Petroleum Property Valuation

JAMES L. SMITH
Southern Methodist University
Dallas, Texas, United States

nondiversifiable risk A cause of unpredictable financial

1. Impact of Property Valuations on the performance that tends to impact all of an investor’s
Petroleum Industry holdings in the same direction or manner.
proved reserves The volume of petroleum resources in a
2. Discounted Cash Flow (DCF) Analysis: The
developed field that are reasonably expected to be
Problem Simplified
recoverable given current technology and prices.
3. Special Characteristics of Petroleum Properties
real options The general phenomenon by which the value
4. Incorporating the Value of Real Options of physical assets depends on, and therefore may be
5. Portfolio Analysis: The Combined Value of enhanced by, management’s ability to modify or
Multiple Properties postpone investment and operating decisions based on
6. Conclusion the receipt of new information.
risk premium That portion of the discount rate that
compensates the investor for the inherent unpredict-
ability of future financial returns, as opposed to the pure
time value of money.

Glossary
capital asset Any equipment, facility, or plant capable of
generating a long-lived stream of future income.
discounted cash flow (DCF) A method for estimating the
present value of future cash flows that adjusts for the Petroleum property valuation refers to the analytical
time value of money and the degree of uncertainty procedure by which the commercial value of oil and
surrounding future receipts. gas fields is assessed. This assessment provides to
discount rate The factor by which expected receipts in a prospective buyers and sellers, and other interested
future period are reduced to reflect the time value of
parties such as lenders and tax assessors, an estimate
money and unpredictable variation in the amount
of the fair market value of underground deposits of
ultimately received.
diversifiable risk A source of unpredictable financial oil and gas—the amount for which they might be
performance that varies randomly from one investment bought or sold. The value of an underground deposit
to another and therefore averages out, rather than is directly related to the ultimate value of whatever
accumulates, over all of an investor’s holdings. petroleum may be extracted in the future, but
fair market value The price expected to be paid for any because the future is uncertain, the value of the
asset in a voluntary exchange between an independent property is subject to various sources of risk that
buyer and independent seller. stem from geological as well as economic factors. To
Monte Carlo analysis A method for assessing the magni- be useful, the valuation procedure must take proper
tude and implications of risk by simulating possible account of the unpredictable fluctuations that would
outcomes via random sampling from the probability
cause field development, operations, and perfor-
distribution that is assumed to control underlying risk
factors.
mance to deviate from the expected outcome. This
net present value (NPV) A measure of the value of a task represents a complex analytical problem that
project obtained by discounting the stream of net cash has challenged traditional valuation methods. To
flows (revenues minus expenditures) to be received over meet the challenge, new and highly innovative
the entire life of the project, based on the time profile techniques have been developed and are finding
and riskiness of net receipts. widespread use.

Encyclopedia of Energy, Volume 4. r 2004 Elsevier Inc. All rights reserved. 811
812 Petroleum Property Valuation

1. IMPACT OF PROPERTY 2.1 Projecting Cash Flows

VALUATIONS ON THE The projected cash flow stream is a composite
PETROLEUM INDUSTRY forecast that results from many separate assumptions
concerning physical attributes of the oil field and the
The petroleum industry’s reliance on accurate and economic environment in which it will be produced.
reliable valuation methods is apparent. The risk of The number of wells and size of facilities required to
paying too much for an acquired property, or selling delineate and develop the field, in conjunction with
for too little, is always present. The frequency and the presumed cost level for drilling services and oil
size of property transactions underscores the im- field equipment, will roughly determine the scope and
portance of getting the valuations right. Since 1979, timing of initial expenditures. The magnitude and
for example, more than 5000 parcels of existing oil duration of cash inflows (sales revenue minus
and gas fields have been sold in the United States, operating cost) are determined by a further set of
bringing more than $600 billion of revenue to the assumptions regarding the flow rate from individual
sellers. The negotiations that support these ex- wells (and the rate at which production will decline as
changes hinge on finding valuation methods that the field is depleted), the quality and price of
both sides deem acceptable. In addition, investments produced oil and gas, necessary operating and
are made by the petroleum industry to acquire leases maintenance costs required to keep wells and field
in raw acreage where no oil or gas field is known to plant facilities in order, and the level of royalties and
exist. The U.S. government is a major source of such taxes that are due to lessors and governmental
properties and has raised in excess of $60 billion authorities. Thus, the projection of net cash flow for
since 1954 (when the federal leasing program was the field as a whole is the combined result of many
initiated) by auctioning petroleum exploration and interrelated but distinct cash flow streams. Some
development rights on federally owned lands and the components are fixed by contract and can typically be
outer continental shelf. The petroleum industry projected with relative certainty (e.g., royalty obliga-
makes comparable investments on a regular basis tions and rental payments), but others require trained
to acquire oil and gas leases from private landowners guesswork that leaves a wide margin of error (e.g.,
and the states, as well. The ability to value petroleum future production rates and oil price trends). It seems
properties accurately therefore plays a critical role in reasonable that those components of the cash flow
determining the financial success or failure of oil and stream that are known with relative certainty be given
gas producers. greater weight in figuring the overall value of the
field, but as discussed further later in this article,
properly executing this aspect of the valuation
procedure was hardly practical until so-called op-
2. DISCOUNTED CASH FLOW tions-based valuation methods were developed.
(DCF) ANALYSIS: THE
PROBLEM SIMPLIFIED 2.2 Discounting Cash Flows
In some respects, an oil field is no different than any A dollar received (or paid) in the future is worth less
other capital asset and valuation techniques for than a dollar received (or paid) today because of the
petroleum properties are therefore similar to time value of money. Cash in hand can be invested to
procedures used in other sectors of the economy. A earn interest, and therefore will have grown in value
capital asset represents any long-lived investment in to outweigh an equivalent amount of cash to be
productive facilities that have the potential to received at any point in the future. If the relevant
generate a future stream of earnings. If those periodic rate of interest is represented by the symbol i
earnings are not large and predictable enough to (e.g., i ¼ 10%), then the present value of a dollar to
justify the initial expenditure, the investment be received t periods hence is given by PV(i, t) ¼
should not be made. Intuitively, the value of the 1/(1 þ i)t. This expression is referred to as the
capital asset may be thought of as the extent to which discount factor, and i is said to be the discount rate.
anticipated cash receipts outweigh the initial expen- The discount factor determines the relative weight to
diture. Measuring and weighing the projected cash be given to cash flows received at different times
flows therefore forms the heart of the valuation during the life of the oil field. Cash flows to be
problem. received immediately are given full weight, since
 Petroleum Property Valuation 813

