The Oil Security Metrics Model

1
The Oil SecurityMetrics Model

• David L. Greene
• Paul N. Leiby
• Oak Ridge National Laboratory
• A presentation to the
• IWG GPRA USDOE
• March 6, 2005
• Washington, DC

2
OSMM estimates oil security benefits of changes
in the U.S. oil market.

• Based on the NAS Committees framework that
  distinguishes energy security benefits from
  normal market benefits.
• Includes uncertainty about future oil market
  conditions via up to four alternative AEO
  projections.
• Simulates supply disruptions calibrated to
  historical deviations of OPEC supply from AEO
  projections.
• Technologies change both the level of demand
  its response to oil prices. Both are estimated.
• Security benefits, are estimated as the
  difference between normal market (AEO Reference
  Case) benefits and benefits in alternative
  futures that allow supply disruptions.

3
The OSMM does not

• Predict technological or market success of DOEs
  RD programs or how technology would change w/o
  DOE RD efforts.
• Predict impact on U.S. petroleum consumption (the
  VISION Model is used for this).
• Estimate impacts of DOE RD on the probability,
  size or timing of oil market disruptions. It
  should.
• Estimate defense or foreign policy costs in
  dollars.
• Incorporate a global demand response to DOEs RD
  achievements.

4
The NRC (2005) made a distinction between normal
and disrupted market benefits.

• Economic net benefits are based on changes in
  the total value of goods and services that can be
  produced in the U.S. economy under normal market
  conditions, (NRC, 2005, p. 29)
• Security net benefits are based on changes in
  the probability or severity of abnormal
  energy-related events that would adversely impact
  the overall economy, public health and safety, or
  the environment. (NRC, 2005, p. 29)
• At present, the OSMM does not estimate the
  changes in the probability of supply disruptions.
• The definition appears to omit insurance costs
  of oil security (e.g. SPR, defense costs) during
  undisrupted periods.

5
The cartelized, volatile oil market produces
three direct costs to the U.S. economy.

• Loss of potential GDP due to greater economic
  scarcity of oil.
• Transfer of wealth due to monopoly pricing and
  price shocks.
• Dislocation losses of GDP due to oil price
  shocks.

Transfer of wealth is not a loss of GDP but a
change in the ownership of GDP. It can occur in
disrupted and undisrupted markets and occurs
whether or not OPEC is the cause of the
disruption.
6
The direct economic costs of oil dependence have
been greatest during supply disruptions.
7
Implementing as a spreadsheet model with Monte
Carlo simulation software allows price shocks,
OPEC reaction parameter uncertainty to be
represented.

• Economic net benefits in normal market are
  obtained by using the undisrupted AEO cases (or
  Reference Case), with and without DOE program
  impacts.
• Economic Security benefits are obtained by
  running Monte Carlo simulations, with and without
  DOE programs, using probability distributions
  for
• Occurrence of a disruption
• Timing of a disruption
• Size of a disruption
• Security net benefits
• (Economic Security benefits)
• Economic net benefits

8
The impacts are tested in 10,000 futures.

• Randomly Choose
• Oil Market Scenario
• Calibrate WOM
• High Oil Price
• Reference Oil Price
• Low Oil Price
• Parameters

• Adjust for Technology Impacts
• Reduced oil demand
• Increased oil supply
• Changes in price elasticities

Generate Stochastic Oil Supply Disruption
Iterate to 10,000

• Compute Oil Dependence Costs
• Transfer of Wealth
• Reduced Potential GDP
• Disruption Costs

Select OPEC Strategy Solve New Oil Market
Equilibrium
Distribution of Oil Dependence Costs by Year
9
The impacts of reducing US oil use, on imports
and oil prices can be bounded.

• The key question is, What will OPEC do?
• Maintain production
• World oil price falls
• US imports depend on elasticities
• OPEC market share increases
• Maintain the price of oil
• US supply unchanged
• US imports fall
• OPEC market share decreases
• Increase production?
• Decrease production more than enough to maintain
  the previous price?

10
The supply shock simulation model creates
projections more consistent with recent history.
Supply shocks are deviations from AEO projected
OPEC supply.
11
Each oil market future chooses an AEO Case in
which there may be oil supply disruptions that
generate price shocks.
12
Expected hybrid vehicle benefits alone 191B
PV, assuming OPEC maintains its production plan
(Ref. Proj. 137 direct oil savings). By NAS
definition, security benefits are 191B - 137B
54B.
13
If OPEC maintains the price path, expected
benefits are 120B PV, versus Ref. Proj. 71B
direct oil savings. Security benefits are 120B
- 71B 49B. (Versus 54B)
14
In addition to monetary benefits, a number of
non-monetary metrics are calculated.
15
The OSMM rigorously estimates the prospective oil
security benefits of EERE RD programs.

• Using the NAS definition of security benefits.
• Reflecting key uncertainties.
• Oil market conditions via AEO Cases
• Supply disruptions
• Parametric uncertainties, too.
• Includes price elasticity as well as demand
  reduction impacts.
• Vehicle technology elasticity impacts included
• Biofuel supply impacts to come.

