6
8
10
12
14
16
18
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
MMb/d
Competitive
Residual supplier (all OPEC)
Residual supplier (only Saudi Arabia)
Residual supplier tight oil cap (only Saudi Arabia)
(a) US$100 billion cap
(b) US$125 billion cap
(c) US$100 billion cap
Competitive
Residual supplier (all OPEC)
Residual supplier (only Saudi Arabia)
Residual supplier tight oil cap (only Saudi Arabia)
Competitive
Residual supplier (all OPEC)
Residual supplier (only Saudi Arabia)
Residual supplier tight oil cap (only Saudi Arabia)
6
8
10
12
14
16
18
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
MMb/d
6
8
10
12
14
16
18
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
MMb/d
6
8
10
12
14
16
18
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
MMb/d
Competitive
Residual supplier (all OPEC)
Residual supplier (only Saudi Arabia)
Residual supplier tight oil cap (only Saudi Arabia)
(a) US$100 billion cap
(b) US$125 billion cap
(c) US$100 billion cap
Competitive
Residual supplier (all OPEC)
Residual supplier (only Saudi Arabia)
Residual supplier tight oil cap (only Saudi Arabia)
Competitive
Residual supplier (all OPEC)
Residual supplier (only Saudi Arabia)
Residual supplier tight oil cap (only Saudi Arabia)
6
8
10
12
14
16
18
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
MMb/d
6
8
10
12
14
16
18
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
MMb/d
6
8
10
12
14
16
18
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
MMb/d
Appendix B. Sensitivity Analysis: Price Elasticity of Demand
42
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
About the Authors
Bertrand Rioux
Bertrand is a senior research associate focusing on the impact of market
regulation and liberalization in energy markets. An experienced energy
systems model developer (linear optimization and mixed complementary
problems), he is working on developing the KAPSARC Energy Model
(KEM) as a decision support tool for analyzing price regulation in energy
economies. Bertrand has contributed to the development of KEM Saudi
Arabia and is the lead developer of KEM China, studying the impact of
government regulation in the coal, power and natural gas markets. He
was previously employed as a research assistant at the Canadian Space
Agency.
Axel Pierru
Axel is the director of KAPSARC’s Energy and Macroeconomics program.
From October 2018 to March 2019, he was KAPSARC’s interim vice
president of research. Axel joined KAPSARC in 2011, after spending
15 years at IFP Energies Nouvelles in France, where he led research,
consulting and training projects. Axel received his Ph.D. in economics
from the Pantheon-Sorbonne University in Paris. He undertakes applied
research that combines methodological innovation with practical relevance
for policymaking. His research is focused on energy economics, policy,
nance, oil pricing, and energy-exporting economies. Axel has been
published extensively, with over 40 peer-reviewed journal papers to his
name.
Fatih Karanl
Fatih is a research fellow at KAPSARC. He received his Ph.D. in economics
from the Pantheon-Sorbonne University in Paris and his M.A. in economic
analysis and modeling jointly from the Pantheon-Sorbonne University and
the École Centrale Paris. Before joining KAPSARC in December 2017,
Fatih was a research fellow at EconomiX-CNRS and an associate professor
of economics at the University of Paris, Nanterre, where he taught
econometrics, energy economics, and environmental economics. Fatih’s
current research focuses mainly on developing economic frameworks
to provide insights into energy policymaking in oil-producing countries.
His research has been published in general-interest economics, journals
(e.g., Applied Economics, Journal of Comparative Economics, and
Macroeconomic Dynamics), as well as journals on energy economics (e.g.,
Energy Economics, Energy Policy, and The Energy Journal).
43
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual SupplierCooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
Shahd Al Rashed
Shahd is a senior research associate in the Markets and Industrial
Development program. A mechanical engineer by background, she
previously worked at Chevron as a are and relief systems engineer and a
facilities engineer supporting upstream and downstream Chevron facilities
worldwide, with consultations and technical services based in Houston,
Texas and Bakerseld, California. Shahd holds a master’s degree in
mechanical engineering with a focus on advanced energy systems
from UC Berkeley and a B.S. degree in mechanical engineering with a
concentration in manufacturing engineering from Boston University.
Abdullah Al Jarboua
Abdullah is a senior research analyst in the Energy and Macroeconomics
program with interest in developing energy systems and energy-economic
models. He holds a master’s degree in Computer Science from King
Abdullah University of Science and Technology and a B.S. degree in
Computer Engineering from Tennessee Technological University.
Colin Ward
Colin was the interim director of the Markets and Industrial Development
program. He has worked in the energy industry for 10 years in various
capacities, including seismic eld work, renery design and consulting for
major international oil companies and national oil companies worldwide.
Colin played a major role in several KAPSARC projects, primarily focusing
on cost estimation for energy projects and environmental impacts of the
global energy industry.
About the Project
This project investigates a world oil market where there is no residual supplier, such as OPEC,
that adjusts its production to inuence prices. A partial equilibrium model is used to simulate oil
supply and demand dynamics assuming perfect competition among all producers. The model
provides a detailed representation of supplies, characterized as longer-term conventional
and shorter-term tight oil production, that can be used to construct a variety of production and
investment scenarios. It is also used to compare the nancial implications of transitioning to a
competitive oil market for the current residual supplier, OPEC, and Saudi Arabia as its largest
contributor.
44
Cooperate or Compete? Insights from Simulating a Global Oil Market with No Residual Supplier
www.kapsarc.org