PV(i, 0) ¼ 1, but the weight assigned to a future from property to property. A completely riskless cash
receipt declines according to the amount of delay. flow stream (which is rare) should be discounted using
Thus, the net present value (NPV) of an arbitrary the risk-free rate, which is approximated by the
cash flow stream represented by the (discrete) series interest rate paid to the holders of long-term govern-
of periodic receipts {CF0, CF1, CF2,..., CFT} is ment bonds. Cash flow streams that are more risky,
computed as the sum of individual present values: like the future earnings of a typical oil field, must be
X T discounted at a higher rate sufficient to compensate
CFt
NPV ¼ t: ð1Þ for the owner’s aversion to bearing that risk. The
t¼0 ð1 þ iÞ degree of compensation required to adequately adjust
It is quite common to perform this computation on for risk is referred to as the risk premium and can be
the basis of continuous discounting, where the estimated from market data using a framework called
periodic intervals are taken to be arbitrarily short the capital asset pricing model.
(a day, a minute,..., an instant), in which case the One important implication of the capital asset
discount factor for cash to be received at future time t pricing model is that diversifiable risks do not
declines exponentially with the length of delay: contribute to the risk premium. A diversifiable risk
PV(i,t) ¼ eit. Therefore, when the cash flow stream is any factor, like the success or failure of a given well,
is expressed as a continuous function of time, NPV is that can be diluted or averaged out by investing in a
reckoned as the area under the discounted cash flow sufficiently large number of separate properties. In
curve: contrast, risks stemming from future fluctuations in
Z T oil prices or drilling costs are nondiversifiable because
NPV ¼ CFt  eit dt: ð2Þ all oil fields would be affected similarly by these
0 common factors. The distinction between diversifi-
It is apparent, whether the problem is formulated able and nondiversifiable risk is critical to accurate
in discrete or continuous time, that correct selection valuation, especially with respect to the exploratory
of the discount rate is critical to the valuation segment of the petroleum industry: although petro-
process. This parameter alone determines the relative leum exploration may be one of the riskiest busi-
weight that will be given to early versus late cash nesses in the world, a substantial portion of those
flows. Since exploration and development of oil and risks are diversifiable, thus the risk premium and
gas fields is typically characterized by large negative discount rate for unexplored petroleum properties is
cash flows early on, to be followed after substantial relatively low in comparison to other industries.
delay by a stream of positive cash flows, the choice of The appropriate discount rate for the type of
a discount rate is decisive in determining whether the petroleum properties typically developed by the
value of a given property is indeed positive. The major U.S. oil and gas producers would be in the
extent to which discounting diminishes the contribu- vicinity of 8 to 14%. This is the nominal rate, to be
tion of future receipts to the value of the property is used for discounting cash flows that are stated in
illustrated in Fig. 1, which shows the time profile of current dollars (dollars of the day). If future cash flow
cash flows from a hypothetical oil field development streams are projected in terms of constant dollars
project. With no other changes to revenues or (where the effect of inflation has already been
expense, the property’s net present value is reduced removed), then the expected rate of inflation must
by a factor of ten, from nearly $1 billion to roughly be deducted from the nominal discount rate, as well.
$100 million, as the discount rate is raised from 8% Cash flow streams derived from properties owned by
(panel a) to 20% (panel b). These panels also smaller or less experienced producers who are unable
illustrate how discounting affects the payback period to diversify their holdings, or in certain foreign lands,
for the property in question (i.e., the time required may be deemed riskier, in which case the discount rate
before the value of discounted receipts finally offsets must be increased in proportion to the added risk.
initial expenditures): 7 versus 11 years at the res- Not only does the appropriate discount rate vary
pective rates of discount. according to property and owner, but the individual
With so much at stake, the selection of a discount components of overall cash flow within any given
rate cannot be made arbitrarily. If the discount rate is project are likely to vary in riskiness and should, in
not properly matched to the riskiness of the particular principle, be discounted at separate rates. It is fair to
cash flow stream being evaluated, the estimate of fair say, however, that methods for disentangling the
market value will be in error. Because no two oil fields separate risk factors are complex and prone to error,
are identical, the appropriate discount rate may vary and it is common practice to discount the overall net
814 Petroleum Property Valuation