16
THANK YOU.
17
U.S. oil imports in the base case, supply shock
and EERE impacted cases.
18
The impact of EERE technology on world oil price.
19
The impact on OPECs market share.
20
The impacts on OPEC gross revenues.
21
The impacts on U.S. oil consumption.
22
The impact on U.S. wealth transfer.
23
The impact on the size of the SPR.
24
And, the impact on the price elasticity of U.S.
oil demand.
25
One can measure how advanced technologies expand
the fuel economy-cost envelope and increase the
price elasticity of oil demand.
26
Technology changes the price elasticity of MPG,
which changes the price elasticity of gasoline
demand.
Energy Efficient technology clearly affects the
long-run price elasticity of demand. The
short-run impact must be carefully considered.
27
The impacts of alternative and replacement fuels
on price elasticity can be similarly estimated.
ßs are price elasticities, bs price slopes, ss
are market shares, g, r, and f indicate gasoline,
replacement fuels, and all motor fuel.

• Given VISION program impact estimates of
  alternative fuel market shares, elasticity
  changes over time can be calculated.
• The time trend in elasticities can be entered
  into the oil market simulation model by modifying
  the price slope of the US oil demand equation.

28
Each direct cost component can be estimated by a
relatively simple equation.

• Total costs
• Wealth Transfer
• Loss of potential GDP (dynamic)
• Disruption losses (dynamic)

29
SPR benefits are readily monetizable either as a
reduction in costs or an increase in insurance
value.(Leiby Bowman, 2000)
30
Reduced costs of an SPR sized to replace 57 days
of imports are expected to be 2.4B, PV.
31
The proposed method doesnt include dollar
estimates for defense or foreign policy costs.

• It should.
• Instead, the model can provide indicators
• OPEC revenues
• OPEC market share
• US petroleum imports
• US petroleum consumption
• When used in simulation mode, these would be
  probability distributions.

32
OPEC oil revenues are an indicator of monopoly
rents to oil producers.
33
Net US imports are an output of the oil market
model. OPEC or Persian Gulf imports are not.
34
The world oil market changed profoundly in 1973.
35
OPEC is an imperfect, partial market monopoly
cartel.

• It influences oil markets via decisions on
• Long-run production capacity
• Short-run production
• Its ability to collude is limited.
• Its control of markets is limited.
• It has a strong incentive to price above MC

P monopoly price C competitive price ß
demand price elasticity s OPEC mkt. share q
ROW supply Qo OPEC supply r rate of demand
growth d rate of ROW supply decline
36
Short- and long-run elasticities bound the region
in which OPEC can operate.
Key Assumptions Linear lagged adj. Supply
Elasticities L.R. 0.60 S.R. 0.06
1-lambda 0.90 Demand Elasticities L.R.
-0.70 S.R. -0.10 1-lambda
0.85 Competitive price 13/bbl.
37
It is not likely that the problem of OPEC market
power will go away soon. (3/11/04).
38
Dermot Gately has convincingly argued that OPEC
will not expand its capacity to meet the worlds
demands at a low price.
39
If Middle East producers do not increase output
by 80, unconventional oil production may have to
expand rapidly. The M.E. can retain a third of
the market as the lowest cost producer through
2050.
40
Security benefits can also be estimated for a
portfolio of EERE technologies, but this has not
yet been done.
41
Expected oil dependence costs under BAU 2 of
GEP with a 90 C.I. of 0.8-3.5 of GDP.(Interior
interval /- 1 std. dev., exterior interval
5 to 95 C.I.)
of GDP
of GDP
42
A one-time single-focus policy is insufficient
Raising LDV fuel economy to 35 MPG by 2017, then
stopping, lowers the cost range to 0.5 to 3.0.
of GDP
of GDP
43
The result is nearly unchanged if OPEC chooses to
maintain output.
of GDP
of GDP
44
The NCEP strategy falls just short of the
independence goal. More is needed, and progress
must be sustained beyond 2030.
45
Oil independence works regardless of OPECs
response strategy.
46
A simple oil market model that can be calibrated
to any AEO scenario can be constructed with four
linear equations.

• US Demand
• ROW (incl. OPEC) Demand
• US Supply
• ROW (excl. OPEC) Supply
• OPEC Supply assumed exogenous.

47
National defense and foreign policy costs are
controversial, difficult to measure, but not zero.

• Two wars since 1990 not entirely unrelated to oil
  security.
• Protection of oil supplies has some impact on US
  military expenditures.
• Transfer of wealth
• Provides surplus that can be spent on military
  build-up but also support for terrorism.
• A blessing or a curse?
• Assuming these costs to be zero is not defensible
  but, at present, OSMM does not estimate national
  security costs.

48
The OSMM does estimate the effects of EEREs RD
programs on the

• Economic costs of oil supply disruptions,
• Wealth transfer
• Potential GDP losses
• GDP dislocation costs
• Potential GDP and wealth transfer costs during
  normal market periods,
• Levels of US oil imports,
• World oil prices,
• OPEC market share and market power and
• SPR costs or benefits.