A
1000
800
600
400

$ Million
200
0
−200
−400
−600
−800
−1000
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Project year

Cash flow Discounted cash flow Cum NPV

B
1000
800
600
400
$ Million

200
0
−200
−400
−600
−800
−1000
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Project year

Cash flow Discounted cash flow Cum NPV

FIGURE 1 Discounting diminishes the contribution of future receipts. The illustrations show a hypothetical net cash flow
stream, heavily negative at the outset, followed by consecutive years of positive operating revenues. When discounted at the
higher rate, it takes longer for the same revenue stream to offset initial expenditures, resulting in a lower cumulative net present
value. (A) Net cash flow and NPV at 8% discount rate. (B) Net cash flow and NPV at 20% discount rate.

cash flow stream at a single rate that reflects the pattern and behavior of cash flows, and therefore
composite risk of the entire project. In many value. Although the following factors are not
applications, the error involved in this approxima- necessarily unique to the valuation of petroleum
tion is probably not large. Moreover, new insights properties, their influence is of sufficient importance
regarding the valuation of real options (to be to justify more detailed discussion.
discussed later) provide a procedure in which it is
appropriate to discount all cash flow streams at the
same rate, which circumvents entirely the problem of 3.1 Exploration and Development Risk
estimating separate risk-adjusted discount factors.
Whether at the stage of exploration or development,
investments made for the purpose of exploiting an
underground petroleum deposit often go awry.
3. SPECIAL CHARACTERISTICS OF Technical failures or economic circumstances may
PETROLEUM PROPERTIES block the recovery of any resources from the
property in question. During the past 30 years,
Most oil and gas fields share certain physical and 72% of all exploration wells and 19% of all
economic characteristics that strongly influence the development wells drilled in the United States have
 Petroleum Property Valuation 815

resulted in so-called dry holes, which is industry reserves are not entirely free of risk due to the
parlance for a well that is unsuccessful. The causes of continuing possibility that price and cost fluctuations
failure are numerous, ranging from barren geological will impact future cash flows. These factors represent
formations, to deficiencies in the quality of the nondiversifiable sources of risk that influence selec-
deposit that preclude recovery at reasonable cost, to tion of the discount rate. In practice, the value of
the technical failure or breakdown of drilling proved reserves reported in a producer’s financial
equipment. In all such cases, the initial investment statements must be reestimated and updated each
is forfeited—written off as the cost of a gamble. year to reflect the impact of price and cost fluctua-
The risk of dry holes is incorporated in the tions and any new conclusions about reservoir
valuation process directly, by assigning appropriate behavior that emerge from ongoing field operations.
weight to a zero-payoff outcome. After this mod- Resources categorized as probable and possible
ification, the NPV expression given previously, are successively further removed from having been
compare Eq. (2), would appear thus: tested, let alone proven, by the drill bit. Their
valuations are accordingly subject to increasing
XT
CFt (and potentially very large) margins of error. In
NPV ¼ CF0 þ ð1  pDH Þ  t; ð3Þ
t¼1 ð1 þ iÞ
addition, the reader must be warned that, although
this resource classification scheme has been endorsed
where CF0 represents the cost of drilling, pDH by the World Petroleum Congress, adherence to the
represents the probability of a dry hole, and the definitions may vary in different parts of the world.
{CF1,..., CFT} represent expected future cash flows
contingent on success of the well. If the risk of failure
is diversifiable, which is certainly true of drilling 3.2 Oil and Gas Equivalents
undertaken by large, publicly owned companies, it
does not contribute to the risk premium associated Many petroleum properties contain natural gas (and
with the property, which means that the discount various natural gas liquids) as well as oil, a
rate (i) would not be affected by the presence or size circumstance that complicates the valuation process.
of pDH. Although the thermal energy content of a barrel of
Drilling risk can take more complex and subtle oil is roughly six times that of an mcf (thousand
forms than the simple dichotomy between success cubic feet) of gas, the market values of the two rarely,
and failure. Outcomes (size of deposit, daily flow if ever, stand in that ratio, as illustrated in Fig. 2.
rates, gas/oil ratio, etc.) depend on many underlying Lack of pipeline facilities is one factor that tends to
factors that are not known with certainty but which depress the value of gas deposits, which are more
are amenable to probabilistic analysis. The influence difficult and costly to bring to market than oil. The
of uncertainty regarding these factors can be unconcentrated form in which natural gas occurs
quantified via Monte Carlo analysis, wherein pro- (low energy density per unit volume) also renders it
jected cash flows and the implied value from Eq. (3)
are recomputed under a broad range of possible
Oil ($/bbl) relative to gas ($/mcf)

scenarios. This exercise yields a probability-weighted 20

average outcome, which is the best single indicator of 18
property value. Monte Carlo analysis also reveals the 16
14
potential range of error in the valuation due to 12
uncertainty in the underlying geological and eco- 10
nomic parameters. 8
A particular nomenclature developed by the 6
4
petroleum industry permits the degree of exploratory 2
and development risk associated with a given 0
property to be quickly assessed. The ‘‘proved
Jan-76
Jan-78
Jan-80
Jan-82
Jan-84
Jan-86
Jan-88
Jan-90
Jan-92
Jan-94
Jan-96
Jan-98
Jan-00
Jan-02

reserves’’ category is the most certain because it

includes only those resources that have already been
delineated and developed and shown to be economic- FIGURE 2 Historical fluctuation in the price of oil relative to
natural gas. If the two fields traded at parity in terms of raw energy
ally recoverable using existing technology under content, one barrel of oil would sell for roughly six times as much
prevailing market conditions. Although the outcome as one mcf of natural gas. Historically, oil has tended to trade at a
of drilling may already have been resolved, proved premium, but the ratio is highly variable.
816 Petroleum Property Valuation

unsuitable for many uses (e.g., as transport fuel), and 22% and 36% of their respective wellhead prices.
this tends to further depress its value. Consequently, Second, the in situ values are much more stable than
the relationship between the value of oil and gas wellhead prices. The year-to-year price change for oil
deposits is not stable but fluctuates markedly through in the ground averages (in absolute value) roughly
time and space, depending on the relative demand 13%, versus 20% for changes in price at the
for, and supply of, the two fuels in regional markets. wellhead. For gas, the contrast is even stronger:
The value of any petroleum property will depend year-to-year price changes averaged 11% in the
on the specific quantities of oil versus gas that are ground versus 24% at the wellhead.
present. While it is common to see properties The relationship between in situ values and
described in terms of the combined amount of ‘‘oil wellhead prices, although complex and ever-chan-
equivalents’’ (which usually means that each mcf of ging, can be understood via a simple model of
gas has been counted as one-sixth of a barrel of oil— production from a developed oil field. Let q0
based on the heat equivalencies of the two fuels), represent the initial level of production, which is
there is no reliable basis for such aggregation, to the presumed to decline continuously at the rate a due to
extent that the chairman of a leading international natural pressure loss in the reservoir as depletion
oil company has proclaimed that oil- and gas- proceeds; thus production at time t is given by qt ¼ q0
equivalents simply do not exist. What was meant, eat. Assume further that the expected wellhead price
of course, is that any calculation of oil equivalents is of oil remains fixed over the relevant horizon at P per
a garbling of information that only serves to obscure barrel, and that unit production costs amount to C
the value of the underlying property. Oil equivalents per barrel. Using Eq. (2), the net present value of the
cannot be compared for purposes of valuation to an property can then be calculated:
equal volume of oil reserves—their values would not Z T
be the same. Nor can the oil equivalents of one NPV ¼ q0 ðP  CÞ  eðaþiÞt dt
property be compared to the oil equivalents of any Z0 N
other property where oil and gas reserves are present E q0 ðP  CÞ  eðaþiÞt dt
in different proportions. To do so risks a gross 0
miscalculation of value. As long as consumers q0 ðP  CÞ
¼ ; ð4Þ
continue to distinguish between the two types of aþi
hydrocarbons, so must the owners and operators of where the approximation is justified by the fact that
petroleum properties. oil fields are long lived. The volume of reserves (R) in
the deposit is given by total production over the life
of the property:
3.3 The Volatility of Commodity Prices Z T Z N
q0
and Property Values R¼ q0  eat dtE q0  eat dt ¼ ; ð5Þ
0 0 a
Compared to most commodities, oil and gas exhibit which means the rate of extraction from reserves is
highly volatile price movements. Daily, annual, and given by the decline rate: q0 ¼ aR. After substituting
monthly swings are unusually large relative to the this expression for q0 into (4) and dividing by R, we
base price levels for both fuels, which puts a large obtain the in situ value (V) of a barrel of reserves:
portion of the value of any petroleum property at
NPV a
risk. V¼ ¼ ðP  CÞ: ð6Þ
The price of a barrel of oil at the wellhead differs, R aþi
however, from the value of a barrel of oil in the Equation (6) says quite a lot about the value of a
ground, primarily because the reserve cannot be producing property. To be concrete, let us set the
produced and delivered to a buyer instantaneously. production decline rate equal to the discount rate
This difference is evident in Fig. 3, which contrasts (10% is a realistic number for both), and set
annual changes in the value of oil and gas reserves production costs equal to one-third of the wellhead
with corresponding changes in the wellhead price price. After simplification, the relationship between
levels of these two commodities. Two things are in situ values and wellhead prices then reduces to
apparent. First, in situ values (i.e., the value of V ¼ P/3, which illustrates the petroleum industry’s
petroleum reserves in the ground) are much smaller traditional one-third rule: The value of a barrel in the
than wellhead prices. Over the past 10 years, oil and ground is worth roughly one-third of the price at the
gas reserves have sold on average for only about wellhead.
 Petroleum Property Valuation 817

A
$40 30%

$35
25%
$30
20%
$25

% of WTI
$/barrel
$20 15%

$15
10%
$10
5%
$5

$0 0%
1991 92 93 94 95 96 97 98 99 00 01
Oil reserves WTI Ratio

B
$5.00 60%
$4.50
50%
$4.00
$3.50

% of Henry hub
40%
$3.00
$/mct

$2.50 30%
$2.00
20%
$1.50
$1.00
10%
$0.50
$0.00 0%
1991 92 93 94 95 96 97 98 99 00 01
Gas reserves Henry hub Ratio

FIGURE 3 Value of oil and natural gas, in situ versus wellhead. In situ valuations are determined by the average price of
reserve transactions tabulated by Cornerstone Ventures, L. P. Wellhead values are determined by the prices of WTI (oil) and
Henry Hub (gas) contracts on the New York Mercantile Exchange. From Exhibits 4 and 5, Cornerstone Ventures, ‘‘Annual
Reserve Report, 2001’’ (dated February 28, 2002), with permission.

Like any rule of thumb, the one-third rule is often What remains to be seen is why in situ values are
wrong, as the numbers in Fig. 3 demonstrate, but it less volatile than wellhead prices. According to the
does point to a general tendency. Moreover, the one-third rule, every 10% rise in wellhead price
derivation provided in Eqs. (4) to (6) allows one to would be matched by a 10% rise in the value of
anticipate when and why deviations would arise. reserves: the volatilities should be the same, but they
Reserves that are extracted more rapidly (like natural are not. The explanation stems from the nature of
gas, for example) would tend to sell for more than commodity price movements, and the difference
one-third of the wellhead price. To see why, simply between random walk and mean-reverting processes,
substitute a4i into (6). This confirms a pattern that as illustrated in Fig. 4. A random walk process tends
was evident in Fig. 3: for gas, the value of reserves is to wander off, rather than to return to its starting
consistently a larger fraction of wellhead price than point. Any chance departure from the existing price
for oil. It is also evident that the value of reserves level tends to become permanent. A mean-reverting
should move inversely with the level of operating process tends to be self-correcting; any succession of
costs, which is why mature fields are eventually upward price movements increases the chance of
abandoned as it becomes more expensive to extract future downward movements, which are required to
the oil. restore the price to its former level.
818 Petroleum Property Valuation

A
$50
$45
$40
$35
$30
$25
$20
$15
$10
$5
$0
0 12 24 36 48 60 72 84 96
Monthly observations

B
$50
$45
$40
$35
$30
$25
$20
$15
$10
$5
$0
0 12 24 36 48 60 72 84 96
Monthly observations
FIGURE 4 The illustrations show five examples each of a random walk sequence and a mean-reverting sequence.
Simulations performed by the author. (A) Random walk price trends. (B) Mean-reverting price trends.

Whereas returns on investments in the stock 3.4 The Relationship between Reservoir
market tend to follow a random walk, the prices of Engineering and Property Valuation
major commodities appear to be mean reverting,
which is consistent with the view that the forces of To this point, we have taken the projection of future
supply and demand tend to keep commodity prices production, and therefore costs and revenues, as
from drifting permanently away from their equili- being determined exogenously. Subject to certain
brium levels. Mean reversion also implies that short- physical constraints, however, the rate of production
term fluctuations in the price of oil and gas at the is actually determined by petroleum engineers who
wellhead are likely to be reversed in due course. Since design and install facilities with a view to maximiz-
the value of a reserve is determined by a combination ing the value of the field. Property valuation there-
of current and future prices, the long-term stability fore rests implicitly on the assumption that
provided by mean reversion tends to dampen the production operations are optimized, and that
impact of short-term commodity price movements. process of optimization must itself be conducted
Equation (6), which uses a single value (P) to within the valuation framework.
represent both the current and future commodity To illustrate, let us return to the previous example
price level, is unrealistic in this regard; if prices do of an oil field subject to exponential decline. Based
not follow a random walk, it gives accurate valua- on our assumption that the discount rate and decline
tions only when the current wellhead price happens rate both equal 10%, and that operating costs
to correspond to the long-term level. amount to one-third of the wellhead price, we
 Petroleum Property Valuation 819

determined the value of the property to be P/3 per method. The options approach is a simple but
barrel of reserves. Now, imagine that our team of brilliant innovation that was devised initially to
reservoir engineers has identified an alternative value certain types of purely financial assets (e.g.,
drilling pattern that would double the extraction stock market puts and calls). When extended to the
rate (a ¼ 20%), with no sacrifice in total reserve problem of valuing tangible investments (bricks and
volume—the same amount of oil would be produced, mortar, steel and concrete), the technique is referred
but faster. However, this alternative development to as the real options approach. The profound
strategy would also require the expenditure of an importance and broad impact of these advances in
additional $2 per barrel of reserves. Should it be valuation methodology were quickly recognized, as
adopted by management? If so, what would be the reflected by the award of the Nobel Memorial Prize
impact on the value of the property? in Economics in 1997.
The valuation framework, represented in this case In some situations, the real options approach
by Eq. (6), supplies the answer to these questions. circumvents all three of the difficulties noted
After incrementing the extraction rate to 20%, but previously: it simplifies the problem of adjusting for
leaving all else unchanged, we find the value of the risk, provides a suitable forecast of future prices, and
reserve to be 4P/9 under the alternative drilling dispenses with any rigid or preconceived timeline for
pattern, an increase of P/9 per barrel. This is a project activities. The last aspect is especially critical
sensible strategy only if the gain outweighs the and gives the method its name. As noted in the
incremental cost of $2 per barrel. Thus, we are preceding section, a portion of the value of any
justified in pursuing the faster, but more costly, property is dependent on managerial flexibility in the
production program if, but only if, the wellhead price design and timing of project components. The
is expected to exceed $18/barrel. options approach assumes not that management will
Although the example may seem overly simplified, precommit to a fixed and rigid schedule of drilling
it illustrates an essential point: the value of a and production, but that instead management will
petroleum property is neither fixed nor guaranteed, react rationally to future events as the project
and certainly it is not determined by geology and unfolds. Pertinent decisions can be taken at numer-
commodity prices alone. Value depends on manage- ous points in the execution of any project, and the
ment’s willingness to identify alternative develop- essence of the options approach is to recognize that
ment concepts and production strategies, and the management will make those decisions when the
ability to adapt flexibly to changes in the economic time comes using the information then on hand—and
environment. Management that falls short in this not before.
regard is bound to leave some portion of a property’s A simple example gives the flavor of this
potential value on the table. approach. Consider an owner who holds a 2-year
lease on an undeveloped oil field, one that has been
appraised as holding 100 million barrels of recover-
4. INCORPORATING THE VALUE able oil, but the required wells and production
OF REAL OPTIONS facilities have not yet been installed. Suppose
installation of productive capacity is estimated to
The discounted cash flow (DCF) technique is cost $5.50 per barrel of reserves. The owner of the
versatile, but not without limitations. To project lease then holds a development option: by incurring
and properly discount future cash flows requires a an expenditure of $550 million, the owner will
forecast of petroleum prices, some disentangling of acquire 100 million barrels of developed reserves. If
myriad risk factors that impinge on individual the value of a developed reserve (in situ) is expected
components of the cash flow stream, and a correct to exceed $5.50 per barrel, the expenditure seems
view as to when each step in the enterprise will justified; otherwise it does not.
transpire. If prices were stable, these requirements As we have seen (see Fig. 3), there is an active
would be less of a burden. For the petroleum market in developed reserves. Suppose those transac-
industry, however, and particularly since the rise of tions reveal the value of developed reserves to be, say,
OPEC in the 1970s, the degree of guesswork and only $5 per barrel. Moreover, suppose the historical
resulting scope for error can be painfully high. volatility seen in that market indicates a 50% chance
To alleviate these problems, advances have that the value of developed reserves will rise or fall
exploited the options approach, a technique devel- by $1 each year (a random walk). Thus, the value of
oped in the 1970s as an alternative to the DCF developed reserves is expected to remain at $5 in the
820 Petroleum Property Valuation

future, albeit with unpredictable variations around information whether or not to exercise the option to
that level. develop the reserves.
It might appear that development of the reserves in At the end of the second year, as the lease is about
question would be inadvisable, and that a 2-year lease to expire, that decision is straightforward. If the
on the property would therefore have no value. It is value has reached $7, then development generates an
certain, at least, that immediate development of the immediate profit of $1.50 per barrel. The only
reserves would effect a $50 million loss ($500 million alternative is to allow the lease to expire, which
minus $550 million), and that prospects for develop- would generate no profit. If the value were lower ($5
ment are not expected to improve on average for the or $3), however, then it would be better to hold off—
duration of the lease. To conclude from these facts, which guarantees zero profit but avoids a loss. Thus,
however, that the lease has no value is to ignore the if the reserves had not already been developed, it is
value of active management. If provisions of the lease clear how management should proceed when the
do not compel management to develop the reserves, lease is about to expire. In the diagram, the boldface
then the lease has considerable option value and entry at each decision node reflects the optimal
should command a relatively high price despite the choice of action, either X for exercise or H for
sometimes unfavorable environment. Indeed, the fair holding off. The number shown beside each symbol
market value of the property would amount to $31 represents the profit that would be earned via that
million. course of action.
This result is obtained by a straightforward Knowing the end point is critical because it allows
application of the options approach, as diagrammed us, by working backward, to evaluate the property at
in Fig. 5. The figure shows a binomial tree that charts each of the earlier decision nodes as well. Suppose,
the possible future values of developed reserves. for example, that we find at the end of the first year
Starting from $5 per barrel, the value would either the value of developed reserves has risen to $6.
rise or fall (with equal probability) to $6 or $4 after Immediate development would generate a profit of
the first year. Continued volatility could carry it to $0.50 per barrel. The alternative is to hold off, to
either $3 or $7 by the end of the second year, but wait and see what the second year may bring. But we
the most likely value would be $5 (because there are have already evaluated the two possibilities: reserve
two price paths leading to that level). Each box in the value will either rise to $7 (which allows a profit of
tree represents a decision node: a point at which $1.50) or fall to $5 (where profits are $0). The two
management must decide on the basis of available possibilities are equally likely, so the expected value
is simply the average, $0.75. Even after discounting
at 10% to compensate for the extra year’s delay, the
Property value = $31 million present value of holding off at the end of year 1 is
$0.68 ( ¼ $0.75/1.10), which exceeds the profit from
V = $7.00
immediate development. Thus, at the decision node
X = $1.50 in question, although the option to develop immedi-
1/2
H = $0.00 ately is said to be ‘‘in the money,’’ it should not be
V = $6.00
X = $0.50 exercised. This illustrates a more general principle
H = $0.68
1/2
that is not so easy to accommodate within the DCF
V = $5.00 1/2 V = $5.00 framework: delay may be advisable even when
X =−$0.50 X =−$0.50
H = $0.31 H = $0.00 immediate action appears profitable.
1/2 1/2
V = $4.00 By the same routine, and always working right to
X =−$1.50 left, the other nodes can be completed. Of particular
H = $0.00 1/2 V = $3.00 interest is the first node, which represents the
Discount rate = 10% X =−$2.50 property’s current valuation based on all available
H = $0.00
Development cost = $5.50/barrel information. While it was evident before we began
Option value = $0.31/barrel that immediate development would bring a loss of
FIGURE 5 Oil field development option. The value of a 2-year $0.50 per barrel, we now see that price volatility
field-development lease is calculated by working backward, from (and management’s capacity to react appropriately
right to left, through the binomial tree. The owner must decide, at to future price changes) adds value to the property.
each node, whether it is more profitable to develop the reserves
immediately or to hold off. In this example, the value of developed The value of $0.31 per barrel that we now assign
reserves is assumed to follow a random walk, starting from the is the present value of the average of the two
level of $55 per barrel. outcomes that will be realized by the end of the
 Petroleum Property Valuation 821

first year (either $0.68 or $0.00), each discounted at of such properties raises additional issues, some of
the rate of 10% to allow for one year’s delay: which push to the very limits of modern techniques.
$0.31 ¼ 12($0.68 þ $0.00)/1.10. It is useful to distinguish three cases, based on the
The options approach provides answers to some extent to which the properties are related and
additional questions that would be difficult to whether or not they can be exploited sequentially.
address via the DCF method. Specifically, the
relationship between the length of lease (term to
expiration of the development option), the degree of 5.1 Properties with Independent Values
price volatility, and property value is developed If the values of the separate properties are believed to
explicitly. Extending the lease term (i.e., adding be statistically independent, then the single-property
nodes) and/or increasing the degree of future price methods described previously can be applied directly
volatility (i.e., spreading the tree) has the effect of and not much else need be said. The value of the
increasing the upside potential of a given property whole portfolio would equal the sum of the parts. To
and can only increase (never decrease) its value. It is satisfy the independence criterion, however, the
a simple matter to reconstruct and recompute the outcome (i.e., net cash flow) of each property must
binomial tree under varied assumptions and thereby be uncorrelated with the others. If there are common
chart the impact of these parameters. economic risk factors (e.g., price and cost levels) on
The method could have been illustrated just as well which all the properties depend, their values are
using a trinomial tree, for which the price movement unlikely to be independent. Common geological
at each node is either up, constant, or down. In factors could also create dependence, as when an
practice, the time step between nodes is taken to be exploratory failure on one property is deemed to
relatively small, in which case the final results are decrease the probability of success on others.
invariant to the particular structure of branching that
is adopted. Regarding the discount rate and volatility
parameters that are required to value the develop- 5.2 Properties with Dependent Values
ment option, it has been shown that if the analyst
follows a certain formulation to measure the volati- This case seems more complex, and therefore
lity (range and probability of future price movements) potentially more interesting. However, dependence
from historical market prices, then it is appropriate to among properties does not by itself necessarily
use the risk-free rate of discount. This aspect of the require any revision to the valuation method. The
options approach frees the analyst from the need to whole will still be equal to the sum of the parts, at
separately figure risk adjustment factors for each least if the properties will be exploited simulta-
component of the cash flow stream using the capital neously. By this, we mean that knowledge of the
asset pricing model, and from the necessity of outcome from any one property is not available in
preparing a subjective forecast of future prices. Many time to alter management’s plan for exploiting the
variations on the basic option framework are others. Thus, the properties are operated as if their
possible, including, for example, applications to outcomes are independent, even if an underlying
unexplored properties, already-producing properties, correlation does exist.
and special formulations that are designed to capture
random walk, mean-reverting, and other types of
5.3 Dependent Properties
price fluctuations in the underlying resource.
Exploited Sequentially
If outcomes are dependent and it is possible to
exploit the properties sequentially, then the valuation
5. PORTFOLIO ANALYSIS: THE problem changes fundamentally. Management will
COMBINED VALUE OF seek to use whatever information is gleaned from
MULTIPLE PROPERTIES earlier outcomes to enhance subsequent decisions.
Statistical dependence implies the existence of
We have so far examined the problem of valuing a relevant information spillovers that facilitate this
single petroleum property—one oil field considered practice. Thus, it is possible for the value of the
in isolation of other similar properties that might be portfolio to exceed the sum of the individual parts—
included in a transaction or already held by a the value of acquired information making up the
potential buyer. Valuing a collection, or portfolio, difference. Application of the techniques discussed
822 Petroleum Property Valuation

previously therefore provides only a lower bound for gambles constantly beset the petroleum business,
the combined value. and such factors will always challenge the accuracy
Valuation models that incorporate information of even the most advanced valuation methods.
spillovers can become enormously complex as the Although margins of error are inherently large, it is
number of properties increases. The central issue can not too much to ask that petroleum property
be illustrated, however, by a simple example. valuations be correct at least on average. Analytical
Consider an oil producer who owns two properties, methods have made some marked progress toward
neither of which has been drilled. Let the cost of each that goal, but the remaining obstacles are not
exploratory well be $2 million and the value of each inconsequential. We can reasonably expect the quest
underground deposit (if confirmed by exploratory for improved valuations of petroleum properties to
success) be $10 million. Finally, we assume that dry sustain basic research into some of the most
hole risk is 40% for each property. What is the value fundamental methods of financial economics well
of this portfolio and of its two components? into the future.
Based on individual analysis, the value of each
component would appear to be $4 million ( ¼ 0.6 
$10$2). By taking the sum of its parts, the value of Acknowledgments
the portfolio would then be appraised at $8 million. If
I sincerely thank G. Campbell Watkins and an anonymous referee
the drilling outcomes of the two properties are for reading an earlier draft of this article and for their many helpful
independent, or if they are dependent but drilled suggestions.
simultaneously, this is a correct analysis and we are
finished. But, suppose the drilling results are highly
dependent (perhaps both geological prospects can be
traced to a common sedimentary source), such that SEE ALSO THE
the outcome of the first foretells the outcome of the FOLLOWING ARTICLES
second. If the first confirms a deposit, of which the
probability is 60%, the other would also be drilled Depletion and Valuation of Energy Resources 
and the second deposit confirmed too, giving a Discount Rates and Energy Efficiency Gap  Energy
combined value of $16 million ( ¼ $10$2 þ Futures and Options  Investment in Fossil Fuels
$10$2). On the other hand, if the first well fails, Industries  Markets for Petroleum  Oil and Natural
of which the probability is 40%, the second would not Gas: Economics of Exploration  Oil and Natural
be drilled, limiting the combined loss to $2 million. Gas Liquids: Global Magnitude and Distribution 
By recognizing and exploiting the information Value Theory and Energy
spillover, management will on average make $8.8
million ( ¼ 0.6  $16–0.4  $2), which exceeds by
10% the combined value of the individual properties. Further Reading
To achieve this result, however, the properties must Adelman, M. A. (1990). Mineral depletion, with special reference
be exploited sequentially, rather than simultaneously. to petroleum. Rev. Econ. Stat. 72(1), 1–10.
Sequential investment creates an option, while Adelman, M. A., and Watkins, G. C. (1997). The value of United
dependence creates the information spillover that States oil and gas reserves: Estimation and application. Adv.
gives value to the option. Although our example is Econ. Energy Res. 10, 131–184.
Dixit, A. K., and Pindyck, R. S. (1995). The options approach to
very simple, the phenomenon it describes is quite capital investment. Harvard Business Rev. 73(3), 105–118.
general. Similar results are obtained whether the Jacoby, H. D., and Laughton, D. G. (1992). Project evaluation:
dependence among properties is complete or partial, A practical asset pricing method. Energy J. 13(2), 19–47.
symmetric or asymmetric, positive or negative. In all Lerche, I., and MacKay, J. A. (1999). ‘‘Economic Risk in
such cases, the sum of property values reckoned Hydrocarbon Exploration.’’ Academic Press, San Diego, CA.
McCray, A. W. (1975). ‘‘Petroleum Evaluations and Economic
individually provides only a minimum valuation for Decisions.’’ Prentice-Hall, Englewood Cliffs, NJ.
the combined portfolio. Paddock, J. L., Siegel, D. R., and Smith, J. L. (1988). Option
valuation of claims on real assets: The case of offshore
petroleum leases. Quarterly J. Economics 103(3), 479–508.
6. CONCLUSION Pickles, E., and Smith, J. L. (1993). Petroleum property valuation:
A binomial lattice implementation of option pricing theory.
Energy J. 14(2), 1–26.
Uncertainties that range from measurable price Pindyck, R. S. (1999). The long-run evolution of energy prices.
volatility to seemingly imponderable geological Energy J. 20(2), 1–27.